WO2020078251A1 - 指示和确定预编码向量的方法以及通信装置 - Google Patents

指示和确定预编码向量的方法以及通信装置 Download PDF

Info

Publication number
WO2020078251A1
WO2020078251A1 PCT/CN2019/110342 CN2019110342W WO2020078251A1 WO 2020078251 A1 WO2020078251 A1 WO 2020078251A1 CN 2019110342 W CN2019110342 W CN 2019110342W WO 2020078251 A1 WO2020078251 A1 WO 2020078251A1
Authority
WO
WIPO (PCT)
Prior art keywords
space
vector
frequency
vectors
matrix
Prior art date
Application number
PCT/CN2019/110342
Other languages
English (en)
French (fr)
Inventor
王潇涵
金黄平
毕晓艳
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN201811281059.7A external-priority patent/CN111064499B/zh
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to BR112021007195-4A priority Critical patent/BR112021007195A2/pt
Priority to KR1020217014727A priority patent/KR102495785B1/ko
Priority to JP2021521259A priority patent/JP7210718B2/ja
Priority to EP19872662.2A priority patent/EP3876429A4/en
Publication of WO2020078251A1 publication Critical patent/WO2020078251A1/zh
Priority to US17/230,523 priority patent/US11456786B2/en
Priority to US17/900,501 priority patent/US11848729B2/en
Priority to US18/502,785 priority patent/US20240072858A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0456—Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting

Definitions

  • the present application relates to the field of wireless communication, and more specifically, to a method and communication device for indicating and determining a precoding vector.
  • Massive MIMO massive multiple-input multiple-output
  • network equipment can reduce interference between multiple users and interference between multiple signal streams of the same user through precoding. It is conducive to improving signal quality, realizing space division multiplexing, and improving spectrum utilization.
  • the terminal device may determine the precoding vector by way of channel measurement, for example, and hopes that through feedback, the network device obtains the same or similar precoding vector as the precoding vector determined by the terminal device.
  • the terminal device may indicate the precoding vector to the network device through a two-stage feedback method of broadband feedback and subband feedback. Specifically, the terminal device may indicate the selected beam vector and the quantized value of the broadband amplitude coefficient of each beam vector through wideband feedback based on each transmission layer, and may indicate the quantization of the combined coefficients available for each subband through subband feedback Value, where the combination coefficient may include, for example, a sub-band amplitude coefficient and a sub-band phase coefficient.
  • the network device can synthesize the broadband feedback information and the subband feedback information to recover the precoding vector corresponding to each subband.
  • the terminal device can indicate the precoding vector to the network device through the two-level feedback method of broadband feedback and subband feedback. Refer to the type II (type II) codebook feedback method defined in the new radio (NR) protocol TS 38.214 .
  • the present application provides a method and a communication device for indicating and determining a precoding vector, in order to reduce feedback overhead.
  • a method for indicating a precoding vector is provided.
  • the method may be executed by a terminal device, or may be executed by a chip configured in the terminal device.
  • the method includes: generating first indication information; and sending the first indication information.
  • a method for determining a precoding vector is provided.
  • the method may be executed by a network device, or may be executed by a chip configured in the network device.
  • the method includes: receiving first indication information.
  • the T 1 space frequency component matrices are selected from M 1 space frequency component matrices, and the M 1 space frequency component matrices are determined by L 1 beam vectors and K 1 frequency domain vectors.
  • the null frequency component T 1 is determined by the matrix M 1 empty frequency vectors T 1 in the empty vector frequency, the frequency of the empty vector M 1 L 1 of the beams and vectors K 1 frequency domain vectors are combined, and each space frequency vector pair is uniquely determined by one beam vector of L 1 beam vectors and one frequency domain vector of K 1 frequency domain vectors.
  • the null frequency component T 1 matrix may exhibit frequency T 1 empty vector combination of T 1 and T 1 vector beams frequency-domain vectors, the T 1 empty vector Frequency The pair is selected from M 1 space frequency vector pairs formed by L 1 beam vectors and K 1 frequency domain vectors.
  • the terminal device indicates a small number of beam vectors, frequency domain vectors, and space-frequency component matrices to the network device, so that the network device can recover the precoding vector.
  • the frequency domain vector can be used to describe the different changing rules of the channel in the frequency domain.
  • the terminal device can simulate the change of the channel in the frequency domain through the linear superposition of one or more frequency domain vectors, fully exploit the relationship between the frequency domain units, utilize the continuity of the frequency domain, and use fewer frequency domain vectors To describe the changing rules of multiple frequency domain units.
  • the feedback overhead will not increase exponentially. Therefore, feedback overhead can be greatly reduced on the basis of ensuring feedback accuracy.
  • the frequency domain vector set may also contain more frequency domain vectors. If the beam vector set and the frequency domain vector set are indicated directly with fewer beam vectors and frequencies The domain vector may bring a large bit overhead, or the terminal device and the network device need to predefine a large number of beam vector combination and index correspondence and a frequency domain vector combination and index correspondence.
  • the terminal device narrows the selection range of the space-frequency component matrix used for weighted summation to the range of M 1 space-frequency component matrix constructed by L 1 beam vector and K 1 frequency domain vector , That is, first select a smaller range from the existing vector set, and then further select and indicate T 1 space-frequency component matrices from the range.
  • it can avoid directly indicating the large feedback overhead caused by the T 1 space-frequency component matrices; on the other hand, it can avoid saving a large number of correspondences between the terminal device and the network device.
  • the T 1 beam vectors are partial beam vectors selected from the L 1 beam vectors, but it does not mean that T 1 must be less than L 1. There may be some or all of the T 1 beam vectors. Is reused. Therefore, the number of beam vectors used to obtain T 1 beam vector pairs is T 1 .
  • the T 1 frequency domain vectors are partial beam vectors selected from the K 1 beam vectors, but this does not mean that T 1 must be less than K 1. There may be some or all of the frequency in the T 1 frequency domain vectors The domain vector is reused. Therefore, the number of frequency domain vectors used to obtain T 1 frequency domain vector pairs is T 1 . For the sake of brevity, the description of the same or similar cases is omitted below.
  • the method further includes: receiving second indication information, where the second indication information is used to indicate one or more of M 1 , L 1, and K 1 Value.
  • the method further includes: sending second indication information, where the second indication information is used to indicate one of M 1 , L 1, and K 1 or The value of multiple items.
  • the value of one or more of M 1 , L 1 and K 1 may be indicated by the network device.
  • the method further includes: sending second indication information, where the second indication information is used to indicate one or more of M 1 , L 1, and K 1 Value.
  • the method further includes: receiving second indication information, where the second indication information is used to indicate one of M 1 , L 1, and K 1 or The value of multiple items.
  • the value of one or more of M 1 , L 1 and K 1 may be determined by the terminal device and reported to the network device.
  • M 1 , L 1 and K 1 may also be defined in advance, as defined in the protocol, which is not limited in this application.
  • the method further includes: receiving third indication information, where the third indication information is used to indicate the value of T 1 .
  • the value of T 1 may be indicated by the network device.
  • the method further includes: sending third indication information, where the third indication information is used to indicate the value of T 1 .
  • the value of T 1 can be determined by the terminal device and reported to the network device.
  • T 1 may also be defined in advance, such as the protocol definition, which is not limited in this application.
  • the first indication information includes position information of the L 1 beam vectors in the beam vector set and the K 1 frequency domain vectors in the frequency domain vector Location information in the collection.
  • the M 1 can be determined Matrix of space-frequency components.
  • the M 1 space frequency component matrices can be indicated by a two-dimensional index.
  • the M 1 space frequency component matrices are selected from a set of space frequency component matrices or a subset of a set of space frequency component matrices, and the space frequency component matrix is determined by the beam vector
  • Each beam vector in the set and each frequency domain vector in the set of frequency domain vectors are determined, and each space frequency component matrix in the set of space frequency component matrices is composed of a beam vector in the set of beam vectors and the frequency domain vector A frequency domain vector in the set is uniquely determined.
  • the first indication information includes position information of the M 1 space frequency component matrices in the space frequency component matrix set, or position information of the M 1 space frequency component matrices in a subset of the space frequency component matrix set .
  • the M 1 space frequency component matrices can be indicated by a one-dimensional index.
  • M 1 space-frequency component matrices are introduced for ease of understanding in this article. This does not mean that the terminal device must generate the M 1 space-frequency component matrix.
  • the terminal device can also use the L 1 beams
  • the vector and K 1 frequency domain vectors are combined to obtain M 1 space frequency vector pairs.
  • the M 1 space frequency component matrix can be constructed. In other words, M 1 space-frequency vector pairs and M 1 space-frequency component matrices can be converted into each other. Therefore, it can be considered that the M 1 space frequency component matrices correspond to L 1 beam vectors and K 1 frequency domain vectors.
  • each space frequency component matrix in the M 1 space frequency component matrices is converted by the conjugate of a beam vector in the L 1 beam vectors and a frequency domain vector in the K 1 frequency domain vectors The product is determined.
  • each space-frequency component matrix of the M 1 space-frequency component matrices is composed of a Krone of a frequency domain vector of the K 1 frequency-domain vectors and a beam vector of the L 1 beam-vectors Clear product determination.
  • each space-frequency component matrix may be determined by the product of a beam vector and a frequency domain vector.
  • each space frequency component matrix may be determined by the Kronecker product of a beam vector and a frequency domain vector. This application does not limit this.
  • the first indication information may indicate T 1 space-frequency component matrices (or T 1 space-frequency vector pairs) in any of the following ways:
  • Method 1 Indicating T 1 space-frequency component matrices among M 1 space-frequency component matrices through a bitmap; or
  • Method 2 Indicate the index of the combination of T 1 space-frequency component matrices in the M 1 space-frequency component matrix
  • Manner 3 indicating the position of the beam vector corresponding to each space frequency component matrix in the T 1 space frequency component matrices in L 1 beam vectors and the position of the frequency domain vector in K 1 frequency domain vectors;
  • T 1 indicates empty space frequency components of each frequency component of the matrix in the matrix M 1 empty frequency component matrix index.
  • T 1 indicates an empty frequency component matrix
  • T 1 may be empty selected frequency range component matrix is reduced, it is possible to reduce the feedback overhead of the T 1 empty frequency component by component matrix in the matrix M 1 in the empty frequency.
  • the weighted sum of the T 1 space frequency component matrices is used to determine the precoding vectors of one or more frequency domain units on the first transmission layer.
  • the first transmission layer may be one transmission layer or multiple transmission layers.
  • the method further includes: generating fourth indication information, where the fourth indication information is used to indicate L 2 beam vectors in the beam vector set and the frequency domain vector set K 2 frequency domain vectors in T and T 2 space frequency component matrices, the weighted sum of the T 2 space frequency component matrices is used to determine the precoding vectors of one or more frequency domain units on the second transmission layer; wherein, The L 2 beam vectors and the K 2 frequency domain vectors correspond to M 2 space frequency component matrices.
  • the T 2 space frequency component matrices are partial space frequency component matrices in the M 2 space frequency component matrices.
  • the method further includes: receiving fourth indication information, where the fourth indication information is used to indicate L 2 beam vectors and T 2 space frequencies in the beam vector set Component matrix, the weighted sum of the T 2 space-frequency component matrices is used to determine the precoding vectors of one or more frequency domain units on the second transmission layer; wherein, the L 2 beam vectors and the K 2 frequency domain vectors and M 2 matrix corresponding to empty frequency component, the frequency component of T 2 empty empty matrix is a part of the frequency component M 2 empty matrix matrix frequency component, the frequency component M 2 matrix empty space of each frequency component
  • the L 2 beam vectors are the beam vector set Partial beam vectors in, and / or, the K 2 frequency domain vectors are partial frequency domain vectors in the set of frequency domain vectors, and M 2 , L 2 , K 2 and T
  • the second transmission layer may be one or more transmission layers except the first transmission layer among the multiple transmission layers.
  • the terminal device may indicate to the network device a precoding vector used to determine one or more frequency domain units on the second transmission layer.
  • L 1 L 2
  • K 1 K 2
  • T 1 T 2 .
  • a communication device including various modules or units for performing the method in any possible implementation manner of the first aspect.
  • a communication device including a processor.
  • the processor is coupled to the memory and can be used to execute instructions in the memory to implement the method in any possible implementation manner of the first aspect.
  • the communication device further includes a memory.
  • the communication device further includes a communication interface, and the processor is coupled to the communication interface.
  • the communication device is a terminal device.
  • the communication interface may be a transceiver or an input / output interface.
  • the communication device is a chip configured in the terminal device.
  • the communication interface may be an input / output interface.
  • the transceiver may be a transceiver circuit.
  • the input / output interface may be an input / output circuit.
  • a communication device including various modules or units for performing the method in any possible implementation manner of the second aspect.
  • a communication device including a processor.
  • the processor is coupled to the memory and can be used to execute instructions in the memory to implement the method in any possible implementation manner of the second aspect.
  • the communication device further includes a memory.
  • the communication device further includes a communication interface, and the processor is coupled to the communication interface.
  • the communication device is a network device.
  • the communication interface may be a transceiver or an input / output interface.
  • the communication device is a chip configured in a network device.
  • the communication interface may be an input / output interface.
  • the transceiver may be a transceiver circuit.
  • the input / output interface may be an input / output circuit.
  • a processor including: an input circuit, an output circuit, and a processing circuit.
  • the processing circuit is configured to receive an input signal through the input circuit and output a signal through the output circuit so that the processor performs the first aspect or the second aspect and any possible implementation of the first aspect or the second aspect The way in the way.
  • the processor may be a chip
  • the input circuit may be an input pin
  • the output circuit may be an output pin
  • the processing circuit may be a transistor, a gate circuit, a flip-flop, and various logic circuits.
  • the input signal received by the input circuit may be received and input by, for example, but not limited to, the receiver
  • the signal output by the output circuit may be, for example, but not limited to, output to and transmitted by the transmitter
  • the circuit may be the same circuit, which is used as an input circuit and an output circuit at different times, respectively.
  • the embodiments of the present application do not limit the specific implementation manner of the processor and various circuits.
  • a processing device including a processor and a memory.
  • the processor is used to read instructions stored in the memory, and can receive signals through the receiver and transmit signals through the transmitter to perform the first aspect or the second aspect and any possible implementation manner of the first aspect or the second aspect Methods.
  • processors there are one or more processors and one or more memories.
  • the memory may be integrated with the processor, or the memory and the processor are provided separately.
  • the memory may be non-transitory (non-transitory) memory, such as read-only memory (read only memory, ROM), which may be integrated with the processor on the same chip, or may be set in different On the chip, the embodiments of the present application do not limit the type of memory and the manner of setting the memory and the processor.
  • non-transitory memory such as read-only memory (read only memory, ROM)
  • ROM read only memory
  • sending instruction information may be a process of outputting instruction information from the processor
  • receiving capability information may be a process of receiving input capability information by the processor.
  • the processed output data may be output to the transmitter, and the input data received by the processor may come from the receiver.
  • the transmitter and the receiver may be collectively referred to as a transceiver.
  • the processing device in the eighth aspect may be one or more chips, and the processor may be implemented by hardware or software.
  • the processor When implemented by hardware, the processor may be a logic circuit, an integrated circuit, etc .;
  • the processor When implemented by software, the processor may be a general-purpose processor, implemented by reading software codes stored in a memory, the memory may be integrated in the processor, and may be located outside the processor and exist independently.
  • a computer program product includes: a computer program (also referred to as code or instructions) that, when the computer program is executed, causes the computer to perform the first aspect or the above The method in the second aspect and any possible implementation manner of the first aspect or the second aspect.
  • a computer program also referred to as code or instructions
  • a computer-readable medium that stores a computer program (also may be referred to as code or instructions) that when executed on a computer, causes the computer to perform the first aspect or the above
  • a computer program also may be referred to as code or instructions
  • a communication system including the foregoing network device and terminal device.
  • FIG. 1 is a schematic diagram of a communication system applicable to the method for indicating and determining a precoding vector provided by an embodiment of the present application;
  • FIG. 2 is a schematic flowchart of a method for indicating and determining a precoding vector provided by an embodiment of the present application
  • FIG. 3 is a schematic flowchart of a method for feeding back a precoding matrix indicator (precoding matrix indicator (PMI)) provided by another embodiment of the present application;
  • PMI precoding matrix indicator
  • FIG. 4 is a schematic block diagram of a communication device provided by an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of a terminal device provided by an embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of a network device provided by an embodiment of the present application.
  • GSM global mobile communication
  • CDMA code division multiple access
  • WCDMA broadband code division multiple access
  • general packet radio service general packet radio service, GPRS
  • LTE long term evolution
  • LTE frequency division duplex FDD
  • TDD time division duplex
  • UMTS universal mobile communication system
  • WiMAX worldwide interoperability for microwave access
  • FIG. 1 is a schematic diagram of a communication system 100 suitable for a method of indicating a precoding vector according to an embodiment of the present application.
  • the communication system 100 may include at least one network device, such as the network device 110 shown in FIG. 1; the communication system 100 may also include at least one terminal device, such as the terminal device 120 shown in FIG. 1.
  • the network device 110 and the terminal device 120 can communicate through a wireless link.
  • Each communication device, such as the network device 110 or the terminal device 120 may be configured with multiple antennas.
  • the configured multiple antennas may include at least one transmit antenna for transmitting signals and at least one receive antenna for receiving signals. Therefore, the communication devices in the communication system 100, such as the network device 110 and the terminal device 120, can communicate through multi-antenna technology.
  • the network device in the communication system may be any device with wireless transceiver function.
  • the network equipment includes but is not limited to: evolved Node B (evolved Node B, eNB), radio network controller (radio network controller, RNC), node B (Node B, NB), base station controller (base station controller, BSC) ), Base transceiver station (BTS), home base station (eg, home evolved NodeB, or home Node B, HNB), baseband unit (BBU), wireless fidelity (WiFi) system Access point (AP), wireless relay node, wireless backhaul node, transmission point (transmission point, TP) or sending and receiving point (transmission and reception point, TRP), etc.
  • 5G such as, NR, gNB in the system, or transmission point (TRP or TP), one or a group (including multiple antenna panels) of the base station in the 5G system, or it can also be a network node that constitutes a gNB or transmission point ,
  • NR gNB
  • gNB may include a centralized unit (CU) and DU.
  • the gNB may also include a radio unit (RU).
  • the CU implements some functions of gNB, and the DU implements some functions of gNB.
  • CU implements the functions of radio resource control (RRC), packet data convergence layer protocol (packet data convergence protocol, PDCP) layer
  • DU implements radio link control (radio link control (RLC), media access control (media access control, MAC) and physical (physical, PHY) layer functions.
  • RRC radio resource control
  • PDCP packet data convergence layer protocol
  • DU implements radio link control (radio link control (RLC), media access control (media access control, MAC) and physical (physical, PHY) layer functions.
  • RRC radio resource control
  • PDCP packet data convergence layer protocol
  • DU implements radio link control (radio link control (RLC), media access control (media access control, MAC) and physical (physical, PHY) layer functions.
  • RLC radio link
  • the network device may be a CU node, or a DU node, or a device including a CU node and a DU node.
  • the CU can be divided into network devices in the radio access network (RAN), and can also be divided into network devices in the core network (CN), which is not limited in this application.
  • RAN radio access network
  • CN core network
  • the terminal equipment in the wireless communication system may also be referred to as user equipment (UE), access terminal, subscriber unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, User terminal, terminal, wireless communication device, user agent or user device.
  • the terminal devices in the embodiments of the present application may be mobile phones, tablet computers, computers with wireless transceiver functions, virtual reality (VR) terminal devices, and augmented reality (AR) terminals Wireless terminals in equipment, industrial control (industrial control), wireless terminals in self-driving (self-driving), wireless terminals in remote medical (remote medical), wireless terminals in smart grid (smart grid), transportation safety ( wireless terminals in transportation, wireless terminals in smart cities, wireless terminals in smart homes, etc.
  • the embodiments of the present application do not limit the application scenarios.
  • FIG. 1 is only a simplified schematic diagram for ease of understanding and examples.
  • the communication system 100 may further include other network devices or other terminal devices, which are not shown in FIG. 1.
  • the processing procedure of the downlink signal at the physical layer before sending may be performed by a network device, or may be performed by a chip configured in the network device. For ease of explanation, they are collectively referred to as network devices below.
  • Network devices can process codewords on physical channels.
  • the codeword may be coded bits that have been coded (eg, including channel coding).
  • the scrambling of the codeword generates scrambling bits.
  • the scrambled bits undergo modulation mapping to obtain modulation symbols.
  • the modulation symbols are mapped to multiple layers (layers) through layer mapping, or transmission layers.
  • the modulation symbols after layer mapping are subjected to precoding to obtain a precoded signal.
  • the pre-encoded signal is mapped to multiple REs after being mapped to resource elements (RE). These REs are then orthogonally multiplexed (orthogonal frequency division multiplexing, OFDM) modulated and transmitted through the antenna port.
  • OFDM orthogonally multiplexed
  • the network device can process the signal to be transmitted with the help of a precoding matrix that matches the channel resource when the channel state is known, so that the precoded signal to be transmitted is adapted to the channel, thereby Reduces the complexity of the receiving device to eliminate the influence between channels. Therefore, through the precoding process of the signal to be transmitted, the received signal quality (for example, signal to interference plus noise ratio (SINR), etc.) can be improved. Therefore, by using precoding technology, transmission devices and multiple receiving devices can be transmitted on the same time-frequency resources, that is, multiple users, multiple inputs, and multiple outputs (MU-MIMO).
  • SINR signal to interference plus noise ratio
  • the sending device may also perform precoding in other ways. For example, when channel information (such as but not limited to a channel matrix) cannot be obtained, pre-coding is performed using a pre-coding matrix or a weighting processing method set in advance. For the sake of brevity, the specific content of this article will not be repeated here.
  • Precoding matrix indicator can be used to indicate the precoding matrix.
  • the precoding matrix may be, for example, a precoding matrix determined by the terminal device based on the channel matrix of each subband.
  • the channel matrix may be determined by the terminal device through channel estimation or the like or based on channel reciprocity.
  • the specific method for the terminal device to determine the precoding matrix is not limited to the above, and the specific implementation may refer to the existing technology. For brevity, it is not listed here one by one.
  • the precoding matrix can be obtained by singular value decomposition (SVD) of the channel matrix or the covariance matrix of the channel matrix, or it can also be obtained by eigenvalue decomposition (eigenvalue) of the covariance matrix of the channel matrix decopomsition, EVD).
  • SVD singular value decomposition
  • eigenvalue eigenvalue
  • EVD covariance matrix of the channel matrix decopomsition
  • the terminal device can quantize the precoding matrix of each subband, and can send the quantized value to the network device through the PMI, so that the network device determines the precoding matrix that is the same or close to the precoding matrix determined by the terminal device according to the PMI .
  • the network device can directly determine the precoding matrix of each subband according to the PMI, or determine the precoding matrix of each subband according to the PMI and then perform further processing, such as orthogonalizing the precoding matrix of different users To determine the final precoding matrix. Therefore, the network device can determine a precoding matrix suitable for the channel of each subband to perform precoding processing on the signal to be transmitted.
  • the specific method for the network device to determine the precoding matrix used in each subband according to the PMI can refer to the prior art. Here, it is only an example for ease of understanding, and should not constitute any limitation to this application.
  • the precoding matrix determined by the terminal device can be understood as the precoding matrix to be fed back.
  • the terminal device may indicate the precoding matrix to be fed back through the PMI, so that the network device can recover the precoding matrix based on the PMI. It can be understood that the precoding matrix recovered by the network device based on the PMI may be the same as or similar to the precoding matrix to be fed back.
  • the following shows a simple example of a precoding matrix through two-stage feedback when the rank is 1.
  • W represents a transmission layer, a subband, and a precoding matrix to be fed back in two polarization directions.
  • W 1 can be fed back through broadband
  • W 2 can be fed back through subband.
  • v 0 to v 3 are beam vectors included in W 1 , and the plurality of beam vectors may be indicated by an index of a combination of the plurality of beam vectors, for example.
  • the precoding matrix shown above the beam vectors in the two polarization directions are the same, and beam vectors v 0 to v 3 are used.
  • a 0 to a 7 are the broadband amplitude coefficients included in W 1 , and can be indicated by the quantized value of the broadband amplitude coefficients.
  • c 0 to c 7 are the sub-band coefficients included in W 2 , and each sub-band coefficient may include a sub-band amplitude coefficient and a sub-band phase coefficient.
  • c 0 to c 7 may include sub-band amplitude coefficients ⁇ 0 to ⁇ 7 and sub-band phase coefficients to And can pass the quantized values of sub-band amplitude coefficients ⁇ 0 to ⁇ 7 and the sub-band phase coefficients to Quantized value to indicate. It can be seen that the precoding matrix to be fed back can be regarded as a weighted sum of multiple beam vectors.
  • the precoding matrix shown above is obtained based on feedback from one transmission layer, and therefore may also be referred to as a precoding vector.
  • the terminal device may separately feedback based on each transmission layer.
  • the precoding vectors fed back from each transmission layer can be constructed to obtain a subband precoding matrix.
  • the number of transmission layers is 4, and the precoding matrix may include 4 precoding vectors, which respectively correspond to 4 transmission layers.
  • the feedback overhead of the terminal equipment also increases. For example, when the number of transmission layers is 4, the feedback overhead of a 0 to a 7 and c 0 to c 7 will be up to 4 times that of one transmission layer. That is to say, if the terminal device performs broadband feedback and subband feedback as described above based on each transmission layer, as the number of transmission layers increases, the feedback overhead caused will increase exponentially. The greater the number of subbands, the greater the increase in feedback overhead. Therefore, it is desirable to provide a method that can reduce the feedback overhead of PMI.
  • the manner of feeding back the precoding matrix through PMI listed above is only an example, and should not constitute any limitation to this application.
  • the terminal device may also feed back the channel matrix to the network device through PMI.
  • the network device may determine the channel matrix according to the PMI, and then determine the precoding matrix, which is not limited in this application.
  • the precoding vector may be determined by a vector in the precoding matrix, for example, a column vector.
  • the precoding matrix may include one or more column vectors, and each column vector may be used to determine a precoding vector.
  • the precoding matrix may also be called a precoding vector.
  • the precoding matrix may be determined by precoding vectors of one or more transmission layers, and each vector in the precoding matrix may correspond to a transmission layer. Assuming that the dimension of the precoding vector can be N 1 ⁇ 1, if the number of transmission layers is R (R is a positive integer), the dimension of the precoding matrix can be N 1 ⁇ R.
  • the number of transmission layers may be indicated by a rank indicator (RI), N 1 may represent the number of antenna ports, and N 1 is a positive integer.
  • the precoding vector may also refer to the components of the precoding matrix in one transmission layer and one polarization direction. Assuming that the number of polarization directions is P (P is a positive integer) and the number of antenna ports in one polarization direction is N 2 , then the dimension of the precoding vector corresponding to one transmission layer is (P ⁇ N 2 ) ⁇ 1, then one The dimension of the precoding vector in the polarization direction may be N 2 ⁇ 1, and N 2 is a positive integer.
  • the precoding vector may correspond to one transmission layer, may also correspond to one polarization direction on one transmission layer, and may also correspond to other parameters. This application does not limit this.
  • Antenna port can be referred to as port. It can be understood as a transmitting antenna recognized by the receiving device, or a transmitting antenna that can be distinguished in space.
  • One antenna port can be configured for each virtual antenna, each virtual antenna can be a weighted combination of multiple physical antennas, and each antenna port can correspond to a reference signal, therefore, each antenna port can be referred to as a reference signal port For example, CSI-RS port, sounding reference signal (SRS) port, etc.
  • SRS sounding reference signal
  • Beam can be understood as the distribution of signal strength formed in a certain direction in space.
  • the technique of forming a beam may be a beam forming (or beamforming) technique or other techniques.
  • the beamforming technology may specifically be a digital beamforming technology, an analog beamforming technology, and a hybrid digital / analog beamforming technology.
  • the beam may be formed by digital beamforming technology.
  • the beam vector may correspond to the beam, may be a precoding vector in a precoding matrix, or may be a beam forming vector.
  • Each element in the beam vector may represent the weight of each antenna port.
  • the weighted signals of each antenna port are superimposed on each other to form an area with strong signal strength.
  • the beam vector may also be referred to as a spatial domain vector.
  • the length (or dimension) of the beam vector is the number of antenna ports in one polarization direction.
  • the beam vector may be a column vector with a dimension of N s ⁇ 1, or a row vector with a dimension of 1 ⁇ N s . This application does not limit this.
  • Frequency domain unit A unit of frequency domain resources, which can represent different frequency domain resource granularities.
  • the frequency domain unit may include, for example but not limited to, subband, resource block (resource block (RB), subcarrier, resource block group (RBG), precoding resource block group (PRG), etc. .
  • Frequency domain vector a vector used in the embodiment of the present application to represent the change rule of the channel in the frequency domain.
  • the frequency domain vector can specifically be used to represent the change rule of the weighting coefficient of each beam vector on each frequency domain unit. This change may be related to multipath delay. Because the signal is transmitted through the wireless channel, there may be different transmission delays on different propagation paths. Therefore, different frequency domain vectors can be used to characterize the variation of delay on different propagation paths.
  • the dimension of the frequency domain vector may be the number of frequency domain units that require CSI measurement. Since the number of frequency domain units that need to perform CSI measurement at different times may be different, the dimension of the frequency domain vector may also change. In other words, the dimension of the frequency domain vector is variable.
  • the length (or dimension) of the frequency domain vector is the number of frequency domain units included in the frequency domain occupied bandwidth of the CSI measurement resource.
  • the frequency domain occupied bandwidth of the CSI measurement resource may be the bandwidth used for transmission of the reference signal, and the reference signal mentioned herein may be a reference signal used for channel measurement, such as CSI-RS.
  • the frequency domain occupied bandwidth of the CSI measurement resource may be, for example, less than or equal to the pilot transmission bandwidth (or measurement bandwidth).
  • the frequency domain occupied bandwidth used to indicate the CSI measurement resource may be, for example, the CSI occupied bandwidth range (CSI-Frequency Occupation).
  • the length of the frequency domain vector is the length of signaling used to indicate the position and number of frequency domain units to be reported.
  • the signaling used to indicate the position and number of frequency domain units to be reported may be reporting bandwidth (reporting bandwidth).
  • the signaling can indicate the position and number of frequency domain units to be reported in the form of a bitmap. Therefore, the dimension of the frequency domain vector can be the number of bits in the bitmap.
  • the length of the frequency domain vector is the number of frequency domain units to be reported.
  • the number of frequency domain units to be reported can be indicated by the above-mentioned signaling reporting bandwidth, for example.
  • the number of frequency domain units to be reported may be all frequency domain units in the frequency domain occupied bandwidth of the CSI measurement resource, or part of the frequency domain units in the bandwidth occupied by the frequency domain of the CSI measurement resource; or, to be reported
  • the number of frequency domain units can be the same as the signaling length of the reported bandwidth, or it can be less than the signaling length of the reported bandwidth. This application does not limit this.
  • the indication information for the length of the frequency domain vector is recorded as the fifth indication information.
  • the fifth indication information may be the above-mentioned signaling used to indicate the frequency domain occupied bandwidth of the CSI measurement resource, may also be used to indicate the location and number of frequency domain units to be reported, or may be the future
  • the newly added signaling in the protocol is not limited in this application.
  • the frequency domain vector may be a column vector with a dimension of N f ⁇ 1, or a row vector with a dimension of 1 ⁇ N f . This application does not limit this.
  • Space-frequency matrix and space-frequency component matrix For convenience of explanation, it is assumed that the number of polarization directions of the transmitting antenna is 1.
  • the space frequency matrix in the polarization direction can be constructed from the precoding vectors on different frequency domain units on a transmission layer.
  • the terminal device may determine the precoding matrix to be fed back on each frequency domain unit through channel measurement or the like, and the precoding matrix to be fed back on each frequency domain unit is processed to obtain The space-frequency matrix corresponding to the layer. For example, for the same transmission layer, a pre-coding vector to be fed back on each frequency domain unit can be combined to obtain a space-frequency matrix.
  • the space-frequency matrix may be called a space-frequency matrix to be fed back.
  • the terminal device may indicate the space-frequency matrix to be fed back through the weighted sum of one or more space-frequency component matrices. In other words, the space-frequency matrix to be fed back can be approximated as a weighted sum of one or more space-frequency component matrices.
  • the one or more space-frequency component matrices can be taken from a predefined set of space-frequency component matrices, or can be determined by the beam vector in the predefined set of beam vectors and the frequency domain vector in the predefined set of frequency domain vectors . This application does not limit this.
  • the space-frequency matrix may be a matrix with dimensions N s ⁇ N f . That is, the space-frequency matrix may include N f column vectors of length N s . The N f column vectors may correspond to N f frequency domain units, and each column vector may be used to determine the corresponding precoding vector of the frequency domain unit.
  • the space-frequency matrix can be written as H, Where h 0 to Are N f column vectors corresponding to N f frequency domain units, and the length of each column vector can be N s .
  • the N f column vectors can be used to determine the precoding vectors of N f frequency domain units, respectively.
  • the space-frequency matrix can be approximated as a weighted sum of one or more space-frequency component matrices.
  • a space frequency component matrix can be uniquely determined by a beam vector and a frequency domain vector.
  • a space-frequency component matrix can be the product of the conjugate transpose of a beam vector and a frequency domain vector; when the beam vector is a column vector and the frequency domain vector is a row
  • a space-frequency component matrix can be the product of a beam vector and a frequency domain vector. Therefore, each space-frequency component matrix may also be a matrix of dimension N s ⁇ N f .
  • the space-frequency matrix may be a matrix of dimension (N s ⁇ N f ) ⁇ 1, or it can also be said that the space-frequency matrix may be a vector of length N s ⁇ N f .
  • the space-frequency matrix may include only one column vector of length N s ⁇ N f .
  • the space-frequency matrix may also be called a space-frequency vector.
  • the space-frequency matrix can be written as H, Each of these vectors has been described in detail above, and for the sake of brevity, it will not be repeated here.
  • the space frequency vector may be approximated as a weighted sum of one or more space frequency component vectors.
  • a space frequency component vector can be uniquely determined by a beam vector and a frequency domain vector.
  • a space-frequency component vector can be the Kronecker product of a beam vector and a frequency domain vector, or it can be the gram of a frequency domain vector and a beam vector. Roneck product. Therefore, each space-frequency component vector may also be a vector of length N s ⁇ N f . In this case, the space-frequency component matrix may also be called a space-frequency component vector.
  • the space-frequency vector determined by the weighted sum of multiple space-frequency component vectors may be sequentially composed of N f column vectors of length N s Connected.
  • the N f column vectors may correspond to N f frequency domain units, and each column vector may be used to determine the corresponding precoding vector of the frequency domain unit.
  • the space-frequency vector determined by the weighted sum of multiple space-frequency component vectors may be sequentially composed of N s column vectors of length N f Connected.
  • N f elements in each column vector may correspond to N f frequency domain units.
  • the n f elements in each column vector of the N s column vectors are connected in sequence to obtain a vector of length N s , which can be used to determine the precoding vector of the n f frequency domain unit.
  • N s N f
  • the space-time frequency matrix with the number of polarization directions being 1, for example, a matrix with dimensions N s ⁇ N f or a vector with length N s ⁇ N f , but This should not constitute any limitation on this application.
  • the space-frequency matrix can still be expressed in several forms listed above, but the dimensions of the space-frequency matrix may also be different depending on the number of polarization directions.
  • the space-frequency matrix may be a matrix with a dimension of 2N s ⁇ N f or a vector with a length of 2N s ⁇ N f .
  • 2 indicates two polarization directions.
  • the space-frequency component matrix can still be a matrix of dimension N s ⁇ N f or a vector of length N s ⁇ N f . Therefore, the space-frequency matrix in each polarization direction can be represented by the weighted sum of multiple space-frequency component matrices. In other words, the space-frequency matrix in each polarization direction can be approximately expressed as a weighted sum of multiple space-frequency component matrices.
  • the multiple space-frequency component matrices used for different polarization directions may be the same, or in other words, the same multiple space-frequency component matrices may be shared by multiple polarization directions.
  • the space-frequency matrix or space-frequency vector of multiple polarization directions on the same transmission layer can be constructed by the same set of beam vectors and the same set of frequency domain vectors.
  • the weighting coefficients of the space-frequency component matrices in different polarization directions may be different.
  • the basic unit obtainable by the operation of a beam vector and a frequency domain vector may be a space-frequency basic unit, for example, it may be a space-frequency component matrix or a space-frequency component vector.
  • the basic unit of space frequency may correspond to a polarization direction.
  • the weighted sum of this basic unit can splice multiple space-frequency matrices of polarization directions.
  • the spatial frequency component matrix is the product of the conjugate transpose of the beam vector and the frequency domain vector
  • the spatial frequency component matrix is the Kronecker product of the frequency domain vector and the beam vector.
  • the form specifies the specific process in which the terminal device instructs the precoding vector and the network device determines the precoding vector. But this should not constitute any limitation on this application. Based on the same conception, those skilled in the art can perform equivalent transformation or replacement on the above-mentioned space-frequency component matrix, and these equivalent transformations and replacements should all fall within the protection scope of the present application.
  • network devices can transmit data to terminal devices through multiple transmission layers.
  • the overhead caused by the terminal device performing feedback based on each transmission layer will also increase exponentially.
  • this application provides a method for indicating and determining a precoding vector, in order to reduce the feedback overhead of PMI.
  • the number of polarization directions of the transmitting antenna is P (P ⁇ 1 and an integer), and the number of transmission layers is R (R ⁇ 1 and an integer).
  • the R transmission layers may include the 0th to R-1 transmission layers
  • the P polarization directions may include the 0th to P-1 polarization directions.
  • consecutive numbers may be started from 1.
  • T indicates the transpose
  • a T indicates the transpose of the matrix (or vector) A
  • the superscript * indicates the conjugate transpose, for example, A * indicates the conjugate transpose of the matrix (or vector) A.
  • the beam vector and the frequency domain vector are both column vectors as an example to illustrate the embodiment provided by this application, but this should not constitute any limitation to this application. Based on the same conception, those skilled in the art can also think of other possible expressions.
  • the embodiments of the present application relate to the Kronecker product operation of the matrix.
  • the Kronecker product operation is available Said.
  • the Kronecker product of matrices A and B can be expressed as
  • the Kronecker product is a block matrix formed by multiplying all elements in a matrix by another matrix.
  • the product of the Kronecker product of the matrix A of dimension k ⁇ l dimension and the matrix B of dimension p ⁇ q dimension is kp ⁇ ql dimension matrix, as follows:
  • projecting vector a onto vector b can be understood as finding the inner product of vector a and vector b.
  • “for indicating” may include both for direct indication and for indirect indication.
  • certain indication information for indicating information I it may include that the indication information directly indicates I or indirectly indicates I, but does not mean that the indication information must carry I.
  • the information indicated by the indication information is called information to be indicated.
  • the information to be indicated can be directly indicated, such as the information to be indicated itself or the Indication index etc.
  • the information to be indicated may also be indirectly indicated by indicating other information, where there is an association relationship between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while other parts of the information to be indicated are known or agreed in advance. For example, it is also possible to realize the indication of specific information by means of the arrangement order of various information pre-agreed (such as stipulated in a protocol), thereby reducing the indication overhead to a certain extent.
  • the precoding matrix is composed of precoding vectors, and each precoding vector in the precoding matrix may have the same part in terms of composition or other attributes.
  • the specific indication method may also be various existing indication methods, such as, but not limited to, the above indication methods and various combinations thereof.
  • the specific details of the various indication methods can refer to the prior art, and will not be repeated here. It can be seen from the foregoing that, for example, when multiple information of the same type needs to be indicated, there may be cases where different information is indicated in different ways.
  • the required indication method can be selected according to specific needs. The embodiments of the present application do not limit the selected indication method. In this way, the indication methods involved in the embodiments of the present application should be understood as covering Fang learns various methods of the information to be indicated.
  • row vectors can be expressed as column vectors
  • a matrix can be represented by the transposed matrix of the matrix
  • a matrix can also be expressed in the form of a vector or an array, which is a vector or an array It can be formed by connecting the row vectors or column vectors of the matrix to each other.
  • the Kronecker product of two vectors can also be expressed in the form of the product of one vector and the transposed vector of another vector.
  • the information to be indicated may be sent together as a whole, or may be divided into multiple sub-information and sent separately, and the sending period and / or sending timing of these sub-information may be the same or different.
  • the specific sending method is not limited in this application.
  • the sending period and / or sending timing of these sub-information may be pre-defined, for example, pre-defined according to the protocol, or may be configured by the transmitting end device by sending configuration information to the receiving end device.
  • the configuration information may include, for example but not limited to, radio resource control signaling, such as RRC signaling, MAC layer signaling, such as MAC-CE signaling, and physical layer signaling, such as downlink control information (DCI).
  • RRC signaling such as RRC signaling
  • MAC layer signaling such as MAC-CE signaling
  • DCI downlink control information
  • pre-acquisition may include signaling indication or pre-defined by the network device, for example, protocol definition.
  • pre-defined can be achieved by pre-storing corresponding codes, tables or other methods that can be used to indicate relevant information in the device (for example, including terminal devices and network devices), and this application does not do for its specific implementation limited.
  • Tenth, "save” involved in the embodiments of the present application may refer to being saved in one or more memories.
  • the one or more memories may be set separately, or may be integrated in an encoder or decoder, a processor, or a communication device.
  • the one or more memories may also be partly set separately and partly integrated in a decoder, processor, or communication device.
  • the type of memory may be any form of storage medium, which is not limited in this application.
  • the "protocol” involved in the embodiments of the present application may refer to a standard protocol in the communication field, for example, it may include the LTE protocol, the NR protocol, and related protocols applied in future communication systems, which are not limited in this application .
  • At least one (a) of a, b, and c may represent: a, or, b, or, c, or, a and b, or, a and c, or, b and c, or, a , B and c.
  • a, b and c may be a single or multiple.
  • the method provided by the embodiments of the present application may be applied to a system that communicates through multi-antenna technology, for example, the communication system 100 shown in FIG. 1.
  • the communication system may include at least one network device and at least one terminal device.
  • Multi-antenna technology can communicate between network equipment and terminal equipment.
  • the embodiments shown below do not specifically limit the specific structure of the execution body of the method provided by the embodiments of the present application, as long as the program that records the code of the method provided by the embodiments of the present application can be executed to
  • the method provided in the embodiment of the application may be used for communication.
  • the execution body of the method provided in the embodiment of the present application may be a terminal device or a network device, or a functional module in the terminal device or network device that can call a program and execute the program.
  • FIG. 2 is a schematic flowchart of a method 200 for indicating and determining a precoding vector provided by an embodiment of the present application from the perspective of device interaction. As shown, the method 200 may include steps 210 to 230. The steps of the method 200 are described in detail below.
  • the terminal device is first described in detail based on one of the one or more transmission layers and one of the one or more polarization directions on the transmission, indicating the precoding vector and the network device to determine the precoding Vector specific process. It should be understood that the application does not limit the number of transmission layers and the number of polarization directions of the transmitting antenna.
  • One transmission layer illustrated in the following examples may be any one of one or more transmission layers, and one polarization direction may be any one of one or more polarization directions.
  • the terminal device In step 210, the terminal device generates first indication information for indicating L 1 (L 1 ⁇ 1 and an integer) beam vectors in the beam vector set and K 1 ( K 1 ⁇ 1 and is an integer) frequency domain vectors and vectors beams to L 1 and K 1 corresponding frequency-domain vector L 1 ⁇ K 1 empty frequency component matrix T 1 (T 1 ⁇ 1 and is an integer ) Space-frequency component matrix.
  • the T 1 space frequency component matrices may be a subset of the M 1 space frequency component matrices.
  • the M 1 space-frequency component matrices here can be obtained by traversing L 1 beam vectors and K 1 frequency domain vectors, respectively.
  • the K 1 frequency domain vectors selected in the frequency domain vector set are denoted as
  • the terminal device may first traverse each beam vector in the range of 0 to L 1 -1, and then traverse each frequency domain vector in the range of 0 to K 1 -1 to obtain M 1 space frequency component matrices.
  • the M 1 space-frequency component matrices may include:
  • the terminal device may also first traverse each frequency domain vector in the range of 0 to K 1 -1, and then traverse each beam vector in the range of 0 to L 1 -1 to obtain M 1 space frequency component matrices.
  • the M 1 space frequency component matrices may include:
  • M 1 space frequency component matrices can also be obtained by the Kronecker product of the frequency domain vector and the beam vector.
  • the first indication information information when used to indicate the T 1 space-frequency component matrix, it can be used to indicate the relative position of the T 1 space-frequency component matrix in the M 1 space-frequency component matrix (for example, relative index or relative number Wait).
  • the terminal device can use the bitmap, or the index of the combination of T 1 space-frequency component matrices in the M 1 space-frequency component matrix, or, the space-frequency component matrices in the T 1 space-frequency component matrix at M 1
  • the index of each space frequency component matrix indicates the T 1 space frequency component matrix.
  • the terminal device uses a bitmap to indicate the T 1 space-frequency component matrixes
  • the relative positions of the T 1 space-frequency component matrixes in the M 1 space-frequency component matrixes can use M 1 bits.
  • terminal equipment can also be used Bits to indicate L 1 beam vector and use Bits to indicate K 1 frequency domain vectors. Therefore, the terminal device can adopt Bits to indicate T 1 space frequency component matrix.
  • beam vector set includes beams L 0 vectors, the frequency domain vector set comprising frequency-domain vector K 0, L 0 ⁇ L 1, K 0 ⁇ K 1, and L 0 ⁇ K 0> L 1 ⁇ K 1.
  • the L 0 beam vectors and K 0 frequency domain vectors may correspond to L 0 ⁇ K 0 space frequency component matrices.
  • the terminal device directly indicates the relative position of the T 1 space-frequency component matrices in the L 0 ⁇ K 0 space-frequency component matrices, it may be necessary to indicate each space-frequency component matrix separately Bits; if indicated by a bitmap, L 0 ⁇ K 0 bits can be used.
  • the feedback overhead can be greatly reduced.
  • the beam vector set includes 16 beam vectors and the frequency domain vector set includes 10 frequency domain vectors
  • 15 space frequency vectors are selected from the 160 space frequency component matrices constructed by the beam vector set and the frequency domain vector set matrix.
  • each space frequency vector matrix needs Bits, that is, 8 bits, the 15 space-frequency vector matrix requires 120 bits for feedback. If a bitmap is used to indicate the 15 space-frequency component matrices, 160 bits are needed for feedback.
  • the terminal device and the network device may need to store a large number of combinations and index correspondence .
  • the 16 beam vectors and the 10 frequency domain vectors may be used to indicate a part of the stronger beam vector and a stronger frequency domain vector, and then from the part of the stronger beam vector and the stronger frequency domain vector. Fifteen space-frequency component matrices are selected in the frequency domain vector.
  • the terminal device may further indicate the selected 15 space frequency component matrices. If the terminal device indicates the 15 space-frequency component matrices through a bitmap, 40 bits can be used for indication. Then, the 15 space-frequency component matrices can be fed back through 62 bits. Compared with the above method, the overhead can be greatly reduced.
  • the weighted sum of the T 1 space frequency component matrices can be used to determine the precoding vectors of one or more frequency domain units.
  • the weighted sum of the T 1 space frequency component matrices can be used to construct a space frequency matrix.
  • the space frequency matrix may include column vectors corresponding to one or more frequency domain units, and each column vector may be used to determine the corresponding precoding vector of the frequency domain unit.
  • the relationship between the space-frequency matrix and the precoding vector has been described in detail above. For the sake of brevity, it will not be repeated here.
  • the space frequency matrix can be approximated as the weighted sum of the T 1 space frequency component matrices.
  • the space-frequency matrix can be written as among them, T 1 represents a frequency component of the first empty empty t 1 frequency component matrix matrix, T 1 represents the frequency component of the matrix is empty Weighting factor.
  • the terminal device may pre-determine the T 1 space-frequency component matrix, or may pre-determine the space-frequency vector pair used to generate the T 1 space-frequency component matrix, and then may further determine the T 1 space-frequency component matrix or T 1 The weight of a pair of space-frequency vectors.
  • the terminal device may indicate the T 1 space-frequency component matrix and the weight of each space-frequency component matrix to the network device, or indicate the T 1 space-frequency vector pair and the weight of each space-frequency vector pair to the terminal device, so as to facilitate
  • the network device recovers the precoding vectors of one or more frequency domain units.
  • the T 1 space-frequency component matrices indicated by the first indication information can be used to determine the precoding vector.
  • the weight of each space-frequency component matrix may be indicated by the first indication information, or may be indicated by other information. This application does not limit this.
  • For the specific indication method of the weights of each space-frequency component matrix reference may be made to the prior art.
  • the T 1 empty frequency component for each space-frequency matrix component matrices may be uniquely determined by the T 1 empty vector frequency vector for a beam, and a frequency domain vector.
  • the terminal apparatus may directly indicate T 1 empty frequency component of the matrix, or may be indirectly indicated T 1 empty frequency component matrix by indicating T 1 empty frequency vectors manner, or may directly indicate that the T 1 th Space frequency vector pair.
  • the T 1 space-frequency vector pairs can be regarded as an equivalent form of the T 1 space-frequency component matrix.
  • the T 1 space frequency component matrices are selected from M 1 space frequency component matrices corresponding to L 1 beam vectors and K 1 frequency domain vectors.
  • the T 1 space frequency component matrices can be determined based on L 1 beam vectors in the beam vector set and K 1 frequency domain vectors in the frequency domain vector set.
  • the terminal device may indicate T 1 space frequency component matrices based on the L 1 beam vectors and K 1 frequency domain vectors.
  • the T 1 space-frequency vector pairs may be selected from M 1 space-frequency vector pairs obtained by combining L 1 beam vectors and K 1 frequency domain vectors.
  • the T 1 space-frequency vector pairs may be selected from M 1 space-frequency vector pairs obtained by combining L 1 beam vectors in the beam vector set and K 1 frequency domain vectors in the frequency domain vector set.
  • the T 1 space-frequency vector pair may be a part of the M 1 space-frequency vector pair.
  • the terminal device may indicate T 1 space frequency vector pairs based on the L 1 beam vectors and K 1 frequency domain vectors, or, based on M 1 space frequency vector pairs.
  • L 1 beam vectors may be partial beam vectors in the beam vector set, and / or, K 1 frequency domain vectors may be partial frequency domain vectors in the frequency domain vector set.
  • K 1 frequency domain vectors may be partial frequency domain vectors in the frequency domain vector set.
  • the K 1 frequency domain vectors are only part of the frequency domain vectors in the frequency domain vector set; when K 1 frequency domain vectors are frequency When all frequency domain vectors in the domain vector set are included, the L 1 beam vectors are only part of the beam vectors in the beam vector set; when L 1 beam vectors are partial beam vectors in the beam vector set, the K 1 frequency
  • the domain vector may be part or all of the frequency domain vectors in the frequency domain vector set; when K 1 frequency domain vectors are part of the frequency domain vectors in the frequency domain vector set, the L 1 beam vectors may be the ones in the beam vector set Part or all of the beam vector.
  • the L 1 beam vectors may be stronger L 1 beam vectors selected from the beam vector set.
  • the K 1 frequency domain vectors are partial frequency domain vectors in the set of frequency domain vectors
  • the K 1 frequency domain vectors may be stronger K 1 frequency domain vectors selected from the set of frequency domain vectors.
  • the stronger L 1 beam vectors can be understood as L beam vectors with larger weighting coefficients
  • the stronger K 1 frequency domain vectors can be understood as K 1 beam vectors with larger weighting coefficients. This is because beam vectors and frequency domain vectors with large weighting coefficients occupy a large weight in the linear combination, and also have a large influence on the approximate accuracy of the precoding vector.
  • the strong L 1 beam vectors and the strong K 1 frequency domain vectors will be described in detail in conjunction with specific implementation methods in the following, and detailed descriptions thereof will be omitted here for the time being.
  • the terminal device can determine L 1 beam vectors from the beam vector set in advance, and K 1 frequency domain vectors from the frequency domain vector set, and select the T 1 space-frequency component matrix to be used as the weighted summation
  • the range is reduced to the range of M 1 space frequency component matrices constructed by L 1 beam vectors and K 1 frequency domain vectors, and T 1 space frequency component matrices are selected and indicated from the M 1 space frequency component matrices, Therefore, it is beneficial to reduce the feedback overhead of the T 1 space-frequency component matrices.
  • M 1 space-frequency component matrices are introduced here only for ease of understanding. This does not mean that the terminal device must generate the M 1 space-frequency component matrix.
  • the terminal device can also use the L 1 beam vector sum K 1 frequency domain vectors are combined to obtain M 1 space frequency vector pairs.
  • the M 1 space frequency component matrix can be constructed. In other words, M 1 space-frequency vector pairs and M 1 space-frequency component matrices can be converted into each other. Therefore, it can be considered that the M 1 space frequency component matrices correspond to L 1 beam vectors and K 1 frequency domain vectors.
  • the introduction of M 1 is only to reflect the correspondence between M 1 space frequency component matrix (or space frequency component vector, or space frequency vector pair) and L 1 beam vector and K 1 frequency domain vector, and should not constitute any limited.
  • L 1 , K 1 and T 1 can be indicated by the network device, or can be pre-defined, such as protocol definition, can also be determined by the terminal device and reported to the network device, or can also be combined with the methods listed above Configuration.
  • the method further includes: the terminal device receives second indication information, and the second indication information is used to indicate at least one of L 1 , K 1 and M 1 The value of two items.
  • the network device sends the second indication information.
  • the second indication information is carried in higher layer signaling, such as an RRC message.
  • the method further includes: the terminal device sends second indication information, where the second indication information is used to indicate L 1 , K 1, and M 1 One or more of the values.
  • the network device receives the second indication information.
  • the second indication information is carried in uplink control information (uplink control information, UCI), such as CSI.
  • uplink control information uplink control information, UCI
  • CSI uplink control information
  • the information indicating the values of L 1 , K 1 and M 1 may be the same information or different information, which is not limited in this application.
  • the value of any one of L 1 and K 1 may also be defined in advance, such as a protocol definition, and the other item is indicated by signaling by the network device.
  • the network device may indicate that the value L 1 through signaling protocol may define the value of K 1, K value for the present application is not limited in size 1.
  • the network device may indicate the value of L 1 through signaling, and the protocol may define the value of K 1 as the value of a certain parameter, or may also define the calculation formula of K 1 .
  • the protocol can define the value of K 1 as the length N f of the frequency domain vector, or it can also define the calculation formula of K 1 , such as, or In this case, the value of K 1 can be understood as implicitly indicated by the fifth indication information. among them, Means round up, Indicates rounding down, [] indicates rounding.
  • the network device may indicate the value of K 1 through signaling, and the protocol may define the value of L 1.
  • the application does not limit the value of L 1 .
  • the network device may indicate the value of K 1 through signaling, and the protocol may define the value of L 1 as the value of a certain parameter, or may also define the calculation formula of L 1 .
  • the value of L 1 can be understood as implicitly indicated by indication information indicating the number of antenna ports in the single polarization direction.
  • the value of any one of L 1 and K 1 may also be defined in advance, such as a protocol definition, and the other item is determined by the terminal device and reported through signaling.
  • the second indication information can indicate L 1 and K The value of another item in K 1 .
  • the protocol defines values of K 1, only the second indication information indicating a value of L; or, L protocol defines the value 1, only the second indication information indicating the value K 1.
  • the value of the other item can also be determined.
  • the second indication information may be indicated indirectly by a value L indicative of the value of M 1; and when L 1 is defined by the protocol, the second indication information can also The value of M 1 is indicated to indirectly indicate the value of K 1 .
  • the method further includes: the terminal device receives third indication information, where the third indication information is used to indicate the value of T 1 .
  • the network device sends the third indication information.
  • the third indication information is carried in higher layer signaling, such as an RRC message.
  • the method further includes: the terminal device sends third indication information, where the third indication information is used to indicate the value of T 1 .
  • the network device receives the third indication information.
  • the third indication information is carried in UCI, such as CSI.
  • the signaling listed above for the third indication information is only an example, and should not constitute any limitation to this application, and this application does not limit the specific signaling that carries the third indication information.
  • second indication information and third indication information may be the same information or different information, which is not limited in this application.
  • the terminal device may determine L 1 beam vectors, K 1 frequency domain vectors and T 1 space frequency component matrices, and then generate first indication information.
  • the K 1 frequency domain vectors are pre-configured.
  • the K 1 frequency domain vectors may be all or part of the frequency domain vectors in the frequency domain vector set.
  • the protocol may predefine the value of K 1 and specify in advance which frequency domain vectors in the set of frequency domain vectors are used as the K 1 frequency domain vectors.
  • the protocol may predefine the value of K 1 , and the network device may indicate the K 1 frequency domain vectors through signaling in advance.
  • the protocol can predefine that the terminal device does not need to report the K 1 frequency domain vectors.
  • the K 1 frequency domain vectors may be specified in advance, such as protocol definition or network device configuration, etc., which is not limited in this application.
  • the protocol may also predefine the terminal device to determine the vector to be reported based on different values of the parameters.
  • the L 1 beam vectors may be the full set of beam vector sets, or the K 1 frequency domain vectors may be the full set of frequency domain vector sets.
  • the terminal device may determine whether to select the full set of beam vector sets or the full set of frequency domain vector sets according to the values of L 1 and K 1 .
  • L 1 beam vectors are the full set of beam vector sets, the terminal device may not report on the L 1 beam vectors; when K 1 frequency domain vectors are the full set of frequency domain vector sets, the terminal device may One frequency domain vector is not reported.
  • the network device may configure the value of K 1 through signaling, such as the foregoing second indication information.
  • the terminal device may use the entire set of frequency domain vector sets as K 1 frequency domain vectors by default.
  • the network device may not indicate the value of K 1 through additional signaling, for example, only the value of L 1 is indicated in the second indication information. That is, optionally, the above-mentioned second indication information is used to indicate the value of L 1 . This can be understood as that the network device implicitly indicates that the value of K 1 is K 0 through the second indication information.
  • the value of K 1 can be defined as a fixed value by the protocol.
  • the terminal device may use the entire set of frequency domain vector sets as K 1 frequency domain vectors by default, without reporting the K 1 frequency domain vectors.
  • the network device may configure the value of L 1 through signaling, such as the foregoing second indication information.
  • the terminal device may use the entire set of beam vector sets as L 1 beam vectors by default.
  • the network device may not indicate the value of L 1 through additional information, for example, only the value of K 1 is indicated in the second indication information. That is, optionally, the second indication information is used to indicate the value of K 1 .
  • the network device implicitly indicates that the value of L 1 is L 0 through the second indication information.
  • the value of L 1 can be defined as a fixed value by the protocol.
  • the terminal device may default the entire set of beam vector sets as L 1 beam vectors without reporting the L 1 beam vectors.
  • the terminal device may determine the vector to be reported according to the parameters configured by the network device.
  • the protocol may also predefine the terminal device to determine the vector to be reported based on different parameters configured by the network device.
  • L 1 ⁇ L 0 and K 1 ⁇ K 0 as examples to describe in detail the specific method in which the terminal device determines and indicates the T 1 space frequency component matrices and the weights of the space frequency component matrices.
  • the terminal device may determine T 1 space-frequency component matrices based on the pre-saved vector set in a corresponding implementation manner. For example, the terminal device may pre-store the beam vector set and the frequency domain vector set, and determine T 1 space-frequency vector pairs based on implementation one; the terminal device may also pre-store the space-frequency component matrix set, and determine based on implementation two T 1 space frequency component matrix.
  • the beam vector set and the frequency domain vector set and the space frequency component matrix set can be converted to each other. Any one of the beam vectors in the set of beam vectors and any one of the frequency domain vectors in the set of frequency domain vectors can determine a space-frequency component matrix in the set of space-frequency component matrices. Any space-frequency component matrix in the set of space-frequency component matrices can be uniquely determined by a beam vector in the beam vector set and a frequency-domain vector in the frequency domain vector set.
  • the index corresponding to each space-frequency vector matrix in the space-frequency vector matrix set can also be converted into the index of the beam vector in the beam vector set and the index of the frequency domain vector in the frequency domain vector set.
  • any space-frequency component vector matrix in the space-frequency vector matrix set can be jointly indicated by a beam vector in the beam vector set and a frequency domain vector in the frequency domain vector set.
  • the terminal device may at least implement T 1 space-frequency component matrices by L 1 beam vectors in the beam vector set and K 1 frequency domain vectors in the frequency domain vector set in at least one of the following implementation manners :
  • T 1 space-frequency component matrixes can be determined by M 1 space-frequency vector pairs.
  • the M 1 space-frequency vector pairs can be obtained by combining L 1 beam vectors and K 1 frequency domain vectors.
  • the terminal device can M 1 from the empty vector frequency for generating the indication of the empty frequency component T 1 T 1 of the matrix vectors empty frequency;
  • T 1 space-frequency component matrices can be selected from M 1 space-frequency component matrices.
  • the M 1 space-frequency component matrices can be determined by L 1 beam vectors and K 1 frequency domain vectors.
  • the terminal device can select from The M 1 space-frequency component matrix indicates the T 1 space-frequency component matrix. It should be noted that this application does not limit the form of the vector set or matrix set pre-stored by the terminal device and the network device.
  • both the terminal device and the network device can save the beam vector set and the frequency domain vector set in advance; or, both the terminal device and the network device can save the space frequency component matrix set in advance; or, the terminal device can save the beam vector set and frequency domain in advance
  • the network device may save the space-frequency component matrix set in advance; or, the terminal device may save the space-frequency component matrix set in advance, and the network device may save the beam vector set and frequency domain vector set in advance.
  • the frequency domain vector set can also be derived, or, after determining In the case of a space frequency component matrix set and a frequency domain vector set, a beam vector set can also be derived. Therefore, the present application does not limit the specific form of the vector set pre-stored by the terminal device and the network device.
  • the terminal device may determine M 1 space frequency vector pairs according to the pre-saved beam vector set, frequency domain vector set and predetermined space frequency matrix, and then determine T 1 space frequency component matrix.
  • the precoding vectors of N f frequency domain units determined by the terminal device are recorded as
  • the precoding vectors of the N f frequency domain units can construct a space frequency matrix H, or,
  • the terminal device can determine the form of the space-frequency matrix to be constructed according to the pre-saved vector set, and then generate the corresponding form of the space-frequency matrix When the terminal device saves the space-frequency component matrix set in advance, a space-frequency matrix can be generated
  • the terminal device may construct a space-frequency matrix H of dimension N s ⁇ N f according to the precoding vectors of N f frequency domain units,
  • the dimension of each beam vector may be N s
  • each beam vector may be taken from a two-dimensional (2dimension, 2D) -DFT matrix.
  • 2D can represent two different directions, such as a horizontal direction and a vertical direction.
  • the N s beam vectors can be written as
  • the L 0 beam vectors can construct a matrix B s ,
  • the set of beam vectors may be expanded to O s ⁇ N s beam vectors by an oversampling factor O s .
  • the dimension of each beam vector in the beam vector set may be N s , and each beam vector may be taken from an oversampled 2D-DFT matrix.
  • the first beam vector set o s (0 ⁇ o s ⁇ O s -1 and o s is an integer) sub-set of beam vectors of N s may each be referred to as e.g. Then a matrix can be constructed based on the N s beam vectors in the o s subset
  • the dimension of each frequency domain vector can be N f , and each frequency domain vector can be taken from a DFT matrix.
  • the N f frequency domain vectors can be written as The N f frequency domain vectors can construct a matrix B f ,
  • the frequency-domain vector set can be extended oversampling factor O f O f ⁇ N f is the frequency-domain vectors.
  • the dimension of each frequency domain vector in the set of frequency domain vectors may be N f , and each frequency domain vector may be taken from an oversampled DFT matrix.
  • the oversampling factor O f is a positive integer.
  • the N f frequency domain vectors in the o f (0 ⁇ o f ⁇ O f -1 and o s are integers) subsets of the frequency domain vector set can be written as O f based on the first sub-set of N s matrix beam vectors can be constructed
  • the terminal device can determine the T 1 space-frequency component matrices and the weighting coefficients of each space-frequency component matrix through steps 1-i to 1-v shown below.
  • Step 1-i The terminal device may determine the weighting coefficient matrix based on the space frequency matrix H, the matrix constructed by the beam vector set, and the matrix constructed by the frequency domain vector set.
  • the beam vector set may include N s beam vectors and the constructed matrix is B s ; the frequency domain vector set may include N f frequency domain vectors and the constructed matrix is B f .
  • This matrix W may be called a weighting coefficient matrix, and its dimension may be N s ⁇ N f .
  • the matrix W of N s rows can set of N s beam vectors corresponding to (or set by the beam vector construction matrix B s) in the beam vector; N f columns of the matrix W may be a vector in the frequency domain N f frequency domain vectors in the set (or matrix B f constructed from the set of frequency domain vectors) correspond.
  • Each coefficient in the matrix may correspond to a space-frequency vector pair.
  • Step 1-ii The terminal device may select a strong L 1 beam vector from the beam vector set, and select a strong K 1 frequency domain vector from the frequency domain vector set.
  • the terminal device may separately take the modulus of the N s rows in the matrix W, and determine the L 1 row with a larger modulus according to the size of each row.
  • the row number of the L 1 row in the matrix W may be the number of the stronger L 1 beam vector in the beam vector set or the column number in B s .
  • the terminal device may also take the modulus of the N f columns in the matrix W, respectively, and determine K 1 columns with a larger modulus according to the size of each column.
  • the column numbers of the K 1 columns in the matrix W may be the sequence numbers of the strong K 1 frequency domain vectors in the frequency domain vector set or the column sequence numbers in B f .
  • the above method of determining L 1 beam vectors and K 1 frequency domain vectors through the weighting coefficient matrix is only one possible implementation manner shown for ease of understanding, and does not mean that the terminal device is determining L 1
  • the weighting coefficient matrix must be generated during the beam vector and K 1 frequency domain vectors. For example, by projecting the precoding vectors of each frequency domain unit to each beam vector in the beam vector set and each frequency domain vector in the frequency domain vector set, an array set consisting of multiple projection values can be obtained Each element may be formed by sequentially connecting elements of each row (or each column) in the above-mentioned weighting coefficient matrix.
  • Step 1-iii The terminal device may determine strong K 1 frequency domain vectors from the frequency domain vector set based on the stronger L 1 beam vectors determined from the beam vector set and combine to obtain M 1 space frequency vectors Correct.
  • each space frequency vector pair of the M 1 space frequency vector pairs may include one beam vector and one frequency domain vector.
  • the beam vector in each space-frequency vector pair can be taken from the above L 1 beam vectors, and the frequency domain vector in each space-frequency vector pair can be taken from the above K 1 frequency domain vectors.
  • a beam vector in the L 1 beam vectors and a frequency domain vector in the K 1 frequency domain vectors can be combined to obtain a unique space-frequency vector pair.
  • the N s beam vectors in the above beam vector set and the N f frequency domain vectors in the frequency domain vector set can be combined to obtain N s ⁇ N f space frequency vector pairs. Therefore, the above M 1 space-frequency vector pairs can be regarded as a subset of the N s ⁇ N f space-frequency vector pairs, and M 1 ⁇ N s ⁇ N f . That is to say, the terminal device can determine T 1 space-frequency vector pairs for linear weighting in the subset of the space-frequency vector pair set obtained by combining the beam vector set and the frequency-domain vector set, or the terminal device needs to report T 1 space-frequency vector pairs. In other words, the selected T 1 space-frequency vector pairs are selected from M 1 space-frequency vector pairs obtained by combining L 1 beam vectors and K 1 frequency domain vectors.
  • the terminal device may determine T 1 space-frequency vector pairs from the M 1 space-frequency vector pairs.
  • the T 1 space frequency vector pairs can be used to determine the T 1 space frequency component matrix.
  • the terminal device may select strong T 1 space-frequency vector pairs from the M 1 space-frequency vector pairs to generate T 1 space-frequency component matrices.
  • the stronger T 1 space-frequency vector pair may be a space-frequency vector pair with a larger modulus of the weighting coefficients in the M 1 space-frequency vector pair. That is, the modulus length of the weighting coefficient of any one of the selected T 1 space-frequency vector pairs is greater than or equal to the weighting coefficient of the weighting coefficient of any one of the remaining M 1 -T 1 space-frequency vector pairs Mold length.
  • the terminal device can determine L 1 ⁇ K 1 (that is, M 1 ) weighting coefficients.
  • the M 1 weighting coefficients may correspond to M 1 space-frequency vectors one-to-one.
  • the terminal device may determine T 1 weighting coefficients with a larger modulus from the M 1 weighting coefficients.
  • the modulus length of any one of the selected T 1 weighting coefficients is greater than or equal to the modulus length of any one of the remaining M 1 -T 1 weighting coefficients.
  • the T 1 weighting coefficients may be weighting coefficients of T 1 space frequency component matrices.
  • the positions of the T 1 weighting coefficients in the L 1 ⁇ K 1 weighting coefficients can be used to determine the T 1 beam vectors and T 1 frequency domain vectors contained in the T 1 space-frequency vector pairs, respectively.
  • the terminal device may extract the above-mentioned L 1 rows with a large modulus and K 1 columns with a large modulus from the matrix W to obtain a matrix with a dimension of L 1 ⁇ K 1 .
  • the matrix with dimensions L 1 ⁇ K 1 is denoted as W ′.
  • This matrix W ' can be regarded as a sub-matrix of the matrix W.
  • the terminal device may modulo each element in the matrix W ′ to select T 1 elements with a larger modulus.
  • the positions of the T 1 elements in the matrix W ′ can be used to determine the positions of the L 1 beam vectors contained in the T 1 space-frequency vector pairs and the positions of the frequency domain vectors in the K 1 frequency domain vectors.
  • the row number of the T 1 element in the matrix W ′ may be the number of the selected T 1 beam vector in the L 1 beam vector, and the column number of the T 1 element in the matrix W ′ It may be the sequence number of the selected T 1 frequency domain vectors in the K 1 frequency domain vectors.
  • the terminal device may extract the above-mentioned L 1 rows with a larger modulus and K 1 columns with a larger modulus from the matrix W according to a predefined rule, for example, in the order of first row and second column. Or, the order of the first column and the second row is sequentially arranged to obtain an array composed of L 1 ⁇ K 1 weighting coefficients.
  • the terminal device may modulo each element in the array to select T 1 elements with a larger modulus.
  • the positions of the T 1 elements in the array can be used to determine the positions of the beam vectors included in the T 1 space-frequency vector pairs in the L 1 beam vectors and the positions of the frequency domain vectors in the K 1 frequency domain vectors.
  • the terminal device does not necessarily generate T 1 space-frequency component matrices based on the T 1 space-frequency vector pairs described above.
  • T 1 space-frequency component matrices based on the T 1 space-frequency vector pairs described above.
  • Step 1-v The terminal device generates first indication information to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency vector pairs.
  • the first indication information may include L 1 beam vectors in the beam vector Position information in the set, position information of K 1 frequency domain vectors in the frequency domain vector set, and information indicating T 1 space frequency vector pairs.
  • the first indication information when used to indicate the L 1 beam vectors, it may indicate the index of the combination of the L 1 beam vectors in the beam vector set.
  • the protocol may predefine multiple combinations of multiple beam vectors, and each combination may correspond to an index.
  • the L 1 beam vectors may be one of the multiple combinations, or close to one of the multiple combinations, and the first indication information may indicate the L 1 by indicating the index of the combination Beam vector. That is, the beams L 1 vector position information beam vector set may be combinations of the beam L 1 in the vector index set of beam vectors.
  • the terminal device can pass Bits to indicate L 1 beam vectors in the beam vector set. Means round up.
  • the first indication information when used to indicate K 1 frequency domain vectors, it may indicate the index of the combination of the K 1 frequency domain vectors in the frequency domain vector set.
  • the protocol may predefine multiple combinations of multiple beam vectors, and each combination may correspond to an index.
  • the K 1 frequency domain vectors may be one of the multiple combinations, or close to one of the multiple combinations, and the first indication information may indicate the K 1 by indicating the index of the combination Frequency domain vectors. That is, the frequency-domain vector K 1 position information in the frequency domain vector set may be a combination of a frequency-domain vector index in the frequency domain in the vector set K.
  • the terminal device can pass Bits to indicate K 1 frequency domain vectors in the frequency domain vector set.
  • the method of indicating L 1 beam vectors by the index indicating the combination of L 1 beam vectors and indicating the K 1 frequency domain vectors by the index indicating the combination of K 1 frequency domain vectors is only one possible
  • the way of implementation should not constitute any limitation on this application.
  • the first indication information indicates the L 1 beam vectors, it may also be used to indicate the index of each beam vector in the L 1 beam vectors in the beam vector set, or, the first indication information indicates the
  • K 1 frequency domain vectors it can also be used to indicate the index of each frequency domain vector in the K 1 frequency domain vectors in the set of frequency domain vectors. This application does not limit the specific manner of indicating L 1 beam vectors and K 1 frequency domain vectors.
  • the first indication information may indicate T 1 space-frequency vector pairs in any of the following ways:
  • Method 1 Indicating T 1 space-frequency vector pairs of M 1 space-frequency vector pairs through a bitmap; or
  • Method 2 Indicate the index of the combination of T 1 space-frequency vector pairs in M 1 space-frequency vector pairs;
  • Manner 3 indicating the position of the beam vector contained in each of the T 1 space frequency vector pairs in the L 1 beam vectors and the position of the frequency domain vector in the K 1 frequency domain vectors;
  • Mode 4 Indicate the position of each space-frequency vector pair in the T 1 space-frequency vector pairs in the M 1 space-frequency vector pair.
  • the terminal device may indicate T 1 space-frequency vector pairs of the M 1 space-frequency vector pairs through the M 1 bit bitmap.
  • Each bit in the bitmap may correspond to one of the M 1 space-frequency vector pairs.
  • Each bit can be used to indicate whether the corresponding space-frequency vector pair is selected as T 1 space-frequency vector pair, or each bit can be used to indicate whether the corresponding space-frequency vector pair belongs to the T 1 space Frequency vector pair. For example, when a bit is set to "0", it means that the corresponding space-frequency vector pair does not belong to the T 1 space-frequency vector pair; when a bit is set to "1", it means that the corresponding space-frequency vector pair belongs to The T 1 space frequency vector pairs.
  • the correspondence between M 1 bits and M 1 space-frequency vector pairs in the bitmap corresponds to the combination mode of the beam vector and the frequency domain vector in the M 1 space-frequency vector pairs.
  • the M 1 space-frequency vector pairs corresponding to the M 1 bits may be arranged in the order of traversing K 1 frequency domain vectors and then L 1 beam vectors, or L 1 beam vectors, Then traverse K 1 frequency domain vectors in sequence.
  • the K 1 frequency domain vectors selected in the frequency domain vector set are denoted as
  • the order of the M 1 space frequency vector pairs may be There are M 1 space-frequency vector pairs. For brevity, I will not list them here.
  • the M 1 bits in the bitmap correspond one-to-one to the M 1 space-frequency vectors described above.
  • Bits 0 to K 1 -1 of the M 1 bits of this bitmap are paired with the space-frequency vector Correspondence; first K 1 through 2K 1 -1 bit and space frequency vectors One-to-one correspondence; and so on, the (L 1 -1) ⁇ K 1 to L 1 ⁇ K 1 -1 bits are paired with the space-frequency vector One to one correspondence.
  • the order of the M 1 space frequency vector pairs may be There are M 1 space-frequency vector pairs. For brevity, I will not list them here.
  • the M 1 bits in the bitmap correspond one-to-one to the M 1 space-frequency vectors described above.
  • the 0th to L 1 -1 bits of the M 1 bits of this bitmap are paired with the space-frequency vector Correspondence; of L 1 through 2L 1 -1 bit and space frequency vectors One-to-one correspondence; by analogy, the L 1 ⁇ (K 1 -1) to L 1 ⁇ K 1 -1 bits are paired with the space-frequency vector One to one correspondence.
  • the one-to-one correspondence between the M 1 bits and M 1 space-frequency vector pairs listed above is only an example, and should not constitute any limitation to this application.
  • This application does not limit the correspondence between M 1 bits and M 1 space-frequency vector pairs.
  • the present application does not limit the arrangement of M 1 space-frequency vector pairs. The above is only for the purpose of explaining the one-to-one correspondence between M 1 bits and M 1 space-frequency vector pairs, and shows the relationship with M 1 There are two possible arrangements of M 1 space-frequency vector pairs corresponding to each bit.
  • the terminal device may indicate the T 1 space-frequency vector pairs through the index of the combination of the T 1 space-frequency vector pairs in the M 1 space-frequency vector pairs.
  • the terminal device can pre-determine multiple combinations of multiple pairs of space-frequency vectors according to the M 1 space-frequency vectors obtained by combining the L 1 beam vectors and K 1 frequency-domain vectors, and each combination can correspond to an index .
  • the T 1 space-frequency vector pairs may be one of the multiple combinations, or close to one of the multiple combinations.
  • the first indication information may indicate the T 1 space-frequency vector pairs by indicating the combined index. Therefore, the terminal device can pass Bits to indicate T 1 space-frequency vector pairs of M 1 space-frequency vector pairs.
  • the terminal device may separately indicate the positions of T 1 beam vectors in L 1 beam vectors and T 1 frequency domain vectors in K 1 frequency domains used to obtain the T 1 space frequency vector pairs The position in the vector. For each beam vector, the terminal device can pass Bits to indicate its position in L 1 beam vectors; for each frequency domain vector, the terminal device can pass Bits to indicate its position in K 1 frequency domain vectors.
  • the terminal device may separately indicate the position of each of the T 1 space-frequency vector pairs in the M 1 space-frequency vector pairs.
  • the position of each space-frequency vector pair in M 1 space-frequency vector pair here can be understood as the relative position of each space-frequency vector pair in M 1 space-frequency vector pair, or local )position.
  • the terminal device may indicate its index in M 1 space-frequency vector pairs.
  • the terminal device can pass Bits to indicate its index in M 1 space-frequency vector pairs.
  • the space-frequency vector pair can be expressed in the form of a space-frequency component matrix (including matrix form or vector form), or in the form of a vector pair obtained by combining a beam vector and a frequency domain vector. This application does not limit this.
  • the terminal device uses the relative positions of the T 1 space-frequency vector pairs in the M 1 space-frequency vector pairs (for example, relative index or Relative number, etc.) to indicate the T 1 space-frequency vector pair, or the relative position (for example, relative index or relative number) of the T 1 space-frequency vector pair in L 1 beam vector and K 1 frequency domain vector Etc.) to indicate T 1 space-frequency vector pairs.
  • the selection range is reduced, the overhead for indicating T 1 space frequency vector pairs is also reduced.
  • each null beam vectors corresponding frequency vector is the number of K 1, L 1 in the corresponding vector beams L 1 ⁇ K 1 to the empty vector frequency, the terminal device based on each beam vector, reported It is selected for weighted summation to determine the pair of space-frequency components of the precoding vector.
  • the terminal device may indicate the T 1 / L based on each beam vector a combination of selected space-frequency vectors in the vector of index and K 1 empty the same beam vectors corresponding to the frequency; if K 1 empty each beam vectors corresponding to the frequency vector of the selected space-frequency vectors The number of is different from each other, the terminal device can report the number of selected space-frequency vector pairs based on each beam vector, and the combination of the selected space-frequency vector pairs in K 1 space corresponding to the same beam vector The index in the frequency vector pair.
  • the method in which the terminal device indicates T 1 space-frequency vector pairs is not limited to the methods listed in this application. For brevity, no one-by-one example will be described here. This application does not limit the specific method in which the terminal device indicates T 1 space-frequency vector pairs.
  • the first indication information further includes quantization information of weighting coefficients of T 1 space frequency vector pairs.
  • the terminal device may be a frequency based on the vector weighting coefficients T 1 empty Step 1-iv determined quantization information to generate the T 1 weighting coefficient vectors empty frequency pair.
  • the terminal device may indicate the T 1 weighting coefficients in a normalized manner according to the weighting coefficients of the T 1 space frequency vector pairs determined in step 1-iv.
  • the terminal device may determine the weighting coefficient with the largest modulus from the T 1 weighting coefficients (for example, referred to as the maximum weighting coefficient), and indicate the position of the maximum weighting coefficient in the matrix W ′. Then, the terminal device may further indicate the relative value of each of the remaining T 1 -1 weighting coefficients with respect to the maximum weighting coefficient. The terminal device may indicate the remaining T 1 -1 weighting coefficients by, for example, the index of the quantization value of each relative value. For example, a one-to-one correspondence between multiple quantization values and multiple indexes may be pre-defined in the codebook, and the terminal device may feedback the relative value of each weighting coefficient relative to the maximum weighting coefficient to the network device based on the one-to-one correspondence.
  • the weighting coefficient fed back by the terminal device may be the same as or similar to the weighting coefficient determined in step 1-iv, so it is called the quantization value of the weighting coefficient.
  • the information for indicating the quantization value of the weighting coefficient may be referred to as the quantization information of the weighting coefficient.
  • the quantization information may be an index of quantization values, for example.
  • the normalization mentioned here can determine the maximum weighting coefficient in units of each polarization direction, each transmission layer, or all transmission layers, so that in each polarization direction, each transmission layer Or all transport layers are normalized in different ranges.
  • the terminal device may also independently determine and feed back the broadband amplitude coefficient to the L 1 beam vectors.
  • the first indication information may further include quantization information of broadband amplitude coefficients of L 1 beam vectors.
  • the terminal device may first select L 1 beam vectors, and then select K 1 frequency domain vectors for each beam vector, and may further determine the composition of each beam vector and its corresponding frequency domain vectors.
  • the weighting coefficients corresponding to the space-frequency vector pair that is, a total of L 1 ⁇ K 1 weighting coefficients.
  • the first indication information is specifically used to indicate each beam vector in the L 1 beam vectors and the frequency domain vector corresponding to each beam vector.
  • This possible design can be applied to scenarios where at least two beam vectors of the selected L 1 beam vectors correspond to different frequency domain vectors, and in particular can be applied to at least two beam vectors of the selected L beam vectors
  • the corresponding frequency domain vectors are different, and the selected beam vectors are few, or in other words, a scene with a small L 1 value (that is, the spatial sparsity is better).
  • the first indication information is also used to indicate the number of frequency domain vectors corresponding to each beam vector.
  • the number of frequency domain vectors corresponding to at least two beam vectors is different.
  • the terminal device determines T 1 space frequency vector pairs from M 1 space frequency vector pairs, it may be based on Each beam vector indicates the selected space-frequency vector pair. For example, based on each beam vector, the terminal device may use any one of the ways 1 to 4 listed above to indicate.
  • the terminal device may further indicate the number of frequency domain vectors corresponding to each beam vector when using the above manner 2 indication.
  • the terminal device may further indicate the space-frequency vector corresponding to each beam vector when using the method 2 above The number of space-frequency vector pairs selected in the pair.
  • the terminal device may first select K frequency domain vectors, and then select L beam vectors for each frequency domain vector, and may further determine the composition of each frequency domain vector and its corresponding beam vector Corresponding to the weighting coefficient of the space-frequency vector pair, that is, a total of L ⁇ K weighting coefficients.
  • the first indication information is specifically used to indicate each of the K frequency domain vectors and the beam vector corresponding to each frequency domain vector.
  • This possible design can be applied to the scenario where at least two frequency domain vectors of the selected K frequency domain vectors correspond to different beam vectors, and in particular can be applied to at least two frequency domains of the selected K frequency domain vectors.
  • the beam vectors corresponding to the domain vectors are different, and the selected frequency domain vectors are fewer, or in other words, the K value is small (that is, the frequency domain has better sparsity) in the scene.
  • the first indication information is also used to indicate the number of beam vectors corresponding to each frequency domain vector.
  • the number of beam vectors corresponding to at least two frequency domain vectors is different.
  • a beam vector is selected for each frequency domain vector and at least two frequency domain vectors have different beam vectors selected, after the end device determines T 1 space frequency vector pairs from M 1 space frequency vector pairs, it may be based on Each frequency domain vector indicates the selected space-frequency vector pair. For example, based on each frequency domain vector, the terminal device may use any one of the methods 1 to 4 listed above to indicate.
  • the terminal device may further indicate the number of beam vectors corresponding to each frequency domain vector when using the above manner 2 indication.
  • the terminal device may further indicate the space-frequency vector corresponding to each beam vector when using the method 2 above The number of space-frequency vector pairs selected in the pair.
  • L 1 beam vectors may be part or all of the beam vectors in the beam vector set, that is, L 1 ⁇ N s ; K 1 frequency domain vectors may be part or all of the frequency domain vector set All frequency domain vectors, that is, K 1 ⁇ N f .
  • the first indication information may only indicate K 1 frequency domain vectors and T 1 space frequency vector pairs without indicating L 1 beam vectors through additional information. In other words, the first indication information is used to indicate K 1 frequency domain vectors and T 1 space frequency vector pairs.
  • the network device may default that the L 1 beam vectors are the full set of beam vector sets.
  • the first indication information may only indicate L 1 beam vectors and T 1 space frequency vector pairs, without indicating K 1 frequency domain vectors through additional information.
  • the network device may default that the K 1 frequency domain vectors are the complete set of frequency domain vector sets.
  • the quantization information of the weighting coefficients of the T 1 space-frequency vector pairs may be carried in the first indication information, or may be carried by additional information, which is not limited in this application.
  • the vectors included in the beam vector set and the frequency domain vector set may have the following three possible situations:
  • Case 1 The beam vector set is expanded to O s ⁇ N s beam vectors by the oversampling factor O s , and the frequency domain vector set is expanded to O f ⁇ N f frequency domain vectors by the oversampling factor O f ;
  • Case 2 The beam vector set is expanded to O s ⁇ N s beam vectors by an oversampling factor O s , and the frequency domain vector set includes N f frequency domain vectors;
  • the beam vector set includes N s beam vectors, and the frequency domain vector set is expanded to O f ⁇ N f frequency domain vectors by the oversampling factor O f .
  • the processing manner of the terminal device may be the same.
  • the specific process of determining the T 1 space-frequency component matrices and the weighting coefficients of each space-frequency component matrix by the terminal device will be described in detail below by taking case 1 as an example.
  • the terminal device may specifically determine the T 1 space-frequency component matrices and the weighting coefficients of each space-frequency component matrix through steps 2-i to 2-vi shown below.
  • Step 2-i The terminal device may determine the weighting coefficient matrix based on the space frequency matrix H, the matrix constructed by the beam vector set, and the matrix constructed by the frequency domain vector set.
  • the beam vector set may include O s subsets, a matrix can be constructed based on the o s subset
  • the set of frequency domain vectors can include O f subsets, and a matrix can be constructed based on the o f subsets Terminal equipment can pass To determine the matrix
  • the matrix It can be regarded as a weighting coefficient matrix corresponding to the o s subset and the o f subset, and its dimension can be N s ⁇ N f .
  • the matrix Of N s rows may be set in the first beam vector subset o s (or by constructing a first matrix subset o s ) Corresponding to N s beam vectors; the matrix The N f columns in can be the same as the f th subset in the frequency domain vector set (or the matrix constructed from the f th subset ) Corresponds to N f frequency domain vectors.
  • the matrix Each coefficient in may correspond to a space-frequency vector pair.
  • the terminal device may determine the O s ⁇ O f group of space-frequency vector pairs based on the O s subsets in the beam vector set and the O f subsets in the frequency domain vector set, and each group of space-frequency vector pairs includes T 1 space frequency vector pair.
  • the terminal device can traverse the value of o s in 0 to O s -1 and the value of o f in 0 to O f -1, and repeat the following steps to determine O s ⁇ O f Group of space-frequency vector pairs: according to the matrix Determine the strong L 1 row and the strong K 1 column, and then determine the strong L 1 beam vector in the o s subset and the strong K 1 frequency domain vector in the o f subset.
  • the L 1 beam vectors and K 1 frequency domain vectors can be combined to obtain M 1 space frequency vector pairs.
  • the terminal device may further L 1 ⁇ K 1 determined by the stronger L 1 row and the stronger K 1 column in the middle determines the stronger T 1 space-frequency vector pair.
  • the terminal device determines the strong L 1 beam vector and the strong K 1 frequency domain vector according to the weighting coefficient matrix W, and then determines the strong The specific process of T 1 space-frequency vector pairs will not be repeated here for brevity.
  • Step 2-iii The terminal device may select the strongest set of space-frequency vector pairs based on the weighting coefficients of O s ⁇ O f sets of space-frequency vector pairs to determine the T 1 space-frequency vector pairs and the Weighting factor.
  • the terminal device may determine the strongest set of space-frequency vector pairs according to the O s ⁇ O f set of space-frequency vector pairs determined in step 2-ii, and T 1 of the strongest set of space-frequency vector pairs Vector pairs can be used to generate T 1 space-frequency component matrices.
  • the terminal device can separately calculate the sum of the modulus lengths of the weighting coefficients of each pair of space frequency vector pairs in the O s ⁇ O f group of space frequency vector pairs, and select the group of space frequency vector pairs with the largest sum of the modulus lengths to generate T 1 space frequency component matrix.
  • the weighting coefficient of the pair of space-frequency vectors is the weighting coefficient of the T 1 space-frequency component matrix.
  • the terminal device Since the frequency of empty vector T 1 T 1 of beams from a beam vector contained in the vector set to the same subset, the terminal device determines that the frequency of T 1 empty vectors at the same time, it can be determined that T 1
  • the subset from which the beam vectors come from can determine the strong L 1 beam vectors in the subset.
  • the terminal device since the T 1 frequency domain vectors included in the T 1 space frequency vector pairs also come from the same subset of the frequency domain vector set, the terminal device, while determining the T 1 space frequency vector pairs, also It is possible to determine the subset from which the T 1 frequency domain vectors come, and to determine the stronger K 1 frequency domain vectors in the subset.
  • the weighting coefficient matrix must be generated during the process of the vector and K 1 frequency domain vectors. For example, by projecting the precoding vectors of each frequency domain unit to each beam vector in any subset of the beam vector set and each frequency domain vector in any subset of the frequency domain vector set, an array set consisting of multiple projection values Each element in the array set may be formed by sequentially connecting elements of each row (or each column) in the above-mentioned weighting coefficient matrix.
  • Step 2-iv The terminal device generates first indication information to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency vector pairs.
  • the first indication information may include L 1 beam vectors in the beam vector Position information in the set, position information of K 1 frequency domain vectors in the frequency domain vector set, and information indicating T 1 space frequency vector pairs.
  • the first indication information when used to indicate the L 1 beam vectors, it may specifically be used to indicate the subset to which the L 1 beam vectors belong and the index of the L 1 beam vectors in the subset; or , Which can be specifically used to indicate the subset to which the L 1 beam vectors belong and the index of the combination of the L 1 beam vectors in the subset.
  • the first indication information when used to indicate the K 1 frequency domain vectors, it may specifically be used to indicate the subset to which the K 1 frequency domain vectors belong and the K 1 frequency domain vectors in the subset Index; or, it can be specifically used to indicate the subset to which the K 1 frequency domain vectors belong and the index of the combination of the K 1 frequency domain vectors in the subset.
  • the specific indication manner may be any one of the manners 1 to 4 described above. Since Mode 1 to Mode 4 have been described in detail above, for the sake of brevity, they will not be repeated here.
  • the first indication information further includes quantization information of weighting coefficients of T 1 space frequency vector pairs.
  • the quantization information of the weighting coefficients of the T 1 space frequency vector pairs may be carried in the first indication information, or may be carried by additional information, which is not limited in this application.
  • the frequency domain vector set includes N f frequency domain vectors, and the terminal device can construct a matrix And matrix B f . After that, the terminal device can traverse the value of o s from 0 to O s -1, by To determine the T 1 space-frequency component matrix and the weighting coefficients of each space-frequency component matrix.
  • the terminal device can construct the matrix B s and the matrix After that, the terminal device can traverse the value of o f from 0 to O f -1, by To determine the T 1 space-frequency component matrix and the weighting coefficients of each space-frequency component matrix.
  • the specific method for the terminal device to determine the T 1 space-frequency component matrices and the weighting coefficients of the space-frequency component matrices in Case 2 and Case 3 is similar to the specific method described in Case 1, and is not repeated here for brevity.
  • the terminal device may generate first indication information to indicate L 1 beam vectors in the beam vector set, K 1 frequency domain vectors in the frequency domain vector set, and T 1 space frequency component matrix.
  • L 1 beam vectors in the beam vector set described above K 1 frequency domain vectors in the frequency domain vector set, and T 1 space frequency component matrix are only examples, and should not be applied to this application. Constitute any limitation.
  • the L 1 beam vectors may be part or all of the beam vectors in the beam vector set, that is, L 1 ⁇ L 0 ; the K 1 frequency domain vectors may be part or all of the frequency in the frequency domain vector set The domain vector, that is, K 1 ⁇ K 0 .
  • the L 1 beam vectors may be part or all of the beam vectors in the beam vector set, that is, L 1 ⁇ L 0 ; the K 1 frequency domain vectors may be part or all of the frequency in the frequency domain vector set The domain vector, that is, K 1 ⁇ K 0 .
  • the L 1 beam vectors may be a subset (eg, an orthogonal group) or a partial beam vector in the subset of the beam vector set;
  • the K 1 frequency domain vectors may also be a subset (for example, an orthogonal group) or a part of frequency domain vectors in the subset of frequency domain vectors.
  • the network device may default that the L 1 beam vectors are the selected subset of the beam vector set.
  • the first indication information may be used to indicate only the frequency domain vector set when indicating the K 1 frequency domain vectors The selected subset without indicating the K 1 frequency domain vectors by additional information.
  • the network device may default that the K 1 frequency domain vectors are the selected subset of the frequency domain vector set.
  • the first indication information may only indicate K 1 frequency domain vectors and T 1 space frequency vector pairs, and The L 1 beam vectors are not indicated by additional information.
  • the network device may default that the L 1 beam vectors are all beam vectors of the beam vector set.
  • the first indication information may indicate only L 1 beam vector and T 1 space frequency vector pair, The K 1 frequency domain vectors are not indicated by additional information.
  • the network device may default that the K 1 frequency domain vectors are all frequency domain vectors in the frequency domain vector set.
  • the terminal device may also make no distinction between the cases listed above, and directly indicate L 1 beam vector, K 1 frequency domain vector, and T 1 space frequency vector pair.
  • the terminal device may determine M 1 space frequency component matrices according to the pre-stored space frequency component matrix set and the predetermined space frequency matrix, and then determine T 1 space frequency component matrices.
  • each space-frequency component matrix in the space-frequency component matrix set pre-stored by the terminal device may be a matrix with a dimension of N s ⁇ N f or a vector with a length of N s ⁇ N f .
  • the specific process in which the terminal device determines T 1 space-frequency component matrices will be described in detail below in conjunction with these two cases.
  • Each space-frequency component matrix in the space-frequency component matrix set is a vector of length N s ⁇ N f .
  • the terminal device may construct a space frequency matrix H of length N s ⁇ N f according to the precoding vectors of the N f frequency domain units,
  • the space-frequency matrix with a length of N s ⁇ N f is called a space-frequency vector.
  • the space frequency component matrix set will be described in detail below.
  • the space frequency component matrix set may include N s ⁇ N f space frequency component matrices.
  • Each space frequency component matrix may be a vector of length N s ⁇ N f .
  • the space-frequency component matrix of length N s ⁇ N f is called a space-frequency component vector.
  • the above-mentioned space-frequency component matrix set may be called a space-frequency component vector set.
  • the weighted sum of T 1 space frequency component vectors selected in the space frequency component vector set can be constructed to obtain a space frequency vector.
  • the space frequency vector obtained by constructing the weighted sum of the T 1 space frequency component vectors may be the same as or similar to the space frequency vector determined by the terminal device above.
  • each space frequency component vector in the space frequency component vector set may be uniquely determined by one beam vector in the beam vector set and one frequency domain vector in the frequency domain vector set.
  • any two space-frequency component vectors in the set of space-frequency component vectors are different, and at least one of the beam vector and the frequency domain vector corresponding to any two space-frequency component vectors is different.
  • each space-frequency component vector in the set of space-frequency component vectors may be a Kronecker product of a beam vector in the set of beam vectors and a frequency-domain vector in the set of frequency domain vectors, or may also be a frequency The Kronecker product of a frequency domain vector in the domain vector set and a beam vector in the beam vector set.
  • each space frequency component vector in the set of space frequency component vectors can be uniquely determined by the Kronecker product of the frequency domain vector and the beam vector.
  • the beam vector set may include N s beam vectors and the frequency domain vector set may include N f frequency domain vectors, then the beam vector set and the frequency domain vector set may determine N s ⁇ N f space frequency component vectors. That is to say, the space frequency component vector set may include N s ⁇ N f space frequency component vectors.
  • Each space frequency component vector may correspond to a beam vector and a frequency domain vector, or each space frequency component vector may correspond to a space frequency vector pair obtained by combining a beam vector and a frequency domain vector.
  • Each space-frequency component vector in the set of space-frequency component vectors may correspond to an index.
  • the N s ⁇ N f space frequency component vectors in the space frequency component vector set can be indicated by the index in the space frequency component vector set, or can be used to generate the space frequency component vector beam vector and frequency domain vector in the beam respectively Index indication in vector set and frequency domain vector set.
  • M 1 empty space frequency components of the vector components of the vector in the set of index can be regarded as a pilot dimensional index M 1 empty vector frequency component, and frequency domain beamforming vectors M 1 empty vector frequency component contained in each vector
  • the indexes in the beam vector set and the frequency domain vector set can be regarded as two-dimensional indexes of M 1 space-frequency component vectors.
  • the one-dimensional index and the two-dimensional index can be converted to each other according to the predefined conversion rules.
  • n s is the index of the beam vector in the set of beam vectors, 0 ⁇ n s ⁇ N s -1, and n s is an integer
  • n f is the index of the frequency domain vector in the frequency domain vector set, 0 ⁇ n f ⁇ N f -1, and n f is an integer.
  • N f frequency-domain vectors of frequency-domain vectors 0 through N f -1, respectively, and frequency-domain vectors of N s Kronecker product of the first beam vectors beam vectors 0 N f determined empty frequency components of the vector may correspond to one-dimensional index of 0 to N f -1; N f frequency-domain vectors of frequency-domain vectors 0 through N f -1, respectively, the first frequency-domain vectors of N s with the beam vectors
  • n s mod (m, N f ). Indicates rounding down, and mod () indicates modulo.
  • s beams vectors and frequency-domain vector set beam vector set of N s beams vectors For N s beams vectors and frequency-domain vector set beam vector set of N s beams vectors, s beams vectors and frequency-domain vector set beam vector set of n in the first n f frequency-domain vectors constructed
  • the space-frequency component vector of can be indicated by the m-th space-frequency component vector.
  • m n f + n s * N f .
  • vector 0 N f frequency-domain frequency-domain vectors respectively a vector of N s beams of beam vectors 0 to N s -1 beams of vectors Kronecker product of the determined N s
  • One space-frequency component vector can correspond to one-dimensional index 0 to N s -1; the first frequency domain vector among N f frequency domain vectors and the 0th beam vector to N s -1 among N s beam vectors, respectively
  • the N s space-frequency component vectors determined by the Kronecker product of the beam vectors can correspond to the one-dimensional index N s to 2N s -1; and so on, the n f frequency domain vector among the N f frequency domain vectors N s beams respectively vector of beam vectors 0 through N s -1 beams
  • Kronecker product vectors of N s of the determined space-frequency component of the vector may correspond to one-dimensional index n f ⁇ N s to (n f +1) ⁇ N s -1.
  • the m-th (0 ⁇ m ⁇ N s ⁇ N f -1) space-frequency component vector can pass the n-th in the beam vector set
  • s beams vectors and frequency-domain vector set beam vector set of N s beams vectors For N s beams vectors and frequency-domain vector set beam vector set of N s beams vectors, s beams vectors and frequency-domain vector set beam vector set of n in the first n f frequency-domain vectors constructed
  • the space-frequency component vector of can be indicated by the m-th space-frequency component vector.
  • m n s + n f * N s .
  • each space-frequency component vector in the two sets of space-frequency component vectors listed above is only an example, and should not constitute any limitation to this application.
  • the terminal device and the network device may number each space-frequency component vector in the set of space-frequency component vectors according to a predetermined rule. The corresponding relationship between each space-frequency component vector defined by the network device and the terminal device and the index is consistent.
  • the vectors included in the space-frequency component vector set may have the following three possible situations:
  • Case 1 The space-frequency component vector set is expanded to O s ⁇ N s ⁇ O f ⁇ N f space-frequency component vectors through the oversampling factors O s and O f , then the space-frequency component vector set may include O s ⁇ O f Subsets, each subset may include N s ⁇ N f space-frequency component vectors;
  • Case 2 The space-frequency component vector set is expanded to O s ⁇ N s ⁇ N f space-frequency component vectors through the oversampling factor O s , then the space-frequency component vector set may include O s subsets, and each subset may include N s ⁇ N f space frequency component vectors; and
  • Case 3 The space-frequency component vector set is expanded to O f ⁇ N s ⁇ N f space-frequency component vectors through the oversampling factor O f , then the space-frequency component vector set may include O f subsets, and each subset may include N s ⁇ N f space frequency component vectors.
  • the oversampling factor O s may be the oversampling factor of the beam vector set
  • each subset of the beam vector set can be determined separately from each subset of the frequency domain vector set to obtain multiple sets of space-frequency component vectors.
  • Each set of space-frequency component vectors includes N s ⁇ N f space frequency component vectors.
  • the numbering rules of the index of N s ⁇ N f space-frequency component vectors in each group of space-frequency component vectors may be the same as the above-mentioned numbering rules when no oversampling rate is considered.
  • the terminal device can determine T 1 space-frequency component vectors and the weighting coefficients of the space-frequency component vectors through steps 3-i to 3-iv shown below.
  • Step 3-i The terminal device may determine a weighting coefficient matrix based on the above-mentioned space-frequency vector and space-frequency component vector set.
  • the space frequency component vector set may include N s ⁇ N f space frequency component vectors.
  • the terminal device may project the predetermined space-frequency vectors to the N s ⁇ N f space-frequency component vectors, respectively. That is, the conjugate transpose of each of the N s ⁇ N f space-frequency component vectors is multiplied by the space-frequency vector, respectively, to obtain N s ⁇ N f projected values.
  • the arrangement order of the N s ⁇ N f projection values corresponds to the arrangement order of the N s ⁇ N f space-frequency component vectors in the space-frequency component vector set.
  • the terminal device may set a N s ⁇ N f space-frequency component of the vector in the order based on the space frequency component of the vector, the N s ⁇ N f projection values are arranged in order of a predetermined dimension is the N s ⁇ N f matrix.
  • the terminal device can start from the first projection value of the N s ⁇ N f projection values, and every N f consecutive projection values are regarded as one line, and N s lines can be obtained, each line including N f Projection values. Arrange the N s rows in order from top to bottom to obtain a matrix W with dimensions N s ⁇ N f .
  • the terminal device can start from the first projection value of the N s ⁇ N f projection values, and every N s consecutive projection values are taken as one column, and N f columns can be obtained, and each column includes N s projection values . Arrange the N f columns in order from left to right to obtain a matrix W with dimensions N s ⁇ N f .
  • the terminal device may directly arrange the N s ⁇ N f space frequency component vectors in a matrix form according to the two-dimensional index . For example, space-frequency component vectors with the same index n s are arranged in the same row, and space-frequency component vectors with the same index n f are arranged in the same column.
  • the above-mentioned matrix W of dimension N s ⁇ N f can be called a weighting coefficient matrix.
  • the matrix W N s ⁇ N f weighting coefficients may be space frequency components of the vector in the set of N s ⁇ N f empty vectors corresponding frequency components, it may represent a N s ⁇ N f empty vector for each frequency component The weighting coefficient of the space-frequency component vector.
  • Step 3-ii The terminal device may determine strong M 1 space frequency component vectors according to the weighting coefficient matrix.
  • the terminal device may separately take the modulus of the N s rows in the matrix W, and determine the L 1 row with a larger modulus according to the size of each row.
  • the L 1 line with the larger modulus is the stronger L 1 line.
  • the terminal device may also take the modulus of the N f columns in the matrix W, respectively, and determine K 1 columns with a larger modulus according to the size of each column.
  • the K 1 column with the larger modulus is the stronger K 1 column.
  • the terminal device can determine the stronger M 1 space frequency from the set of space frequency component vectors according to the position of the stronger L 1 row and the stronger K 1 column in the matrix W, and the predefined conversion rule Component vector.
  • each space frequency component vector in the set of space frequency component vectors can be determined by each beam vector in the beam vector set and each frequency domain vector in the frequency domain vector set.
  • the M 1 space-frequency component vectors may be determined by the stronger L 1 beam vectors in the beam vector set and the stronger K 1 frequency domain vectors in the frequency domain vector set.
  • the row numbers of the stronger L 1 rows in the matrix W determined above can be the numbers of the stronger L 1 beam vectors in the beam vector set, and the stronger K 1 columns in the matrix W
  • the sequence number may be the sequence number of the strong K 1 frequency domain vector in the frequency domain vector set.
  • the terminal device may determine strong T 1 space frequency component vectors from the strong M 1 space frequency component vectors.
  • the terminal device may determine the strong T 1 space frequency component vectors according to the modulus length of the weighting coefficients of the M 1 space frequency component vectors. For example, the modulus length of the weighting coefficient of any one of the selected T 1 space frequency component vectors is greater than or equal to the modulus length of any one of the remaining M 1 -T 1 space frequency component vectors . At the same time, the weighting coefficients of the T 1 space-frequency component vectors can also be determined.
  • Step 3-iv The terminal device generates first indication information to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component vectors.
  • the first indication information may include M 1 space-frequency component vectors in the space-frequency component vector set Or location information in a subset of the set of space-frequency component vectors and information indicating T 1 space-frequency component vectors.
  • the first indication information when used to indicate M 1 space-frequency component vectors, it may specifically be used to indicate a two-dimensional index of the M 1 space-frequency component vectors. That is, the index of the L 1 beam vectors included in the M 1 space frequency component vectors in the beam vector set and the index of the K 1 frequency domain vectors included in the frequency domain vector set.
  • the first indication information when used to indicate the M 1 space frequency component vectors, it may specifically be used to indicate that the M 1 space frequency component vectors are in the space frequency component vector set or a subset of the space frequency component vector set index of.
  • the protocol can predefine the numbering rules of multiple space frequency component vector indexes, and the terminal device and the network device can determine the index of each space frequency component vector in the space frequency component vector set based on the same numbering rule. That is, the position information of the M 1 space frequency component vectors may be the index of each space frequency component vector in the space frequency component vector set.
  • the terminal device can pass Bits to indicate each of the M 1 space-frequency component vectors.
  • the first indication information when used to indicate the M 1 space frequency component vectors, it may specifically be used to indicate the index of the combination of the M 1 space frequency component vectors in the space frequency component vector set.
  • the protocol may predefine multiple combinations of multiple space-frequency component vectors, and each combination corresponds to an index.
  • the M 1 space-frequency component vectors may be one of the multiple combinations, or close to one of the multiple combinations.
  • the first indication information may indicate the M 1 space frequency component vectors by indicating the combined index. That is, the frequency component M 1 empty vector position information may be combinations of the M 1 space-frequency component of the vector components of the vector space in the set of frequency index.
  • the terminal device can pass Bits to indicate M 1 space frequency component vectors in the space frequency component vector set.
  • the first indication information is used to indicate that M 1 space-frequency component vectors are The position in the frequency component vector set may also be understood to indirectly indicate the position of the beam vector and the frequency domain vector corresponding to each space frequency component vector in the beam vector set and the position in the frequency domain vector set, respectively.
  • the position information of the M 1 space frequency component vectors and the position information of the L 1 beam vectors and the position information of the K 1 frequency domain vectors can be converted to each other, or equivalently.
  • the first indication information when the first indication information is used to indicate M 1 space frequency component vectors, it indirectly indicates L 1 beam vectors and K 1 frequency domain vectors.
  • the two methods for indicating M 1 space frequency component vectors listed above are only examples, and should not constitute any limitation to this application.
  • the first indication information may also indicate the M 1 space frequency component vectors in other ways.
  • the first indication information may indicate T 1 space-frequency component vectors in any of the following ways:
  • Method 2 Indicate the index of the combination of T 1 space-frequency component vectors in M 1 space-frequency component vectors
  • Manner 3 Instruct the position of the beam vector corresponding to the T 1 space frequency component vectors in the L 1 beam vectors and the position of the frequency domain vector corresponding to the T 1 space frequency component vectors in the K 1 frequency domain vectors; or
  • Mode 4 Indicate the position of each space-frequency vector pair in the T 1 space-frequency vector pairs in the M 1 space-frequency vector pair.
  • the first indication information further includes quantization information of weighting coefficients of T 1 space frequency vector pairs.
  • the specific process in which the first indication information indicates the weighting coefficients of T 1 space-frequency component vectors has been described in detail.
  • the specific manner in which the first indication information indicates the weighting coefficients of the T 1 space-frequency component vectors may be the same as the specific manner provided in the first implementation manner, and for the sake of brevity, details are not described here.
  • the quantization information of the weighting coefficients of the T 1 space frequency vector pairs may be carried in the first indication information, or may be carried by additional information, which is not limited in this application.
  • the terminal device may specifically determine the T 1 space-frequency component vectors and the weighting coefficients of the space-frequency component vectors through steps 4-i to 4-iv shown below.
  • Step 4-i The terminal device may determine multiple weighting coefficient matrices based on each subset in the set of space-frequency vectors and space-frequency component vectors.
  • the space-frequency component vector set may include O c subsets.
  • the terminal device may separately project the predetermined space-frequency vectors to N s ⁇ N f space-frequency component vectors of each subset to obtain O c sets of projection values, and each set of projection values includes N s ⁇ N f projection values. For each set of projection values, wherein the N s ⁇ N f in the order of the null projection values for each frequency sub-components of the vector in the set of N s ⁇ N f concentrated empty in the order corresponding to the frequency component of the vector.
  • the terminal device can concentrate the respective sub-N s ⁇ N f null frequency component of the vector in the order, the N s ⁇ N f projection values are arranged in order of the predetermined dimensions of N s ⁇ N f matrix.
  • O c matrices corresponding to O c subsets can be obtained, each matrix can correspond to a subset, and each matrix can be called a weighting coefficient matrix corresponding to the subset. Since the numbering rule of the index of each space-frequency component vector in the space-frequency component vector set has been described in detail above, the specific method of constructing the matrix has been described in detail, and for the sake of brevity, it will not be repeated here.
  • the N s ⁇ N f weighting coefficients may be space frequency components of the vector in the first set subsets o c N s ⁇ N f empty vectors corresponding frequency components, may represent a weighting for each empty subset of the frequency component of the vector coefficient.
  • the terminal device may determine O c groups of space frequency component vectors based on the O c subsets in the space frequency component set, and each group of space frequency component vectors may include T 1 space frequency component vectors.
  • the terminal device can traverse the value of o c from 0 to O c -1, and repeatedly perform the following steps to determine the O c group of space-frequency component vectors: according to the matrix W oc, determine the strong L 1 row and the strong K 1 column, and then determine the strong M 1 space frequency component vectors in the o c subset.
  • the terminal device may further determine strong T 1 space frequency component vectors from the M 1 space frequency component vectors.
  • step 3-ii and step 3-iii above the terminal device determines a strong M 1 space-frequency component vector according to the weighting coefficient matrix and determines a strong T from the M 1 space-frequency component vector space-frequency component of the vector a specific process, for brevity, not repeated here.
  • Step 4-iii The terminal device may select the strongest set of space-frequency component vectors based on the weighting coefficients of the O c set of space-frequency component vectors to determine T 1 space-frequency component vectors.
  • the terminal device may determine the strongest set of space-frequency component vectors according to the O c set of space-frequency component vectors determined in step 4-ii. For example, the terminal device may separately calculate the sum of the modulus lengths of the weighting coefficients of each group of space frequency component vectors in the O c group of space frequency component vectors, and determine the group of space frequency component vectors with the largest sum of modulus lengths as the strongest Group of space frequency component vectors.
  • the T 1 space-frequency component vectors and the weighting coefficients of the space-frequency component vectors can be determined.
  • the terminal device T 1 Since the T 1 empty vector frequency components from the same subset, the terminal device T 1 is determined empty frequency component of the vector, while it may be determined that a subset of T 1 empty vector from the frequency component, it can be Determine the strong M 1 space-frequency component vectors in this subset.
  • Step 4-iv The terminal device generates first indication information to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component vectors.
  • the first indication information may include M 1 space-frequency component vectors in the space-frequency component vector set Or location information in a subset of the set of space-frequency component vectors and information indicating T 1 space-frequency vector pairs.
  • the first indication information when used to indicate the M 1 space frequency component vectors, it may specifically be used to indicate the position of the L 1 beam vectors included in the M 1 space frequency component vectors in the beam vector set Information and the position information of the K 1 frequency domain vectors contained in the frequency domain vector set; or, it can be specifically used to indicate the subset to which the M 1 space frequency component vectors belong and the M 1 space frequency component vectors in the The index in the subset; or, it can be specifically used to indicate the index of the subset to which the M 1 space-frequency component vectors belong and the combination of the M 1 space-frequency component vectors in the subset.
  • L 1 beams vector position information beam vector set may refer to the L 1 beams vector index in the beam vector set, or the L 1 combinations beam vectors index in the beam vector set, Or, the subset to which the L 1 beam vectors belong and the index of the L 1 beam vectors in the subset, or, the combination of the subset to which the L 1 beam vectors belong and the combination of the L 1 beam vectors in the sub Centralized index.
  • K 1 frequency-domain vector position information in the frequency domain vector set may mean that the K 1 frequency-domain vector index in the frequency domain vector set, or the K 1 combinations of frequency-domain vectors of the set of frequency domain vector Index in, or, the subset to which the K 1 frequency domain vectors belong and the index of the K 1 frequency domain vectors in the subset, or, the subset to which the K 1 frequency domain vectors belong and the K 1 The index of the combination of frequency domain vectors in this subset.
  • the first indication information is used to indicate that M 1 space-frequency component vectors are The position in the frequency component vector set may also be understood to indirectly indicate the position of the beam vector and the frequency domain vector corresponding to each space frequency component vector in the beam vector set and the position in the frequency domain vector set, respectively.
  • the position information of the M 1 space frequency component vectors and the position information of the L 1 beam vectors and the position information of the K 1 frequency domain vectors can be converted to each other, or equivalently.
  • the first indication information when the first indication information is used to indicate M 1 space frequency component vectors, it indirectly indicates L 1 beam vectors and K 1 frequency domain vectors.
  • the specific indication manner may be any one of the manners 1 to 4 described above.
  • the specific implementation process of indicating the T 1 space-frequency component vectors based on Mode 1, Mode 2, Mode 3, and Mode 4 and the bit overhead brought by them respectively has been described in detail in Implementation Mode 1 above. .
  • the first indication information further includes quantization information of weighting coefficients of T 1 space frequency component vectors.
  • the specific process in which the first indication information indicates the weighting coefficients of T 1 space-frequency component vectors has been described in detail.
  • the specific manner in which the first indication information indicates the weighting coefficients of the T 1 space-frequency component vectors may be the same as the specific manner provided in the first implementation manner, and for the sake of brevity, details are not described here.
  • the quantization information of the weighting coefficients of the T 1 space frequency component vectors may be carried in the first indication information, or may be carried by additional information, which is not limited in this application.
  • Each space-frequency component matrix in the set of space-frequency component matrices is a matrix of dimension N s ⁇ N f .
  • the terminal device may construct a space-frequency matrix H of dimension N s ⁇ N f according to the precoding vectors of the N f frequency domain units,
  • the space frequency component matrix set will be described in detail below.
  • the space frequency component matrix set may include N s ⁇ N f space frequency component matrices.
  • Each space frequency component matrix may be a matrix of dimension N s ⁇ N f .
  • the weighted sum of T 1 space frequency component matrices selected in the space frequency component matrix set can be constructed to obtain a space frequency matrix.
  • the space frequency matrix obtained by weighting and constructing the T 1 space frequency component matrices may be the same as or similar to the space frequency matrix determined by the terminal device above.
  • each space-frequency component matrix in the space-frequency component matrix set may be uniquely determined by one beam matrix in the beam matrix set and one frequency domain matrix in the frequency domain matrix set.
  • any two space-frequency component matrices in the set of space-frequency component matrices are different, and at least one of the beam matrix and the frequency domain matrix corresponding to any two space-frequency component matrices is different.
  • each space-frequency component matrix in the space-frequency component vector matrix may be a product of the conjugate transpose of one beam vector in the set of beam vectors and one frequency domain vector in the set of frequency domain vectors.
  • the space frequency component matrix set may include N s ⁇ N f space frequency component matrices.
  • Each space frequency component matrix may correspond to a beam vector and a frequency domain vector, or each space frequency component matrix may correspond to a space frequency vector pair obtained by combining a beam vector and a frequency domain vector.
  • Each space-frequency component matrix in the set of space-frequency component matrices may correspond to a one-dimensional index or a two-dimensional index. That is, the N s ⁇ N f space frequency component matrices in the space frequency component matrix set can be indicated by the index in the space frequency component matrix set or a subset of the space frequency component matrix set, or can be used to generate the space frequency component matrix The index of the beam vector and frequency domain vector in the beam vector set and frequency domain vector set respectively.
  • the corresponding relationship between each space-frequency component matrix and the one-dimensional index and the conversion rule between the one-dimensional index and the two-dimensional index have been explained in detail in combination with the numbering rule 1 and numbering rule 2. .
  • the set of space frequency component matrices can be expanded to O c ⁇ N s ⁇ N f space frequency component matrices by the oversampling factor O c .
  • the space frequency component matrix set may include O c subsets, and each subset may include N s ⁇ N f space frequency component matrices.
  • the numbering rule of the index of the space-frequency component matrix in each subset may be the same as the numbering rule when no oversampling rate is considered, and for the sake of brevity, no further description is given here.
  • the terminal device may determine T 1 space-frequency component matrices and the weighting coefficients of each space-frequency component matrix through steps 5-i to 5-iv shown below.
  • Step 5-i The terminal device may determine a weighting coefficient based on the above-mentioned space-frequency matrix and space-frequency component matrix set.
  • the space frequency component matrix set may include N s ⁇ N f space frequency component matrices, and the dimension of each space frequency component matrix may be N s ⁇ N f .
  • the terminal device may determine N s ⁇ N f weighting coefficients according to the predetermined space-frequency matrix and the N s ⁇ N f space-frequency component matrices.
  • each terminal device may each space-frequency component of each element of the matrix of the conjugate of the corresponding element of the matrix and the sum-frequency space, and to give N s ⁇ N f null frequency component matrix corresponding to N s ⁇ N f values.
  • the elements in the space-frequency matrix are denoted as b p, q , then the product of the conjugate of each element in each space-frequency component matrix and the corresponding element in the space-frequency matrix can be expressed as among them, Represents the conjugation of elements a p, q .
  • N s ⁇ N f space-frequency component matrices in the space-frequency component matrix can obtain N s ⁇ N f values.
  • the N s ⁇ N f values can be regarded as N s ⁇ N f weighting coefficients.
  • the above steps can be implemented by matrix operations.
  • the above N s ⁇ N f values can be obtained by finding the trace of the product of the conjugate transpose of each space-frequency component matrix and the space-frequency matrix.
  • the terminal device may set a matrix N s ⁇ N f number of the order space frequency components of the matrix in accordance with space-frequency component, the N s ⁇ N f values are arranged in order of a predetermined dimension is the N s ⁇ N f matrix.
  • the specific process in which the terminal device arranges the N s ⁇ N f values into a matrix with dimensions N s ⁇ N f according to a predetermined order has been described in detail in case A above, and for the sake of brevity, it will not be repeated here.
  • the above-mentioned matrix W of dimension N s ⁇ N f can be called a weighting coefficient matrix.
  • the matrix W N s ⁇ N f weighting coefficients may be space frequency component matrix set N s ⁇ N f null frequency component corresponding to the matrix, may represent the N s ⁇ N f null matrix of each frequency component The weighting coefficient of the space-frequency component matrix.
  • Step 5-ii The terminal device may determine strong M 1 space-frequency component matrices according to the weighting coefficient matrix.
  • the specific method for the terminal device to determine the M 1 space-frequency component matrices according to the weighting coefficient matrix is the same as the specific method for the terminal device to determine the strong M 1 space-frequency component vectors in step 3-ii in case A above. Since the specific method has been described in detail above, it will not be repeated here for brevity.
  • Step 5-iii The terminal device may determine a strong T 1 space-frequency component matrix from the M 1 space-frequency component matrix.
  • the terminal device may determine a strong T 1 space frequency component matrix according to the modulus length of the weighting coefficients of the M 1 space frequency component matrix. For example, the modulus length of the weighting coefficient of any one of the selected T 1 space frequency component matrices is greater than or equal to the modulus length of any one of the remaining M 1 -T 1 space frequency component matrices . At the same time, the weighting coefficients of the T 1 space-frequency component matrices can also be determined.
  • Step 5-iv The terminal device generates first indication information to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrix.
  • the first indication information may include M 1 space-frequency component matrices in the space-frequency component matrix set Or position information in a subset of the set of space-frequency component matrices and information indicating T 1 space-frequency component matrices.
  • the first indication information further includes quantization information of weighting coefficients of T 1 space-frequency component matrices.
  • the first indication information indicates that M 1 space-frequency component matrix, T 1 space-frequency component matrix, and the specific method and bit overhead of the weighting coefficients of each space-frequency component matrix are the same as the method shown in case A For the sake of brevity, I will not repeat them here.
  • the first indication information is used to indicate that M 1 space-frequency component matrix is in the empty
  • the position in the frequency component matrix set can also be understood as indirectly indicating the position of the beam vector and the frequency domain vector corresponding to each space frequency component matrix in the beam vector set and the position in the frequency domain vector set, respectively.
  • the position information of the M 1 space frequency component matrix and the position information of the L 1 beam vectors and the position information of the K 1 frequency domain vectors can be converted to each other, or equivalently.
  • the first indication information when the first indication information is used to indicate M 1 space frequency component matrices, it indirectly indicates L 1 beam vectors and K 1 frequency domain vectors.
  • the quantization information of the weighting coefficients of the T 1 space frequency component matrices may be carried in the first indication information, or may be carried by additional information, which is not limited in this application.
  • the terminal device may specifically determine T 1 space-frequency component matrices and the weighting coefficients of each space-frequency component matrix through steps 6-i to 6-iv shown below.
  • Step 6-i The terminal device may determine a plurality of weighting coefficient matrices based on the space frequency vector and each subset in the space frequency component matrix set.
  • the space frequency component matrix set may include O c subsets.
  • the terminal device may determine O c groups of weighting coefficients according to the predetermined space frequency matrix and O c subsets, and each group of weighting coefficients includes N s ⁇ N f weighting coefficients.
  • each group of weighting coefficients includes N s ⁇ N f weighting coefficients.
  • the arrangement order of the N s ⁇ N f values corresponds to the arrangement order of the N s ⁇ N f space frequency component matrices in each subset of the space frequency component vector set.
  • the terminal device can concentrate the order N s ⁇ N f null frequency component in accordance with each sub-matrix, N s ⁇ N f the weighting coefficients arranged in the order of the predetermined dimensions of N s ⁇ N f matrix.
  • N s ⁇ N f the weighting coefficients arranged in the order of the predetermined dimensions of N s ⁇ N f matrix.
  • the N s ⁇ N f weighting coefficients may be null frequency component in the first set matrix subsets o c N s ⁇ N f null frequency component corresponding to the matrix, it may represent the weighting for each empty subset of frequency components of the matrix coefficient.
  • the terminal device may determine O c groups of space frequency component matrices based on the O c subsets in the space frequency component set.
  • Each group of space frequency component matrices may include T 1 space frequency component matrices.
  • the terminal device can traverse the value of o c from 0 to O c -1, and repeatedly perform the following steps to determine the O c group space-frequency component matrix: according to the matrix Determine the strong L 1 row and the strong K 1 column, and then determine the strong M 1 space-frequency component matrix in the o c subset.
  • the terminal device may further determine strong T 1 space-frequency component matrices from the M 1 space-frequency component matrices.
  • the terminal device determines a strong M 1 space-frequency component matrix according to the weighting coefficient matrix and determines a strong T from the M 1 space-frequency component matrix the specific process components of a space-frequency matrix, for brevity, not repeated here.
  • Step 6-iii The terminal device may select the strongest set of space-frequency component matrices based on the weighting coefficients of the O c sets of space-frequency component matrices to determine T 1 space-frequency component matrices.
  • the terminal device may determine the strongest set of space-frequency component matrices according to the O c set of space-frequency component matrices determined in step 6-ii. For example, the terminal device may separately calculate the sum of the modulus lengths of the weighting coefficients of each group of space frequency component matrices in the O c group of space frequency component matrices, and determine the group of space frequency component matrices with the largest sum of modulus lengths as the strongest Group space frequency component matrix.
  • T 1 space-frequency component matrices and the weighting coefficients of each space-frequency component matrix can be determined.
  • the terminal device T 1 Since the T 1 empty frequency component matrix from the same subset, the terminal device T 1 is determined in the frequency component matrix is empty, while it may be determined that the T 1 empty subset of frequency components from the matrix, it can be Determine the strong M 1 space-frequency component matrices in this subset.
  • Step 6-iv The terminal device generates first indication information to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrix.
  • the first indication information may include M 1 space frequency component matrices in the space frequency component matrix set Or location information in a subset of the space frequency component matrix set and information indicating T 1 space frequency matrix matrices.
  • the first indication information further includes quantization information of weighting coefficients of T 1 space-frequency component matrices.
  • the first indication information indicates that M 1 space-frequency component matrix, T 1 space-frequency component matrix, and the specific method and bit overhead of the weighting coefficients of each space-frequency component matrix are the same as the method shown in case A For the sake of brevity, I will not repeat them here.
  • the first indication information is used to indicate that M 1 space-frequency component matrix is in the empty
  • the position in the frequency component matrix set can also be understood as indirectly indicating the position of the beam vector and the frequency domain vector corresponding to each space frequency component matrix in the beam vector set and the position in the frequency domain vector set, respectively.
  • the position information of the M 1 space frequency component matrix and the position information of the L 1 beam vectors and the position information of the K 1 frequency domain vectors can be converted to each other, or equivalently.
  • the first indication information when the first indication information is used to indicate M 1 space frequency component matrices, it indirectly indicates L 1 beam vectors and K 1 frequency domain vectors.
  • the quantization information of the weighting coefficients in the T 1 space-frequency component directions may be carried in the first indication information, or may be carried by additional information, which is not limited in this application.
  • the terminal device may determine multiple weighting coefficients according to the precoding vector of each frequency domain unit and each space-frequency component matrix in the space-frequency component matrix set.
  • the plurality of weighting coefficients may form an array set, and each element in the array set may be formed by sequentially connecting elements of each row (or each column) in the above-mentioned weighting coefficient matrix.
  • the method for the terminal device to obtain the beam vector can refer to the type II codebook feedback method defined in the NR protocol.
  • the frequency domain vector can be obtained by comparing at least one component element of the precoding vector of each frequency domain unit in the same transmission layer and one polarization direction (such as but not limited to the weighting coefficient of the beam vector constituting the precoding vector, etc.) In order to obtain the frequency domain variation law, the same set of frequency domain vectors can be used for different polarization directions.
  • the specific methods for the terminal device listed above to determine the strong T 1 space-frequency component matrices from the M 1 space-frequency component matrices are only examples, and should not constitute any limitation to this application.
  • the terminal device may refer to the method in the prior art to determine the strong T 1 space-frequency component matrix from the M 1 space-frequency component matrix. For brevity, no detailed description is provided here.
  • the first indication information is used to indicate that M 1 space-frequency component matrix is in the empty
  • the position in the frequency component matrix set can also be understood as indirectly indicating the position of the beam vector and the frequency domain vector corresponding to each space frequency component matrix in the beam vector set and the position in the frequency domain vector set, respectively.
  • the position information of the M 1 space frequency component matrix and the position information of the L 1 beam vectors and the position information of the K 1 frequency domain vectors can be converted to each other.
  • the terminal device may generate first indication information to indicate the selected L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrix And the weighting coefficient of each space-frequency component matrix.
  • step 220 the terminal device sends the first indication information.
  • the network device receives the first indication information.
  • the first indication information may be PMI, or some information elements in the PMI, or other information. This application does not limit this.
  • the first indication information may be carried in one or more messages in the prior art and sent by the terminal device to the network device, or may be carried in one or more messages newly designed in the present application and sent by the terminal device to the network device.
  • the terminal device may send the first indication information to the network device through physical uplink resources, such as a physical uplink shared channel (physical uplink shared channel (PUSCH) or physical uplink control channel (PUCCH), for example, to facilitate the network device
  • PUSCH physical uplink shared channel
  • PUCCH physical uplink control channel
  • the specific method for the terminal device to send the first indication information to the network device through the physical uplink resource may be the same as that in the prior art. For brevity, a detailed description of the specific process is omitted here.
  • step 230 the network device determines the precoding vectors of one or more frequency domain units according to the first indication information.
  • the terminal device can generate the first indication information based on two different implementations.
  • the information used to indicate L 1 beam vectors and the information used to indicate K 1 frequency domain vectors in the first indication information may be different or the same.
  • the specific process of the network device determining the precoding vectors of one or more frequency domain units according to the first indication information is described in detail below.
  • the first indication information includes position information of L 1 beam vectors, position information of K 1 frequency domain vectors, and information indicating T 1 space frequency component matrices.
  • the network device may determine the selected L 1 beam vector and K 1 frequency domain vector according to the position information of L 1 beam vector and the position information of K 1 frequency domain vector.
  • the network device may determine that the beam vector set is selected according to the index of the combination of L 1 beam vectors indicated in the first indication information in the beam vector set, and the correspondence between the predefined combination of beam vectors and the index L 1 beam vector.
  • the network device may also determine the frequency domain vector set according to the index of the K 1 frequency domain vector combination indicated in the first indication information in the frequency domain vector set, and the correspondence between the predefined frequency domain vector combination and the index K 1 frequency domain vectors selected.
  • the L 1 beam vectors and K 1 frequency domain vectors can be combined to obtain M 1 space frequency vector pairs.
  • the network device may indicate empty T 1 according to the frequency information component matrix, T 1 is determined null frequency component from the matrix M 1 in the empty frequency vectors.
  • the first indication information may indicate the T 1 space-frequency vector pairs to the network device in different ways.
  • the first indication information indicates the selected T 1 space-frequency vector pairs among the M 1 space-frequency vector pairs through a bitmap, and the network device may determine the number of M 1 space-frequency vector pairs according to each bit in the bitmap.
  • the one-to-one correspondence between the space-frequency vector pairs determines the selected T 1 space-frequency vector pairs; in mode 2, the first indication information is included in the M 1 space-frequency vector pairs through the combination of T 1 space-frequency vector pairs To indicate the M 1 space-frequency vector pairs, the network device may determine the T 1 space-frequency vector pairs according to the correspondence between the predefined combination of space-frequency vector pairs and the index; in mode 3, the first indication index position information contained in the beam vector and the frequency domain vector to indicate that the T 1 empty frequency vectors T 1 by the empty-frequency vector for each vector space-frequency, the network device may be the position vector of the beam The position information of the information and the frequency domain vector, determine T 1 beam vector and T 1 frequency domain vector, and further combine to obtain T 1 space frequency vector pair.
  • the network device may determine the quantization value of the weighting coefficient of each space-frequency vector pair based on the quantization information based on the weighting coefficient of each space-frequency vector pair.
  • the weighting coefficients of each space-frequency vector pair can be carried by the first indication information or other information.
  • the network device may determine the quantization value of the weighting coefficient of each space-frequency vector pair according to the predefined one-to-one correspondence between the plurality of quantization values and the multiple indexes.
  • the space frequency matrix is determined based on the T 1 space frequency vector pairs and the quantized values of the weighting coefficients of the space frequency vector pairs.
  • the network device may determine the space-frequency matrix H based on the following formula:
  • T 1 represents the frequency-domain frequency-domain vectors T 1 vectors.
  • the space-frequency matrix H in the formula may be the same as or similar to the space-frequency matrix determined by the terminal device, and is a space-frequency matrix recovered by the network device according to the first indication information. Since the space-frequency matrix can be constructed from the precoding vectors corresponding to the N f frequency domain units, the network device can determine the precoding vector corresponding to the n f th frequency domain unit according to the n f column vector in the matrix H.
  • the network device may further generate T 1 space-frequency component matrices after determining T 1 space-frequency vector pairs. For example, for the t 1st space-frequency component matrix of T 1 space-frequency component matrices, it can be written as or, Wait.
  • the network device may determine the space-frequency matrix based on the above T 1 space-frequency component matrix and the quantized values of the weighting coefficients of each space-frequency component matrix. For example, the network device may determine the space-frequency matrix H according to the following formula:
  • the network device may determine the precoding vector corresponding to each frequency domain unit according to the space-frequency matrix.
  • each column vector in the matrix can correspond to a frequency domain unit, and can be used for the precoding vector of the corresponding frequency domain unit. If the space-frequency matrix is a vector of length N s ⁇ N f , then the column vector obtained by sequentially connecting the n f ⁇ N s element to the (n f +1) N s -1 element in the vector n f can be the first frequency domain corresponding to column vector cell.
  • network devices can process the normalized column vectors of n f, n f to determine the first frequency-domain precoding vector corresponding to the unit.
  • the normalization process may be, for example, the column vector of n f is multiplied by a normalization coefficient, so that the power of the column vector elements is equal to one.
  • the normalization coefficient may be, for example, the reciprocal of the square root of the sum of the modulus lengths of the elements in this column. This application does not limit the specific value of the normalization coefficient and the specific method of the normalization process.
  • the first indication information includes position information of M 1 space frequency component matrices and information indicating T 1 space frequency component matrices.
  • the frequency components of M 1 empty matrix position information may be the frequency component M 1 empty one-dimensional index matrix, M 1 may be the empty two-dimensional frequency component matrix index.
  • the network device may be directly based on the one-dimensional index , Determine the M 1 space-frequency component matrices from the space-frequency component matrix set.
  • the network device If the location information of M 1 space-frequency component of a matrix M 1 for the empty one-dimensional index-frequency component, and the network device the pre-stored set of vectors and a set of frequency-domain vectors beam vector set, the network device according to a predefined The conversion rule of the one-dimensional index and the two-dimensional index determines from the beam vector set and the frequency domain vector set L 1 beam vectors and K 1 frequency domain vectors used to generate the M 1 space frequency component matrix.
  • the network device may be directly based on the The two-dimensional index determines L 1 beam vectors and K 1 frequency domain vectors used to generate the M 1 space frequency component matrixes from the beam vector set and the frequency domain vector set.
  • the network device may be pre-defined two-dimensional The conversion rule of the index and the one-dimensional index determines M 1 space frequency component matrices from the space frequency component matrix set.
  • the network device may further determine T 1 space-frequency component matrices from the M 1 space-frequency component matrices, and may further determine the weighting coefficient of each space-frequency component matrix based on the quantization information of the weighting coefficients of each space-frequency component matrix Quantified value.
  • the specific method for the network device to determine the space-frequency matrix according to the T 1 space-frequency component matrix and the quantized values of the weighting coefficients of each space-frequency component matrix has been described in detail above, and the network device may, for example, use the formula Determine the space frequency matrix.
  • the network device may determine the precoding vector of one or more frequency domain units. Since the specific method for the network device to determine the precoding vector according to the space-frequency matrix has been described in detail above, it will not be repeated here for brevity.
  • the network device may not generate a space frequency matrix, based on the instructions of the terminal device L 1 beam vectors and K 1 frequency domain vectors directly determine the precoding vectors on one or more frequency domain units.
  • the first network device may determine precoding vectors n f in the frequency domain based on the following unit formula
  • T 1 represents the first vector T 1 th beam beams of vector is selected
  • T 1 represents the frequency-domain frequency-domain vectors T 1 in the vector is selected
  • n fth element in Represents the t 1st beam vector T 1 and frequency-domain vectors
  • Corresponding weighting coefficients which may include amplitude coefficients and phase coefficients, for example.
  • the length N f of the frequency domain vector may be the number of frequency domain units included in the frequency domain occupied bandwidth of the CSI measurement resource allocated to the terminal device, or the signaling length of the reporting band, Or, the number of frequency domain units to be reported.
  • the length of the frequency domain vector is the number of frequency domain units included in the frequency domain occupied bandwidth of the CSI measurement resource allocated to the terminal device or the signaling length of the reporting band
  • the number of frequency domain units to be reported may be less than or equal to N f 's. Therefore, the network device may determine the precoding vector of each frequency domain unit according to the position of the frequency domain unit to be reported indicated by the reporting band or other signaling.
  • the length of the frequency domain vector is determined according to the number of frequency domain units included in the frequency domain occupied bandwidth of the CSI measurement resource or the signaling length of the reporting band, and the change rule of the channel on multiple consecutive frequency domain units can be The frequency domain vector is reflected.
  • the number of frequency domain units in the frequency domain occupied bandwidth of the CSI measurement resource or reporting information can more accurately reflect the changing law of the channel in the frequency domain, and the precoding vector recovered based on the feedback can also be more adapted to the channel.
  • the network device listed above and n f the specific method of precoding vectors corresponding to the unit of the frequency domain are merely examples according to the first indication information, not constitute any limitation on the present application deal with. This application does not exclude a network device according to the first indication information, other means of determining a precoding vector corresponding to the second sub-band may be n f.
  • the terminal device indicates a small number of beam vectors, frequency domain vectors, and space-frequency component matrices to the network device, so that the network device can recover the precoding vector.
  • the frequency domain vector can be used to describe the different changing rules of the channel in the frequency domain.
  • the terminal device can simulate the change of the channel in the frequency domain through the linear superposition of one or more frequency domain vectors, fully exploit the relationship between the frequency domain units, utilize the continuity of the frequency domain, and use fewer frequency domain vectors To describe the changing rules of multiple frequency domain units.
  • the feedback overhead will not increase exponentially. Therefore, feedback overhead can be greatly reduced on the basis of ensuring feedback accuracy.
  • the frequency domain vector set may also contain more frequency domain vectors. If the beam vector set and the frequency domain vector set are indicated directly with fewer beam vectors and frequencies The domain vector may bring a large bit overhead, or the terminal device and the network device need to predefine a large number of beam vector combination and index correspondence and a frequency domain vector combination and index correspondence.
  • the terminal device reduces the selection range of the beam vector and the frequency domain vector used for weighted summation to the M 1 space-frequency component matrix constructed by L 1 beam vector and K 1 frequency domain vector.
  • the range that is, first select a smaller range from the existing vector set, and then further select and indicate T 1 space-frequency component matrices from the range.
  • it can avoid directly indicating the large feedback overhead caused by the T 1 space-frequency component matrices; on the other hand, it can avoid saving a large number of correspondences between the terminal device and the network device.
  • the terminal device indicates a precoding vector on a transmission layer and a polarization direction and the network device determines the precoding vector is described in detail above in conjunction with FIG. 2.
  • this method is not only applicable to the case where the transmission layer is 1 or the number of polarization directions is 1, but is also applicable to the case of multiple transmission layers or multiple polarization directions.
  • T 1 space frequency component matrices (or T 1 space frequency vector pairs) selected in multiple polarization directions may be the same. That is, multiple polarization directions share the same T 1 space-frequency component matrix (or T 1 space-frequency vector pair), or different space-frequency component matrices (or space-frequency vector pair) may be used respectively.
  • a space frequency component matrix is taken as an example to illustrate the case of multiple polarization directions or multiple transmission layers.
  • the space-frequency component matrix may include the matrix form or the vector form listed above. It can be understood that the space-frequency component matrix is only one possible form, and may also be expressed in the form of space-frequency vector pairs. That is, the space-frequency component matrix referred to below can also be replaced with a space-frequency vector pair.
  • the first indication information is used to indicate the precoding vector of each frequency domain unit in one or more polarization directions.
  • L 1 beam vectors and K 1 frequency domain vectors used to determine the T 1 space frequency component matrix may also be multiple polarization directions Shared.
  • information for indicating L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrices may be common.
  • the terminal device may use a bitmap of length L 1 ⁇ K 1 to indicate T 1 space-frequency component matrices used in each polarization direction among the multiple polarization directions.
  • the terminal device may only generate and send one first indication information for multiple polarization directions . If the first indication information is also used to indicate the weighting coefficients of T 1 space frequency component matrices, the terminal device may send one piece of first indication information for each polarization direction.
  • the information used to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrices may be common, and the terminal device may only Indicate the L 1 beam vectors, K 1 frequency domain vectors, and T 1 space-frequency component matrices once, and the weighting coefficients of the space-frequency component matrices corresponding to different polarization directions can be obtained through different first indication information Instructions.
  • the description of the same or similar cases is omitted in the following.
  • the terminal device may use L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrices determined based on a certain polarization direction for multiple polarization directions. Based on which polarization direction the terminal device determines L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrices may be pre-defined, as defined by the protocol, which is not limited in this application.
  • the terminal device may also determine T 1 space-frequency component matrices based on each polarization direction to obtain multiple sets of space-frequency component matrices. And select one set from the multiple sets of space-frequency component matrix for multiple polarization directions.
  • the sum of the modulus lengths of the selected weighting coefficients of the set of space frequency component matrices may be greater than or equal to the sum of the modulus lengths of the weighting coefficients of any one set of space frequency component matrices in the remaining one or more sets of space frequency component matrices.
  • the number of space frequency component vector matrices used in different polarization directions may be the same or different.
  • the number of beam vectors used to determine the space-frequency component matrix in each polarization direction may be the same or different; the number of frequency domain vectors used to determine the space-frequency component matrix in each polarization direction may also be the same, or It can be different, and this application does not limit it.
  • the information of the beam vector, the frequency domain vector, and the space-frequency component matrix corresponding to the different polarization directions may be different from each other.
  • the terminal device may indicate the selected beam vector, frequency domain vector, and space-frequency component matrix used as the weighted summation based on each polarization direction, respectively.
  • the specific manner in which the terminal device determines the beam vector, the frequency domain vector, and the space-frequency component matrix used as the weighted sum based on each polarization direction is the same as the above-described determination of L 1 beam vector, K 1 based on one polarization direction
  • the specific methods of the frequency domain vector and the T 1 space-frequency component matrix are the same, and for the sake of brevity, they will not be repeated here.
  • multiple polarization directions can share the same L 1 beam vector, K 1 frequency domain vector, and T 1 space frequency component matrix.
  • the following uses an example in which multiple polarization directions share the same L 1 beam vector, K 1 frequency domain vector, and T 1 space frequency component matrix.
  • T 1 space-frequency component matrices used as weighted summation in the two polarization directions It's the same.
  • Network device may determine a precoding vector n f of the frequency-domain units based on the following equation:
  • the weighting coefficient corresponding to the n fth element in Represents the t 1st beam vector in the second polarization direction T 1 and frequency-domain vectors The n fth element in Corresponding weighting factor.
  • the t 1 beam vector and the broadband amplitude coefficient determine
  • From the selected T 1 beam vector in the first polarization direction, the t 1 beam vector and the broadband amplitude coefficient determine Represents the direction of the first polarization T 1 and frequency-domain vectors
  • the weighting coefficient corresponding to the n fth element in Indicates the direction of the second polarization T 1 and frequency-domain vectors
  • the n fth element in Corresponding weighting factor.
  • the two polarization directions share the same L 1 beam vector, K 1 frequency domain vector, and T 1 space frequency component matrix.
  • the terminal device uses the bitmap described in Mode 1 to indicate T 1 space-frequency component matrices, it can specifically be indicated by a bitmap with a length of L 1 ⁇ K 1 .
  • the two polarization directions share the same L 1 beam vector and K 1 frequency domain vector
  • the terminal device may determine the direction of each polarization direction based on the L 1 beam vector and K 1 frequency domain vector.
  • the value of S 1 may be 2T 1, the terminal device may determine a strong T 1 th frequency component of the matrix are empty on each direction of polarization of the two polarization directions of the first polarization direction T 1 of th
  • the space frequency component matrix and the T 1 space frequency component matrices in the second polarization direction may be the same or different, which is not limited in this application; the value of S 1 may not be 2T 1 , and the terminal device may be based on two The polarization directions together determine S 1 strong space-frequency component matrix.
  • the total number of space-frequency component matrices in the first polarization direction and the second polarization direction may be S 1 , and the first The number of space-frequency component matrices in the polarization direction and the number of space-frequency component matrices in the second polarization direction may be the same or different, which is not limited in this application.
  • the terminal device When the terminal device indicates the S 1 space-frequency component matrices, it can still use any one of the methods 1 to 4 listed above to indicate.
  • the bitmap described in Mode 1 is used for indication, the length of the bitmap may be 2L 1 ⁇ K 1 bit, or may also be indicated by bitmaps corresponding to two polarization directions respectively.
  • the third indication information may also be used to indicate the value S 1.
  • S 1 can be an even number, such as 2T 1 ; if the terminal device determines and reports the space-frequency component matrix based on the two polarization directions, then S 1 It can be odd or even, which is not limited in this application.
  • the two polarization directions share the same L 1 beam vector
  • the terminal device may determine K 1 frequency domain vectors and T 1 space frequency component matrices based on each polarization direction.
  • the two polarization directions share the same L 1 beam vector
  • the terminal device may determine K 1 frequency domain vectors based on each polarization direction, and jointly determine 2T 1 stronger based on the two polarization directions.
  • Space frequency component matrix may be used to determine K 1 frequency domain vectors based on each polarization direction, and jointly determine 2T 1 stronger based on the two polarization directions.
  • the T 1 space frequency component matrices indicated by the first indication information are associated with the first transmission layer among the multiple transmission layers. That is, the T 1 space frequency component matrices indicated by the first indication information can be used to determine the precoding vectors on one or more frequency domain units on the first transmission layer.
  • the first transmission layer may be one transmission layer or multiple transmission layers, which is not limited in this application.
  • the first indication information can be used to determine the precoding vector of each frequency domain unit on one or more transmission layers.
  • the first indication information may be used to indicate one or more polarization directions, and / or, precoding vectors of each frequency domain unit on one or more transmission layers. That is, the first indication information can be used to determine the precoding vector of each frequency domain unit in one or more polarization directions, or can be used to determine the precoding vector of each frequency domain unit on one or more transmission layers, and can also be used To determine the precoding vector of each frequency domain unit in one or more polarization directions on each transmission layer in one or more transmission layers.
  • the method further includes: the terminal device generates fourth indication information, which is used to indicate L 2 beam vectors in the beam vector set, K 2 frequency domain vectors in the frequency domain vector set, and T 2 Matrix of space-frequency components.
  • the weighted sum of the T 2 space frequency component matrices can be used to determine the precoding vectors of one or more frequency domain units on the second transmission layer.
  • L 2 beam vectors, K 2 frequency domain vectors, and T 2 space frequency component matrices indicated by the fourth indication information are associated with the second transmission layer.
  • the K 2 frequency domain vectors are pre-configured.
  • the K 2 frequency domain vectors may be all or part of the frequency domain vectors in the set of frequency domain vectors.
  • the protocol may predefine the value of K 2 and specify in advance which frequency domain vectors in the frequency domain vector set are used as the K 2 frequency domain vectors.
  • the protocol may predefine the value of K 1 , and the network device may indicate the K 2 frequency domain vectors through signaling in advance.
  • the predefined terminal device does not need to report the K 2 frequency domain vectors.
  • the K 2 frequency domain vectors may be specified in advance, such as protocol definition or network device configuration, etc., which is not limited in this application.
  • the K 2 frequency domain vectors are the same as K 1 frequency domain vectors.
  • the K 2 frequency domain vectors are different from K 1 frequency domain vectors.
  • the K 2 frequency domain vectors are a subset of K 1 frequency domain vectors.
  • the vectors reported by the terminal device in the case where different parameter values and different parameters are configured have been explained in detail in conjunction with the first indication information.
  • the terminal device When the terminal device generates and sends the fourth indication information, it may be based on the same Way to deal with. For brevity, I will not repeat them here.
  • the second transmission layer is one or more transmission layers of the plurality of transmission layers except the first transmission layer.
  • the fourth indication information can be used to determine the precoding vector of each frequency domain unit on one or more transmission layers.
  • the fourth indication information may be used to indicate the precoding vector of each frequency domain unit in one or more polarization directions on each transmission layer in the second transmission layer.
  • the number of beam vectors determined on any two transmission layers is the same, the number of frequency domain vectors determined on any two transmission layers is the same, and on any two transmission layers The number of determined spatial frequency component matrices is the same.
  • each transmission layer may share the same L 1 beam vector, K 1 frequency domain vector, and T 1 space frequency component matrix.
  • the fourth indication information and the information for indicating L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrices in the first indication information may be common.
  • the first indication information and the fourth indication information may be the same indication information.
  • each transmission layer may share the same L 1 beam vector and K 1 frequency domain vector, but each uses its own T 1 space frequency component matrix. That is, a first transport layer is different from a T space frequency components of the matrix and the second transport layer T a space frequency component matrix.
  • the fourth indication information and the information for indicating L 1 beam vectors and K 1 frequency domain vectors in the first indication information may be common, and the first indication information and the fourth indication information may be used separately Indicates the T 1 space-frequency component matrix on the corresponding transmission layer.
  • each transmission layer uses its own L 1 beam vector, K 1 frequency domain vector, or T 1 space frequency component matrix, respectively. That is, L 1 beam vectors of the first transmission layer are different from L 1 beam vectors of the second transmission layer, K 1 frequency domain vectors of the first transmission layer are different from L 1 beam vectors of the second transmission layer, and first transport layer T of a space-frequency component matrix T of the second transport layer frequency component different from a null matrix.
  • the first indication information and the second indication information may be used to indicate L 1 beam vectors, K 1 frequency domain vectors, and T 1 space frequency component matrices on the corresponding transmission layer, respectively.
  • L 1 , K 1 and T 1 may change as the number of transmission layers increases.
  • the protocol may be defined in advance, and when the number of transmission layers is greater than a certain preset threshold, at least one of L 1 , K 1 and T 1 decreases.
  • T 1 when the number of transmission layers is greater than 2, T 1 can be reduced. As, T 1 may be reduced to T 1/2, or reduced to T 1/3.
  • both L 1 and T 1 can be reduced.
  • L 1 may be reduced to L 1/2
  • T 1 may be reduced to T 1/2
  • L 1 may be reduced to L 1/3
  • T 1 may be reduced to T 1/3.
  • both K 1 and T 1 can be reduced.
  • K 1 may be reduced to K 1/2
  • T 1 may be reduced to T 1/2
  • K 1 may be reduced to K 1/3
  • T 1 may be reduced to T 1/3.
  • L 1 , K 1 and T 1 can all be reduced.
  • L 1 may be reduced to L 1/2
  • K 1 may be reduced to K 1/2
  • T 1 may be reduced to T 1/2
  • L 1 may be reduced to L 1/3
  • K 1 may be reduced to K 1/3
  • T 1 may be reduced to T 1/3.
  • first transport layer and the second transport layer are taken as examples to illustrate the size relationships of L 1 and L 2 , K 1 and K 2, and T 1 and T 2 on different transport layers, this This application should not constitute any limitation.
  • the number of transmission layers is not limited to 2, it may be greater than 2.
  • the number of transmission layers is 3, 4, etc., which is not limited in this application.
  • the number of transmission layers is greater than a preset threshold, the number of at least one of the space frequency component matrix, the beam vector, and the frequency domain vector in some transmission layers may be reduced.
  • the number of spatial frequency component matrices in some transmission layers may be halved. That is, the number T 1 of space frequency component matrices of the first transmission layer may be twice the number T 2 of space frequency component matrices of the second transmission layer.
  • the first transmission layer may, for example, correspond to the precoding vector determined by the feature vector corresponding to the larger feature value in the SVD process
  • the second transmission layer may, for example, correspond to the feature vector determined by the smaller feature value in the SVD process.
  • Precoding vector may represent one transmission layer or multiple transmission layers with the same characteristics; the second transmission layer may represent one transmission layer or multiple transmission layers with the same characteristics.
  • the number of first indication information used to indicate the precoding vectors of the first transmission layer may be multiple to correspond to multiple transmission layers. If the multiple transmission layers share the beam vector, the frequency domain vector, and the space-frequency component matrix, the information used to indicate the beam vector, the frequency domain vector, and the space-frequency component matrix in the plurality of first indication information can be shared. At this time, only one first indication message may be generated and sent.
  • the number of fourth indication information used to indicate the precoding vectors of the second transmission layer may be multiple to correspond to multiple transmission layers. If the multiple transmission layers share beam vectors, frequency domain vectors, or space-frequency component matrices, the information used to indicate the beam vectors, frequency domain vectors, or space-frequency component matrices in the multiple fourth indication information can be shared. At this time, only one fourth indication information may be generated and sent.
  • the protocol may also predefine other criteria for dividing the transport layer, which is not limited in this application.
  • the preset threshold is 2. Then the number of space-frequency component matrices of the second and third transmission layers in the four transmission layers can be halved, and the number of space-frequency component matrices of the 0th transmission layer and the first transmission layer The number is unchanged.
  • the 0th transport layer and the 1st transport layer may be two examples of the first transport layer
  • the 2nd transport layer and the 3rd transport layer may be two examples of the second transport layer.
  • the number of beam vectors and / or the number of frequency domain vectors in some transmission layers can also be reduced.
  • no more examples are given here.
  • the terminal device may divide part of the first transmission layer The space frequency component matrix is used for the second transmission layer.
  • the terminal device may indicate only part of the space-frequency component matrix used for the second transmission layer in the fourth indication information, for example, by means of a bitmap, an index of the space-frequency component matrix, etc., without repeating the indication The selected beam vector and frequency domain vector, thereby reducing feedback overhead.
  • the terminal device may indicate the relative position of the partial space-frequency component matrix used for the second transmission layer in the space-frequency component matrix used for the first transmission layer, or a local position, such as relative number or relative Index etc.
  • the terminal device can indicate the selected beam vector, frequency domain vector, space-frequency component matrix and the space-frequency component matrix of the selected transmission layer on the corresponding transmission layer through the first indication information and the fourth indication information respectively Weighting factor.
  • the number of transmission layers is 4, and the 0th transmission layer and the 1st transmission layer may both be the first transmission layer.
  • the 0th transmission layer and the 1st transmission layer can share the same beam vector and frequency domain vector.
  • Two first indication information corresponding to the 0th transmission layer and the 1st transmission layer are used to indicate the 0th
  • indication information of L 1 beam vectors and indication information of K 1 frequency domain vectors may be generated and indicated only once.
  • the space frequency component matrices used by the 0th transmission layer and the 1st transmission layer can be indicated by different bitmaps.
  • the first indication information corresponding to the 0th transmission layer may include a bitmap for indicating the space frequency component matrix of the 0th transmission layer; the first indication information corresponding to the 1st transmission layer may include Bitmap of the space-frequency component matrix of the first transmission layer.
  • Both the second transport layer and the third transport layer may be the second transport layer.
  • the second transmission layer and the third transmission layer can also share the same beam vector and frequency domain vector, and the beam vector used by the second transmission layer and the third transmission layer can be the 0th transmission layer and the first A subset of the beam vectors used by each transmission layer, the frequency domain vectors used by the second transmission layer and the third transmission layer can also be the frequency domain vectors used by the 0th transmission layer and the first transmission layer Subset.
  • the beam vectors used by the second transmission layer and the third transmission layer may also be determined by the terminal device based on the second transmission layer and / or the third transmission layer, not necessarily the 0th transmission layer and A subset of the beam vectors used by the first transmission layer; the frequency domain vectors used by the second transmission layer and the third transmission layer may also be based on the second transmission layer and / or the third transmission layer of the terminal device Determined, but not necessarily a subset of the frequency domain vectors used by the 0th transmission layer and the 1st transmission layer. This application does not limit this.
  • the second transmission layer and the third transmission layer are a subset of the beam vectors used by the 0th transmission layer and the first transmission layer
  • the second transmission layer and the third transmission layer When the frequency domain vector used by the layer is a subset of the frequency domain vector used by the 0th transmission layer and the 1st transmission layer, two fourth indicators corresponding to the 2nd transmission layer and the 3rd transmission layer respectively
  • indication information of L 2 beam vectors and K 2 frequency domain vectors may be generated and indicated only once.
  • the L 2 beam vectors and K 2 frequency domain vectors can be indicated in L 1 beam vectors and K 1 frequency domain vectors, for example, the relative position of the L 2 beam vectors in L 1 beam vectors can be indicated , It can also indicate the relative position of K 2 frequency domain vectors in K 1 frequency domain vectors.
  • the space-frequency component matrices used by the second transmission layer and the third transmission layer can be indicated by different bitmaps.
  • the fourth indication information corresponding to the second transmission layer may include a bitmap indicating the space-frequency component matrix of the second transmission layer; the fourth indication information corresponding to the third transmission layer may include an indication Bitmap of the space-frequency component matrix of the third transmission layer.
  • the network device can indicate a value for at least two of L 1 , K 1 and T 1 respectively, and the terminal device can determine whether L 1 , K 1 or T 1 needs to be changed according to the number of transmission layers and predefined rules Value.
  • the first network device may determine the r th transport layer precoding vector f n frequency-domain units based on the following equation:
  • n fth element in Represents the t 1 th beam vector in the first polarization direction with the r th transmission layer T 1 and frequency-domain vectors
  • the n fth element in The corresponding weighting factor Represents the t 1 th beam vector in the second polarization direction with the r th transmission layer T 1 and frequency-domain vectors
  • the network device may determine a precoding matrix n f of the frequency domain based on the following formula units
  • 0 represents a transport layer of the first precoding vector f n frequency domain units
  • the first R-1 represents the first transmission layer precoding vector f n frequency domain elements.
  • the terminal device may indicate to the network device the precoding matrix of one or more frequency domain units on one or more transmission layers.
  • the terminal device indicates a small number of beam vectors, frequency domain vectors, and space-frequency component matrices to the network device, so that the network device recovers the precoding vector.
  • the frequency domain vector can be used to describe the different changing rules of the channel in the frequency domain.
  • the terminal device can simulate the change of the channel in the frequency domain through the linear superposition of one or more frequency domain vectors, fully exploit the relationship between the frequency domain units, utilize the continuity of the frequency domain, and use fewer frequency domain vectors To describe the changing rules of multiple frequency domain units.
  • the feedback overhead will not increase exponentially. Therefore, feedback overhead can be greatly reduced on the basis of ensuring feedback accuracy.
  • the terminal device may feed back weighting coefficients to the M 1 space frequency component matrices corresponding to the L 1 beam vectors and K 1 frequency domain vectors, respectively. .
  • the terminal device may be empty frequency component M 1 T 1 in the matrix weighting coefficients matrix space-frequency component with a higher feedback accuracy, while the remaining M 1 -T 1 weighting coefficient matrix using space-frequency component than Low feedback accuracy to reduce feedback overhead.
  • the number of quantization bits of the weighting coefficient of any one of the T 1 space-frequency component matrices may be greater than the number of quantization bits of the weighting coefficient of any one of the remaining M 1 -T 1 space-frequency component matrices.
  • the resulting feedback overhead is slightly larger than the feedback overhead brought by the method provided above, the precoding matrix recovered by the network device may be closer to the precoding matrix determined by the terminal device, therefore, the approximate accuracy higher.
  • the instructions and the method for determining the precoding matrix provided above are particularly suitable for the case where the number of frequency domain units is large and continuous. This method can make full use of the correlation between frequency domain units and compress the feedback overhead. The huge overhead caused by independent feedback of multiple frequency domain units is reduced.
  • the number of frequency domain units is not very large, or the frequency domain units are not continuous. If the existing frequency domain units are independently fed back to indicate the precoding vectors of each frequency domain unit, May not bring a lot of feedback overhead. In addition, in the case where the frequency domain units are discontinuous, a high approximation accuracy can be ensured through the independent feedback of each frequency domain unit. Therefore, the present application also provides a method for feeding back PMI, which can combine different scenarios and adopt a feedback mode to feed back precoding vectors, which can not only ensure approximate accuracy, but also save feedback overhead.
  • FIG. 3 is a schematic flowchart of a method 300 for feedback PMI provided by another embodiment of the present application from the perspective of device interaction. As shown, the method 300 may include steps 310 to 340. The steps of the method 300 are described in detail below.
  • the terminal device may indicate the precoding matrix to the network device through PMI.
  • the PMI may include the first indication information and / or the fourth indication information in the method 200 above, and may also include other information other than the first indication information and the fourth indication information that may be used to indicate the precoding matrix.
  • PMI is only a name used to indicate the information of the precoding matrix, and should not constitute any limitation to this application. This application does not exclude the possibility of defining other names in the future agreement to indicate the same or similar functions as PMI.
  • step 310 the network device generates sixth indication information, which is used to indicate the feedback mode of the PMI.
  • the feedback mode of the PMI may be the feedback mode provided above or other feedback modes.
  • the feedback mode of the PMI may be the first feedback mode or the second feedback mode.
  • the first feedback mode may be a mode for feeding back PMI based on beam vector sets only
  • the second feedback mode may be a mode for feeding back PMI based on beam vector sets and frequency domain vector sets
  • the feedback PMI based on the beam vector set and the frequency domain vector set and the feedback PMI based on the space frequency component matrix set can be Think it is equivalent.
  • the beam vector set and the frequency domain vector set and the space frequency component matrix set can be converted between each other, it can also be considered that the second feedback mode is based on the beam vector set and the space frequency component matrix set feedback PMI, or, based on the frequency domain vector Set and space frequency component matrix set feedback PMI. This application does not limit this.
  • the feedback PMI based on the beam vector set only is referred to the first feedback mode relative to the second feedback mode.
  • the first feedback mode may only feedback the PMI based on the beam vector set without providing an additional vector set or matrix set.
  • the difference between the first feedback mode and the second feedback mode is that the first feedback mode is not based on the set of frequency domain vectors, and the second feedback mode is based on the set of frequency domain vectors.
  • the first feedback mode may be a mode of independent feedback by frequency domain units
  • the second feedback mode may be a mode of joint feedback by frequency domain units
  • the first feedback mode may refer to the feedback mode based on the type II codebook feedback PMI defined in the NR protocol TS38.214R15.
  • the second feedback mode may be, for example, the feedback mode described above in connection with the method 200.
  • the second feedback mode can be understood as a feedback mode of joint feedback of frequency domain units.
  • the second feedback mode is based on the continuity of the frequency domain, and uses the relationship between the frequency domain units to jointly feed back multiple frequency domain units, thereby compressing the frequency domain feedback overhead.
  • the second feedback mode can greatly reduce the feedback overhead compared to the first feedback mode.
  • the sixth indication information may display the feedback mode.
  • an indication bit or an indication field may indicate the feedback mode. For example, when the indication bit is set to "0", the first feedback mode is used, when the indication bit is set to "1", the second feedback mode is used; or, when the indication bit is set to "1", the first feedback mode is used, indicating that the bit is set to " 0 "indicates that the second feedback mode is used. This application does not limit this.
  • the sixth indication information may also implicitly indicate the feedback mode through other information. For example, when the network device indicates the length of the frequency domain vector to the terminal device, it may be considered that the network device requires the terminal device to feed back the precoding vector based on the second feedback mode. At this time, the fifth indication information for indicating the length of the frequency domain vector described above can be understood as an example of the sixth indication information.
  • the length of the space-frequency component vector can be determined by the length of the frequency domain vector and the length of the beam vector. Therefore, when the network device indicates the length of the frequency domain vector to the terminal device, the terminal device can select the vector and
  • the frequency domain vector feedback PMI may also be based on the space frequency vector feedback PMI, which is not limited in this application.
  • step 320 the network device sends sixth indication information. Accordingly, the terminal device receives the sixth indication information.
  • the sixth indication information is carried in the RRC message.
  • the specific method for the network device to send the sixth indication information to the terminal device may be the same as the method for the network device to send signaling to the terminal device in the prior art. For brevity, a detailed description of the specific process is omitted here.
  • step 330 the terminal device generates a PMI based on the feedback mode indicated by the sixth indication information.
  • the terminal device may generate the PMI based on the feedback mode indicated by the sixth indication information.
  • the specific process of generating the PMI by the terminal device may be the same as that in the prior art, and for the sake of brevity, details are not described here.
  • the terminal device When the terminal device generates the PMI based on the second feedback mode, its specific implementation process has been described in detail in the method 200 above, and for the sake of brevity, it will not be repeated here.
  • step 340 the terminal device sends PMI. Accordingly, the network device receives the PMI.
  • the terminal device may send the PMI to the network device, so that the network device determines the precoding matrix.
  • the network device may be the network device that sent the sixth indication information above, or may be another network device, which is not limited in this application. It should be understood that the steps of the terminal device shown in the figure to send the PMI to the network device are only schematic, and should not constitute any limitation to this application.
  • the network device can determine the precoding matrix according to the PMI, and then determine the precoding matrix used for data transmission.
  • the network device can determine the precoding matrix according to the PMI based on different feedback modes.
  • the specific process for the network device to determine the precoding matrix according to the PMI may be the same as that in the prior art, and for the sake of brevity, details are not described here.
  • the terminal device When the terminal device generates the PMI based on the second feedback mode, the specific process for the network device to determine the precoding matrix according to the PMI has been described in detail in the method 200 above, and for the sake of brevity, no more details are provided here.
  • the terminal device may feedback the PMI using a corresponding feedback mode based on the instruction of the network device.
  • the terminal device may feedback the PMI using a corresponding feedback mode based on the instruction of the network device.
  • the size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application .
  • the communication device 1000 may include a communication unit 1100 and a processing unit 1200.
  • the communication device 1000 may correspond to the terminal device in the foregoing method embodiment, for example, it may be a terminal device, or a chip configured in the terminal device.
  • the communication device 1000 may correspond to the terminal device in the method 200 or the method 300 according to an embodiment of the present application, and the communication device 1000 may include the method 200 in FIG. 2 or the method 300 in FIG. 3. A unit of a method performed by a terminal device.
  • each unit in the communication device 1000 and the other operations and / or functions described above are for implementing the corresponding processes of the method 200 in FIG. 2 or the method 300 in FIG. 3, respectively.
  • the communication unit 1100 can be used to perform step 220 in the method 200
  • the processing unit 1200 can be used to perform step 210 in the method 200.
  • the communication unit 1100 is further configured to receive second indication information, where the second indication information is used to indicate the value of one or more of M 1 , L 1 and K 1 .
  • the communication unit 1100 is further configured to receive third indication information, where the third indication information is used to indicate the value of T 1 .
  • the M 1 space-frequency component matrices are selected from a space-frequency component matrix set or a subset of the space-frequency component matrix set, and the space-frequency component matrix is composed of each beam vector in the beam vector set and the frequency domain vector set
  • Each frequency domain vector in is determined, and each space frequency component matrix in the space frequency component matrix set is uniquely determined by a beam vector in the beam vector set and a frequency domain vector in the frequency domain vector set;
  • the first indication information includes position information of the M 1 space frequency component matrices in the space frequency component matrix set, or position information of the M 1 space frequency component matrices in a subset of the space frequency component matrix set.
  • each space frequency component matrix in the M 1 space frequency component matrices is converted by the conjugate of a beam vector in the L 1 beam vectors and a frequency domain vector in the K 1 frequency domain vectors The product is determined.
  • each space-frequency component matrix of the M 1 space-frequency component matrices is composed of a Krone of a frequency domain vector of the K 1 frequency-domain vectors and a beam vector of the L 1 beam-vectors Clear product determination.
  • the weighted sum of the T 1 space frequency component matrices is used to determine the precoding vectors of one or more frequency domain units on the first transmission layer.
  • L 1 L 2
  • K 1 K 2
  • T 1 T 2 .
  • the communication unit 1100 may be used to perform steps 320 and 340 in the method 300, and the processing unit 1200 may be used to perform step 330 in the method 300.
  • the communication unit 1100 in the communication device 1000 may correspond to the transceiver 2020 in the terminal device 2000 shown in FIG. 5, and the processing unit 1200 in the communication device 1000 may It corresponds to the processor 2010 in the terminal device 2000 shown in FIG. 5.
  • the communication unit 1100 in the communication device 1000 may be an input / output interface.
  • the communication device 1000 may correspond to the network device in the foregoing method embodiment, for example, it may be a network device, or a chip configured in the network device.
  • the communication device 1000 may correspond to the network device in the method 200 or the method 300 according to an embodiment of the present application.
  • the communication device 1000 may include the method 200 in FIG. 2 or the method 300 in FIG. 3. A unit of a method performed by a network device.
  • each unit in the communication device 1000 and the other operations and / or functions described above are for implementing the corresponding processes of the method 200 in FIG. 2 or the method 300 in FIG. 3, respectively.
  • the communication unit 1100 can be used to perform step 220 in the method 200 and the processing unit 1200 can be used to perform step 230 in the method 200.
  • the processing unit 1200 may be used to determine precoding vectors of one or more frequency domain units based on the first indication information.
  • the communication unit 1100 is further configured to send second indication information, where the second indication information is used to indicate the value of one or more of M 1 , L 1 and K 1 .
  • the communication unit 1100 is further configured to send third indication information, where the third indication information is used to indicate the value of T 1 .
  • the M 1 space-frequency component matrices are selected from a space-frequency component matrix set or a subset of the space-frequency component matrix set, and the space-frequency component matrix is composed of each beam vector in the beam vector set and the frequency domain vector set
  • Each frequency domain vector in is determined, and each space frequency component matrix in the space frequency component matrix set is uniquely determined by a beam vector in the beam vector set and a frequency domain vector in the frequency domain vector set;
  • the first indication information includes position information of the M 1 space frequency component matrices in the space frequency component matrix set, or position information of the M 1 space frequency component matrices in a subset of the space frequency component matrix set.
  • each space frequency component matrix in the M 1 space frequency component matrices is converted by the conjugate of a beam vector in the L 1 beam vectors and a frequency domain vector in the K 1 frequency domain vectors The product is determined.
  • each space-frequency component matrix of the M 1 space-frequency component matrices is composed of a Krone of a frequency domain vector of the K 1 frequency-domain vectors and a beam vector of the L 1 beam-vectors Clear product determination.
  • the weighted sum of the T 1 space frequency component matrices is used to determine the precoding vectors of one or more frequency domain units on the first transmission layer.
  • L 1 L 2
  • K 1 K 2
  • T 1 T 2 .
  • the communication unit 1100 may be used to perform steps 320 and 340 in the method 300, and the processing unit 1200 may be used to perform step 310 in the method 300.
  • the communication unit in the communication device 1000 may correspond to the transceiver 3200 in the network device 3000 shown in FIG. 6, and the processing unit 1200 in the communication device 1000 may This corresponds to the processor 3100 in the network device 3000 shown in FIG. 6.
  • the communication unit 1100 in the communication device 1000 may be an input / output interface.
  • FIG. 5 is a schematic structural diagram of a terminal device 2000 provided by an embodiment of the present application.
  • the terminal device 2000 can be applied to the system shown in FIG. 1 to perform the functions of the terminal device in the above method embodiments.
  • the terminal device 2000 includes a processor 2010 and a transceiver 2020.
  • the terminal device 2000 further includes a memory 2030.
  • the processor 2010, the transceiver 2002 and the memory 2030 can communicate with each other through an internal connection path to transfer control and / or data signals.
  • the memory 2030 is used to store a computer program, and the processor 2010 is used from the memory 2030 Call and run the computer program to control the transceiver 2020 to send and receive signals.
  • the terminal device 2000 may further include an antenna 2040 for sending uplink data or uplink control signaling output by the transceiver 2020 through a wireless signal.
  • the processor 2010 and the memory 2030 may be combined into a processing device.
  • the processor 2010 is used to execute the program code stored in the memory 2030 to implement the above-mentioned functions.
  • the memory 2030 may also be integrated in the processor 2010 or independent of the processor 2010.
  • the processor 2010 may correspond to the processing unit in FIG. 4.
  • the above-mentioned transceiver 2020 may correspond to the communication unit in FIG. 4 and may also be called a transceiver unit.
  • the transceiver 2020 may include a receiver (or receiver, receiving circuit) and a transmitter (or transmitter, transmitting circuit). Among them, the receiver is used to receive signals, and the transmitter is used to transmit signals.
  • the terminal device 2000 shown in FIG. 5 can implement various processes involving the terminal device in the method embodiment shown in FIG. 2.
  • the operations and / or functions of each module in the terminal device 2000 are respectively to implement the corresponding processes in the above method embodiments.
  • the above-mentioned processor 2010 may be used to perform the actions described in the foregoing method embodiments that are internally implemented by the terminal device, and the transceiver 2020 may be used to perform the operations described in the foregoing method embodiments by the terminal device to or from the network device. action.
  • the transceiver 2020 may be used to perform the operations described in the foregoing method embodiments by the terminal device to or from the network device. action.
  • the terminal device 2000 may further include a power supply 2050, which is used to provide power to various devices or circuits in the terminal device.
  • a power supply 2050 which is used to provide power to various devices or circuits in the terminal device.
  • the terminal device 2000 may further include one or more of an input unit 2060, a display unit 2070, an audio circuit 2080, a camera 2090, a sensor 2100, etc. It may also include a speaker 2082, a microphone 2084, and so on.
  • FIG. 6 is a schematic structural diagram of a network device provided by an embodiment of the present application, for example, may be a schematic structural diagram of a base station.
  • the base station 3000 can be applied to the system shown in FIG. 1 to perform the functions of the network device in the above method embodiments.
  • the base station 3000 may include one or more radio frequency units, such as a remote radio unit (RRU) 3100 and one or more baseband units (BBU) (also called a distributed unit (DU) )) 3200.
  • RRU 3100 may be referred to as a transceiver unit, which corresponds to the communication unit 1200 in FIG. 4.
  • the transceiver unit 3100 may also be called a transceiver, a transceiver circuit, or a transceiver, etc., which may include at least one antenna 3101 and a radio frequency unit 3102.
  • the transceiving unit 3100 may include a receiving unit and a transmitting unit, the receiving unit may correspond to a receiver (or receiver, receiving circuit), and the transmitting unit may correspond to a transmitter (or transmitter, transmitting circuit).
  • the RRU 3100 part is mainly used for the transmission and reception of radio frequency signals and the conversion of radio frequency signals and baseband signals, for example, for sending instruction information to terminal devices.
  • the BBU 3200 part is mainly used for baseband processing and controlling the base station.
  • the RRU 3100 and the BBU 3200 may be physically arranged together, or may be physically separated, that is, distributed base stations.
  • the BBU 3200 is the control center of the base station, and may also be referred to as a processing unit, which may correspond to the processing unit 1100 in FIG. 4 and is mainly used to complete baseband processing functions, such as channel coding, multiplexing, modulation, spread spectrum, and so on.
  • the BBU processing unit
  • the BBU may be used to control the base station to perform the operation flow on the network device in the above method embodiment, for example, to generate the above instruction information.
  • the BBU 3200 may be composed of one or more boards, and multiple boards may jointly support a wireless access network (such as an LTE network) of a single access standard, or may support different access standards respectively. Wireless access network (such as LTE network, 5G network or other networks).
  • the BBU 3200 also includes a memory 3201 and a processor 3202.
  • the memory 3201 is used to store necessary instructions and data.
  • the processor 3202 is used to control the base station to perform necessary actions, for example, to control the base station to execute the operation flow of the network device in the foregoing method embodiment.
  • the memory 3201 and the processor 3202 may serve one or more single boards. In other words, the memory and processor can be set separately on each board. It is also possible that multiple boards share the same memory and processor. In addition, each board can also be provided with necessary circuits.
  • the base station 3000 shown in FIG. 6 can implement various processes involving network devices in the method embodiment of FIG. 2.
  • the operations and / or functions of each module in the base station 3000 are to implement the corresponding processes in the above method embodiments.
  • the above-mentioned BBU 3200 can be used to perform the actions described in the foregoing method embodiments that are internally implemented by the network device, and the RRU 3100 can be used to perform the actions described in the previous method embodiments that the network device sends to or receives from the terminal device.
  • the RRU 3100 can be used to perform the actions described in the previous method embodiments that the network device sends to or receives from the terminal device.
  • An embodiment of the present application further provides a processing device, including a processor and an interface; the processor is used to execute the method of any of the foregoing method embodiments.
  • the above processing device may be one or more chips.
  • the processing device may be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or a system chip (SoC), or It is a central processor (CPU), it can also be a network processor (NP), it can also be a digital signal processing circuit (digital signal processor, DSP), or a microcontroller (micro controller) , MCU), can also be a programmable controller (programmable logic device, PLD) or other integrated chips.
  • FPGA field programmable gate array
  • ASIC application specific integrated circuit
  • SoC system chip
  • CPU central processor
  • NP network processor
  • DSP digital signal processor
  • microcontroller micro controller
  • MCU microcontroller
  • PLD programmable logic device
  • each step of the above method may be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software.
  • the steps of the method disclosed in conjunction with the embodiments of the present application may be directly embodied and executed by a hardware processor, or may be executed and completed by a combination of hardware and software modules in the processor.
  • the software module may be located in a mature storage medium in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, and a register.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. In order to avoid repetition, they are not described in detail here.
  • the processor in the embodiment of the present application may be an integrated circuit chip with signal processing capabilities.
  • each step of the foregoing method embodiment may be completed by an integrated logic circuit of hardware in a processor or instructions in the form of software.
  • the aforementioned processor may be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components .
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present application may be implemented or executed.
  • the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the steps of the method disclosed in conjunction with the embodiments of the present application may be directly embodied and executed by a hardware decoding processor, or may be executed and completed by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in a mature storage medium in the art, such as a random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, and register.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
  • the memory in the embodiments of the present application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (read-only memory, ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), electronically Erase programmable EPROM (EEPROM) or flash memory.
  • the volatile memory may be a random access memory (random access memory, RAM), which is used as an external cache.
  • RAM random access memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate synchronous dynamic random access memory double data SDRAM, DDR SDRAM
  • enhanced synchronous dynamic random access memory enhanced SDRAM, ESDRAM
  • serial link DRAM SLDRAM
  • direct RAMbus RAM direct RAMbus RAM
  • the present application also provides a computer program product, the computer program product includes: computer program code, when the computer program code runs on the computer, the computer is caused to execute the embodiment shown in FIG. 2 The method of any one of the embodiments.
  • the present application also provides a computer-readable medium that stores program code, and when the program code runs on a computer, the computer is caused to execute the embodiment shown in FIG. 2 The method of any one of the embodiments.
  • the present application further provides a system, which includes the foregoing one or more terminal devices and one or more network devices.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be from a website site, computer, server or data center Transmission to another website, computer, server or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device including a server, a data center, and the like integrated with one or more available media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, or a magnetic tape), an optical medium (for example, a high-density digital video disc (DVD)), or a semiconductor medium (for example, a solid state disk, SSD)) etc.
  • a magnetic medium for example, a floppy disk, a hard disk, or a magnetic tape
  • an optical medium for example, a high-density digital video disc (DVD)
  • DVD high-density digital video disc
  • SSD solid state disk
  • the network device in each of the above device embodiments corresponds exactly to the network device or terminal device in the terminal device and method embodiments, and the corresponding steps are performed by the corresponding modules or units, for example, the communication unit (transceiver) performs the receiving or The steps of sending, other than sending and receiving, can be executed by the processing unit (processor).
  • the function of the specific unit can refer to the corresponding method embodiment. There may be one or more processors.
  • a component may be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer.
  • the application running on the computing device and the computing device can be components.
  • One or more components can reside in a process and / or thread of execution, and a component can be localized on one computer and / or distributed between 2 or more computers.
  • these components can execute from various computer readable media having various data structures stored thereon.
  • the component may, for example, be based on a signal having one or more data packets (eg, data from two components that interact with another component between the local system, the distributed system, and / or the network, such as the Internet that interacts with other systems through signals) Communicate through local and / or remote processes.
  • data packets eg, data from two components that interact with another component between the local system, the distributed system, and / or the network, such as the Internet that interacts with other systems through signals
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the units is only a division of logical functions.
  • there may be other divisions for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
  • each functional unit may be implemented in whole or in part by software, hardware, firmware, or any combination thereof.
  • software When implemented using software, it can be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be from a website site, computer, server or data center Transmit to another website, computer, server or data center by wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device including a server, a data center, and the like integrated with one or more available media.
  • the usable medium may be a magnetic medium (eg, floppy disk, hard disk, magnetic tape), optical medium (eg, DVD), or semiconductor medium (eg, solid state disk (SSD)), or the like.
  • the function is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.
  • the technical solution of the present application essentially or part of the contribution to the existing technology or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to enable a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Radio Transmission System (AREA)

Abstract

本申请提供了一种指示和确定预编码向量的方法和通信装置，以减少反馈开销。该方法包括：终端设备生成并发送第一指示信息，网络设备根据该第一指示信息确定一个或多个频域单元的预编码向量。该第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量及T 1个空频分量矩阵。该T 1个空频分量矩阵的加权和用于确定各频域单元的预编码向量。其中，T 1个空频分量矩阵选自与L 1个波束向量和K 1个频域向量对应的M 1个空频分量矩阵，每个空频分量矩阵由一个波束向量和一个频域向量唯一确定，M 1＝L 1×K 1。该L 1个波束向量为波束向量集合中的部分波束向量，和/或，该K 1个频域向量为频域向量集合中的部分频域向量。

Description

指示和确定预编码向量的方法以及通信装置
本申请要求于2018年10月16日提交中国专利局、申请号为201811205381.1、申请名称为“指示和确定预编码向量的方法以及通信装置”的中国专利申请的优先权，以及于2018年10月30日提交中国专利局、申请号为201811281059.7、申请名称为“指示和确定预编码向量的方法以及通信装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及无线通信领域，并且更具体地，涉及指示和确定预编码向量的方法以及通信装置。
背景技术
在大规模多输入多输出(massive multiple-input multiple-output，Massive MIMO)技术中，网络设备可通过预编码减小多用户之间的干扰以及同一用户的多个信号流之间的干扰，有利于提高信号质量，实现空分复用，提高频谱利用率。
终端设备例如可以通过信道测量等方式确定预编码向量，并希望通过反馈，使得网络设备获得与终端设备确定的预编码向量相同或者相近的预编码向量。在一种实现方式中，终端设备可以通过宽带反馈和子带反馈的两级反馈方式来向网络设备指示预编码向量。具体地，终端设备可以基于每个传输层，通过宽带反馈指示被选择的波束向量以及各波束向量的宽带幅度系数的量化值，并可以通过子带反馈指示可用于各个子带的组合系数的量化值，其中组合系数例如可包括子带幅度系数和子带相位系数。网络设备可以综合宽带反馈的信息和子带反馈的信息恢复出各子带对应的预编码向量。终端设备通过宽带反馈和子带反馈的两级反馈方式来向网络设备指示预编码向量的具体方法可以参考新无线(new radio，NR)协议TS 38.214中定义的类型II(type II)码本反馈方式。
然而，随着传输层数的增加，上述反馈模式所带来的反馈开销会成倍增加。而子带数量越多，反馈开销增加的幅度也越大。
发明内容
本申请提供一种指示和确定预编码向量的方法以及通信装置，以期减小反馈开销。
第一方面，提供了一种指示预编码向量的方法，该方法可以由终端设备执行，或者，也可以由配置于终端设备中的芯片执行。
具体地，该方法包括：生成第一指示信息；发送该第一指示信息。该第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，该T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，该L 1个波束向量和该K 1个频域向量与M 1个空频分量矩阵对应，该T 1个空频分量 矩阵是该M 1个空频分量矩阵中的部分空频分量矩阵，该M 1个空频分量矩阵中的每个空频分量矩阵由L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；该L 1个波束向量为波束向量集合中的部分波束向量，和/或，该K 1个频域向量为频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数。
第二方面，提供了一种确定预编码向量的方法，该方法可以由网络设备执行，或者，也可以由配置于网络设备中的芯片执行。
具体地，该方法包括：接收第一指示信息。该第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，该T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，该L 1个波束向量和该K 1个频域向量与M 1个空频分量矩阵对应，该T 1个空频分量矩阵是该M 1个空频分量矩阵中的部分空频分量矩阵，该M 1个空频分量矩阵中的每个空频分量矩阵由L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；该L 1个波束向量为波束向量集合中的部分波束向量，和/或，该K 1个频域向量为频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；基于该第一指示信息，确定一个或多个频域单元的预编码向量。
在一种实现方式中，该T 1个空频分量矩阵选自M 1个空频分量矩阵，该M 1个空频分量矩阵由L 1个波束向量和K 1个频域向量确定。在另一种实现方式中，该T 1个空频分量矩阵由M 1个空频向量对中的T 1个空频向量对确定，该M 1个空频向量对由L 1个波束向量和K 1个频域向量组合得到，每个空频向量对由L 1个波束向量中的一个波束向量和K 1个频域向量中的一个频域向量唯一确定。在又一种实现方式中，该T 1个空频分量矩阵可以表现为T 1个波束向量和T 1个频域向量组合而成的T 1个空频向量对，该T 1个空频向量对选自由L 1个波束向量和K 1个频域向量而成的M 1个空频向量对。
基于上述技术方案，终端设备通过向网络设备指示少量的波束向量、频域向量和空频分量矩阵，以便于网络设备恢复预编码向量。其中，频域向量可用于描述信道在频域上不同的变化规律。终端设备可以通过一个或多个频域向量的线性叠加来模拟信道在频域上的变化，充分挖掘了频域单元之间的关系，利用了频域的连续性，用较少的频域向量来描述多个频域单元的变化规律。相比于现有技术而言，无需基于每个频域单元独立地上报加权系数，在频域单元数量增加的情况下，并不会造成反馈开销的成倍增加。因此，可以在保证反馈精度的基础上大大减小反馈开销。
然而，由于波束向量集合中可能包含较多的波束向量，频域向量集合中也可能包含较多的频域向量，如果直接从波束向量集合和频域向量集合中指示较少的波束向量和频域向量，可能会带来较大的比特开销，或者，需要终端设备和网络设备预先定义大量的波束向量组合与索引的对应关系以及频域向量组合与索引的对应关系。
而本申请实施例中，终端设备将用作加权求和的空频分量矩阵的选择范围缩小至由L 1个波束向量和K 1个频域向量构建的M 1个空频分量矩阵的范围中，也就是从已有的向量集合中先选择一个较小的范围，再进一步从该范围中选择和指示T 1个空频分量矩阵。一方面可以避免直接指示该T 1个空频分量矩阵带来的较大的反馈开销，另一方面可以避免在终端设备和网络设备两端保存大量的对应关系。
需要说明的是，该T 1个波束向量是L 1个波束向量中选择的部分波束向量，但并不代 表T 1一定小于L 1，该T 1个波束向量中可能有部分或全部的波束向量是重复使用的。因此用于组合得到T 1个波束向量对的波束向量的个数为T 1。同理，该T 1个频域向量是K 1个波束向量中选择的部分波束向量，但这并不代表T 1一定小于K 1，该T 1个频域向量中可能有部分或全部的频域向量是重复使用的。因此用于组合得到T 1个频域向量对的频域向量的个数为T 1。下文中为了简洁，省略对相同或相似情况的说明。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：接收第二指示信息，该第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
相应地，结合第二方面，在第二方面的某些实现方式中，该方法还包括：发送第二指示信息，该第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
即，M 1、L 1和K 1中一项或多项的取值可以由网络设备指示。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：发送第二指示信息，该第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
相应地，结合第二方面，在第二方面的某些实现方式中，该方法还包括：接收第二指示信息，该第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
即，M 1、L 1和K 1中一项或多项的取值可以由终端设备确定并上报网络设备。
应理解，M 1、L 1和K 1中一项或多项的取值还可以预先定义，如协议定义，本申请对此不作限定。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：接收第三指示信息，该第三指示信息用于指示T 1的取值。
即，T 1的取值可以由网络设备指示。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：发送第三指示信息，该第三指示信息用于指示T 1的取值。
即，T 1的取值可以由终端设备确定并上报网络设备。
应理解，T 1的取值还可以预先定义，如协议定义，本申请对此不作限定。
结合第一方面或第二方面，在某些实现方式中，该第一指示信息包括该L 1个波束向量在该波束向量集合中的位置信息和该K 1个频域向量在该频域向量集合中的位置信息。
也就是说，由于该L 1个波束向量和K 1个频域向量与M 1个空频分量矩阵对应，通过指示该L 1个波束向量和K 1个频域向量，便可以确定该M 1个空频分量矩阵。换句话说，该M 1个空频分量矩阵可以通过二维索引指示。
结合第一方面或第二方面，在某些实现方式中，该M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，该空频分量矩阵由该波束向量集合中的各波束向量和该频域向量集合中的各频域向量确定，且该空频分量矩阵集合中的每个空频分量矩阵由该波束向量集合中的一个波束向量和该频域向量集合中的一个频域向量唯一确定。其中，该第一指示信息包括该M 1个空频分量矩阵在该空频分量矩阵集合中的位置信息，或，该M 1个空频分量矩阵在空频分量矩阵集合的子集中的位置信息。
也就是说，该M 1个空频分量矩阵可以通过一维索引指示。
需要说明的是，文中仅为便于理解而引入M 1个空频分量矩阵的概念，这并不代表终端设备一定会生成该M 1个空频分量矩阵，终端设备也可以由该L 1个波束向量和K 1个频域向量组合得到M 1个空频向量对。但可以理解的是，由该L 1个波束向量和K 1个频域向 量，或者，由该M 1个空频向量对，可以构建得到该M 1个空频分量矩阵。或者说，M 1个空频向量对与M 1个空频分量矩阵可以相互转换。因此可以认为该M 1个空频分量矩阵是与L 1个波束向量和K 1个频域向量对应。
可选地，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量的共轭转置的乘积确定。
可选地，该M 1个空频分量矩阵中的每个空频分量矩阵由该K 1个频域向量中的一个频域向量和该L 1个波束向量中的一个波束向量的克罗内克积确定。
本实施例中，为方便理解，分别以波束向量和频域向量均为列向量为例来说明M 1个空频分量矩阵与L 1个波束向量和K 1个频域向量的关系，但这不应对本申请构成任何限定。例如，频域向量也可以为行向量，此时，每个空频分量矩阵可以由一个波束向量和一个频域向量的乘积确定。又例如，每个空频分量矩阵可以由一个波束向量和一个频域向量的克罗内克积确定。本申请对此不作限定。
结合第一方面或第二方面，在某些实现方式中，该第一指示信息可以通过以下任意一种方式来指示T 1个空频分量矩阵(或者说T 1个空频向量对)：
方式1、通过位图(bitmap)来指示M 1个空频分量矩阵中的T 1个空频分量矩阵；或
方式2、指示T 1个空频分量矩阵的组合在M 1个空频分量矩阵中的索引；或
方式3、指示T 1个空频分量矩阵中每个空频分量矩阵所对应的波束向量在L 1个波束向量中的位置和频域向量在K 1个频域向量中的位置；或
方式4、指示T 1个空频分量矩阵中每个空频分量矩阵在M 1个空频分量矩阵中的索引。
通过在M 1个空频分量矩阵中指示T 1个空频分量矩阵，可以将T 1个空频分量矩阵的选择范围缩小，也就可以减少该T 1个空频分量矩阵的反馈开销。
结合第一方面或第二方面，在某些实现方式中，该T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
其中，该第一传输层可以是一个传输层，也可以是多个传输层。
结合第一方面，在某些可能的实现方式中，该方法还包括：生成第四指示信息，该第四指示信息用于指示该波束向量集合中的L 2个波束向量、该频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，该T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，该L 2个波束向量和该K 2个频域向量与M 2个空频分量矩阵对应，该T 2个空频分量矩阵是该M 2个空频分量矩阵中的部分空频分量矩阵，该M 2个空频分量矩阵中的每个空频分量矩阵由该L 2个波束向量中的一个波束向量和该K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；该L 2个波束向量为该波束向量集合中的部分波束向量，和/或，该K 2个频域向量为该频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；发送该第四指示信息。
结合第二方面，在某些可能的实现方式中，该方法还包括：接收第四指示信息，该第四指示信息用于指示该波束向量集合中的L 2个波束向量和T 2个空频分量矩阵，该T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，该L 2个波束向量和该K 2个频域向量与M 2个空频分量矩阵对应，该T 2个空频分量矩阵是该M 2个空频分量矩阵中的部分空频分量矩阵，该M 2个空频分量矩阵中的每个空频分量矩阵由该L 2个波束向量中的一个波束向量和K 2个频域向量中的一个频域向量唯一确定，且 M 2＝L 2×K 2；该L 2个波束向量为该波束向量集合中的部分波束向量，和/或，该K 2个频域向量为该频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；基于该第四指示信息，确定该第二传输层上一个或多个频域单元的预编码向量。
其中，该第二传输层可以是多个传输层中除第一传输层之外的一个或多个传输层。
基于上述第四指示信息，终端设备可以向网络设备指示用于确定第二传输层上一个或多个频域单元的预编码向量。
结合第一方面或第二方面，在某些实现方式中，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
结合第一方面或第二方面，在某些实现方式中，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
第三方面，提供了一种通信装置，包括用于执行第一方面任一种可能实现方式中的方法的各个模块或单元。
第四方面，提供了一种通信装置，包括处理器。该处理器与存储器耦合，可用于执行存储器中的指令，以实现上述第一方面任一种可能实现方式中的方法。可选地，该通信装置还包括存储器。可选地，该通信装置还包括通信接口，处理器与通信接口耦合。
在一种实现方式中，该通信装置为终端设备。当该通信装置为终端设备时，所述通信接口可以是收发器，或，输入/输出接口。
在另一种实现方式中，该通信装置为配置于终端设备中的芯片。当该通信装置为配置于终端设备中的芯片时，所述通信接口可以是输入/输出接口。
可选地，所述收发器可以为收发电路。可选地，所述输入/输出接口可以为输入/输出电路。
第五方面，提供了一种通信装置，包括用于执行第二方面任一种可能实现方式中的方法的各个模块或单元。
第六方面，提供了一种通信装置，包括处理器。该处理器与存储器耦合，可用于执行存储器中的指令，以实现上述第二方面任一种可能实现方式中的方法。可选地，该通信装置还包括存储器。可选地，该通信装置还包括通信接口，处理器与通信接口耦合。
在一种实现方式中，该通信装置为网络设备。当该通信装置为网络设备时，所述通信接口可以是收发器，或，输入/输出接口。
在另一种实现方式中，该通信装置为配置于网络设备中的芯片。当该通信装置为配置于网络设备中的芯片时，所述通信接口可以是输入/输出接口。
可选地，所述收发器可以为收发电路。可选地，所述输入/输出接口可以为输入/输出电路。
第七方面，提供了一种处理器，包括：输入电路、输出电路和处理电路。所述处理电路用于通过所述输入电路接收输入信号，并通过所述输出电路输出信号，使得所述处理器执行第一方面或第二方面以及第一方面或第二方面任一种可能实现方式中的方法。
在具体实现过程中，上述处理器可以为芯片，输入电路可以为输入管脚，输出电路可以为输出管脚，处理电路可以为晶体管、门电路、触发器和各种逻辑电路等。输入电路所接收的输入的信号可以是由例如但不限于接收器接收并输入的，输出电路所输出的信号可以是例如但不限于输出给发射器并由发射器发射的，且输入电路和输出电路可以是同一电路，该电路在不同的时刻分别用作输入电路和输出电路。本申请实施例对处理器及各种电路的具体实现方式不做限定。
第八方面，提供了一种处理装置，包括处理器和存储器。该处理器用于读取存储器中存储的指令，并可通过接收器接收信号，通过发射器发射信号，以执行第一方面或第二方面以及第一方面或第二方面任一种可能实现方式中的方法。
可选地，所述处理器为一个或多个，所述存储器为一个或多个。
可选地，所述存储器可以与所述处理器集成在一起，或者所述存储器与处理器分离设置。
在具体实现过程中，存储器可以为非瞬时性(non-transitory)存储器，例如只读存储器(read only memory，ROM)，其可以与处理器集成在同一块芯片上，也可以分别设置在不同的芯片上，本申请实施例对存储器的类型以及存储器与处理器的设置方式不做限定。
应理解，相关的数据交互过程例如发送指示信息可以为从处理器输出指示信息的过程，接收能力信息可以为处理器接收输入能力信息的过程。具体地，处理输出的数据可以输出给发射器，处理器接收的输入数据可以来自接收器。其中，发射器和接收器可以统称为收发器。
上述第八方面中的处理装置可以是一个或多个芯片，该处理器可以通过硬件来实现也可以通过软件来实现，当通过硬件实现时，该处理器可以是逻辑电路、集成电路等；当通过软件来实现时，该处理器可以是一个通用处理器，通过读取存储器中存储的软件代码来实现，该存储器可以集成在处理器中，可以位于该处理器之外，独立存在。
第九方面，提供了一种计算机程序产品，所述计算机程序产品包括：计算机程序(也可以称为代码，或指令)，当所述计算机程序被运行时，使得计算机执行上述第一方面或第二方面以及第一方面或第二方面中任一种可能实现方式中的方法。
第十方面，提供了一种计算机可读介质，所述计算机可读介质存储有计算机程序(也可以称为代码，或指令)当其在计算机上运行时，使得计算机执行上述第一方面或第二方面以及第一方面或第二方面中任一种可能实现方式中的方法。
第十一方面，提供了一种通信系统，包括前述的网络设备和终端设备。
附图说明
图1是适用于本申请实施例提供的指示和确定预编码向量的方法的通信系统的示意图；
图2是本申请实施例提供的指示和确定预编码向量的方法的示意性流程图；
图3是本申请另一实施例提供的反馈预编码矩阵指示(precoding matrix indicator，PMI)的方法的示意性流程图；
图4是本申请实施例提供的通信装置的示意性框图；
图5是本申请实施例提供的终端设备的结构示意图；
图6是本申请实施例提供的网络设备的结构示意图。
具体实施方式
下面将结合附图，对本申请中的技术方案进行描述。
本申请实施例的技术方案可以应用于各种通信系统，例如：全球移动通信(global  system for mobile communications，GSM)系统、码分多址(code division multiple access，CDMA)系统、宽带码分多址(wideband code division multiple access，WCDMA)系统、通用分组无线业务(general packet radio service，GPRS)、长期演进(long term evolution，LTE)系统、LTE频分双工(frequency division duplex，FDD)系统、LTE时分双工(time division duplex，TDD)、通用移动通信系统(universal mobile telecommunication system，UMTS)、全球互联微波接入(worldwide interoperability for microwave access，WiMAX)通信系统、未来的第五代(5th generation，5G)系统或新无线(new radio，NR)等。
为便于理解本申请实施例，首先以图1中示出的通信系统为例详细说明适用于本申请实施例的通信系统。图1是适用于本申请实施例的指示预编码向量的方法的通信系统100的示意图。如图1所示，该通信系统100可以包括至少一个网络设备，例如图1所示的网络设备110；该通信系统100还可以包括至少一个终端设备，例如图1所示的终端设备120。网络设备110与终端设备120可通过无线链路通信。各通信设备，如网络设备110或终端设备120，均可以配置多个天线。对于该通信系统100中的每一个通信设备而言，所配置的多个天线可以包括至少一个用于发送信号的发射天线和至少一个用于接收信号的接收天线。因此，该通信系统100中的各通信设备之间，如网络设备110与终端设备120之间，可通过多天线技术通信。
应理解，该通信系统中的网络设备可以是任意一种具有无线收发功能的设备。该网络设备包括但不限于：演进型节点B(evolved Node B，eNB)、无线网络控制器(radio network controller，RNC)、节点B(Node B，NB)、基站控制器(base station controller，BSC)、基站收发台(base transceiver station，BTS)、家庭基站(例如，home evolved NodeB，或home Node B，HNB)、基带单元(baseband unit，BBU)，无线保真(wireless fidelity，WiFi)系统中的接入点(access point，AP)、无线中继节点、无线回传节点、传输点(transmission point，TP)或者发送接收点(transmission and reception point，TRP)等，还可以为5G，如，NR，系统中的gNB，或，传输点(TRP或TP)，5G系统中的基站的一个或一组(包括多个天线面板)天线面板，或者，还可以为构成gNB或传输点的网络节点，如基带单元(BBU)，或，分布式单元(distributed unit，DU)等。
在一些部署中，gNB可以包括集中式单元(centralized unit，CU)和DU。gNB还可以包括射频单元(radio unit，RU)。CU实现gNB的部分功能，DU实现gNB的部分功能。比如，CU实现无线资源控制(radio resource control，RRC)，分组数据汇聚层协议(packet data convergence protocol，PDCP)层的功能，DU实现无线链路控制(radio link control，RLC)、媒体接入控制(media access control，MAC)和物理(physical，PHY)层的功能。由于RRC层的信息最终会变成PHY层的信息，或者，由PHY层的信息转变而来，因而，在这种架构下，高层信令，如RRC层信令，也可以认为是由DU发送的，或者，由DU+CU发送的。可以理解的是，网络设备可以为CU节点、或DU节点、或包括CU节点和DU节点的设备。此外，CU可以划分为接入网(radio access network，RAN)中的网络设备，也可以将CU划分为核心网(core network，CN)中的网络设备，本申请对此不做限定。
还应理解，该无线通信系统中的终端设备也可以称为用户设备(user equipment，UE)、接入终端、用户单元、用户站、移动站、移动台、远方站、远程终端、移动设备、用户终 端、终端、无线通信设备、用户代理或用户装置。本申请的实施例中的终端设备可以是手机(mobile phone)、平板电脑(pad)、带无线收发功能的电脑、虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、工业控制(industrial control)中的无线终端、无人驾驶(self driving)中的无线终端、远程医疗(remote medical)中的无线终端、智能电网(smart grid)中的无线终端、运输安全(transportation safety)中的无线终端、智慧城市(smart city)中的无线终端、智慧家庭(smart home)中的无线终端等等。本申请的实施例对应用场景不做限定。
还应理解，图1仅为便于理解而示例的简化示意图，该通信系统100中还可以包括其他网络设备或者还可以包括其他终端设备，图1中未予以画出。
为了便于理解本申请实施例，下面简单说明下行信号在发送之前在物理层的处理过程。应理解，下文所描述的对下行信号的处理过程可以由网络设备执行，也可以由配置于网络设备中的芯片执行。为方便说明，下文统称为网络设备。
网络设备在物理信道可对码字(code word)进行处理。其中，码字可以为经过编码(例如包括信道编码)的编码比特。码字经过加扰(scrambling)，生成加扰比特。加扰比特经过调制映射(modulation mapping)，得到调制符号。调制符号经过层映射(layer mapping)，被映射到多个层(layer)，或者称，传输层。经过层映射后的调制符号经过预编码(precoding)，得到预编码后的信号。预编码后的信号经过资源元素(resource element，RE)映射后，被映射到多个RE上。这些RE随后经过正交复用(orthogonal frequency division multiplexing，OFDM)调制后通过天线端口(antenna port)发射出去。
应理解，上文所描述的对下行信号的处理过程仅为示例性描述，不应对本申请构成任何限定。对下行信号的处理过程具体可以参考现有技术，为了简洁，这里省略对其具体过程的详细说明。
为了便于理解本申请实施例，下面先对本申请实施例中涉及的术语做简单说明。
1、预编码技术：网络设备可以在已知信道状态的情况下，借助与信道资源相匹配的预编码矩阵来对待发送信号进行处理，使得经过预编码的待发送信号与信道相适配，从而使得接收设备消除信道间影响的复杂度降低。因此，通过对待发送信号的预编码处理，接收信号质量(例如信号与干扰加噪声比(signal to interference plus noise ratio，SINR)等)得以提升。因此，采用预编码技术，可以实现发送设备与多个接收设备在相同的时频资源上传输，也就是实现了多用户多输入多输出(multiple user multiple input multiple output，MU-MIMO)。应注意，有关预编码技术的相关描述仅为便于理解而示例，并非用于限制本申请实施例的保护范围。在具体实现过程中，发送设备还可以通过其他方式进行预编码。例如，在无法获知信道信息(例如但不限于信道矩阵)的情况下，采用预先设置的预编码矩阵或者加权处理方式进行预编码等。为了简洁，其具体内容本文不再赘述。
2、预编码矩阵指示(precoding matrix indicator，PMI)：可用于指示预编码矩阵。其中，该预编码矩阵例如可以是终端设备基于各个子带的信道矩阵确定的预编码矩阵。该信道矩阵可以是终端设备通过信道估计等方式或者基于信道互易性确定。但应理解，终端设备确定预编码矩阵的具体方法并不限于上文所述，具体实现方式可参考现有技术，为了简洁，这里不再一一列举。
例如，预编码矩阵可以通过对信道矩阵或信道矩阵的协方差矩阵进行奇异值分解 (singular value decomposition，SVD)的方式获得，或者，也可以通过对信道矩阵的协方差矩阵进行特征值分解(eigenvalue decopomsition，EVD)的方式获得。本申请对此不作限定。应理解，上文中列举的预编码矩阵的确定方式仅为示例，不应对本申请构成任何限定。预编码矩阵的确定方式可以参考现有技术，为了简洁，这里省略对其具体过程的详细说明。
终端设备可以对每个子带的预编码矩阵进行量化，并可以通过PMI将量化值发送给网络设备，以便于网络设备根据PMI确定出于终端设备所确定的预编码矩阵相同或相近的预编码矩阵。例如，网络设备可以根据PMI直接确定每个子带的预编码矩阵，也可以根据PMI确定每个子带的预编码矩阵后通过进一步的处理，如将不同用户的预编码矩阵做正交化处理等，以确定最终使用的预编码矩阵。因此，网络设备能够确定出与每个子带的信道相适配的预编码矩阵来对待发送信号进行预编码处理。应理解，网络设备根据PMI确定在每个子带使用的预编码矩阵的具体方法可以参考现有技术，这里仅为便于理解而示例，不应对本申请构成任何限定。
综上，终端设备所确定的预编码矩阵可以理解为待反馈的预编码矩阵。终端设备可以通过PMI指示待反馈的预编码矩阵，以便于网络设备基于PMI恢复出该预编码矩阵。可以理解，网络设备基于PMI恢复出的预编码矩阵可以与上述待反馈的预编码矩阵相同或相近。
下文示出了秩(rank)为1时通过两级反馈的预编码矩阵的简单示例。
Figure PCTCN2019110342-appb-000001
其中，W表示一个传输层、一个子带、两个极化方向上待反馈的预编码矩阵。W 1可以通过宽带反馈，W 2可以通过子带反馈。v 0至v 3为W 1中包含的波束向量，该多个波束向量例如可通过该多个波束向量的组合的索引来指示。在上文中示出的预编码矩阵中，两个极化方向上的波束向量是相同的，均使用了波束向量v 0至v 3。a 0至a 7为W 1中包含的宽带幅度系数，可通过宽带幅度系数的量化值来指示。c 0至c 7为W 2中包含的子带系数，每个子带系数可以包括子带幅度系数和子带相位系数。如c 0至c 7可以分别包括子带幅度系数α 0至α 7以及子带相位系数
Figure PCTCN2019110342-appb-000002
至
Figure PCTCN2019110342-appb-000003
并可分别通过子带幅度系数α 0至α 7的量化值和子带相位系数
Figure PCTCN2019110342-appb-000004
至
Figure PCTCN2019110342-appb-000005
的量化值来指示。可以看到，该待反馈的预编码矩阵可以视为多个波束向量的加权和。
应理解，上文示出的预编码矩阵基于一个传输层的反馈得到，因此也可以称为预编码向量。当传输层数增加时，终端设备可以基于每个传输层分别反馈。由每个传输层反馈而得到的预编码向量可以构建得到一个子带的预编码矩阵。例如传输层数为4，该预编码矩 阵可以包括4个预编码向量，分别与4个传输层对应。
随着传输层数的增加，终端设备的反馈开销也会增加。例如传输层数为4时，a 0至a 7以及c 0至c 7的反馈开销最多将达到一个传输层时的4倍。也就是说，如果终端设备基于每个传输层进行如上所述的宽带反馈和子带反馈，则随着传输层数的增加，所带来的反馈开销会成倍增加。而子带数量越多，反馈开销增加的幅度也越大。因此，希望提供一种方法，能够降低PMI的反馈开销。
应理解，上文所列举的通过PMI反馈预编码矩阵的方式仅为示例，而不应对本申请构成任何限定。例如，终端设备也可以通过PMI来向网络设备反馈信道矩阵，网络设备可以根据PMI确定信道矩阵，进而确定预编码矩阵，本申请对此不作限定。
3、预编码向量：在本申请实施例中，预编码向量可以由预编码矩阵中的一个向量确定，如，列向量。换句话说，预编码矩阵可以包括一个或多个列向量，每个列向量可用于确定一个预编码向量。当预编码矩阵仅包括一个列向量时，该预编码矩阵也可以称为预编码向量。
预编码矩阵可以是由一个或多个传输层的预编码向量确定，预编码矩阵中的每个向量可以对应于一个传输层。假设预编码向量的维度可以为N 1×1，若传输层数为R(R为正整数)，则预编码矩阵的维度可以为N 1×R。其中，传输层数可以由秩指示(rank indicator，RI)来指示，N 1可以表示天线端口数，N 1为正整数。
当发射天线被配置多个极化方向时，预编码向量还可以是指预编码矩阵在一个传输层、一个极化方向上的分量。假设极化方向数为P(P为正整数)，一个极化方向上天线端口数为N 2，则与一个传输层对应的预编码向量的维度为(P×N 2)×1，则一个极化方向上的预编码向量的维度可以为N 2×1，N 2为正整数。
因此，预编码向量可以与一个传输层对应，也可以与一个传输层上的一个极化方向对应，还可以与其他参量对应。本申请对此不作限定。
4、天线端口：可简称端口。可以理解为被接收设备所识别的发射天线，或者在空间上可以区分的发射天线。针对每个虚拟天线可以配置一个天线端口，每个虚拟天线可以为多个物理天线的加权组合，每个天线端口可以与一个参考信号对应，因此，每个天线端口可以称为一个参考信号的端口，例如，CSI-RS端口、探测参考信号(sounding reference signal，SRS)端口等。
5、波束与波束向量：波束可以理解为在空间某一方向上形成的信号强度的分布。形成波束的技术可以是波束赋形(或者称，波束成形)技术或者其他技术。波束成形技术具体可以是数字波束赋形技术、模拟波束赋形技术以及混合数字/模拟波束赋形技术。在本申请实施例中，波束可以通过数字波束赋形技术形成。
波束向量可以与波束对应，可以是预编码矩阵中的预编码向量，也可以是波束赋形向量。波束向量中的各个元素可以表示各个天线端口的权重。各个天线端口经加权后的信号相互叠加，可以形成一个信号强度较强的区域。
在本申请实施例中，波束向量也可以称为空域向量。可选地，该波束向量的长度(或者说，维度)是一个极化方向上的天线端口数。
假设该波束向量的长度为N s。波束向量可以是维度为N s×1的列向量，也可以是维度为1×N s的行向量。本申请对此不做限定。
6、频域单元：频域资源的单位，可表示不同的频域资源粒度。频域单元例如可以包括但不限于，子带、资源块(resource block，RB)、子载波、资源块组(resource block group,RBG)、预编码资源块组(precoding resource block group，PRG)等。
7、频域向量：本申请实施例中提出的用于表示信道在频域的变化规律的向量。频域向量具体可用于表示各波束向量的加权系数在各个频域单元上的变化规律。这种变化规律可能与多径时延相关。由于信号在经过无线信道传输时，在不同的传播路径可能上存在不同的传输时延。因此可以通过不同的频域向量来表征不同传播路径上时延的变化规律。
频域向量的维度可以是需要进行CSI测量的频域单元的数量。由于在不同的时刻需要进行CSI测量的频域单元的数量可能不同，因此，频域向量的维度也可能会变化。换句话说，频域向量的维度是可变的。
可选地，该频域向量的长度(或者说，维度)是CSI测量资源的频域占用带宽中包含的频域单元数量。
其中，CSI测量资源的频域占用带宽CSI测量资源的频域占用带宽可以是用于传输参考信号的带宽，这里所说的参考信号可以为用作信道测量的参考信号，如CSI-RS。CSI测量资源的频域占用带宽例如可以小于或等于导频传输带宽(或者称，测量带宽)。在NR中，用于指示CSI测量资源的频域占用带宽例如可以是CSI占用带宽范围(CSI-Frequency Occupation)。
应理解，CSI测量资源的频域占用带宽仅为便于描述而命名，不应对本申请构成任何限定，本申请并不排除通过其他命名来表达相同含义的可能。
可选地，该频域向量的长度是用于指示待上报的频域单元的位置及个数的信令的长度。
在NR中，用于指示待上报的频域单元的位置及个数的信令可以是上报带宽(reporting band)。该信令例如可以通过位图的形式来指示待上报的频域单元的位置及个数。因此，频域向量的维度可以为该位图的比特数。应理解，reporting band仅为该信令的一种可能的命名，不应对本申请构成让任何限定。本申请并不排除通过其他名称来命名该信令以实现相同或相似功能的可能。
可选地，频域向量的长度是待上报的频域单元数。
其中，待上报的频域单元数例如可以通过上述上报带宽这一信令指示。待上报的频域单元数可以为该CSI测量资源的频域占用带宽中的全部频域单元，或者，也可以为该CSI测量资源的频域占用带宽中的部分频域单元；或者，待上报的频域单元数可以与上报带宽的信令长度相同，或者，也可以小于上报带宽的信令长度。本申请对此不作限定。
当协议定义了频域向量的长度为上述列举的某一项时，用于指示上述CSI测量资源的频域占用带宽的信令或用于指示待上报的频域单元的位置及个数的信令中的某一项可以认为隐式地指示了频域向量的长度。为便于区分和说明，将用于频域向量的长度的指示信息记作第五指示信息。该第五指示信息可能是上述用于指示上述CSI测量资源的频域占用带宽的信令，也可能是用于指示待上报的频域单元的位置及个数的信令，或者还可能是未来协议中新增的信令，本申请对此不作限定。
假设该频域向量的长度为N f，则频域向量可以是维度为N f×1的列向量，也可以是维度为1×N f的行向量。本申请对此不做限定。
8、空频矩阵与空频分量矩阵：为方便说明，下文中假设发射天线的极化方向数为1。
若发射天线的极化方向数为1，则由一个传输层上不同频域单元上的预编码向量可以构建得到该极化方向上的空频矩阵。
在本申请实施例中，终端设备例如可以通过信道测量等方式确定各频域单元上待反馈的预编码矩阵，该各频域单元上待反馈的预编码矩阵进行处理，可以得到与每个传输层对应的空频矩阵。例如，对于同一个传输层，将各频域单元上待反馈的预编码向量组合可以得到空频矩阵。该空频矩阵可以称为待反馈的空频矩阵。终端设备可以通过一个或多个空频分量矩阵的加权和指示该待反馈的空频矩阵。换句话说，待反馈的空频矩阵可以近似为一个或多个空频分量矩阵的加权和。该一个或多个空频分量矩阵可以取自预先定义的空频分量矩阵集合，或者，也可以由预先定义的波束向量集合中的波束向量和预先定义的频域向量集合中的频域向量确定。本申请对此不作限定。
在一种可能的设计中，空频矩阵可以是维度为N s×N f的矩阵。即，该空频矩阵可以包括N f个长度为N s的列向量。该N f个列向量可以与N f个频域单元对应，每个列向量可用于确定所对应的频域单元的预编码向量。
例如，空频矩阵可以记作H，
Figure PCTCN2019110342-appb-000006
其中，h 0至
Figure PCTCN2019110342-appb-000007
是与N f个频域单元对应的N f个列向量，各列向量的长度均可以为N s。该N f个列向量可分别用于确定N f个频域单元的预编码向量。
空频矩阵可以近似为一个或多个空频分量矩阵的加权和。
在本实施例中，一个空频分量矩阵可以由一个波束向量和一个频域向量唯一确定。例如，当波束向量和频域向量均为列向量时，一个空频分量矩阵可以为一个波束向量和一个频域向量的共轭转置的乘积；当波束向量为列向量、频域向量为行向量时，一个空频分量矩阵可以为一个波束向量和一个频域向量的乘积。因此，每个空频分量矩阵也可以是维度为N s×N f的矩阵。
在另一种可能的设计中，空频矩阵可以是维度为(N s×N f)×1的矩阵，或者也可以说，空频矩阵可以是长度为N s×N f的向量。也就是说，该空频矩阵可以仅包括一个长度为N s×N f的列向量。此情况下，该空频矩阵也可以称为空频向量。
例如，空频矩阵可以记作H，
Figure PCTCN2019110342-appb-000008
其中的各向量在上文中已经做了详细说明，为了简洁，这里不再赘述。
空频向量可以近似为一个或多个空频分量向量的加权和。
在本实施例中，一个空频分量向量可以由一个波束向量和一个频域向量唯一确定。例如，当波束向量和频域向量均为列向量时，一个空频分量向量可以是一个波束向量和一个频域向量的克罗内克积，也可以是一个频域向量与一个波束向量的克罗内克积。因此，每个空频分量向量也可以是长度为N s×N f的向量。在此情况下，空频分量矩阵也可以称为空频分量向量。
若空频分量向量由频域向量和波束向量的克罗内克积确定，由多个空频分量向量的加权和所确定的空频向量可以是由N f个长度为N s的列向量依次连接而成。该N f个列向量可以与N f个频域单元对应，每个列向量可用于确定所对应的频域单元的预编码向量。
若空频分量向量由波束向量和频域向量的克罗内克积确定，由多个空频分量向量的加权和所确定的空频向量可以是由N s个长度为N f的列向量依次连接而成。每个列向量中的 N f个元素可以与N f个频域单元对应。该N s个列向量的每个列向量中的第n f个元素依次连接可以得到长度为N s的向量，该向量可用于确定第n f个频域单元的预编码向量。其中，0≤n f≤N f-1，且n f为整数。
应理解，上文仅为便于理解说明了极化方向数为1时空频矩阵的几种可能的形式，如，维度为N s×N f的矩阵或长度为N s×N f的向量，但这不应对本申请构成任何限定。当极化方向数大于1时，空频矩阵仍然可以表示为上文所列举的几种形式，只是极化方向数的不同，空频矩阵的维度也可能不同。例如，当极化方向数为2时，空频矩阵可以是维度为2N s×N f的矩阵，也可以是长度为2N s×N f的向量。其中，2表示两个极化方向。
然而，空频分量矩阵仍然可以是维度为N s×N f的矩阵或长度为N s×N f的向量。因此，对于每个极化方向上的空频矩阵，可以分别由多个空频分量矩阵加权求和来表示。换句话说，每个极化方向上的空频矩阵可以近似表示为多个空频分量矩阵的加权和。用于不同极化方向的多个空频分量矩阵可以是相同的，或者说，多个极化方向可以共用相同的多个空频分量矩阵。换句话说，同一传输层上多个极化方向的空频矩阵或空频向量可以由同一组波束向量和同一组频域向量来构建。但不同极化方向上各空频分量矩阵的加权系数可能不同。
在本申请实施例中，通过一个波束向量和一个频域向量的运算可以得到的基本单位可以是一个空频基本单位，例如可以是空频分量矩阵，也可以是空频分量向量。该空频基本单位可以对应一个极化方向。这个基本单位的加权和可以拼接出多个极化方向的空频矩阵。
还应理解，空频矩阵的具体形式并不限于上文所列举，为了简洁，这里不再一一举例说明。在下文示出的实施例中，分别结合空频分量矩阵为波束向量和频域向量的共轭转置的乘积以及空频分量矩阵为频域向量和波束向量的克罗内克积这两种形式，详细说明了终端设备指示预编码向量和网络设备确定预编码向量的具体过程。但这不应对本申请构成任何限定。本领域的技术人员基于相同的构思，可以对上述空频分量矩阵进行等价的变形或替换，这些等价的变形和替换均应落入本申请的保护范围内。
如前所述，在下行信道测量中，网络设备根据PMI确定出的预编码矩阵与终端设备所确定的预编码矩阵的近似度越高，其确定出的用于数据传输的预编码矩阵也就越能够与信道状态相适配，因此也就能够提高信号的接收质量。
为了提高频谱资源的利用率，提高通信系统的数据传输能力，网络设备可以通过多个传输层向终端设备传输数据。然而，当传输层数增加时，终端设备基于每个传输层进行反馈所带来的开销也会成倍增加。而子带数量越多，反馈开销增加的幅度也会越大。因此，希望提供一种方法，能够降低反馈开销。
有鉴于此，本申请提供一种指示和确定预编码向量的方法，以期降低PMI的反馈开销。
为了便于理解本申请实施例，作出以下几点说明。
第一，在本申请实施例中，假设发射天线的极化方向数为P(P≥1且为整数)，传输层数为R(R≥1且为整数)。
在本实施例中，为便于描述，在涉及编号时，可以从0开始连续编号。例如，R个传输层可以包括第0个传输层至第R-1个传输层，P个极化方向可以包括第0个极化方向至 第P-1个极化方向。当然，具体实现时不限于此，例如，可以从1开始连续编号。应理解，上文所述均为便于描述本申请实施例提供的技术方案而进行的设置，而并非用于限制本申请的范围。
第二，在本申请实施例中，多处涉及矩阵和向量的变换。为便于理解，这里做统一说明。上角标T表示转置，如A T表示矩阵(或向量)A的转置；上角标*表示共轭转置，如，A *表示矩阵(或向量)A的共轭转置。后文中为了简洁，省略对相同或相似情况的说明。
第三，在下文示出的实施例中，以波束向量和频域向量均为列向量为例来说明本申请提供的实施例，但这不应对本申请构成任何限定。基于相同的构思，本领域的技术人员还可以想到其他更多可能的表现方式。
第四，本申请实施例涉及矩阵的克罗内克(Kronecker)积运算。在本申请实施例中，克罗内克积运算可用
Figure PCTCN2019110342-appb-000009
表示。例如，矩阵A和B的克罗内克积可表示为
Figure PCTCN2019110342-appb-000010
克罗内克积是一个矩阵中的所有元素分别乘以另一矩阵组成的分块矩阵。例如，维度为k×l维的矩阵A和p×q维的矩阵B的克罗内克尔积乘积得到kp×ql维的矩阵，具体如下：
Figure PCTCN2019110342-appb-000011
有关克罗内克积的具体定义可参考现有技术。为了简洁，本文不再赘述。
第五，本申请实施例中多处涉及向量间的投影。例如，将向量a投影至向量b，可以理解为求向量a与向量b的内积。
第六，在本申请实施例中，“用于指示”可以包括用于直接指示和用于间接指示。例如，当描述某一指示信息用于指示信息I时，可以包括该指示信息直接指示I或间接指示I，而并不代表该指示信息中一定携带有I。
将指示信息所指示的信息称为待指示信息，则具体实现过程中，对待指示信息进行指示的方式有很多种，例如但不限于，可以直接指示待指示信息，如待指示信息本身或者该待指示信息的索引等。也可以通过指示其他信息来间接指示待指示信息，其中该其他信息与待指示信息之间存在关联关系。还可以仅仅指示待指示信息的一部分，而待指示信息的其他部分则是已知的或者提前约定的。例如，还可以借助预先约定(例如协议规定)的各个信息的排列顺序来实现对特定信息的指示，从而在一定程度上降低指示开销。同时，还可以识别各个信息的通用部分并统一指示，以降低单独指示同样的信息而带来的指示开销。例如，本领域的技术人员应当明白，预编码矩阵是由预编码向量组成的，预编码矩阵中的各个预编码向量，在组成或者其他属性方面，可能存在相同的部分。
此外，具体的指示方式还可以是现有各种指示方式，例如但不限于，上述指示方式及其各种组合等。各种指示方式的具体细节可以参考现有技术，本文不再赘述。由上文所述可知，举例来说，当需要指示相同类型的多个信息时，可能会出现不同信息的指示方式不相同的情形。具体实现过程中，可以根据具体的需要选择所需的指示方式，本申请实施例对选择的指示方式不做限定，如此一来，本申请实施例涉及的指示方式应理解为涵盖可以使得待指示方获知待指示信息的各种方法。
此外，待指示信息可能存在其他等价形式，例如行向量可以表现为列向量，一个矩阵 可以通过该矩阵的转置矩阵来表示，一个矩阵也可以表现为向量或者数组的形式，该向量或者数组可以由该矩阵的各个行向量或者列向量相互连接而成，两个向量的克罗内克尔积也可以通过一个向量与另一个向量的转置向量的乘积等形式来表现等。本申请实施例提供的技术方案应理解为涵盖各种形式。举例来说，本申请实施例涉及的部分或者全部特性，应理解为涵盖该特性的各种表现形式。
待指示信息可以作为一个整体一起发送，也可以分成多个子信息分开发送，而且这些子信息的发送周期和/或发送时机可以相同，也可以不同。具体发送方法本申请不进行限定。其中，这些子信息的发送周期和/或发送时机可以是预先定义的，例如根据协议预先定义的，也可以是发射端设备通过向接收端设备发送配置信息来配置的。其中，该配置信息可以例如但不限于包括无线资源控制信令，例如RRC信令、MAC层信令，例如MAC-CE信令和物理层信令，例如下行控制信息(downlink control information，DCI)中的一种或者至少两种的组合。
第七，本申请对很多特性(例如克罗内克积、PMI、频域单元、波束、波束向量以及波束向量的加权系数等)所列出的定义仅用于以举例方式来解释该特性的功能，其详细内容可以参考现有技术。
第八，在下文示出的实施例中第一、第二、第三、第四以及各种数字编号仅为描述方便进行的区分，并不用来限制本申请实施例的范围。例如，区分不同的指示信息、不同的传输层等。
第九，在下文示出的实施例中，“预先获取”可包括由网络设备信令指示或者预先定义，例如，协议定义。其中，“预先定义”可以通过在设备(例如，包括终端设备和网络设备)中预先保存相应的代码、表格或其他可用于指示相关信息的方式来实现，本申请对于其具体的实现方式不做限定。
第十，本申请实施例中涉及的“保存”，可以是指的保存在一个或者多个存储器中。所述一个或者多个存储器，可以是单独的设置，也可以是集成在编码器或者译码器，处理器、或通信装置中。所述一个或者多个存储器，也可以是一部分单独设置，一部分集成在译码器、处理器、或通信装置中。存储器的类型可以是任意形式的存储介质，本申请并不对此限定。
第十一，本申请实施例中涉及的“协议”可以是指通信领域的标准协议，例如可以包括LTE协议、NR协议以及应用于未来的通信系统中的相关协议，本申请对此不做限定。
第十二，“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B的情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a、b和c中的至少一项(个)，可以表示：a，或，b，或，c，或，a和b，或，a和c，或，b和c，或，a、b和c。其中a、b和c分别可以是单个，也可以是多个。
下面将结合附图详细说明本申请实施例提供的指示和确定预编码向量的方法。
应理解，本申请实施例提供的方法可以应用于通过多天线技术通信的系统，例如，图1中所示的通信系统100。该通信系统可以包括至少一个网络设备和至少一个终端设备。 网络设备和终端设备之间可通过多天线技术通信。
还应理解，下文示出的实施例并未对本申请实施例提供的方法的执行主体的具体结构特别限定，只要能够通过运行记录有本申请实施例的提供的方法的代码的程序，以根据本申请实施例提供的方法进行通信即可，例如，本申请实施例提供的方法的执行主体可以是终端设备或网络设备，或者，是终端设备或网络设备中能够调用程序并执行程序的功能模块。
以下，不失一般性，以网络设备与终端设备之间的交互为例详细说明本申请实施例提供的指示和确定预编码向量的方法。
图2是从设备交互的角度示出的本申请实施例提供的指示和确定预编码向量的方法200的示意性流程图。如图所示，该方法200可以包括步骤210至步骤230。下面详细说明方法200中的各步骤。
为便于理解，首先详细说明终端设备基于一个或多个传输层中的一个传输层、该传输上的一个或多个极化方向中的一个极化方向，指示预编码向量和网络设备确定预编码向量的具体过程。应理解，本申请对于传输层数以及发射天线的极化方向数并不做限定。下文所示例说明的一个传输层可以为一个或多个传输层中的任意一个传输层，一个极化方向可以为一个或多个极化方向中的任意一个极化方向。
在步骤210中，终端设备生成第一指示信息，该第一指示信息用于指示波束向量集合中的L 1(L 1≥1且为整数)个波束向量、频域向量集合中的K 1(K 1≥1且为整数)个频域向量以及与L 1个波束向量和K 1个频域向量对应的L 1×K 1个空频分量矩阵中的T 1(T 1≥1且为整数)个空频分量矩阵。其中，该T 1个空频分量矩阵可以是从与L 1个波束向量和K 1个频域向量对应的M 1(M 1＝L 1×K 1)个空频分量矩阵中确定的。换句话说，该T 1个空频分量矩阵可以是该M 1个空频分量矩阵的子集。这里的M 1个空频分量矩阵可以是由分别遍历L 1个波束向量和K 1个频域向量运算得到。
假设波束向量集合中被选择的L 1个波束向量记作
Figure PCTCN2019110342-appb-000012
频域向量集合中被选择的K 1个频域向量记作
Figure PCTCN2019110342-appb-000013
终端设备可以先在0至L 1-1的范围内遍历各波束向量，再在0至K 1-1的范围内遍历各频域向量，得到M 1个空频分量矩阵。
以波束向量与频域向量的共轭转置的乘积为例，该M 1个空频分量矩阵可以包括：
Figure PCTCN2019110342-appb-000014
终端设备也可以先在0至K 1-1的范围内遍历各频域向量，再在0至L 1-1的范围内遍历各波束向量，得到M 1个空频分量矩阵。
仍以波束向量与频域向量的共轭转置的乘积为例，该M 1个空频分量矩阵可以包括：
Figure PCTCN2019110342-appb-000015
应理解，上文列举的空频分量矩阵的形式仅为示例，而不应对本申请构成任何限定。基于上述规则，通过频域向量和波束向量的克罗内克积的方式也可以得到M 1个空频分量矩阵。
可以看到，该M 1个空频分量矩阵的每个空频分量矩阵可以由一个波束向量和一个频域向量唯一确定，且M 1＝L 1×K 1。
该第一指示信息信息在用于指示该T 1个空频分量矩阵时，可用于指示该T 1个空频分量矩阵在M 1个空频分量矩阵中的相对位置(例如相对索引或相对编号等)。例如，终端设备可以通过位图，或，T 1个空频分量矩阵的组合在M 1个空频分量矩阵中的索引，或，T 1个空频分量矩阵中各空频分量矩阵在M 1个空频分量矩阵中的索引等方式来指示该T 1个空频分量矩阵。
若终端设备采用位图来指示该T 1个空频分量矩阵，则该T 1个空频分量矩阵在M 1个空频分量矩阵中的相对位置可以使用M 1个比特。此外，终端设备还可以采用
Figure PCTCN2019110342-appb-000016
个比特来指示L 1个波束向量，并采用
Figure PCTCN2019110342-appb-000017
个比特来指示K 1个频域向量。因此，终端设备可以采用
Figure PCTCN2019110342-appb-000018
个比特来指示T 1个空频分量矩阵。
然而，如果直接在波束向量集合和频域向量集合中指示该T 1个空频分量矩阵，所带来的开销是不同的。例如，波束向量集合包括L 0个波束向量，频域向量集合包括K 0个频域向量，L 0≥L 1，K 0≥K 1，且L 0×K 0＞L 1×K 1。该L 0个波束向量和K 0个频域向量可以对应L 0×K 0个空频分量矩阵。
若终端设备直接指示该T 1个空频分量矩阵在L 0×K 0个空频分量矩阵中的相对位置，若对每个空频分量矩阵分别指示，则可能需要
Figure PCTCN2019110342-appb-000019
个比特；若采用位图来指示，则可以使用L 0×K 0个比特。
在某些情况下，如L 1和/或K 1的取值较小的情况下，可以大大减小反馈开销。
举例而言，假设波束向量集合包括16个波束向量，频域向量集合包括10个频域向量，从该波束向量集合和频域向量集合构建的160个空频分量矩阵中选择15个空频向量矩阵。
在当前技术中，若对每个空频分量矩阵分别指示，则对于每个空频向量矩阵需要
Figure PCTCN2019110342-appb-000020
个比特，即，8个比特，则15个空频向量矩阵需要120个比特来反馈。若采用位图方式来指示该15个空频分量矩阵，则需要160个比特来反馈。
若反馈15个空频分量矩阵的组合的索引，则由于波束向量集合和频域向量集合中的各向量可以两两组合，终端设备和网络设备可能需要预存大量的组合与索引的一一对应关系。
而在本实施例中，可以先从该16个波束向量和10个频域向量中指示部分较强的波束向量和较强的频域向量，再从该部分较强的波束向量和较强的频域向量中选择15个空频分量矩阵。
例如，假设选择8个波束向量和5个频域向量，则可以分别通过
Figure PCTCN2019110342-appb-000021
个比特和
Figure PCTCN2019110342-appb-000022
个比特来反馈，即，分别为14个比特和8个比特。在该8个波束向量和5个频域向量组合得到的40个空频分量矩阵中，终端设备可以进一步指示被选择的15个空频分量矩阵。若终端设备通过位图来指示该15个空频分量矩阵，则可以使用40个比特来指示。则该15个空频分量矩阵可以通过62个比特来反馈。相比于上述方式而言，可以大大降低开销。
应理解，上文列举的波束向量、频域向量和空频分量矩阵的个数以及所带来的比特开销仅为示例，而不应对本申请构成任何限定。
还应理解，上文列举的用于指示T 1个空频向量对的方法仅为示例，不应对本申请构成任何限定。本申请对于终端设备指示T 1个空频向量对的具体方法不作限定。
在本实施例中，该T 1个空频分量矩阵的加权和可用于确定一个或多个频域单元的预编码向量。具体地，该T 1个空频分量矩阵的加权和可用于构建空频矩阵。该空频矩阵可以包括与一个或多个频域单元对应的列向量，每个列向量可用于确定所对应的频域单元的预编码向量。上文中已经对空频矩阵与预编码向量的关系做了详细说明，为了简洁，这里不再赘述。
该空频矩阵可以近似为上述T 1个空频分量矩阵的加权和。例如，空频矩阵可以记作
Figure PCTCN2019110342-appb-000023
其中，
Figure PCTCN2019110342-appb-000024
表示T 1个空频分量矩阵中的第t 1个空频分量矩阵，
Figure PCTCN2019110342-appb-000025
表示第t 1个空频分量矩阵
Figure PCTCN2019110342-appb-000026
的加权系数。
终端设备可以预先确定该T 1个空频分量矩阵，也可以预先确定用于生成该T 1个空频分量矩阵的空频向量对，然后可进一步确定该T 1个空频分量矩阵或T 1个空频向量对的权重。终端设备可以将该T 1个空频分量矩阵及各空频分量矩阵的权重指示给网络设备，或者将该T 1个空频向量对及各空频向量对的权重指示给终端设备，以便于网络设备恢复出一个或多个频域单元的预编码向量。
因此，上述第一指示信息所指示的T 1个空频分量矩阵，结合各空频分量矩阵的权重，可用于确定预编码向量。其中，各空频分量矩阵的权重可以通过该第一指示信息来指示，也可以通过其他信息指示。本申请对此不作限定。各空频分量矩阵的权重的具体指示方法可以参考现有技术。
在本实施例中，上述T 1个空频分量矩阵中的每个空频分量矩阵可以是由T 1个空频向量对中的一个波束向量和一个频域向量唯一确定。因此，终端设备可以直接指示T 1个空频分量矩阵，也可以通过指示T 1个空频向量对的方式来间接地指示T 1个空频分量矩阵,或者，也可以直接指示该T 1个空频向量对。该T 1个空频向量对可以视为该T 1个空频分量矩阵的等价形式。
在一种实现方式中，该T 1个空频分量矩阵选自与L 1个波束向量和K 1个频域向量对应的M 1个空频分量矩阵。该M 1个空频分量矩阵由L 1个波束向量和K 1个频域向量确定，每个空频分量矩阵可以由一个波束向量和一个频域向量唯一确定，且M 1＝L 1×K 1。
因此，该T 1个空频分量矩阵可以基于由波束向量集合中的L 1个波束向量和频域向量集合中的K 1个频域向量确定。终端设备可以基于该L 1个波束向量和K 1个频域向量中指示T 1个空频分量矩阵。
在另一种实现方式中，该T 1个空频向量对可以选自由L 1个波束向量和K 1个频域向量组合得到的M 1个空频向量对。该M 1个空频向量对中的每个空频向量对可以由一个波束向量和一个频域向量唯一确定，且M 1＝L 1×K 1。
因此，该T 1个空频向量对可以选自由波束向量集合中的L 1个波束向量和频域向量集合中的K 1个频域向量组合得到的M 1个空频向量对。该T 1个空频向量对可以是该M 1个空频向量对中的部分空频向量对。终端设备可以基于该L 1个波束向量和K 1个频域向量，或者，基于M 1个空频向量对，指示T 1个空频向量对。
其中，L 1个波束向量可以是波束向量集合中的部分波束向量，和/或，K 1个频域向量 可以是频域向量集合中的部分频域向量。换句话说，当L 1个波束向量为波束向量集合中的全部波束向量时，该K 1个频域向量仅为频域向量集合中的部分频域向量；当K 1个频域向量为频域向量集合中的全部频域向量时，该L 1个波束向量仅为波束向量集合中的部分波束向量；当L 1个波束向量为波束向量集合中的部分波束向量时，该K 1个频域向量可以为频域向量集合中的部分或全部频域向量；当K 1个频域向量为频域向量集合中的部分频域向量时，该L 1个波束向量可以为波束向量集合中的部分或全部波束向量。
假设波束向量集合包括L 0个波束向量，频域向量集合包括K 0个频域向量。则L 0≥L 1，K 0≥K 1，且L 0、L 1、K 0和K 1不同时满足L 0＝L 1和K 0＝K 1。
当该L 1个波束向量为波束向量集合中的部分波束向量时，该L 1个波束向量可以是从波束向量集合中选出的较强的L 1个波束向量。当该K 1个频域向量为频域向量集合中的部分频域向量时，该K 1个频域向量可以是从频域向量集合中选出的较强的K 1个频域向量。其中，较强的L 1个波束向量可以理解为加权系数较大的L个波束向量，较强的K 1个频域向量可以理解为加权系数较大的K 1个波束向量。这是因为加权系数较大的波束向量和频域向量在线性组合中所占的权重较大，对预编码向量的近似精度的影响也较大。后文中会结合具体的实现方式详细说明较强的L 1个波束向量和较强的K 1个频域向量，这里暂且省略对其的详细说明。
综上，终端设备可以预先从波束向量集合中确定L 1个波束向量，并从频域向量集合中确定K 1个频域向量，将用作加权求和的T 1个空频分量矩阵的选择范围缩小至由L 1个波束向量和K 1个频域向量构建的M 1个空频分量矩阵的范围中，并从M 1个空频分量矩阵中选择和指示T 1个空频分量矩阵，从而有利于减小该T 1个空频分量矩阵的反馈开销。
应注意，这里仅为便于理解而引入M 1个空频分量矩阵的概念，这并不代表终端设备一定会生成该M 1个空频分量矩阵，终端设备也可以由该L 1个波束向量和K 1个频域向量组合得到M 1个空频向量对。但可以理解的是，由该L 1个波束向量和K 1个频域向量，或者，由该M 1个空频向量对，可以构建得到该M 1个空频分量矩阵。或者说，M 1个空频向量对与M 1个空频分量矩阵可以相互转换。因此可以认为该M 1个空频分量矩阵是与L 1个波束向量和K 1个频域向量对应。引入M 1仅仅为了体现M 1个空频分量矩阵(或空频分量向量、或空频向量对)与L 1个波束向量和K 1个频域向量的对应关系，而不应对本申请构成任何限定。
其中，L 1、K 1以及T 1的取值可以由网络设备指示，也可以预先定义，如协议定义，还可以由终端设备确定后上报网络设备，或者，还可以将上述列举的方法结合来配置。
若L 1和K 1的取值由网络设备指示，则可选地，该方法还包括：终端设备接收第二指示信息，该第二指示信息用于指示L 1、K 1和M 1中至少两项的取值。相应地，网络设备发送该第二指示信息。
可选地，该第二指示信息携带在高层信令中，如RRC消息。
若L 1和K 1的取值由终端设备确定并上报，则可选地，该方法还包括：终端设备发送第二指示信息，该第二指示信息用于指示L 1、K 1和M 1中一项或多项取值。相应地，网络设备接收该第二指示信息。
可选地，该第二指示信息携带在上行控制信息(uplink control information，UCI)中，如CSI。
由于L 1、K 1和M 1满足M 1＝L 1×K 1，在确定了其中任意两项的取值的情况下，另一项的取值也可以确定。
应理解，用于指示L 1、K 1和M 1的取值的信息可以是同一信息，也可以是不同信息，本申请对此不作限定。
可选地，该L 1和K 1中任意一项的取值也可以是预先定义，如协议定义，而另一项由网络设备通过信令指示。
例如，网络设备可以通过信令指示L 1的取值，协议可以定义K 1的取值，本申请对于K 1的取值大小不作限定。
又例如，网络设备可以通过信令指示L 1的取值，协议可以定义K 1的取值为某一参数的取值，或者，也可以定义K 1的计算式。如，协议可以定义K 1的取值为频域向量的长度N f，或者，也可以定义K 1的计算式，如，
Figure PCTCN2019110342-appb-000027
或
Figure PCTCN2019110342-appb-000028
此情况下，该K 1的取值可以理解为由上述第五指示信息隐式指示。其中，
Figure PCTCN2019110342-appb-000029
表示向上取整，
Figure PCTCN2019110342-appb-000030
表示向下取整，[ ]表示四舍五入取整。
再例如，网络设备可以通过信令指示K 1的取值，协议可以定义L 1的取值，本申请对于L 1的取值大小不作限定。
还例如，网络设备可以通过信令指示K 1的取值，协议可以定义L 1的取值为某一参数的取值，或者，也可以定义L 1的计算式。如，协议可以定义L 1的取值为一个极化方向上的天线端口数N s，或者，也可以定义L 1的计算式，如，
Figure PCTCN2019110342-appb-000031
或L 1＝[N s/2]。此情况下，该L 1的取值可以理解为由用于指示单极化方向上的天线端口数的指示信息隐式指示。
应理解，上文列举的协议定义的L 1或K 1的取值和计算式仅为示例，而不应对本申请构成任何限定。
可选地，该L 1和K 1中任意一项的取值也可以是预先定义，如协议定义，而另一项由终端设备确定并通过信令上报。
上文中已经对结合协议定义和网络设备指示的方式来确定L 1和K 1的过程做了详细说明，结合协议定义和终端设备指示的方式来确定L 1和K 1的过程与之相似，为了简洁，这里不再赘述。
由于L 1和K 1中的某一项也可以由协议定义，或者，由协议定义为某一参数的取值，或者，由协议定义计算式，因此，该第二指示信息可以指示L 1和K 1中另一项的取值。例如，协议定义K 1的取值，第二指示信息仅指示L 1的取值；或，协议定义L 1的取值，第二指示信息仅指示K 1的取值。
由于L 1、K 1和M 1满足M 1＝L 1×K 1，在确定了其中任意两项的取值的情况下，另一项的取值也可以确定。例如，当K 1由协议定义时，该第二指示信息也可以通过指示M 1的取值来间接地指示L 1的取值；当L 1由协议定义时，该第二指示信息也可以通过指示M 1的取值来间接地指示K 1的取值。
应理解，上文列举的用于携带第二指示信息的信令仅为示例，而不应对本申请构成任何限定，本申请对于携带第二指示信息的具体信令不作限定。
若T 1的取值由网络设备指示，则可选地，该方法还包括：终端设备接收第三指示信息，该第三指示信息用于指示T 1的取值。相应地，网络设备发送该第三指示信息。
可选地，该第三指示信息携带在高层信令中，如RRC消息。
若T 1的取值由终端设备确定并上报，则可选地，该方法还包括：终端设备发送第三指示信息，该第三指示信息用于指示T 1的取值。相应地，网络设备接收该第三指示信息。
可选地，该第三指示信息携带在UCI中，如CSI。
应理解，上文列举的用于第三指示信息的信令仅为示例，而不应对本申请构成任何限定，本申请对于携带第三指示信息的具体信令不作限定。
还应理解，上述第二指示信息和第三指示信息可以是同一信息，也可以是不同信息，本申请对此不作限定。
在确定了L 1、K 1和T 1的取值后，终端设备可以确定L 1个波束向量、K 1个频域向量和T 1个空频分量矩阵，进而生成第一指示信息。
可选地，该第一指示信息信息用于指示L 1个波束向量和T 1个空频分量矩阵，该T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，该L 1个波束向量和K 1个频域向量与M 1个空频分量矩阵对应，该T 1个空频分量矩阵是M 1个空频分量矩阵中的部分空频分量矩阵，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 1个波束向量中的一个波束向量和K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；该L 1个波束向量为该波束向量集合中的部分波束向量，和/或，该K 1个频域向量为该频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数。
可选地，该K 1个频域向量是预先配置的。如，该K 1个频域向量可以为频域向量集合的全部或部分频域向量。
例如，协议可以预先定义K 1＝K 0，即，协议默认将该频域向量集合的全集作为该K 1个频域向量。又例如，协议可以预先定义K 1的取值，并预先指定将频域向量集合中的哪些频域向量作为该K 1个频域向量。再例如，协议可以预先定义K 1的取值，网络设备可以预先通过信令指示将该K 1个频域向量。
换句话说，协议可以预先定义终端设备不需要上报该K 1个频域向量。该K 1个频域向量可以是预先指定的，如协议定义或网络设备配置等，本申请对此不作限定。
可选地，协议也可以预先定义终端设备基于参数的不同取值确定需要上报的向量。
如前所述，L 0≥L 1，K 0≥K 1，且L 0、L 1、K 0和K 1不同时满足L 0＝L 1和K 0＝K 1。该L 1个波束向量有可能是波束向量集合的全集，或者，K 1个频域向量有可能是频域向量集合的全集。可选地，终端设备可以根据L 1和K 1的取值确定是否选择波束向量集合的全集或频域向量集合的全集。当L 1个波束向量为波束向量集合的全集时，终端设备可以对该L 1个波束向量不做上报；当K 1个频域向量为频域向量集合的全集时，终端设备可以对该K 1个频域向量不做上报。
在一种实现方式中，网络设备可以通过信令配置K 1的取值，如上述第二指示信息。当K 1被配置为与K 0相等时，终端设备可以默认将频域向量集合中的全集作为K 1个频域向量。或者，网络设备也可以不通过额外的信令指示K 1的取值，如，在上述第二指示信息中仅指示L 1的取值。即，可选地，上述第二指示信息用于指示L 1的取值。这可以理解为该网络设备通过第二指示信息隐式地指示K 1的取值为K 0。
终端设备可以根据接收到的信令判断是否上报K 1个频域向量。如，当确定K 1＝K 0时，可以不上报K 1个频域向量。网络设备也可以在确定K 1＝K 0的情况下，直接将频域向量集 合中的频域向量当作K 1个频域向量。
在另一种实现方式中，该K 1的取值可以由协议定义为固定值。当K 1的取值被定义为K 0时，终端设备可以默认将频域向量集合的全集作为K 1个频域向量，而不需要上报该K 1个频域向量。
与之相似地，在一种实现方式中，网络设备可以通过信令配置L 1的取值，如上述第二指示信息。当L 1被配置为与L 0相等时，终端设备可以默认将波束向量集合的全集作为L 1个波束向量。或者，网络设备也可以不通过额外的信息指示L 1的取值，如，在上述第二指示信息中仅指示K 1的取值。即，可选地，上述第二指示信息用于指示K 1的取值。此时可理解为该网络设备通过第二指示信息隐式地指示L 1的取值为L 0。
终端设备可以根据接收到的信令判断是否上报L 1个波束向量。如，当确定L 1＝L 0时，可以不上报L 1个波束向量。网络设备也可以在确定L 1＝L 0的情况下，直接将波束向量集合中的波束向量当作L 1个波束向量。
在另一种实现方式中，该L 1的取值可以由协议定义为固定值。当L 1的取值被定义为L 0时，则终端设备可以默认将波束向量集合的全集作为L 1个波束向量，而不需要上报该L 1个波束向量。
再进一步地，当L 1、K 1和T 1的取值均由网络设备配置时，终端设备可以根据网络设备所配置的参数来确定需要上报的向量。可选地，协议也可以预先定义终端设备基于网络设备配置的不同参数确定需要上报的向量。
例如，若网络设备通过信令指示L 1、K 1和T 1的取值，则终端设备可以默认需要上报L 1个波束向量、K 1个频域向量和T 1个空频向量对；若网络设备通过信令指示L 1和T 1的取值，则终端设备可以默认K 1＝K 0，即，可以将频域向量集合中的K 0个频域向量均用作K 1个频域向量，可以仅上报L 1个波束向量和T 1个空频向量对；若网络设备通过信令指示K 1和T 1的取值，则终端设备可以默认L 1＝L 0，即，可以将波束向量集合中的L 0个波束向量均用作L 1个波束向量，可以仅上报K 1个频域向量和T 1个空频向量对；若网络设备通过信令指示L 1和K 1的取值，则终端设备可以默认直接将该L 1个波束向量和K 1个频域向量用来做加权求和，可以仅上报该L 1个波束向量和K 1个频域向量，而不需要上报T 1个空频向量对。
应理解，上文列举了网络设备在配置不同的参数的情况下终端设备需要确定和上报的向量，但这不应对本申请构成任何限定。
下面以L 1≠L 0且K 1≠K 0为例，详细说明终端设备确定和指示该T 1个空频分量矩阵及各空频分量矩阵的权重的具体方法。
终端设备可以基于预先保存的向量集合的形式，采用相应的实现方式来确定T 1个空频分量矩阵。例如，终端设备可以预先保存有波束向量集合和频域向量集合，基于实现方式一来确定T 1个空频向量对；终端设备也可以预先保存有空频分量矩阵集合，基于实现方式二来确定T 1个空频分量矩阵。
需要说明的是，波束向量集合和频域向量集合与空频分量矩阵集合之间可以互相转换。波束向量集合中的任意一个波束向量和频域向量集合中的任意一个频域向量可以确定得到空频分量矩阵集合中的一个空频分量矩阵。空频分量矩阵集合中的任意一个空频分量矩阵可以由波束向量集合中的一个波束向量和频域向量集合中的一个频域向量唯一确定。
因此，空频向量矩阵集合中与各空频向量矩阵对应的索引也可以转换为波束向量集合中的波束向量的索引和频域向量集合中频域向量的索引。换句话说，空频向量矩阵集合中任意一个空频分量向量矩阵可以通过波束向量集合中的一个波束向量和频域向量集合中的一个频域向量来联合指示。
在本实施例中，终端设备至少可以通过以下任意一种实现方式，由波束向量集合中的L 1个波束向量和频域向量集合中的K 1个频域向量指示T 1个空频分量矩阵：
实现方式一、T 1个空频分量矩阵可以由M 1个空频向量对确定，该M 1个空频向量对可以由L 1个波束向量和K 1个频域向量组合得到，终端设备可以从该M 1个空频向量对中指示用于生成该T 1个空频分量矩阵的T 1个空频向量对；
实现方式二、T 1个空频分量矩阵可以选自M 1个空频分量矩阵，该M 1个空频分量矩阵可以由L 1个波束向量和K 1个频域向量确定，终端设备可以从该M 1个空频分量矩阵中指示该T 1个空频分量矩阵。需要说明的是，本申请对于终端设备和网络设备预先保存的向量集合或矩阵集合的形式并不做限定。例如，终端设备和网络设备均可以预先保存波束向量集合和频域向量集合；或者，终端设备和网络设备均可以预先保存空频分量矩阵集合；或者，终端设备可以预先保存波束向量集合和频域向量集合，网络设备可以预先保存空频分量矩阵集合；或者，终端设备可以预先保存空频分量矩阵集合，网络设备可以预先保存波束向量集合和频域向量集合。
由于波束向量集合和频域向量集合与空频分量矩阵集合之间可以互相转换，在确定了空频分量矩阵集合和波束向量矩阵的情况下，也可以推出频域向量集合，或者，在确定了空频分量矩阵集合和频域向量集合的情况下，也可以推出波束向量集合。因此，本申请对于终端设备和网络设备预先保存的向量集合的具体形式并不做限定。
下文中仅为便于理解本申请实施例，结合具体的向量集合形式和矩阵集合形式，基于上文列举的两种实现方式来说明终端设备生成第一指示信息的具体过程。
实现方式一、
终端设备可以根据预先保存的波束向量集合、频域向量集合以及预先确定的空频矩阵，确定M 1个空频向量对，进而确定T 1个空频分量矩阵。
假设终端设备所确定的N f个频域单元的预编码向量分别记作
Figure PCTCN2019110342-appb-000032
该N f个频域单元的预编码向量可以构建空频矩阵H，
Figure PCTCN2019110342-appb-000033
或，
Figure PCTCN2019110342-appb-000034
需要说明的是，上文列举的两种空频矩阵的形式之间可以相互转换。终端设备可以根据预先保存的向量集合确定需要构建的空频矩阵的形式，进而生成相应形式的空频矩阵例如，当终端设备预先保存波束向量集合和频域向量集合时，可以生成空频矩阵
Figure PCTCN2019110342-appb-000035
当终端设备预先保存空频分量矩阵集合时，可以生成空频矩阵
Figure PCTCN2019110342-appb-000036
在本实施例中，终端设备可以根据N f个频域单元的预编码向量构建维度为N s×N f的空频矩阵H，
Figure PCTCN2019110342-appb-000037
对于波束向量集合，在一种可能的设计中，该波束向量集合可以包括N s个波束向量。即，L 0＝N s。每个波束向量的维度可以为N s，且每个波束向量可以取自二维(2dimension， 2D)-DFT矩阵。其中，2D可以表示两个不同的方向，如，水平方向和垂直方向。
该N s个波束向量例如可以记作
Figure PCTCN2019110342-appb-000038
该L 0个波束向量可以构建矩阵B s，
Figure PCTCN2019110342-appb-000039
在另一种可能的设计中，该波束向量集合可以通过过采样因子O s扩展为O s×N s个波束向量。此情况下，该波束向量集合可以包括O s个子集，每个子集可以包括N s个波束向量。即，L 0＝O s×N s。该波束向量集合中的每个波束向量的维度可以为N s，且每个波束向量可以取自过采样2D-DFT矩阵。其中，过采样因子O s为正整数。具体地，O s＝O 1×O 2，O 1可以是水平方向的过采样因子，O 2可以是垂直方向的过采样因子。O 1≥1，O 2≥1，O 1、O 2不同时为1，且均为整数。
波束向量集合中的第o s(0≤o s≤O s-1且o s为整数)个子集中的N s个波束向量例如可以分别记作
Figure PCTCN2019110342-appb-000040
则基于该第o s个子集中的N s个波束向量可以构造矩阵
Figure PCTCN2019110342-appb-000041
对于频域向量集合，在一种可能的设计中，该频域向量集合可以包括N f个频域向量。即，K 0＝N f。每个频域向量的维度可以为N f，且每个频域向量可以取自DFT矩阵。
该N f个频域向量例如可以记作
Figure PCTCN2019110342-appb-000042
该N f个频域向量可以构建矩阵B f，
Figure PCTCN2019110342-appb-000043
在另一种可能的设计中，该频域向量集合可以通过过采样因子O f扩展为O f×N f个频域向量。此情况下，该频域向量集合可以包括O f个子集，每个子集可以包括N f个频域向量。即，K 0＝O f×N f。该频域向量集合中的每个频域向量的维度可以为N f，且每个频域向量可以取自过采样DFT矩阵。其中，过采样因子O f为正整数。
频域向量集合中的第o f(0≤o f≤O f-1且o s为整数)个子集中的N f个频域向量例如可以分别记作
Figure PCTCN2019110342-appb-000044
则基于该第o f个子集中的N s个波束向量可以构造矩阵
Figure PCTCN2019110342-appb-000045
下面分别说明在考虑和不考虑过采样率的情况下，终端设备确定和指示T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体方法。
若不考虑过采样率，终端设备可以通过下文所示的步骤1-i至步骤1-v来确定T 1个空频分量矩阵及各空频分量矩阵的加权系数。
步骤1-i、终端设备可以基于上述空频矩阵H、波束向量集合构造的矩阵和频域向量集合构造的矩阵，确定加权系数矩阵。
若不考虑过采样率，则波束向量集合可以包括N s个波束向量，所构建的矩阵为B s；频域向量集合可以包括N f个频域向量，所构建的矩阵为B f。终端设备可以通过W＝B s *HB f来确定矩阵W。该矩阵W可以称为加权系数矩阵，其维度可以是N s×N f。其中，矩阵W中的N s个行可以与波束向量集合(或者由波束向量集合构建的矩阵B s)中的N s个波束向量对应；该矩阵W中的N f个列可以与频域向量集合(或者由频域向量集合构建的矩阵B f)中的N f个频域向量对应。该矩阵中的每个系数可对应于一个空频向量对。
步骤1-ii、终端设备可以从波束向量集合中选择较强的L 1个波束向量，并从频域向量集合中选择较强的K 1个频域向量。
终端设备可以分别对该矩阵W中的N s个行分别取模，根据各行的模长大小，确定模 较大的L 1个行。该L 1个行在矩阵W中的行序号可以为较强的L 1个波束向量在波束向量集合中的序号或在B s中的列序号。终端设备还可以分别对该矩阵W中的N f个列取模，根据各列的模长大小，确定模较大的K 1个列。该K 1个列在矩阵W中的列序号可以为较强的K 1个频域向量在频域向量集合的序号或在B f中的列序号。
应理解，上文所描述的终端设备从波束向量集合中确定较强的L 1个行、从频域向量集合中确定较强的K 1个列的具体方法仅为便于理解而示例，不应对本申请构成任何限定。终端设备从波束向量集合中确定较强的L 1个行、从频域向量集合中确定较强的K 1个列的具体方法可以参考现有技术，为了简洁，这里不做详细说明。
还应理解，上文通过加权系数矩阵确定L 1个波束向量和K 1个频域向量的方法仅为便于理解而示出的一种可能的实现方式，并不代表终端设备在确定L 1个波束向量和K 1个频域向量的过程中一定生成了该加权系数矩阵。例如，将各频域单元的预编码向量分别投影至波束向量集合中的各个波束向量和频域向量集合中的各个频域向量，可以得到多个投影值组成的数组集合，该数组集合中的各元素可以是上述加权系数矩阵中的各行(或各列)元素依次连接而成。
步骤1-iii、终端设备可以基于从波束向量集合中确定的较强的L 1个波束向量和从频域向量集合中确定较强的K 1个频域向量，组合得到M 1个空频向量对。
其中，该M 1个空频向量对中的每个空频向量对可以包括一个波束向量和一个频域向量。每个空频向量对中的波束向量可以取自上述L 1个波束向量，每个空频向量对中的频域向量可以取自上述K 1个频域向量。该L 1个波束向量中的一个波束向量和K 1个频域向量中的一个频域向量可组合得到唯一的一个空频向量对。或者说，该M 1个空频向量对中的每个空频向量对由L 1个波束向量中的一个波束向量和K 1个频域向量中的一个频域向量唯一确定。任意两个空频向量对包含的波束向量和频域向量中至少有一项是不同的。因此，由上述L 1个波束向量和K 1个频域向量可以确定得到L 1×K 1个空频向量对。即，M 1＝L 1×K 1。
另一方面，上述波束向量集合中的N s个波束向量和频域向量集合中的N f个频域向量可以组合得到N s×N f个空频向量对。因此，上述M 1个空频向量对可以视为该N s×N f个空频向量对的子集，且M 1＜N s×N f。也就是说，终端设备可以在波束向量集合和频域向量集合组合得到的空频向量对集合的子集中确定空频用于线性加权的T 1个空频向量对，或者说，终端设备需要上报的T 1个空频向量对。换句话说，被选择的T 1个空频向量对选自由L 1个波束向量和K 1个频域向量组合得到的M 1个空频向量对。
步骤1-iv、终端设备可以从该M 1个空频向量对中确定T 1个空频向量对。该T 1个空频向量对可用于确定T 1个空频分量矩阵。
终端设备可以从该M 1个空频向量对中选择较强的T 1个空频向量对，以生成T 1个空频分量矩阵。该较强的T 1个空频向量对可以是M 1个空频向量对中加权系数的模较大的空频向量对。即，被选择的T 1个空频向量对中任意一个空频向量对的加权系数的模长大于或等于剩余M 1-T 1个空频向量对中任意一个空频向量对的加权系数的模长。
基于上文描述所确定的矩阵W中较强的L 1个行和较强的K 1个列，终端设备可以确定L 1×K 1(也就是M 1)个加权系数。该M 1个加权系数可以与M 1个空频向量对一一对应。终端设备可以从该M 1个加权系数中确定模较大的T 1个加权系数。被选择的T 1个加权系数中的任意一个加权系数的模长大于或等于剩余M 1-T 1个加权系数中的任意一个加权系数 的模长。该T 1个加权系数可以是T 1个空频分量矩阵的加权系数。
该T 1个加权系数在L 1×K 1个加权系数中的位置可用于确定T 1个空频向量对所分别包含的T 1个波束向量和T 1个频域向量。
在一种实现方式中，终端设备可以从矩阵W中抽取出上述模较大的L 1个行和模较大的K 1个列可以得到维度为L 1×K 1的矩阵。为方便区分和说明，将该维度为L 1×K 1的矩阵记作W'。该矩阵W'可以视为矩阵W的子矩阵。终端设备可以将该矩阵W'中的各元素分别取模，以选择出模较大的T 1个元素。该T 1个元素在该矩阵W'中的位置可用于确定T 1个空频向量对所包含的L 1个波束向量中的位置和频域向量在K 1个频域向量中的位置。具体地，该T 1个元素在该矩阵W'中的行序号可以为被选择T 1个波束向量在L 1个波束向量中的序号，该T 1个元素在该矩阵W'中的列序号可以为被选择的T 1个频域向量在K 1个频域向量中的序号。
在另一种实现方式中，终端设备可以将矩阵W中抽取出上述模较大的L 1个行和模较大的K 1个列按照预先定义的规则依次排列，例如按照先行后列的顺序或者先列后行的顺序依次排列，得到L 1×K 1个加权系数构成的数组。终端设备可以将该数组中的各元素分别取模，以选择模较大的T 1个元素。该T 1个元素在该数组中的位置可用于确定T 1个空频向量对所包含的波束向量在L 1个波束向量中的位置和频域向量在K 1个频域向量中的位置。
该T 1个空频向量对可用于确定T 1个空频分量矩阵。假设终端设备在步骤1-iv中选择的T 1个空频向量对中的T 1个波束向量分别记作
Figure PCTCN2019110342-appb-000046
T 1个空频向量对中的T 1个频域向量分别记作
Figure PCTCN2019110342-appb-000047
则空频分量矩阵可以为
Figure PCTCN2019110342-appb-000048
或
Figure PCTCN2019110342-appb-000049
t 1＝0，1，……，T 1-1。其中，
Figure PCTCN2019110342-appb-000050
可以是维度为N s×N f的矩阵，
Figure PCTCN2019110342-appb-000051
可以是长度为N s×N f的向量。
应理解，终端设备并不一定会基于上文所述的T 1个空频向量对生成T 1个空频分量矩阵。这里仅为便于理解，示出了空频向量对与空频分量矩阵之间的几种可能的转换方式。
步骤1-v、终端设备生成第一指示信息，以指示L 1个波束向量、K 1个频域向量和T 1个空频向量对。
基于上述步骤1-i至步骤1-iv所确定的L 1个波束向量、K 1个频域向量以及T 1个空频向量对，该第一指示信息可以包括L 1个波束向量在波束向量集合中的位置信息、K 1个频域向量在频域向量集合中的位置信息，以及用于指示T 1个空频向量对的信息。
可选地，该第一指示信息在用于指示该L 1个波束向量时，可以指示该L 1个波束向量的组合在波束向量集合中的索引。例如，协议可以预先定义多个波束向量的多种组合，每种组合可对应一个索引。该L 1个波束向量可以为该多种组合中的一种，或者，接近于该多种组合中的一种，该第一指示信息可通过指示该组合的索引的方式来指示该L 1个波束向量。即，该L 1个波束向量在波束向量集合中的位置信息可以是该L 1个波束向量的组合在波束向量集合中的索引。此时，该终端设备可以通过
Figure PCTCN2019110342-appb-000052
个比特来指示波束向量集合中的L 1个波束向量。
Figure PCTCN2019110342-appb-000053
表示向上取整。
可选地，该第一指示信息在用于指示K 1个频域向量时，可以指示该K 1个频域向量的组合在频域向量集合中的索引。例如，协议可以预先定义多个波束向量的多种组合，每种组合可对应一个索引。该K 1个频域向量可以为该多种组合中的一种，或者，接近于该多种组合中的一种，该第一指示信息可通过指示该组合的索引的方式来指示该K 1个频域向 量。即，该K 1个频域向量在频域向量集合中的位置信息可以是该K 1个频域向量的组合在频域向量集合中的索引。此时，该终端设备可以通过
Figure PCTCN2019110342-appb-000054
个比特来指示频域向量集合中的K 1个频域向量。
应理解，通过指示L 1个波束向量的组合的索引来指示L 1个波束向量以及通过指示K 1个频域向量的组合的索引来指示K 1个频域向量的方法仅为一种可能的实现方式，不应对本申请构成任何限定。例如，该第一指示信息指示该L 1个波束向量时，也可用于指示该L 1个波束向量中的每个波束向量在该波束向量集合中的索引，或者，该第一指示信息指示该K 1个频域向量时，也可用于指示该K 1个频域向量中的每个频域向量在频域向量集合中的索引。本申请对于指示L 1个波束向量和K 1个频域向量的具体方式不作限定。
可选地，该第一指示信息可以通过以下任意一种方式来指示T 1个空频向量对：
方式1、通过位图(bitmap)来指示M 1个空频向量对中的T 1个空频向量对；或
方式2、指示T 1个空频向量对的组合在M 1个空频向量对中的索引；或
方式3、指示T 1个空频向量对中每个空频向量对所包含的波束向量在L 1个波束向量中的位置和频域向量在K 1个频域向量中的位置；或
方式4、指示T 1个空频向量对中的每个空频向量对在M 1个空频向量对中的位置。
下面结合上述四种方式详细说明该第一指示信息指示T 1个空频向量对的具体方法。
在方式1中，终端设备可以通过M 1个比特的位图来指示M 1个空频向量对中的T 1个空频向量对。该位图中的每个比特可以对应M 1个空频向量对中的一个空频向量对。每个比特位可用于指示所对应的空频向量对是否被选择作为T 1个空频向量对，或者说，每个比特位可用于指示所对应的空频向量对是否属于上述T 1个空频向量对。例如，当某一比特置“0”时，表示所对应的空频向量对不属于该T 1个空频向量对；当某一比特置“1”时，表示所对应的空频向量对属于该T 1个空频向量对。
其中，位图中M 1个比特与M 1个空频向量对的对应关系与M 1个空频向量对中波束向量与频域向量的组合方式相对应。例如，该M 1个比特所对应的M 1个空频向量对可以按照先遍历K 1个频域向量、再遍历L 1个波束向量的顺序排列，也可以按照先遍历L 1个波束向量、再遍历K 1个频域向量的顺序排列。
假设波束向量集合中被选择的L 1个波束向量记作
Figure PCTCN2019110342-appb-000055
频域向量集合中被选择的K 1个频域向量记作
Figure PCTCN2019110342-appb-000056
若先遍历K 1个频域向量再遍历L 1个波束向量，则该M 1个空频向量对的排列顺序可以是
Figure PCTCN2019110342-appb-000057
Figure PCTCN2019110342-appb-000058
共M 1个空频向量对。为了简洁，这里不再一一列举。该位图中的M 1个比特与上述M 1个空频向量对一一对应。
该位图的M 1个比特中的第0至第K 1-1个比特与空频向量对
Figure PCTCN2019110342-appb-000059
Figure PCTCN2019110342-appb-000060
一一对应；第K 1至第2K 1-1个比特与空频向量对
Figure PCTCN2019110342-appb-000061
Figure PCTCN2019110342-appb-000062
一一对应；以此类推，第(L 1-1)×K 1至第L 1×K 1-1个比特与空频向量对
Figure PCTCN2019110342-appb-000063
Figure PCTCN2019110342-appb-000064
一一对应。
若先遍历L 1个波束向量再遍历K 1个频域向量，则该M 1个空频向量对的排列顺序可以是
Figure PCTCN2019110342-appb-000065
Figure PCTCN2019110342-appb-000066
共M 1个空频向量对。为了简洁，这里不再一一列举。该位图中的M 1 个比特与上述M 1个空频向量对一一对应。
该位图的M 1个比特中的第0至第L 1-1个比特与空频向量对
Figure PCTCN2019110342-appb-000067
Figure PCTCN2019110342-appb-000068
一一对应；第L 1至第2L 1-1个比特与空频向量对
Figure PCTCN2019110342-appb-000069
Figure PCTCN2019110342-appb-000070
一一对应；以此类推，第L 1×(K 1-1)至第L 1×K 1-1个比特与空频向量对
Figure PCTCN2019110342-appb-000071
Figure PCTCN2019110342-appb-000072
一一对应。
应理解，上文列举的M 1个比特与M 1个空频向量对的一一对应关系仅为示例，不应对本申请构成任何限定。本申请对于M 1个比特与M 1个空频向量对的对应关系并不做限定。此外，本申请并不限定M 1个空频向量对的排列方式，上文仅为便于说明M 1个比特与M 1个空频向量对的一一对应关系，而示出了与M 1个比特一一对应的M 1个空频向量对的两种可能的排列方式。
在方式2中，终端设备可以通过该T 1个空频向量对的组合在M 1个空频向量对中的索引来指示该T 1个空频向量对。也就是说，终端设备可以根据上述L 1个波束向量和K 1个频域向量组合得到的M 1个空频向量预先确定多个空频向量对的多种组合，每种组合可对应一个索引。该T 1个空频向量对可以为该多种组合中的一种，或者，接近该多种组合中的一种。该第一指示信息可通过指示该组合的索引的方式来指示该T 1个空频向量对。因此，该终端设备可以通过
Figure PCTCN2019110342-appb-000073
个比特来指示M 1个空频向量对中的T 1个空频向量对。
在方式3中，终端设备可以分别指示用于组合得到该T 1个空频向量对的T 1个波束向量在L 1个波束向量中的位置和T 1个频域向量在K 1个频域向量中的位置。对于每个波束向量，该终端设备可以通过
Figure PCTCN2019110342-appb-000074
个比特来指示其在L 1个波束向量中的位置；对于每个频域向量，该终端设备可以通过
Figure PCTCN2019110342-appb-000075
个比特来指示其在K 1个频域向量中的位置。
在方式4中，终端设备可以分别指示该T 1个空频向量对中的每个空频向量对在M 1个空频向量对中的位置。这里所说的每个空频向量对在M 1个空频向量对中的位置，可以理解为每个空频向量对在M 1个空频向量对中的相对位置，或者说，局部(local)位置。例如，对应每个空频向量对，该终端设备可以指示其在M 1个空频向量对中的索引。此时，终端设备可以通过
Figure PCTCN2019110342-appb-000076
个比特来指示其在M 1个空频向量对中的索引。
在上文列举的几种方式中，空频向量对可以表现为空频分量矩阵(包括矩阵形式或向量形式)的形式，也可以表现为波束向量和频域向量组合得到的向量对的形式。本申请对此不作限定。
可以看到，上文列举的用于指示T 1个空频向量对的方法中，终端设备通过该T 1个空频向量对在M 1个空频向量对中的相对位置(例如相对索引或相对编号等)来指示该T 1个空频向量对，或者，通过该T 1个空频向量对在L 1个波束向量和K 1个频域向量中的相对位置(例如相对索引或相对编号等)来指示T 1个空频向量对。由于将选择的范围缩小了，用于指示T 1个空频向量对而带来的开销也得以降低。
应理解，上文列举的用于指示T 1个空频向量对的方法仅为示例，不应对本申请构成任何限定。例如，与每个波束向量对应的空频向量对的个数为K 1，在L 1个波束向量对应的L 1×K 1个空频向量对中，终端设备可以基于每个波束向量，上报被选择用作加权求和以确定预编码向量的空频分量对。如，若各波束向量对应的空频向量对中被选择的空频向量对的个数相同，即为T 1/L 1个，则终端设备可以分别基于每个波束向量指示该T 1/L 1个被选择的空频向量对的组合在与同一波束向量对应的K 1个空频向量对中的索引；若各波束向量对应的K 1个空频向量对中被选择的空频向量对的个数彼此互不相同，终端设备可以分 别基于每个波束向量上报被选择的空频向量对的个数，以及被选择的空频向量对的组合在与同一波束向量对应的K 1个空频向量对中的索引。
应理解，终端设备指示T 1个空频向量对的方法并不仅限于本申请所列举的方法，为了简洁，这里不再一一举例说明。本申请对于终端设备指示T 1个空频向量对的具体方法不作限定。
可选地，该第一指示信息还包括T 1个空频向量对的加权系数的量化信息。
终端设备可以基于步骤1-iv确定的T 1个空频向量对的加权系数，生成该T 1个空频向量对的加权系数的量化信息。
可选地，终端设备可以根据在步骤1-iv中确定的T 1个空频向量对的加权系数，通过归一化方式来指示该T 1个加权系数。
具体地，终端设备可以从该T 1个加权系数中确定模最大的加权系数(例如记作最大加权系数)，并指示该最大加权系数在矩阵W'中的位置。然后，终端设备可以进一步指示剩余的T 1-1个加权系数中的每个加权系数相对于该最大加权系数的相对值。终端设备例如可以通过各相对值的量化值的索引来指示上述剩余的T 1-1个加权系数。例如，码本中可以预先定义多个量化值与多个索引的一一对应关系，终端设备可以基于该一一对应关系，将上述各加权系数相对于最大加权系数的相对值反馈给网络设备。因此，由终端设备所反馈的加权系数可能与步骤1-iv中所确定的加权系数可能相同，也可能相近，故称为加权系数的量化值。用于指示该加权系数的量化值的信息可以称为加权系数的量化信息。该量化信息例如可以是量化值的索引。
应理解，基于归一化方式来指示各加权系数仅为一种可能的实现方式，而不应对本申请构成任何限定。终端设备指示加权系数的具体方法可以参考现有技术的方法，为了简洁，这里不再赘述。
需要说明的是，这里提到的归一化，可以是以每个极化方向、每个传输层或所有传输层为单位来确定最大加权系数，从而在每个极化方向、每个传输层或所有传输层等不同的范围内进行归一化。
还应理解，上文所述的终端设备确定L 1个波束向量、K 1个频域向量、T 1个空频向量对以及各空频向量对的加权系数的方法仅为示例，而不应对本申请构成任何限定。
例如，可选地，终端设备也可以对该L 1个波束向量单独确定和反馈宽带幅度系数。此情况下，该第一指示信息还可进一步包括L 1个波束向量的宽带幅度系数的量化信息。
可选地，终端设备可以先选择L 1个波束向量，然后对每个波束向量分别选择K 1个频域向量，并可以进一步确定每个波束向量及其所对应的各频域向量所构成的空频向量对所对应的加权系数，即，共L 1×K 1个加权系数。
此情况下，该第一指示信息具体用于指示L 1个波束向量中的每个波束向量和每个波束向量对应的频域向量。该可能的设计可以应用于被选择的L 1个波束向量中至少有两个波束向量对应的频域向量不同的场景中，尤其可以应用于被选择的L个波束向量中至少有两个波束向量对应的频域向量不同，且被选择的波束向量较少，或者说，L 1值较小(即空域稀疏性较好)的场景中。可选地，该第一指示信息还用于指示每个波束向量对应的频域向量的个数。可选地，至少两个波束向量对应的频域向量的个数不同。
若对每个波束向量分别选择频域向量，且至少两个波束向量选择的频域向量不同，终 端设备在从M 1个空频向量对中确定了T 1个空频向量对之后，可以基于每个波束向量，分别指示被选择的空频向量对。如，基于每个波束向量，终端设备可以采用上文所列举的方式1至方式4中的任意一种来指示。
若至少两个波束向量对应的频域向量的个数不同，终端设备在采用上述方式2指示时，还可进一步指示与每个波束向量对应的频域向量的个数。
若至少两个波束向量对应的空频向量对中，被选择的空频向量对的个数不同，终端设备在采用上述方式2指示时，还可进一步指示与每个波束向量对应的空频向量对中被选择的空频向量对的个数。
可选地，终端设备也可以先选择K个频域向量，然后对每个频域向量分别选择L个波束向量，并可以进一步确定与每个频域向量及其所对应的各波束向量所构成的空频向量对所对应的加权系数，即，共L×K个加权系数。
此情况下，该第一指示信息具体用于指示K个频域向量中的每个频域向量和每个频域向量对应的波束向量。该可能的设计可以应用于被选择的K个频域向量中至少有两个频域向量对应的波束向量不同的场景中，尤其可以应用于被选择的K个频域向量中至少有两个频域向量对应的波束向量不同，且被选择的频域向量较少，或者说，K值较小(即频域稀疏性较好)场景中。可选地，该第一指示信息还用于指示每个频域向量对应的波束向量的个数。可选地，至少两个频域向量对应的波束向量的个数不同。
若对每个频域向量分别选择波束向量，且至少两个频域向量选择的波束向量不同，终设备在从M 1个空频向量对中确定了T 1个空频向量对之后，可以基于每个频域向量，分别指示被选择的空频向量对。如，基于每个频域向量，终端设备可以采用上文所列举的方式1至方式4中的任意一种来指示。
若至少两个频域向量对应的波束向量的个数不同，终端设备在采用上述方式2指示时，还可进一步指示与每个频域向量对应的波束向量的个数。
若至少两个波束向量对应的空频向量对中，被选择的空频向量对的个数不同，终端设备在采用上述方式2指示时，还可进一步指示与每个波束向量对应的空频向量对中被选择的空频向量对的个数。
此外，如前所述，上述L 1个波束向量可以是波束向量集合中的部分或全部波束向量，即，L 1≤N s；K 1个频域向量可以是频域向量集合中的部分或全部频域向量，即，K 1≤N f。但L 1、K 1、N s和N f不同时满足L 1＝N s，K 1＝N f。
当L 1＝N s时，该第一指示信息可以仅指示K 1个频域向量以及T 1个空频向量对，而不通过额外的信息来指示L 1个波束向量。换句话说，该第一指示信息用于指示K 1个频域向量和T 1个空频向量对。网络设备在接收到该第一指示信息时，可以默认该L 1个波束向量即为波束向量集合的全集。
当K 1＝N f时，该第一指示信息可以仅指示L 1个波束向量以及T 1个空频向量对，而不通过额外的信息来指示K 1个频域向量。L 1个波束向量和T 1个空频向量对网络设备在接收到该第一指示信息时，可以默认该K 1个频域向量即为频域向量集合的全集。
还应理解，该T 1个空频向量对的加权系数的量化信息可以携带在第一指示信息中，也可以通过额外的信息来携带，本申请对此不作限定。
若考虑过采样率，则波束向量集合和频域向量集合中所包含的向量可能存在如下三种 可能的情况：
情况一、波束向量集合通过过采样因子O s扩展为O s×N s个波束向量，频域向量集合通过过采样因子O f扩展为O f×N f个频域向量；
情况二、波束向量集合通过过采样因子O s扩展为O s×N s个波束向量，频域向量集合包括N f个频域向量；以及
情况三、波束向量集合包括N s个波束向量，频域向量集合通过过采样因子O f扩展为O f×N f个频域向量。
对于上述三种可能的情况，终端设备处理的方式可以是相同的。下面就以情况一为例详细说明终端设备确定T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体过程。
终端设备具体可以通过下文所示的步骤2-i至步骤2-vi来确定T 1个空频分量矩阵及各空频分量矩阵的加权系数。
步骤2-i、终端设备可以基于上述空频矩阵H、波束向量集合构造的矩阵和频域向量集合构造的矩阵，确定加权系数矩阵。
若波束向量集合通过过采样因子O s扩展为O s×N s个波束向量，频域向量集合通过过采样因子O f扩展为O f×N f个频域向量，则波束向量集合可以包括O s个子集，基于第o s个子集可以构建矩阵
Figure PCTCN2019110342-appb-000077
频域向量集合可以包括O f个子集，基于第o f个子集可以构建矩阵
Figure PCTCN2019110342-appb-000078
终端设备可以通过
Figure PCTCN2019110342-appb-000079
来确定矩阵
Figure PCTCN2019110342-appb-000080
该矩阵
Figure PCTCN2019110342-appb-000081
可以视为与第o s个子集、第o f个子集对应的加权系数矩阵，其维度可以是N s×N f。其中，矩阵
Figure PCTCN2019110342-appb-000082
中的N s个行可以与波束向量集合中的第o s个子集(或者由第o s个子集构建的矩阵
Figure PCTCN2019110342-appb-000083
)中的N s个波束向量对应；该矩阵
Figure PCTCN2019110342-appb-000084
中的N f个列可以与频域向量集合中的第o f个子集(或者由第o f个子集构建的矩阵
Figure PCTCN2019110342-appb-000085
)中的N f个频域向量对应。该矩阵
Figure PCTCN2019110342-appb-000086
中的每个系数可对应于一个空频向量对。
步骤2-ii、终端设备可以基于波束向量集合中的O s个子集、频域向量集合中的O f个子集，确定O s×O f组空频向量对，每组空频向量对包括T 1个空频向量对。
具体地，终端设备可以分别对o s在0至O s-1中遍历取值、对o f在0至O f-1中遍历取值，重复执行下述步骤，以确定O s×O f组空频向量对：根据矩阵
Figure PCTCN2019110342-appb-000087
确定较强的L 1个行和较强的K 1个列，进而确定第o s个子集中较强的L 1个波束向量和第o f个子集中较强的K 1个频域向量。该L 1个波束向量和K 1个频域向量可以组合得到M 1个空频向量对。终端设备可进一步根据该矩阵
Figure PCTCN2019110342-appb-000088
中较强的L 1个行和较强的K 1个列所确定的L 1×K 1确定较强的T 1个空频向量对。
由于在上文步骤1-ii至步骤1-v中已经详细说明了终端设备根据加权系数矩阵W确定较强的L 1个波束向量和较强的K 1个频域向量、进而确定较强的T 1个空频向量对的具体过程，为了简洁，这里不再赘述。
步骤2-iii、终端设备可以基于O s×O f组空频向量对的加权系数，选择最强的一组空频向量对，以确定T 1个空频向量对及各空频向量对的加权系数。
终端设备可以根据步骤2-ii中确定的O s×O f组空频向量对，确定最强的一组空频向量对，该最强的一组空频向量对中的T 1个空频向量对可以用于生成T 1个空频分量矩阵。例如，终端设备可以分别对O s×O f组空频向量对中的每组空频向量对的加权系数求模长之和，选择模长之和最大的一组空频向量对来生成T 1个空频分量矩阵。这组空频向量对的 加权系数即为该T 1个空频分量矩阵的加权系数。
终端设备基于T 1个空频向量对生成T 1个空频分量矩阵的具体过程在上文步骤1-v中已经做了详细说明，为了简洁，这里不再赘述。
由于该T 1个空频向量对所包含的T 1个波束向量来自波束向量集合中的同一个子集，终端设备在确定该T 1个空频向量对的同时，也就可以确定该T 1个波束向量所来自的子集，也就可以确定该子集中较强的的L 1个波束向量。
同理，由于该T 1个空频向量对所包含的T 1个频域向量也来自频域向量集合中的同一个子集，终端设备在确定该T 1个空频向量对的同时，也就可以确定该T 1个频域向量所来自的子集，也就可以确定该子集中较强的K 1个频域向量。
应理解，上文通过加权系数矩阵确定L 1个波束向量和K 1个频域向量的方法仅为便于理解而示出的一种可能的实现方式，并不代表终端设备在确定L 1个波束向量和K 1个频域向量的过程中一定生成了该加权系数矩阵。例如，将各频域单元的预编码向量分别投影至波束向量集合的任意子集中的各个波束向量和频域向量集合的任意子集中的各个频域向量，可以得到多个投影值组成的数组集合，该数组集合中的各元素可以是上述加权系数矩阵中的各行(或各列)元素依次连接而成。
步骤2-iv、终端设备生成第一指示信息，以指示L 1个波束向量、K 1个频域向量和T 1个空频向量对。
基于上述步骤2-i至步骤2-iii中确定的L 1个波束向量、K 1个频域向量以及T 1个空频向量对，该第一指示信息可以包括L 1个波束向量在波束向量集合中的位置信息、K 1个频域向量在频域向量集合中的位置信息，以及用于指示T 1个空频向量对的信息。
可选地，该第一指示信息在用于指示该L 1个波束向量时，具体可用于指示该L 1个波束向量所属的子集以及该L 1个波束向量在该子集中的索引；或者，具体可用于指示该L 1个波束向量所属的子集以及该L 1个波束向量的组合在该子集中的索引。
可选地，该第一指示信息在用于指示该K 1个频域向量时，具体可用于指示该K 1个频域向量所属的子集以及该K 1个频域向量在该子集中的索引；或者，具体可用于指示该K 1个频域向量所属的子集以及该K 1个频域向量的组合在该子集中的索引。
该第一指示信息在用于指示该T 1个空频向量对时，其具体的指示方式可以是上文中所述的方式1至方式4中的任意一种。由于上文已经对方式1至方式4做了详细说明，为了简洁，这里不再赘述。
可选地，该第一指示信息还包括T 1个空频向量对的加权系数的量化信息。
应理解，该T 1个空频向量对的加权系数的量化信息可以携带在第一指示信息中，也可以通过额外的信息来携带，本申请对此不作限定。
在情况二中，由于波束向量集合通过过采样因子O s扩展为O s×N s个波束向量，频域向量集合包括N f个频域向量，终端设备可以构建矩阵
Figure PCTCN2019110342-appb-000089
和矩阵B f。此后，终端设备可以对o s在0至O s-1中在遍历取值，通过
Figure PCTCN2019110342-appb-000090
来确定T 1个空频分量矩阵及各空频分量矩阵的加权系数。
在情况三中，由于波束向量集合包括包括N f个频域向量，频域向量集合通过过采样因子O f扩展为O f×N f个频域向量，终端设备可以构建矩阵B s和矩阵
Figure PCTCN2019110342-appb-000091
Figure PCTCN2019110342-appb-000092
此后，终端设备可以对o f在0至O f-1中在遍历取值，通过
Figure PCTCN2019110342-appb-000093
来确定 T 1个空频分量矩阵及各空频分量矩阵的加权系数。
终端设备在情况二和情况三中确定T 1个空频分量矩阵及各空频分量矩阵的加权系数的具体方法与情况一中所描述的具体方法相似，为了简洁，这里不再赘述。
基于上述技术方案，终端设备可以生成第一指示信息，以指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量和T 1个空频分量矩阵。
应理解，上文所描述的波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量和T 1个空频分量矩阵的方法仅为示例，而不应对本申请构成任何限定。如前所述，上述L 1个波束向量可以是波束向量集合中的部分或全部波束向量，即，L 1≤L 0；K 1个频域向量可以是频域向量集合中的部分或全部频域向量，即，K 1≤K 0。但L 1、K 1、L 0和K 0不同时满足L 1＝L 0，K 1＝K 0。
如前所述，上述L 1个波束向量可以是波束向量集合中的部分或全部波束向量，即，L 1≤L 0；K 1个频域向量可以是频域向量集合中的部分或全部频域向量，即，K 1≤K 0。但L 1、K 1、L 0和K 0不同时满足L 1＝L 0，K 1＝K 0。
在分别对波束向量集合和频域向量集合过采样的情况下，该L 1个波束向量可以是波束向量集合中的一个子集(例如，一个正交组)或子集中的部分波束向量；该K 1个频域向量也可以是频域向量集合中的一个子集(例如，一个正交组)或子集中的部分频域向量。但，L 1、K 1、L 0和K 0不同时满足L 1＝L 0，K 1＝K 0。
可选地，当L 1个波束向量为波束向量集合中的一个子集时，该第一指示信息在用于指示该L 1个波束向量时，可以仅指示该波束向量集合中被选择的子集，而不通过额外的信息来指示该L 1个波束向量。网络设备在接收到该第一指示信息时，可以默认该L 1个波束向量即为该波束向量集合中被选择的子集。
可选地，当K 1个频域向量为频域向量集合中的一个子集时，该第一指示信息在用于指示该K 1个频域向量时，可以仅指示该频域向量集合中被选择的子集，而不通过额外的信息来指示该K 1个频域向量。网络设备在接收到该第一指示信息时，可以默认该K 1个频域向量即为该频域向量集合中被选择的子集。
可选地，当L 1个波束向量为波束向量集合的全集时，即L 1＝L 0时，该第一指示信息可以仅指示K 1个频域向量和T 1个空频向量对，而不通过额外的信息来指示该L 1个波束向量。网络设备在接收到该第一指示信息时，可以默认该L 1个波束向量即为波束向量集合的全部波束向量。可选地，当K 1个频域向量为频域向量集合的全集时，即K 1＝K 0时，该第一指示信息可以仅指示L 1个波束向量和T 1个空频向量对，而不通过额外的信息来指示该K 1个频域向量。网络设备在接收到该第一指示信息时，可以默认该K 1个频域向量即为频域向量集合中的全部频域向量。
应理解，在上文所列举的在不同的情况下第一指示信息所指示的具体内容仅为示例，不应对本申请构成任何限定。例如，在终端设备也可以对上文列举的情况不作区分，直接指示L 1个波束向量、K 1个频域向量和T 1个空频向量对。
实现方式二、
终端设备可以根据预先保存的空频分量矩阵集合以及预先确定的空频矩阵，确定M 1个空频分量矩阵，进而确定T 1个空频分量矩阵。
在本实施例中，终端设备预先保存的空频分量矩阵集合中的各空频分量矩阵可以是维 度为N s×N f的矩阵，也可以是长度为N s×N f的向量。下面将分别结合这两种情况详细说明终端设备确定T 1个空频分量矩阵的具体过程。
情况A、空频分量矩阵集合中的各空频分量矩阵是长度为N s×N f的向量。
假设终端设备所确定的N f个频域单元的预编码向量分别记作
Figure PCTCN2019110342-appb-000094
在情况A中，终端设备可以根据该N f个频域单元的预编码向量构建长度为N s×N f的空频矩阵H，
Figure PCTCN2019110342-appb-000095
下文中为方便区分，将该长度为N s×N f的空频矩阵称为空频向量。
下面详细说明空频分量矩阵集合。
在一种可能的设计中，该空频分量矩阵集合可以包括N s×N f个空频分量矩阵。每个空频分量矩阵可以是长度为N s×N f的向量。下文中为方便区分和说明，将该长度为N s×N f的空频分量矩阵称为空频分量向量。相应地，上述空频分量矩阵集合可以称为空频分量向量集合。该空频分量向量集合中选择的T 1个空频分量向量的加权和可以构建得到空频向量。该T 1个空频分量向量的加权和构建得到的空频向量可以与上文终端设备所确定的空频向量相同或相近。
在本申请实施例中，该空频分量向量集合中的每个空频分量向量可以是由波束向量集合中的一个波束向量和频域向量集合中的一个频域向量唯一确定。换句话说，该空频分量向量集合中的任意两个空频分量向量不同，且任意两个空频分量向量所对应的波束向量和频域向量中至少有一项不同。
具体地，空频分量向量集合中的每个空频分量向量可以是波束向量集合中的一个波束向量和频域向量集合中的一个频域向量的克罗内克积，或者，也可以是频域向量集合中的一个频域向量和波束向量集合中的一个波束向量的克罗内克积。为了与上文所构建的空频向量相对应，该空频分量向量集合中的每个空频分量向量可以由频域向量和波束向量的克罗内克尔积唯一确定。
如前所述，若不考虑过采样率，波束向量集合可以包括N s个波束向量，频域向量集合可以包括N f个频域向量，则由该波束向量集合和频域向量集合可以确定N s×N f个空频分量向量。也就是说，该空频分量向量集合中可以包括N s×N f个空频分量向量。每个空频分量向量可以与一个波束向量和一个频域向量对应，或者说，每个空频分量向量可以与一个波束向量和一个频域向量组合得到的空频向量对相对应。
该空频分量向量集合中的每个空频分量向量可对应一个索引。该空频分量向量集合中的N s×N f个空频分量向量可以通过空频分量向量集合中的索引指示，也可以通过可用于生成空频分量向量的波束向量和频域向量分别在波束向量集合和频域向量集合中的索引指示。其中，M 1个空频分量向量在空频分量向量集合中的索引可以视为M 1个空频分量向量的一维索引，M 1个空频分量向量所包含的波束向量和频域向量分别在波束向量集合和频域向量集合中的索引可视为M 1个空频分量向量的二维索引。一维索引和二维索引之间可以根据预先定义的转换规则相互转换。
具体地，假设m为空频分量向量集合中的空频分量向量的索引，0≤m≤N s×N f-1且m为整数；n s为波束向量集合中的波束向量的索引，0≤n s≤N s-1，且n s为整数；n f为频域向量集合中的频域向量的索引，0≤n f≤N f-1，且n f为整数。
例如，N f个频域向量中第0个频域向量至第N f-1个频域向量分别与N s个波束向量中 的第0个波束向量的克罗内克积所确定的N f个空频分量向量可对应一维索引0至N f-1；N f个频域向量中第0个频域向量至第N f-1个频域向量分别与N s个波束向量中的第1个波束向量的克罗内克积所确定的N f个空频分量向量可对应一维索引N f至2N f-1；以此类推，N f个频域向量中第0个频域向量至第N f-1个频域向量分别与N s个波束向量中的第n s个波束向量的克罗内克积所确定的N f个空频分量向量可对应一维索引n s×N f至(n s+1)×N f-1。为方便区分和说明，可以将这种编号规则记作规则一。
因此，对于空频分量向量集合中的N s×N f个空频分量向量，假设第m个空频分量向量可通过波束向量集合中的第n s个波束向量和频域向量集合中的第n f个频域向量指示。其中，
Figure PCTCN2019110342-appb-000096
n s＝mod(m，N f)。
Figure PCTCN2019110342-appb-000097
表示向下取整，mod()表示求模。
对于波束向量集合中的N s个波束向量和频域向量集合中的N s个波束向量，波束向量集合中的n s个波束向量和频域向量集合中的第n f个频域向量所构成的空频分量向量可以通过第m个空频分量向量指示。其中，m＝n f+n s*N f。又例如，N f个频域向量中第0个频域向量分别与N s个波束向量中的第0个波束向量至第N s-1个波束向量的克罗内克积所确定的N s个空频分量向量可对应一维索引0至N s-1；N f个频域向量中第1个频域向量分别与N s个波束向量中的第0个波束向量至第N s-1个波束向量的克罗内克积所确定的N s个空频分量向量可对应一维索引N s至2N s-1；以此类推，N f个频域向量中第n f个频域向量分别与N s个波束向量中的第0个波束向量至第N s-1个波束向量的克罗内克积所确定的N s个空频分量向量可对应一维索引n f×N s至(n f+1)×N s-1。为方便区分和说明，可以将这种编号规则记作规则二。
因此，对于空频分量向量集合中的N s×N f个空频分量向量，第m(0≤m≤N s×N f-1)个空频分量向量可通过波束向量集合中的第n s个波束向量和频域向量集合中的第n f个频域向量指示。其中，
Figure PCTCN2019110342-appb-000098
n f＝mod(m，N s)。
对于波束向量集合中的N s个波束向量和频域向量集合中的N s个波束向量，波束向量集合中的n s个波束向量和频域向量集合中的第n f个频域向量所构成的空频分量向量可以通过第m个空频分量向量指示。其中，m＝n s+n f*N s。
应理解，上文列举的两种空频分量向量集合中各空频分量向量的编号方式仅为示例，不应对本申请构成任何限定。终端设备和网络设备可以根据预先约定的规则对空频分量向量集合中的各空频分量向量编号。网络设备和终端设备所定义的各空频分量向量与索引的对应关系是一致的。
若考虑过采样率，则空频分量向量集合中包含的向量可能存在如下三种可能的情况：
情况一、空频分量向量集合通过过采样因子O s和O f扩展为O s×N s×O f×N f个空频分量向量，则该空频分量向量集合可以包括O s×O f个子集，每个子集可以包括N s×N f个空频分量向量；
情况二、空频分量向量集合通过过采样因子O s扩展为O s×N s×N f个空频分量向量，则该空频分量向量集合可以包括O s个子集，每个子集可以包括N s×N f个空频分量向量；以及
情况三、空频分量向量集合通过过采样因子O f扩展为O f×N s×N f个空频分量向量，则该空频分量向量集合可以包括O f个子集，每个子集可以包括N s×N f个空频分量向量。
在上述三种情况中，过采样因子O s可以是波束向量集合的过采样因子，过采样因子 O f可以是频域向量集合的过采样因子。若将空频分量向量集合的过采样因子记为O c，则O c＝O s×O f，O c＞1，且为正整数。在存在过采样率的情况下，O s和O f不同时为1。
在考虑过采样率的情况下，波束向量集合中的每个子集可以分别与频域向量集合中的每个子集确定得到多组空频分量向量，每组空频分量向量中包括N s×N f个空频分量向量。每组空频分量向量中N s×N f个空频分量向量的索引的编号规则与上文中不考虑过采样率时的编号规则可以是相同的，为了简洁，这里不再赘述。
下面分别说明在考虑和不考虑过采样率的情况下，终端设备确定和指示T 1个空频分量向量以及各空频分量向量的加权系数的具体方法。
若不考虑过采样率，终端设备可以通过下文所示的步骤3-i至步骤3-iv确定T 1个空频分量向量以及各空频分量向量的加权系数。
步骤3-i、终端设备可以基于上述空频向量和空频分量向量集合确定加权系数矩阵。
若不考虑过采样率，则空频分量向量集合可以包括N s×N f个空频分量向量。终端设备可以将预先确定的空频向量分别投影至该N s×N f个空频分量向量。即，将N s×N f个空频分量向量中的每个空频分量向量的共轭转置分别乘以空频向量，以得到N s×N f个投影值。该N s×N f个投影值的排列顺序与空频分量向量集合中N s×N f个空频分量向量的排列顺序相对应。
终端设备可以根据该空频分量向量集合中N s×N f个空频分量向量的排列顺序，将该N s×N f个投影值按照预先规定的顺序排列成维度为N s×N f的矩阵。
具体地，若上述空频分量向量集合中N s×N f个空频分量向量的索引为一维索引，且该N s×N f个空频分量向量的索引的编号规则基于上文所述的规则一确定，则终端设备可以从该N s×N f个投影值的首个投影值开始，每N f个连续的投影值作为一行，可以得到N s个行，每个行包括N f个投影值。将该N s个行按照从上至下的顺序排列，可以得到维度为N s×N f的矩阵W。
若上述空频分量向量集合中N s×N f个空频分量向量的索引为一维索引，且该N s×N f个空频分量向量的索引的编号规则基于上文所述的规则二确定，则终端设备可以从该N s×N f个投影值的首个投影值开始，每N s个连续的投影值作为一列，可以得到N f个列，每个列包括N s个投影值。将该N f个列按照从左至右的顺序排列，可以得到维度为N s×N f的矩阵W。
若上述空频分量向量集合中N s×N f个空频分量向量的索引为二维索引，则终端设备可以直接根据二维索引将该N s×N f个空频分量向量排成矩阵形式。例如，将索引n s相同的空频分量向量排在同一行，将索引n f相同的空频分量向量排在同一列。
上述维度为N s×N f的矩阵W可以称为加权系数矩阵。其中，矩阵W中的N s×N f个加权系数可以与空频分量向量集合中的N s×N f个空频分量向量对应，可以表示N s×N f个空频分量向量中每个空频分量向量的加权系数。
步骤3-ii、终端设备可以根据加权系数矩阵确定较强的M 1个空频分量向量。
终端设备可以分别对该矩阵W中的N s个行分别取模，根据各行的模长大小，确定模较大的L 1个行。该模较大的L 1个行即为较强的L 1个行。终端设备还可以分别对该矩阵W中的N f个列取模，根据各列的模长大小，确定模较大的K 1个列。该模较大的K 1个列即为较强的K 1个列。终端设备可以根据该较强的L 1个行和较强的K 1个列在矩阵W中的位 置，以及预先定义的转换规则，从空频分量向量集合中确定较强的M 1个空频分量向量。
事实上，该空频分量向量集合中的各空频分量向量可以由波束向量集合中的各波束向量和频域向量集合中的各频域向量确定。该M 1个空频分量向量可以是由波束向量集合中较强的L 1个波束向量和频域向量集合中较强的K 1个频域向量确定。上文所确定的较强的L 1个行在矩阵W中的行序号可以为较强的L 1个波束向量在波束向量集合中的序号，较强的K 1个列在矩阵W中的列序号可以为较强的K 1个频域向量在频域向量集合的序号。
应理解，上文所描述的终端设备根据加权系数矩阵确定较强的L 1个行和较强的K 1个列、进而确定较强的M 1个空频分量向量的具体方法仅为便于理解而示例，不应对本申请构成任何限定。本申请并不排除终端设备采用其他方式来确定较强的M 1个空频分量向量的可能。只要终端设备所确定出的较强的M 1个空频分量向量可以由L 1个波束向量和K 1个频域向量构建，均应落入本申请的保护范围内。
步骤3-iii、终端设备可以从较强的M 1个空频分量向量中确定较强的T 1个空频分量向量。
终端设备可以根据该M 1个空频分量向量的加权系数的模长大小确定较强的T 1个空频分量向量。例如，被选择的T 1个空频分量向量中的任意一个空频分量向量的加权系数的模长大于或等于剩余M 1-T 1个空频分量向量中任意一个空频分量向量的模长。同时，该T 1个空频分量向量的加权系数也可以确定。
步骤3-iv、终端设备生成第一指示信息，以指示L 1个波束向量、K 1个频域向量和T 1个空频分量向量。
基于上述步骤3-i至步骤3-iii所确定的M 1个空频分量向量和T 1个空频分量向量，该第一指示信息可以包括M 1个空频分量向量在空频分量向量集合或空频分量向量集合的子集中的位置信息以及用于指示T 1个空频分量向量的信息。
可选地，该第一指示信息在用于指示M 1个空频分量向量时，具体可用于指示该M 1个空频分量向量的二维索引。即，该M 1个空频分量向量所包含的L 1个波束向量在波束向量集合中的索引以及所包含的K 1个频域向量在频域向量集合中的索引。
上文实现方式一中已经详细说明了该第一指示信息指示L 1个波束向量和K 1个频域向量的具体方法和比特开销，为了简洁，这里不再赘述。
可选地，该第一指示信息在用于指示该M 1个空频分量向量时，具体可用于指示该M 1个空频分量向量在空频分量向量集合或空频分量向量集合的子集中的索引。前文中已经说明，协议可以预先定义多个空频分量向量的索引的编号规则，终端设备和网络设备可以基于相同的编号规则确定空频分量向量集合中各空频分量向量的索引。即，该M 1个空频分量向量的位置信息可以是每个空频分量向量在空频分量向量集合中的索引。此时，终端设备可以通过
Figure PCTCN2019110342-appb-000099
个比特来指示M 1个空频分量向量中的每个空频分量向量。
可选地，该第一指示信息在用于指示该M 1个空频分量向量时，具体可用于指示该M 1个空频分量向量的组合在空频分量向量集合中的索引。例如，协议可以预先定义多个空频分量向量的多种组合，每种组合对应一个索引。该M 1个空频分量向量可以为该多种组合中的一种，或者，接近于该多种组合中的一种。该第一指示信息可以通过指示该组合的索引的方式来指示该M 1个空频分量向量。即，该M 1个空频分量向量的位置信息可以是该M 1个空频分量向量的组合在空频分量向量集合中的索引。此时，终端设备可以通过
Figure PCTCN2019110342-appb-000100
个比特来指示空频分量向量集合中的M 1个空频分量向量。
需要说明的是，由于空频分量向量集合中的每个空频分量向量是由一个波束向量和一个频域向量唯一确定，因此，第一指示信息用于指示M 1个空频分量向量在空频分量向量集合中的位置时，也可以理解为间接地指示了各空频分量向量所对应的波束向量和频域向量分别在波束向量集合中的位置和在频域向量集合中的位置。换句话说，M 1个空频分量向量的位置信息和L 1个波束向量的位置信息与K 1个频域向量的位置信息之间可以互相转换，或者说，是等价的。换句话说说，该第一指示信息在用于指示M 1个空频分量向量时，间接地指示了L 1个波束向量与K 1个频域向量。
应理解，上文列举的两种指示M 1个空频分量向量的方法仅为示例，而不应对本申请构成任何限定。该第一指示信息也可以通过其他方式来指示该M 1个空频分量向量。
可选地，该第一指示信息可以通过以下任意一种方式来指示T 1个空频分量向量：
方式1、通过位图(bitmap)来指示M 1个空频分量向量中的T 1个空频分量向量；或
方式2、指示T 1个空频分量向量的组合在M 1个空频分量向量中的索引；或
方式3、指示T 1个空频分量向量所对应的波束向量在L 1个波束向量中的位置以及T 1个空频分量向量所对应的频域向量在K 1个频域向量中的位置；或
方式4、指示T 1个空频向量对中的每个空频向量对在M 1个空频向量对中的位置。
上文实现方式一中已经详细说明了基于方式1、方式2、方式3和方式4来指示T 1个空频分量向量的具体过程以及分别带来的比特开销，为了简洁，这里不再赘述。
可选地，该第一指示信息还包括T 1个空频向量对的加权系数的量化信息。
上文实现方式一中已经详细说明了第一指示信息指示T 1个空频分量向量的加权系数的具体过程。在实现方式二中，第一指示信息指示T 1个空频分量向量的加权系数的具体方式可以与实现方式一中所提供的具体方式相同，为了简洁，这里不再赘述。
应理解，该T 1个空频向量对的加权系数的量化信息可以携带在第一指示信息中，也可以通过额外的信息来携带，本申请对此不作限定。
若考虑过采样率，终端设备具体可以通过下文所示的步骤4-i至步骤4-iv确定T 1个空频分量向量及各空频分量向量的加权系数。
步骤4-i、终端设备可以基于空频向量和空频分量向量集合中的各子集确定多个加权系数矩阵。
若空频分量向量集合通过过采样因子O c扩展为O c×N s×N f个空频分量向量，则空频分量向量集合可以包括O c个子集。终端设备可以将预先确定的空频向量分别投影至每个子集的N s×N f个空频分量向量，以得到O c组投影值，每组投影值包括N s×N f个投影值。对于每组投影值，其中的N s×N f个投影值的排列顺序与空频分量向量集合中的各子集中N s×N f个空频分量向量的排列顺序相对应。
终端设备可以根据各子集中N s×N f个空频分量向量的排列顺序，将N s×N f个投影值按照预先规定的顺序排列成维度为N s×N f的矩阵。由此可以得到与O c个子集对应的O c个矩阵，每个矩阵可对应一个子集，每个矩阵可以称为与子集对应的加权系数矩阵。由于上文中已经结合空频分量向量集合中各空频分量向量的索引的编号规则详细说明了构建矩阵的具体方法，为了简洁，这里不再赘述。
上述与O c个子集对应的O c个加权系数矩阵可以记作
Figure PCTCN2019110342-appb-000101
o c＝0，1，……，O c-1。对于矩阵
Figure PCTCN2019110342-appb-000102
其中的N s×N f个加权系数可以与空频分量向量集合中第o c个子集的N s×N f 个空频分量向量对应，可以表示该子集中的每个空频分量向量的加权系数。
步骤4-ii、终端设备可以基于空频分量集合中的O c个子集，确定O c组空频分量向量，每组空频分量向量可以包括T 1个空频分量向量。
终端设备可以对o c在0至O c-1中遍历取值，重复执行下述步骤，以确定O c组空频分量向量：根据矩阵W oc确定较强的L 1个行和较强的K 1个列，进而确定第o c个子集中较强的M 1个空频分量向量。终端设备可进一步从该M 1个空频分量向量中确定较强的T 1个空频分量向量。
由于在上文步骤3-ii和步骤3-iii中已经详细说明了终端设备根据加权系数矩阵确定较强的M 1个空频分量向量以及从M 1个空频分量向量中确定较强的T 1个空频分量向量的具体过程，为了简洁，这里不再赘述。
步骤4-iii、终端设备可以基于O c组空频分量向量的加权系数，选择最强的一组空频分量向量，以确定T 1个空频分量向量。
终端设备可以根据步骤4-ii中确定的O c组空频分量向量，确定最强的一组空频分量向量。例如，终端设备可以分别对O c组空频分量向量中的每组空频分量向量的加权系数求模长之和，将模长之和最大的一组空频分量向量确定为最强的一组空频分量向量。由此可以确定T 1个空频分量向量以及各空频分量向量的加权系数。
由于该T 1个空频分量向量来自同一个子集，终端设备在确定T 1个空频分量向量的同时，也就可以确定该T 1个空频分量向量所来自的子集，也就可以确定该子集中较强的M 1个空频分量向量。
步骤4-iv、终端设备生成第一指示信息，以指示L 1个波束向量、K 1个频域向量和T 1个空频分量向量。
基于上述步骤4-i至步骤4-iii中确定的M 1个空频分量向量和T 1个空频分量向量，该第一指示信息可以包括M 1个空频分量向量在空频分量向量集合或空频分量向量集合的子集中的位置信息和用于指示T 1个空频向量对的信息。
可选地，该第一指示信息在用于指示该M 1个空频分量向量时，具体可用于指示该M 1个空频分量向量所包含的L 1个波束向量在波束向量集合中的位置信息以及所包含的K 1个频域向量在频域向量集合中的位置信息；或者，具体可用于指示该M 1个空频分量向量所属的子集以及该M 1个空频分量向量在该子集中的索引；或者，具体可用于指示该M 1个空频分量向量所属的子集以及该M 1个空频分量向量的组合在该子集中的索引。
其中，L 1个波束向量在波束向量集合中的位置信息可以是指该L 1个波束向量在波束向量集合中的索引，或，该L 1个波束向量的组合在波束向量集合中的索引，或，该L 1个波束向量所属的子集以及该L 1个波束向量在该子集中的索引，或，该L 1个波束向量所属的子集以及该L 1个波束向量的组合在该子集中的索引。
K 1个频域向量在频域向量集合中的位置信息可以是指该K 1个频域向量在频域向量集合中的索引，或，该K 1个频域向量的组合在频域向量集合中的索引，或，该K 1个频域向量所属的子集以及该K 1个频域向量在该子集中的索引，或，该K 1个频域向量所属的子集以及该K 1个频域向量的组合在该子集中的索引。
需要说明的是，由于空频分量向量集合中的每个空频分量向量是由一个波束向量和一个频域向量唯一确定，因此，第一指示信息用于指示M 1个空频分量向量在空频分量向量 集合中的位置时，也可以理解为间接地指示了各空频分量向量所对应的波束向量和频域向量分别在波束向量集合中的位置和在频域向量集合中的位置。换句话说，M 1个空频分量向量的位置信息和L 1个波束向量的位置信息与K 1个频域向量的位置信息之间可以互相转换，或者说，是等价的。换句话说说，该第一指示信息在用于指示M 1个空频分量向量时，间接地指示了L 1个波束向量与K 1个频域向量。
该第一指示信息在用于指示T 1个空频分量向量时，其具体的指示方式可以是上文中所述的方式1至方式4中的任意一种。上文实现方式一中已经详细说明了分别基于方式1、方式2、方式3和方式4来指示T 1个空频分量向量的具体过程以及分别带来的比特开销，为了简洁，这里不再赘述。
可选地，该第一指示信息还包括T 1个空频分量向量的加权系数的量化信息。
上文实现方式一中已经详细说明了第一指示信息指示T 1个空频分量向量的加权系数的具体过程。在实现方式二中，第一指示信息指示T 1个空频分量向量的加权系数的具体方式可以与实现方式一中所提供的具体方式相同，为了简洁，这里不再赘述。
应理解，该T 1个空频分量向量的加权系数的量化信息可以携带在第一指示信息中，也可以通过额外的信息来携带，本申请对此不作限定。
情况B、空频分量矩阵集合中的各空频分量矩阵是维度为N s×N f的矩阵。
假设终端设备所确定的N f个频域单元的预编码向量分别记作
Figure PCTCN2019110342-appb-000103
在情况B中，终端设备可以根据该N f个频域单元的预编码向量构建维度为N s×N f的空频矩阵H，
Figure PCTCN2019110342-appb-000104
下面详细说明空频分量矩阵集合。
在一种可能的设计中，该空频分量矩阵集合可以包括N s×N f个空频分量矩阵。每个空频分量矩阵可以是维度为N s×N f的矩阵。该空频分量矩阵集合中选择的T 1个空频分量矩阵的加权和可以构造得到空频矩阵。该T 1个空频分量矩阵的加权和构造得到的空频矩阵可以与上文终端设备所确定的空频矩阵相同或相近。
在本申请实施例中，该空频分量矩阵集合中的每个空频分量矩阵可以是由波束矩阵集合中的一个波束矩阵和频域矩阵集合中的一个频域矩阵唯一确定。换句话说，该空频分量矩阵集合中的任意两个空频分量矩阵不同，且任意两个空频分量矩阵所对应的波束矩阵和频域矩阵中至少有一项不同。具体地，空频分量向量矩阵中的每个空频分量矩阵可以是波束向量集合中的一个波束向量和频域向量集合中的一个频域向量的共轭转置的乘积。
如前所述，若不考虑过采样率，该空频分量矩阵集合中可以包括N s×N f个空频分量矩阵。每个空频分量矩阵可以与一个波束向量和一个频域向量对应，或者说，每个空频分量矩阵可以与一个波束向量和一个频域向量组合得到的空频向量对相对应。
该空频分量矩阵集合中的每个空频分量矩阵可对应一个一维索引，也可对应一个二维索引。即，该空频分量矩阵集合中的N s×N f个空频分量矩阵可以通过空频分量矩阵集合或空频分量矩阵集合的子集中的索引指示，也可以通过可用于生成空频分量矩阵的波束向量和频域向量分别在波束向量集合和频域向量集合中的索引指示。上文情况A中已经结合编号规则一和编号规则二详细说明了各空频分量矩阵与一维索引的对应关系以及一维索引和二维索引之间的转换规则，为了简洁，这里不再赘述。
若考虑过采样率，则空频分量矩阵集合可以通过过采样因子O c扩展为O c×N s×N f个 空频分量矩阵。该空频分量矩阵集合可以包括O c个子集，每个子集可以包括N s×N f个空频分量矩阵。各子集内的空频分量矩阵的索引的编号规则与不考虑过采样率时的编号规则可以是相同的，为了简洁，这里不再赘述。
下面分别说明在考虑和不考虑过采样率的情况下，终端设备确定和指示T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体方法。
若不考虑过采样率，终端设备可以通过下文所示的步骤5-i至步骤5-iv确定T 1个空频分量矩阵以及各空频分量矩阵的加权系数。
步骤5-i、终端设备可以基于上述空频矩阵和空频分量矩阵集合确定加权系数。
若不考虑过采样率，该空频分量矩阵集合可以包括N s×N f个空频分量矩阵，每个空频分量矩阵的维度可以为N s×N f。终端设备可以根据预先确定的空频矩阵和该N s×N f个空频分量矩阵确定N s×N f个加权系数。
具体地，终端设备可以分别将每个空频分量矩阵中的各元素的共轭与空频矩阵中的相应元素的乘积求和，得到与N s×N f个空频分量矩阵对应的N s×N f个值。例如，空频分量矩阵中的一个空频分量矩阵中的元素记作a p,q(p＝0，1，……，N s-1；q＝0，1，……，N f-1)，空频矩阵中的元素记作b p,q，则将每个空频分量矩阵中的各元素的共轭与该空频矩阵中的相应元素的乘积求和可以表示为
Figure PCTCN2019110342-appb-000105
其中，
Figure PCTCN2019110342-appb-000106
表示的元素a p,q的共轭。对空频分量矩阵中的N s×N f个空频分量矩阵重复执行该步骤，可以得到N s×N f个值。该N s×N f个值可视为N s×N f个加权系数。
上述步骤可以通过矩阵运算来实现。例如，可以通过求每个空频分量矩阵的共轭转置与空频矩阵的乘积的迹，来得到上述N s×N f个值。
此后，终端设备可以根据空频分量矩阵集合中N s×N f个空频分量矩阵的排列顺序，将该N s×N f个值按照预先规定的顺序排列成维度为N s×N f的矩阵。终端设备根据预先规定的顺序将该N s×N f个值排列成维度为N s×N f的矩阵的具体过程在上文情况A中已经做了详细说明，为了简洁，这里不再赘述。
上述维度为N s×N f的矩阵W可以称为加权系数矩阵。其中，矩阵W中的N s×N f个加权系数可以与空频分量矩阵集合中的N s×N f个空频分量矩阵对应，可以表示N s×N f个空频分量矩阵中每个空频分量矩阵的加权系数。
步骤5-ii、终端设备可以根据加权系数矩阵确定较强的M 1个空频分量矩阵。
终端设备根据加权系数矩阵确定M 1个空频分量矩阵的具体方法与上文情况A中步骤3-ii终端设备根据加权系数矩阵确定较强的M 1个空频分量向量的具体方法相同。由于上文中已经对该具体方法做了详细说明，为了简洁，这里不再赘述。
步骤5-iii、终端设备可以从M 1个空频分量矩阵中确定较强的T 1个空频分量矩阵。
终端设备可以根据该M 1个空频分量矩阵的加权系数的模长大小确定较强的T 1个空频分量矩阵。例如，被选择的T 1个空频分量矩阵中的任意一个空频分量矩阵的加权系数的模长大于或等于剩余M 1-T 1个空频分量矩阵中任意一个空频分量矩阵的模长。同时，该T 1个空频分量矩阵的加权系数也可以确定。
步骤5-iv、终端设备生成第一指示信息，以指示L 1个波束向量、K 1个频域向量和T 1个空频分量矩阵。
基于上述步骤5-i至步骤5-iii所确定的M 1个空频分量矩阵和T 1个空频分量矩阵，该第一指示信息可以包括M 1个空频分量矩阵在空频分量矩阵集合或空频分量矩阵集合的子集中的位置信息和用于指示T 1个空频分量矩阵的信息。
可选地，该第一指示信息还包括T 1个空频分量矩阵的加权系数的量化信息。
由于上文情况A中已经详细说明了该第一指示信息用于指示M 1个空频分量矩阵、T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体方法以及比特开销。在情况B中，该第一指示信息指示M 1个空频分量矩阵、T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体方法以及比特开销与情况A中示出的方法相同，为了简洁，这里不再赘述。
需要说明的是，由于空频分量矩阵集合中的每个空频分量矩阵是由一个波束向量和一个频域向量唯一确定，因此，第一指示信息用于指示M 1个空频分量矩阵在空频分量矩阵集合中的位置时，也可以理解为间接地指示了各空频分量矩阵所对应的波束向量和频域向量分别在波束向量集合中的位置和在频域向量集合中的位置。换句话说，M 1个空频分量矩阵的位置信息和L 1个波束向量的位置信息与K 1个频域向量的位置信息之间可以互相转换，或者说，是等价的。换句话说说，该第一指示信息在用于指示M 1个空频分量矩阵时，间接地指示了L 1个波束向量与K 1个频域向量。
应理解，该T 1个空频分量矩阵的加权系数的量化信息可以携带在第一指示信息中，也可以通过额外的信息来携带，本申请对此不作限定。
若考虑过采样率，终端设备具体可以通过下文所示的步骤6-i至步骤6-iv确定T 1个空频分量矩阵以及各空频分量矩阵的加权系数。
步骤6-i、终端设备可以基于空频向量和空频分量矩阵集合中的各子集确定多个加权系数矩阵。
若空频分量向量集合通过过采样因子O c扩展为O c×N s×N f个空频分量矩阵，则空频分量矩阵集合可以包括O c个子集。终端设备可以根据预先确定的空频矩阵和O c个子集，确定O c组加权系数，每组加权系数包括N s×N f个加权系数。终端设备确定各组加权系数的具体方法可以参考上文中步骤5-i中的实现方式，为了简洁，这里不再赘述。对于每组加权系数，其中的N s×N f个值的排列顺序与空频分量向量集合中的各子集中N s×N f个空频分量矩阵的排列顺序相对应。
终端设备可以根据各子集中N s×N f个空频分量矩阵的排列顺序，将N s×N f个加权系数按照预先规定的顺序排列成维度为N s×N f的矩阵。由此可以得到与O c个子集对应的O c个矩阵，每个矩阵可对应一个子集，每个矩阵可以称为与子集对应的加权系数矩阵。由于上文中已经结合空频分量矩阵集合中各空频分量矩阵的索引的编号规则详细说明了构建矩阵的具体方法，为了简洁，这里不再赘述。
上述与O c个子集对应的O c个加权系数矩阵可以记作
Figure PCTCN2019110342-appb-000107
o c＝0，1，……，O c-1。对于矩阵
Figure PCTCN2019110342-appb-000108
其中的N s×N f个加权系数可以与空频分量矩阵集合中第o c个子集的N s×N f个空频分量矩阵对应，可以表示该子集中的每个空频分量矩阵的加权系数。
步骤6-ii、终端设备可以基于空频分量集合中的O c个子集，确定O c组空频分量矩阵，每组空频分量矩阵可以包括T 1个空频分量矩阵。
终端设备可以对o c在0至O c-1中遍历取值，重复执行下述步骤，以确定O c组空频分量矩阵：根据矩阵
Figure PCTCN2019110342-appb-000109
确定较强的L 1个行和较强的K 1个列，进而确定第o c个子集中较强 的M 1个空频分量矩阵。终端设备可进一步从该M 1个空频分量矩阵中确定较强的T 1个空频分量矩阵。
由于在上文步骤3-ii和步骤3-iii中已经详细说明了终端设备根据加权系数矩阵确定较强的M 1个空频分量矩阵以及从M 1个空频分量矩阵中确定较强的T 1个空频分量矩阵的具体过程，为了简洁，这里不再赘述。
步骤6-iii、终端设备可以基于O c组空频分量矩阵的加权系数，选择最强的一组空频分量矩阵，以确定T 1个空频分量矩阵。
终端设备可以根据步骤6-ii中确定的O c组空频分量矩阵，确定最强的一组空频分量矩阵。例如，终端设备可以分别对O c组空频分量矩阵中的每组空频分量矩阵的加权系数求模长之和，将模长之和最大的一组空频分量矩阵确定为最强的一组空频分量矩阵。由此可以确定T 1个空频分量矩阵以及各空频分量矩阵的加权系数。
由于该T 1个空频分量矩阵来自同一个子集，终端设备在确定T 1个空频分量矩阵的同时，也就可以确定该T 1个空频分量矩阵所来自的子集，也就可以确定该子集中较强的M 1个空频分量矩阵。
步骤6-iv、终端设备生成第一指示信息，以指示L 1个波束向量、K 1个频域向量和T 1个空频分量矩阵。
基于上述步骤6-i至步骤6-iii中确定的M 1个空频分量矩阵和T 1个空频分量矩阵，该第一指示信息可以包括M 1个空频分量矩阵在空频分量矩阵集合或空频分量矩阵集合的子集中的位置信息和用于指示T 1个空频矩阵矩阵的信息。
可选地，该第一指示信息还包括T 1个空频分量矩阵的加权系数的量化信息。
由于上文情况A中已经详细说明了该第一指示信息用于指示M 1个空频分量矩阵、T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体方法以及比特开销。在情况B中，该第一指示信息指示M 1个空频分量矩阵、T 1个空频分量矩阵以及各空频分量矩阵的加权系数的具体方法以及比特开销与情况A中示出的方法相同，为了简洁，这里不再赘述。
需要说明的是，由于空频分量矩阵集合中的每个空频分量矩阵是由一个波束向量和一个频域向量唯一确定，因此，第一指示信息用于指示M 1个空频分量矩阵在空频分量矩阵集合中的位置时，也可以理解为间接地指示了各空频分量矩阵所对应的波束向量和频域向量分别在波束向量集合中的位置和在频域向量集合中的位置。换句话说，M 1个空频分量矩阵的位置信息和L 1个波束向量的位置信息与K 1个频域向量的位置信息之间可以互相转换，或者说，是等价的。换句话说说，该第一指示信息在用于指示M 1个空频分量矩阵时，间接地指示了L 1个波束向量与K 1个频域向量。
应理解，该T 1个空频分量向的加权系数的量化信息可以携带在第一指示信息中，也可以通过额外的信息来携带，本申请对此不作限定。
还应理解，上文通过加权系数矩阵确定较强的L 1个行和较强的K 1个列的方法仅为便于理解而示出的一种可能的实现方式，并不代表终端设备在确定较强的L 1个行和较强的K 1个列的过程中一定生成了该加权系数矩阵。例如，终端设备可以根据各频域单元的预编码向量和空频分量矩阵集合中的各空频分量矩阵，确定多个加权系数。该多个加权系数可以组成一个数组集合，该数组集合中的各元素可以是上述加权系数矩阵中的各行(或各列)元素依次连接而成。
又例如，终端设备获取波束向量的方法可以参考NR协议中定义的type II码本反馈方式。通过将同一传输层、一个极化方向上各个频域单元的预编码向量的至少一个组成元素(例如但不限于构成该预编码向量的波束向量的加权系数等)进行比较，可以获得频域向量，从而获得频域的变化规律，不同极化方向可以使用相同的一组频域向量。
还应理解，上文所列举的终端设备从M 1个空频分量矩阵中确定较强的T 1个空频分量矩阵的具体方法仅为示例，而不应对本申请构成任何限定。终端设备可以参考现有技术中的方法从M 1个空频分量矩阵中确定较强的T 1个空频分量矩阵，为了简洁，这里不做详细说明。
还应理解，上文中列举的用于确定和指示T 1个空频向量对(或T 1个空频分量向量，或T 1个空频分量矩阵)的加权系数的具体方法和信令仅为示例。本申请对于确定和指示T 1个空频向量对(或T 1个空频分量向量，或T 1个空频分量矩阵)的加权系数的具体方法和信令不作限定，其具体实现方式和指示信令可以与现有技术相同。
需要说明的是，由于空频分量矩阵集合中的每个空频分量矩阵是由一个波束向量和一个频域向量唯一确定，因此，第一指示信息用于指示M 1个空频分量矩阵在空频分量矩阵集合中的位置时，也可以理解为间接地指示了各空频分量矩阵所对应的波束向量和频域向量分别在波束向量集合中的位置和在频域向量集合中的位置。换句话说，M 1个空频分量矩阵的位置信息和L 1个波束向量的位置信息与K 1个频域向量的位置信息之间可以互相转换。
基于上文所列举的终端设备在不同情况下的实现方式，终端设备可以生成第一指示信息，以指示被选择的L 1个波束向量、K 1个频域向量、T 1个空频分量矩阵以及各空频分量矩阵的加权系数。
在步骤220中，终端设备发送该第一指示信息。相应地，在步骤220中，网络设备接收该第一指示信息。
具体地，该第一指示信息可以是PMI，也可以是PMI中的部分信元，还可以是其他信息。本申请对此不作限定。该第一指示信息可以携带在现有技术中的一个或者多个消息中由终端设备发送给网络设备，也可以携带在本申请新设计的一个或者多个消息中由终端设备发送给网络设备。终端设备例如可以通过物理上行资源，如物理上行共享信道(physical uplink share channel，PUSCH)或物理上行控制信道(physical uplink control channel，PUCCH)，向网络设备发送该第一指示信息，以便于网络设备基于该第一指示信息恢复预编码向量。
终端设备通过物理上行资源向网络设备发送第一指示信息的具体方法可以与现有技术相同，为了简洁，这里省略对其具体过程的详细说明。
在步骤230中，网络设备根据该第一指示信息确定一个或多个频域单元的预编码向量。
在步骤210中已经说明，终端设备可以基于两种不同的实现方式生成第一指示信息。在这种实现方式中，第一指示信息中用于指示L 1个波束向量的信息和用于指示K 1个频域向量的信息有可能是不同的，也有可能是相同的。下面详细说明网络设备根据第一指示信息确定一个或多个频域单元的预编码向量的具体过程。
可选地，该第一指示信息包括L 1个波束向量的位置信息、K 1个频域向量的位置信息 和用于指示T 1个空频分量矩阵的信息。
首先，网络设备可以根据L 1个波束向量的位置信息和K 1个频域向量的位置信息确定被选择的L 1个波束向量和K 1个频域向量。
例如，网络设备可以根据该第一指示信息所指示的L 1个波束向量的组合在波束向量集合中的索引，以及预先定义的波束向量的组合与索引的对应关系，确定波束向量集合中被选择的L 1个波束向量。网络设备也可以根据第一指示信息所指示的K 1个频域向量的组合在频域向量集合中的索引，以及预先定义的频域向量的组合与索引的对应关系，确定频域向量集合中被选择的K 1个频域向量。该L 1个波束向量和K 1个频域向量可以组合得到M 1个空频向量对。
此后，网络设备可以根据用于指示T 1个空频分量矩阵的信息，从该M 1个空频向量对中确定T 1个空频分量矩阵。
如前所述，第一指示信息可以通过不同的方式向网络设备指示该T 1个空频向量对。例如，在方式1中，第一指示信息通过位图来指示M 1个空频向量对中被选择的T 1个空频向量对，网络设备可以根据该位图中的各比特与M 1个空频向量对的一一对应关系，确定被选择的T 1个空频向量对；在方式2中，第一指示信息通过T 1个空频向量对的组合在M 1个空频向量对中的索引来指示该M 1个空频向量对，网络设备可以根据预先定义的空频向量对的组合与索引的对应关系，确定该T 1个空频向量对；在方式3中，第一指示信息通过该T 1个空频向量对中的每个空频向量对所包含的波束向量的位置和频域向量的索引来指示该T 1个空频向量对，网络设备可以根据波束向量的位置信息和频域向量的位置信息，确定T 1个波束向量和T 1个频域向量，并进一步组合得到T 1个空频向量对。
此后，网络设备可以基于基于各空频向量对的加权系数的量化信息确定各空频向量对的加权系数的量化值。如前所述，各空频向量对的加权系数可以通过第一指示信息或其他信息携带。例如，网络设备可以根据预先定义的多个量化值与多个索引的一一对应关系，确定各空频向量对的加权系数的量化值。
基于上述T 1个空频向量对及各空频向量对的加权系数的量化值确定空频矩阵。例如，网络设备可以基于如下公式确定空频矩阵H：
Figure PCTCN2019110342-appb-000110
或，
Figure PCTCN2019110342-appb-000111
其中，
Figure PCTCN2019110342-appb-000112
表示T 1个波束向量中的第t 1个波束向量，
Figure PCTCN2019110342-appb-000113
表示T 1个频域向量中的第t 1个频域向量。
Figure PCTCN2019110342-appb-000114
表示与第t 1个空频分量矩阵对应的加权系数。式中的空频矩阵H可以与终端设备所确定的空频矩阵相同或相近，是网络设备根据第一指示信息恢复得到的空频矩阵。由于空频矩阵可以是N f个频域单元对应的预编码向量构造而得，网络设备可以根据矩阵H中的第n f个列向量确定与第n f个频域单元对应的预编码向量。
或者，网络设备也可以在确定了T 1个空频向量对后，进一步生成T 1个空频分量矩阵。例如，对于T 1个空频分量矩阵中的第t 1个空频分量矩阵，可以记作
Figure PCTCN2019110342-appb-000115
或，
Figure PCTCN2019110342-appb-000116
等。
此后，网络设备可以基于上述T 1个空频分量矩阵和各空频分量矩阵的加权系数的量化值确定空频矩阵。例如，网络设备可以根据如下公式确定空频矩阵H：
Figure PCTCN2019110342-appb-000117
式中各参数在上文中已经做了详细说明，为了简洁，这里不再赘述。
网络设备可以根据该空频矩阵中确定与各个频域单元对应的预编码向量。
若该空频矩阵为维度为N s×N f的矩阵，则该矩阵中的每个列向量可以与一个频域单元对应，可用于所对应的频域单元的预编码向量。若该空频矩阵为长度为N s×N f的向量，则该向量中的第n f×N s个元素至第(n f+1)N s-1个元素依次连接而得到的列向量可以是与第n f个频域单元对应的列向量。
以空频矩阵中的第n f个列向量为例，网络设备可以对该第n f个列向量进行归一化处理，以确定与第n f个频域单元对应的预编码向量。其中，归一化处理例如可以是将该第n f个列向量乘以归一化系数，以使得该列向量中各元素的功率之和等于1。归一化系数例如可以是这一列中各元素的模长之和的平方根的倒数。本申请对于归一化系数的具体取值以及归一化处理的具体方式均不做限定。
可选地，该第一指示信息包括M 1个空频分量矩阵的位置信息和用于指示T 1个空频分量矩阵的信息。
如前所述，M 1个空频分量矩阵的位置信息可以是该M 1个空频分量矩阵的一维索引，也可以是该M 1个空频分量矩阵的二维索引。
若M 1个空频分量矩阵的位置信息为该M 1个空频分量矩阵的一维索引，而网络设备预先保存的向量集合为空频分量矩阵集合，则网络设备可以直接根据该一维索引，从空频分量矩阵集合中确定该M 1个空频分量矩阵。
若M 1个空频分量矩阵的位置信息为该M 1个空频分量矩阵的一维索引，而网络设备预先保存的向量集合波束向量集合和频域向量集合，则网络设备可以根据预先定义的一维索引和二维索引的转换规则，从波束向量集合和频域向量集合中确定用于生成该M 1个空频分量矩阵的L 1个波束向量和K 1个频域向量。
若M 1个空频分量矩阵的位置信息为该M 1个空频分量矩阵的二维索引，而网络设备预先保存的向量集合为波束向量集合和频域向量集合，则网络设备可以直接根据该二维索引，从波束向量集合和频域向量集合中确定用于生成该M 1个空频分量矩阵的L 1个波束向量和K 1个频域向量。
若M 1个空频分量矩阵的位置信息为该M 1个空频分量矩阵的二维索引，而网络设备预先保存的向量集合为空频分量矩阵集合，则网络设备可以根据预先定义的二维索引和一维索引的转换规则，从空频分量矩阵集合中确定M 1个空频分量矩阵。
此后，网络设备可以进一步从该M 1个空频分量矩阵中确定T 1个空频分量矩阵，并可进一步基于各空频分量矩阵的加权系数的量化信息确定各空频分量矩阵的加权系数的量化值。
网络设备根据T 1个空频分量矩阵和各空频分量矩阵的加权系数的量化值确定空频矩阵的具体方法在上文中已经做了详细说明，网络设备例如可以根据公式
Figure PCTCN2019110342-appb-000118
确定空频矩阵。
此后，网络设备可以确定一个或多个频域单元的预编码向量。由于上文中已经对网络 设备根据空频矩阵确定预编码向量的具体方法做了详细说明，为了简洁，这里不再赘述。
应理解，上文所列举的两种确定各频域单元上的预编码向量的具体方法仅为示例，而不应对本申请构成任何限定，网络设备可以不生成空频矩阵，基于终端设备所指示的L 1个波束向量、K 1个频域向量直接确定一个或多个频域单元上的预编码向量。
例如，网络设备可以基于如下公式确定第n f个频域单元上的预编码向量
Figure PCTCN2019110342-appb-000119
Figure PCTCN2019110342-appb-000120
其中，
Figure PCTCN2019110342-appb-000121
为归一化系数，
Figure PCTCN2019110342-appb-000122
表示被选择的T 1个波束向量中的第t 1个波束向量，
Figure PCTCN2019110342-appb-000123
表示
Figure PCTCN2019110342-appb-000124
的共轭，
Figure PCTCN2019110342-appb-000125
表示被选择的T 1个频域向量中的第t 1个频域向量
Figure PCTCN2019110342-appb-000126
中的第n f个元素，
Figure PCTCN2019110342-appb-000127
表示与第t 1个波束向量
Figure PCTCN2019110342-appb-000128
和第t 1个频域向量
Figure PCTCN2019110342-appb-000129
中的第n f个元素
Figure PCTCN2019110342-appb-000130
对应的加权系数，该加权系数例如可以包括幅度系数和相位系数。
上文公式还可进一步变形为：
Figure PCTCN2019110342-appb-000131
其中，
Figure PCTCN2019110342-appb-000132
由被选择的T 1个波束向量中的第t 1个波束向量与宽带幅度系数
Figure PCTCN2019110342-appb-000133
确定，则加权系数
Figure PCTCN2019110342-appb-000134
可以满足：
Figure PCTCN2019110342-appb-000135
需要说明的是，如前所述，频域向量的长度N f可以是配置给终端设备的CSI测量资源的频域占用带宽中包含的频域单元的数量，或，reporting band的信令长度，或，待上报的频域单元数。当频域向量的长度为配置给终端设备的CSI测量资源的频域占用带宽中包含的频域单元的数量或reporting band的信令长度时，待上报的频域单元数可以是小于或者等于N f的。因此，网络设备可以根据reporting band或者其他信令所指示的待上报的频域单元的位置，确定各频域单元的预编码向量。
其中，根据CSI测量资源的频域占用带宽中包含的频域单元的数量或reporting band的信令长度来确定频域向量的长度，可以将信道在多个连续的频域单元上的变化规律通过频域向量来体现，相比于根据待上报的频域单元数确定频域向量的长度这种方法而言，根据CSI测量资源的频域占用带宽中的频域单元的数量或reporting band的信令长度确定的频域向量更能够准确地反映信道在频域的变化规律，基于反馈所恢复的预编码向量也更能够与信道适配。
应理解，上文所列举的网络设备根据第一指示信息确定与第n f个频域单元对应的预编码向量的具体方法仅为示例，而不应对本申请构成任何限定。本申请并不排除网络设备根据该第一指示信息，采用其他方式确定与第n f个子带对应的预编码向量的可能。
基于上述技术方案，终端设备通过向网络设备指示少量的波束向量、频域向量和空频分量矩阵，以便于网络设备恢复预编码向量。其中，频域向量可用于描述信道在频域上不同的变化规律。终端设备可以通过一个或多个频域向量的线性叠加来模拟信道在频域上的变化，充分挖掘了频域单元之间的关系，利用了频域的连续性，用较少的频域向量来描述多个频域单元的变化规律。相比于现有技术而言，无需基于每个频域单元独立地上报加权系数，在频域单元数量增加的情况下，并不会造成反馈开销的成倍增加。因此，可以在保 证反馈精度的基础上大大减小反馈开销。
然而，由于波束向量集合中可能包含较多的波束向量，频域向量集合中也可能包含较多的频域向量，如果直接从波束向量集合和频域向量集合中指示较少的波束向量和频域向量，可能会带来较大的比特开销，或者，需要终端设备和网络设备预先定义大量的波束向量组合与索引的对应关系以及频域向量组合与索引的对应关系。
而本申请实施例中，终端设备将用作加权求和的波束向量和频域向量的选择范围缩小至由L 1个波束向量和K 1个频域向量构建的M 1个空频分量矩阵的范围中，也就是从已有的向量集合中先选择一个较小的范围，再进一步从该范围中选择和指示T 1个空频分量矩阵。一方面可以避免直接指示该T 1个空频分量矩阵带来的较大的反馈开销，另一方面可以避免在终端设备和网络设备两端保存大量的对应关系。
应理解，上文结合图2详细说明了终端设备指示一个传输层上、一个极化方向上的预编码向量以及网络设备确定预编码向量的具体过程。但应理解，该方法并不仅仅适用于传输层为1或极化方向数为1的情况，对于多个传输层或多个极化方向的情况同样适用。
对于同一传输层，多个极化方向上被选择的T 1个空频分量矩阵(或者说，T 1个空频向量对)可以是相同的。即，多个极化方向共用相同的T 1个空频分量矩阵(或者说，T 1个空频向量对)，也可以分别使用不同的空频分量矩阵(或者说，空频向量对)。
下文中为方便说明，以空频分量矩阵为例来说明多个极化方向或多个传输层的情况。其中，空频分量矩阵可以包括上文所列举的矩阵形式或向量形式。可以理解，该空频分量矩阵仅为一种可能的形式，也可以表现为空频向量对的形式。即，下文中所涉及的空频分量矩阵也可以替换为空频向量对。
可选地，该第一指示信息用于指示一个或多个极化方向上各频域单元的预编码向量。
当多个极化方向共用相同的T 1个空频分量矩阵时，用于确定该T 1个空频分量矩阵的L 1个波束向量和K 1个频域向量也可以是多个极化方向共用的。此情况下，与多个极化方向对应的多个第一指示信息中，用于指示L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵的信息可以是共用的。例如，终端设备可以采用一个长度为L 1×K 1位图来指示多个极化方向中每个极化方向上使用的T 1个空频分量矩阵。
若该第一指示信息仅用于指示L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵，则终端设备对于多个极化方向可以仅生成并发送一个第一指示信息。若该第一指示信息还用于指示T 1个空频分量矩阵的加权系数，则该终端设备可以针对每个极化方向发送一个第一指示信息。与多个极化方向对应的多个第一指示信息中，用于指示L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵的信息可以是共用的，终端设备可以仅指示一次上述L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵，而与不同极化方向对应的各空频分量矩阵的加权系数可以分别通过不同的第一指示信息来指示。为了简洁，后文中省略对相同或相似情况的说明。
终端设备可以将基于某一个极化方向确定的L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵用于多个极化方向上。终端设备基于哪个极化方向确定L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵可以是预先定义的，如协议定义，本申请对此不作限定。
终端设备也可以基于每个极化方向确定T 1个空频分量矩阵，得到多组空频分量矩阵。 并从该多组空频分量矩阵中选择一组用于多个极化方向上。被选择的这一组空频分量矩阵的加权系数的模长之和可以大于或等于剩余的一组或多组空频分量矩阵中任意一组空频分量矩阵的加权系数的模长之和。
当多个极化方向分别使用不同的空频分量矩阵时，不同极化方向上使用的空频分量向量矩阵的数量可以相同，也可以不同。用于确定各极化方向上的空频分量矩阵的波束向量的个数可以相同，也可以不同；用于确定各极化方向上的空频分量矩阵的频域向量的个数可以相同，也可以不同，本申请对此不作限定。此情况下，与多个极化方向对应的多个第一指示信息中，与不同极化方向对应的波束向量、频域向量以及空频分量矩阵的信息可以是互不相同的。终端设备可以基于每个极化方向分别指示被选择的波束向量、频域向量以及用作加权求和的空频分量矩阵。终端设备基于每个极化方向确定波束向量、频域向量以及用作加权求和的空频分量矩阵的具体方式与上文所述的基于一个极化方向确定L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵的具体方式相同，为了简洁，这里不再赘述。
通常情况下，多个极化方向上可以共用相同的L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵。下面以多个极化方向共用相同的L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵为例来说明。
假设极化方向数为2，且两个极化方向上共用相同的L 1个波束向量和K 1个频域向量，两个极化方向上用作加权求和的T 1个空频分量矩阵也是相同的。
网络设备可以基于如下公式确定第n f个频域单元的预编码向量：
Figure PCTCN2019110342-appb-000136
其中，
Figure PCTCN2019110342-appb-000137
表示与第一极化方向上的第t 1个波束向量和第t 1个频域向量
Figure PCTCN2019110342-appb-000138
中的第n f个元素对应的加权系数，
Figure PCTCN2019110342-appb-000139
表示与第二极化方向上的第t 1个波束向量
Figure PCTCN2019110342-appb-000140
和第t 1个频域向量
Figure PCTCN2019110342-appb-000141
中的第n f个元素
Figure PCTCN2019110342-appb-000142
对应的加权系数。
上述公式还可进一步变形为：
Figure PCTCN2019110342-appb-000143
其中，
Figure PCTCN2019110342-appb-000144
由第一极化方向上被选择的T 1个波束向量中的第t 1个波束向量与宽带幅度系数
Figure PCTCN2019110342-appb-000145
确定，
Figure PCTCN2019110342-appb-000146
由第二极化方向上被选择的T 1个波束向量中的第t 1个波束向量与宽带幅度系数
Figure PCTCN2019110342-appb-000147
确定，
Figure PCTCN2019110342-appb-000148
表示与第一极化方向上的
Figure PCTCN2019110342-appb-000149
和第t 1个频域向量
Figure PCTCN2019110342-appb-000150
中的第n f个元素对应的加权系数，
Figure PCTCN2019110342-appb-000151
表示与第二极化方向上的
Figure PCTCN2019110342-appb-000152
和第t 1个频域向量
Figure PCTCN2019110342-appb-000153
中的第n f个元素
Figure PCTCN2019110342-appb-000154
对应的加权系数。
可选地，两个极化方向共用相同的L 1个波束向量、K 1个频域向量和T 1个空频分量矩阵。当终端设备采用方式1中所述的位图来指示T 1个空频分量矩阵时，具体可以通过一 个长度为L 1×K 1的位图来指示。
可选地，两个极化方向共用相同的L 1个波束向量和K 1个频域向量，终端设备可以基于该L 1个波束向量和K 1个频域向量，确定各极化方向上的一个或多个空频分量矩阵。该两个极化方向上确定的空频分量矩阵的总数例如可以记作S 1。该S 1的取值可以为2T 1，终端设备可以基于每个极化方向上分别确定T 1个较强的空频分量矩阵，该两个极化方向中第一极化方向的T 1个空频分量矩阵与第二极化方向的T 1个空频分量矩阵可以相同，也可以不同，本申请对此不作限定；该S 1的取值也可以不为2T 1，终端设备可以基于两个极化方向共同确定S 1个较强的空频分量矩阵。当终端设备基于两个极化方向共同确定S 1个较强的空频分量矩阵时，第一极化方向和第二极化方向上的空频分量矩阵的总数可以为S 1，且第一极化方向上的空频分量矩阵的数量与第二极化方向上的空频分量矩阵的数量可以相同，也可以不同，本申请对此不作限定。
终端设备在指示该S 1个空频分量矩阵时，仍然可以采用上文所列举的方式1至方式4中的任意一项来指示。当采用方式1中所述的位图来指示时，该位图的长度可以是2L 1×K 1个比特，或者也可以通过与两个极化方向分别对应的位图来分别指示。
可选地，当极化方向数为2时，上述第三指示信息也可以用于指示S 1的取值。
若终端设备基于每个极化方向分别确定并上报空频分量矩阵，则S 1可以为偶数，如2T 1；若终端设备基于两个极化方向共同确定并上报空频分量矩阵，则S 1可以为奇数，也可以为偶数，本申请对此不作限定。
可选地，两个极化方向共用相同的L 1个波束向量，终端设备可以基于每个极化方向确定K 1个频域向量以及T 1个空频分量矩阵。
可选地，两个极化方向共用相同的L 1个波束向量，终端设备可以基于每个极化方向确定K 1个频域向量，并基于两个极化方向共同确定2T 1个较强的空频分量矩阵。
可选地，上述第一指示信息所指示的T 1个空频分量矩阵与多个传输层中的第一传输层关联。即，该第一指示信息所指示的T 1个空频分量矩阵可用于确定第一传输层上一个或多个频域单元上的预编码向量。
其中，该第一传输层可以是一个传输层，也可以是多个传输层，本申请对此不作限定。换句话说，该第一指示信息可用于确定一个或多个传输层上各频域单元的预编码向量。
综上，该第一指示信息可用于指示一个或多个极化方向，和/或，一个或多个传输层上各频域单元的预编码向量。即，第一指示信息可用于确定一个或多个极化方向上各频域单元的预编码向量，也可以用于确定一个或多个传输层上各频域单元的预编码向量，还可以用于确定一个或多个传输层中各传输层上一个或多个极化方向上各频域单元的预编码向量。
进一步地，该方法还包括：终端设备生成第四指示信息，该第四指示信息用于指示波束向量集合中的L 2个波束向量、频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵。该T 2个空频分量矩阵的加权和可用于确定第二传输层上一个或多个频域单元的预编码向量。换句话说，该第四指示信息所指示的L 2个波束向量、K 2个频域向量以及T 2个空频分量矩阵与第二传输层关联。
可选地，该第四指示信息用于指示L 2个波束向量和T 2个空频分量矩阵，该T 2个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，该L 2个波束向量 和K 2个频域向量与M 2个空频分量矩阵对应，该T 2个空频分量矩阵是M 2个空频分量矩阵中的部分空频分量矩阵，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 2个波束向量中的一个波束向量和K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；该L 2个波束向量为该波束向量集合中的部分波束向量，和/或，该K 2个频域向量为该频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数。
可选地，该K 2个频域向量是预先配置的。如，该K 2个频域向量可以为频域向量集合的全部或部分频域向量。
例如，协议可以预先定义K 2＝K 0，即，协议默认将该频域向量集合的全集作为该K 2个频域向量。又例如，协议可以预先定义K 2的取值，并预先指定将频域向量集合中的哪些频域向量作为该K 2个频域向量。再例如，协议可以预先定义K 1的取值，网络设备可以预先通过信令指示将该K 2个频域向量。
换句话说，预先定义终端设备不需要上报该K 2个频域向量。该K 2个频域向量可以是预先指定的，如协议定义或网络设备配置等，本申请对此不作限定。
可选地，该K 2个频域向量与K 1个频域向量相同。可选地，该K 2个频域向量与K 1个频域向量不同。
可选地，该K 2个频域向量为K 1个频域向量的子集。
上文中已经结合第一指示信息详细说明了不同的参数取值以及配置不同参数的情况下终端设备上报的向量，终端设备在生成和发送第四指示信息时，可以基于与上文所述相同的方式处理。为了简洁，这里不再赘述。
可选地，该第二传输层为多个传输层中除第一传输层之外的一个或多个传输层。换句话说，该第四指示信息可用于确定一个或多个传输层上各频域单元的预编码向量。具体地，该第四指示信息可用于指示第二传输层中各传输层上一个或多个极化方向上各频域单元的预编码向量。
在一种可能的设计中，L 1＝L 2，K 1＝K 2，且T 1＝T 2。也就是说，对于多个传输层来说，任意两个传输层上确定的波束向量的个数相同、任意两个传输层上确定的频域向量的个数相同，且任意两个传输层上确定的空频分量矩阵的个数相同。
在这种设计中，可选地，各传输层可以共用相同的L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵。此情况下，该第四指示信息与上述第一指示信息中用于指示L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵的信息可以是共用的。此时，该第一指示信息和第四指示信息可以是同一指示信息。
并且，当不同传输层和/或不同极化方向存在共用的上述信息时，可以仅生成并发送一个指示信息。
可选地，各传输层可以共用相同的L 1个波束向量和K 1个频域向量，但分别使用各自的T 1个空频分量矩阵。即，第一传输层的T 1个空频分量矩阵和第二传输层的T 1个空频分量矩阵不同。此情况下，该第四指示信息与上述第一指示信息中用于指示L 1个波束向量和K 1个频域向量的信息可以是共用的，第一指示信息和第四指示信息可分别用于指示所对应的传输层上的T 1个空频分量矩阵。
可选地，各传输层分别使用各自的L 1个波束向量、K 1个频域向量或T 1个空频分量矩阵。即，第一传输层的L 1个波束向量与第二传输层的L 1个波束向量不同，第一传输层的 K 1个频域向量与第二传输层的L 1个波束向量不同，且第一传输层的T 1个空频分量矩阵与第二传输层的T 1个空频分量矩阵不同。此情况下，第一指示信息和第二指示信息可分别用于指示所对应的传输层上的L 1个波束向量、K 1个频域向量以及T 1个空频分量矩阵。
进一步地，L 1、K 1和T 1的取值可以随传输层数的增加而变化。例如，协议可以预先定义，当传输层数大于某一预设门限时，L 1、K 1和T 1中的至少一项减小。
例如，当传输层数大于2时，T 1可以减小。如，T 1可以减小为T 1/2，或者减小为T 1/3。
又例如，当传输层数大于2时，L 1和T 1均可以减小。如，L 1可以减小为L 1/2，T 1也可以减小为T 1/2，或者，L 1可以减小为L 1/3，T 1也可以减小为T 1/3。
再例如，当传输层数大于2时，K 1和T 1均可以减小。如，K 1可以减小为K 1/2，T 1也可以减小为T 1/2，或者，K 1可以减小为K 1/3，T 1也可以减小为T 1/3。
还例如，当传输层数大于2时，L 1、K 1和T 1均可以减小。如，L 1可以减小为L 1/2，K 1也可以减小为K 1/2，T 1也可以减小为T 1/2，或者，L 1可以减小为L 1/3，K 1也可以减小为K 1/3，T 1也可以减小为T 1/3。
应理解，上文所列举的预设门限仅为示例，而不应对本申请构成任何限定。还应理解，上文所列举的对L 1、K 1和T 1分别减小的方法仅为示例，而不应对本申请构成任何限定。上述预设门限以及L 1、K 1和T 1减小后的具体取值可以由协议预先定义。
需要说明的是，虽然L 1、K 1和T 1的取值可以随传输层数而变化，变化后的L 1、K 1和T 1和与变化后的L 2、K 2和T 2仍然可以满足：L 1＝L 2，K 1＝K 2，且T 1＝T 2。
还应理解，上文虽然以第一传输层和第二传输层为例来说明了不同传输层上的L 1与L 2、K 1与K 2以及T 1与T 2的大小关系，但这不应对本申请构成任何限定。传输层数并不仅限于2，还有可能大于2。例如，传输层数为3、4等，本申请对此不作限定。
在另一种可能的设计中，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。也就是说，对于多个传输层来说，至少两个传输层上确定的波束向量的个数不同，或者，至少两个传输层上确定的频域向量的个数不同，或者，至少两个传输层上确定的空频分量矩阵的个数不同。
可选地，当传输层数大于预设门限时，可以减小部分传输层中的空频分量矩阵、波束向量和频域向量中至少一项的个数。
例如，当传输层数大于预设门限时，可以将部分传输层中的空频分量矩阵的个数减半。即，第一传输层的空频分量矩阵的个数T 1可以是第二传输层的空频分量矩阵的个数T 2的两倍。其中，第一传输层例如可以对应于SVD过程中较大特征值所对应的特征向量确定的预编码向量，第二传输层例如可以对应于SVD过程中较小特征值所对应的特征向量确定的预编码向量。第一传输层可以表示一个传输层，也可以表示具有相同特性的多个传输层；第二传输层可以表示一个传输层，也可以表示具有相同特性的多个传输层。
当第一传输层表示多个传输层时，用于指示第一传输层的预编码向量的第一指示信息的数量可以为多个，以与多个传输层对应。若该多个传输层共用波束向量、频域向量和空频分量矩阵，则该多个第一指示信息中用于指示波束向量、频域向量个空频分量矩阵的信息可以共用。此时，可以仅生成并发送一个第一指示信息。
同理，当第二传输层表示多个传输层时，用于指示第二传输层的预编码向量的第四指示信息的数量可以为多个，以与多个传输层对应。若该多个传输层共用波束向量、频域向量或空频分量矩阵，则该多个第四指示信息中用于指示波束向量、频域向量或空频分量矩 阵的信息可以共用。此时，可以仅生成并发送一个第四指示信息。
上文中基于特征值的大小来对传输层进行划分仅为一种可能的实现方式，而不应对本申请构成任何限定。例如，协议也可以预先定义对传输层进行划分的其他准则，本申请对此不作限定。
如传输层数为4，预设门限为2。则可以将4个传输层中的第2个传输层和第3个传输层的空频分量矩阵的个数减半，而第0个传输层和第1个传输层的空频分量矩阵的个数不变。其中，第0个传输层和第1个传输层可以是第一传输层的两例，第2个传输层和第3个传输层可以是第二传输层的两例。
基于相同的方式，也可以减小部分传输层中波束向量的个数和/或频域向量的个数。为了简洁，这里不再一一举例说明。
再进一步地，当传输层数大于预设门限，减小部分传输层中的空频分量矩阵、波束向量和频域向量中至少一项的个数时，终端设备可以将第一传输层的部分空频分量矩阵用于第二传输层。此时，终端设备可以在第四指示信息中仅指示被用于第二传输层的部分空频分量矩阵，例如，通过位图、空频分量矩阵的索引等方式来指示，而不再重复指示被选择的波束向量和频域向量，由此可以减小反馈开销。此时，终端设备可以指示用于第二传输层的部分空频分量矩阵在用于第一传输层的空频分量矩阵中的相对位置，或者说，局部(local)位置，如相对编号或相对索引等。
可以理解的是，至少两个传输层的波束向量的个数不同，或者，至少两个传输层上确定的频域向量的个数不同，或者，至少两个传输层上确定的空频分量矩阵的个数不同时，终端设备可以分别通过第一指示信息和第四指示信息来分别指示所对应的传输层上被选择的波束向量、频域向量、空频分量矩阵以及各空频分量矩阵的加权系数。
应理解，上述方法还可以结合使用。例如，传输层数为4，其中的第0个传输层和第1个传输层均可以为第一传输层。第0个传输层和第1个传输层可以共用相同的波束向量和频域向量，与第0个传输层和第1个传输层分别对应的两个第一指示信息在用于指示该第0个传输层和第1个传输层的波束向量和频域向量时，可以仅生成并指示一次L 1个波束向量的指示信息和K 1个频域向量的指示信息。但第0个传输层和第1个传输层使用的空频分量矩阵可以通过不同的位图分别指示。与第0个传输层对应的第一指示信息中可以包括用于指示第0个传输层的空频分量矩阵的位图；与第1个传输层对应的第一指示信息中可以包括用于指示第1个传输层的空频分量矩阵的位图。
第2个传输层和第3个传输层均可以为第二传输层。第2个传输层和第3个传输层也可以共用相同的波束向量和频域向量，且第2个传输层和第3个传输层所使用的波束向量可以是第0个传输层和第1个传输层所使用的波束向量的子集，第2个传输层和第3个传输层所使用的频域向量也可以是第0个传输层和第1个传输层所使用的频域向量的子集。当然，第2个传输层和第3个传输层所使用的波束向量也可以是终端设备基于第2个传输层和/或第3个传输层确定的，而不一定是第0个传输层和第1个传输层所使用的波束向量的子集；第2个传输层和第3个传输层所使用的频域向量也可以是终端设备基于第2个传输层和/或第3个传输层确定的，而不一定是第0个传输层和第1个传输层所使用的频域向量的子集。本申请对此不作限定。
当第2个传输层和第3个传输层所使用的波束向量是第0个传输层和第1个传输层所 使用的波束向量的子集，且，第2个传输层和第3个传输层所使用的频域向量是第0个传输层和第1个传输层所使用的频域向量的子集时，与第2个传输层和第3个传输层分别对应的两个第四指示信息在用于指示该第2个传输层和第3个传输层的波束向量和频域向量时，可以仅生成并指示一次L 2个波束向量和K 2个频域向量的指示信息。该L 2个波束向量和K 2个频域向量可以在L 1个波束向量和K 1个频域向量中指示，如，可以指示该L 2个波束向量在L 1个波束向量中的相对位置，也可以指示K 2个频域向量在K 1个频域向量中的相对位置。但第2个传输层和第3个传输层使用的空频分量矩阵可以通过不同的位图分别指示。与第2个传输层对应的第四指示信息中可以包括用于指示第2个传输层的空频分量矩阵的位图；与第3个传输层对应的第四指示信息中可以包括用于指示第3个传输层的空频分量矩阵的位图。
需要说明的是，由于传输层数可以由终端设备确定，网络设备在通过高层信令指示L 1、K 1以及T 1的时候，并不能预先确定传输层数。因此网络设备可以对L 1、K 1和T 1中的至少两项分别指示一个取值，终端设备可以根据传输层数以及预先定义的规则，确定是否需要改变的L 1、K 1或T 1的取值。
假设传输层数为R，极化方向数为2，网络设备可以基于如下公式确定第r个传输层上第n f个频域单元的预编码向量：
Figure PCTCN2019110342-appb-000155
其中，
Figure PCTCN2019110342-appb-000156
为与第r个传输层对应的归一化系数，
Figure PCTCN2019110342-appb-000157
为第r个传输层上的第t 1个波束向量，
Figure PCTCN2019110342-appb-000158
为第r个传输层上的第t 1个频域向量
Figure PCTCN2019110342-appb-000159
中的第n f个元素，
Figure PCTCN2019110342-appb-000160
表示与第r个传输层、第一极化方向上的第t 1个波束向量
Figure PCTCN2019110342-appb-000161
和第t 1个频域向量
Figure PCTCN2019110342-appb-000162
中的第n f个元素
Figure PCTCN2019110342-appb-000163
对应的加权系数，
Figure PCTCN2019110342-appb-000164
表示与第r个传输层、第二极化方向上的第t 1个波束向量
Figure PCTCN2019110342-appb-000165
和第t 1个频域向量
Figure PCTCN2019110342-appb-000166
中的第n f个元素
Figure PCTCN2019110342-appb-000167
对应的加权系数。
若各传输层上共用相同的L 1个波束向量和K 1个频域向量，则上式可以简化为：
Figure PCTCN2019110342-appb-000168
再进一步地，网络设备可以基于如下公式确定第n f个频域单元的预编码矩阵
Figure PCTCN2019110342-appb-000169
Figure PCTCN2019110342-appb-000170
其中，
Figure PCTCN2019110342-appb-000171
表示第0个传输层上第n f个频域单元的预编码向量，
Figure PCTCN2019110342-appb-000172
表示第R-1个传输层上第n f个频域单元的预编码向量。
基于上述技术方案，终端设备可以向网络设备指示一个或多个传输层上的一个或多个频域单元的预编码矩阵。
因此，终端设备通过向网络设备指示少量的波束向量、频域向量和空频分量矩阵，以便于网络设备恢复预编码向量。其中，频域向量可用于描述信道在频域上不同的变化规律。终端设备可以通过一个或多个频域向量的线性叠加来模拟信道在频域上的变化，充分挖掘了频域单元之间的关系，利用了频域的连续性，用较少的频域向量来描述多个频域单元的变化规律。相比于现有技术而言，无需基于每个频域单元独立地上报加权系数，在频域单元数量增加的情况下，并不会造成反馈开销的成倍增加。因此，可以在保证反馈精度的基础上大大减小反馈开销。
应理解，上文所提供的指示预编码向量的方法仅为一种可能的实现方式，而不应对本申请构成任何限定。例如，终端设备在确定了L 1个波束向量和K 1个频域向量之后，可以对该L 1个波束向量和K 1个频域向量所对应的M 1个空频分量矩阵分别反馈加权系数。终端设备可以对该M 1个空频分量矩阵中的T 1个空频分量矩阵的加权系数采用较高的反馈精度，而对剩余的M 1-T 1个空频分量矩阵的加权系数采用较低的反馈精度，以减小反馈开销。具体地，该T 1个空频分量矩阵中任意一个的加权系数的量化比特数可以大于剩余的M 1-T 1个空频分量矩阵中任意一个的加权系数的量化比特数。由此带来的反馈开销虽然较上文所提供的方法带来的反馈开销略大一些，但网络设备所恢复的预编码矩阵可能更接近于终端设备所确定的预编码矩阵，因此，近似精度更高。
需要说明的是，上文提供的指示和确定预编码矩阵的方法特别适用于频域单元数量多且连续的情况，该方法可以充分利用频域单元之间的相关性，对反馈开销进行压缩，减小了多个频域单元独立反馈带来的巨大开销。
然而，在某些情况下，频域单元数量并不是很多，或者频域单元之间并不连续，若沿用现有的各频域单元独立反馈的方式来指示各频域单元的预编码向量，可能并不会带来很大的反馈开销。并且在频域单元不连续的情况下，通过各频域单元独立反馈的方式可以保证较高的近似精度。因此，本申请另提供一种反馈PMI的方法，能够结合不同的场景，采用反馈模式反馈预编码向量，既能够保证近似精度，又节省了反馈开销。
下面结合图3详细说明本申请另一实施例提供的反馈PMI的方法。
图3是从设备交互的角度示出的本申请另一实施例提供的反馈PMI的方法300的示意性流程图。如图所示，该方法300可以包括步骤310至步骤340。下面详细说明方法300中的各步骤。
需要说明的是，在本实施例中，终端设备可以通过PMI来向网络设备指示预编码矩阵。该PMI可以包括上文中方法200中的第一指示信息和/或第四指示信息，还可以包括除第一指示信息和第四指示信息之外的其他可用于指示预编码矩阵的信息，本申请对此不作限定。此外，PMI仅为用于指示预编码矩阵的信息的名称，而不应对本申请构成任何限定，本申请并不排除在未来的协议中定义其他的名称来表示与PMI相同或相似功能的可能。
在步骤310中，网络设备生成第六指示信息，该第六指示信息用于指示PMI的反馈模式。
在本实施例中，PMI的反馈模式可以是上文所提供的反馈模式，也可以是其他反馈模式。具体地，该PMI的反馈模式可以为第一反馈模式或第二反馈模式。其中，第一反馈模式可以为仅基于波束向量集合反馈PMI的模式，第二反馈模式可以为基于波束向量集 合和频域向量集合反馈PMI的模式，或者可以为基于空频分量矩阵集合反馈PMI的模式。
由于空频分量矩阵集合与波束向量集合频域向量集合是相关联的，因此，对于第二反馈模式而言，基于波束向量集合和频域向量集合反馈PMI与基于空频分量矩阵集合反馈PMI可以认为是等价的。并且由于波束向量集合和频域向量集合与空频分量矩阵集合之间可以互相转换，还可以认为该第二反馈模式是基于波束向量集合和空频分量矩阵集合反馈PMI，或者，基于频域向量集合和空频分量矩阵集合反馈PMI。本申请对此不作限定。
其中，对于第一反馈模式所说的仅基于波束向量集合反馈PMI是相对于第二反馈模式而言。与第二反馈模式相比，该第一反馈模式可以仅基于波束向量集合反馈PMI，而不需要提供额外的向量集合或矩阵集合。换句话说，该第一反馈模式和第二反馈模式的区别在于，第一反馈模式未基于频域向量集合，而第二反馈模式基于频域向量集合。
从另一个角度来说，该第一反馈模式可以是频域单元独立反馈的模式，第二反馈模式可以是频域单元联合反馈的模式。
作为一种可能的实现方式，该第一反馈模式例如可以参考NR协议TS38.214R15中定义的基于type II码本反馈PMI的反馈模式。该第二反馈模式例如可以是上文中结合方法200描述的反馈模式。相比于第一反馈模式而言，该第二反馈模式可以理解为频域单元联合反馈的反馈模式。基于上文中的描述可以看到，第二反馈模式基于频域的连续性，利用频域单元之间的关系，将多个频域单元联合反馈，由此压缩频域反馈开销。尤其是在待上报的频域单元数较多的情况下，第二反馈模式相比于第一反馈模式，能够大大减小反馈开销。
在本实施例中，该第六指示信息可以显示地指示反馈模式。例如可以通过一个指示比特或指示字段指示反馈模式。如，指示比特置“0”时表示采用第一反馈模式，指示比特置“1”时表示采用第二反馈模式；或者，指示比特置“1”时表示采用第一反馈模式，指示比特置“0”时表示采用第二反馈模式。本申请对此不做限定。
该第六指示信息也可以通过其他信息来隐式地指示反馈模式。例如，当网络设备向终端设备指示频域向量的长度时，可以认为网络设备需要终端设备基于第二反馈模式反馈预编码向量。此时上文所述的用于指示频域向量的长度的第五指示信息可以理解为该第六指示信息的一例。
需要说明的是，空频分量向量的长度可以由频域向量的长度和波束向量的长度共同决定，因此，当网络设备向终端设备指示频域向量的长度时，终端设备可以基于端口选择向量和频域向量反馈PMI，也可以基于空频向量反馈PMI，本申请对此不作限定。
在步骤320中，网络设备发送第六指示信息。相应地，终端设备接收该第六指示信息。
可选地，该第六指示信息携带在RRC消息中。
网络设备向终端设备发送第六指示信息的具体方法可以与现有技术网络设备向终端设备发送信令的方式相同，为了简洁，这里省略对其具体过程的详细说明。
在步骤330中，终端设备基于第六指示信息所指示的反馈模式，生成PMI。
终端设备可以基于第六指示信息所指示的反馈模式，生成PMI。当终端设备基于第一反馈模式生成PMI时，终端设备生成PMI的具体过程可以与现有技术相同，为了简洁，这里不再赘述。当终端设备基于第二反馈模式生成PMI时，其具体实现过程在上文中方法200中已经做了详细说明，为了简洁，这里不再赘述。
在步骤340中，终端设备发送PMI。相应地，网络设备接收PMI。
终端设备可以将PMI发送给网络设备，以便网络设备确定预编码矩阵。该网络设备可以为上文中发送第六指示信息的网络设备，也可以是其他网络设备，本申请对此不作限定。应理解，图中示出的终端设备向网络设备发送PMI的步骤仅为示意，不应对本申请构成任何限定。
此后，网络设备便可以根据PMI确定预编码矩阵，进而确定用于数据传输的预编码矩阵。网络设备可以基于不同的反馈模式，根据PMI确定预编码矩阵。当终端设备基于第一反馈模式生成PMI时，网络设备根据PMI确定预编码矩阵的具体过程可以与现有技术相同，为了简洁，这里不再赘述。当终端设备基于第二反馈模式生成PMI时，网络设备根据PMI确定预编码矩阵的具体过程在上文中方法200中已经做了详细说明，为了简洁，这里不再赘述。
基于上述方法，终端设备可以基于网络设备的指示，采用相应的反馈模式反馈PMI。通过引入多种反馈模式，可以使用不同的测量情况，可以兼顾反馈精度和反馈开销，从而在两者间获得平衡。
应理解，上述实施例中，各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
以上，结合图2和图3详细说明了本申请实施例提供的指示和确定预编码向量的方法。以下，结合图4至图6详细说明本申请实施例提供的通信装置。
图4是本申请实施例提供的通信装置的示意性框图。如图所示，该通信装置1000可以包括通信单元1100和处理单元1200。
在一种可能的设计中，该通信装置1000可对应于上文方法实施例中的终端设备，例如，可以为终端设备，或者配置于终端设备中的芯片。
具体地，该通信装置1000可对应于根据本申请实施例的方法200或方法300中的终端设备，该通信装置1000可以包括用于执行图2中的方法200或图3中的方法300中的终端设备执行的方法的单元。并且，该通信装置1000中的各单元和上述其他操作和/或功能分别为了实现图2中的方法200或图3中的方法300的相应流程。
其中，当该通信装置1000用于执行图2中的方法200时，通信单元1100可用于执行方法200中的步骤220，处理单元1200可用于执行方法200中的步骤210。
具体地，该处理单元1200可用于生成第一指示信息，该第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，该T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，该L 1个波束向量和该K 1个频域向量与M 1个空频分量矩阵对应，该T 1个空频分量矩阵是该M 1个空频分量矩阵中的部分空频分量矩阵，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；该L 1个波束向量为该波束向量集合中的部分波束向量，和/或，该K 1个频域向量为该频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；该通信单元1100可用于发送该第一指示信息。
可选地，该通信单元1100还用于接收第二指示信息，该第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
可选地，该通信单元1100还用于接收第三指示信息，该第三指示信息用于指示T 1的取值。
可选地，该M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，该空频分量矩阵由该波束向量集合中的各波束向量和该频域向量集合中的各频域向量确定，且该空频分量矩阵集合中的每个空频分量矩阵由该波束向量集合中的一个波束向量和该频域向量集合中的一个频域向量唯一确定；
该第一指示信息包括该M 1个空频分量矩阵在该空频分量矩阵集合中的位置信息，或，该M 1个空频分量矩阵在空频分量矩阵集合的子集中的位置信息。
可选地，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量的共轭转置的乘积确定。
可选地，该M 1个空频分量矩阵中的每个空频分量矩阵由该K 1个频域向量中的一个频域向量和该L 1个波束向量中的一个波束向量的克罗内克积确定。
可选地，该T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
可选地，该处理单元1200还用于生成第四指示信息，该第四指示信息用于指示该波束向量集合中的L 2个波束向量、该频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，该T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，该L 2个波束向量和该K 2个频域向量与M 2个空频分量矩阵对应，该T 2个空频分量矩阵是该M 2个空频分量矩阵中的部分空频分量矩阵，该M 2个空频分量矩阵中的每个空频分量矩阵由该L 2个波束向量中的一个波束向量和该K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；该L 2个波束向量为该波束向量集合中的部分波束向量，和/或，该K 2个频域向量为该频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；该通信单元1100还用于发送该第四指示信息。
可选地，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
可选地，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
当该通信装置1000用于执行图3中的方法300时，通信单元1100可用于执行方法300中的步骤320和步骤340，处理单元1200可用于执行方法300中的步骤330。
应理解，各单元执行上述相应步骤的具体过程在上述方法实施例中已经详细说明，为了简洁，在此不再赘述。
还应理解，该通信装置1000为终端设备时，该通信装置1000中的通信单元1100可对应于图5中示出的终端设备2000中的收发器2020，该通信装置1000中的处理单元1200可对应于图5中示出的终端设备2000中的处理器2010。
还应理解，该通信装置1000为配置于终端设备中的芯片时，该通信装置1000中的通信单元1100可以为输入/输出接口。
在另一种可能的设计中，该通信装置1000可对应于上文方法实施例中的网络设备，例如，可以为网络设备，或者配置于网络设备中的芯片。
具体地，该通信装置1000可对应于根据本申请实施例的方法200或方法300中的网络设备，该通信装置1000可以包括用于执行图2中的方法200或图3中的方法300中的网络设备执行的方法的单元。并且，该通信装置1000中的各单元和上述其他操作和/或功 能分别为了实现图2中的方法200或图3中的方法300的相应流程。
其中，当该通信装置1000用于执行图4中的方法300时，通信单元1100可用于执行方法200中的步骤220，处理单元1200可用于执行方法200中的步骤230。
具体地，该通信单元1100可用于接收第一指示信息，该第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，该T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，该L 1个波束向量和该K 1个频域向量与M 1个空频分量矩阵对应，该T 1个空频分量矩阵是该M 1个空频分量矩阵中的部分空频分量矩阵，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；该L 1个波束向量为该波束向量集合中的部分波束向量，和/或，该K 1个频域向量为该频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；
该处理单元1200可用于基于该第一指示信息，确定一个或多个频域单元的预编码向量。
可选地，该通信单元1100还用于发送第二指示信息，该第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
可选地，该通信单元1100还用于发送第三指示信息，该第三指示信息用于指示T 1的取值。
可选地，该M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，该空频分量矩阵由该波束向量集合中的各波束向量和该频域向量集合中的各频域向量确定，且该空频分量矩阵集合中的每个空频分量矩阵由该波束向量集合中的一个波束向量和该频域向量集合中的一个频域向量唯一确定；
该第一指示信息包括该M 1个空频分量矩阵在该空频分量矩阵集合中的位置信息，或，该M 1个空频分量矩阵在空频分量矩阵集合的子集中的位置信息。
可选地，该M 1个空频分量矩阵中的每个空频分量矩阵由该L 1个波束向量中的一个波束向量和该K 1个频域向量中的一个频域向量的共轭转置的乘积确定。
可选地，该M 1个空频分量矩阵中的每个空频分量矩阵由该K 1个频域向量中的一个频域向量和该L 1个波束向量中的一个波束向量的克罗内克积确定。
可选地，该T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
可选地，该通信单元1100还用于接收第四指示信息，该第四指示信息用于指示该波束向量集合中的L 2个波束向量、该频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，该T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，该L 2个波束向量和该K 2个频域向量与M 2个空频分量矩阵对应，该T 2个空频分量矩阵是该M 2个空频分量矩阵中的部分空频分量矩阵，该M 2个空频分量矩阵中的每个空频分量矩阵由该L 2个波束向量中的一个波束向量和该K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；该L 2个波束向量为该波束向量集合中的部分波束向量，和/或，该K 2个频域向量为该频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；该处理单元1200还用于基于该第四指示信息，确定第二传输层上一个或多个频域单元的预编码向量。
可选地，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
可选地，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
当该通信装置1000用于执行图3中的方法300时，通信单元1100可用于执行方法300中的步骤320和步骤340，处理单元1200可用于执行方法300中的步骤310。
还应理解，该通信装置1000为网络设备时，该通信装置1000中的通信单元为可对应于图6中示出的网络设备3000中的收发器3200，该通信装置1000中的处理单元1200可对应于图6中示出的网络设备3000中的处理器3100。
还应理解，该通信装置1000为配置于网络设备中的芯片时，该通信装置1000中的通信单元1100可以为输入/输出接口。
图5是本申请实施例提供的终端设备2000的结构示意图。该终端设备2000可应用于如图1所示的系统中，执行上述方法实施例中终端设备的功能。如图所示，该终端设备2000包括处理器2010和收发器2020。可选地，该终端设备2000还包括存储器2030。其中，处理器2010、收发器2002和存储器2030之间可以通过内部连接通路互相通信，传递控制和/或数据信号，该存储器2030用于存储计算机程序，该处理器2010用于从该存储器2030中调用并运行该计算机程序，以控制该收发器2020收发信号。可选地，终端设备2000还可以包括天线2040，用于将收发器2020输出的上行数据或上行控制信令通过无线信号发送出去。
上述处理器2010可以和存储器2030可以合成一个处理装置，处理器2010用于执行存储器2030中存储的程序代码来实现上述功能。具体实现时，该存储器2030也可以集成在处理器2010中，或者独立于处理器2010。该处理器2010可以与图4中的处理单元对应。
上述收发器2020可以与图4中的通信单元对应，也可以称为收发单元。收发器2020可以包括接收器(或称接收机、接收电路)和发射器(或称发射机、发射电路)。其中，接收器用于接收信号，发射器用于发射信号。
应理解，图5所示的终端设备2000能够实现图2所示方法实施例中涉及终端设备的各个过程。终端设备2000中的各个模块的操作和/或功能，分别为了实现上述方法实施例中的相应流程。具体可参见上述方法实施例中的描述，为避免重复，此处适当省略详细描述。
上述处理器2010可以用于执行前面方法实施例中描述的由终端设备内部实现的动作，而收发器2020可以用于执行前面方法实施例中描述的终端设备向网络设备发送或从网络设备接收的动作。具体请见前面方法实施例中的描述，此处不再赘述。
可选地，上述终端设备2000还可以包括电源2050，用于给终端设备中的各种器件或电路提供电源。
除此之外，为了使得终端设备的功能更加完善，该终端设备2000还可以包括输入单元2060、显示单元2070、音频电路2080、摄像头2090和传感器2100等中的一个或多个，所述音频电路还可以包括扬声器2082、麦克风2084等。
图6是本申请实施例提供的网络设备的结构示意图，例如可以为基站的结构示意图。该基站3000可应用于如图1所示的系统中，执行上述方法实施例中网络设备的功能。如图所示，该基站3000可以包括一个或多个射频单元，如远端射频单元(remote radio unit， RRU)3100和一个或多个基带单元(BBU)(也可称为分布式单元(DU))3200。所述RRU 3100可以称为收发单元，与图4中的通信单元1200对应。可选地，该收发单元3100还可以称为收发机、收发电路、或者收发器等等，其可以包括至少一个天线3101和射频单元3102。可选地，收发单元3100可以包括接收单元和发送单元，接收单元可以对应于接收器(或称接收机、接收电路)，发送单元可以对应于发射器(或称发射机、发射电路)。所述RRU 3100部分主要用于射频信号的收发以及射频信号与基带信号的转换，例如用于向终端设备发送指示信息。所述BBU 3200部分主要用于进行基带处理，对基站进行控制等。所述RRU 3100与BBU 3200可以是物理上设置在一起，也可以物理上分离设置的，即分布式基站。
所述BBU 3200为基站的控制中心，也可以称为处理单元，可以与图4中的处理单元1100对应，主要用于完成基带处理功能，如信道编码，复用，调制，扩频等等。例如所述BBU(处理单元)可以用于控制基站执行上述方法实施例中关于网络设备的操作流程，例如，生成上述指示信息等。
在一个示例中，所述BBU 3200可以由一个或多个单板构成，多个单板可以共同支持单一接入制式的无线接入网(如LTE网)，也可以分别支持不同接入制式的无线接入网(如LTE网，5G网或其他网)。所述BBU 3200还包括存储器3201和处理器3202。所述存储器3201用以存储必要的指令和数据。所述处理器3202用于控制基站进行必要的动作，例如用于控制基站执行上述方法实施例中关于网络设备的操作流程。所述存储器3201和处理器3202可以服务于一个或多个单板。也就是说，可以每个单板上单独设置存储器和处理器。也可以是多个单板共用相同的存储器和处理器。此外每个单板上还可以设置有必要的电路。
应理解，图6所示的基站3000能够实现图2的方法实施例中涉及网络设备的各个过程。基站3000中的各个模块的操作和/或功能，分别为了实现上述方法实施例中的相应流程。具体可参见上述方法实施例中的描述，为避免重复，此处适当省略详细描述。
上述BBU 3200可以用于执行前面方法实施例中描述的由网络设备内部实现的动作，而RRU 3100可以用于执行前面方法实施例中描述的网络设备向终端设备发送或从终端设备接收的动作。具体请见前面方法实施例中的描述，此处不再赘述。
本申请实施例还提供了一种处理装置，包括处理器和接口；所述处理器用于执行上述任一方法实施例的方法。
应理解，上述处理装置可以是一个或多个芯片。例如，该处理装置可以是现场可编程门阵列(field programmable gate array，FPGA)，可以是专用集成芯片(application specific integrated circuit，ASIC)，还可以是系统芯片(system on chip，SoC)，还可以是中央处理器(central processor unit，CPU)，还可以是网络处理器(network processor，NP)，还可以是数字信号处理电路(digital signal processor，DSP)，还可以是微控制器(micro controller unit，MCU)，还可以是可编程控制器(programmable logic device，PLD)或其他集成芯片。
在实现过程中，上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。结合本申请实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器， 闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
应注意，本申请实施例中的处理器可以是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法实施例的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现场可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。
可以理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。应注意，本文描述的系统和方法的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
根据本申请实施例提供的方法，本申请还提供一种计算机程序产品，该计算机程序产品包括：计算机程序代码，当该计算机程序代码在计算机上运行时，使得该计算机执行图2所示实施例中任意一个实施例的方法。
根据本申请实施例提供的方法，本申请还提供一种计算机可读介质，该计算机可读介质存储有程序代码，当该程序代码在计算机上运行时，使得该计算机执行图2所示实施例中任意一个实施例的方法。
根据本申请实施例提供的方法，本申请还提供一种系统，其包括前述的一个或多个终端设备以及一个或多个网络设备。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中， 或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line，DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，高密度数字视频光盘(digital video disc，DVD))、或者半导体介质(例如，固态硬盘(solid state disc，SSD))等。
上述各个装置实施例中网络设备与终端设备和方法实施例中的网络设备或终端设备完全对应，由相应的模块或单元执行相应的步骤，例如通信单元(收发器)执行方法实施例中接收或发送的步骤，除发送、接收外的其它步骤可以由处理单元(处理器)执行。具体单元的功能可以参考相应的方法实施例。其中，处理器可以为一个或多个。
在本说明书中使用的术语“部件”、“模块”、“系统”等用于表示计算机相关的实体、硬件、固件、硬件和软件的组合、软件、或执行中的软件。例如，部件可以是但不限于，在处理器上运行的进程、处理器、对象、可执行文件、执行线程、程序和/或计算机。通过图示，在计算设备上运行的应用和计算设备都可以是部件。一个或多个部件可驻留在进程和/或执行线程中，部件可位于一个计算机上和/或分布在2个或更多个计算机之间。此外，这些部件可从在上面存储有各种数据结构的各种计算机可读介质执行。部件可例如根据具有一个或多个数据分组(例如来自与本地系统、分布式系统和/或网络间的另一部件交互的二个部件的数据，例如通过信号与其它系统交互的互联网)的信号通过本地和/或远程进程来通信。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各种说明性逻辑块(illustrative logical block)和步骤(step)，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
在上述实施例中，各功能单元的功能可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令(程序)。在计算机上加载和执行所述计算机程序指令(程序)时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘(solid state disk，SSD))等。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (52)

  1. 一种指示预编码向量的方法，其特征在于，包括：
    生成第一指示信息，所述第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，所述T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，所述L 1个波束向量和所述K 1个频域向量与M 1个空频分量矩阵对应，所述T 1个空频分量矩阵是所述M 1个空频分量矩阵中的部分空频分量矩阵，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；所述L 1个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 1个频域向量为所述频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；
    发送所述第一指示信息。
  2. 如权利要求1所述的方法，其特征在于，所述方法还包括：
    接收第二指示信息，所述第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
  3. 如权利要求1或2所述的方法，其特征在于，所述方法还包括：
    接收第三指示信息，所述第三指示信息用于指示T 1的取值。
  4. 如权利要求1至3中任一项所述的方法，其特征在于，所述第一指示信息包括所述L 1个波束向量在所述波束向量集合中的位置信息和所述K 1个频域向量在所述频域向量集合中的位置信息。
  5. 如权利要求1至3中任一项所述的方法，其特征在于，所述M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，所述空频分量矩阵由所述波束向量集合中的各波束向量和所述频域向量集合中的各频域向量确定，且所述空频分量矩阵集合中的每个空频分量矩阵由所述波束向量集合中的一个波束向量和所述频域向量集合中的一个频域向量唯一确定；
    所述第一指示信息包括所述M 1个空频分量矩阵在所述空频分量矩阵集合中的位置信息，或，所述M 1个空频分量矩阵在所述空频分量矩阵集合的子集中的位置信息。
  6. 如权利要求1至5中任一项所述的方法，其特征在于，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量的共轭转置的乘积确定；或
    所述M 1个空频分量矩阵中的每个空频分量矩阵由所述K 1个频域向量中的一个频域向量和所述L 1个波束向量中的一个波束向量的克罗内克积确定。
  7. 如权利要求1至6中任一项所述的方法，其特征在于，所述T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
  8. 如权利要求1至7中任一项所述的方法，其特征在于，所述方法还包括：
    生成第四指示信息，所述第四指示信息用于指示所述波束向量集合中的L 2个波束向量、所述频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，所述T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，所述L 2个波束向量和所述K 2个频域向量与M 2个空频分量矩阵对应，所述T 2个空频分量矩阵是所 述M 2个空频分量矩阵中的部分空频分量矩阵，所述M 2个空频分量矩阵中的每个空频分量矩阵由所述L 2个波束向量中的一个波束向量和所述K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；所述L 2个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 2个频域向量为所述频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；
    发送所述第四指示信息。
  9. 如权利要求8所述的方法，其特征在于，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
  10. 如权利要求8所述的方法，其特征在于，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
  11. 一种确定预编码向量的方法，其特征在于，包括：
    接收第一指示信息，所述第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，所述T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，所述L 1个波束向量和所述K 1个频域向量与M 1个空频分量矩阵对应，所述T 1个空频分量矩阵是所述M 1个空频分量矩阵中的部分空频分量矩阵，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；所述L 1个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 1个频域向量为所述频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；
    基于所述第一指示信息，确定一个或多个频域单元的预编码向量。
  12. 如权利要求11所述的方法，其特征在于，所述方法还包括：
    发送第二指示信息，所述第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
  13. 如权利要求11或12所述的方法，其特征在于，所述方法还包括：
    发送第三指示信息，所述第三指示信息用于指示T 1的取值。
  14. 如权利要求11至13中任一项所述的方法，其特征在于，所述第一指示信息包括所述L 1个波束向量在所述波束向量集合中的位置信息和所述K 1个频域向量在所述频域向量集合中的位置信息。
  15. 如权利要求11至13中任一项所述的方法，其特征在于，所述M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，所述空频分量矩阵由所述波束向量集合中的各波束向量和所述频域向量集合中的各频域向量确定，且所述空频分量矩阵集合中的每个空频分量矩阵由所述波束向量集合中的一个波束向量和所述频域向量集合中的一个频域向量唯一确定；
    所述第一指示信息包括所述M 1个空频分量矩阵在所述空频分量矩阵集合中的位置信息，或，所述M 1个空频分量矩阵在所述空频分量矩阵集合的子集中的位置信息。
  16. 如权利要求11至15中任一项所述的方法，其特征在于，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量的共轭转置的乘积确定；或
    所述M 1个空频分量矩阵中的每个空频分量矩阵由所述K 1个频域向量中的一个频域向量和所述L 1个波束向量中的一个波束向量的克罗内克积确定。
  17. 如权利要求11至16中任一项所述的方法，其特征在于，所述T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
  18. 如权利要求11至17中任一项所述的方法，其特征在于，所述方法还包括：
    接收第四指示信息，所述第四指示信息用于指示所述波束向量集合中的L 2个波束向量、所述频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，所述T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，所述L 2个波束向量和所述K 2个频域向量与M 2个空频分量矩阵对应，所述T 2个空频分量矩阵是所述M 2个空频分量矩阵中的部分空频分量矩阵，所述M 2个空频分量矩阵中的每个空频分量矩阵由所述L 2个波束向量中的一个波束向量和所述K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；所述L 2个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 2个频域向量为所述频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；
    基于所述第四指示信息，确定所述第二传输层上一个或多个频域单元的预编码向量。
  19. 如权利要求18所述的方法，其特征在于，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
  20. 如权利要求18所述的方法，其特征在于，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
  21. 一种通信装置，其特征在于，包括：
    处理单元，用于生成第一指示信息，所述第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，所述T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，所述L 1个波束向量和所述K 1个频域向量与M 1个空频分量矩阵对应，所述T 1个空频分量矩阵是所述M 1个空频分量矩阵中的部分空频分量矩阵，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；所述L 1个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 1个频域向量为所述频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；
    通信单元，用于发送所述第一指示信息。
  22. 如权利要求21所述的装置，其特征在于，所述通信单元还用于接收第二指示信息，所述第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
  23. 如权利要求21或22所述的方法，其特征在于，所述通信单元还用于接收第三指示信息，所述第三指示信息用于指示T 1的取值。
  24. 如权利要求21至23中任一项所述的装置，其特征在于，所述第一指示信息包括所述L 1个波束向量在所述波束向量集合中的位置信息和所述K 1个频域向量在所述频域向量集合中的位置信息。
  25. 如权利要求21至23中任一项所述的装置，其特征在于，所述M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，所述空频分量矩阵由所述波束向量集合中的各波束向量和所述频域向量集合中的各频域向量确定，且所述空频分量矩阵集合中的每个空频分量矩阵由所述波束向量集合中的一个波束向量和所述频域向量集合中的一个频域向量唯一确定；
    所述第一指示信息包括所述M 1个空频分量矩阵在所述空频分量矩阵集合中的位置信息，或，所述M 1个空频分量矩阵在所述空频分量矩阵集合的子集中的位置信息。
  26. 如权利要求21至25中任一项所述的装置，其特征在于，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中 的一个频域向量的共轭转置的乘积确定；或
    所述M 1个空频分量矩阵中的每个空频分量矩阵由所述K 1个频域向量中的一个频域向量和所述L 1个波束向量中的一个波束向量的克罗内克积确定。
  27. 如权利要求21至26中任一项所述的装置，其特征在于，所述T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
  28. 如权利要求21至27中任一项所述的装置，其特征在于，所述处理单元还用于生成第四指示信息，所述第四指示信息用于指示所述波束向量集合中的L 2个波束向量、所述频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，所述T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，所述L 2个波束向量和所述K 2个频域向量与M 2个空频分量矩阵对应，所述T 2个空频分量矩阵是所述M 2个空频分量矩阵中的部分空频分量矩阵，所述M 2个空频分量矩阵中的每个空频分量矩阵由所述L 2个波束向量中的一个波束向量和所述K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；所述L 2个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 2个频域向量为所述频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；
    所述通信单元还用于发送所述第四指示信息。
  29. 如权利要求28所述的装置，其特征在于，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
  30. 如权利要求28所述的装置，其特征在于，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
  31. 一种通信装置，其特征在于，包括：
    通信单元，用于接收第一指示信息，所述第一指示信息用于指示波束向量集合中的L 1个波束向量、频域向量集合中的K 1个频域向量以及T 1个空频分量矩阵，所述T 1个空频分量矩阵的加权和用于确定一个或多个频域单元的预编码向量；其中，所述L 1个波束向量和所述K 1个频域向量与M 1个空频分量矩阵对应，所述T 1个空频分量矩阵是所述M 1个空频分量矩阵中的部分空频分量矩阵，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量唯一确定，且M 1＝L 1×K 1；所述L 1个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 1个频域向量为所述频域向量集合中的部分频域向量，M 1、L 1、K 1和T 1均为正整数；
    处理单元，用于基于所述第一指示信息，确定一个或多个频域单元的预编码向量。
  32. 如权利要求31所述的装置，其特征在于，所述通信单元还用于发送第二指示信息，所述第二指示信息用于指示M 1、L 1和K 1中一项或多项的取值。
  33. 如权利要求31或32所述的装置，其特征在于，所述通信单元还用于发送第三指示信息，所述第三指示信息用于指示T 1的取值。
  34. 如权利要求31至33中任一项所述的装置，其特征在于，所述第一指示信息包括所述L 1个波束向量在所述波束向量集合中的位置信息和所述K 1个频域向量在所述频域向量集合中的位置信息。
  35. 如权利要求31至33中任一项所述的装置，其特征在于，所述M 1个空频分量矩阵选自空频分量矩阵集合或空频分量矩阵集合的子集，所述空频分量矩阵由所述波束向量集合中的各波束向量和所述频域向量集合中的各频域向量确定，且所述空频分量矩阵集合中的每个空频分量矩阵由所述波束向量集合中的一个波束向量和所述频域向量集合中的一个频域向量唯一确定；
    所述第一指示信息包括所述M 1个空频分量矩阵在所述空频分量矩阵集合中的位置信息，或，所述M 1个空频分量矩阵在所述空频分量矩阵集合的子集中的位置信息。
  36. 如权利要求31至35中任一项所述的装置，其特征在于，所述M 1个空频分量矩阵中的每个空频分量矩阵由所述L 1个波束向量中的一个波束向量和所述K 1个频域向量中的一个频域向量的共轭转置的乘积确定；或
    所述M 1个空频分量矩阵中的每个空频分量矩阵由所述K 1个频域向量中的一个频域向量和所述L 1个波束向量中的一个波束向量的克罗内克积确定。
  37. 如权利要求31至36中任一项所述的装置，其特征在于，所述T 1个空频分量矩阵的加权和用于确定第一传输层上一个或多个频域单元的预编码向量。
  38. 如权利要求31至37中任一项所述的装置，其特征在于，所述通信单元还用于接收第四指示信息，所述第四指示信息用于指示所述波束向量集合中的L 2个波束向量、所述频域向量集合中的K 2个频域向量以及T 2个空频分量矩阵，所述T 2个空频分量矩阵的加权和用于确定第二传输层上一个或多个频域单元的预编码向量；其中，所述L 2个波束向量和所述K 2个频域向量与M 2个空频分量矩阵对应，所述T 2个空频分量矩阵是所述M 2个空频分量矩阵中的部分空频分量矩阵，所述M 2个空频分量矩阵中的每个空频分量矩阵由所述L 2个波束向量中的一个波束向量和所述K 2个频域向量中的一个频域向量唯一确定，且M 2＝L 2×K 2；所述L 2个波束向量为所述波束向量集合中的部分波束向量，和/或，所述K 2个频域向量为所述频域向量集合中的部分频域向量，M 2、L 2、K 2和T 2均为正整数；
    基于所述第四指示信息，确定所述第二传输层上一个或多个频域单元的预编码向量。
  39. 如权利要求38所述的装置，其特征在于，L 1＝L 2，K 1＝K 2，且T 1＝T 2。
  40. 如权利要求38所述的装置，其特征在于，L 1＞L 2，或K 1＞K 2，或T 1＞T 2。
  41. 一种通信装置，其特征在于，包括至少一个处理器，所述至少一个处理器用于执行如权利要求1至10中任一项所述的方法。
  42. 一种通信装置，其特征在于，包括至少一个处理器，所述至少一个处理器用于执行如权利要求11至20中任一项所述的方法。
  43. 一种通信装置，其特征在于，所述装置用于实现如权利要求1至10中任一项所述的方法。
  44. 一种通信装置，其特征在于，所述装置用于实现如权利要求11至20中任一项所述的方法。
  45. 一种处理装置，其特征在于，包括：
    处理器，所述处理器用于执行存储器中存储的计算机程序，以使得所述装置实现如权利要求1至10中任一项所述的方法。
  46. 一种处理装置，其特征在于，包括：
    处理器，所述处理器用于执行存储器中存储的计算机程序，以使得所述装置实现如权利要求11至20中任一项所述的方法。
  47. 一种处理装置，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于从所述存储器调用并运行所述计算机程序，以使得所述装置实现如权利要求1至10中任一项所述的方法。
  48. 一种处理装置，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于从所述存储器调用并运行所述计算机程序，以使得所述装置实现如权利要求11至20中任一项所述的方法。
  49. 一种计算机可读介质，其特征在于，包括计算机程序，当所述计算机程序在计算机上运行时，使得所述计算机执行如权利要求1至10中任一项所述的方法。
  50. 一种计算机可读介质，其特征在于，包括计算机程序，当所述计算机程序在计算机上运行时，使得所述计算机执行如权利要求11至20中任一项所述的方法。
  51. 一种计算机程序产品，所述计算机程序产品包括计算机程序，当所述计算机程序在计算机上运行时，使得计算机执行如权利要求1至10中任一项所述的方法。
  52. 一种计算机程序产品，所述计算机程序产品包括计算机程序，当所述计算机程序在计算机上运行时，使得计算机执行如权利要求11至20中任一项所述的方法。
PCT/CN2019/110342 2018-10-16 2019-10-10 指示和确定预编码向量的方法以及通信装置 WO2020078251A1 (zh)

Priority Applications (7)

Application Number Priority Date Filing Date Title
BR112021007195-4A BR112021007195A2 (pt) 2018-10-16 2019-10-10 método para indicar vetor de pré-codificação, método para determinar vetor de pré-codificação, e aparelho de comunicações
KR1020217014727A KR102495785B1 (ko) 2018-10-16 2019-10-10 프리코딩 벡터를 지시하는 방법, 프리코딩 벡터를 결정하는 방법 및 통신 장치
JP2021521259A JP7210718B2 (ja) 2018-10-16 2019-10-10 プリコーディングベクトルを示すための方法、プリコーディングベクトルを決定するための方法、および通信装置
EP19872662.2A EP3876429A4 (en) 2018-10-16 2019-10-10 PRECODING VECTOR INDICATION PROCESS, PRECODING VECTOR DETERMINATION PROCESS, AND COMMUNICATION APPARATUS
US17/230,523 US11456786B2 (en) 2018-10-16 2021-04-14 Method for indicating precoding vector, method for determining precoding vector, and communications apparatus
US17/900,501 US11848729B2 (en) 2018-10-16 2022-08-31 Method for indicating precoding vector, method for determining precoding vector, and communications apparatus
US18/502,785 US20240072858A1 (en) 2018-10-16 2023-11-06 Method for indicating precoding vector, method for determining precoding vector, and communications apparatus

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201811205381 2018-10-16
CN201811205381.1 2018-10-16
CN201811281059.7A CN111064499B (zh) 2018-10-16 2018-10-30 指示和确定预编码向量的方法以及通信装置
CN201811281059.7 2018-10-30

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/230,523 Continuation US11456786B2 (en) 2018-10-16 2021-04-14 Method for indicating precoding vector, method for determining precoding vector, and communications apparatus

Publications (1)

Publication Number Publication Date
WO2020078251A1 true WO2020078251A1 (zh) 2020-04-23

Family

ID=70284405

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/110342 WO2020078251A1 (zh) 2018-10-16 2019-10-10 指示和确定预编码向量的方法以及通信装置

Country Status (1)

Country Link
WO (1) WO2020078251A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050265290A1 (en) * 2004-05-28 2005-12-01 Hochwald Bertrand M Feedback method for channel state information of a wireless link
CN102271031A (zh) * 2011-08-09 2011-12-07 中兴通讯股份有限公司 一种信道信息反馈的方法和系统
CN106130614A (zh) * 2016-07-15 2016-11-16 上海华为技术有限公司 一种低开销反馈的方法、用户设备以及基站

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050265290A1 (en) * 2004-05-28 2005-12-01 Hochwald Bertrand M Feedback method for channel state information of a wireless link
CN102271031A (zh) * 2011-08-09 2011-12-07 中兴通讯股份有限公司 一种信道信息反馈的方法和系统
CN106130614A (zh) * 2016-07-15 2016-11-16 上海华为技术有限公司 一种低开销反馈的方法、用户设备以及基站

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HUAWEI ET AL.: "Discussion on CSI enhancement for MU-MIMO", 3GPP TSG RAN WG1 MEETING #94BIS R1-1810103, 12 October 2018 (2018-10-12), pages 3 - 5, XP051517518 *
HUAWEI ET AL.: "Enhancements on CSI reporting and codebook design", 3GPP TSG RAN WG1 MEETING #94 RL-1808949, 24 August 2018 (2018-08-24), XP051516324 *

Similar Documents

Publication Publication Date Title
CN110855336B (zh) 指示和确定预编码向量的方法以及通信装置
WO2020125510A1 (zh) 一种信道测量方法和通信装置
CN111064499B (zh) 指示和确定预编码向量的方法以及通信装置
WO2020125534A1 (zh) 一种信道测量方法和通信装置
WO2020083057A1 (zh) 指示和确定预编码向量的方法以及通信装置
WO2020143577A1 (zh) 参数配置方法和通信装置
JP7238167B2 (ja) プリコーディング行列表示及び決定方法、及び通信装置
WO2020125511A1 (zh) 一种信道测量方法和通信装置
WO2020143580A1 (zh) 用于构建预编码向量的向量指示方法和通信装置
US11362707B2 (en) Precoding vector indicating and determining method and communications apparatus
CN111757382A (zh) 指示信道状态信息的方法以及通信装置
WO2020244496A1 (zh) 一种信道测量方法和通信装置
WO2020073788A1 (zh) 指示和确定预编码向量的方法以及通信装置
WO2020135101A1 (zh) 用于构建预编码向量的向量指示方法以及通信装置
CN110875767B (zh) 指示和确定预编码向量的方法和通信装置
WO2020078251A1 (zh) 指示和确定预编码向量的方法以及通信装置
CN112436874A (zh) 用于构建预编码向量的向量指示方法以及通信装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19872662

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021521259

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112021007195

Country of ref document: BR

ENP Entry into the national phase

Ref document number: 20217014727

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2019872662

Country of ref document: EP

Effective date: 20210514

ENP Entry into the national phase

Ref document number: 112021007195

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20210415