WO2022160344A1 - 基站和终端 - Google Patents

基站和终端 Download PDF

Info

Publication number
WO2022160344A1
WO2022160344A1 PCT/CN2021/074690 CN2021074690W WO2022160344A1 WO 2022160344 A1 WO2022160344 A1 WO 2022160344A1 CN 2021074690 W CN2021074690 W CN 2021074690W WO 2022160344 A1 WO2022160344 A1 WO 2022160344A1
Authority
WO
WIPO (PCT)
Prior art keywords
channel
information
base station
terminal
network
Prior art date
Application number
PCT/CN2021/074690
Other languages
English (en)
French (fr)
Inventor
王新
侯晓林
李安新
陈岚
原田浩树
Original Assignee
株式会社Ntt都科摩
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
Application filed by 株式会社Ntt都科摩 filed Critical 株式会社Ntt都科摩
Priority to PCT/CN2021/074690 priority Critical patent/WO2022160344A1/zh
Priority to CN202180089526.XA priority patent/CN116746077A/zh
Priority to US18/258,641 priority patent/US20240056142A1/en
Publication of WO2022160344A1 publication Critical patent/WO2022160344A1/zh

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
    • H04B7/0478—Special codebook structures directed to feedback optimisation
    • 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
    • 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
    • H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • 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
    • H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0617—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal for beam forming

Definitions

  • the present disclosure relates to the field of wireless communication, and more particularly, to a base station and a corresponding terminal for performing channel reconstruction according to feedback from a terminal.
  • a base station sends a downlink reference signal to a UE, and the UE performs channel estimation according to the downlink reference signal and sends type I or type II precoding matrix indication (PMI) information to the base station.
  • the base station may determine the corresponding codeword and the combination coefficients related to the symbols according to the PMI information sent by the UE to reconstruct the channel, and use the reconstructed channel to perform downlink precoding.
  • the type II PMI information sent by the UE is the PMI information at the subband level, which leads to a larger quantization granularity in the spatial domain and frequency domain, and a larger quantization granularity of the combined coefficient.
  • the base station reconstructs the subband-level channel based on the subband-level PMI information sent by the UE.
  • the base station can use the reconstructed channel for precoding in units of physical resource block bundling (PRB bundling), and the granularity of physical resource block bundling is usually much smaller than that of PMI information.
  • PRB bundling physical resource block bundling
  • one subband may include 16 resource blocks (RBs).
  • the UE may transmit Type II PMI information at the subband (ie, 16 RBs) level.
  • the minimum size of physical resource block bundling may be 2 RBs, which is much smaller than the granularity of PMI information.
  • a base station includes: a receiving unit configured to receive precoding matrix indication information of a first granularity from a terminal; a processing unit configured to perform channel reconstruction according to the precoding matrix indication information to obtain a first channel, using a super-resolution network Perform interpolation and denoising processing on the channel of the first granularity to obtain a second channel, and perform downlink precoding on the second channel, wherein the first channel has the first granularity, the second channel having a second particle size, the second particle size being finer than the first particle size.
  • the base station includes: a receiving unit configured to receive precoding matrix indication information of a first granularity from a terminal; a processing unit configured to perform channel reconstruction through a first sub-network according to the precoding matrix indication information of the first granularity, Interpolating and denoising to obtain a second channel, and performing downlink precoding on the channel of the second granularity, wherein the second granularity is finer than the first granularity.
  • the base station includes: a sending unit configured to send first channel state information reference information of a first density to a terminal; a receiving unit configured to receive first feedback information for the first channel state information reference information from the terminal ; wherein the first feedback information includes the first channel response information obtained by the terminal performing the first channel estimation according to the first channel state information reference information, and the first channel state information reference information processed by the terminal according to the down-sampling process The determined precoding matrix indication information.
  • a terminal includes: a receiving unit configured to receive first channel state information reference information of a first density; a processing unit configured to perform first channel estimation according to the first channel state information reference information to obtain first channel response information , and perform down-sampling processing on the first channel state information reference information, and determine precoding matrix indication information according to the down-sampled channel state information reference information; and a sending unit, configured to send the first channel to the base station.
  • a receiving unit configured to receive first channel state information reference information of a first density
  • a processing unit configured to perform first channel estimation according to the first channel state information reference information to obtain first channel response information , and perform down-sampling processing on the first channel state information reference information, and determine precoding matrix indication information according to the down-sampled channel state information reference information
  • a sending unit configured to send the first channel to the base station.
  • Response information and precoding matrix indication information configured to send the first channel to the base station.
  • FIG. 1 is a schematic diagram illustrating that in a communication system, a base station performs channel reconstruction according to feedback from a terminal.
  • FIG. 2 is a schematic diagram illustrating channel state information (CSI) and ideal CSI determined according to PMI information fed back by a terminal in an existing communication system.
  • CSI channel state information
  • FIG. 3 is a schematic block diagram illustrating a base station according to one embodiment of the present disclosure.
  • FIG. 4 is a schematic block diagram illustrating a base station according to another embodiment of the present disclosure.
  • FIG. 5 is a schematic block diagram illustrating a base station according to another embodiment of the present disclosure.
  • FIG. 6 is a schematic diagram illustrating training of a neural network using a PMI training dataset and a corresponding channel response dataset, according to one example of the present disclosure.
  • FIG. 7 is a schematic block diagram illustrating a terminal according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic diagram illustrating that the processing unit processes the first channel state information reference information according to an example of the present disclosure.
  • FIG. 9 is a flowchart of a channel processing method according to one embodiment of the present disclosure.
  • FIG. 10 is a flowchart of a channel processing method according to another embodiment of the present disclosure.
  • FIG. 11 is a flowchart of a reference signal transmission method according to an embodiment of the present disclosure.
  • FIG. 12 is a flowchart of an information transmission method according to an embodiment of the present disclosure.
  • FIG. 13 is a schematic diagram of a hardware structure of a device involved according to an embodiment of the present disclosure.
  • 14A-14C are schematic diagrams illustrating a first sub-network structure according to an embodiment of the present disclosure.
  • the same reference numbers refer to the same elements throughout.
  • the terminals described herein are illustrative only and should not be construed as limiting the scope of the present disclosure.
  • the terminals described here may include various types of terminals, such as user terminals (User Equipment, UE), mobile terminals (or referred to as mobile stations) or fixed terminals.
  • UE User Equipment
  • mobile terminals or referred to as mobile stations
  • fixed terminals for the sake of convenience, in the following, sometimes mutual The terminal and the UE are used interchangeably.
  • FIG. 1 is a schematic diagram illustrating that in a communication system, a base station performs channel reconstruction according to feedback from a terminal.
  • the terminal 110 performs channel estimation according to the downlink reference signal, and obtains type I or type II precoding matrix indication (PMI) information according to the channel estimation result to send to the base station 120 .
  • the base station 120 reconstructs the channel according to the PMI information sent by the UE, and performs downlink precoding using the reconstructed channel.
  • the ideal space-time channel H expected to be obtained by the base station can be expressed by the following formula (1):
  • the space-time domain channel H can be regarded as the superposition of N multipath components, and the amplitude of each multipath component is ⁇ i , which can be written as the delay ⁇ i , and the horizontal angle of arrival Vertical angle of arrival ⁇ i , a function F of phase ⁇ i .
  • ⁇ i the delay
  • F the horizontal angle of arrival
  • FIG. 2 is a schematic diagram illustrating channel state information (CSI) and ideal CSI determined according to PMI information fed back by a terminal in an existing communication system.
  • CSI channel state information
  • the CSI determined according to the PMI information fed back by the terminal is represented by a diamond
  • the ideal channel state information is represented by a circle.
  • the granularity of the CSI determined according to the PMI information fed back by the terminal is far coarser than the ideal channel state information.
  • the base station uses the CSI directly reconstructed based on the PMI, the spectral efficiency is lost by more than 50% compared to using the ideal CSI.
  • the base station can use the rebuilt channel for precoding in units of physical resource block binding, and the granularity of physical resource block binding is usually much smaller than that of PMI information.
  • FIG. 3 is a schematic block diagram illustrating a base station according to one embodiment of the present disclosure.
  • the base station 300 may include a receiving unit 310 and a processing unit 320 .
  • the base station 300 may further include other components, however, since these components are not related to the content of the embodiments of the present disclosure, their illustration and description are omitted here.
  • the receiving unit 310 of the base station 300 receives the precoding matrix indication information of the first granularity from the terminal.
  • the processing unit 320 performs channel reconstruction according to the precoding matrix indication information to obtain a first channel, where the first channel has a first granularity.
  • the receiving unit 310 may receive the type II PMI information of the subband level from the terminal.
  • the type II PMI information may include spatial-domain codeword selection information, amplitude information and phase information of codeword combining coefficients at the broadband and subband levels, and may also include frequency-domain codeword selection information and spatial-frequency domain codeword combining coefficients.
  • the processing unit 320 may use the amplitude information and the phase information in the PMI information received by the receiving unit 310 to respectively perform amplitude and phase weighting on the spatial domain (also referred to as "beam domain")-frequency domain channel codewords of the multiple beams, And the weighted vectors are combined to obtain a first channel with subband level.
  • the first channel of the subband level obtained after the combination can be represented in the form of a space-frequency domain channel matrix, wherein the space-domain value of the space-frequency domain channel matrix can be the number of antennas through which the base station 300 communicates with the terminal, and the frequency-domain value It may be determined according to the number of subbands in which the base station 300 communicates with the terminal.
  • the processing unit 320 performs interpolation and denoising processing on the channel of the first granularity using a super-resolution network to obtain a second channel, wherein the second channel has a second granularity, and the second granularity is finer than the first granularity .
  • the channels of the first granularity may be subband-level channels.
  • the channel of the second granularity may be a sub-carrier level or a resource block (RB) level channel.
  • the processing unit 320 may preprocess the first channel to facilitate subsequent operations of the super-resolution network. For example, in the case that the first channel is a space-frequency-domain channel, the processing unit 320 may perform Fourier transform on the first channel, so as to convert the space-frequency-domain channel into a beam-delay-domain channel. In addition, since in the beam-delay domain, the channel delay components are mainly concentrated in the head of the delay-domain channel matrix, the processing unit 320 can truncate the delay-domain channel, retain the head, and divide the truncated data into real and imaginary parts Two channels are used as input to the super-resolution network. By transforming the first channel to be processed into the delay domain and staging the data, the computational complexity of the super-resolution network can be reduced.
  • the processing unit 320 may also use zero-filling or linear interpolation, nearest-neighbor interpolation and other existing difference methods to interpolate the first channel
  • a pre-difference value is performed to obtain a channel to the desired frequency domain accuracy (eg, RB level or subcarrier level).
  • the pre-difference value operation only formally increases the channel dimension to the output dimension to facilitate subsequent network processing, but does not substantially improve the channel accuracy.
  • Super-resolution interpolation of the channel matrix is done through a super-resolution network.
  • the processing unit 320 may use various super-resolution networks.
  • the first channel of the first granularity may be interpolated and denoised in a similar manner as the super-resolution network is used to interpolate and denoise the image.
  • the super-resolution network can be pre-trained with a high density of reference signals.
  • the base station 300 may further include a sending unit, so as to send a high-density reference signal to at least one of the user equipment and the data collection apparatus, and to receive the first channel state information reference by at least one of the user equipment and the data collection apparatus information feedback.
  • the processing unit 320 may use the feedback information for the first channel state information reference information to train the super-resolution network, so that the super-resolution network learns to obtain the above-mentioned formula (1) corresponding to the specific feedback information through interpolation and denoising processing. ) with respect to the parameters of the channel and the candidate set of the function F. Therefore, in actual deployment, the processing unit 320 can input the first channel obtained according to the precoding matrix indication information fed back by the terminal into the trained super-resolution network to recover the channel accurately.
  • the processing unit 320 may use at least one of a Very Deep Super Resolution Network (VDSR) and a Cascading Residual Network (CARN) to perform interpolation and denoising processing on the channels of the first granularity to obtain the second channel.
  • VDSR Very Deep Super Resolution Network
  • CARN Cascading Residual Network
  • the processing unit 320 may use a VDSR with 16-20 layers and a convolution kernel size of 3 to perform interpolation and denoising processing on the channels of the first granularity to obtain the second channel. Since a large depth network is beneficial to learn the features in the channel, using a large depth network can better perform channel recovery.
  • a residual network structure may be applied in a large depth super-resolution network according to an example of the present disclosure. Specifically, the input can be superimposed before the output layer to enhance the correspondence between the output and the input.
  • the processing unit 320 may also use a cascaded residual convolutional network formed by introducing a plurality of small convolutional networks into the residual structure with each other.
  • each small convolutional network can be a 3-layer convolutional network, and the convolution kernel size is 3.
  • the cascaded residual network can achieve better performance with less complexity.
  • the processing unit 320 performs downlink precoding on the second channel for sending to the terminal.
  • the super-resolution network is used to perform interpolation and denoising processing on the channel of the first granularity to obtain a channel with a finer granularity the second channel for downlink precoding.
  • FIG. 4 is a schematic block diagram illustrating a base station according to another embodiment of the present disclosure.
  • the base station 300 may include a receiving unit 410 and a processing unit 420 .
  • the base station 400 may further include other components, however, since these components are irrelevant to the content of the embodiments of the present disclosure, their illustration and description are omitted here.
  • the receiving unit 410 of the base station 400 receives the precoding matrix indication information of the first granularity from the terminal.
  • the receiving unit 410 may receive the type II PMI information of the subband level from the terminal.
  • the type II PMI information may include magnitude information and phase information at the subband level.
  • the processing unit 420 performs channel reconstruction, interpolation and de-noising processing through the first sub-network according to the precoding matrix indication information of the first granularity to obtain a second channel, and performs downlink precoding on the channel of the second granularity, wherein the second particle size is finer than the first particle size.
  • the first sub-network may be a first sub-neural network.
  • the input dimension of the first sub-network may be higher than the output dimension of the first sub-network.
  • the first sub-network is designed with high-dimensional input and low-dimensional output. Due to the use of high-dimensional input, the original information from the information indicated by the precoding matrix can be saved, and because of the use of low-dimensional output, the network complexity and training difficulty can be reduced through dimensionality reduction during network processing.
  • the input of the first sub-network may be precoding matrix indication information or pre-processed precoding matrix indication information from the terminal, and the first sub-network may perform input reconstruction of the precoding matrix indication information.
  • the first sub-network may weightedly combine the magnitude and phase of the input data.
  • the precoding matrix indication information may include amplitude information and phase information.
  • the amplitude information may include wideband beam information, wideband amplitude information, subband amplitude information of each subband, and subband phase information of each subband of each beam that the base station 400 communicates with the terminal.
  • the first sub-network may combine the wideband beam information, the wideband amplitude information, and the subband amplitude information of each subband of each beam to obtain a channel amplitude matrix.
  • the first sub-network can separately obtain the channel amplitude matrix in each polarization direction according to the polarization directions of the array elements of the antenna array.
  • the first sub-network can obtain the real part matrix and the imaginary part matrix according to the broadband beam information of each beam and the phase information of each subband.
  • the first sub-network can separately obtain the real part matrix and the imaginary part matrix in each polarization direction according to the polarization direction of the beam.
  • the first sub-network can multiply the channel amplitude matrix in the polarization direction by the real part matrix and the imaginary part matrix respectively to obtain the beam-frequency domain channel matrix (also referred to as "beam” for short). - frequency domain channel”).
  • the base station 400 communicates with the terminal using beams in two polarization directions.
  • the first sub-network can obtain channel amplitude matrices A1, A2 in two polarization directions, and channel phase matrices (ie, real part matrix and imaginary part matrix) Pr1, Pr2, Pi1, Pi2 in two polarization directions.
  • the first sub-network can obtain beam-frequency domain channel matrices Hr1, Hr2, Hi1, Hi2 by the following formula (2):
  • the first sub-network may Fourier transform the beam-frequency domain channel to convert the beam-frequency domain channel into a beam-delay domain channel.
  • the channel delay components are mainly concentrated in the head of the delay domain channel matrix, the first sub-network can truncate the delay domain channel to reduce the dimension of the output.
  • the first sub-network may include a fully connected layer (Dense layer), and the input reconstruction of the precoding matrix indication information is performed through the fully connected layer.
  • the fully connected layer weights and combines the amplitude and phase of the input data to obtain a beam-frequency domain channel, converts the beam-frequency domain channel into a beam-delay domain channel, and truncates the beam-delay domain channel to reduce the network output dimension .
  • the corresponding fully-connected layer can be replaced with a partially-connected layer that only connects elements that need to be directly multiplied.
  • the first sub-network may further include one or more super-resolution networks for interpolation and denoising.
  • One or more super-resolution networks may be placed before or after the above-mentioned fully-connected or partially-connected layers.
  • the above-mentioned fully-connected layers or partially-connected layers can also be arranged between multiple super-resolution networks.
  • the super-resolution network according to the embodiment of the present disclosure has been described in detail above with reference to the example shown in FIG. 3 , so it will not be repeated here.
  • FIG. 14A-14C are schematic diagrams illustrating a first sub-network structure according to an embodiment of the present disclosure.
  • a super-resolution network may be set up before the fully-connected layer or the partially-connected layer to interpolate and de-noise the precoding matrix indication information from the terminal.
  • the interpolated and denoised data are then input to fully connected or partially connected layers for channel reconstruction and dimensionality reduction.
  • the precoding matrix indication information from the terminal can be first input to the fully connected layer or the partially connected layer for channel reconstruction and dimensionality reduction processing, and then the obtained channel is input to the super-resolution network for Interpolate and denoise.
  • FIG. 14A a super-resolution network may be set up before the fully-connected layer or the partially-connected layer to interpolate and de-noise the precoding matrix indication information from the terminal.
  • the interpolated and denoised data are then input to fully connected or partially connected layers for channel reconstruction and dimensionality reduction.
  • the precoding matrix indication information from the terminal can
  • the precoding matrix indication information from the terminal may be input to the first super-resolution network for denoising processing.
  • the denoised data is then input to a fully connected layer or a partially connected layer for channel reconstruction and dimensionality reduction.
  • the channel after dimensionality reduction is input to the second super-resolution network for interpolation processing.
  • the processing unit 400 may further perform time-domain channel estimation enhancement on multiple second channels obtained according to precoding matrix indication information sent multiple times from the same terminal through the second sub-network and at least one of time-domain predictions.
  • the first sub-network may process the precoding matrix indication information sent by the terminal at one time (eg, sent in a single time slot), and perform channel reconstruction, interpolation and denoising based on the precoding matrix indication information sent by the terminal at one time.
  • the second sub-network includes at least one of a recurrent neural network (Recurrent Neural Network, RNN) and a Long and Short-Term Memory (Long and Short-Term Memory, LSTM) network, and the second sub-network
  • RNN recurrent Neural Network
  • LSTM Long and Short-Term Memory
  • the coding matrix indicates that the information is input to the RNN/LSTM network to implement at least one of temporal channel estimation enhancement and temporal prediction.
  • the reference information of the channel state information sent by the base station will be described below with reference to FIG. 5 .
  • the channel state information reference information described below in conjunction with FIG. 5 may be applied to the base stations described in conjunction with FIGS. 3 and 4 .
  • FIG. 5 is a schematic block diagram illustrating a base station according to another embodiment of the present disclosure.
  • a base station 500 may include a sending unit 510 and a receiving unit 520 .
  • the base station 500 may further include other components, however, since these components are irrelevant to the content of the embodiments of the present disclosure, their illustration and description are omitted here.
  • the sending unit 510 sends the reference information of the first channel state information of the first density to the terminal.
  • the first density may be a higher density.
  • the sending unit 510 sends the first channel state information reference information over the entire communication bandwidth of the base station 500 at a first density.
  • the sending unit 510 may send high-density first channel state information reference information over the entire communication bandwidth.
  • the first channel state information reference information may be sent using resource blocks or resource elements occupied by ports 1-12.
  • the first channel state information reference information may occupy all subcarriers in the frequency domain, that is, the density of the first channel state information reference information may reach 12 resource elements (REs) per resource Blocks (RBs) per port, and each port's reference signal uses one OFDM symbol.
  • multiple ports can be multiplexed on the same OFDM symbol using a cyclic shift (cyclic shift) manner, or transmitted on different OFDM symbols in a TDM manner.
  • the density of the first channel state information reference information may reach 6 resource elements (REs) per resource block (RB) per port, and every two ports are multiplexed in the form of interleaved combs in the frequency domain , occupying one OFDM symbol.
  • multiplexing in cyclic shift and TDM is also possible.
  • the sending unit 510 further sends the channel state information reference information configuration information for indicating the first channel state information reference information, and uses the channel state information reference information configuration information in the training data collection time period.
  • the indicated resource is used to send the first channel state information reference information.
  • the training data collection period may include multiple time slots.
  • the receiving unit 520 receives the first feedback information for the first channel state information reference information sent by the terminal.
  • the terminal may be at least one of a user equipment and a data collection apparatus.
  • the data collection device may send the first feedback information for the first channel state information reference information to the base station 500 through a dedicated interface.
  • the sending unit 510 may also send control signaling to the UE to schedule the UE to send the first feedback information for the first channel state information reference information to the base station 500 through a data channel .
  • the first feedback information may include that the terminal performs first channel estimation according to high-density first channel state information reference information to obtain first channel response information (hereinafter referred to as "channel response data set"), and
  • the channel state information reference information obtained by performing down-sampling processing on the channel state information reference information determines precoding matrix indication information (hereinafter referred to as "PMI training data set").
  • the base station shown in FIG. 5 may further include a processing unit 530 .
  • the processing unit 530 may train the neural networks such as the super-resolution network, the first sub-network, and the second sub-network in the base station described above in conjunction with FIG. 3 and FIG. 4 according to the first feedback information, so that the neural network can
  • the state reference signal performs high precision channel estimation, ie, channel reconstruction, denoising and interpolation.
  • Processing unit 530 may use the PMI training dataset and the corresponding channel response dataset to train the neural network.
  • FIG. 6 is a schematic diagram illustrating training of a neural network using a PMI training dataset and a corresponding channel response dataset, according to one example of the present disclosure.
  • the processing unit of the base station takes the PMI training data set as the first input of the neural network, so as to use the neural network to perform channel reconstruction, denoising and interpolation processing according to the first input.
  • the PMI training data set can be used to simulate the precoding matrix indication information sent by the terminal during actual deployment.
  • the processing unit of the base station may use the high-density channel response data set of the first channel state information reference information as the second input of the neural network for network optimization, thereby, for example, providing the network with the information for the specific PMI training data set.
  • Target channel response is a schematic diagram illustrating training of a neural network using a PMI training dataset and a corresponding channel response dataset, according to one example of the present disclosure.
  • Base station 500 may be, for example, a dedicated base station for neural network training. After completing the training, the private base station 500 may provide the trained neural network to the base station that actually communicates with the UE in the communication network, so that the trained neural network can be used for channel reconstruction, denoising and/or actual communication during the actual communication. Interpolation and other operations.
  • the base station 500 may be used to actually communicate with the UE after completing the training.
  • the sending unit 510 sends the second channel state information reference information of the second density to the UE.
  • the receiving unit 520 receives the second feedback information of the UE for the second channel state information reference information.
  • the second channel state information reference information may be existing channel state information reference information used for channel estimation in actual communication, and the first density is greater than the second density. That is, the second channel state information reference information is sparser than the first channel state information reference information.
  • the processing unit 530 may use a pre-trained network to perform channel reconstruction, denoising and interpolation according to the second channel state information reference information, so as to obtain a high-precision channel to be used for downlink precoding in actual communication. Further, during actual deployment, the processing unit 530 may perform operations similar to those of the processing unit 320 and the processing unit 420 described above, which will not be repeated here.
  • FIG. 7 is a schematic block diagram illustrating a terminal according to an embodiment of the present disclosure.
  • a terminal 700 may include a receiving unit 710 , a processing unit 720 , and a sending unit 730 .
  • the terminal 700 may further include other components, however, since these components are not related to the content of the embodiments of the present disclosure, their illustration and description are omitted here.
  • the receiving unit 710 receives the first channel state information reference information of the first density.
  • the first channel state information reference information can be used to train the neural network of the base station.
  • the first channel state information reference information may be transmitted over the entire communication bandwidth of the base, and the first channel state information reference information has a higher channel state information reference than the channel state information reference used for channel measurement in the actual deployment stage higher density of information.
  • the processing unit 720 performs first channel estimation according to the first channel state information reference information to obtain the first channel response information, and performs downsampling processing on the first channel state information reference information, and performs downsampling processing on the channel state information reference information according to the downsampling process.
  • Determine precoding matrix indication information FIG. 8 is a schematic diagram illustrating that the processing unit 720 processes the first channel state information reference information according to an example of the present disclosure. As shown in FIG. 8 , on the one hand, the processing unit performs first channel estimation on the first channel state information reference information with high density to obtain high-precision first channel response information.
  • the processing unit downsamples the first channel state information reference information to simulate the density of the channel state information reference information received in the actual deployment stage, and obtains a low-density channel. Then, the processing unit calculates the PMI feedback amount according to the channel state information reference information after the down-sampling process to determine the precoding matrix indication information.
  • the sending unit 730 sends the first channel response information and the precoding matrix indication information to the base station. Therefore, the base station can train its neural network according to the received first channel response information and the precoding matrix indication information.
  • the terminal 700 may be at least one of a user equipment (UE) and a data collection apparatus.
  • the sending unit 730 can send the first channel response information and the precoding matrix indication information to the base station in offline mode or through an air interface.
  • the sending unit 730 may use the precoding matrix indication feedback channel to send the precoding matrix indication information to the base station, and may use the uplink data channel or the uplink control channel to transmit the first channel response information to the base station after modulation and coding.
  • the base station can perform channel reconstruction, interpolation, and denoising training on its neural network according to the first channel response information and the precoding matrix indication information.
  • FIG. 9 is a flowchart of a channel processing method 900 according to one embodiment of the present disclosure. Since the steps of the channel processing method 900 correspond to the operations of the base station 300 described above with reference to the figures, detailed descriptions of the same contents are omitted here for simplicity.
  • step S901 the precoding matrix indication information of the first granularity is received from the terminal.
  • step S902 channel reconstruction is performed according to the precoding matrix indication information to obtain a first channel, where the first channel has a first granularity.
  • the amplitude and phase information in the received PMI information may be used to respectively perform amplitude and phase weighting on the spatial domain (also referred to as "beam domain")-frequency domain channel codewords of multiple beams, And the weighted vectors are combined to obtain a first channel with subband level.
  • the first channel of the subband level obtained after the combination can be expressed in the form of a space-frequency domain channel matrix, wherein the space-domain value of the space-frequency domain channel matrix can be the number of antennas for communication between the base station and the terminal, and the frequency-domain value can be It is determined according to the number of subbands in which the base station communicates with the terminal.
  • step S903 the super-resolution network is used to perform interpolation and denoising processing on the channel of the first granularity to obtain a second channel, wherein the second channel has a second granularity, and the second granularity is smaller than the first granularity Fine-grained.
  • the method shown in FIG. 9 may further include that before inputting the first channel of the first granularity to the super-resolution network, the first channel may be preprocessed to facilitate subsequent super-resolution network operation.
  • the first channel when the first channel is a space-frequency-domain channel, Fourier transform may be performed on the first channel to convert the space-frequency-domain channel into a beam-delay-domain channel.
  • the channel delay components are mainly concentrated in the head of the delay-domain channel matrix, the delay-domain channel can be truncated, the head can be retained, and the truncated data can be divided into two parts, the real part and the imaginary part. channel as the input to the super-resolution network.
  • the method shown in Fig. 9 may also include the use of existing difference methods such as zero-filling or linear interpolation, nearest-neighbor interpolation, etc., before transforming it into a beam-delay domain channel. Interpolation pre-diffs the first channel to obtain a channel to the desired frequency domain accuracy (eg, RB level or subcarrier level).
  • the desired frequency domain accuracy eg, RB level or subcarrier level
  • step S904 downlink precoding is performed on the second channel to send to the terminal.
  • the super-resolution network is used to perform interpolation and denoising processing on the channel of the first granularity to obtain the second granularity with finer granularity. channel for downlink precoding.
  • FIG. 10 is a flowchart of a channel processing method 1000 according to another embodiment of the present disclosure. Since the steps of the channel processing method 1000 correspond to the operations of the base station 400 described above with reference to the figures, the detailed description of the same content is omitted here for simplicity.
  • step S1001 the precoding matrix indication information of the first granularity is received from the terminal.
  • type II PMI information at the subband level may be received from the terminal.
  • the type II PMI information may include magnitude information and phase information at the subband level.
  • step S1002 through the first sub-network, perform channel reconstruction, interpolation and denoising processing according to the precoding matrix indication information of the first granularity to obtain a second channel, and perform downlink precoding on the channel of the second granularity encoding, wherein the second granularity is finer than the first granularity.
  • the first sub-network may be a first sub-neural network.
  • the input dimension of the first sub-network may be higher than the output dimension of the first sub-network.
  • the first sub-network is designed with high-dimensional input and low-dimensional output. Due to the use of high-dimensional input, the original information from the information indicated by the precoding matrix can be saved, and because of the use of low-dimensional output, the network complexity and training difficulty can be reduced through dimensionality reduction during network processing.
  • the input of the first sub-network may be precoding matrix indication information from the terminal or pre-processing precoding matrix indication information, and the first sub-network may perform input reconstruction of the precoding matrix indication information.
  • the first sub-network may weight and combine the amplitude and phase of the input data.
  • the precoding matrix indication information may include amplitude information and phase information.
  • the amplitude information may include wideband beam information, wideband amplitude information, subband amplitude information of each subband, and subband phase information of each subband of each beam that the base station communicates with the terminal.
  • the first sub-network may combine the broadband beam information of each beam, the broadband amplitude information and the subband amplitude information of each subband to obtain a channel amplitude matrix. Specifically, the first sub-network can separately obtain the channel amplitude matrix in each polarization direction according to the polarization direction of the beam. In addition, in step S1002, the first sub-network may also obtain a real part matrix and an imaginary part matrix according to the broadband beam information of each beam and the phase information of each subband. Similarly, the first sub-network can separately obtain the real part matrix and the imaginary part matrix in each polarization direction according to the polarization direction of the beam.
  • the first sub-network can multiply the channel amplitude matrix in the polarization direction by the real part matrix and the imaginary part matrix respectively to obtain the beam-frequency domain channel matrix (also referred to as "beam” for short). - frequency domain channel”).
  • the first sub-network may perform Fourier transform on the beam-frequency domain channel to convert the beam-frequency domain channel into a beam-delay domain channel.
  • the channel delay components are mainly concentrated in the head of the delay domain channel matrix, the first sub-network can truncate the delay domain channel to reduce the dimension of the output.
  • the first sub-network may include a fully connected layer (Dense layer), and the input reconstruction of the precoding matrix indication information is performed through the fully connected layer.
  • the fully connected layer weights and combines the amplitude and phase of the input data to obtain a beam-frequency domain channel, converts the beam-frequency domain channel into a beam-delay domain channel, and truncates the beam-delay domain channel to reduce the network output dimension .
  • the corresponding fully-connected layer can be replaced with a partially-connected layer that only connects elements that need to be directly multiplied.
  • the first sub-network may further include one or more super-resolution networks for interpolation and denoising.
  • One or more super-resolution networks may be placed before or after the above-mentioned fully-connected or partially-connected layers.
  • the above-mentioned fully-connected layers or partially-connected layers can also be arranged between multiple super-resolution networks.
  • a super-resolution network can be set up before the fully-connected layer or the partially-connected layer to interpolate and de-noise the precoding matrix indication information from the terminal.
  • the interpolated and denoised data are then input to the fully-connected or partially-connected layers for channel reconstruction and dimensionality reduction.
  • the precoding matrix indication information from the terminal can be first input to the fully connected layer or the partially connected layer for channel reconstruction and dimensionality reduction processing, and then the obtained channel is input to the super-resolution network for interpolation and denoising processing.
  • the precoding matrix indication information from the terminal may be input to the first super-resolution network for denoising processing.
  • the denoised data is then input to a fully connected layer or a partially connected layer for channel reconstruction and dimensionality reduction.
  • the channel after dimensionality reduction is input to the second super-resolution network for interpolation processing.
  • the method shown in FIG. 10 may further include, through the second sub-network, performing time-domain channel estimation enhancement and time-domain channel estimation enhancement on multiple second channels obtained according to the precoding matrix indication information sent from the same terminal multiple times. at least one of the predictions.
  • the first sub-network may process the precoding matrix indication information sent by the terminal at one time (eg, sent in a single time slot), and perform channel reconstruction, interpolation and denoising based on the precoding matrix indication information sent by the terminal at one time.
  • the second sub-network includes at least one of an RNN and an LSTM network, and the second sub-network can input the precoding matrix indication information sent by the same terminal multiple times into the value RNN/LSTM network, so as to realize time-domain channel estimation enhancement and time-domain prediction at least one of the.
  • FIG. 11 is a flowchart of a reference signal transmission method 1100 according to one embodiment of the present disclosure. Since the steps of the reference signal transmission method 1100 correspond to the operations of the base station 500 described above with reference to the drawings, the detailed description of the same content is omitted here for simplicity.
  • the first channel state information reference information of the first density is sent to the terminal.
  • the first density may be a higher density.
  • the first channel state information reference information is sent over the entire communication bandwidth of the base station at a first density.
  • high-density first channel state information reference information may be sent over the entire communication bandwidth.
  • the method 1100 may further include sending channel state information reference information configuration information for indicating the first channel state information reference information, and using the channel state information reference information configuration information during the training data collection period
  • the indicated resource is used to send the first channel state information reference information.
  • the training data collection period may include multiple time slots.
  • the terminal may be at least one of a user equipment and a data collection apparatus.
  • the data collection device can send the first feedback information for the first channel state information reference information to the base station through a dedicated interface.
  • the method 1100 may further include sending control signaling to the UE to schedule the UE to send the first feedback information for the first channel state information reference information to the base station 500 through a data channel .
  • the first feedback information may include that the terminal performs first channel estimation according to high-density first channel state information reference information to obtain first channel response information (hereinafter referred to as "channel response data set"), and
  • the channel state information reference information obtained by performing down-sampling processing on the channel state information reference information determines precoding matrix indication information (hereinafter referred to as "PMI training data set").
  • Processing unit 530 may use the PMI training dataset and the corresponding channel response dataset to train the neural network.
  • the method 1100 may further include performing neural networks such as the super-resolution network, the first sub-network, and the second sub-network in the base station described above in conjunction with FIG. 3 and FIG. 4 according to the first feedback information.
  • the training is performed so that the neural network can perform high-precision channel estimation based on the state reference signal fed back by the terminal, that is, channel reconstruction, denoising and interpolation.
  • the base station may be, for example, a dedicated base station used for neural network training. After completing the training, the private base station can provide the trained neural network to the base station that actually communicates with the UE in the communication network, so that the trained neural network can be used for channel reconstruction, denoising and/or interpolation during actual communication. processing, etc.
  • the base station may be used to actually communicate with the UE after completing the training.
  • the second channel state information reference information of the second density is sent to the UE.
  • the second feedback information of the UE for the second channel state information reference information is received.
  • the second channel state information reference information may be existing channel state information reference information used for channel estimation in actual communication, and the first density is greater than the second density. That is, the second channel state information reference information is sparser than the first channel state information reference information.
  • step S1105 using the pre-trained network, channel reconstruction, denoising and interpolation are performed according to the second channel state information reference information to obtain a high-precision channel to be used for downlink precoding in actual communication.
  • FIG. 12 is a flowchart of an information transmission method 1200 according to one embodiment of the present disclosure. Since the steps of the reference signal transmission method 1200 correspond to the operations of the terminal 700 described above with reference to the drawings, the detailed description of the same content is omitted here for simplicity.
  • the first channel state information reference information of the first density is received.
  • the first channel state information reference information can be used to train the neural network of the base station.
  • the first channel state information reference information may be transmitted over the entire communication bandwidth of the base, and the first channel state information reference information has a higher channel state information reference than the channel state information reference used for channel measurement in the actual deployment stage higher density of information.
  • step S1202 first channel estimation is performed according to the first channel state information reference information to obtain first channel response information. And in step S1203, down-sampling processing is performed on the first channel state information reference information, and precoding matrix indication information is determined according to the channel state information reference information after the down-sampling processing.
  • step S1202 is performed first and then step S1203 is performed for illustration, the present disclosure is not limited thereto.
  • step S1203 may also be performed first and then step S1201, or step S1202 and step S1203 may be performed simultaneously.
  • step S1204 the first channel response information and the precoding matrix indication information are sent to the base station. Therefore, the base station can train its neural network according to the received first channel response information and the precoding matrix indication information.
  • each functional block may be implemented by one device that is physically and/or logically combined, or two or more devices that are physically and/or logically separated may be directly and/or indirectly (for example, By wired and/or wireless) connection, it is realized by the above-mentioned multiple devices.
  • FIG. 13 is a schematic diagram of a hardware structure of an involved device 1300 (base station, terminal) according to an embodiment of the present disclosure.
  • the above-mentioned device 1300 (base station, terminal) can be configured as a computer device that physically includes a processor 1310, a memory 1320, a memory 1330, a communication device 1340, an input device 1350, an output device 1360, a bus 1370, and the like.
  • the word “device” may be replaced with a circuit, a device, a unit, or the like.
  • the hardware structure of the first network element may include one or more devices shown in the figures, or may not include some devices.
  • processor 1310 For example, only one processor 1310 is shown, but there may be multiple processors. Furthermore, processing may be performed by one processor, or by more than one processor simultaneously, sequentially, or in other ways. In addition, the processor 1310 may be mounted on more than one chip.
  • Each function of the device 1300 is realized, for example, by reading predetermined software (programs) into hardware such as the processor 1310 and the memory 1320, thereby causing the processor 1310 to perform calculations and to control the communication by the communication device 1340. , and controls the reading and/or writing of data in the memory 1320 and the memory 330 .
  • the processor 1310 controls the entire computer by operating an operating system, for example.
  • the processor 1310 may be constituted by a central processing unit (CPU, Central Processing Unit) including an interface with peripheral devices, a control device, an arithmetic device, a register, and the like.
  • CPU Central Processing Unit
  • the above-mentioned processing units and the like can be implemented by the processor 1310 .
  • the processor 1310 reads out programs (program codes), software modules, data, etc. from the memory 1330 and/or the communication device 1340 to the memory 1320, and executes various processes according to them.
  • programs program codes
  • software modules software modules
  • data etc.
  • the program a program for causing a computer to execute at least a part of the operations described in the above-described embodiments may be employed.
  • the processing unit of the first network element can be implemented by a control program stored in the memory 1320 and operated by the processor 1310, and other functional blocks can also be implemented similarly.
  • the memory 1320 is a computer-readable recording medium. Random access memory (RAM, Random Access Memory) and at least one of other suitable storage media. Memory 1320 may also be referred to as registers, cache, main memory (main storage), and the like. The memory 1320 may store executable programs (program codes), software modules, and the like for implementing the method according to an embodiment of the present disclosure.
  • RAM Random Access Memory
  • Memory 1320 may also be referred to as registers, cache, main memory (main storage), and the like.
  • the memory 1320 may store executable programs (program codes), software modules, and the like for implementing the method according to an embodiment of the present disclosure.
  • the memory 1330 is a computer-readable recording medium, and can be composed of, for example, a flexible disk, a floppy (registered trademark) disk, a magneto-optical disk (for example, a CD-ROM (Compact Disc ROM), etc.), Digital versatile discs, Blu-ray (registered trademark) discs), removable disks, hard drives, smart cards, flash memory devices (eg, cards, sticks, key drivers), magnetic stripes, databases , a server, and at least one of other suitable storage media.
  • Memory 1330 may also be referred to as secondary storage.
  • the communication device 1340 is a hardware (transmitting and receiving device) used for communication between computers through a wired and/or wireless network, and is also called, for example, a network device, a network controller, a network card, a communication module, and the like.
  • the communication device 1340 may include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, and the like.
  • the above-mentioned sending unit, receiving unit, etc. can be implemented by the communication device 1340 .
  • the input device 1350 is an input device (eg, keyboard, mouse, microphone, switch, button, sensor, etc.) that accepts input from the outside.
  • the output device 1360 is an output device (eg, a display, a speaker, a Light Emitting Diode (LED, Light Emitting Diode) lamp, etc.) that implements output to the outside.
  • the input device 1350 and the output device 1360 may also have an integrated structure (eg, a touch panel).
  • each device such as the processor 1310 and the memory 1320 is connected by a bus 1370 for communicating information.
  • the bus 1370 may be constituted by a single bus, or may be constituted by different buses between devices.
  • the electronic device may include a microprocessor, a digital signal processor (DSP, Digital Signal Processor), an application specific integrated circuit (ASIC, Application Specific Integrated Circuit), a programmable logic device (PLD, Programmable Logic Device), a field programmable gate Array (FPGA, Field Programmable Gate Array) and other hardware, can realize part or all of each functional block through the hardware.
  • DSP digital signal processor
  • ASIC Application Specific Integrated Circuit
  • PLD programmable logic device
  • FPGA Field Programmable Gate Array
  • the processor 710 may be installed by at least one of these pieces of hardware.
  • channels and/or symbols may also be signals (signaling).
  • signals can also be messages.
  • the reference signal may also be referred to as RS (Reference Signal) for short, and may also be referred to as a pilot (Pilot), a pilot signal, etc. according to the applicable standard.
  • a component carrier CC, Component Carrier
  • CC Component Carrier
  • the information, parameters, etc. described in this specification may be expressed by absolute values, may be expressed by relative values with respect to predetermined values, or may be expressed by corresponding other information.
  • the radio resource may be indicated by a prescribed index.
  • the formulas and the like using these parameters may also be different from those explicitly disclosed in this specification.
  • the information, signals, etc. described in this specification may be represented using any of a variety of different technologies.
  • data, commands, instructions, information, signals, bits, symbols, chips, etc. may be mentioned throughout the above description may be generated by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, light fields or photons, or any of them. combination to represent.
  • information, signals, etc. may be output from the upper layer to the lower layer, and/or from the lower layer to the upper layer.
  • Information, signals, etc. can be input or output via multiple network nodes.
  • Input or output information, signals, etc. can be stored in a specific place (eg, memory), and can also be managed through a management table. Input or output information, signals, etc. can be overwritten, updated or supplemented. Output messages, signals, etc. can be deleted. Input information, signals, etc. can be sent to other devices.
  • a specific place eg, memory
  • Input or output information, signals, etc. can be overwritten, updated or supplemented.
  • Output messages, signals, etc. can be deleted.
  • Input information, signals, etc. can be sent to other devices.
  • Notification of information is not limited to the mode/embodiment described in this specification, and may be performed by other methods.
  • the notification of information may be through physical layer signaling (eg, Downlink Control Information (DCI, Downlink Control Information), Uplink Control Information (UCI, Uplink Control Information)), upper layer signaling (eg, Radio Resource Control Information) (RRC, Radio Resource Control) signaling, broadcast information (Master Information Block (MIB, Master Information Block), System Information Block (SIB, System Information Block), etc.), Media Access Control (MAC, Medium Access Control) signaling ), other signals, or a combination thereof.
  • DCI Downlink Control Information
  • UCI Uplink Control Information
  • RRC Radio Resource Control Information
  • RRC Radio Resource Control
  • MAC Media Access Control
  • the physical layer signaling may also be referred to as L1/L2 (Layer 1/Layer 2) control information (L1/L2 control signal), L1 control information (L1 control signal), or the like.
  • the RRC signaling may also be called an RRC message, for example, an RRC connection setup (RRC Connection Setup) message, an RRC connection reconfiguration (RRC Connection Reconfiguration) message, and the like.
  • the MAC signaling can be notified by, for example, a MAC control element (MAC CE (Control Element)).
  • notification of predetermined information is not limited to being performed explicitly, and may be performed implicitly (eg, by not performing notification of the predetermined information, or by notification of other information).
  • the determination can be performed by a value (0 or 1) represented by 1 bit, by a true or false value (Boolean value) represented by true (true) or false (false), or by a numerical comparison ( For example, a comparison with a predetermined value) is performed.
  • software, commands, information, etc. may be sent or received via a transmission medium.
  • a transmission medium For example, when sending from a website, server, or other remote source using wireline technology (coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL, Digital Subscriber Line, etc.) and/or wireless technology (infrared, microwave, etc.)
  • wireline technology coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL, Digital Subscriber Line, etc.
  • wireless technology infrared, microwave, etc.
  • system and “network” are used interchangeably in this specification.
  • Base station BS, Base Station
  • radio base station eNB
  • gNB gNodeB
  • cell gNodeB
  • cell group femtocell
  • carrier femtocell
  • a base station may house one or more (eg, three) cells (also referred to as sectors). When the base station accommodates multiple cells, the entire coverage area of the base station can be divided into multiple smaller areas, and each smaller area can also pass through the base station subsystem (for example, indoor small base stations (Remote Radio Heads (RRH, RRH) Remote Radio Head)) to provide communication services.
  • the terms "cell” or “sector” refer to a portion or the entirety of the coverage area of the base station and/or base station subsystem in which the communication service is performed.
  • mobile station MS, Mobile Station
  • user terminal user terminal
  • UE User Equipment
  • terminal mobile station
  • a mobile station is also sometimes referred to by those skilled in the art as subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless Terminal, remote terminal, handset, user agent, mobile client, client, or some other appropriate term.
  • the radio base station in this specification may also be replaced with a user terminal.
  • each aspect/embodiment of the present disclosure can also be applied to a structure in which the communication between the radio base station and the user terminal is replaced by the communication between a plurality of user terminals (D2D, Device-to-Device).
  • the functions possessed by the above-mentioned electronic device may be regarded as functions possessed by the user terminal.
  • words like "up” and "down” can also be replaced with "side”.
  • the upstream channel can also be replaced by a side channel.
  • the user terminal in this specification can also be replaced with a wireless base station.
  • the functions possessed by the above-mentioned user terminal may be regarded as functions possessed by the first communication device or the second communication device.
  • a specific operation performed by a base station may also be performed by an upper node thereof depending on circumstances.
  • various actions performed for communication with a terminal can be performed through the base station, one or more networks other than the base station Nodes (for example, Mobility Management Entity (MME, Mobility Management Entity), Serving-Gateway (S-GW, Serving-Gateway), etc. can be considered, but not limited thereto), or a combination thereof.
  • MME Mobility Management Entity
  • S-GW Serving-Gateway
  • Serving-Gateway Serving-Gateway
  • LTE Long Term Evolution
  • LTE-A Long Term Evolution Advanced
  • LTE-B Long Term Evolution Beyond
  • LTE-Beyond Long Term Evolution Beyond
  • IMT-Advanced 4th Generation Mobile Communication System
  • 4G 4th generation mobile communication system
  • 5G 5th Generation Mobile Communication System
  • Future Radio Access Future Radio Access
  • New-RAT New Radio Access Technology
  • New Radio New Radio
  • NR New Radio
  • New Radio Access New Radio Access
  • NX New radio access
  • Future Generation Radio Access Future Generation Radio Access
  • GSM Global System for Mobile Communications
  • CDMA3000 Code Division Multiple Access 3000
  • UMB Ultra Mobile Broadband
  • IEEE 920.11 Wi-Fi (registered trademark)
  • IEEE 920.11 Wi-Fi (registered trademark)
  • any reference in this specification to an element using the designation "first”, “second” etc. is not intended to comprehensively limit the number or order of such elements. These names may be used in this specification as a convenient method of distinguishing two or more units. Thus, a reference to a first element and a second element does not imply that only two elements may be employed or that the first element must precede the second element in some form.
  • determining (determining) used in this specification may include various operations. For example, with regard to “judging (determining)”, calculating, computing, processing, deriving, investigating, looking up (eg, tables, databases, or other Searching in the data structure), confirming (ascertaining), etc. are regarded as “judgment (determination)”. In addition, regarding “judgment (determination)”, receiving (for example, receiving information), transmitting (for example, transmitting information), input (input), output (output), accessing (accessing) (for example, access to data in the memory), etc., are regarded as “judgment (determination)".
  • connection refers to any connection or combination, direct or indirect, between two or more units, which can be It includes the following situations: between two units “connected” or “combined” with each other, there are one or more intermediate units.
  • the combination or connection between the units may be physical or logical, or may also be a combination of the two.
  • connecting can also be replaced by "accessing”.
  • two units may be considered to be electrically connected through the use of one or more wires, cables, and/or printed, and as a number of non-limiting and non-exhaustive examples, by using a radio frequency region , the microwave region, and/or the wavelengths of electromagnetic energy in the light (both visible and invisible) region, etc., are “connected” or “combined” with each other.

Landscapes

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

Abstract

本公开提供一种基站和终端。所述基站，包括：接收单元，配置为从终端接收第一粒度的预编码矩阵指示信息；处理单元，配置为根据所述预编码矩阵指示信息进行信道重建以获得第一信道，使用超分辨率网络对于所述第一粒度的信道进行插值和去噪处理以获得第二信道，以及对所述第二信道进行下行预编码，其中所述第一信道具有所述第一粒度，所述第二信道具有第二粒度，所述第二粒度比所述第一粒度细。

Description

基站和终端 技术领域
本公开涉及无线通信领域，并且更具体地涉及一种用于根据终端的反馈进行信道重建的基站以及相应的终端。
背景技术
在通信系统中，基站向UE发送下行参考信号，UE根据下行参考信号进行信道估计并向基站发送类型I或类型II的预编码矩阵指示(PMI)信息。基站可根据UE发送的PMI信息确定相应的码字和码子相关的组合系数以重建信道，并且使用重建的信道进行下行预编码。目前UE发送的类型II的PMI信息为子带级的PMI信息，这导致空域和频域量化粒度较大，并且组合系数量化粒度较大。相应地，基站基于UE发送的子带级的PMI信息重建子带级信道。另一方面，在5G NR系统中，基站可以以物理资源块绑定(PRB bundling)为单位，使用重建的信道进行预编码，而物理资源块绑定的粒度通常远小于PMI信息的粒度。
例如，在基站和UE之间的通信带宽为100MHz，并且子载波间隔(Subcarrier Spacing，SCS)为30kHz的情况下，一个子带中可包括16个资源块(Resource Block,RB)。UE可发送子带(即，16个RB)级的类型II的PMI信息。另一方面，物理资源块绑定的最小尺寸可为2个RB，其远小于PMI信息的粒度。
此外，在根据子PMI信息重建子带级信道时，会对于空域和系数的量化噪声带来较大误差，这导致降低了各种多输入多输出(MIMO)预编码消除用户间干扰的能力，因此，需要降低量化噪声。
发明内容
根据本公开的一个方面，提供了一种基站。所述基站包括：接收单元，配置为从终端接收第一粒度的预编码矩阵指示信息；处理单元，配置为根据所述预编码矩阵指示信息进行信道重建以获得第一信道，使用超分辨率网络 对于所述第一粒度的信道进行插值和去噪处理以获得第二信道，以及对所述第二信道进行下行预编码，其中所述第一信道具有所述第一粒度，所述第二信道具有第二粒度，所述第二粒度比所述第一粒度细。
根据本公开的另一方面，提供了另一基站。所述基站包括：接收单元，配置为从终端接收第一粒度的预编码矩阵指示信息；处理单元，配置为通过第一子网络，根据所述第一粒度的预编码矩阵指示信息进行信道重建、插值和去噪处理以获得第二信道，以及对所述第二粒度的信道进行下行预编码，其中所述第二粒度比所述第一粒度细。
根据本公开的另一方面，提供了另一基站。所述基站包括：发送单元，配置为向终端发送第一密度的第一信道状态信息参考信息；接收单元，配置为从所述终端接收对于所述第一信道状态信息参考信息的第一反馈信息；其中所述第一反馈信息包括终端根据所述第一信道状态信息参考信息进行第一信道估计所获得第一信道响应信息，以及所述终端根据降采样处理后的第一信道状态信息参考信息确定的预编码矩阵指示信息。
根据本公开的另一方面，提供了一种终端。所述终端包括：接收单元，配置为接收第一密度的第一信道状态信息参考信息；处理单元，配置为根据所述第一信道状态信息参考信息进行第一信道估计以获得第一信道响应信息，以及对所述第一信道状态信息参考信息进行降采样处理，并根据降采样处理后的信道状态信息参考信息确定预编码矩阵指示信息；以及发送单元，配置为向基站发送所述第一信道响应信息和预编码矩阵指示信息。
附图说明
通过结合附图对本公开实施例进行更详细的描述，本公开的上述以及其它目的、特征和优势将变得更加明显。附图用来提供对本公开实施例的进一步理解，并且构成说明书的一部分，与本公开实施例一起用于解释本公开，并不构成对本公开的限制。在附图中，相同的参考标号通常代表相同部件或步骤。
图1是示出在通信系统中，基站根据终端的反馈进行信道重建的示意图。
图2是示出了现有的通信系统中，根据终端反馈的PMI信息确定的信道状态信息(CSI)和理想CSI的示意图。
图3是示出根据本公开一个实施例的基站的示意性框图。
图4是示出根据本公开另一实施例的基站的示意性框图。
图5是示出根据本公开另一实施例的基站的示意性框图。
图6是示出根据本公开的一个示例，使用PMI训练数据集和对应的信道响应数据集对神经网络进行训练的示意图。
图7是示出根据本公开一个实施例的终端的示意性框图。
图8是示出根据本公开一个示例，处理单元对第一信道状态信息参考信息进行处理的示意图。
图9是根据本公开的一个实施例的信道处理方法的流程图。
图10是根据本公开的另一实施例的信道处理方法的流程图。
图11是根据本公开的一个实施例的参考信号发送方法的流程图。
图12是根据本公开的一个实施例的信息发送方法的流程图。
图13是根据本公开的实施例的所涉及的设备的硬件结构的示意图。
图14A-图14C是示出了根据本公开实施例的第一子网络结构的示意图。
具体实施方式
为了使得本公开的目的、技术方案和优点更为明显，下面将参照附图详细描述根据本公开的示例实施例。在附图中，相同的参考标号自始至终表示相同的元件。应当理解：这里描述的实施例仅仅是说明性的，而不应被解释为限制本公开的范围。此外，这里所述的终端可以包括各种类型的终端，例如用户终端(User Equipment，UE)、移动终端(或称为移动台)或者固定终端，然而，为方便起见，在下文中有时候可互换地使用终端和UE。
图1是示出在通信系统中，基站根据终端的反馈进行信道重建的示意图。如图1所示，终端110根据下行参考信号进行信道估计，并且根据信道估计结果获得类型I或类型II的预编码矩阵指示(PMI)信息以发送给基站120。基站120根据UE发送的PMI信息重建信道，并且使用重建的信道进行下行预编码。可通过以下公式(1)表示基站期望获得的理想空域-时域信道H：
Figure PCTCN2021074690-appb-000001
其中，空域-时域信道H可以看作N条多径分量的叠加，每条多径分量 幅度为α i，其可以写为延迟τ i,，水平到达角
Figure PCTCN2021074690-appb-000002
垂直到达角θ i，相位φ i的函数F。目前基于下行参考信号逐一估计以上公式中的各个参数非常困难，因此难以精确恢复信道。
具体地，在目前的在通信系统中，终端110发送子带级的类型II的PMI信息。这导致空域和频域量化粒度较大，并且组合系数量化粒度较大。相应地，基站120在根据PMI信息重建的信道也是子带级的信道，这使得重建的信道的粒度较粗。图2是示出了现有的通信系统中，根据终端反馈的PMI信息确定的信道状态信息(CSI)和理想CSI的示意图。在图2所示的示例中，以菱形表示根据终端反馈的PMI信息确定的CSI，以圆形表示理想的信道状态信息。如图2所示，根据终端反馈的PMI信息确定的CSI的粒度远比理想的信道状态信息粗糙。当基站使用基于PMI直接重建的CSI时，频谱效率相对于使用理想CSI有50％以上的损失。
另一方面，在5G NR系统中，基站可以以物理资源块绑定为单位，使用重建的信道进行预编码，而物理资源块绑定的粒度通常远小于PMI信息的粒度。
此外，在根据子PMI信息重建子带级信道时，会对于空域和系数的量化噪声带来较大误差，这导致降低了各种多输入多输出(MIMO)预编码消除用户间干扰的能力，因此，需要降低量化噪声。
以下，参考图3来说明根据本公开的一个实施例的基站。图3是示出根据本公开一个实施例的基站的示意性框图。如图3所示，根据本公开一个实施例的基站300可包括接收单元310、和处理单元320。除了接收单元和处理单元，基站300还可以包括其他部件，然而，由于这些部件与本公开实施例的内容无关，因此在这里省略其图示和描述。
如图3所示，基站300的接收单元310从终端接收第一粒度的预编码矩阵指示信息。处理单元320根据所述预编码矩阵指示信息进行信道重建以获得第一信道，其中所述第一信道具有第一粒度。例如，接收单元310可从终端接收子带级的类型II的PMI信息。具体地，类型II的PMI信息可包括空域码字选择信息，宽带和子带级的码字组合系数的幅度信息和相位信息,还可以包括频域码字选择信息以及空频域码字组合系数。处理单元320可使用接收单元310所接收的PMI信息中的幅度信息和相位信息，分别对多个波束的 空域(也可称为“波束域”)-频域信道码字进行幅度和相位加权，并且将加权后的向量合并以得到具有子带级的第一信道。合并后得到的子带级的第一信道可以以空域-频域信道矩阵的形式表示，其中空域-频域信道矩阵的空域值可以是基站300与终端进行通信的天线的数量，而频域值可以根据基站300与终端进行通信的子带的数量确定。
然后，处理单元320使用超分辨率网络对于所述第一粒度的信道进行插值和去噪处理以获得第二信道，其中所述第二信道具有第二粒度，并且第二粒度比第一粒度细。例如，如上所述，第一粒度的信道可以是子带级的信道。在此情况下，第二粒度的信道可以是子载波级或者资源块(RB)级的信道。
根据本公开的一个示例，在将第一粒度的第一信道输入到超分辨率网络之前，处理单元320可对第一信道进行预处理，以便于后续超分辨率网络的操作。例如，在第一信道为空域-频域信道的情况下，处理单元320可对第一信道进行傅里叶变换，以将空域-频域信道为波束-延迟域信道。此外，由于在波束-延迟域中，信道延迟分量主要集中在延迟域信道矩阵头部，处理单元320可对延迟域信道进行截断，保留头部，并且将截断后数据分为实部和虚部两个通道作为超分辨率网络的输入。通过对将要处理的第一信道变换至延迟域并对数据进行阶段，可减少超分辨率网络计算的复杂度。
此外为了进一步简化超分辨率网络的操作，在将其变换至波束-延迟域信道之前，处理单元320还可使用填零或线性插值、最邻近插值等已有的差值方法插值对第一信道进行预差值，以获得到期望的频域精度(例如RB级或子载波级)的信道。应理解，该预差值操作只是形式上将信道维度提高到输出的维度便于后续网络进行处理，而没有实质上提升信道的精度。对信道矩阵的超分辨率插值是通过超分辨网络完成的。
根据本公开的一个示例，处理单元320可使用各种超分辨率网络。可与使用超分辨率网络对图像进行插值和去噪类似的方法对第一粒度的第一信道进行插值和去噪处理。此外，可预先使用高密度的参考信号对超分辨率网络进行训练。例如，基站300还可包括发送单元，以便向用户设备和数据采集装置中的至少一个发送高密度的参考信号，并且接收用户设备和数据采集装置中的至少一个对于所述第一信道状态信息参考信息的反馈信息。处理单元320可使用对于所述第一信道状态信息参考信息的反馈信息对超分辨率网络 进行训练，以使得超分辨率网络学习通过插值和去噪处理获得与特定反馈信息对应的上述公式(1)所示的关于信道的参数和函数F的候选集合。从而在实际部署时，处理单元320可将根据终端反馈的预编码矩阵指示信息获得的第一信道输入到训练好的超分辨率网络进行精确的恢复信道。
例如，处理单元320可使用大深度超分辨率网络(Very Deep Super Resolution，VDSR)和级联残差网络(Cascading Residual Network，CARN)中的至少一个对于第一粒度的信道进行插值和去噪处理以获得第二信道。具体地，处理单元320可使用16-20层、卷积核大小为3的VDSR对于第一粒度的信道进行插值和去噪处理以获得第二信道。由于大深度的网络有利于学习信道中的特征，因此使用大深度的网络可更好进行信道恢复。此外，根据本公开的一个示例可在大深度超分辨率网络中应用残差网络结构。具体地，可输出层前叠加输入，以增强输出和输入之间的对应关系。
可替换地，处理单元320还可使用通过将多个小卷积网络互相引入残差结构形成的级联的残差卷积网络。在级联残差网络中每个小卷积网络可以为3层的卷积网络，并且卷积核大小为3。与大深度超分辨率网络相比，级联残差网络可以在较小复杂度情况下得到更好的性能。
最后，处理单元320对第二信道进行下行预编码，以发送给终端。在结合图3描述的实施例中，基站的处理单元根据预编码矩阵指示信息进行传统的信道重建后，使用超分辨率网络对于第一粒度的信道进行插值和去噪处理以获得具有更精细粒度的第二信道，以用于下行预编码。
根据本公开的另一实施例，除了插值和去噪处理以外，还可使用神经网络来根据从终端接收第一粒度的预编码矩阵指示信息进行信道重建。以下，参考图4来说明根据本公开的另一实施例的基站。图4是示出根据本公开另一实施例的基站的示意性框图。如图4所示，根据本公开一个实施例的基站300可包括接收单元410、和处理单元420。除了接收单元和处理单元，基站400还可以包括其他部件，然而，由于这些部件与本公开实施例的内容无关，因此在这里省略其图示和描述。
如图4所示，基站400的接收单元410从终端接收第一粒度的预编码矩阵指示信息。例如，接收单元410可从终端接收子带级的类型II的PMI信息。具体地，类型II的PMI信息可包括子带级的幅度信息和相位信息。
处理单元420通过第一子网络，根据所述第一粒度的预编码矩阵指示信息进行信道重建、插值和去噪处理以获得第二信道，以及对所述第二粒度的信道进行下行预编码，其中所述第二粒度比所述第一粒度细。具体地，第一子网络可以是第一子神经网络。
根据本公开的一个示例，第一子网络的输入维度可比第一子网络的输出维度高。换言之，第一子网络采用高维输入低维输出设计。由于采用高维输入，可保存来自预编码矩阵指示信息的原始信息，并且由于采用低维输出，从而通过网络处理过程中降维，可降低网络复杂度和训练难度。
具体地，第一子网络的输入可是来自终端的预编码矩阵指示信息或者经过预处理的预编码矩阵指示信息，并且第一子网络可对该预编码矩阵指示信息进行输入重建。根据本公开的一个示例，第一子网络可加权合并输入数据的幅度和相位。例如，预编码矩阵指示信息可包括幅度信息和相位信息。进一步地，幅度信息可包括基站400与终端进行通信的各个波束的宽带波束信息、宽带幅度信息和各个子带的子带幅度信息，以及各个子带的子带相位信息。第一子网络可将各个波束的宽带波束信息、宽带幅度信息和各个子带的子带幅度信息合并，以获得信道幅度矩阵。具体地，第一子网络可根据天线阵的阵元极化方向，分别获得在各个极化方向上的信道幅度矩阵。此外，第一子网络可根据各个波束的宽带波束信息以及各子带相位信息获得实部矩阵和虚部矩阵。类似地，第一子网络可根据波束的极化方向，分别获得在各个极化方向上的实部矩阵和虚部矩阵。然后第一子网络可在各个极化方向上，将该极化方向上的信道幅度矩阵与实部矩阵和虚部矩阵分别相乘，以获得波束-频域信道矩阵(也可简称为“波束-频域信道”)。
例如，基站400以2个极化方向的波束与终端进行通信。第一子网络可获得两个极化方向的信道幅度矩阵A1、A2，两个极化方向的信道相位矩阵(即，实部矩阵和虚部矩阵)Pr1、Pr2、Pi1、Pi2。第一子网络可以通过以下公式(2)得到波束-频域信道矩阵Hr1、Hr2、Hi1、Hi2：
Hr1＝A1*Pr1,Hr2＝A2*Pr2；
Hi1＝A1*Pi1,Hi2＝A2*Pi2        ……(2)
其中公式(2)中的“*”表示矩阵对应元素相乘。
接下来，第一子网络可对波束-频域信道进行傅里叶变换，以将波束-频域 信道为波束-延迟域信道。此外，由于在波束-延迟域中，信道延迟分量主要集中在延迟域信道矩阵头部，第一子网络可对延迟域信道进行截断，以减少输出的维度。
根据本公开的一个示例，第一子网络可包括全连接层(Dense layer)，并且通过全连接层来对预编码矩阵指示信息进行输入重建。具体地，全连接层加权合并输入数据的幅度和相位以获得波束-频域信道，将波束-频域信道为波束-延迟域信道，并且对波束-延迟域信道进行截断，以降低网络输出维度。可替换地，由于加权合并处理中所需的操作仅为将各个矩阵对应元素相乘，因此可以将对应全连接层替换为部分连接层，该部分连接层仅连接需要直接进行乘法的元素。
根据本公开的另一个示例，第一子网络还可包括一个或多个超分辨率网络，以进行插值和去噪处理。一个或多个超分辨率网络可设置在上述全连接层或部分连接层之前或之后。此外，还可在多个超分辨率网络之间设置上述全连接层或部分连接层。以上已结合图3所示的示例对于根据本公开实施例的超分辨率网络进行了详细描述，故在此不再赘述。
图14A-图14C是示出了根据本公开实施例的第一子网络结构的示意图。例如，如图14A所示，可在全连接层或部分连接层之前设置超分辨率网络以对来自终端的预编码矩阵指示信息进行插值和去噪处理。然后将经过插值和去噪处理的数据输入到全连接层或部分连接层以进行信道重建和降维处理。又例如，如图14B所示，可将来自终端的预编码矩阵指示信息首先输入到全连接层或部分连接层以进行信道重建和降维处理，然后将得到的信道输入到超分辨率网络以进行插值和去噪处理。再例如，如图14C所示，可将来自终端的预编码矩阵指示信息输入至第一超分辨率网络以进行去噪处理。然后将经过去噪的数据输入至全连接层或部分连接层以进行信道重建和降维处理。最后将经过降维处理的信道输入至第二超分辨率网络以进行插值处理。
可选择地，根据本公开的另一示例，处理单元400还可通过第二子网络来对根据来自同一终端多次发送的预编码矩阵指示信息获得的多个第二信道进行时域信道估计增强和时域预测中的至少一个。例如，第一子网络可处理终端一次发送的(例如，在单个时隙内发送的)预编码矩阵指示信息，并基于终端一次发送的预编码矩阵指示信息进行信道重构、插值和去噪。第二子网 络包括递归神经网络(Recurrent Neural Network,RNN)和长短时记忆机制(Long and Short-Term Memory,LSTM)网络中的至少一个，并且第二子网络可将同一终端多次发送的预编码矩阵指示信息输入值RNN/LSTM网络，以实现时域信道估计增强和时域预测中的至少一个。
以下将结合图5对根据本公开的一个实施例，基站发送的信道状态信息参考信息进行说明。以下结合图5描述的信道状态信息参考信息可应用于结合图3和图4描述的基站。
图5是示出根据本公开另一实施例的基站的示意性框图。如图5所示，根据本公开另一实施例的基站500可包括发送单元510和接收单元520。除了发送单元和接收单元，基站500还可以包括其他部件，然而，由于这些部件与本公开实施例的内容无关，因此在这里省略其图示和描述。
如图5所示，发送单元510向终端发送第一密度的第一信道状态信息参考信息。第一密度可以是较高的密度。此外，根据本公开的一个示例，发送单元510以第一密度在基站500的整个通信带宽上发送第一信道状态信息参考信息。换言之，发送单元510可在整个通信带宽上发送高密度的第一信道状态信息参考信息。例如，可使用端口1-12所占用的资源块或资源元素发送第一信道状态信息参考信息。
根据本公开的一个施例，该第一信道状态信息参考信息可在频域上占满所有子载波，即，该第一信道状态信息参考信息的密度可达到12个资源元素(RE)每资源块(RB)每端口，并且每个端口的参考信号使用一个OFDM码元。此外，多个端口可以使用循环移位(cyclic shift)的方式在同一个OFDM码元上复用，或者以TDM的方式在不同OFDM码元上传输。或者，该第一信道状态信息参考信息的密度可达到6个资源元素(RE)每资源块(RB)每端口,并且每两个端口在频域上以交错梳状(comb)的形式复用，占用一个OFDM码元。类似地，也可以以循环移位和TDM的方式复用。
根据公开的另一示例，发送单元510还发送用于指示所述第一信道状态信息参考信息的信道状态信息参考信息配置信息，并且在训练数据采集时间段使用该信道状态信息参考信息配置信息所指示的资源来发送第一信道状态信息参考信息。例如，训练数据采集时间段可包括多个时隙。
然后，接收单元520接收终端发送的对于第一信道状态信息参考信息的 第一反馈信息。根据本公开的一个示例，终端可以是用户设备和数据采集装置中的至少一个。当终端为数据采集设备时，该数据采集设备可以通过专用接口将对于第一信道状态信息参考信息的第一反馈信息发送给基站500。另一方面，当终端为用户设备(UE)时，发送单元510还可向UE发送控制信令，以调度UE将对于第一信道状态信息参考信息的第一反馈信息通过数据信道发送给基站500。
例如，第一反馈信息可包括终端根据高密度的第一信道状态信息参考信息进行第一信道估计以获得第一信道响应信息(以下称为“信道响应数据集”)，以及终端根据对第一信道状态信息参考信息进行降采样处理后的信道状态信息参考信息确定预编码矩阵指示信息(以下称为“PMI训练数据集”)。
根据本公开的一个示例，图5中所示的基站还可包括处理单元530。处理单元530可根据第一反馈信息对以上结合图3和图4描述的基站中的超分辨率网络、第一子网络、第二子网络等神经网络进行训练，以便神经网络能够根据终端反馈的状态参考信号进行高精度信道估计，即，进行信道重建、去噪和插值。
处理单元530可使用PMI训练数据集和对应的信道响应数据集对神经网络进行训练。
图6是示出根据本公开的一个示例，使用PMI训练数据集和对应的信道响应数据集对神经网络进行训练的示意图。如图6所示，基站的处理单元将PMI训练数据集作为神经网络的第一输入，以使用神经网络根据该第一输入进行信道重建、去噪和插值处理。PMI训练数据集可用于模拟实际部署时终端发送的预编码矩阵指示信息。另一方面，基站的处理单元可将高密度的第一信道状态信息参考信息的信道响应数据集作为神经网络的第二输入以进行网络优化，从而例如，向网络提供对于特定PMI训练数据集的目标信道响应。
基站500可以是，例如，用于神经网络训练的专用基站。专用基站500可以在完成训练后，将训练好的神经网络的提供给通信网络中实际与UE进行通信的基站，以便实际进行通信时可使用训练好的神经网络进行信道重建、去噪和/或插值处理等操作。
又例如，基站500在完成训练后可用于实际与UE进行通信。在实际部署时，发送单元510向UE发送第二密度的第二信道状态信息参考信息。并 且接收单元520接收UE对于第二信道状态信息参考信息的第二反馈信息。第二信道状态信息参考信息可以是现有的、在实际通信用于信道估计的信道状态信息参考信息，并且所述第一密度大于所述第二密度。也就是说，第二信道状态信息参考信息比第一信道状态信息参考信息稀疏。处理单元530可使用预先训练的网络，根据第二信道状态信息参考信息进行信道重建、去噪和插值，以获得将用于在实际通信中要进行下行预编码的高精度信道。进一步地，在实际部署时，处理单元530可执行与以上描述的处理单元320和处理单元420类似的操作，在此不再赘述。
以下将结合图7对根据本公开的一个实施例、与图5中所述的基站对应的终端进行说明。图7是示出根据本公开一个实施例的终端的示意性框图。如图7所示，根据本公开另一实施例的终端700可包括接收单元710、处理单元720、和发送单元730。除了发送单元、接收单元和处理单元，终端700还可以包括其他部件，然而，由于这些部件与本公开实施例的内容无关，因此在这里省略其图示和描述。
如图7所示，接收单元710接收第一密度的第一信道状态信息参考信息。如上所述，第一信道状态信息参考信息可用于对基站的神经网络进行训练。根据本公开的一个示例，第一信道状态信息参考信息可以是在基的整个通信带宽上发送的，并且第一信道状态信息参考信息具有比在实际部署阶段用于进行信道测量的信道状态信息参考信息更高的密度。
处理单元720根据第一信道状态信息参考信息进行第一信道估计以获得第一信道响应信息，以及对第一信道状态信息参考信息进行降采样处理，并根据降采样处理后的信道状态信息参考信息确定预编码矩阵指示信息。图8是示出根据本公开一个示例，处理单元720对第一信道状态信息参考信息进行处理的示意图。如图8所示，一方面，处理单元对具有高密度的第一信道状态信息参考信息进行第一信道估计以获得高精度第一信道响应信息。另一方面，处理单元对第一信道状态信息参考信息进行降采样以模拟中的在实际部署阶段接收的信道状态信息参考信息密度，并获得低密度信道。然后处理单元根据降采样处理后的信道状态信息参考信息进行PMI反馈量计算来确定预编码矩阵指示信息。
返回图7，发送单元730向基站发送所述第一信道响应信息和预编码矩 阵指示信息。从而基站可根据所接收的第一信道响应信息和预编码矩阵指示信息对其神经网络进行训练。根据本公开的一个示例，终端700可以是用户设备(UE)和数据采集装置中的至少一个。当终端700为用于训练基站的神经网络的数据采集设备时，发送单元730可通过离线方式或经过空口等多种方式将第一信道响应信息和预编码矩阵指示信息发送到基站。当终端700为UE时，发送单元730可使用预编码矩阵指示反馈信道将预编码矩阵指示信息发送给基站，并且可在调制编码后使用上行数据信道或上行控制信道将第一信道响应信息传输给基站。基站可根据第一信道响应信息和预编码矩阵指示信息对其神经网络进行信道重建、插值、去噪等训练。
下面，参照图9来描述根据本公开实施例的信道处理方法。图9是根据本公开的一个实施例的信道处理方法900的流程图。由于信道处理方法900的步骤与上文参照图描述的基站300的操作对应，因此在这里为了简单起见，省略对相同内容的详细描述。
如图9所示，在步骤S901中，从终端接收第一粒度的预编码矩阵指示信息。在步骤S902中，根据所述预编码矩阵指示信息进行信道重建以获得第一信道，其中所述第一信道具有第一粒度。例如，在步骤S902中可使用所接收的PMI信息中的幅度信息和相位信息，分别对多个波束的空域(也可称为“波束域”)-频域信道码字进行幅度和相位加权，并且将加权后的向量合并以得到具有子带级的第一信道。合并后得到的子带级的第一信道可以以空域-频域信道矩阵的形式表示，其中空域-频域信道矩阵的空域值可以是基站与终端进行通信的天线的数量，而频域值可以根据基站与终端进行通信的子带的数量确定。
然后，在步骤S903中，使用超分辨率网络对于所述第一粒度的信道进行插值和去噪处理以获得第二信道，其中所述第二信道具有第二粒度，并且第二粒度比第一粒度细。
根据本公开的一个示例，图9中所示的方法还可包括在将第一粒度的第一信道输入到超分辨率网络之前，可对第一信道进行预处理，以便于后续超分辨率网络的操作。例如，在第一信道为空域-频域信道的情况下，可对第一信道进行傅里叶变换，以将空域-频域信道为波束-延迟域信道。此外，由于在波束-延迟域中，信道延迟分量主要集中在延迟域信道矩阵头部，因此可对延 迟域信道进行截断，保留头部，并且将截断后数据分为实部和虚部两个通道作为超分辨率网络的输入。通过对将要处理的第一信道变换至延迟域并对数据进行阶段，可减少超分辨率网络计算的复杂度。
此外为了进一步简化超分辨率网络的操作，图9中所示的方法还可包括在将其变换至波束-延迟域信道之前，使用填零或线性插值、最邻近插值等已有的差值方法插值对第一信道进行预差值，以获得到期望的频域精度(例如RB级或子载波级)的信道。
最后，在步骤S904中对第二信道进行下行预编码，以发送给终端。在结合图9描述的信道处理方法中，根据预编码矩阵指示信息进行传统的信道重建后，使用超分辨率网络对于第一粒度的信道进行插值和去噪处理以获得具有更精细粒度的第二信道，以用于下行预编码。
根据本公开的另一实施例，除了插值和去噪处理以外，还可使用神经网络来根据从终端接收第一粒度的预编码矩阵指示信息进行信道重建。下面，参照图10来描述根据本公开另一实施例的信道处理方法。图10是根据本公开的另一实施例的信道处理方法1000的流程图。由于信道处理方法1000的步骤与上文参照图描述的基站400的操作对应，因此在这里为了简单起见，省略对相同内容的详细描述。
如图10所示，在步骤S1001中，从终端接收第一粒度的预编码矩阵指示信息。例如，可从终端接收子带级的类型II的PMI信息。具体地，类型II的PMI信息可包括子带级的幅度信息和相位信息。
在步骤S1002中，通过第一子网络，根据所述第一粒度的预编码矩阵指示信息进行信道重建、插值和去噪处理以获得第二信道，以及对所述第二粒度的信道进行下行预编码，其中所述第二粒度比所述第一粒度细。具体地，第一子网络可以是第一子神经网络。
根据本公开的一个示例，第一子网络的输入维度可比第一子网络的输出维度高。换言之，第一子网络采用高维输入低维输出设计。由于采用高维输入，可保存来自预编码矩阵指示信息的原始信息，并且由于采用低维输出，从而通过网络处理过程中降维，可降低网络复杂度和训练难度。
具体地，第一子网络的输入可是来自终端的预编码矩阵指示信息或者经过预处理的预编码矩阵指示信息，并且第一子网络可对该预编码矩阵指示信 息进行输入重建。根据本公开的一个示例，在步骤S1002中，第一子网络可加权合并输入数据的幅度和相位。例如，预编码矩阵指示信息可包括幅度信息和相位信息。进一步地，幅度信息可包括基站与终端进行通信的各个波束的宽带波束信息、宽带幅度信息和各个子带的子带幅度信息，以及各个子带的子带相位信息。在步骤S1002中，第一子网络可将各个波束的宽带波束信息、宽带幅度信息和各个子带的子带幅度信息合并，以获得信道幅度矩阵。具体地，第一子网络可根据波束的极化方向，分别获得在各个极化方向上的信道幅度矩阵。此外，在步骤S1002中，第一子网络还可根据各个波束的宽带波束信息以及各子带相位信息获得实部矩阵和虚部矩阵。类似地，第一子网络可根据波束的极化方向，分别获得在各个极化方向上的实部矩阵和虚部矩阵。然后第一子网络可在各个极化方向上，将该极化方向上的信道幅度矩阵与实部矩阵和虚部矩阵分别相乘，以获得波束-频域信道矩阵(也可简称为“波束-频域信道”)。
接下来，在步骤S1002中，第一子网络可对波束-频域信道进行傅里叶变换，以将波束-频域信道为波束-延迟域信道。此外，由于在波束-延迟域中，信道延迟分量主要集中在延迟域信道矩阵头部，第一子网络可对延迟域信道进行截断，以减少输出的维度。
根据本公开的一个示例，第一子网络可包括全连接层(Dense layer)，并且通过全连接层来对预编码矩阵指示信息进行输入重建。具体地，全连接层加权合并输入数据的幅度和相位以获得波束-频域信道，将波束-频域信道为波束-延迟域信道，并且对波束-延迟域信道进行截断，以降低网络输出维度。可替换地，由于加权合并处理中所需的操作仅为将各个矩阵对应元素相乘，因此可以将对应全连接层替换为部分连接层，该部分连接层仅连接需要直接进行乘法的元素。
根据本公开的另一个示例，第一子网络还可包括一个或多个超分辨率网络，以进行插值和去噪处理。一个或多个超分辨率网络可设置在上述全连接层或部分连接层之前或之后。此外，还可在多个超分辨率网络之间设置上述全连接层或部分连接层。
例如，可在全连接层或部分连接层之前设置超分辨率网络以对来自终端的预编码矩阵指示信息进行插值和去噪处理。然后将经过插值和去噪处理的 数据输入到全连接层或部分连接层以进行信道重建和降维处理。又例如，可将来自终端的预编码矩阵指示信息首先输入到全连接层或部分连接层以进行信道重建和降维处理，然后将得到的信道输入到超分辨率网络以进行插值和去噪处理。再例如，可将来自终端的预编码矩阵指示信息输入至第一超分辨率网络以进行去噪处理。然后将经过去噪的数据输入至全连接层或部分连接层以进行信道重建和降维处理。最后将经过降维处理的信道输入至第二超分辨率网络以进行插值处理。
可选择地，图10中所示的方法还可包括通过第二子网络来对根据来自同一终端多次发送的预编码矩阵指示信息获得的多个第二信道进行时域信道估计增强和时域预测中的至少一个。例如，第一子网络可处理终端一次发送的(例如，在单个时隙内发送的)预编码矩阵指示信息，并基于终端一次发送的预编码矩阵指示信息进行信道重构、插值和去噪。第二子网络包括RNN和LSTM网络中的至少一个，并且第二子网络可将同一终端多次发送的预编码矩阵指示信息输入值RNN/LSTM网络，以实现时域信道估计增强和时域预测中的至少一个。
以下将结合图11对根据本公开的一个实施例的参考信号发送方法进行说明。图11是根据本公开的一个实施例的参考信号发送方法1100的流程图。由于参考信号发送方法1100的步骤与上文参照图描述的基站500的操作对应，因此在这里为了简单起见，省略对相同内容的详细描述。
如图11所示，在步骤S1101中，向终端发送第一密度的第一信道状态信息参考信息。第一密度可以是较高的密度。此外，根据本公开的一个示例，在步骤S1101中，以第一密度在基站的整个通信带宽上发送第一信道状态信息参考信息。换言之，在步骤S1101中，可在整个通信带宽上发送高密度的第一信道状态信息参考信息。
根据公开的另一示例，方法1100还可包括发送用于指示所述第一信道状态信息参考信息的信道状态信息参考信息配置信息，并且在训练数据采集时间段使用该信道状态信息参考信息配置信息所指示的资源来发送第一信道状态信息参考信息。例如，训练数据采集时间段可包括多个时隙。
然后，在步骤S1102中，接收终端发送的对于第一信道状态信息参考信息的第一反馈信息。根据本公开的一个示例，终端可以是用户设备和数据采 集装置中的至少一个。当终端为数据采集设备时，该数据采集设备可以通过专用接口将对于第一信道状态信息参考信息的第一反馈信息发送给基站。另一方面，当终端为用户设备(UE)时，方法1100还可包括向UE发送控制信令，以调度UE将对于第一信道状态信息参考信息的第一反馈信息通过数据信道发送给基站500。
例如，第一反馈信息可包括终端根据高密度的第一信道状态信息参考信息进行第一信道估计以获得第一信道响应信息(以下称为“信道响应数据集”)，以及终端根据对第一信道状态信息参考信息进行降采样处理后的信道状态信息参考信息确定预编码矩阵指示信息(以下称为“PMI训练数据集”)。处理单元530可使用PMI训练数据集和对应的信道响应数据集对神经网络进行训练。
此外，根据本公开的另一示例，方法1100还可包括根据第一反馈信息对以上结合图3和图4描述的基站中的超分辨率网络、第一子网络、第二子网络等神经网络进行训练，以便神经网络能够根据终端反馈的状态参考信号进行高精度信道估计，即，进行信道重建、去噪和插值。
基站可以是，例如，用于神经网络训练的专用基站。专用基站可以在完成训练后，将训练好的神经网络的提供给通信网络中实际与UE进行通信的基站，以便实际进行通信时可使用训练好的神经网络进行信道重建、去噪和/或插值处理等操作。
又例如，基站在完成训练后可用于实际与UE进行通信。在此情况下在实际部署时，如图11所示，在步骤S1103中，向UE发送第二密度的第二信道状态信息参考信息。并且在步骤S1104中，接收UE对于第二信道状态信息参考信息的第二反馈信息。第二信道状态信息参考信息可以是现有的、在实际通信用于信道估计的信道状态信息参考信息，并且所述第一密度大于所述第二密度。也就是说，第二信道状态信息参考信息比第一信道状态信息参考信息稀疏。然后在步骤S1105中，使用预先训练的网络，根据第二信道状态信息参考信息进行信道重建、去噪和插值，以获得将用于在实际通信中要进行下行预编码的高精度信道。
以下将结合图12对根据本公开的一个实施例的、由终端执行的信息发送方法进行说明。图12是根据本公开的一个实施例的信息发送方法1200的流 程图。由于参考信号发送方法1200的步骤与上文参照图描述的终端700的操作对应，因此在这里为了简单起见，省略对相同内容的详细描述。
如图12所示，在步骤S1201中，接收第一密度的第一信道状态信息参考信息。如上所述，第一信道状态信息参考信息可用于对基站的神经网络进行训练。根据本公开的一个示例，第一信道状态信息参考信息可以是在基的整个通信带宽上发送的，并且第一信道状态信息参考信息具有比在实际部署阶段用于进行信道测量的信道状态信息参考信息更高的密度。
在步骤S1202中，根据第一信道状态信息参考信息进行第一信道估计以获得第一信道响应信息。并且在步骤S1203中对第一信道状态信息参考信息进行降采样处理，并根据降采样处理后的信道状态信息参考信息确定预编码矩阵指示信息。虽然在图12中以先执行步骤S1202再执行步骤S1203为例进行了说明，但是本公开不限于此。例如，也可先执行步骤S1203再执行步骤S1201，或者同时执行步骤S1202和步骤S1203。
然后在步骤S1204中向基站发送所述第一信道响应信息和预编码矩阵指示信息。从而基站可根据所接收的第一信道响应信息和预编码矩阵指示信息对其神经网络进行训练。
<硬件结构>
另外，上述实施方式的说明中使用的框图示出了以功能为单位的块。这些功能块(结构单元)通过硬件和/或软件的任意组合来实现。此外，各功能块的实现手段并不特别限定。即，各功能块可以通过在物理上和/或逻辑上相结合的一个装置来实现，也可以将在物理上和/或逻辑上相分离的两个以上装置直接地和/或间接地(例如通过有线和/或无线)连接从而通过上述多个装置来实现。
例如，本公开的一个实施例的第一网络元件可以作为执行本公开的无线通信方法的处理的计算机来发挥功能。图13是根据本公开的实施例的所涉及的设备1300(基站、终端)的硬件结构的示意图。上述的设备1300(基站、终端)可以作为在物理上包括处理器1310、内存1320、存储器1330、通信装置1340、输入装置1350、输出装置1360、总线1370等的计算机装置来构成。
另外，在以下的说明中，“装置”这样的文字也可替换为电路、设备、单元等。第一网络元件的硬件结构可以包括一个或多个图中所示的各装置，也可以不包括部分装置。
例如，处理器1310仅图示出一个，但也可以为多个处理器。此外，可以通过一个处理器来执行处理，也可以通过一个以上的处理器同时、依次、或采用其它方法来执行处理。另外，处理器1310可以通过一个以上的芯片来安装。
设备1300的各功能例如通过如下方式实现：通过将规定的软件(程序)读入到处理器1310、内存1320等硬件上，从而使处理器1310进行运算，对由通信装置1340进行的通信进行控制，并对内存1320和存储器330中的数据的读出和/或写入进行控制。
处理器1310例如使操作系统进行工作从而对计算机整体进行控制。处理器1310可以由包括与周边装置的接口、控制装置、运算装置、寄存器等的中央处理器(CPU，Central Processing Unit)构成。例如，上述的处理单元等可以通过处理器1310实现。
此外，处理器1310将程序(程序代码)、软件模块、数据等从存储器1330和/或通信装置1340读出到内存1320，并根据它们执行各种处理。作为程序，可以采用使计算机执行在上述实施方式中说明的动作中的至少一部分的程序。例如，第一网络元件的处理单元可以通过保存在内存1320中并通过处理器1310来工作的控制程序来实现，对于其它功能块，也可以同样地来实现。
内存1320是计算机可读取记录介质，例如可以由只读存储器(ROM，Read Only Memory)、可编程只读存储器(EPROM，Erasable Programmable ROM)、电可编程只读存储器(EEPROM，Electrically EPROM)、随机存取存储器(RAM，Random Access Memory)、其它适当的存储介质中的至少一个来构成。内存1320也可以称为寄存器、高速缓存、主存储器(主存储装置)等。内存1320可以保存用于实施本公开的一实施方式所涉及的方法的可执行程序(程序代码)、软件模块等。
存储器1330是计算机可读取记录介质，例如可以由软磁盘(flexible disk)、软(注册商标)盘(floppy disk)、磁光盘(例如，只读光盘(CD-ROM(Compact Disc ROM)等)、数字通用光盘、蓝光(Blu-ray，注册商标)光盘)、可移 动磁盘、硬盘驱动器、智能卡、闪存设备(例如，卡、棒(stick)、密钥驱动器(key driver))、磁条、数据库、服务器、其它适当的存储介质中的至少一个来构成。存储器1330也可以称为辅助存储装置。
通信装置1340是用于通过有线和/或无线网络进行计算机间的通信的硬件(发送接收装置)，例如也称为网络设备、网络控制器、网卡、通信模块等。通信装置1340为了实现例如频分双工(FDD，Frequency Division Duplex)和/或时分双工(TDD，Time Division Duplex)，可以包括高频开关、双工器、滤波器、频率合成器等。例如，上述的发送单元、接收单元等可以通过通信装置1340来实现。
输入装置1350是接受来自外部的输入的输入设备(例如，键盘、鼠标、麦克风、开关、按钮、传感器等)。输出装置1360是实施向外部的输出的输出设备(例如，显示器、扬声器、发光二极管(LED，Light Emitting Diode)灯等)。另外，输入装置1350和输出装置1360也可以为一体的结构(例如触控面板)。
此外，处理器1310、内存1320等各装置通过用于对信息进行通信的总线1370连接。总线1370可以由单一的总线构成，也可以由装置间不同的总线构成。
此外，电子设备可以包括微处理器、数字信号处理器(DSP，Digital Signal Processor)、专用集成电路(ASIC，Application Specific Integrated Circuit)、可编程逻辑器件(PLD，Programmable Logic Device)、现场可编程门阵列(FPGA，Field Programmable Gate Array)等硬件，可以通过该硬件来实现各功能块的部分或全部。例如，处理器710可以通过这些硬件中的至少一个来安装。
(变形例)
另外，关于本说明书中说明的用语和/或对本说明书进行理解所需的用语，可以与具有相同或类似含义的用语进行互换。例如，信道和/或符号也可以为信号(信令)。此外，信号也可以为消息。参考信号也可以简称为RS(Reference Signal)，根据所适用的标准，也可以称为导频(Pilot)、导频信号等。此外，分量载波(CC，Component Carrier)也可以称为小区、频率载波、载波频率等。
此外，本说明书中说明的信息、参数等可以用绝对值来表示，也可以用与规定值的相对值来表示，还可以用对应的其它信息来表示。例如，无线资源可以通过规定的索引来指示。进一步地，使用这些参数的公式等也可以与本说明书中明确公开的不同。
在本说明书中用于参数等的名称在任何方面都并非限定性的。例如，各种各样的信道(物理上行链路控制信道(PUCCH，Physical Uplink Control Channel)、物理下行链路控制信道(PDCCH，Physical Downlink Control Channel)等)和信息单元可以通过任何适当的名称来识别，因此为这些各种各样的信道和信息单元所分配的各种各样的名称在任何方面都并非限定性的。
本说明书中说明的信息、信号等可以使用各种各样不同技术中的任意一种来表示。例如，在上述的全部说明中可能提及的数据、命令、指令、信息、信号、比特、符号、芯片等可以通过电压、电流、电磁波、磁场或磁性粒子、光场或光子、或者它们的任意组合来表示。
此外，信息、信号等可以从上层向下层、和/或从下层向上层输出。信息、信号等可以经由多个网络节点进行输入或输出。
输入或输出的信息、信号等可以保存在特定的场所(例如内存)，也可以通过管理表进行管理。输入或输出的信息、信号等可以被覆盖、更新或补充。输出的信息、信号等可以被删除。输入的信息、信号等可以被发往其它装置。
信息的通知并不限于本说明书中说明的方式/实施方式，也可以通过其它方法进行。例如，信息的通知可以通过物理层信令(例如，下行链路控制信息(DCI，Downlink Control Information)、上行链路控制信息(UCI，Uplink Control Information))、上层信令(例如，无线资源控制(RRC，Radio Resource Control)信令、广播信息(主信息块(MIB，Master Information Block)、系统信息块(SIB，System Information Block)等)、媒体存取控制(MAC，Medium Access Control)信令)、其它信号或者它们的组合来实施。
另外，物理层信令也可以称为L1/L2(第1层/第2层)控制信息(L1/L2控制信号)、L1控制信息(L1控制信号)等。此外，RRC信令也可以称为RRC消息，例如可以为RRC连接建立(RRC Connection Setup)消息、RRC连接重设定(RRC Connection Reconfiguration)消息等。此外，MAC信令例如可以通过MAC控制单元(MAC CE(Control Element))来通知。
此外，规定信息的通知(例如，“为X”的通知)并不限于显式地进行，也可以隐式地(例如，通过不进行该规定信息的通知，或者通过其它信息的通知)进行。
关于判定，可以通过由1比特表示的值(0或1)来进行，也可以通过由真(true)或假(false)表示的真假值(布尔值)来进行，还可以通过数值的比较(例如与规定值的比较)来进行。
软件无论被称为软件、固件、中间件、微代码、硬件描述语言，还是以其它名称来称呼，都应宽泛地解释为是指命令、命令集、代码、代码段、程序代码、程序、子程序、软件模块、应用程序、软件应用程序、软件包、例程、子例程、对象、可执行文件、执行线程、步骤、功能等。
此外，软件、命令、信息等可以经由传输介质被发送或接收。例如，当使用有线技术(同轴电缆、光缆、双绞线、数字用户线路(DSL，Digital Subscriber Line)等)和/或无线技术(红外线、微波等)从网站、服务器、或其它远程资源发送软件时，这些有线技术和/或无线技术包括在传输介质的定义内。
本说明书中使用的“系统”和“网络”这样的用语可以互换使用。
在本说明书中，“基站(BS，Base Station)”、“无线基站”、“eNB”、“gNB”、“小区”、“扇区”、“小区组”、“载波”以及“分量载波”这样的用语可以互换使用。基站有时也以固定台(fixed station)、NodeB、eNodeB(eNB)、接入点(access point)、发送点、接收点、毫微微小区、小小区等用语来称呼。
基站可以容纳一个或多个(例如三个)小区(也称为扇区)。当基站容纳多个小区时，基站的整个覆盖区域可以划分为多个更小的区域，每个更小的区域也可以通过基站子系统(例如，室内用小型基站(射频拉远头(RRH，Remote Radio Head)))来提供通信服务。“小区”或“扇区”这样的用语是指在该覆盖中进行通信服务的基站和/或基站子系统的覆盖区域的一部分或整体。
在本说明书中，“移动台(MS，Mobile Station)”、“用户终端(user terminal)”、“用户装置(UE，User Equipment)”以及“终端”这样的用语可以互换使用。移动台有时也被本领域技术人员以用户台、移动单元、用户单元、无线单元、远程单元、移动设备、无线设备、无线通信设备、远程设备、移动用户台、接入终端、移动终端、无线终端、远程终端、手持机、用户代理、移动客户端、客户端或者若干其它适当的用语来称呼。
此外，本说明书中的无线基站也可以用用户终端来替换。例如，对于将无线基站和用户终端间的通信替换为多个用户终端间(D2D，Device-to-Device)的通信的结构，也可以应用本公开的各方式/实施方式。此时，可以将上述的电子设备所具有的功能当作用户终端所具有的功能。此外，“上行”和“下行”等文字也可以替换为“侧”。例如，上行信道也可以替换为侧信道。
同样，本说明书中的用户终端也可以用无线基站来替换。此时，可以将上述的用户终端所具有的功能当作第一通信设备或第二通信设备所具有的功能。
在本说明书中，设为通过基站进行的特定动作根据情况有时也通过其上级节点(upper node)来进行。显然，在具有基站的由一个或多个网络节点(network nodes)构成的网络中，为了与终端间的通信而进行的各种各样的动作可以通过基站、除基站之外的一个以上的网络节点(可以考虑例如移动管理实体(MME，Mobility Management Entity)、服务网关(S-GW，Serving-Gateway)等，但不限于此)、或者它们的组合来进行。
本说明书中说明的各方式/实施方式可以单独使用，也可以组合使用，还可以在执行过程中进行切换来使用。此外，本说明书中说明的各方式/实施方式的处理步骤、序列、流程图等只要没有矛盾，就可以更换顺序。例如，关于本说明书中说明的方法，以示例性的顺序给出了各种各样的步骤单元，而并不限定于给出的特定顺序。
本说明书中说明的各方式/实施方式可以应用于利用长期演进(LTE，Long Term Evolution)、高级长期演进(LTE-A，LTE-Advanced)、超越长期演进(LTE-B，LTE-Beyond)、超级第3代移动通信系统(SUPER 3G)、高级国际移动通信(IMT-Advanced)、第4代移动通信系统(4G，4th generation mobile communication system)、第5代移动通信系统(5G，5th generation mobile communication system)、未来无线接入(FRA，Future Radio Access)、新无线接入技术(New-RAT，Radio Access Technology)、新无线(NR，New Radio)、新无线接入(NX，New radio access)、新一代无线接入(FX，Future generation radio access)、全球移动通信系统(GSM(注册商标)，Global System for Mobile communications)、码分多址接入3000(CDMA3000)、超级移动宽带(UMB，Ultra Mobile Broadband)、IEEE 920.11(Wi-Fi(注册商标))、IEEE 920.16 (WiMAX(注册商标))、IEEE 920.20、超宽带(UWB，Ultra-WideBand)、蓝牙(Bluetooth(注册商标))、其它适当的无线通信方法的系统和/或基于它们而扩展的下一代系统。
本说明书中使用的“根据”这样的记载，只要未在其它段落中明确记载，则并不意味着“仅根据”。换言之，“根据”这样的记载是指“仅根据”和“至少根据”这两者。
本说明书中使用的对使用“第一”、“第二”等名称的单元的任何参照，均非全面限定这些单元的数量或顺序。这些名称可以作为区别两个以上单元的便利方法而在本说明书中使用。因此，第一单元和第二单元的参照并不意味着仅可采用两个单元或者第一单元必须以若干形式占先于第二单元。
本说明书中使用的“判断(确定)(determining)”这样的用语有时包含多种多样的动作。例如，关于“判断(确定)”，可以将计算(calculating)、推算(computing)、处理(processing)、推导(deriving)、调查(investigating)、搜索(looking up)(例如表、数据库、或其它数据结构中的搜索)、确认(ascertaining)等视为是进行“判断(确定)”。此外，关于“判断(确定)”，也可以将接收(receiving)(例如接收信息)、发送(transmitting)(例如发送信息)、输入(input)、输出(output)、存取(accessing)(例如存取内存中的数据)等视为是进行“判断(确定)”。此外，关于“判断(确定)”，还可以将解决(resolving)、选择(selecting)、选定(choosing)、建立(establishing)、比较(comparing)等视为是进行“判断(确定)”。也就是说，关于“判断(确定)”，可以将若干动作视为是进行“判断(确定)”。
本说明书中使用的“连接的(connected)”、“结合的(coupled)”这样的用语或者它们的任何变形是指两个或两个以上单元间的直接的或间接的任何连接或结合，可以包括以下情况：在相互“连接”或“结合”的两个单元间，存在一个或一个以上的中间单元。单元间的结合或连接可以是物理上的，也可以是逻辑上的，或者还可以是两者的组合。例如，“连接”也可以替换为“接入”。在本说明书中使用时，可以认为两个单元是通过使用一个或一个以上的电线、线缆、和/或印刷电气连接，以及作为若干非限定性且非穷尽性的示例，通过使用具有射频区域、微波区域、和/或光(可见光及不可见光这两者)区域的波长的电磁能等，被相互“连接”或“结合”。
在本说明书或权利要求书中使用“包括(including)”、“包含(comprising)”、以及它们的变形时，这些用语与用语“具备”同样是开放式的。进一步地，在本说明书或权利要求书中使用的用语“或(or)”并非是异或。
以上对本公开进行了详细说明，但对于本领域技术人员而言，显然，本公开并非限定于本说明书中说明的实施方式。本公开在不脱离由权利要求书的记载所确定的本公开的宗旨和范围的前提下，可以作为修改和变更方式来实施。因此，本说明书的记载是以示例说明为目的，对本公开而言并非具有任何限制性的意义。

Claims (10)

  1. 一种基站，包括：
    接收单元，配置为从终端接收第一粒度的预编码矩阵指示信息；
    处理单元，配置为根据所述预编码矩阵指示信息进行信道重建以获得第一信道，使用超分辨率网络对于所述第一粒度的信道进行插值和去噪处理以获得第二信道，以及对所述第二信道进行下行预编码，其中
    所述第一信道具有所述第一粒度，
    所述第二信道具有第二粒度，
    所述第二粒度比所述第一粒度细。
  2. 如权利要求1所述的基站，其中
    所述第一信道为空域-频域信道；以及
    所述处理单元还配置为将重建的所述第一信道变换至延迟域以获得延迟域信道，以及将所述延迟域信道分为实部和虚部以输入到所述超分辨率网络。
  3. 一种基站，包括：
    接收单元，配置为从终端接收第一粒度的预编码矩阵指示信息；
    处理单元，配置为通过第一子网络，根据所述第一粒度的预编码矩阵指示信息进行信道重建、插值和去噪处理以获得第二信道，以及对所述第二粒度的信道进行下行预编码，其中
    所述第二粒度比所述第一粒度细。
  4. 如权利要求3所述的基站，其中
    所述第一子网络根据所述预编码矩阵指示信息进行幅度与相位的加权合并以获得波束-频域信道，将所述波束-频域信道变换为波束-延迟域信道，以及对所述波束-延迟域信道进行截断。
  5. 如权利3或4所述的基站，其中
    所述处理单元还配置为通过第二子网络，对根据来自同一终端多次发送 的预编码矩阵指示信息获得的多个第二信道进行时域信道估计增强和时域预测中的至少一个。
  6. 一种基站，包括：
    发送单元，配置为向终端发送第一密度的第一信道状态信息参考信息；
    接收单元，配置为从所述终端接收对于所述第一信道状态信息参考信息的第一反馈信息；其中
    所述第一反馈信息包括终端根据所述第一信道状态信息参考信息进行第一信道估计所获得第一信道响应信息，以及所述终端根据降采样处理后的第一信道状态信息参考信息确定的预编码矩阵指示信息。
  7. 如权利要求6所述的基站，其中
    所述发送单元以所述第一密度在所述基站的整个通信带宽上发送所述第一信道状态信息参考信息。
  8. 如权利要求6或7所述的基站，其中
    所述发送单元还配置为发送用于指示所述第一信道状态信息参考信息的信道状态信息参考信息配置信息；以及
    所述发送单元还配置为在特定训练数据采集时间出发该CSI-RS的测量。
  9. 一种终端，包括：
    接收单元，配置为接收第一密度的第一信道状态信息参考信息；
    处理单元，配置为根据所述第一信道状态信息参考信息进行第一信道估计以获得第一信道响应信息，以及对所述第一信道状态信息参考信息进行降采样处理，并根据降采样处理后的信道状态信息参考信息确定预编码矩阵指示信息；以及
    发送单元，配置为向基站发送所述第一信道响应信息和预编码矩阵指示信息。
  10. 如权利要求9所述的终端，其中
    所述终端为用户设备和数据采集装置中的至少一个。
PCT/CN2021/074690 2021-02-01 2021-02-01 基站和终端 WO2022160344A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
PCT/CN2021/074690 WO2022160344A1 (zh) 2021-02-01 2021-02-01 基站和终端
CN202180089526.XA CN116746077A (zh) 2021-02-01 2021-02-01 基站和终端
US18/258,641 US20240056142A1 (en) 2021-02-01 2021-02-01 Base station and terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/074690 WO2022160344A1 (zh) 2021-02-01 2021-02-01 基站和终端

Publications (1)

Publication Number Publication Date
WO2022160344A1 true WO2022160344A1 (zh) 2022-08-04

Family

ID=82652938

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/074690 WO2022160344A1 (zh) 2021-02-01 2021-02-01 基站和终端

Country Status (3)

Country Link
US (1) US20240056142A1 (zh)
CN (1) CN116746077A (zh)
WO (1) WO2022160344A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101257367A (zh) * 2007-02-28 2008-09-03 皇家飞利浦电子股份有限公司 选择预编码的方法和装置
EP2469729A1 (en) * 2009-08-17 2012-06-27 Alcatel Lucent Method and apparatus for keeping the precoding channel coherency in a communication network
CN104782071A (zh) * 2013-10-18 2015-07-15 华为技术有限公司 信道状态信息的测量和反馈方法、终端及基站
US20180254814A1 (en) * 2015-09-14 2018-09-06 Lg Electronics Inc. Method for transmitting and receiving channel state information (csi) in wireless communication system, and apparatus therefor
CN110463072A (zh) * 2017-01-31 2019-11-15 Lg电子株式会社 在无线通信系统中报告信道状态信息的方法和设备
WO2020012736A1 (ja) * 2018-07-10 2020-01-16 パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ 基地局、端末及び通信方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101257367A (zh) * 2007-02-28 2008-09-03 皇家飞利浦电子股份有限公司 选择预编码的方法和装置
EP2469729A1 (en) * 2009-08-17 2012-06-27 Alcatel Lucent Method and apparatus for keeping the precoding channel coherency in a communication network
CN104782071A (zh) * 2013-10-18 2015-07-15 华为技术有限公司 信道状态信息的测量和反馈方法、终端及基站
US20180254814A1 (en) * 2015-09-14 2018-09-06 Lg Electronics Inc. Method for transmitting and receiving channel state information (csi) in wireless communication system, and apparatus therefor
CN110463072A (zh) * 2017-01-31 2019-11-15 Lg电子株式会社 在无线通信系统中报告信道状态信息的方法和设备
WO2020012736A1 (ja) * 2018-07-10 2020-01-16 パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ 基地局、端末及び通信方法

Also Published As

Publication number Publication date
CN116746077A (zh) 2023-09-12
US20240056142A1 (en) 2024-02-15

Similar Documents

Publication Publication Date Title
CN117044125A (zh) 用于支持在fdd mimo系统中进行csi反馈的机器学习或人工智能技术的方法和装置
WO2021253937A1 (zh) 无线通信系统的终端、基站以及由终端和基站执行的方法
CN112751592B (zh) 上报信道状态信息的方法和通信装置
US10243634B2 (en) Method and device for dual layer beamforming
WO2019047827A9 (zh) 一种指示及确定预编码矩阵的方法和设备
CN111342913B (zh) 一种信道测量方法和通信装置
WO2021063178A1 (zh) 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
CN114642019A (zh) 一种信道信息获取的方法
WO2021083157A1 (zh) 一种预编码矩阵的处理方法和通信装置
CN112533295A (zh) 参数配置方法和通信装置
CN112054824B (zh) 一种信道测量方法和通信装置
CN111757382B (zh) 指示信道状态信息的方法以及通信装置
WO2023160247A1 (zh) 下行传输的方法及装置
WO2022160344A1 (zh) 基站和终端
CN111435848B (zh) 指示和确定预编码向量的方法以及通信装置
WO2019109570A1 (zh) 信道测量方法和用户设备
WO2022160345A1 (zh) 信道重建方法、基站和终端
WO2019037694A1 (zh) 波束指示和上报方法、网络设备、终端
US20210126688A1 (en) Method and apparatus for channel estimation for ofdm based single carrier system
WO2022141079A1 (zh) 终端以及基站
WO2022141078A1 (zh) 通信系统中的终端以及基站
WO2023040618A1 (en) Method and apparatus for mimo csi feedback
WO2024065183A1 (zh) 通信方法、装置及存储介质
WO2024075097A1 (en) Efficiently transmitting a set of samples to another device

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: 21921944

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18258641

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 202180089526.X

Country of ref document: CN

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21921944

Country of ref document: EP

Kind code of ref document: A1