WO2020227941A1 - 上报信道状态信息的方法、装置及相关设备 - Google Patents

上报信道状态信息的方法、装置及相关设备 Download PDF

Info

Publication number
WO2020227941A1
WO2020227941A1 PCT/CN2019/086909 CN2019086909W WO2020227941A1 WO 2020227941 A1 WO2020227941 A1 WO 2020227941A1 CN 2019086909 W CN2019086909 W CN 2019086909W WO 2020227941 A1 WO2020227941 A1 WO 2020227941A1
Authority
WO
WIPO (PCT)
Prior art keywords
matrix
indication information
indicate
rows
dimensional
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2019/086909
Other languages
English (en)
French (fr)
Inventor
李雪茹
曲秉玉
刘鹍鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to PCT/CN2019/086909 priority Critical patent/WO2020227941A1/zh
Publication of WO2020227941A1 publication Critical patent/WO2020227941A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management

Definitions

  • This application relates to the field of communication technology, and in particular to methods, devices and related equipment for reporting channel state information.
  • MIMO technology is widely used in communication systems, such as Long Term Evolution (LTE) systems.
  • LTE Long Term Evolution
  • MIMO technology refers to the use of multiple transmitting antennas and receiving antennas at the transmitting end and the receiving end, so that signals are sent and received through multiple antennas at the transmitting end and the receiving end.
  • MIMO technology can improve communication quality and system channel capacity.
  • the MIMO system can use precoding (Precoding) technology to improve signal transmission quality and rate.
  • Precoding precoding
  • the network equipment determines the precoding matrix required for downlink data transmission according to the channel state information (CSI) fed back by the terminal equipment, and then The network device uses the precoding matrix to send downlink data to the terminal device.
  • CSI channel state information
  • the feedback CSI includes at least one parameter of the rank indicator (Rank Index, RI) of the channel matrix, the precoding matrix index (PMI), and the channel quality indicator (Channel Quality Index, CQI), wherein the PMI includes PMI1 and PMI2 , PMI1 is used to indicate the matrix W 1 , and PMI2 is used to indicate the matrix W 2 , and the product of W 1 and W 2 represents the precoding matrix W recommended by the terminal device.
  • RI rank indicator
  • PMI precoding matrix index
  • CQI Channel Quality Index
  • Non-linear precoding methods such as Dirty Paper Coding (DPC) and Iterative Precoding (Iterative Precoding) are superior to linear precoding in terms of improving communication quality.
  • DPC Dirty Paper Coding
  • Iterative Precoding Iterative Precoding
  • network equipment needs to obtain its connection with the terminal Channel matrix H for device communication.
  • the prediction accuracy of the channel matrix H is better than the prediction accuracy of its eigenvector V.
  • the network device In order to achieve high-precision channel information prediction on the network device side, the network device also needs to obtain the channel matrix H for communication with the terminal device. .
  • the network equipment can only obtain the precoding matrix W recommended by the terminal equipment, which generally does not directly reflect the channel matrix H itself, so the network equipment cannot be based on the PMI To obtain the channel matrix H.
  • the existing technical solutions for terminal equipment to feed back CSI to network equipment cannot support high-performance nonlinear precoding such as dirty paper coding, iterative precoding, and channel information prediction based on the channel matrix H.
  • the embodiments of the present application provide a method, an apparatus, and related equipment for reporting channel state information, which can reduce the overhead of the first communication device to feed back CSI, and the second communication device can obtain the channel matrix H (r) based on the CSI sent by the first communication device, To support non-linear precoding and channel information prediction.
  • a method for reporting CSI including: a second communication device sends configuration information to a first communication device, the configuration information is used to configure the CSI reported by the first communication device, and then the first communication device reports the CSI.
  • the configuration information is used to indicate the configuration parameters of the CSI.
  • the configuration parameters include one or more of the following information: CSI report parameters, frequency domain granularity of the reported parameters, codebook type used, and reference on which CSI measurement is based Signal information, CSI report bandwidth information, etc.
  • the frequency domain granularity of the reported parameters can be broadband reporting or subband reporting or other forms.
  • the codebook type used may include the first type Type I codebook or the second type Type II codebook.
  • the foregoing configuration information may be sent by the second communication device to the first communication device through high-level signaling, and the high-level signaling carries the configuration information.
  • the high-level signaling may be radio resource control (Radio Resource Control, RRC) signaling.
  • RRC Radio Resource Control
  • the CSI is used to indicate the matrix H (r)
  • the matrix H (r) is used to indicate the channel of the rth antenna port of the first communication device
  • the matrix H (r) is a matrix with N t rows and N sb columns.
  • N sb is the number of frequency domain units included in the report bandwidth of the CSI
  • N t is a positive integer
  • N r is the number of antenna ports of the first communication device.
  • the frequency domain unit may be a sub-carrier, a resource block (Resource Block, RB), a sub-band (Sub Band), etc., which are not limited in the embodiment of the present application.
  • CSI includes first indication information and second indication information.
  • the second indication information is used to indicate the matrix matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t .
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • the values can be the same; or, the N r antenna ports are grouped according to a certain principle, and the antenna ports in each group The values are the same.
  • the reporting overhead of the first indication information can be reduced.
  • the values can be the same; or, the N r antenna ports are grouped according to a certain principle, and the antenna ports in each group The values are the same.
  • the reporting overhead of the first indication information can be reduced.
  • the channel matrix H (r) containing N t N sb elements is compressed into a matrix containing 2LP elements
  • the number of elements that need to be fed back is reduced, thereby reducing the overhead of feeding back CSI.
  • the first communication device sends the CSI including the first indication information and the second indication information to the second communication device
  • the second communication device may determine the channel matrix H (r) based on the first indication information and the second indication information to Support non-linear precoding, and better channel information prediction.
  • L bases are arranged in the following order into a diagonal block matrix
  • the L bases may be derived from a set of predefined bases.
  • the predefined bases are Discrete Fourier Transformation (DFT) bases.
  • the first communication device can report To report the index of each base in
  • K bases are constructed as follows K bases can be derived from a set of predefined bases, optionally, the predefined bases are DFT bases.
  • the first communication device can report To report the index of each base in
  • the second indication information includes third indication information and fourth indication information, and the third indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of M 1 page with L rows and P columns, M 1 is a positive integer less than 2N r ;
  • the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column, or, the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column.
  • the fourth indication information is used to indicate the matrix When the value of l is different, Is the same matrix, through this scheme can reduce feedback s expenses.
  • the fourth indication information is used to indicate the matrix In the case of different values of k, Is the same matrix, through this scheme can reduce feedback s expenses. at this time, or Can be represented by W Rx,1 .
  • the matrix or The column vector in is the compressed base
  • 2N r is the dimension of the compressed base
  • M 1 is the number of compressed bases.
  • the fourth indication information indication matrix or Elements in the matrix or The compressed base in is not predefined, but calculated by the first communication device and reported to the second communication device. Compared with pre-defined base, non-pre-defined base or It can match the channel to be compressed more flexibly.
  • the non-predefined base can be better matched.
  • the non-predefined base can be determined according to the feature vector, which corresponds to the multi-antenna port r The feature vector of the determined feature space. The feature vector can better match the corresponding multiple antenna ports r of the first communication device Therefore, the compression effect is better, that is, the technical solution can be accurately represented by less compression base
  • the DFT predefined base is only applicable to the channel matrix compression of multiple antenna ports with the same polarization direction.
  • the antenna port of the first communication device or the second communication device has multiple polarization directions, it needs to be based on different The polarization direction splits the channel matrix and compresses them with the DFT predefined base.
  • a non-predefined base is more flexible and can be calculated according to the correlation between antenna port channel matrices of different polarization directions to support joint compression of antenna port channel matrices of different polarization directions and improve the compression effect.
  • the first communication device may construct W Rx,1 as follows: the first communication device receives the reference signal and performs measurement to obtain with with Both are matrices of size L ⁇ P (that is, L rows and P columns), with Is a coefficient matrix determined according to the channels between the two polarization direction antenna ports of the second communication device and the rth antenna port of the first communication device.
  • the first communication device divides 2N r two-dimensional matrices Spliced into a three-dimensional matrix of L ⁇ P ⁇ 2N r (that is, L rows, P columns and 2N r pages) And the three-dimensional matrix Expand into a two-dimensional matrix with a scale of 2N r ⁇ PL (that is, 2N r rows and PL columns) First communication device order Complex matrix of this size, and Covariance matrix Perform eigen decomposition, and the eigenvectors corresponding to the largest M 1 eigenvalues form a compressed matrix
  • the first communication device may also quantize the eigenvectors corresponding to the largest M 1 eigenvalues to form a compression matrix
  • the three-dimensional matrix generate or Or, W Rx,1 can be generated in other ways, which is not limited in the embodiment of the present application.
  • the first communication device may use the first correlation and the second correlation to generate or
  • the first correlation is the correlation between the channels of the multiple antenna ports of the first communication device
  • the second correlation is the correlation between the channels of the antenna ports of the two polarization directions of the second communication device.
  • 2N r matrices To three-dimensional matrix
  • the mapping between the pages can be given by the function g(r,i), namely And for any (r′,i′) ⁇ (r′′,i′′), g(r′,i′) ⁇ g(r′′,i′′);
  • Means taking matrix Fix the third two-dimensional matrix with L rows and P columns with an index of g(r,i), and the elements in the lth row and kth column of the L rows and P columns of the two-dimensional matrix are
  • the channel matrix The relationship with the third instruction information and the fourth instruction information is: here It is the lossy channel information after compression.
  • Representation matrix After two front fixing index one-dimensional vector of length M 1 of the one-dimensional vector length M 1 of the m-th element is Representation matrix
  • Two front fixing the index length of a 2N r-dimensional vector, and the one-dimensional vector of length 2N r g (r, i) th element ⁇ is a one-dimensional vector As a matrix multiplication of column vectors.
  • the matrix Compressed as a complex matrix as a whole a matrix containing 2L ⁇ P ⁇ N r elements Compressed into a matrix containing only 1 element of L ⁇ P ⁇ M
  • a matrix containing 2L ⁇ P ⁇ N r elements Compressed into a matrix containing only 1 element of L ⁇ P ⁇ M
  • the matrix Magnitude matrix Said The elements in row l and column k are based on the The value determined by the amplitude of the element in the lth row and kth column, for example, the The element in row l and column k is the The magnitude or square of the magnitude of the element in the lth row and kth column;
  • Means taking matrix Fix the first two one-dimensional vectors with index length M 1 , and the m-th element in the one-dimensional vector with length M 1 is Means taking matrix Fix the first two one-dimensional vectors with an index length of 2N r , and the g(r, i)th element of the one-dimensional vector with length 2N r is ⁇ is a one-dimensional vector As a matrix multiplication of column vectors.
  • the amplitude matrix containing 2L ⁇ P ⁇ N r elements Compressed into a matrix containing only 1 element of L ⁇ P ⁇ M
  • the amplitude is affected by the spatial transmission distance of the channel, the scattering environment, etc. Since the distances between the multiple antenna ports of the first communication device and the multiple antenna ports of the second communication device are relatively close, the amplitude matrix With strong correlation, the present invention can greatly reduce the CSI reporting overhead without causing significant performance loss.
  • the second indication information further includes fifth indication information, and the fifth indication information is used to indicate the The phase of the element.
  • the matrix when the scattering environment between the first communication device and the second communication device is more complicated, such as the non-line of sight (NLoS) scattering environment, the matrix The phase correlation is weak, right The phase of the elements in the middle is not compressed, that is, the matrix is directly reported The phase of the middle element can ensure the feedback accuracy.
  • NoS non-line of sight
  • the fifth indication information is used to indicate the The phase of elements in the middle, including:
  • the fifth indication information includes sixth indication information and seventh indication information, and the sixth indication information is used to indicate a three-dimensional matrix Is a three-dimensional matrix with L rows and P columns and M 2 pages, where M 2 is a positive integer less than 2N r ;
  • the seventh indication information is used to indicate the matrix Is a matrix of 2N r rows M 2 columns, or the seventh indication information is used to indicate the matrix Is a matrix of 2N r rows and M 2 columns;
  • the matrix Phase matrix Said The elements in row l and column k are based on the The value determined by the phase of the element in the lth row and kth column, for example, the The element in row l and column k is the The phase or the square of the phase of the element in the lth row and kth column;
  • the phase matrix containing 2L ⁇ P ⁇ N r elements Compressed into a matrix containing only L ⁇ P ⁇ M 2 elements
  • the matrix The phase correlation is strong, and the matrix Phase matrix Compressed reporting can greatly reduce the overhead of CSI reporting without causing significant performance loss.
  • the matrix Each column vector of is the basis for compressing matrix H (r′)
  • the matrix Each column vector of is the basis for compression of matrix H (r′′)
  • the constraints make the matrix And matrix The same column as possible corresponds to the same base.
  • the above for the matrix And matrix The constraint can improve the vector Correlation between, thereby enhancing the compression effect.
  • the third indication information includes instructions for indicating Middle element X 1 l,k,g bits of X 1 l,k,g are positive integers, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Y 1 l,c,d bits of Y 1 l,c,d , the Y 1 l,c,d are positive integers, or if the fourth indication information is used to indicate Then the fourth indication information includes an indication matrix Element in row c and column d in Z 1 k, c, d bits, Z 1 k, c, d are positive integers.
  • Y 1 l,c,d is greater than or equal to X 1 l,k,g .
  • Z 1 k,c,d is greater than or equal to X 1 l,k,g .
  • the quantization accuracy of the elements of the compressed base is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 1 , and is beneficial to reduce the overhead of reporting CSI.
  • the sixth indication information includes instructions for indicating Middle element X 2 l,k,f bits of X 2 l,k,f are positive integers, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Y 2 l,c,d bits of Y 2 l,c,d , Y 2 l,c,d are positive integers, or, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Z 2 k,c,d bits, Z 2 k,c,d are positive integers.
  • the seventh indication information is used to indicate a matrix Then Y 2 l,c,d is greater than or equal to X 2 l,k,f . Optionally, if the seventh indication information is used to indicate a matrix Then Z 2 k,c,d is greater than or equal to X 2 l,k,f .
  • the quantization accuracy of the elements of the compressed substrate is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 2 , and is beneficial to reduce the cost of reporting CSI.
  • the first communication device responds to the matrix Magnitude matrix And phase matrix Compressed separately, at this time, used for the amplitude matrix Compressed matrix or And for the phase matrix Compressed matrix or It can be different and is calculated independently.
  • Amplitude matrix The compression rate is determined by M 1
  • the compression rate is determined by M 2
  • the matrix can be realized by setting different M 1 and M 2 Different compression ratios of amplitude and phase, preferably, M 1 is smaller than M 2 .
  • the reporting of CSI by the first communication device includes:
  • the first communication device reports the third indication information and the fourth indication information at the same time; or
  • the first communication device reports the third indication information and the fourth indication information at different times.
  • the first communication device when the channel time domain changes rapidly, the first communication device separately calculates the compression matrix for the channel H (r) at each moment or Compression matrix calculated by the first communication device or It is used exclusively for the channel H (r) at this moment. In this case, the first communication device simultaneously reports the third and fourth indication information; when the channel time domain changes slowly, the compression calculated by the first communication device matrix or Shared by channel H (r) at multiple times to reduce the compression matrix or In this case, the first communication device reports the third indication information and the fourth indication information at different times.
  • the reporting of CSI by the first communication device includes:
  • the first communication device reports the sixth indication information and the seventh indication information at the same time; or
  • the first communication device reports the sixth indication information and the seventh indication information at different times.
  • the first communication device when the channel time domain changes rapidly, the first communication device separately calculates the compression matrix for the channel H (r) at each moment or Compression matrix calculated by the first communication device or It is used exclusively for the channel H (r) at this moment. In this case, the first communication device reports the sixth indication information and the seventh indication information at the same time; when the channel time domain changes slowly, the compression calculated by the first communication device matrix or Shared by channel H (r) at multiple times to reduce the compression matrix or In this case, the first communication device reports the sixth indication information and the seventh indication information at different times.
  • the reporting of the third indication information is an aperiodic report
  • the reporting of the fourth indication information is a periodic report.
  • the first communication device may periodically report the compression matrix or The size of the cycle determines that the second communication device obtains the compression matrix or Real-time, the first communication device can report the three-dimensional matrix acyclically The reporting may be triggered by the first communication device or the second communication device.
  • the report of the sixth indication information is an aperiodic report
  • the report of the seventh indication information is a periodic report.
  • the first communication device may periodically report the compression matrix or The size of the cycle determines that the second communication device obtains the compression matrix or Real-time, the first communication device can report the three-dimensional matrix acyclically The reporting may be triggered by the first communication device or the second communication device.
  • the M 1 may be configured, or the network device through the downlink control information or a RRC control signaling medium access control layer (MediaAccess ControlControl Element, MACCE) ( Downlink Control Information, DCI) letter Let the value of M 1 be configured.
  • MACCE MediaAccess Control Control Element, MACCE
  • the M 2 is configurable, and the network device configures the value of M 2 through RRC or MACCE or DCI signaling.
  • an embodiment of the present application provides a communication device, which includes a unit for implementing the method for reporting CSI in the first aspect.
  • an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a computer program or instruction, and when the program or instruction is executed by a processor, the processor executes The method for reporting CSI as described in the first aspect.
  • an embodiment of the present application provides a terminal device, including a receiver and a transmitter, the receiver is coupled with the transmitter, and is characterized in that:
  • the receiver is configured to receive configuration information from a second communication device, where the configuration information is used to configure the CSI reported by the transmitter;
  • the transmitter is used to report CSI
  • the CSI is used to indicate a matrix H (r)
  • the matrix H (r) is used to indicate the channel of the rth antenna port of the terminal device
  • the matrix H (r ) Is a matrix of N t rows and N sb columns
  • N sb is the number of frequency domain units included in the reported bandwidth of the CSI
  • N t is a positive integer
  • N r is The number of antenna ports of the terminal device
  • the CSI includes first indication information and second indication information;
  • the second indication information is used to indicate the matrix
  • the matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t ;
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • an embodiment of the present application provides a network device, including a receiver and a transmitter, the receiver is coupled with the transmitter, and is characterized in that:
  • the transmitter is configured to send configuration information to a first communication device, where the configuration information is used to configure the CSI reported by the first communication device;
  • the receiver is configured to receive CSI, the CSI is used to indicate a matrix H (r) , the matrix H (r) is used to indicate the channel of the rth antenna port of the first communication device, and the matrix H (r) is a matrix with N t rows and N sb columns, N sb is the number of frequency domain units included in the reported bandwidth of the CSI, N t is a positive integer, r ⁇ 1,2,...,N r ⁇ , N r is the number of antenna ports of the first communication device, and the CSI includes first indication information and second indication information;
  • the second indication information is used to indicate the matrix
  • the matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t ;
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • an embodiment of the present application provides a chip system, wherein the chip system includes a processor and an interface circuit, and the interface circuit is coupled with the processor,
  • the processor is configured to execute computer programs or instructions to implement the method according to the first aspect
  • the interface circuit is used to communicate with other modules outside the chip system.
  • FIG. 1 is a schematic structural diagram of a communication system provided by an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of a method for reporting CSI according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a matrix compression provided by an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of another communication system provided by an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a terminal device according to an embodiment of the present invention.
  • Fig. 6 is a schematic structural diagram of a network device provided by an embodiment of the present invention.
  • FIG. 1 is a schematic structural diagram of a communication system disclosed in an embodiment of the present application.
  • the communication system may include at least one first communication device and at least one second communication device, where the first communication device and the second communication device The communication device establishes a communication connection.
  • linear pre-coding and channel information prediction based on the channel matrix H in particular: a terminal device fed back CSI comprises RI, PMI and CQI channel matrix, wherein the PMI includes PMI1 and the PMI2, PMI1 for all elements indicating matrix W 1 of, PMI2 is used to indicate all the elements in the matrix W 2, W 1 and W 2 product characterization eigenvector of the channel matrix H V.
  • the first communication device receives configuration information from the second communication device, and the configuration information is used to configure the CSI reported by the first communication device. Then, the first communication device may report the CSI, and the CSI includes the first indication information and the second indication information.
  • CSI is used to indicate the matrix H (r)
  • the matrix H (r) is used to indicate the channel of the rth antenna port of the first communication device, r ⁇ ⁇ 1,2,...,N r ⁇ , N r is the first The number of antenna ports of the communication device.
  • the matrix H (r) is a matrix with N t rows and N sb columns.
  • the matrix H (r) may be a complex matrix with N t rows and N sb columns.
  • N sb is the number of frequency domain units included in the CSI report bandwidth.
  • N t is the number of antenna ports of the second communication device.
  • the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, Is a matrix with N sb rows and K columns, L is a positive integer, 2L is less than N t , K is a positive integer, and K is less than N sb , then the second indication information is used to indicate the matrix The matrix Is a matrix of 2L rows and K columns,
  • L bases are arranged in the following order into a diagonal block matrix,
  • the L bases can be derived from a set of predefined bases.
  • the first communication device can report To report the index of each base in
  • K bases are constructed as follows
  • the K bases can be derived from a set of predefined bases.
  • the first communication device can report To report the index of each base in
  • the channel matrix H (r) containing N t N sb elements is compressed into a matrix containing 2LK elements
  • the number of elements that need to be fed back is reduced, thereby reducing the overhead of feeding back CSI.
  • the first communication device sends the CSI including the first indication information and the second indication information to the second communication device
  • the second communication device may determine the channel matrix H (r) based on the first indication information and the second indication information to Support non-linear precoding, and better channel information prediction.
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, L is a positive integer, 2L is less than N t , then the second indication information is used to indicate the matrix matrix Is a matrix with 2L rows and N sb columns,
  • the channel matrix H (r) containing N t N sb elements is compressed into a matrix containing 2LN sb elements
  • the number of elements that need to be fed back is reduced, thereby reducing the overhead of feeding back CSI.
  • the first communication device sends the CSI including the first indication information and the second indication information to the second communication device
  • the second communication device may determine the channel matrix H (r) based on the first indication information and the second indication information to Support non-linear precoding, and better channel information prediction.
  • GSM Global System of Mobile Communication
  • CDMA Code Division Multiple Access
  • WCDMA Wideband Code Division Multiple Access
  • GSM Global System of Mobile Communication
  • GPRS General Packet Radio Service
  • LTE Long Term Evolution
  • FDD Frequency Division Duplex
  • TDD Time Division Duplex
  • UMTS Universal Mobile Telecommunication System
  • WiMAX Worldwide Interoperability for Microwave Access
  • 5G fifth-generation mobile communication technology
  • 4G New Radio (NR) communication system 4G New Radio (NR) communication system
  • NR URLLC
  • NB-IoT or eMTC system etc.
  • the second communication device may be a network device, and the network device may specifically be an access point (AP) in a wireless local area network (Wireless Local Area Networks, WLAN), a global system for mobile communication (GSM) or Base Station Transceiver Station (BTS) in Code Division Multiple Access (Code Division Multiple Access, CDMA), Base Station (NodeB, NB) in Wideband Code Division Multiple Access (WCDMA), LTE
  • AP access point
  • WLAN wireless Local Area Networks
  • GSM Global System for mobile communication
  • BTS Base Station Transceiver Station
  • CDMA Code Division Multiple Access
  • NodeB, NB Base Station
  • WCDMA Wideband Code Division Multiple Access
  • LTE LTE
  • PLMN Public Land Mobile Network
  • the first communication device may be a terminal device, which may also be called User Equipment (UE), mobile station, access terminal, user unit, user station, mobile station, remote station, remote terminal, mobile equipment, terminal , Wireless communication equipment, user agent or user device, etc., which can specifically be Station (ST), cellular phone, cordless phone, Session Initiation Protocol (SIP) phone, wireless local loop (Wireless local loop) in WLAN Local Loop (WLL) station, Personal Digital Assistant (PDA), handheld devices with wireless communication functions, computing devices, other processing devices connected to wireless modems, in-vehicle devices, wearable devices, and future 4G networks Any of mobile stations and terminal equipment in the future evolved PLMN network.
  • UE User Equipment
  • ST Station
  • IP Session Initiation Protocol
  • WLL WLAN Local Loop
  • PDA Personal Digital Assistant
  • FIG. 2 is a schematic flowchart of a method for reporting CSI according to an embodiment of the present invention. Specifically, as shown in FIG. 2, the method for reporting CSI in the embodiment of the present invention may include the following steps:
  • the second communication device sends configuration information to the first communication device, where the configuration information is used to configure the CSI reported by the first communication device.
  • the first communication device acquires CSI.
  • the CSI includes first indication information and second indication information, and the first indication information is used to indicate the matrix And matrix
  • the second indication information is used to indicate the matrix
  • the first communication device can transfer the matrix Compress as a complex matrix as a whole, or compress the matrix To compress the amplitude matrix, or the matrix To compress the phase matrix of The magnitude matrix and the matrix To compress the phase matrix, or report the matrix directly The phase matrix of The phase matrix is compressed.
  • the second indication information may include third indication information and fourth indication information.
  • the third indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ K ⁇ M 1 (that is, L rows, K columns, M 1 pages), and M 1 is a positive integer.
  • the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column.
  • the second indication information may include third indication information and fourth indication information.
  • the third indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ K ⁇ M 1 (that is, L rows, K columns, M 1 pages), and M 1 is a positive integer.
  • the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column.
  • the three-dimensional matrix Can also be called a three-dimensional tensor
  • the fourth indication information is used to indicate the matrix When the value of l is different, Is the same matrix, through this scheme can reduce feedback s expenses. Or, the fourth indication information is used to indicate the matrix In the case of different values of k, Is the same matrix, through this scheme can reduce feedback s expenses. at this time, or Can be represented by W Rx,1 .
  • the first communication device may construct W Rx,1 as follows: the first communication device receives the reference signal and performs measurement to obtain with with All are matrices of size L ⁇ K (that is, L rows and K columns), with Is a coefficient matrix determined according to the channels between the two polarization direction antenna ports of the second communication device and the rth antenna port of the first communication device.
  • the first communication device divides 2N r two-dimensional matrices Concatenated into a three-dimensional matrix of L ⁇ K ⁇ 2N r (that is, L rows and K columns and 2N r pages) And the three-dimensional matrix Expand to a two-dimensional matrix with a scale of 2N r ⁇ KL (that is, 2N r rows and KL columns)
  • First communication device order Complex matrix of this size, and Covariance matrix Perform eigen decomposition, and the eigenvectors corresponding to the largest M 1 eigenvalues form a compressed matrix
  • the first communication device may also quantize the eigenvectors corresponding to the largest M 1 eigenvalues to form a compression matrix
  • the first communication device can perform matrix transformation. Taking Figure 3 as an example, the first communication device can transform Split into 2N r matrices with a size of L ⁇ K (that is, L rows and K columns) are formed. The first communication device Put on Later, a newly constructed three-dimensional matrix is obtained, and the newly constructed three-dimensional matrix contains KL vectors with a length of 2N r . The first communication device arranges these KL vectors with a length of 2N r in any order to form a two-dimensional matrix with a scale of 2N r rows and KL columns
  • the first communication device may construct a compression matrix, for example, the first communication device may make Complex matrix of this scale, right Covariance matrix Perform feature decomposition to get Among them, D is a matrix with a scale of 2N r ⁇ 2N r (that is, 2N r rows and 2N r columns), and diagonal elements are eigenvalues arranged in descending order. V is a matrix with a scale of 2N r ⁇ 2N r (that is, 2N r rows and 2N r columns), and each column is an eigenvector, which corresponds to the eigenvalue in D.
  • the third indication information and the fourth indication information indicate a complex matrix by get.
  • the channel sharing compression matrix W Rx1 at T s times in order to further reduce the overhead of the feedback compression matrix W Rx,1 , the channel sharing compression matrix W Rx1 at T s times, where
  • the compressed matrix W Rx1 and the combined coefficient matrix There are two ways to report:
  • Compression matrix W Rx,1 and merge coefficient matrix It can be reported at different times. For example, compressed matrix W Rx, 1 cycle report, combined coefficient matrix Aperiodic reporting, compressed matrix W Rx,1 can be combined by multiple coefficient matrices Shared.
  • the matrix In the case of compression of the amplitude matrix, the third indication information and the fourth indication information indicate the matrix Magnitude matrix
  • the second indication information includes fifth indication information, and the fifth indication information is used to indicate the matrix The phase of the element.
  • the fifth indication information may include sixth indication information and seventh indication information.
  • the sixth indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ K ⁇ M 2 (that is, L rows, K columns, M 2 pages), and M 2 is a positive integer.
  • the seventh indication information is used to indicate the matrix It is a matrix of 2N r rows and M 2 columns.
  • the fifth indication information may include sixth indication information and seventh indication information.
  • the sixth indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ K ⁇ M 2 , and M 2 is a positive integer.
  • the seventh indication information is used to indicate the matrix It is a matrix of 2N r rows and M 2 columns.
  • phase matrix The description of and the complex matrix The description is similar to that of the complex number matrix The description of the embodiment of this application will not be repeated.
  • the three-dimensional matrix Can also be called a three-dimensional tensor
  • the matrix And three-dimensional matrix There are two ways to report:
  • matrix And three-dimensional matrix It can be reported at different times.
  • the matrix Periodic reporting, three-dimensional matrix Aperiodic reporting, matrix Can be multiple 3D matrices Shared.
  • the matrix And three-dimensional matrix There are two ways to report:
  • matrix And three-dimensional matrix It can be reported at different times.
  • the matrix Periodic reporting, three-dimensional matrix Aperiodic reporting, matrix Can be multiple 3D matrices Shared.
  • the second indication information includes third indication information, fourth indication information, and fifth indication information.
  • the third indication information and the fourth indication information in the embodiment of this application can refer to the amplitude matrix
  • the fifth indication information in the embodiment of this application please refer to the comparison of the phase matrix The specific description in the case of compression is not repeated in the embodiment of the present application.
  • the magnitude matrix The compression rate is determined by M 1
  • the phase matrix The compression rate is determined by M 2
  • the matrix can be realized by setting different M 1 and M 2 Different compression ratios of amplitude and phase, preferably, M 1 is smaller than M 2 .
  • the matrix Each column vector of is the basis for compressing matrix H (r′)
  • the matrix Each column vector of is the basis for compression of matrix H (r′′)
  • the constraints make the matrix And matrix The same column as possible corresponds to the same base.
  • the constraint can improve the vector Correlation between, thereby enhancing the compression effect.
  • the third indication information includes instructions for indicating Middle element X 1 l,k,g bits of X 1 l,k,g are positive integers, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Y 1 l,c,d bits of Y 1 l,c,d , Y 1 l,c,d are positive integers, or, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Z 1 k, c, d bits, Z 1 k, c, d are positive integers.
  • the fourth indication information is used to indicate a matrix
  • the Y 1 l,c,d is greater than or equal to X 1 l,k,g .
  • the fourth indication information is used to indicate a matrix Z 1 k,c,d is greater than or equal to X 1 l,k,g .
  • the quantization accuracy of the elements of the compressed base is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 1 , and is beneficial to reduce the overhead of reporting CSI.
  • the sixth indication information includes instructions for indicating Middle element X 2 l,k,f bits of X 2 l,k,f are positive integers, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Y 2 l,c,d bits of Y 2 l,c,d , Y 2 l,c,d are positive integers, or, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Z 2 k,c,d bits, Z 2 k,c,d are positive integers.
  • the seventh indication information is used to indicate a matrix
  • the Y 2 l, c, d is greater than or equal to X 2 l, k, f .
  • the seventh indication information is used to indicate a matrix Z 2 k,c,d is greater than or equal to X 2 l,k,f .
  • the quantization accuracy of the elements of the compressed substrate is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 2 , and is beneficial to reduce the cost of reporting CSI.
  • the first communication device reports CSI.
  • the amount of feedback is among them Is the number of bits for each parameter to be fed back.
  • the third instruction information and the fourth instruction information indicate the complex matrix
  • the compressed matrix W Rx,1 is shared at T s times, and the feedback amount includes two parts: the feedback amount of the fourth indication information And the feedback amount of the third instruction information among them with Is the number of bits for each parameter to be fed back.
  • Compression rate refers to the amount of feedback of the third and fourth indication information and the direct feedback matrix The ratio between the amount of feedback.
  • the channel matrix H (r) containing N t N sb elements is compressed into a matrix containing 2LK elements
  • the number of elements that need to be fed back is reduced, thereby reducing the overhead of feeding back CSI.
  • Second, by adding the matrix Compressed as a complex matrix as a whole, a matrix containing 2L ⁇ K ⁇ N r elements Compressed into a matrix containing only 1 element of L ⁇ K ⁇ M Significantly reduce the overhead of CSI reporting without significant performance loss, or by Magnitude matrix
  • the amplitude matrix containing 2L ⁇ K ⁇ N r elements Compressed into a matrix containing only 1 element of L ⁇ K ⁇ M It can greatly reduce the overhead of CSI reporting without causing significant performance loss, or by comparing the matrix Phase matrix For compression, the phase matrix containing 2L ⁇ K ⁇ N r elements Compressed into a matrix containing only L ⁇ K ⁇ M 2 elements It can greatly reduce the CSI reporting overhead without causing significant performance loss.
  • the first communication device sends the CSI including the first indication information and the second indication information to the second communication device
  • the second communication device may determine the channel matrix H (r) based on the first indication information and the second indication information to Support non-linear precoding, and better channel information prediction.
  • an embodiment of the present invention provides another method for reporting CSI.
  • the method for reporting CSI in the embodiment of the present invention may include the following steps:
  • the second communication device sends configuration information to the first communication device, where the configuration information is used to configure the CSI reported by the first communication device;
  • the first communication device reports CSI.
  • the CSI includes first indication information and second indication information.
  • the first indication information is used to indicate the matrix
  • the second indication information is used to indicate the matrix
  • the first communication device can transfer the matrix Compress as a complex matrix as a whole, or compress the matrix To compress the amplitude matrix, or the matrix To compress the phase matrix of The magnitude matrix and the matrix To compress the phase matrix, or report the matrix directly The phase matrix of The phase matrix is compressed.
  • the second indication information may include third indication information and fourth indication information.
  • the third indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ N sb ⁇ M 1 , and M 1 is a positive integer.
  • the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column.
  • the second indication information may include third indication information and fourth indication information.
  • the third indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ N sb ⁇ M 1 , and M 1 is a positive integer.
  • the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column.
  • the three-dimensional matrix Can also be called a three-dimensional tensor
  • the fourth indication information is used to indicate the matrix When the value of l is different, Is the same matrix, through this scheme can reduce feedback s expenses. Or, the fourth indication information is used to indicate the matrix In the case of different values of k, Is the same matrix, through this scheme can reduce feedback s expenses. at this time, or Can be represented by W Rx,1 .
  • the first communication device may construct W Rx,1 as follows: the first communication device receives the reference signal and performs measurement to obtain with with Both are matrices of size L ⁇ N sb (that is, L rows and N sb columns), with Is a coefficient matrix determined according to the channels between the two polarization direction antenna ports of the second communication device and the rth antenna port of the first communication device.
  • the first communication device divides 2N r two-dimensional matrices Spliced into a three-dimensional matrix of L ⁇ N sb ⁇ 2N r (that is, L rows, N sb columns and 2N r pages) And the three-dimensional matrix Expand into a two-dimensional matrix with a scale of 2N r ⁇ N sb L (that is, 2N r rows and N sb L columns)
  • First communication device order Complex matrix of this size, and Covariance matrix Perform eigen decomposition, and the eigenvectors corresponding to the largest M 1 eigenvalues form a compressed matrix
  • the first communication device may also quantize the eigenvectors corresponding to the largest M 1 eigenvalues to form a compression matrix
  • the first communication device can perform matrix transformation. For example, the first communication device can transform Split into 2N r matrices with a size of L ⁇ N sb (that is, L rows and N sb columns) are formed. The first communication device Put on Later, a newly constructed three-dimensional matrix is obtained, and the newly constructed three-dimensional matrix contains N sb L vectors with a length of 2N r . The first communication device arranges these N sb L vectors with a length of 2N r in any order to form a two-dimensional matrix with a scale of 2N r rows and N sb L columns
  • the first communication device may construct a compression matrix, for example, the first communication device may make Complex matrix of this scale, right Covariance matrix Perform feature decomposition to get Among them, D is a matrix with a scale of 2N r ⁇ 2N r (that is, 2N r rows and 2N r columns), and diagonal elements are eigenvalues arranged in descending order. V is a matrix with a scale of 2N r ⁇ 2N r (that is, 2N r rows and 2N r columns), and each column is an eigenvector, which corresponds to the eigenvalue in D.
  • the third indication information and the fourth indication information indicate a complex matrix by get.
  • the channel sharing compression matrix W Rx,1 at T s times in order to further reduce the overhead of the feedback compression matrix W Rx,1 , the channel sharing compression matrix W Rx,1 at T s times, where
  • the compressed matrix W Rx1 and the combined coefficient matrix There are two ways to report:
  • Compression matrix W Rx,1 and merge coefficient matrix It can be reported at different times. For example, compressed matrix W Rx, 1 cycle report, combined coefficient matrix Aperiodic reporting, compressed matrix W Rx,1 can be combined by multiple coefficient matrices Shared.
  • the matrix In the case of compression of the amplitude matrix, the third indication information and the fourth indication information indicate the matrix Magnitude matrix
  • the second indication information includes fifth indication information, and the fifth indication information is used to indicate the matrix The phase of the element.
  • the fifth indication information may include sixth indication information and seventh indication information.
  • the sixth indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ N sb ⁇ M 2 (ie, L rows, N sb columns, M 2 pages), and M 2 is a positive integer.
  • the seventh indication information is used to indicate the matrix It is a matrix of 2N r rows and M 2 columns.
  • the fifth indication information may include sixth indication information and seventh indication information.
  • the sixth indication information is used to indicate the three-dimensional matrix It is a three-dimensional matrix of L ⁇ N sb ⁇ M 2 , and M 2 is a positive integer.
  • the seventh indication information is used to indicate the matrix It is a matrix of 2N r rows and M 2 columns.
  • the three-dimensional matrix Can also be called a three-dimensional tensor
  • the matrix And three-dimensional matrix There are two ways to report:
  • matrix And three-dimensional matrix It can be reported at different times.
  • the matrix Periodic reporting, three-dimensional matrix Aperiodic reporting, matrix Can be multiple 3D matrices Shared.
  • the matrix And three-dimensional matrix There are two ways to report:
  • matrix And three-dimensional matrix It can be reported at different times.
  • the matrix Periodic reporting, three-dimensional matrix Aperiodic reporting, matrix Can be multiple 3D matrices Shared.
  • phase matrix The description of and the complex matrix The description is similar to that of the complex number matrix The description of the embodiment of this application will not be repeated.
  • the second indication information includes third indication information, fourth indication information, and fifth indication information.
  • the third indication information and the fourth indication information in the embodiment of this application can refer to the amplitude matrix
  • the fifth indication information in the embodiment of this application please refer to the comparison of the phase matrix The specific description in the case of compression is not repeated in the embodiment of the present application.
  • the magnitude matrix The compression rate is determined by M 1
  • the phase matrix The compression rate is determined by M 2
  • the matrix can be realized by setting different M 1 and M 2 Different compression ratios of amplitude and phase, preferably, M 1 is smaller than M 2 .
  • the matrix Each column vector of is the basis for compressing matrix H (r′)
  • the matrix Each column vector of is the basis for compression of matrix H (r′′)
  • the above is for the matrix
  • the constraint can improve the vector Correlation between, thereby enhancing the compression effect.
  • the third indication information includes instructions for indicating Middle element X 1 l,k,g bits of X 1 l,k,g are positive integers, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Y 1 l,c,d bits of Y 1 l,c,d , Y 1 l,c,d are positive integers, or, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Z 1 k, c, d bits, Z 1 k, c, d are positive integers.
  • the fourth indication information is used to indicate a matrix
  • the Y 1 l,c,d is greater than or equal to X 1 l,k,g .
  • the fourth indication information is used to indicate a matrix Z 1 k,c,d is greater than or equal to X 1 l,k,g .
  • the quantization accuracy of the elements of the compressed base is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 1 , and is beneficial to reduce the overhead of reporting CSI.
  • the sixth indication information includes instructions for indicating Middle element X 2 l,k,f bits of X 2 l,k,f are positive integers, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Y 2 l,c,d bits of Y 2 l,c,d , Y 2 l,c,d are positive integers, or, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Z 2 k,c,d bits, Z 2 k,c,d are positive integers.
  • the seventh indication information is used to indicate a matrix
  • the Y 2 l, c, d is greater than or equal to X 2 l, k, f .
  • the seventh indication information is used to indicate a matrix Z 2 k,c,d is greater than or equal to X 2 l,k,f .
  • the quantization accuracy of the elements of the compressed substrate is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 2 , and is beneficial to reduce the cost of reporting CSI.
  • the amount of feedback is among them Is the number of bits for each parameter to be fed back.
  • the third instruction information and the fourth instruction information indicate the complex matrix
  • the compressed matrix W Rx,1 is shared at T s times, and the feedback amount includes two parts: the feedback amount of the fourth indication information And the feedback amount of the third instruction information among them with Is the number of bits for each parameter to be fed back.
  • the channel matrix H (r) containing N t N sb elements is compressed into a matrix containing 2LN sb elements
  • the number of elements that need to be fed back is reduced, thereby reducing the overhead of feeding back CSI.
  • a matrix containing 2L ⁇ N sb ⁇ N r elements Compressed into a matrix containing only L ⁇ N sb ⁇ M 1 element Significantly reduce the overhead of CSI reporting without significant performance loss, or by Magnitude matrix
  • the amplitude matrix containing 2L ⁇ N sb ⁇ N r elements Compressed into a matrix containing only L ⁇ N sb ⁇ M 1 element It can greatly reduce the overhead of CSI reporting without causing significant performance loss, or by comparing the matrix Phase matrix
  • the phase matrix containing 2L ⁇ N sb ⁇ N r elements Compressed into a matrix containing only L ⁇ N sb ⁇ M 2 elements It can greatly reduce the CSI reporting overhead without causing significant performance loss.
  • the first communication device sends the CSI including the first indication information and the second indication information to the second communication device
  • the second communication device may determine the channel matrix H (r) based on the first indication information and the second indication information to Support non-linear precoding, and better channel information prediction.
  • an embodiment of the present invention provides a method for reporting CSI.
  • the method for reporting CSI in the embodiment of the present invention may include the following steps:
  • the second communication device sends configuration information to the first communication device, where the configuration information is used to configure the CSI reported by the first communication device;
  • the first communication device reports CSI
  • the CSI is used to indicate a matrix H (r)
  • the matrix H (r) is used to indicate the channel of the rth antenna port of the first communication device
  • the matrix H ( r) is a matrix with N t rows and N sb columns
  • N sb is the number of frequency domain units included in the reported bandwidth of the CSI
  • N t is a positive integer, r ⁇ 1,2,...,N r ⁇
  • the CSI includes first indication information and second indication information
  • the second indication information is used to indicate the matrix
  • the matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t ;
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • the second indication information includes third indication information and fourth indication information, and the third indication information is used to indicate a three-dimensional matrix It is a three-dimensional matrix of M 1 page with L rows and P columns, M 1 is a positive integer;
  • the fourth indication information is used to indicate the matrix Is a matrix of 2N r rows M 1 column, or the fourth indication information is used to indicate the matrix It is a matrix of 2N r rows M 1 column.
  • the embodiments of this application can be Split into To form 2N r matrices with L rows and P columns, the function g(r,i) realizes these 2N r matrices to a three-dimensional matrix Mapping between pages.
  • Means taking matrix Fix the first two one-dimensional vectors with index length M 1 , and the m-th element in the one-dimensional vector with length M 1 is Means taking matrix Fix the first two one-dimensional vectors with an index length of 2N r , and the g(r, i)th element of the one-dimensional vector with length 2N r is ⁇ is a one-dimensional vector As a matrix multiplication of column vectors.
  • the matrix Magnitude matrix Said The elements in row l and column k are based on the The value determined by the amplitude of the element in the lth row and kth column, for example, the The element in row l and column k is the The magnitude or square of the magnitude of the element in the lth row and kth column;
  • the embodiments of this application can be Split into To form 2N r matrices with L rows and P columns, the function g(r,i) realizes these 2N r matrices to a three-dimensional matrix Mapping between pages.
  • Means taking matrix Fix the first two one-dimensional vectors with index length M 1 , and the m-th element in the one-dimensional vector with length M 1 is Means taking matrix Fix the first two one-dimensional vectors with an index length of 2N r , and the g(r, i)th element of the one-dimensional vector with length 2N r is ⁇ is a one-dimensional vector As a matrix multiplication of column vectors.
  • the second indication information further includes fifth indication information, and the fifth indication information is used to indicate the The phase of the element.
  • the fifth indication information is used to indicate the The phase of elements in the middle, including:
  • the fifth indication information includes sixth indication information and seventh indication information, and the sixth indication information is used to indicate a three-dimensional matrix It is a three-dimensional matrix with L rows and P columns and M 2 pages, where M 2 is a positive integer;
  • the seventh indication information is used to indicate the matrix Is a matrix of 2N r rows M 2 columns, or the seventh indication information is used to indicate the matrix Is a matrix of 2N r rows and M 2 columns;
  • the matrix Phase matrix Said The elements in row l and column k are based on the The value determined by the phase of the element in the lth row and kth column, such as the The element in row l and column k is the The phase or the square of the phase of the element in the lth row and kth column;
  • the embodiments of this application can be Split into To form 2N r matrices with L rows and P columns, the function f(r,i) realizes these 2N r matrices to a three-dimensional matrix Mapping between pages.
  • the matrix Each column vector of is the basis for compressing matrix H (r′)
  • the matrix Each column vector of is the basis for compression of matrix H (r′′)
  • the above is for the matrix
  • the constraint can improve the vector Correlation between, thereby enhancing the compression effect.
  • the matrix Each column vector of is the basis for compressing matrix H (r′)
  • the matrix Each column vector of is the basis for compression of matrix H (r′′)
  • the constraints make the matrix And matrix The same column as possible corresponds to the same base.
  • the above for the matrix And matrix The constraint can improve the vector Correlation between, thereby enhancing the compression effect.
  • the third indication information includes instructions for indicating Middle element X 1 l,k,g bits, X 1 l,k,g are positive integers, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Y 1 l,c,d bits of Y 1 l,c,d , Y 1 l,c,d are positive integers, or, if the fourth indication information is used to indicate the matrix Then the fourth indication information includes an indication matrix Element in row c and column d in Z 1 k, c, d bits, Z 1 k, c, d are positive integers.
  • the fourth indication information is used to indicate a matrix
  • the Y 1 l,c,d is greater than or equal to X 1 l,k,g .
  • the fourth indication information is used to indicate a matrix Z 1 k,c,d is greater than or equal to X 1 l,k,g .
  • the quantization accuracy of the elements of the compressed base is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 1 , and is beneficial to reduce the overhead of reporting CSI.
  • the sixth indication information includes instructions for indicating Middle element X 2 l,k,f bits of X 2 l,k,f are positive integers, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Y 2 l,c,d bits of Y 2 l,c,d , Y 2 l,c,d are positive integers, or, if the seventh indication information is used to indicate the matrix Then the seventh indication information includes an indication matrix Element in row c and column d in Z 2 k,c,d bits, Z 2 k,c,d are positive integers.
  • the seventh indication information is used to indicate a matrix
  • the Y 2 l, c, d is greater than or equal to X 2 l, k, f .
  • the seventh indication information is used to indicate a matrix Z 2 k,c,d is greater than or equal to X 2 l,k,f .
  • the quantization accuracy of the elements of the compressed substrate is not lower than the accuracy of the merging coefficients, which is beneficial to reduce the compression loss, that is, the compressed channel matrix more matches the real Channel matrix.
  • the base matrix is the same when the value of l is different, or when That is, when the base matrix is the same when the value of k is different, it is more necessary to improve the quantization accuracy of the compressed base element.
  • improving the quantization accuracy of compressed base elements is beneficial to reduce the number of selected compressed bases M 2 , and is beneficial to reduce the cost of reporting CSI.
  • Amplitude matrix The compression rate is determined by M 1
  • the compression rate is determined by M 2
  • the matrix can be realized by setting different M 1 and M 2 Different compression ratios of amplitude and phase, preferably, M 1 is smaller than M 2 .
  • the communication system includes a first communication device 401 and a second communication device 402.
  • the first communication device 401 may include a first receiving unit 4011 and a first sending unit.
  • Unit 4012 the second communication device 402 may include a second sending unit 4021 and a second receiving unit 4022.
  • the second sending unit 4021 is configured to send configuration information to the first receiving unit 4011, where the configuration information is used to configure the CSI reported by the first sending unit 4012;
  • the first sending unit 4012 is configured to report CSI to the second receiving unit 4022, where the CSI is used to indicate a matrix H (r) , and the matrix H (r) is used to indicate the rth antenna port of the first communication device
  • the matrix H (r) is a matrix with N t rows and N sb columns, N sb is the number of frequency domain units included in the reported bandwidth of the CSI, N t is a positive integer, r ⁇ ⁇ 1,2, ..., N r ⁇ , N r is the number of antenna ports of the first communication device, and the CSI includes first indication information and second indication information;
  • the second indication information is used to indicate the matrix
  • the matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t ;
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • FIG. 5 is a schematic structural diagram of a terminal device provided by an embodiment of the present invention.
  • the receiver 503 and the transmitter 504 can call the program code stored in the memory 502 to perform the following operations:
  • the receiver 503 receives configuration information from the second communication device, where the configuration information is used to configure the CSI reported by the transmitter;
  • the transmitter 504 reports CSI, where the CSI is used to indicate a matrix H (r) , the matrix H (r) is used to indicate the channel of the rth antenna port of the terminal device, and the matrix H (r) is N t A matrix with rows N sb and columns, N sb is the number of frequency domain units included in the reported bandwidth of the CSI, N t is a positive integer, r ⁇ ⁇ 1,2,...,N r ⁇ , N r is the first The number of antenna ports of the communication device, where the CSI includes first indication information and second indication information;
  • the second indication information is used to indicate the matrix
  • the matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t ;
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • FIG. 6 is a schematic structural diagram of a network device provided by an embodiment of the present invention.
  • the receiver 603 and the transmitter 604 can call the program code stored in the memory 602 to perform the following operations:
  • the transmitter 604 sends configuration information to the first communication device, where the configuration information is used to instruct the first communication device to report CSI;
  • the receiver 603 receives CSI from the first communication device, where the CSI is used to indicate a matrix H (r) , and the matrix H (r) is used to indicate the channel of the rth antenna port of the first communication device ,
  • the matrix H (r) is a matrix with N t rows and N sb columns, N sb is the number of frequency domain units included in the reported bandwidth of the CSI, N t is a positive integer, r ⁇ 1,2,..., N r ⁇ , N r is the number of antenna ports of the first communication device, and the CSI includes first indication information and second indication information;
  • the second indication information is used to indicate the matrix
  • the matrix It is a matrix with 2L rows and P columns, L is a positive integer, and 2L is less than N t ;
  • the first indication information is used to indicate the matrix Is a matrix with N t rows and 2L columns, and P is N sb , or the first indication information is used to indicate the matrix And matrix Is a matrix with N t rows and 2L columns, It is a matrix with N sb rows and K columns, K is a positive integer, K is less than N sb , and P is K.
  • the processor may be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (ASIC), a ready-made programmable gate array (Field Programmable Gate Array, FPGA) Or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.
  • DSP Digital Signal Processor
  • ASIC application specific integrated circuit
  • FPGA ready-made programmable gate array
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present invention may be directly embodied as being executed and completed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in Random Access Memory (RAM), Flash memory, Read-Only Memory (Read-Only Memory, ROM), programmable read-only memory, or electrically erasable programmable memory, registers, etc. mature in the field Storage medium.
  • RAM Random Access Memory
  • ROM Read-Only Memory
  • programmable read-only memory or electrically erasable programmable memory, registers, etc. mature in the field Storage medium.
  • the storage medium is located in the memory, and the processor reads the instructions in the memory and completes the steps of the above method in combination with its hardware.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or It can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of this application essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code .

Landscapes

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

Abstract

一种上报信道状态信息的方法、装置及相关设备。其中，该方法包括：第一通信装置接收来自于第二通信装置的配置信息，配置信息用于指示第一通信装置上报CSI；第一通信装置上报CSI，CSI包括第一指示信息和第二指示信息，第二指示信息用于指示矩阵第一指示信息用于指示矩阵或，第一指示信息用于指示矩阵和矩阵采用本申请，可降低第一通信装置反馈CSI的开销，且第二通信装置可以基于第一通信装置发送的CSI获取信道矩阵H (r)，以支持非线性预编码以及信道信息预测。

Description

上报信道状态信息的方法、装置及相关设备 技术领域
本申请涉及通信技术领域，尤其涉及上报信道状态信息的方法、装置及相关设备。
背景技术
目前，多输入多输出(Multiple Input and Multiple Output，MIMO)技术被广泛应用于通信系统中，例如长期演进(Long Term Evolution，LTE)系统。MIMO技术是指在发射端和接收端分别使用多个发射天线和接收天线，使得信号通过发射端和接收端的多个天线进行发送和接收，MIMO技术能够提高通信质量和系统信道容量。
MIMO系统可以采用预编码(Precoding)技术来提高信号传输质量和速率，具体来讲，网络设备根据终端设备反馈的信道状态信息(Channel State Information，CSI)确定下行数据传输需要的预编码矩阵，进而网络设备利用该预编码矩阵向终端设备进行下行数据发送。针对LTE标准版本14(Rel-14)和新无线接入技术(New RadioAccess Technology，NR)中定义的线性预编码方法，现有的一种终端设备向网络设备反馈CSI的技术方案中，终端设备反馈的CSI包括信道矩阵的秩指示(Rank Index，RI)、预编码矩阵指示(Precoding Matrix Index，PMI)和信道质量指示(Channel Quality Index，CQI)中的至少一个参数，其中PMI包括PMI1和PMI2，PMI1用于指示矩阵W 1，PMI2用于指示矩阵W 2，W 1和W 2的乘积表征终端设备推荐的预编码矩阵W。脏纸编码(Dirty Paper Coding，DPC)、迭代预编码(Iterative Precoding)等非线性预编码方法在提高通信质量方面优于线性预编码，为了支持上述非线性预编码，网络设备需要获取其与终端设备通信的信道矩阵H。另外，在信道信息预测方面，信道矩阵H的预测精度优于其特征向量V的预测精度，为了实现网络设备侧的高精度信道信息预测，网络设备同样需要获取其与终端设备通信的信道矩阵H。然而，在现有终端设备向网络设备反馈CSI的技术方案中，网络设备仅能获取终端设备推荐的预编码矩阵W，该预编码矩阵一般不直接反应信道矩阵H本身，因此网络设备无法根据PMI来获取信道矩阵H。
综上所述，现有终端设备向网络设备反馈CSI的技术方案无法支持脏纸编码、迭代预编码等高性能非线性预编码以及基于信道矩阵H的信道信息预测。
发明内容
本申请实施例提供上报信道状态信息的方法、装置及相关设备，可降低第一通信装置反馈CSI的开销，且第二通信装置可以基于第一通信装置发送的CSI获取信道矩阵H (r)，以支持非线性预编码以及信道信息预测。
第一方面，提供了一种上报CSI的方法，包括：第二通信装置向第一通信装置发送配置信息，配置信息用于配置第一通信装置上报的CSI，然后第一通信装置上报CSI。
其中，配置信息用于指示CSI的配置参数，例如，配置参数包括如下信息中的一个或多个：CSI的上报参数、上报参数的频域粒度、使用的码本类型、CSI测量所基于的参考信号信息、CSI的上报带宽的信息等。上报参数的频域粒度可以是宽带上报或者子带上报或 者其它形式。使用的码本类型可以包括第一类型Type I码本或第二类型Type II码本。上述配置信息可以是第二通信装置通过高层信令发送给第一通信装置的，高层信令携带配置信息。示例性的，高层信令可以为无线资源控制(Radio Resource Control，RRC)信令。
其中，CSI用于指示矩阵H (r)，矩阵H (r)用于指示第一通信装置第r个天线端口的信道，矩阵H (r)为N t行N sb列的矩阵。N sb是CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是第一通信装置的天线端口数量。其中，频域单元可以是子载波、资源块(Resource Block，RB)、子带(Sub Band)等，本申请实施例不做限制。
其中，CSI包括第一指示信息和第二指示信息。
其中，第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000001
矩阵
Figure PCTCN2019086909-appb-000002
为2L行P列的矩阵，L为正整数，2L小于N t。
其中，第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000003
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000004
和矩阵
Figure PCTCN2019086909-appb-000005
Figure PCTCN2019086909-appb-000006
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000007
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
可选的，针对不同的天线端口r，
Figure PCTCN2019086909-appb-000008
可以取值相同；或者，N r个天线端口按照某种原则分组，每一组中的天线端口的
Figure PCTCN2019086909-appb-000009
取值相同。通过该方案可以降低第一指示信息的上报开销。
可选的，针对不同的天线端口r，
Figure PCTCN2019086909-appb-000010
可以取值相同；或者，N r个天线端口按照某种原则分组，每一组中的天线端口的
Figure PCTCN2019086909-appb-000011
取值相同。通过该方案可以降低第一指示信息的上报开销。
在该技术方案中，通过
Figure PCTCN2019086909-appb-000012
或者
Figure PCTCN2019086909-appb-000013
和
Figure PCTCN2019086909-appb-000014
包含N tN sb个元素的信道矩阵H (r)被压缩为包含2LP个元素的矩阵
Figure PCTCN2019086909-appb-000015
降低了需要反馈的元素的个数，进而降低反馈CSI的开销。另外，第一通信装置将包括第一指示信息和第二指示信息的CSI发送给第二通信装置，第二通信装置可以基于第一指示信息和第二指示信息确定信道矩阵H (r)，以支持非线性预编码，以及更好的信道信息预测。
可选的，
Figure PCTCN2019086909-appb-000016
的形式为：由L个基底按如下顺序排列为对角块矩阵，
Figure PCTCN2019086909-appb-000017
Figure PCTCN2019086909-appb-000018
L个基底可以来源于一组预定义的基底，可选的，预定义基底为离散傅里叶变换(Discrete Fourier Transformation，DFT)基底。第一通信装置可以通过上报
Figure PCTCN2019086909-appb-000019
中每个基底的索引来上报
Figure PCTCN2019086909-appb-000020
可选的，
Figure PCTCN2019086909-appb-000021
的形式为：由K个基底按如下方式构成
Figure PCTCN2019086909-appb-000022
K 个基底可以来源于一组预定义的基底，可选的，预定义基底为DFT基底。第一通信装置可以通过上报
Figure PCTCN2019086909-appb-000023
中每个基底的索引来上报
Figure PCTCN2019086909-appb-000024
在一种实现方式中，第二指示信息包括第三指示信息和第四指示信息，第三指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000025
为L行P列M 1页的三维矩阵，M 1为小于2N r的正整数；
第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000026
为2N r行M 1列矩阵，或，第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000027
为2N r行M 1列矩阵。
可选的，在第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000028
的情况下，l取值不同时，
Figure PCTCN2019086909-appb-000029
是相同的矩阵，通过该方案可降低反馈
Figure PCTCN2019086909-appb-000030
的开销。可选的，在第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000031
的情况下，k取值不同时，
Figure PCTCN2019086909-appb-000032
是相同的矩阵，通过该方案可降低反馈
Figure PCTCN2019086909-appb-000033
的开销。此时，
Figure PCTCN2019086909-appb-000034
或
Figure PCTCN2019086909-appb-000035
可以由W Rx,1表示。
其中，矩阵
Figure PCTCN2019086909-appb-000036
或
Figure PCTCN2019086909-appb-000037
中的列向量为压缩基底，2N r为压缩基底的维度，M 1为压缩基底的个数。
可选的，第四指示信息指示矩阵
Figure PCTCN2019086909-appb-000038
或
Figure PCTCN2019086909-appb-000039
中的元素，即矩阵
Figure PCTCN2019086909-appb-000040
或
Figure PCTCN2019086909-appb-000041
中的压缩基底不是预定义的，而是由第一通信装置计算并上报第二通信装置。相较于预定义基底，非预定义基底
Figure PCTCN2019086909-appb-000042
或
Figure PCTCN2019086909-appb-000043
可以更灵活地匹配待压缩信道。
在该技术方案中，首先，当第一通信装置为移动终端时，由于移动终端周围的散射环境较为复杂，第一通信装置天线端口侧的信道角度扩展较大，导致使用DFT基底作为压缩矩阵对第一通信装置的多个天线端口r对应的
Figure PCTCN2019086909-appb-000044
进行压缩时，压缩率很低，也就是说，很难通过较少的DFT基底准确的表示多个天线端口r对应的
Figure PCTCN2019086909-appb-000045
但是，非预定义基底可以更好地匹配，例如该非预定义的基底可以是根据特征向量确定的，该特征向量是通过多天线端口r对应的
Figure PCTCN2019086909-appb-000046
确定的特征空间的特征向量。该特征向量可以更好的匹配第一通信装置的多个天线端口r对应的
Figure PCTCN2019086909-appb-000047
因此压缩效果更好，也就是说，该技术方案通过较少的压缩基底即可精确的表示
Figure PCTCN2019086909-appb-000048
其次，DFT预定义基底仅适用于相同极化方向的多个天线端口的信道矩阵压缩，在第一通信装置或第二通信装置的天线端口有多个极化方向的场景下，需要根据不同的极化方向对信道矩阵进行拆分，并分别用DFT预定义基底压缩。而非预定义基底则更加灵活，可 根据不同极化方向天线端口信道矩阵之间的相关性计算得到，以支持不同极化方向天线端口信道矩阵的联合压缩，提高压缩效果。
再次，通过
Figure PCTCN2019086909-appb-000049
或
Figure PCTCN2019086909-appb-000050
包含2L×P×N r(即2L行P列N r页)个元素的
Figure PCTCN2019086909-appb-000051
Figure PCTCN2019086909-appb-000052
被压缩为包含L×P×M 1(即L行P列M 1页)个元素的三维矩阵
Figure PCTCN2019086909-appb-000053
降低了需要反馈的元素的个数，降低反馈CSI的开销。
可选的，第一通信装置可以按如下方法构造W Rx,1：第一通信装置接收参考信号进行测量，得到
Figure PCTCN2019086909-appb-000054
和
Figure PCTCN2019086909-appb-000055
和
Figure PCTCN2019086909-appb-000056
均为规模L×P(即L行P列)的矩阵，
Figure PCTCN2019086909-appb-000057
和
Figure PCTCN2019086909-appb-000058
是分别根据第二通信装置两个极化方向天线端口与第一通信装置第r个天线端口之间的信道确定的系数矩阵。然后第一通信装置将2N r个二维矩阵
Figure PCTCN2019086909-appb-000059
拼接为L×P×2N r(即L行P列2N r页)的三维矩阵
Figure PCTCN2019086909-appb-000060
并将三维矩阵
Figure PCTCN2019086909-appb-000061
展开为规模为2N r×PL(即2N r行PL列)的二维矩阵
Figure PCTCN2019086909-appb-000062
第一通信装置令
Figure PCTCN2019086909-appb-000063
这个规模的复数矩阵，并对
Figure PCTCN2019086909-appb-000064
的协方差矩阵
Figure PCTCN2019086909-appb-000065
进行特征分解，其最大的M 1个特征值对应的特征向量构成压缩矩阵
Figure PCTCN2019086909-appb-000066
可选的，第一通信装置还可以将最大的M 1个特征值对应的特征向量进行量化后构成压缩矩阵
Figure PCTCN2019086909-appb-000067
在该技术方案中，可以由三维矩阵
Figure PCTCN2019086909-appb-000068
生成
Figure PCTCN2019086909-appb-000069
或
Figure PCTCN2019086909-appb-000070
或由其他方式生成W Rx,1，本申请实施例不做限制。例如，第一通信装置可以利用第一相关性和第二相关性生成
Figure PCTCN2019086909-appb-000071
或
Figure PCTCN2019086909-appb-000072
第一相关性为第一通信装置多个天线端口的信道之间的相关性，第二相关性为第二通信装置两个极化方向的天线端口的信道之间的相关性。
其中，2N r个矩阵
Figure PCTCN2019086909-appb-000073
到三维矩阵
Figure PCTCN2019086909-appb-000074
各页之间的映射可以由函数g(r,i)给定，即
Figure PCTCN2019086909-appb-000075
且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
Figure PCTCN2019086909-appb-000076
表示取矩阵
Figure PCTCN2019086909-appb-000077
固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
Figure PCTCN2019086909-appb-000078
可选的，g(r,1)＝r，g(r,2)＝r+N r，也就是说，将第二通信装置同一极化方向天线端口组与第一通信装置多个天线端口之间的信道放在一起；
可选的，g(r,1)＝2×(r-1)+1，g(r,2)＝2×(r-1)+2，也就是说，将第一通信装置同一天线端口与第二通信装置两个极化方向天线端口组之间的信道放在一起。
在上述基底
Figure PCTCN2019086909-appb-000079
或
Figure PCTCN2019086909-appb-000080
定义方式下，信道矩阵
Figure PCTCN2019086909-appb-000081
和第三指示信息、第四指示信息的关系为：
Figure PCTCN2019086909-appb-000082
这里的
Figure PCTCN2019086909-appb-000083
为压缩后的有损信道信息。
若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000084
则所述
Figure PCTCN2019086909-appb-000085
Figure PCTCN2019086909-appb-000086
得到，或若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000087
则所述
Figure PCTCN2019086909-appb-000088
Figure PCTCN2019086909-appb-000089
得到；
Figure PCTCN2019086909-appb-000090
表示矩阵
Figure PCTCN2019086909-appb-000091
固定前两个索引后长度为M 1的一维向量，所述长度M 1的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000092
表示矩阵
Figure PCTCN2019086909-appb-000093
固定前两个索引后长度为2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
Figure PCTCN2019086909-appb-000094
×为把一维向量
Figure PCTCN2019086909-appb-000095
当做列向量的矩阵乘运算。
在该技术方案中，通过将矩阵
Figure PCTCN2019086909-appb-000096
作为复数矩阵整体进行压缩，将包含2L×P×N r个元素的矩阵
Figure PCTCN2019086909-appb-000097
压缩为仅包含L×P×M 1个元素的矩阵
Figure PCTCN2019086909-appb-000098
大幅度降低CSI上报开销，同时不带来明显的性能损失。
在一种实现方式中，所述矩阵
Figure PCTCN2019086909-appb-000099
的幅度矩阵
Figure PCTCN2019086909-appb-000100
所述
Figure PCTCN2019086909-appb-000101
的第l行第k列的元素是根据所述
Figure PCTCN2019086909-appb-000102
的第l行第k列的元素的幅度确定的值，例如，所述
Figure PCTCN2019086909-appb-000103
的第l行第k列的元素是所述
Figure PCTCN2019086909-appb-000104
的第l行第k列的元素的幅度或幅度的平方；
其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
所述
Figure PCTCN2019086909-appb-000105
为L行P列2N r页的三维矩阵，
Figure PCTCN2019086909-appb-000106
表示取矩阵
Figure PCTCN2019086909-appb-000107
固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行k列元素为
Figure PCTCN2019086909-appb-000108
若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000109
则所述
Figure PCTCN2019086909-appb-000110
Figure PCTCN2019086909-appb-000111
或若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000112
则所述
Figure PCTCN2019086909-appb-000113
Figure PCTCN2019086909-appb-000114
得到；
Figure PCTCN2019086909-appb-000115
表示取矩阵
Figure PCTCN2019086909-appb-000116
固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000117
表示取矩阵
Figure PCTCN2019086909-appb-000118
固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
Figure PCTCN2019086909-appb-000119
×为把一维向量
Figure PCTCN2019086909-appb-000120
当做列向量的矩阵乘运算。
在该技术方案中，通过对矩阵
Figure PCTCN2019086909-appb-000121
的幅度矩阵
Figure PCTCN2019086909-appb-000122
进行压缩，将包含2L×P×N r个元素的幅度矩阵
Figure PCTCN2019086909-appb-000123
压缩为仅包含L×P×M 1个元素的矩阵
Figure PCTCN2019086909-appb-000124
在矩阵
Figure PCTCN2019086909-appb-000125
中，幅度受信道的空间传输距离、散射环境等影响，由于第一通信装置的多个天线端口之间、第二通信装置的多个天线端口之间距离均较近，故幅度矩阵
Figure PCTCN2019086909-appb-000126
具有较 强的相关性，可通过本发明在大幅度降低CSI上报开销的同时，不带来明显的性能损失。
在一种实现方式中，所述第二指示信息还包括第五指示信息，所述第五指示信息用于指示所述
Figure PCTCN2019086909-appb-000127
中元素的相位。
在该技术方案中，当第一通信装置和第二通信装置之间散射环境较复杂时，例如非视距(Non Line of Sight，NLoS)散射环境，矩阵
Figure PCTCN2019086909-appb-000128
的相位相关性较弱，对
Figure PCTCN2019086909-appb-000129
中元素的相位不进行压缩，即直接上报矩阵
Figure PCTCN2019086909-appb-000130
中元素的相位，可保证反馈精度。
可选的，所述第五指示信息用于指示所述
Figure PCTCN2019086909-appb-000131
中元素的相位，包括：
所述第五指示信息包括第六指示信息和第七指示信息，所述第六指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000132
为L行P列M 2页的三维矩阵，M 2为小于2N r的正整数；
所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000133
为2N r行M 2列矩阵，或，所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000134
为2N r行M 2列矩阵；
所述矩阵
Figure PCTCN2019086909-appb-000135
的相位矩阵
Figure PCTCN2019086909-appb-000136
所述
Figure PCTCN2019086909-appb-000137
的第l行第k列的元素是根据所述
Figure PCTCN2019086909-appb-000138
的第l行第k列的元素的相位确定的值，例如，所述
Figure PCTCN2019086909-appb-000139
的第l行第k列的元素是所述
Figure PCTCN2019086909-appb-000140
的第l行第k列的元素的相位或相位的平方；
其中，f(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，f(r′,i′)≠f(r″,i″)；
所述
Figure PCTCN2019086909-appb-000141
为L行P列2N r页的三维矩阵，
Figure PCTCN2019086909-appb-000142
表示取矩阵
Figure PCTCN2019086909-appb-000143
固定第三个索引为f(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
Figure PCTCN2019086909-appb-000144
若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000145
则所述
Figure PCTCN2019086909-appb-000146
Figure PCTCN2019086909-appb-000147
或若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000148
则所述
Figure PCTCN2019086909-appb-000149
Figure PCTCN2019086909-appb-000150
Figure PCTCN2019086909-appb-000151
表示取矩阵
Figure PCTCN2019086909-appb-000152
固定前两个索引长度M 2的一维向量，所述长度M 2的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000153
表示取矩阵
Figure PCTCN2019086909-appb-000154
固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第f(r,i)个元素为
Figure PCTCN2019086909-appb-000155
×为把一维向量
Figure PCTCN2019086909-appb-000156
当做列向量的矩阵乘运算。
在该技术方案中，通过对矩阵
Figure PCTCN2019086909-appb-000157
的相位矩阵
Figure PCTCN2019086909-appb-000158
进行压缩，将包含2L×P×N r个元素的相位矩阵
Figure PCTCN2019086909-appb-000159
压缩为仅包含L×P×M 2个元素的矩阵
Figure PCTCN2019086909-appb-000160
当第一通信装置和第二通信装置之间散射环境较简单时，例如视距(Line of Sight，LoS)散射环境，矩阵
Figure PCTCN2019086909-appb-000161
的相位相关性较强，对矩阵
Figure PCTCN2019086909-appb-000162
的相位矩阵
Figure PCTCN2019086909-appb-000163
进行压缩上报，可在大幅度降低CSI上报开销的同时，不带来明显的性能损失。
在一种实现方式中，在所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000164
的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
Figure PCTCN2019086909-appb-000165
则l′＝l″；或
在所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000166
和矩阵
Figure PCTCN2019086909-appb-000167
的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
Figure PCTCN2019086909-appb-000168
则l′＝l″，对于任意r′,r″∈{1,2,…,N r}，k′,k″∈{1,2,…,K}，若r′≠r″且
Figure PCTCN2019086909-appb-000169
则k′＝k″。
在该技术方案中，矩阵
Figure PCTCN2019086909-appb-000170
的各列向量为对矩阵H (r′)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000171
的各列向量为对矩阵H (r″)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000172
的各列向量和矩阵
Figure PCTCN2019086909-appb-000173
的各列向量可以来源于同一组预定义的DFT基底，若矩阵
Figure PCTCN2019086909-appb-000174
和矩阵
Figure PCTCN2019086909-appb-000175
选取了同一个DFT基底，即
Figure PCTCN2019086909-appb-000176
Figure PCTCN2019086909-appb-000177
则将该DFT基底在矩阵
Figure PCTCN2019086909-appb-000178
和矩阵
Figure PCTCN2019086909-appb-000179
中放在相同的位置，即l′＝l″，这样矩阵H (r′)和矩阵H (r″)压缩后的矩阵
Figure PCTCN2019086909-appb-000180
和矩阵
Figure PCTCN2019086909-appb-000181
的第l′＝l″行对应同一个基底；类似地，矩阵
Figure PCTCN2019086909-appb-000182
和矩阵
Figure PCTCN2019086909-appb-000183
的约束使得矩阵
Figure PCTCN2019086909-appb-000184
和矩阵
Figure PCTCN2019086909-appb-000185
的同一列尽量对应同一基底。上述对于矩阵
Figure PCTCN2019086909-appb-000186
和矩阵
Figure PCTCN2019086909-appb-000187
的约束可提高向量
Figure PCTCN2019086909-appb-000188
之间的相关性，从而提升压缩效果。
在一种实现方式中，所述第三指示信息包括用于指示
Figure PCTCN2019086909-appb-000189
中元素
Figure PCTCN2019086909-appb-000190
的X 1 l,k,g个比特，X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000191
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000192
中第c行第d列的元素
Figure PCTCN2019086909-appb-000193
的Y 1 l,c,d个比特，所述Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000194
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000195
中第c行第d列的元素
Figure PCTCN2019086909-appb-000196
的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000197
则Y 1 l,c,d大于等于X 1 l,k,g。可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000198
则Z 1 k,c,d大于等于X 1 l,k,g。
在该技术方案中，由于所述第四指示信息指示的矩阵
Figure PCTCN2019086909-appb-000199
或
Figure PCTCN2019086909-appb-000200
Figure PCTCN2019086909-appb-000201
为对所述矩阵
Figure PCTCN2019086909-appb-000202
压缩的基底矩阵，所述第三指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000203
为对所述 矩阵
Figure PCTCN2019086909-appb-000204
压缩的合并系数矩阵，在选择相同的压缩基底个数M 1的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000205
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000206
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 1，有利于降低上报CSI的开销。
在一种实现方式中，所述第六指示信息包括用于指示
Figure PCTCN2019086909-appb-000207
中元素
Figure PCTCN2019086909-appb-000208
的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000209
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000210
中第c行第d列的元素
Figure PCTCN2019086909-appb-000211
的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000212
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000213
中第c行第d列的元素
Figure PCTCN2019086909-appb-000214
的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000215
则Y 2 l,c,d大于等于X 2 l,k,f。可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000216
则Z 2 k,c,d大于等于X 2 l,k,f。
在该技术方案中，由于所述第七指示信息指示的矩阵
Figure PCTCN2019086909-appb-000217
或
Figure PCTCN2019086909-appb-000218
Figure PCTCN2019086909-appb-000219
为对所述矩阵
Figure PCTCN2019086909-appb-000220
压缩的基底矩阵，所述第六指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000221
为对所述矩阵
Figure PCTCN2019086909-appb-000222
压缩的合并系数矩阵，在选择相同的压缩基底个数M 2的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000223
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000224
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 2，有利于降低上报CSI的开销。
可选的，第一通信装置对所述矩阵
Figure PCTCN2019086909-appb-000225
的幅度矩阵
Figure PCTCN2019086909-appb-000226
和相位矩阵
Figure PCTCN2019086909-appb-000227
分别压缩，此时，用于所述幅度矩阵
Figure PCTCN2019086909-appb-000228
的压缩矩阵
Figure PCTCN2019086909-appb-000229
或
Figure PCTCN2019086909-appb-000230
与用于所述相位矩阵
Figure PCTCN2019086909-appb-000231
的压缩矩阵
Figure PCTCN2019086909-appb-000232
或
Figure PCTCN2019086909-appb-000233
可以不同，是独立计算的。
在该技术方案中，矩阵
Figure PCTCN2019086909-appb-000234
的幅度矩阵
Figure PCTCN2019086909-appb-000235
具有较强的相关性，相位矩阵
Figure PCTCN2019086909-appb-000236
相关性往往弱于幅度矩阵
Figure PCTCN2019086909-appb-000237
故幅度矩阵
Figure PCTCN2019086909-appb-000238
相较相位矩阵
Figure PCTCN2019086909-appb-000239
往往易于压缩。幅度矩阵
Figure PCTCN2019086909-appb-000240
的压缩率由M 1决定，相位矩阵
Figure PCTCN2019086909-appb-000241
的压缩率由M 2决定，可通过设定不同的M 1和M 2实现矩阵
Figure PCTCN2019086909-appb-000242
幅度和相位的不同压缩率，优选的，M 1小于M 2。
在一种实现方式中，所述第一通信装置上报CSI，包括：
所述第一通信装置在相同的时刻上报所述第三指示信息和所述第四指示信息；或
所述第一通信装置在不同的时刻上报所述第三指示信息和所述第四指示信息。
在该技术方案中，当信道时域变化较快时，第一通信装置为每一个时刻的信道H (r)单独计算压缩矩阵
Figure PCTCN2019086909-appb-000243
或
Figure PCTCN2019086909-appb-000244
第一通信装置计算的压缩矩阵
Figure PCTCN2019086909-appb-000245
或
Figure PCTCN2019086909-appb-000246
为该时刻的信道H (r)独用，在这种情况下，第一通信装置同时上报第三指示信息和第四指示信息；当信道时域变化较慢时，第一通信装置计算的压缩矩阵
Figure PCTCN2019086909-appb-000247
或
Figure PCTCN2019086909-appb-000248
为多个时刻的信道H (r)共用，以降低压缩矩阵
Figure PCTCN2019086909-appb-000249
或
Figure PCTCN2019086909-appb-000250
的反馈量，在这种情况下，第一通信装置在不同时刻上报第三指示信息和第四指示信息。
在一种实现方式中，所述第一通信装置上报CSI，包括：
所述第一通信装置在相同的时刻上报所述第六指示信息和所述第七指示信息；或
所述第一通信装置在不同的时刻上报所述第六指示信息和所述第七指示信息。
在该技术方案中，当信道时域变化较快时，第一通信装置为每一个时刻的信道H (r)单独计算压缩矩阵
Figure PCTCN2019086909-appb-000251
或
Figure PCTCN2019086909-appb-000252
第一通信装置计算的压缩矩阵
Figure PCTCN2019086909-appb-000253
或
Figure PCTCN2019086909-appb-000254
为该时刻的信道H (r)独用，在这种情况下，第一通信装置同时上报第六指示信息和第七指示信息；当信道时域变化较慢时，第一通信装置计算的压缩矩阵
Figure PCTCN2019086909-appb-000255
或
Figure PCTCN2019086909-appb-000256
为多个时刻的信道H (r)共用，以降低压缩矩阵
Figure PCTCN2019086909-appb-000257
或
Figure PCTCN2019086909-appb-000258
的反馈量，在这种情况下，第一通信装置在不同时刻上报第六指示信息和第七指示信息。
在一种实现方式中，所述第三指示信息的上报为非周期上报，所述第四指示信息的上报为周期上报。
在该技术方案中，在所述第一通信装置在不同的时刻上报所述第三指示信息和所述第四指示信息情况下，第一通信装置可以周期上报压缩矩阵
Figure PCTCN2019086909-appb-000259
或
Figure PCTCN2019086909-appb-000260
周期的大小决定第二通信装置获取压缩矩阵
Figure PCTCN2019086909-appb-000261
或
Figure PCTCN2019086909-appb-000262
的实时性，第一通信装置可以非周期上报三维矩阵
Figure PCTCN2019086909-appb-000263
上报可以由第一通信装置或第二通信装置触发。
在一种实现方式中，所述第六指示信息的上报为非周期上报，所述第七指示信息的上报为周期上报。
在该技术方案中，在所述第一通信装置在不同的时刻上报所述第六指示信息和所述第七指示信息情况下，第一通信装置可以周期上报压缩矩阵
Figure PCTCN2019086909-appb-000264
或
Figure PCTCN2019086909-appb-000265
周期的大小决定第二通信装置获取压缩矩阵
Figure PCTCN2019086909-appb-000266
或
Figure PCTCN2019086909-appb-000267
的实时性，第一通信装置可以非周期上报三维矩阵
Figure PCTCN2019086909-appb-000268
上报可以由第一通信装置或第二通信装置触发。
在一种实现方式中，所述M 1可配置，所述网络设备通过RRC或媒体接入控制层控制信令(MediaAccess ControlControl Element，MACCE)或下行链路控制信息(Downlink Control Information，DCI)信令配置M 1的取值。
在一种实现方式中，所述M 2可配置，所述网络设备通过RRC或MACCE或DCI信令配置M 2的取值。
第二方面，本申请实施例提供了一种通信装置，所述装置包括用于实现第一方面所述的上报CSI的方法的单元。
第三方面，本申请实施例提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序或指令，当所述程序或指令被处理器执行时，使所述处理器执行如第一方面所述的上报CSI的方法。
第四方面，本申请实施例提供一种终端设备，包括接收器和发射器，所述接收器与发射器耦合，其特征在于，
所述接收器，用于接收来自于第二通信装置的配置信息，所述配置信息用于配置所述发射器上报的CSI；
所述发射器，用于上报CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述终端设备第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述终端设备的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000269
所述矩阵
Figure PCTCN2019086909-appb-000270
为2L行P列的矩阵，L为正整数，2L小于N t；
所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000271
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000272
和矩阵
Figure PCTCN2019086909-appb-000273
Figure PCTCN2019086909-appb-000274
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000275
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
第五方面，本申请实施例提供一种网络设备，包括接收器和发射器，所述接收器与发射器耦合，其特征在于，
所述发射器，用于向第一通信装置发送配置信息，所述配置信息用于配置所述第一通 信装置上报的CSI；
所述接收器，用于接收CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000276
所述矩阵
Figure PCTCN2019086909-appb-000277
为2L行P列的矩阵，L为正整数，2L小于N t；
所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000278
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000279
和矩阵
Figure PCTCN2019086909-appb-000280
Figure PCTCN2019086909-appb-000281
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000282
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
第六方面，本申请实施例提供一种芯片系统，其特征在于，所述芯片系统包括处理器和接口电路，所述接口电路与所述处理器耦合，
所述处理器用于执行计算机程序或指令，以实现如第一方面所述的方法；
所述接口电路用于与所述芯片系统之外的其它模块进行通信。
附图说明
图1是本发明实施例提供的一种通信系统的结构示意图；
图2是本发明实施例提供的一种上报CSI的方法的流程示意图；
图3是本发明实施例提供的一种矩阵压缩的示意图；
图4是本发明实施例提供的另一种通信系统的结构示意图；
图5是本发明实施例提供的一种终端设备的结构示意图；
图6是本发明实施例提供的一种网络设备的结构示意图。
具体实施方式
下面结合本发明实施例中的附图对本发明实施例进行描述。
为了更好的理解本申请实施例公开的上报信道状态信息的方法、装置及相关设备，下面首先对本申请实施的通信系统进行描述。请参见图1，图1是本申请实施例公开的一种通信系统的结构示意图，该通信系统可以包括至少一个第一通信装置以及至少一个第二通信装置，其中，第一通信装置与第二通信装置建立通信连接。
相对现有的一种终端设备向网络设备反馈CSI的技术方案，网络设备仅能获取信道矩阵H的特征向量V，无法获取信道矩阵H，继而无法支持脏纸编码、迭代预编码等高性能非线性预编码以及基于信道矩阵H的信道信息预测，具体为：终端设备反馈的CSI包括信道矩阵的RI、PMI和CQI，其中PMI包括PMI1和PMI2，PMI1用于指示矩阵W 1中的全部元素，PMI2用于指示矩阵W 2中的全部元素，W 1和W 2的乘积表征信道矩阵H的特征向量V。
本申请实施例中，第一通信装置接收来自于第二通信装置的配置信息，配置信息用于配置第一通信装置上报的CSI。然后，第一通信装置可以上报CSI，CSI包括第一指示信息和第二指示信息。
其中，CSI用于指示矩阵H (r)，矩阵H (r)用于指示第一通信装置第r个天线端口的信道，r∈{1,2,…,N r}，N r是第一通信装置的天线端口数量。矩阵H (r)为N t行N sb列的矩阵，示例性的，矩阵H (r)可以为N t行N sb列的复矩阵。
其中，N sb是CSI的上报带宽所包括的频域单元数量。N t是第二通信装置的天线端口数量。
例如，如果第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000283
和矩阵
Figure PCTCN2019086909-appb-000284
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000285
为N sb行K列的矩阵，L为正整数，2L小于N t，K为正整数，K小于N sb，则所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000286
所述矩阵
Figure PCTCN2019086909-appb-000287
为2L行K列的矩阵，
Figure PCTCN2019086909-appb-000288
Figure PCTCN2019086909-appb-000289
其中，
Figure PCTCN2019086909-appb-000290
的形式为：由L个基底按如下顺序排列为对角块矩阵，
Figure PCTCN2019086909-appb-000291
Figure PCTCN2019086909-appb-000292
L个基底可以来源于一组预定义的基底。第一通信装置可以通过上报
Figure PCTCN2019086909-appb-000293
中每个基底的索引来上报
Figure PCTCN2019086909-appb-000294
其中，
Figure PCTCN2019086909-appb-000295
的形式为：由K个基底按如下方式构成
Figure PCTCN2019086909-appb-000296
K个基底可以来源于一组预定义的基底。第一通信装置可以通过上报
Figure PCTCN2019086909-appb-000297
中每个基底的索引来上报
Figure PCTCN2019086909-appb-000298
在该实施例中，通过
Figure PCTCN2019086909-appb-000299
和
Figure PCTCN2019086909-appb-000300
包含N tN sb个元素的信道矩阵H (r)被压缩为包含2LK个元素的矩阵
Figure PCTCN2019086909-appb-000301
降低了需要反馈的元素的个数，进而降低反馈CSI的开销。另外，第一通信装置将包括第一指示信息和第二指示信息的CSI发送给第二通信装置，第二通信装置可以基于第一指示信息和第二指示信息确定信道矩阵H (r)，以支持非线性预编码，以及更好的信道信息预测。
又如，如果第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000302
为N t行2L列的矩阵，L为正整数，2L小于N t，则第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000303
矩阵
Figure PCTCN2019086909-appb-000304
为2L行N sb列的矩阵，
Figure PCTCN2019086909-appb-000305
Figure PCTCN2019086909-appb-000306
在该实施例中，通过
Figure PCTCN2019086909-appb-000307
包含N tN sb个元素的信道矩阵H (r)被压缩为包含2LN sb个元素的矩阵
Figure PCTCN2019086909-appb-000308
降低了需要反馈的元素的个数，进而降低反馈CSI的开销。另外，第一通信装置将包括第一指示信息和第二指示信息的CSI发送给第二通信装置，第二通信装置可以 基于第一指示信息和第二指示信息确定信道矩阵H (r)，以支持非线性预编码，以及更好的信道信息预测。
本发明实施例的技术方案可以应用于各种通信系统，例如：全球移动通讯(Global System of Mobile communication，GSM)系统、码分多址(Code Division Multiple Access，CDMA)系统、宽带码分多址(Wideband Code Division Multiple Access，WCDMA)系统、通用分组无线业务(General Packet Radio Service，GPRS)、长期演进(Long Term Evolution，LTE)系统、LTE频分双工(Frequency Division Duplex，FDD)系统、LTE时分双工(Time Division Duplex，TDD)、通用移动通信系统(Universal Mobile Telecommunication System，UMTS)、全球互联微波接入(Worldwide Interoperability for Microwave Access，WiMAX)通信系统、第五代移动通信技术(4th-Generation，4G)通信系统、4G新空口(New Radio，NR)通信系统、NR、URLLC、NB-IoT或者eMTC系统等。
第二通信装置可以为网络设备，网络设备具体可以是无线局域网(Wireless Local Area Networks，WLAN)中的接入点(Access Point，AP)、全球移动通信系统(Global System for Mobile Communication，GSM)或码分多址(Code Division Multiple Access，CDMA)中的基站收发信台(Base Transceiver Station，BTS)、宽带码分多址(Wideband Code Division Multiple Access，WCDMA)中的基站(NodeB，NB)、LTE系统中的演进型基站(Evolutional Node B，eNB)、中继站或接入点、车载设备、可穿戴设备、未来4G网络中的接入网设备以及未来演进的公共陆地移动网络(Public Land Mobile Network，PLMN)中的接入网设备等中的任意一种。
第一通信装置可以为终端设备，终端设备也可以称为用户设备(User Equipment，UE)、移动台、接入终端、用户单元、用户站、移动站、远方站、远程终端、移动设备、终端、无线通信设备、用户代理或用户装置等，其具体可以是WLAN中的站点(Station，ST)、蜂窝电话、无绳电话、会话启动协议(Session Initiation Protocol，SIP)电话、无线本地环路(Wireless Local Loop，WLL)站、个人数字处理(Personal Digital Assistant，PDA)、具有无线通信功能的手持设备、计算设备、连接到无线调制解调器的其它处理设备、车载设备、可穿戴设备、未来4G网络中的移动台以及未来演进的PLMN网络中的终端设备等中的任意一种。
基于图1所示的通信系统的结构示意图，请参见图2，图2是本发明实施例提供的一种上报CSI的方法的流程示意图。具体的，如图2所示，本发明实施例的上报CSI的方法可以包括以下步骤：
201、第二通信装置向第一通信装置发送配置信息，配置信息用于配置第一通信装置上报的CSI。
202、第一通信装置获取CSI，CSI包括第一指示信息和第二指示信息，第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000309
和矩阵
Figure PCTCN2019086909-appb-000310
第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000311
第一通信装置可以将矩阵
Figure PCTCN2019086909-appb-000312
作为复数矩阵整体进行压缩，或者对矩阵
Figure PCTCN2019086909-appb-000313
的幅度矩阵进行压缩，或者对矩阵
Figure PCTCN2019086909-appb-000314
的相位矩阵进行压缩，或者分别对矩阵
Figure PCTCN2019086909-appb-000315
的幅度矩阵以及矩阵
Figure PCTCN2019086909-appb-000316
的相位矩阵进行压缩，或者直接上报矩阵
Figure PCTCN2019086909-appb-000317
的相位矩阵，即不对矩阵
Figure PCTCN2019086909-appb-000318
的相位矩阵进行压缩。
一、在将矩阵
Figure PCTCN2019086909-appb-000319
作为复数矩阵整体进行压缩的情况下，第二指示信息可以包括第三指示信息和第四指示信息。第三指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000320
为L×K×M 1(即L行K列M 1页)的三维矩阵，M 1为正整数。第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000321
为2N r行M 1列矩阵。或者，在将矩阵
Figure PCTCN2019086909-appb-000322
作为复数矩阵整体进行压缩的情况下，第二指示信息可以包括第三指示信息和第四指示信息。第三指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000323
为L×K×M 1(即L行K列M 1页)的三维矩阵，M 1为正整数。第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000324
Figure PCTCN2019086909-appb-000325
为2N r行M 1列矩阵。
其中，三维矩阵
Figure PCTCN2019086909-appb-000326
也可以称为三维张量
Figure PCTCN2019086909-appb-000327
在一种实现方式中，在第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000328
的情况下，l取值不同时，
Figure PCTCN2019086909-appb-000329
是相同的矩阵，通过该方案可降低反馈
Figure PCTCN2019086909-appb-000330
的开销。或者，在第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000331
的情况下，k取值不同时，
Figure PCTCN2019086909-appb-000332
是相同的矩阵，通过该方案可降低反馈
Figure PCTCN2019086909-appb-000333
的开销。此时，
Figure PCTCN2019086909-appb-000334
或
Figure PCTCN2019086909-appb-000335
可以由W Rx,1表示。
举例来说，第一通信装置可以按如下方法构造W Rx,1：第一通信装置接收参考信号进行测量，得到
Figure PCTCN2019086909-appb-000336
和
Figure PCTCN2019086909-appb-000337
和
Figure PCTCN2019086909-appb-000338
均为规模L×K(即L行K列)的矩阵，
Figure PCTCN2019086909-appb-000339
和
Figure PCTCN2019086909-appb-000340
是分别根据第二通信装置两个极化方向天线端口与第一通信装置第r个天线端口之间的信道确定的系数矩阵。然后第一通信装置将2N r个二维矩阵
Figure PCTCN2019086909-appb-000341
Figure PCTCN2019086909-appb-000342
拼接为L×K×2N r(即L行K列2N r页)的三维矩阵
Figure PCTCN2019086909-appb-000343
并将三维矩阵
Figure PCTCN2019086909-appb-000344
展开为规模为2N r×KL(即2N r行KL列)的二维矩阵
Figure PCTCN2019086909-appb-000345
第一通信装置令
Figure PCTCN2019086909-appb-000346
这个规模的复数矩阵，并对
Figure PCTCN2019086909-appb-000347
的协方差矩阵
Figure PCTCN2019086909-appb-000348
进行特征分解，其最大的M 1个特征值对应的特征向量构成压缩矩阵
Figure PCTCN2019086909-appb-000349
可选的，第一通信装置还可以将最大的M 1个特征值对应的特征向量进行量化后构成压缩矩阵
Figure PCTCN2019086909-appb-000350
首先，第一通信装置可以进行矩阵变形，以图3为例，第一通信装置可以将
Figure PCTCN2019086909-appb-000351
拆分为
Figure PCTCN2019086909-appb-000352
构成2N r个规模为L×K(即L行K列)的矩阵。第一通信装置把
Figure PCTCN2019086909-appb-000353
Figure PCTCN2019086909-appb-000354
放在
Figure PCTCN2019086909-appb-000355
后面，得到新构造的三维矩阵，新构造的三维矩阵中包含KL个长度为2N r的向量。第一通信装置将这KL个长度为2N r的向量按任意顺序排列，构成规模为2N r行KL列的二维矩阵
Figure PCTCN2019086909-appb-000356
进一步的，第一通信装置可以构造压缩矩阵，例如第一通信装置可以令
Figure PCTCN2019086909-appb-000357
Figure PCTCN2019086909-appb-000358
这个规模的复数矩阵，对
Figure PCTCN2019086909-appb-000359
的协方差矩阵
Figure PCTCN2019086909-appb-000360
进行特征分解得到
Figure PCTCN2019086909-appb-000361
其中，D是规模为2N r×2N r(即2N r行2N r列)的矩阵，对角线元素为按降序排列的特征值。V是规模为2N r×2N r(即2N r行2N r列)的矩阵，每一列为一个特征向量，与D中的特征值对应。第一通信装置可以通过W Rx,1＝V(：,[1:M 1])得到压缩矩阵W Rx,1。可选的，第一通信装置还可以将W Rx,1＝V(：,[1:M 1])进行量化后构成压缩矩阵
Figure PCTCN2019086909-appb-000362
在上述W Rx,1的构造方式下，第三指示信息和第四指示信息指示复数矩阵
Figure PCTCN2019086909-appb-000363
由
Figure PCTCN2019086909-appb-000364
得到。
其中，
Figure PCTCN2019086909-appb-000365
为L行K列的二维矩阵，由式
Figure PCTCN2019086909-appb-000366
Figure PCTCN2019086909-appb-000367
得到，
Figure PCTCN2019086909-appb-000368
为L行K列的二维矩阵，由式
Figure PCTCN2019086909-appb-000369
得到。
所述
Figure PCTCN2019086909-appb-000370
是长度为2N r的一维向量，由式
Figure PCTCN2019086909-appb-000371
得到，
Figure PCTCN2019086909-appb-000372
表示取矩阵
Figure PCTCN2019086909-appb-000373
固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000374
×为把一维向量
Figure PCTCN2019086909-appb-000375
当做列向量的矩阵乘运算。
在一种实现方式中，为了进一步降低反馈压缩矩阵W Rx,1的开销，T s个时刻的信道共享压缩矩阵W Rx1，其中
Figure PCTCN2019086909-appb-000376
在一种实现方式中，压缩矩阵W Rx1和合并系数矩阵
Figure PCTCN2019086909-appb-000377
的上报方式可以包括如下两种：
压缩矩阵W Rx,1和合并系数矩阵
Figure PCTCN2019086909-appb-000378
可以在相同的时刻上报。
压缩矩阵W Rx,1和合并系数矩阵
Figure PCTCN2019086909-appb-000379
可以在不同的时刻上报。例如，压缩矩阵W Rx,1周期上报，合并系数矩阵
Figure PCTCN2019086909-appb-000380
非周期上报，压缩矩阵W Rx,1可以被多个合并系数矩阵
Figure PCTCN2019086909-appb-000381
共用。
二、在对矩阵
Figure PCTCN2019086909-appb-000382
的幅度矩阵进行压缩的情况下，第三指示信息和第四指示信息指示矩阵
Figure PCTCN2019086909-appb-000383
的幅度矩阵
Figure PCTCN2019086909-appb-000384
其中，本申请实施例对幅度矩阵
Figure PCTCN2019086909-appb-000385
的描述方式和对复数矩阵
Figure PCTCN2019086909-appb-000386
的描述类似， 具体可参见本申请实施例对复数矩阵
Figure PCTCN2019086909-appb-000387
的描述，本申请实施例不再赘述。
三、在直接上报矩阵
Figure PCTCN2019086909-appb-000388
的相位矩阵的情况下，第二指示信息包括第五指示信息，所述第五指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000389
中元素的相位。
四、在对矩阵
Figure PCTCN2019086909-appb-000390
的相位矩阵进行压缩的情况下，第五指示信息可以包括第六指示信息和第七指示信息。第六指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000391
为L×K×M 2(即L行K列M 2页)的三维矩阵，M 2为正整数。第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000392
为2N r行M 2列矩阵。
或者，第五指示信息可以包括第六指示信息和第七指示信息。第六指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000393
为L×K×M 2的三维矩阵，M 2为正整数。第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000394
Figure PCTCN2019086909-appb-000395
为2N r行M 2列矩阵。
其中，本申请实施例对相位矩阵
Figure PCTCN2019086909-appb-000396
的描述方式和对复数矩阵
Figure PCTCN2019086909-appb-000397
的描述类似，具体可参见本申请实施例对复数矩阵
Figure PCTCN2019086909-appb-000398
的描述，本申请实施例不再赘述。
其中，三维矩阵
Figure PCTCN2019086909-appb-000399
也可以称为三维张量
Figure PCTCN2019086909-appb-000400
在一种实现方式中，矩阵
Figure PCTCN2019086909-appb-000401
和三维矩阵
Figure PCTCN2019086909-appb-000402
的上报方式可以包括如下两种：
矩阵
Figure PCTCN2019086909-appb-000403
和三维矩阵
Figure PCTCN2019086909-appb-000404
可以在相同的时刻上报。
矩阵
Figure PCTCN2019086909-appb-000405
和三维矩阵
Figure PCTCN2019086909-appb-000406
可以在不同的时刻上报。例如，矩阵
Figure PCTCN2019086909-appb-000407
周期上报，三维矩阵
Figure PCTCN2019086909-appb-000408
非周期上报，矩阵
Figure PCTCN2019086909-appb-000409
可以被多个三维矩阵
Figure PCTCN2019086909-appb-000410
共用。
在一种实现方式中，矩阵
Figure PCTCN2019086909-appb-000411
和三维矩阵
Figure PCTCN2019086909-appb-000412
的上报方式可以包括如下两种：
矩阵
Figure PCTCN2019086909-appb-000413
和三维矩阵
Figure PCTCN2019086909-appb-000414
可以在相同的时刻上报。
矩阵
Figure PCTCN2019086909-appb-000415
和三维矩阵
Figure PCTCN2019086909-appb-000416
可以在不同的时刻上报。例如，矩阵
Figure PCTCN2019086909-appb-000417
周期上报，三维矩阵
Figure PCTCN2019086909-appb-000418
非周期上报，矩阵
Figure PCTCN2019086909-appb-000419
可以被多个三维矩阵
Figure PCTCN2019086909-appb-000420
共用。
五、在分别对矩阵
Figure PCTCN2019086909-appb-000421
的幅度矩阵和矩阵
Figure PCTCN2019086909-appb-000422
的相位矩阵进行压缩的情况下，第二指示信息包括第三指示信息、第四指示信息以及第五指示信息。
其中，本申请实施例中的第三指示信息和第四指示信息可参见本申请实施例对幅度矩阵
Figure PCTCN2019086909-appb-000423
进行压缩的情况下的具体描述，另外，本申请实施例中的第五指示信息可参见 本申请实施例对相位矩阵
Figure PCTCN2019086909-appb-000424
进行压缩的情况下的具体描述，本申请实施例不再赘述。
在一种实现方式中，幅度矩阵
Figure PCTCN2019086909-appb-000425
的压缩率由M 1决定，相位矩阵
Figure PCTCN2019086909-appb-000426
的压缩率由M 2决定，可通过设定不同的M 1和M 2实现矩阵
Figure PCTCN2019086909-appb-000427
幅度和相位的不同压缩率，优选的，M 1小于M 2。
在一种实现方式中，在第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000428
和矩阵
Figure PCTCN2019086909-appb-000429
的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
Figure PCTCN2019086909-appb-000430
则l′＝l″，对于任意r′,r″∈{1,2,…,Nr}，k′,k″∈{1,2,…,K}，若r′≠r″且
Figure PCTCN2019086909-appb-000431
则k′＝k″。
本申请实施例中，矩阵
Figure PCTCN2019086909-appb-000432
的各列向量为对矩阵H (r′)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000433
的各列向量为对矩阵H (r″)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000434
的各列向量和矩阵
Figure PCTCN2019086909-appb-000435
的各列向量可以来源于同一组预定义的DFT基底，若矩阵
Figure PCTCN2019086909-appb-000436
和矩阵
Figure PCTCN2019086909-appb-000437
选取了同一个DFT基底，即
Figure PCTCN2019086909-appb-000438
Figure PCTCN2019086909-appb-000439
则将该DFT基底在矩阵
Figure PCTCN2019086909-appb-000440
和矩阵
Figure PCTCN2019086909-appb-000441
中放在相同的位置，即l′＝l″，这样矩阵H (r′)和矩阵H (r″)压缩后的矩阵
Figure PCTCN2019086909-appb-000442
和矩阵
Figure PCTCN2019086909-appb-000443
的第l′＝l″行对应同一个基底；类似地，矩阵
Figure PCTCN2019086909-appb-000444
和矩阵
Figure PCTCN2019086909-appb-000445
的约束使得矩阵
Figure PCTCN2019086909-appb-000446
和矩阵
Figure PCTCN2019086909-appb-000447
的同一列尽量对应同一基底。上述对于矩阵
Figure PCTCN2019086909-appb-000448
和矩阵
Figure PCTCN2019086909-appb-000449
的约束可提高向量
Figure PCTCN2019086909-appb-000450
之间的相关性，从而提升压缩效果。
在一种实现方式中，第三指示信息包括用于指示
Figure PCTCN2019086909-appb-000451
中元素
Figure PCTCN2019086909-appb-000452
的X 1 l,k,g个比特，X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000453
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000454
中第c行第d列的元素
Figure PCTCN2019086909-appb-000455
的Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000456
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000457
中第c行第d列的元素
Figure PCTCN2019086909-appb-000458
的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000459
所述Y 1 l,c,d大于等于X 1 l,k,g。可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000460
Z 1 k,c,d大于等于X 1 l,k,g。
本申请实施例中，由于所述第四指示信息指示的矩阵
Figure PCTCN2019086909-appb-000461
或
Figure PCTCN2019086909-appb-000462
Figure PCTCN2019086909-appb-000463
为对所述矩阵
Figure PCTCN2019086909-appb-000464
压缩的基底矩阵，所述第三指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000465
为对所述矩阵
Figure PCTCN2019086909-appb-000466
压缩的合并系数矩阵，在选择相同的压缩基底个数M 1的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000467
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000468
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 1，有利于降低上报CSI的开销。
在一种实现方式中，第六指示信息包括用于指示
Figure PCTCN2019086909-appb-000469
中元素
Figure PCTCN2019086909-appb-000470
的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000471
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000472
中第c行第d列的元素
Figure PCTCN2019086909-appb-000473
的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000474
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000475
中第c行第d列的元素
Figure PCTCN2019086909-appb-000476
的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000477
所述Y 2 l,c,d大于等于X 2 l,k,f。可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000478
Z 2 k,c,d大于等于X 2 l,k,f。
本申请实施例中，由于所述第七指示信息指示的矩阵
Figure PCTCN2019086909-appb-000479
或
Figure PCTCN2019086909-appb-000480
Figure PCTCN2019086909-appb-000481
为对所述矩阵
Figure PCTCN2019086909-appb-000482
压缩的基底矩阵，所述第六指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000483
为对所述矩阵
Figure PCTCN2019086909-appb-000484
压缩的合并系数矩阵，在选择相同的压缩基底个数M 2的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000485
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000486
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 2，有利于降低上报CSI的开销。
203、第一通信装置上报CSI。
若第一通信装置直接反馈矩阵
Figure PCTCN2019086909-appb-000487
反馈量为
Figure PCTCN2019086909-appb-000488
Figure PCTCN2019086909-appb-000489
其中
Figure PCTCN2019086909-appb-000490
为反馈每一个参数的比特数。在第三指示信息和第四指示信息指示复数矩阵
Figure PCTCN2019086909-appb-000491
的实施方法中，压缩矩阵W Rx,1为T s个时刻公用，反馈量包括两部分：反馈第四指示信息的反馈量
Figure PCTCN2019086909-appb-000492
以及反馈第三指示信息的反馈量
Figure PCTCN2019086909-appb-000493
其中
Figure PCTCN2019086909-appb-000494
和
Figure PCTCN2019086909-appb-000495
为反馈每一个参数的比特数。
当N r＝4、L＝4、K＝4、
Figure PCTCN2019086909-appb-000496
时，压缩率如表1所示：
表1
  T s＝1 T s＝3 T s＝5 T s＝9
M 1＝2 0.3400 0.2733 0.2556 0.2511
M 1＝3 0.4240 0.4240 0.4000 0.3916
M 1＝4 0.6000 0.4556 0.4333 0.4222
M 1＝4 0.7640 0.6073 0.5556 0.5427
M 1＝5 1.0400 0.7400 0.7000 0.6733
压缩率指的是：反馈第三指示信息和第四指示信息的反馈量与直接反馈矩阵
Figure PCTCN2019086909-appb-000497
的反馈量之间的比值。
在本发明实施例中，通过
Figure PCTCN2019086909-appb-000498
和
Figure PCTCN2019086909-appb-000499
包含N tN sb个元素的信道矩阵H (r)被压缩为包含2LK个元素的矩阵
Figure PCTCN2019086909-appb-000500
降低了需要反馈的元素的个数，进而降低反馈CSI的开销。其次，通过将矩阵
Figure PCTCN2019086909-appb-000501
作为复数矩阵整体进行压缩，将包含2L×K×N r个元素的矩阵
Figure PCTCN2019086909-appb-000502
Figure PCTCN2019086909-appb-000503
压缩为仅包含L×K×M 1个元素的矩阵
Figure PCTCN2019086909-appb-000504
大幅度降低CSI上报开销，同时不带来明显的性能损失，或者通过对矩阵
Figure PCTCN2019086909-appb-000505
的幅度矩阵
Figure PCTCN2019086909-appb-000506
进行压缩，将包含2L×K×N r个元素的幅度矩阵
Figure PCTCN2019086909-appb-000507
压缩为仅包含L×K×M 1个元素的矩阵
Figure PCTCN2019086909-appb-000508
可在大幅度降低CSI上报开销的同时，不带来明显的性能损失，或者通过对矩阵
Figure PCTCN2019086909-appb-000509
的相位矩阵
Figure PCTCN2019086909-appb-000510
进行压缩，将包含2L×K×N r个元素的相位矩阵
Figure PCTCN2019086909-appb-000511
压缩为仅包含L×K×M 2个元素的矩阵
Figure PCTCN2019086909-appb-000512
可在大幅度降低CSI上报开销的同时，不带来明显的性能损失。另外，第一通信装置将包括第一指示信息和第二指示信息的CSI发送给第二通信装置，第二通信装置可以基于第一指示信息和第二指示信息确定信道矩阵H (r)，以 支持非线性预编码，以及更好的信道信息预测。
基于图1所示的通信系统的结构示意图，本发明实施例提供另一种上报CSI的方法。具体的，本发明实施例的上报CSI的方法可以包括以下步骤：
第二通信装置向第一通信装置发送配置信息，配置信息用于配置第一通信装置上报的CSI；
第一通信装置上报CSI，CSI包括第一指示信息和第二指示信息，第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000513
第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000514
第一通信装置可以将矩阵
Figure PCTCN2019086909-appb-000515
作为复数矩阵整体进行压缩，或者对矩阵
Figure PCTCN2019086909-appb-000516
的幅度矩阵进行压缩，或者对矩阵
Figure PCTCN2019086909-appb-000517
的相位矩阵进行压缩，或者分别对矩阵
Figure PCTCN2019086909-appb-000518
的幅度矩阵以及矩阵
Figure PCTCN2019086909-appb-000519
的相位矩阵进行压缩，或者直接上报矩阵
Figure PCTCN2019086909-appb-000520
的相位矩阵，即不对矩阵
Figure PCTCN2019086909-appb-000521
的相位矩阵进行压缩。
一、在将矩阵
Figure PCTCN2019086909-appb-000522
作为复数矩阵整体进行压缩的情况下，第二指示信息可以包括第三指示信息和第四指示信息。第三指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000523
为L×N sb×M 1的三维矩阵，M 1为正整数。第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000524
为2N r行M 1列矩阵。或者，在将矩阵
Figure PCTCN2019086909-appb-000525
作为复数矩阵整体进行压缩的情况下，第二指示信息可以包括第三指示信息和第四指示信息。第三指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000526
为L×N sb×M 1的三维矩阵，M 1为正整数。第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000527
为2N r行M 1列矩阵。
其中，三维矩阵
Figure PCTCN2019086909-appb-000528
也可以称为三维张量
Figure PCTCN2019086909-appb-000529
在一种实现方式中，在第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000530
的情况下，l取值不同时，
Figure PCTCN2019086909-appb-000531
是相同的矩阵，通过该方案可降低反馈
Figure PCTCN2019086909-appb-000532
的开销。或者，在第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000533
的情况下，k取值不同时，
Figure PCTCN2019086909-appb-000534
是相同的矩阵，通过该方案可降低反馈
Figure PCTCN2019086909-appb-000535
的开销。此时，
Figure PCTCN2019086909-appb-000536
或
Figure PCTCN2019086909-appb-000537
可以由W Rx,1表示。
举例来说，第一通信装置可以按如下方法构造W Rx,1：第一通信装置接收参考信号进行测量，得到
Figure PCTCN2019086909-appb-000538
和
Figure PCTCN2019086909-appb-000539
和
Figure PCTCN2019086909-appb-000540
均为规模L×N sb(即L行N sb列)的矩阵，
Figure PCTCN2019086909-appb-000541
和
Figure PCTCN2019086909-appb-000542
是分别根据第二通信装置两个极化方向天线端口与第一通信装置第r个天线端口之间的信道确定的系数矩阵。然后第一通信装置将2N r个二维矩阵
Figure PCTCN2019086909-appb-000543
Figure PCTCN2019086909-appb-000544
拼接为L×N sb×2N r(即L行N sb列2N r页)的三维矩阵
Figure PCTCN2019086909-appb-000545
并将三维矩阵
Figure PCTCN2019086909-appb-000546
展开为规模为2N r×N sbL(即2N r行N sbL列)的二维矩阵
Figure PCTCN2019086909-appb-000547
第一通信装置令
Figure PCTCN2019086909-appb-000548
Figure PCTCN2019086909-appb-000549
这个规模的复数矩阵，并对
Figure PCTCN2019086909-appb-000550
的协方差矩阵
Figure PCTCN2019086909-appb-000551
进行特征分解，其最大的M 1个特征值对应的特征向量构成压缩矩阵
Figure PCTCN2019086909-appb-000552
可选的，第一通信装置还可以将最大的M 1个特征值对应的特征向量进行量化后构成压缩矩阵
Figure PCTCN2019086909-appb-000553
Figure PCTCN2019086909-appb-000554
首先，第一通信装置可以进行矩阵变形，例如，第一通信装置可以将
Figure PCTCN2019086909-appb-000555
拆分为
Figure PCTCN2019086909-appb-000556
构成2N r个规模为L×N sb(即L行N sb列)的矩阵。第一通信装置把
Figure PCTCN2019086909-appb-000557
放在
Figure PCTCN2019086909-appb-000558
后面，得到新构造的三维矩阵，新构造的三维矩阵中包含N sbL个长度为2N r的向量。第一通信装置将这N sbL个长度为2N r的向量按任意顺序排列，构成规模为2N r行N sbL列的二维矩阵
Figure PCTCN2019086909-appb-000559
进一步的，第一通信装置可以构造压缩矩阵，例如第一通信装置可以令
Figure PCTCN2019086909-appb-000560
Figure PCTCN2019086909-appb-000561
这个规模的复数矩阵，对
Figure PCTCN2019086909-appb-000562
的协方差矩阵
Figure PCTCN2019086909-appb-000563
进行特征分解得到
Figure PCTCN2019086909-appb-000564
其中，D是规模为2N r×2N r(即2N r行2N r列)的矩阵，对角线元素为按降序排列的特征值。V是规模为2N r×2N r(即2N r行2N r列)的矩阵，每一列为一个特征向量，与D中的特征值对应。第一通信装置可以通过W Rx,1＝V(：,[1:M 1])得到压缩矩阵W Rx,1。可选的，第一通信装置还可以将W Rx,1＝V(：,[1:M 1])进行量化后构成压缩矩阵
Figure PCTCN2019086909-appb-000565
在上述W Rx,1的构造方式下，第三指示信息和第四指示信息指示复数矩阵
Figure PCTCN2019086909-appb-000566
由
Figure PCTCN2019086909-appb-000567
得到。
其中，
Figure PCTCN2019086909-appb-000568
为L行N sb列的二维矩阵，由式
Figure PCTCN2019086909-appb-000569
Figure PCTCN2019086909-appb-000570
得到，
Figure PCTCN2019086909-appb-000571
为L行N sb列的二维矩阵，由式
Figure PCTCN2019086909-appb-000572
Figure PCTCN2019086909-appb-000573
得到。
所述
Figure PCTCN2019086909-appb-000574
是长度为2N r的一维向量，由式
Figure PCTCN2019086909-appb-000575
得到，
Figure PCTCN2019086909-appb-000576
表示取矩阵
Figure PCTCN2019086909-appb-000577
固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000578
×为 把一维向量
Figure PCTCN2019086909-appb-000579
当做列向量的矩阵乘运算。
在一种实现方式中，为了进一步降低反馈压缩矩阵W Rx,1的开销，T s个时刻的信道共享压缩矩阵W Rx,1，其中
Figure PCTCN2019086909-appb-000580
在一种实现方式中，压缩矩阵W Rx1和合并系数矩阵
Figure PCTCN2019086909-appb-000581
的上报方式可以包括如下两种：
压缩矩阵W Rx，1和合并系数矩阵
Figure PCTCN2019086909-appb-000582
可以在相同的时刻上报。
压缩矩阵W Rx,1和合并系数矩阵
Figure PCTCN2019086909-appb-000583
可以在不同的时刻上报。例如，压缩矩阵W Rx,1周期上报，合并系数矩阵
Figure PCTCN2019086909-appb-000584
非周期上报，压缩矩阵W Rx,1可以被多个合并系数矩阵
Figure PCTCN2019086909-appb-000585
共用。
二、在对矩阵
Figure PCTCN2019086909-appb-000586
的幅度矩阵进行压缩的情况下，第三指示信息和第四指示信息指示矩阵
Figure PCTCN2019086909-appb-000587
的幅度矩阵
Figure PCTCN2019086909-appb-000588
其中，本申请实施例对幅度矩阵
Figure PCTCN2019086909-appb-000589
的描述方式和对复数矩阵
Figure PCTCN2019086909-appb-000590
的描述类似，具体可参见本申请实施例对复数矩阵
Figure PCTCN2019086909-appb-000591
的描述，本申请实施例不再赘述。
三、在直接上报矩阵
Figure PCTCN2019086909-appb-000592
的相位矩阵的情况下，第二指示信息包括第五指示信息，所述第五指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000593
中元素的相位。
四、在对矩阵
Figure PCTCN2019086909-appb-000594
的相位矩阵进行压缩的情况下，第五指示信息可以包括第六指示信息和第七指示信息。第六指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000595
为L×N sb×M 2(即L行N sb列M 2页)的三维矩阵，M 2为正整数。第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000596
为2N r行M 2列矩阵。
或者，第五指示信息可以包括第六指示信息和第七指示信息。第六指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000597
为L×N sb×M 2的三维矩阵，M 2为正整数。第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000598
为2N r行M 2列矩阵。
其中，三维矩阵
Figure PCTCN2019086909-appb-000599
也可以称为三维张量
Figure PCTCN2019086909-appb-000600
在一种实现方式中，矩阵
Figure PCTCN2019086909-appb-000601
和三维矩阵
Figure PCTCN2019086909-appb-000602
的上报方式可以包括如下两种：
矩阵
Figure PCTCN2019086909-appb-000603
和三维矩阵
Figure PCTCN2019086909-appb-000604
可以在相同的时刻上报。
矩阵
Figure PCTCN2019086909-appb-000605
和三维矩阵
Figure PCTCN2019086909-appb-000606
可以在不同的时刻上报。例如，矩阵
Figure PCTCN2019086909-appb-000607
周期上报，三维矩阵
Figure PCTCN2019086909-appb-000608
非周期上报，矩阵
Figure PCTCN2019086909-appb-000609
可以被多个三维矩阵
Figure PCTCN2019086909-appb-000610
共用。
在一种实现方式中，矩阵
Figure PCTCN2019086909-appb-000611
和三维矩阵
Figure PCTCN2019086909-appb-000612
的上报方式可以包括如下两种：
矩阵
Figure PCTCN2019086909-appb-000613
和三维矩阵
Figure PCTCN2019086909-appb-000614
可以在相同的时刻上报。
矩阵
Figure PCTCN2019086909-appb-000615
和三维矩阵
Figure PCTCN2019086909-appb-000616
可以在不同的时刻上报。例如，矩阵
Figure PCTCN2019086909-appb-000617
周期上报，三维矩 阵
Figure PCTCN2019086909-appb-000618
非周期上报，矩阵
Figure PCTCN2019086909-appb-000619
可以被多个三维矩阵
Figure PCTCN2019086909-appb-000620
共用。
其中，本申请实施例对相位矩阵
Figure PCTCN2019086909-appb-000621
的描述方式和对复数矩阵
Figure PCTCN2019086909-appb-000622
的描述类似，具体可参见本申请实施例对复数矩阵
Figure PCTCN2019086909-appb-000623
的描述，本申请实施例不再赘述。
五、在分别对矩阵
Figure PCTCN2019086909-appb-000624
的幅度矩阵和矩阵
Figure PCTCN2019086909-appb-000625
的相位矩阵进行压缩的情况下，第二指示信息包括第三指示信息、第四指示信息以及第五指示信息。
其中，本申请实施例中的第三指示信息和第四指示信息可参见本申请实施例对幅度矩阵
Figure PCTCN2019086909-appb-000626
进行压缩的情况下的具体描述，另外，本申请实施例中的第五指示信息可参见本申请实施例对相位矩阵
Figure PCTCN2019086909-appb-000627
进行压缩的情况下的具体描述，本申请实施例不再赘述。
在一种实现方式中，幅度矩阵
Figure PCTCN2019086909-appb-000628
的压缩率由M 1决定，相位矩阵
Figure PCTCN2019086909-appb-000629
的压缩率由M 2决定，可通过设定不同的M 1和M 2实现矩阵
Figure PCTCN2019086909-appb-000630
幅度和相位的不同压缩率，优选的，M 1小于M 2。
在一种实现方式中，在第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000631
的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
Figure PCTCN2019086909-appb-000632
则l′＝l″。
本申请实施例中，矩阵
Figure PCTCN2019086909-appb-000633
的各列向量为对矩阵H (r′)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000634
的各列向量为对矩阵H (r″)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000635
的各列向量和矩阵
Figure PCTCN2019086909-appb-000636
的各列向量可以来源于同一组预定义的DFT基底，若矩阵
Figure PCTCN2019086909-appb-000637
和矩阵
Figure PCTCN2019086909-appb-000638
选取了同一个DFT基底，即
Figure PCTCN2019086909-appb-000639
Figure PCTCN2019086909-appb-000640
则将该DFT基底在矩阵
Figure PCTCN2019086909-appb-000641
和矩阵
Figure PCTCN2019086909-appb-000642
中放在相同的位置，即l′＝l″，这样矩阵H (r′)和矩阵H (r″)压缩后的矩阵
Figure PCTCN2019086909-appb-000643
和矩阵
Figure PCTCN2019086909-appb-000644
的第l′＝l″行对应同一个基底。上述对于矩阵
Figure PCTCN2019086909-appb-000645
的约束可提高向量
Figure PCTCN2019086909-appb-000646
之间的相关性，从而提升压缩效果。
在一种实现方式中，第三指示信息包括用于指示
Figure PCTCN2019086909-appb-000647
中元素
Figure PCTCN2019086909-appb-000648
的X 1 l,k,g个比特，X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000649
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000650
中第c行第d列的元素
Figure PCTCN2019086909-appb-000651
的Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000652
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000653
中第c行第d列的元素
Figure PCTCN2019086909-appb-000654
的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000655
所述Y 1 l,c,d大于等于X 1 l,k,g。可选的， 若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000656
Z 1 k,c,d大于等于X 1 l,k,g。
本申请实施例中，由于所述第四指示信息指示的矩阵
Figure PCTCN2019086909-appb-000657
或
Figure PCTCN2019086909-appb-000658
Figure PCTCN2019086909-appb-000659
为对所述矩阵
Figure PCTCN2019086909-appb-000660
压缩的基底矩阵，所述第三指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000661
为对所述矩阵
Figure PCTCN2019086909-appb-000662
压缩的合并系数矩阵，在选择相同的压缩基底个数M 1的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000663
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000664
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 1，有利于降低上报CSI的开销。
在一种实现方式中，第六指示信息包括用于指示
Figure PCTCN2019086909-appb-000665
中元素
Figure PCTCN2019086909-appb-000666
的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000667
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000668
中第c行第d列的元素
Figure PCTCN2019086909-appb-000669
的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000670
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000671
中第c行第d列的元素
Figure PCTCN2019086909-appb-000672
的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000673
所述Y 2 l,c,d大于等于X 2 l,k,f。可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000674
Z 2 k,c,d大于等于X 2 l,k,f。
本申请实施例中，由于所述第七指示信息指示的矩阵
Figure PCTCN2019086909-appb-000675
或
Figure PCTCN2019086909-appb-000676
Figure PCTCN2019086909-appb-000677
为对所述矩阵
Figure PCTCN2019086909-appb-000678
压缩的基底矩阵，所述第六指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000679
为对所述矩阵
Figure PCTCN2019086909-appb-000680
压缩的合并系数矩阵，在选择相同的压缩基底个数M 2的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000681
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000682
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 2，有利于降低上报CSI的开销。
若第一通信装置直接反馈矩阵
Figure PCTCN2019086909-appb-000683
反馈量为
Figure PCTCN2019086909-appb-000684
Figure PCTCN2019086909-appb-000685
其中
Figure PCTCN2019086909-appb-000686
为反馈每一个参数的比特数。在第三指示信息和第四指示信息指示复数矩阵
Figure PCTCN2019086909-appb-000687
的实施方法中，压缩矩阵W Rx,1为T s个时刻公用，反馈量包括两部分：反馈第四指示信息的反馈量
Figure PCTCN2019086909-appb-000688
以及反馈第三指示信息的反馈量
Figure PCTCN2019086909-appb-000689
Figure PCTCN2019086909-appb-000690
其中
Figure PCTCN2019086909-appb-000691
和
Figure PCTCN2019086909-appb-000692
为反馈每一个参数的比特数。
在本发明实施例中，通过
Figure PCTCN2019086909-appb-000693
包含N tN sb个元素的信道矩阵H (r)被压缩为包含2LN sb个元素的矩阵
Figure PCTCN2019086909-appb-000694
降低了需要反馈的元素的个数，进而降低反馈CSI的开销。其次，通过将矩阵
Figure PCTCN2019086909-appb-000695
作为复数矩阵整体进行压缩，将包含2L×N sb×N r个元素的矩阵
Figure PCTCN2019086909-appb-000696
Figure PCTCN2019086909-appb-000697
压缩为仅包含L×N sb×M 1个元素的矩阵
Figure PCTCN2019086909-appb-000698
大幅度降低CSI上报开销，同时不带来明显的性能损失，或者通过对矩阵
Figure PCTCN2019086909-appb-000699
的幅度矩阵
Figure PCTCN2019086909-appb-000700
进行压缩，将包含2L×N sb×N r个元素的幅度矩阵
Figure PCTCN2019086909-appb-000701
压缩为仅包含L×N sb×M 1个元素的矩阵
Figure PCTCN2019086909-appb-000702
可在大幅度降低CSI上报开销的同时，不带来明显的性能损失，或者通过对矩阵
Figure PCTCN2019086909-appb-000703
的相位矩阵
Figure PCTCN2019086909-appb-000704
进行压缩，将包含2L×N sb×N r个元素的相位矩阵
Figure PCTCN2019086909-appb-000705
Figure PCTCN2019086909-appb-000706
压缩为仅包含L×N sb×M 2个元素的矩阵
Figure PCTCN2019086909-appb-000707
可在大幅度降低CSI上报开销的同时，不带来明显的性能损失。另外，第一通信装置将包括第一指示信息和第二指示信息的CSI发送给第二通信装置，第二通信装置可以基于第一指示信息和第二指示信息确定信道矩阵H (r)，以支持非线性预编码，以及更好的信道信息预测。
基于图1所示的通信系统的结构示意图，本发明实施例提供一种上报CSI的方法。具体的，本发明实施例的上报CSI的方法可以包括以下步骤：
第二通信装置向第一通信装置发送配置信息，所述配置信息用于配置所述第一通信装置上报的CSI；
所述第一通信装置上报CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000708
所述矩阵
Figure PCTCN2019086909-appb-000709
为2L行P列的矩阵，L为正整数，2L小于N t；
所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000710
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000711
和矩阵
Figure PCTCN2019086909-appb-000712
Figure PCTCN2019086909-appb-000713
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000714
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
在一种实现方式中，所述第二指示信息包括第三指示信息和第四指示信息，所述第三指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000715
为L行P列M 1页的三维矩阵，M 1为正整数；
所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000716
为2N r行M 1列矩阵，或，所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000717
为2N r行M 1列矩阵。
在一种实现方式中，所述矩阵
Figure PCTCN2019086909-appb-000718
其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
所述
Figure PCTCN2019086909-appb-000719
为L行P列2N r页的三维矩阵，
Figure PCTCN2019086909-appb-000720
表示取矩阵
Figure PCTCN2019086909-appb-000721
固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
Figure PCTCN2019086909-appb-000722
本申请实施例可将
Figure PCTCN2019086909-appb-000723
拆分为
Figure PCTCN2019086909-appb-000724
构成2N r个规模为L行P列的矩阵，函数g(r,i)实现这2N r个矩阵到三维矩阵
Figure PCTCN2019086909-appb-000725
各页之间的映射。
若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000726
则所述
Figure PCTCN2019086909-appb-000727
是由式
Figure PCTCN2019086909-appb-000728
Figure PCTCN2019086909-appb-000729
得到，或若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000730
则所述
Figure PCTCN2019086909-appb-000731
是由式
Figure PCTCN2019086909-appb-000732
得到；
Figure PCTCN2019086909-appb-000733
表示取矩阵
Figure PCTCN2019086909-appb-000734
固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000735
表示取矩阵
Figure PCTCN2019086909-appb-000736
固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
Figure PCTCN2019086909-appb-000737
×为把一维向量
Figure PCTCN2019086909-appb-000738
当做列向量的矩阵乘运算。
在一种实现方式中，所述矩阵
Figure PCTCN2019086909-appb-000739
的幅度矩阵
Figure PCTCN2019086909-appb-000740
所述
Figure PCTCN2019086909-appb-000741
的第l行第k列的元素是根据所述
Figure PCTCN2019086909-appb-000742
的第l行第k列的元素的幅度确定的值，例如所述
Figure PCTCN2019086909-appb-000743
的第l行第k列的元素是所述
Figure PCTCN2019086909-appb-000744
的第l行第k列的元素的幅度或幅度的平方；
其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
所述
Figure PCTCN2019086909-appb-000745
为L行P列2N r页的三维矩阵，
Figure PCTCN2019086909-appb-000746
表示取矩阵
Figure PCTCN2019086909-appb-000747
固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
Figure PCTCN2019086909-appb-000748
本申请实施例可将
Figure PCTCN2019086909-appb-000749
拆分为
Figure PCTCN2019086909-appb-000750
构成2N r个规模为L行P列的矩阵，函数g(r,i)实现这2N r个矩阵到三维矩阵
Figure PCTCN2019086909-appb-000751
各页之间的映射。
若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000752
则所述
Figure PCTCN2019086909-appb-000753
是由式
Figure PCTCN2019086909-appb-000754
Figure PCTCN2019086909-appb-000755
得到，或若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000756
则所述
Figure PCTCN2019086909-appb-000757
是由式
Figure PCTCN2019086909-appb-000758
得到；
Figure PCTCN2019086909-appb-000759
表示取矩阵
Figure PCTCN2019086909-appb-000760
固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000761
表示取矩阵
Figure PCTCN2019086909-appb-000762
固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
Figure PCTCN2019086909-appb-000763
×为把一维向量
Figure PCTCN2019086909-appb-000764
当做列向量的矩阵乘运算。
在一种实现方式中，所述第二指示信息还包括第五指示信息，所述第五指示信息用于指示所述
Figure PCTCN2019086909-appb-000765
中元素的相位。
在一种实现方式中，所述第五指示信息用于指示所述
Figure PCTCN2019086909-appb-000766
中元素的相位，包括：
所述第五指示信息包括第六指示信息和第七指示信息，所述第六指示信息用于指示三维矩阵
Figure PCTCN2019086909-appb-000767
为L行P列M 2页的三维矩阵，M 2为正整数；
所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000768
为2N r行M 2列矩阵，或，所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000769
为2N r行M 2列矩阵；
所述矩阵
Figure PCTCN2019086909-appb-000770
的相位矩阵
Figure PCTCN2019086909-appb-000771
所述
Figure PCTCN2019086909-appb-000772
的第l行第k列的元素是根据所述
Figure PCTCN2019086909-appb-000773
的第l行第k列的元素的相位确定的值，例如所述
Figure PCTCN2019086909-appb-000774
的第l行第k列的元素是所述
Figure PCTCN2019086909-appb-000775
的第l行第k列的元素的相位或相位的平方；
其中，f(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，f(r′,i′)≠f(r″,i″)；
所述
Figure PCTCN2019086909-appb-000776
为L行P列2N r页的三维矩阵，
Figure PCTCN2019086909-appb-000777
表示取矩阵
Figure PCTCN2019086909-appb-000778
固定第三个索引为f(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
Figure PCTCN2019086909-appb-000779
本申请实施例可将
Figure PCTCN2019086909-appb-000780
拆分为
Figure PCTCN2019086909-appb-000781
构成2N r个规模为L行P列的矩阵，函数f(r,i)实现这2N r个矩阵到三维矩阵
Figure PCTCN2019086909-appb-000782
各页之间的映射。
若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000783
则所述
Figure PCTCN2019086909-appb-000784
是由式
Figure PCTCN2019086909-appb-000785
Figure PCTCN2019086909-appb-000786
得到，或若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000787
则所述
Figure PCTCN2019086909-appb-000788
是由式
Figure PCTCN2019086909-appb-000789
得到；
Figure PCTCN2019086909-appb-000790
表示取矩阵
Figure PCTCN2019086909-appb-000791
固定前两个索引长度M 2的一维向量，所述长度M 2的一维向量中第m个元素为
Figure PCTCN2019086909-appb-000792
表示取矩阵
Figure PCTCN2019086909-appb-000793
固定前两个索引长度2N r的一维向量，所述长度2N r 的一维向量第f(r,i)个元素为
Figure PCTCN2019086909-appb-000794
×为把一维向量
Figure PCTCN2019086909-appb-000795
当做列向量的矩阵乘运算。
在一种实现方式中，在所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000796
的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
Figure PCTCN2019086909-appb-000797
则l′＝l″。
在该技术方案中，矩阵
Figure PCTCN2019086909-appb-000798
的各列向量为对矩阵H (r′)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000799
的各列向量为对矩阵H (r″)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000800
的各列向量和矩阵
Figure PCTCN2019086909-appb-000801
的各列向量可以来源于同一组预定义的DFT基底，若矩阵
Figure PCTCN2019086909-appb-000802
和矩阵
Figure PCTCN2019086909-appb-000803
选取了同一个DFT基底，即
Figure PCTCN2019086909-appb-000804
Figure PCTCN2019086909-appb-000805
则将该DFT基底在矩阵
Figure PCTCN2019086909-appb-000806
和矩阵
Figure PCTCN2019086909-appb-000807
中放在相同的位置，即l′＝l″，这样矩阵H (r′)和矩阵H (r″)压缩后的矩阵
Figure PCTCN2019086909-appb-000808
和矩阵
Figure PCTCN2019086909-appb-000809
的第l′＝l″行对应同一个基底。上述对于矩阵
Figure PCTCN2019086909-appb-000810
的约束可提高向量
Figure PCTCN2019086909-appb-000811
之间的相关性，从而提升压缩效果。
在一种实现方式中，在所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000812
和矩阵
Figure PCTCN2019086909-appb-000813
的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
Figure PCTCN2019086909-appb-000814
则l′＝l″，对于任意r′,r″∈{1,2,…,N r}，k′,k″∈{1,2,…,K}，若r′≠r″且
Figure PCTCN2019086909-appb-000815
则k′＝k″。
在该技术方案中，矩阵
Figure PCTCN2019086909-appb-000816
的各列向量为对矩阵H (r′)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000817
的各列向量为对矩阵H (r″)压缩的基底，矩阵
Figure PCTCN2019086909-appb-000818
的各列向量和矩阵
Figure PCTCN2019086909-appb-000819
的各列向量可以来源于同一组预定义的DFT基底，若矩阵
Figure PCTCN2019086909-appb-000820
和矩阵
Figure PCTCN2019086909-appb-000821
选取了同一个DFT基底，即
Figure PCTCN2019086909-appb-000822
Figure PCTCN2019086909-appb-000823
则将该DFT基底在矩阵
Figure PCTCN2019086909-appb-000824
和矩阵
Figure PCTCN2019086909-appb-000825
中放在相同的位置，即l′＝l″，这样矩阵H (r′)和矩阵H (r″)压缩后的矩阵
Figure PCTCN2019086909-appb-000826
和矩阵
Figure PCTCN2019086909-appb-000827
的第l′＝l″行对应同一个基底；类似地，矩阵
Figure PCTCN2019086909-appb-000828
和矩阵
Figure PCTCN2019086909-appb-000829
的约束使得矩阵
Figure PCTCN2019086909-appb-000830
和矩阵
Figure PCTCN2019086909-appb-000831
的同一列尽量对应同一基底。上述对于矩阵
Figure PCTCN2019086909-appb-000832
和矩阵
Figure PCTCN2019086909-appb-000833
的约束可提高向量
Figure PCTCN2019086909-appb-000834
之间的相关性，从而提升压缩效果。
在一种实现方式中，所述第三指示信息包括用于指示
Figure PCTCN2019086909-appb-000835
中元素
Figure PCTCN2019086909-appb-000836
的X 1 l,k,g个比特， X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000837
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000838
中第c行第d列的元素
Figure PCTCN2019086909-appb-000839
的Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000840
则所述第四指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000841
中第c行第d列的元素
Figure PCTCN2019086909-appb-000842
的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000843
所述Y 1 l,c,d大于等于X 1 l,k,g。可选的，若所述第四指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000844
Z 1 k,c,d大于等于X 1 l,k,g。
在该技术方案中，由于所述第四指示信息指示的矩阵
Figure PCTCN2019086909-appb-000845
或
Figure PCTCN2019086909-appb-000846
Figure PCTCN2019086909-appb-000847
为对所述矩阵
Figure PCTCN2019086909-appb-000848
压缩的基底矩阵，所述第三指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000849
为对所述矩阵
Figure PCTCN2019086909-appb-000850
压缩的合并系数矩阵，在选择相同的压缩基底个数M 1的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000851
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000852
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 1，有利于降低上报CSI的开销。
在一种实现方式中，所述第六指示信息包括用于指示
Figure PCTCN2019086909-appb-000853
中元素
Figure PCTCN2019086909-appb-000854
的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000855
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000856
中第c行第d列的元素
Figure PCTCN2019086909-appb-000857
的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000858
则所述第七指示信息包括用于指示矩阵
Figure PCTCN2019086909-appb-000859
中第c行第d列的元素
Figure PCTCN2019086909-appb-000860
的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000861
所述Y 2 l,c,d大于等于X 2 l,k,f。可选的，若所述第七指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000862
Z 2 k,c,d大于等于X 2 l,k,f。
在该技术方案中，由于所述第七指示信息指示的矩阵
Figure PCTCN2019086909-appb-000863
或
Figure PCTCN2019086909-appb-000864
Figure PCTCN2019086909-appb-000865
为对所述矩阵
Figure PCTCN2019086909-appb-000866
压缩的基底矩阵，所述第六指示信息指示的三维矩阵
Figure PCTCN2019086909-appb-000867
为对所述矩阵
Figure PCTCN2019086909-appb-000868
压缩的合并系数矩阵，在选择相同的压缩基底个数M 2的情况下，压缩基底的元素的量化精度不低于合并系数的精度有利于降低压缩损失，即压缩后的信道矩阵更匹配真实的信道矩阵。当
Figure PCTCN2019086909-appb-000869
即l取值不同时基底矩阵相同，或当
Figure PCTCN2019086909-appb-000870
即k取值不同时基底矩阵相同时，提高压缩基底元素的量化精度更有必要。换另外一个角度说，在压缩损失一定的情况下，提高压缩基底元素的量化精度有利于减少选择的压缩基底个数M 2，有利于降低上报CSI的开销。
在该技术方案中，矩阵
Figure PCTCN2019086909-appb-000871
的幅度矩阵
Figure PCTCN2019086909-appb-000872
具有较强的相关性，相位矩阵
Figure PCTCN2019086909-appb-000873
相关性往往弱于幅度矩阵
Figure PCTCN2019086909-appb-000874
故幅度矩阵
Figure PCTCN2019086909-appb-000875
相较相位矩阵
Figure PCTCN2019086909-appb-000876
往往易于压缩。幅度矩阵
Figure PCTCN2019086909-appb-000877
的压缩率由M 1决定，相位矩阵
Figure PCTCN2019086909-appb-000878
的压缩率由M 2决定，可通过设定不同的M 1和M 2实现矩阵
Figure PCTCN2019086909-appb-000879
幅度和相位的不同压缩率，优选的，M 1小于M 2。
图4是本申请实施例提供的另一种通信系统的结构示意图，该通信系统包括第一通信装置401和第二通信装置402，第一通信装置401可以包括第一接收单元4011和第一发送单元4012，第二通信装置402可以包括第二发送单元4021和第二接收单元4022。
第二发送单元4021，用于向第一接收单元4011发送配置信息，配置信息用于配置第一发送单元4012上报的CSI；
第一发送单元4012，用于向第二接收单元4022上报CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000880
所述矩阵
Figure PCTCN2019086909-appb-000881
为2L行P列的矩阵，L为正整数，2L小于N t；
所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000882
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000883
和矩阵
Figure PCTCN2019086909-appb-000884
Figure PCTCN2019086909-appb-000885
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000886
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
需要说明的是，图4对应的实施例中未提及的内容以及各个单元执行步骤的具体实现方式可参见图1和图2所示实施例以及前述内容，这里不再赘述。
在一种实现方式中，图4中的第一接收单元4011和第一发送单元4012所实现的相关功能可以结合处理器、接收器以及发射器来实现。图5是本发明实施例提供的一种终端设备的结构示意图，接收器503以及发射器504可以调用存储器502中存储的程序代码以执行以下操作：
接收器503接收来自于第二通信装置的配置信息，所述配置信息用于配置所述发射器 上报的CSI；
发射器504上报CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述终端设备第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000887
所述矩阵
Figure PCTCN2019086909-appb-000888
为2L行P列的矩阵，L为正整数，2L小于N t；
所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000889
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000890
和矩阵
Figure PCTCN2019086909-appb-000891
Figure PCTCN2019086909-appb-000892
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000893
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
需要说明的是，图5对应的实施例中未提及的内容以及各个器件执行步骤的具体实现方式可参见图1和图2所示实施例以及前述内容，这里不再赘述。
在一种实现方式中，图4中的第二发送单元4021和第二接收单元4022所实现的相关功能可以结合处理器、接收器以及发射器来实现。图6是本发明实施例提供的一种网络设备的结构示意图，接收器603以及发射器604可以调用存储器602中存储的程序代码以执行以下操作：
发射器604向第一通信装置发送配置信息，所述配置信息用于指示所述第一通信装置上报CSI；
接收器603接收来自于所述第一通信装置的CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
所述第二指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000894
所述矩阵
Figure PCTCN2019086909-appb-000895
为2L行P列的矩阵，L为正整数，2L小于N t；
所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000896
为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
Figure PCTCN2019086909-appb-000897
和矩阵
Figure PCTCN2019086909-appb-000898
Figure PCTCN2019086909-appb-000899
为N t行2L列的矩阵，
Figure PCTCN2019086909-appb-000900
为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
需要说明的是，图6对应的实施例中未提及的内容以及各个器件执行步骤的具体实现方式可参见图1和图2所示实施例以及前述内容，这里不再赘述。
上述各个实施例中处理器可以是通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本发明实施例中的公开的各方法、步骤及逻辑 框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本发明实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存取存储器(Random Access Memory，RAM)、闪存、只读存储器(Read-Only Memory，ROM)、可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的指令，结合其硬件完成上述方法的步骤。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到的变化或替换，都应涵盖在本申请的保护范围之内，因此本申请的保护范围应以权利要求的保护范围为准。

Claims (56)

  1. 一种上报信道状态信息CSI的方法，其特征在于，包括：
    第一通信装置接收来自于第二通信装置的配置信息，所述配置信息用于配置所述第一通信装置上报的CSI；
    所述第一通信装置上报CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
    所述第二指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100001
    所述矩阵
    Figure PCTCN2019086909-appb-100002
    为2L行P列的矩阵，L为正整数，2L小于N t；
    所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100003
    为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100004
    和矩阵
    Figure PCTCN2019086909-appb-100005
    Figure PCTCN2019086909-appb-100006
    为N t行2L列的矩阵，
    Figure PCTCN2019086909-appb-100007
    为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
  2. 如权利要求1所述的方法，其特征在于，所述第二指示信息包括第三指示信息和第四指示信息，所述第三指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100008
    为L行P列M 1页的三维矩阵，M 1为正整数；
    所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100009
    为2N r行M 1列矩阵，或，所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100010
    为2N r行M 1列矩阵。
  3. 如权利要求2所述的方法，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100011
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100012
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100013
    表示取矩阵
    Figure PCTCN2019086909-appb-100014
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100015
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100016
    则所述
    Figure PCTCN2019086909-appb-100017
    是由式
    Figure PCTCN2019086909-appb-100018
    Figure PCTCN2019086909-appb-100019
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100020
    则所述
    Figure PCTCN2019086909-appb-100021
    是由式
    Figure PCTCN2019086909-appb-100022
    得到；
    Figure PCTCN2019086909-appb-100023
    表示取矩阵
    Figure PCTCN2019086909-appb-100024
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100025
    表示取矩阵
    Figure PCTCN2019086909-appb-100026
    固定前两个索引长度2N r的一维向量，所述长度2N r 的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100027
    ×为把一维向量
    Figure PCTCN2019086909-appb-100028
    当做列向量的矩阵乘运算。
  4. 如权利要求2所述的方法，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100029
    的幅度矩阵
    Figure PCTCN2019086909-appb-100030
    Figure PCTCN2019086909-appb-100031
    所述
    Figure PCTCN2019086909-appb-100032
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100033
    的第l行第k列的元素的幅度确定的值；
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100034
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100035
    表示取矩阵
    Figure PCTCN2019086909-appb-100036
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100037
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100038
    则所述
    Figure PCTCN2019086909-appb-100039
    是由式
    Figure PCTCN2019086909-appb-100040
    Figure PCTCN2019086909-appb-100041
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100042
    则所述
    Figure PCTCN2019086909-appb-100043
    是由式
    Figure PCTCN2019086909-appb-100044
    得到；
    Figure PCTCN2019086909-appb-100045
    表示取矩阵
    Figure PCTCN2019086909-appb-100046
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100047
    表示取矩阵
    Figure PCTCN2019086909-appb-100048
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100049
    ×为把一维向量
    Figure PCTCN2019086909-appb-100050
    当做列向量的矩阵乘运算。
  5. 如权利要求4所述的方法，其特征在于，所述
    Figure PCTCN2019086909-appb-100051
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100052
    的第l行第k列的元素的幅度或幅度的平方。
  6. 如权利要求1-5任一项所述的方法，其特征在于，所述第二指示信息包括第五指示信息，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100053
    中元素的相位。
  7. 如权利要求6所述的方法，其特征在于，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100054
    中元素的相位，包括：
    所述第五指示信息包括第六指示信息和第七指示信息，所述第六指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100055
    为L行P列M 2页的三维矩阵，M 2为正整数；
    所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100056
    为2N r行M 2列矩阵，或，所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100057
    为2N r行M 2列矩阵；
    所述矩阵
    Figure PCTCN2019086909-appb-100058
    的相位矩阵
    Figure PCTCN2019086909-appb-100059
    所述
    Figure PCTCN2019086909-appb-100060
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100061
    的第l行第k列的元素的相位确定的值；
    其中，f(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，f(r′,i′)≠f(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100062
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100063
    表示取矩阵
    Figure PCTCN2019086909-appb-100064
    固定第三个索引为f(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100065
    若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100066
    则所述
    Figure PCTCN2019086909-appb-100067
    是由式
    Figure PCTCN2019086909-appb-100068
    Figure PCTCN2019086909-appb-100069
    得到，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100070
    则所述
    Figure PCTCN2019086909-appb-100071
    是由式
    Figure PCTCN2019086909-appb-100072
    得到；
    Figure PCTCN2019086909-appb-100073
    表示取矩阵
    Figure PCTCN2019086909-appb-100074
    固定前两个索引长度M 2的一维向量，所述长度M 2的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100075
    表示取矩阵
    Figure PCTCN2019086909-appb-100076
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第f(r,i)个元素为
    Figure PCTCN2019086909-appb-100077
    ×为把一维向量
    Figure PCTCN2019086909-appb-100078
    当做列向量的矩阵乘运算。
  8. 如权利要求7所述的方法，其特征在于，所述
    Figure PCTCN2019086909-appb-100079
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100080
    的第l行第k列的元素的相位或相位的平方。
  9. 如权利要求1-8任一项所述的方法，其特征在于，在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100081
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100082
    Figure PCTCN2019086909-appb-100083
    则l′＝l″；或
    在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100084
    和矩阵
    Figure PCTCN2019086909-appb-100085
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100086
    则l′＝l″，对于任意r′,r″∈{1,2,…,N r}，k′,k″∈{1,2,…,K}，若r′≠r″且
    Figure PCTCN2019086909-appb-100087
    则k′＝k″。
  10. 如权利要求2-9任一项所述的方法，其特征在于，所述第三指示信息包括用于指示
    Figure PCTCN2019086909-appb-100088
    中元素
    Figure PCTCN2019086909-appb-100089
    的X 1 l，k，g个比特，X 1 l，k，g是正整数，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100090
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100091
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100092
    的Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100093
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100094
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100095
    的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
  11. 如权利要求10所述的方法，其特征在于，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100096
    则Y 1 l,c,d大于等于X 1 l,k,g，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100097
    则Z 1 k,c,d大于等于X 1 l,k,g。
  12. 如权利要求7-11任一项所述的方法，其特征在于，所述第六指示信息包括用于指 示
    Figure PCTCN2019086909-appb-100098
    中元素
    Figure PCTCN2019086909-appb-100099
    的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100100
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100101
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100102
    的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100103
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100104
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100105
    的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
  13. 如权利要求12所述的方法，其特征在于，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100106
    则Y 2 l,c,d大于等于X 2 l,k,f，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100107
    则Z 2 k,c,d大于等于X 2 l,k,f。
  14. 一种获取CSI的方法，其特征在于，包括：
    第二通信装置向第一通信装置发送配置信息，所述配置信息用于配置所述第一通信装置上报的CSI；
    所述第二通信装置接收CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
    所述第二指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100108
    所述矩阵
    Figure PCTCN2019086909-appb-100109
    为2L行P列的矩阵，L为正整数，2L小于N t；
    所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100110
    为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100111
    和矩阵
    Figure PCTCN2019086909-appb-100112
    Figure PCTCN2019086909-appb-100113
    为N t行2L列的矩阵，
    Figure PCTCN2019086909-appb-100114
    为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
  15. 如权利要求14所述的方法，其特征在于，所述第二指示信息包括第三指示信息和第四指示信息，所述第三指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100115
    为L行P列M 1页的三维矩阵，M 1为正整数；
    所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100116
    为2N r行M 1列矩阵，或，所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100117
    为2N r行M 1列矩阵。
  16. 如权利要求15所述的方法，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100118
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}和，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100119
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100120
    表示取矩阵
    Figure PCTCN2019086909-appb-100121
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100122
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100123
    则所述
    Figure PCTCN2019086909-appb-100124
    是由式
    Figure PCTCN2019086909-appb-100125
    Figure PCTCN2019086909-appb-100126
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100127
    则所述
    Figure PCTCN2019086909-appb-100128
    是由式
    Figure PCTCN2019086909-appb-100129
    得到；
    Figure PCTCN2019086909-appb-100130
    表示取矩阵
    Figure PCTCN2019086909-appb-100131
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100132
    表示取矩阵
    Figure PCTCN2019086909-appb-100133
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100134
    ×为把一维向量
    Figure PCTCN2019086909-appb-100135
    当做列向量的矩阵乘运算。
  17. 如权利要求15所述的方法，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100136
    的幅度矩阵
    Figure PCTCN2019086909-appb-100137
    Figure PCTCN2019086909-appb-100138
    所述
    Figure PCTCN2019086909-appb-100139
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100140
    的第l行第k列的元素的幅度确定的值；
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100141
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100142
    表示取矩阵
    Figure PCTCN2019086909-appb-100143
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行k列元素为
    Figure PCTCN2019086909-appb-100144
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100145
    则所述
    Figure PCTCN2019086909-appb-100146
    是由式
    Figure PCTCN2019086909-appb-100147
    Figure PCTCN2019086909-appb-100148
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100149
    则所述
    Figure PCTCN2019086909-appb-100150
    是由式
    Figure PCTCN2019086909-appb-100151
    得到；
    Figure PCTCN2019086909-appb-100152
    表示取矩阵
    Figure PCTCN2019086909-appb-100153
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100154
    表示取矩阵
    Figure PCTCN2019086909-appb-100155
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100156
    ×为把一维向量
    Figure PCTCN2019086909-appb-100157
    当做列向量的矩阵乘运算。
  18. 如权利要求17所述的方法，其特征在于，所述
    Figure PCTCN2019086909-appb-100158
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100159
    的第l行第k列的元素的幅度或幅度的平方。
  19. 如权利要求14-18任一项所述的方法，其特征在于，所述第二指示信息包括第五指示信息，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100160
    中元素的相位。
  20. 如权利要求19所述的方法，其特征在于，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100161
    中元素的相位，包括：
    所述第五指示信息包括第六指示信息和第七指示信息，所述第六指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100162
    为L行P列M 2页的三维矩阵，M 2为正整数；
    所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100163
    为2N r行M 2列矩阵，或，所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100164
    为2N r行M 2列矩阵；
    所述矩阵
    Figure PCTCN2019086909-appb-100165
    的相位矩阵
    Figure PCTCN2019086909-appb-100166
    所述
    Figure PCTCN2019086909-appb-100167
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100168
    的第l行第k列的元素的相位确定的值；
    其中，f(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，
    f(r′,i′)≠f(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100169
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100170
    表示取矩阵
    Figure PCTCN2019086909-appb-100171
    固定第三个索引为f(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100172
    若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100173
    则所述
    Figure PCTCN2019086909-appb-100174
    是由式
    Figure PCTCN2019086909-appb-100175
    Figure PCTCN2019086909-appb-100176
    得到，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100177
    则所述
    Figure PCTCN2019086909-appb-100178
    是由式
    Figure PCTCN2019086909-appb-100179
    得到；
    Figure PCTCN2019086909-appb-100180
    表示取矩阵
    Figure PCTCN2019086909-appb-100181
    固定前两个索引长度M 2的一维向量，所述长度M 2的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100182
    表示取矩阵
    Figure PCTCN2019086909-appb-100183
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第f(r,i)个元素为
    Figure PCTCN2019086909-appb-100184
    ×为把一维向量
    Figure PCTCN2019086909-appb-100185
    当做列向量的矩阵乘运算。
  21. 如权利要求20所述的方法，其特征在于，所述
    Figure PCTCN2019086909-appb-100186
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100187
    的第l行第k列的元素的相位或相位的平方。
  22. 如权利要求14-21任一项所述的方法，其特征在于，在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100188
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100189
    Figure PCTCN2019086909-appb-100190
    则l′＝l″；或
    在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100191
    和矩阵
    Figure PCTCN2019086909-appb-100192
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100193
    则l′＝l″，对于任意r′,r″∈{1,2,…,Nr}，k′,k″∈{1,2,…,K}，若r′≠r″且
    Figure PCTCN2019086909-appb-100194
    则k′＝k″。
  23. 如权利要求15-22任一项所述的方法，其特征在于，所述第三指示信息包括用于指示
    Figure PCTCN2019086909-appb-100195
    中元素
    Figure PCTCN2019086909-appb-100196
    的X 1 l,k,g个比特，X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100197
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100198
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100199
    的 Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100200
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100201
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100202
    的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
  24. 如权利要求23所述的方法，其特征在于，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100203
    则Y 1 l,c,d大于等于X 1 l,k,g，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100204
    则Z 1 k,c,d大于等于X 1 l,k,g。
  25. 如权利要求20-24任一项所述的方法，其特征在于，所述第六指示信息包括用于指示
    Figure PCTCN2019086909-appb-100205
    中元素
    Figure PCTCN2019086909-appb-100206
    的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100207
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100208
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100209
    的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100210
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100211
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100212
    的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
  26. 如权利要求25所述的方法，其特征在于，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100213
    则Y 2 l,c,d大于等于X 2 l,k,f，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100214
    则Z 2 k,c,d大于等于X 2 l,k,f。
  27. 一种通信装置，其特征在于，所述装置包括用于实现权利要求1-13中任一项所述的上报CSI的方法的单元。
  28. 一种通信装置，其特征在于，所述装置包括用于实现权利要求14-26中任一项所述的获取CSI的方法的单元。
  29. 一种终端设备，包括接收器和发射器，所述接收器与所述发射器耦合，其特征在于，
    所述接收器，用于接收来自于第二通信装置的配置信息，所述配置信息用于配置所述发射器上报的CSI；
    所述发射器，用于上报CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述终端设备第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述终端设备的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
    所述第二指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100215
    所述矩阵
    Figure PCTCN2019086909-appb-100216
    为2L行P列的矩阵，L为正整数，2L小于N t；
    所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100217
    为N t行2L列的矩阵，P为 N sb，或，所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100218
    和矩阵
    Figure PCTCN2019086909-appb-100219
    Figure PCTCN2019086909-appb-100220
    为N t行2L列的矩阵，
    Figure PCTCN2019086909-appb-100221
    为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
  30. 如权利要求29所述的终端设备，其特征在于，所述第二指示信息包括第三指示信息和第四指示信息，所述第三指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100222
    为L行P列M 1页的三维矩阵，M 1为正整数；
    所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100223
    为2N r行M 1列矩阵，或所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100224
    为2N r行M 1列矩阵。
  31. 如权利要求30所述的终端设备，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100225
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100226
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100227
    表示取矩阵
    Figure PCTCN2019086909-appb-100228
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100229
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100230
    则所述
    Figure PCTCN2019086909-appb-100231
    是由式
    Figure PCTCN2019086909-appb-100232
    Figure PCTCN2019086909-appb-100233
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100234
    则所述
    Figure PCTCN2019086909-appb-100235
    是由式
    Figure PCTCN2019086909-appb-100236
    得到；
    Figure PCTCN2019086909-appb-100237
    表示取矩阵
    Figure PCTCN2019086909-appb-100238
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100239
    表示取矩阵
    Figure PCTCN2019086909-appb-100240
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100241
    ×为把一维向量
    Figure PCTCN2019086909-appb-100242
    当做列向量的矩阵乘运算。
  32. 如权利要求30所述的终端设备，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100243
    的幅度矩阵
    Figure PCTCN2019086909-appb-100244
    Figure PCTCN2019086909-appb-100245
    所述
    Figure PCTCN2019086909-appb-100246
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100247
    的第l行第k列的元素的幅度确定的值；
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100248
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100249
    表示取矩阵
    Figure PCTCN2019086909-appb-100250
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100251
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100252
    则所述
    Figure PCTCN2019086909-appb-100253
    是由式
    Figure PCTCN2019086909-appb-100254
    Figure PCTCN2019086909-appb-100255
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100256
    则所述
    Figure PCTCN2019086909-appb-100257
    是由式
    Figure PCTCN2019086909-appb-100258
    得到；
    Figure PCTCN2019086909-appb-100259
    表示取矩阵
    Figure PCTCN2019086909-appb-100260
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100261
    表示取矩阵
    Figure PCTCN2019086909-appb-100262
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100263
    ×为把一维向量
    Figure PCTCN2019086909-appb-100264
    当做列向量的矩阵乘运算。
  33. 如权利要求32所述的终端设备，其特征在于，所述
    Figure PCTCN2019086909-appb-100265
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100266
    的第l行第k列的元素的幅度或幅度的平方。
  34. 如权利要求29-33任一项所述的终端设备，其特征在于，所述第二指示信息包括第五指示信息，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100267
    中元素的相位。
  35. 如权利要求34所述的终端设备，其特征在于，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100268
    中元素的相位，包括：
    所述第五指示信息包括第六指示信息和第七指示信息，所述第六指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100269
    为L行P列M 2页的三维矩阵，M 2为正整数；
    所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100270
    为2N r行M 2列矩阵，或，所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100271
    为2N r行M 2列矩阵；
    所述矩阵
    Figure PCTCN2019086909-appb-100272
    的相位矩阵
    Figure PCTCN2019086909-appb-100273
    所述
    Figure PCTCN2019086909-appb-100274
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100275
    的第l行第k列的元素的相位确定的值；
    其中，f(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，f(r′,i′)≠f(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100276
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100277
    表示取矩阵
    Figure PCTCN2019086909-appb-100278
    固定第三个索引为f(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100279
    若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100280
    则所述
    Figure PCTCN2019086909-appb-100281
    是由式
    Figure PCTCN2019086909-appb-100282
    Figure PCTCN2019086909-appb-100283
    得到，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100284
    则所述
    Figure PCTCN2019086909-appb-100285
    是由式
    Figure PCTCN2019086909-appb-100286
    得到；
    Figure PCTCN2019086909-appb-100287
    表示取矩阵
    Figure PCTCN2019086909-appb-100288
    固定前两个索引长度M 2的一维向量，所述长度M 2的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100289
    表示取矩阵
    Figure PCTCN2019086909-appb-100290
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第f(r,i)个元素为
    Figure PCTCN2019086909-appb-100291
    ×为把一维向量
    Figure PCTCN2019086909-appb-100292
    当做列向量的矩阵乘运算。
  36. 如权利要求35所述的终端设备，其特征在于，所述
    Figure PCTCN2019086909-appb-100293
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100294
    的第l行第k列的元素的相位或相位的平方。
  37. 如权利要求29-36任一项所述的终端设备，其特征在于，在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100295
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100296
    则l′＝l″；或
    在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100297
    和矩阵
    Figure PCTCN2019086909-appb-100298
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100299
    则l′＝l″，对于任意r′,r″∈{1,2,…,N r}，k′,k″∈{1,2,…,K}，若r′≠r″且
    Figure PCTCN2019086909-appb-100300
    则k′＝k″。
  38. 如权利要求30-37任一项所述的终端设备，其特征在于，所述第三指示信息包括用于指示
    Figure PCTCN2019086909-appb-100301
    中元素
    Figure PCTCN2019086909-appb-100302
    的X 1 l,k,g个比特，X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100303
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100304
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100305
    的Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100306
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100307
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100308
    的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
  39. 如权利要求38所述的终端设备，其特征在于，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100309
    则Y 1 l,c,d大于等于X 1 l,k,g，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100310
    则Z 1 k,c,d大于等于X 1 l,k,g。
  40. 如权利要求35-39任一项所述的终端设备，其特征在于，所述第六指示信息包括用于指示
    Figure PCTCN2019086909-appb-100311
    中元素
    Figure PCTCN2019086909-appb-100312
    的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100313
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100314
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100315
    的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100316
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100317
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100318
    的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
  41. 如权利要求40所述的终端设备，其特征在于，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100319
    则Y 2 l,c,d大于等于X 2 l,k,f，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100320
    则Z 2 k,c,d大于等于X 2 l,k,f。
  42. 一种网络设备，包括接收器和发射器，所述接收器与所述发射器耦合，其特征在于，
    所述发射器，用于向第一通信装置发送配置信息，所述配置信息用于配置所述第一通信装置上报的CSI；
    所述接收器，用于接收CSI，所述CSI用于指示矩阵H (r)，所述矩阵H (r)用于指示所述第一通信装置第r个天线端口的信道，所述矩阵H (r)为N t行N sb列的矩阵，N sb是所述CSI的上报带宽所包括的频域单元数量，N t为正整数，r∈{1,2,…,N r}，N r是所述第一通信装置的天线端口数量，所述CSI包括第一指示信息和第二指示信息；
    所述第二指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100321
    所述矩阵
    Figure PCTCN2019086909-appb-100322
    为2L行P列的矩阵，L为正整数，2L小于N t；
    所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100323
    为N t行2L列的矩阵，P为N sb，或，所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100324
    和矩阵
    Figure PCTCN2019086909-appb-100325
    Figure PCTCN2019086909-appb-100326
    为N t行2L列的矩阵，
    Figure PCTCN2019086909-appb-100327
    为N sb行K列的矩阵，K为正整数，K小于N sb，P为K。
  43. 如权利要求42所述的网络设备，其特征在于，所述第二指示信息包括第三指示信息和第四指示信息，所述第三指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100328
    为L行P列M 1页的三维矩阵，M 1为正整数；
    所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100329
    为2N r行M 1列矩阵，或，所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100330
    为2N r行M 1列矩阵。
  44. 如权利要求43所述的网络设备，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100331
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}和，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100332
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100333
    表示取矩阵
    Figure PCTCN2019086909-appb-100334
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100335
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100336
    则所述
    Figure PCTCN2019086909-appb-100337
    是由式
    Figure PCTCN2019086909-appb-100338
    Figure PCTCN2019086909-appb-100339
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100340
    则所述
    Figure PCTCN2019086909-appb-100341
    是由式
    Figure PCTCN2019086909-appb-100342
    得到；
    Figure PCTCN2019086909-appb-100343
    表示取矩阵
    Figure PCTCN2019086909-appb-100344
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100345
    表示取矩阵
    Figure PCTCN2019086909-appb-100346
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100347
    ×为把一维向量
    Figure PCTCN2019086909-appb-100348
    当做列向量的矩阵乘运算。
  45. 如权利要求43所述的网络设备，其特征在于，所述矩阵
    Figure PCTCN2019086909-appb-100349
    的幅度矩阵
    Figure PCTCN2019086909-appb-100350
    Figure PCTCN2019086909-appb-100351
    所述
    Figure PCTCN2019086909-appb-100352
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100353
    的第l行第k列的元素的幅度确定的值；
    其中，g(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，g(r′,i′)≠g(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100354
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100355
    表示取矩阵
    Figure PCTCN2019086909-appb-100356
    固定第三个索引为g(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行k列元素为
    Figure PCTCN2019086909-appb-100357
    若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100358
    则所述
    Figure PCTCN2019086909-appb-100359
    是由式
    Figure PCTCN2019086909-appb-100360
    Figure PCTCN2019086909-appb-100361
    得到，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100362
    则所述
    Figure PCTCN2019086909-appb-100363
    是由式
    Figure PCTCN2019086909-appb-100364
    得到；
    Figure PCTCN2019086909-appb-100365
    表示取矩阵
    Figure PCTCN2019086909-appb-100366
    固定前两个索引长度M 1的一维向量，所述长度M 1的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100367
    表示取矩阵
    Figure PCTCN2019086909-appb-100368
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第g(r,i)个元素为
    Figure PCTCN2019086909-appb-100369
    ×为把一维向量
    Figure PCTCN2019086909-appb-100370
    当做列向量的矩阵乘运算。
  46. 如权利要求45所述的网络设备，其特征在于，所述
    Figure PCTCN2019086909-appb-100371
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100372
    的第l行第k列的元素的幅度或幅度的平方。
  47. 如权利要求42-46任一项所述的网络设备，其特征在于，所述第二指示信息包括第五指示信息，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100373
    中元素的相位。
  48. 如权利要求47所述的网络设备，其特征在于，所述第五指示信息用于指示所述
    Figure PCTCN2019086909-appb-100374
    中元素的相位，包括：
    所述第五指示信息包括第六指示信息和第七指示信息，所述第六指示信息用于指示三维矩阵
    Figure PCTCN2019086909-appb-100375
    为L行P列M 2页的三维矩阵，M 2为正整数；
    所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100376
    为2N r行M 2列矩阵，或，所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100377
    为2N r行M 2列矩阵；
    所述矩阵
    Figure PCTCN2019086909-appb-100378
    的相位矩阵
    Figure PCTCN2019086909-appb-100379
    所述
    Figure PCTCN2019086909-appb-100380
    的第l行第k列的元素是根据所述
    Figure PCTCN2019086909-appb-100381
    的第l行第k列的元素的相位确定的值；
    其中，f(r,i)∈{1,2,…,2N r}，r∈{1,2,…,N r}，i∈{1,2}，且对于任意(r′,i′)≠(r″,i″)，f(r′,i′)≠f(r″,i″)；
    所述
    Figure PCTCN2019086909-appb-100382
    为L行P列2N r页的三维矩阵，
    Figure PCTCN2019086909-appb-100383
    表示取矩阵
    Figure PCTCN2019086909-appb-100384
    固定第三个索引为f(r,i)的L行P列的二维矩阵，所述L行P列的二维矩阵中第l行第k列的元素为
    Figure PCTCN2019086909-appb-100385
    若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100386
    则所述
    Figure PCTCN2019086909-appb-100387
    是由式
    Figure PCTCN2019086909-appb-100388
    Figure PCTCN2019086909-appb-100389
    得到，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100390
    则所述
    Figure PCTCN2019086909-appb-100391
    是由式
    Figure PCTCN2019086909-appb-100392
    得到；
    Figure PCTCN2019086909-appb-100393
    表示取矩阵
    Figure PCTCN2019086909-appb-100394
    固定前两个索引长度M 2的一维向量，所述长度M 2的一维向量中第m个元素为
    Figure PCTCN2019086909-appb-100395
    表示取矩阵
    Figure PCTCN2019086909-appb-100396
    固定前两个索引长度2N r的一维向量，所述长度2N r的一维向量第f(r,i)个元素为
    Figure PCTCN2019086909-appb-100397
    ×为把一维向量
    Figure PCTCN2019086909-appb-100398
    当做列向量的矩阵乘运算。
  49. 如权利要求48所述的网络设备，其特征在于，所述
    Figure PCTCN2019086909-appb-100399
    的第l行第k列的元素是所述
    Figure PCTCN2019086909-appb-100400
    的第l行第k列的元素的相位或相位的平方。
  50. 如权利要求42-49任一项所述的网络设备，其特征在于，在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100401
    的情况下，对于任意r′,r″∈{1,2,…,Nr}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100402
    则l′＝l″；或
    在所述第一指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100403
    和矩阵
    Figure PCTCN2019086909-appb-100404
    的情况下，对于任意r′,r″∈{1,2,…,N r}，l′,l″∈{1,2,…,2L}，若r′≠r″且
    Figure PCTCN2019086909-appb-100405
    则l′＝l″，对于任意r′,r″∈{1,2,…,N r}，k′,k″∈{1,2,…,K}，若r′≠r″且
    Figure PCTCN2019086909-appb-100406
    则k′＝k″。
  51. 如权利要求43-50任一项所述的网络设备，其特征在于，所述第三指示信息包括用于指示
    Figure PCTCN2019086909-appb-100407
    中元素
    Figure PCTCN2019086909-appb-100408
    的X 1 l,k,g个比特，X 1 l,k,g是正整数，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100409
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100410
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100411
    的Y 1 l,c,d个比特，Y 1 l,c,d是正整数，或，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100412
    则所述第四指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100413
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100414
    的Z 1 k,c,d个比特，Z 1 k,c,d是正整数。
  52. 如权利要求51所述的网络设备，其特征在于，若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100415
    则Y 1 l,c,d大于等于X 1 l,k,g，或若所述第四指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100416
    则Z 1 k,c,d大于等于X 1 l,k,g。
  53. 如权利要求48-52任一项所述的网络设备，其特征在于，所述第六指示信息包括用于指示
    Figure PCTCN2019086909-appb-100417
    中元素
    Figure PCTCN2019086909-appb-100418
    的X 2 l,k,f个比特，X 2 l,k,f是正整数，若所述第七指示信息用于指示矩 阵
    Figure PCTCN2019086909-appb-100419
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100420
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100421
    的Y 2 l,c,d个比特，Y 2 l,c,d是正整数，或，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100422
    则所述第七指示信息包括用于指示矩阵
    Figure PCTCN2019086909-appb-100423
    中第c行第d列的元素
    Figure PCTCN2019086909-appb-100424
    的Z 2 k,c,d个比特，Z 2 k,c,d是正整数。
  54. 如权利要求53所述的网络设备，其特征在于，若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100425
    则Y 2 l,c,d大于等于X 2 l,k,f，或若所述第七指示信息用于指示矩阵
    Figure PCTCN2019086909-appb-100426
    则Z 2 k,c,d大于等于X 2 l,k,f。
  55. 一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序或指令，当所述程序或指令被处理器执行时，使所述处理器执行如权利要求1-13中任一项所述的上报CSI的方法。
  56. 一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序或指令，当所述程序或指令被处理器执行时，使所述处理器执行如权利要求14-26中任一项所述的获取CSI的方法。
PCT/CN2019/086909 2019-05-14 2019-05-14 上报信道状态信息的方法、装置及相关设备 Ceased WO2020227941A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/086909 WO2020227941A1 (zh) 2019-05-14 2019-05-14 上报信道状态信息的方法、装置及相关设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/086909 WO2020227941A1 (zh) 2019-05-14 2019-05-14 上报信道状态信息的方法、装置及相关设备

Publications (1)

Publication Number Publication Date
WO2020227941A1 true WO2020227941A1 (zh) 2020-11-19

Family

ID=73289982

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/086909 Ceased WO2020227941A1 (zh) 2019-05-14 2019-05-14 上报信道状态信息的方法、装置及相关设备

Country Status (1)

Country Link
WO (1) WO2020227941A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090252091A1 (en) * 2008-04-04 2009-10-08 Futurewei Technologies, Inc. System and Method for Multi-Stage Zero Forcing Beamforming in a Wireless Communications System
CN102412941A (zh) * 2010-09-21 2012-04-11 华为技术有限公司 周期信道状态信息的传输配置和传输方法及装置
EP2665198A1 (en) * 2012-05-14 2013-11-20 BlackBerry Limited Codebook based downlink multi-user interference alignment scheme
CN108631837A (zh) * 2017-03-24 2018-10-09 华为技术有限公司 信息的传输方法和设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090252091A1 (en) * 2008-04-04 2009-10-08 Futurewei Technologies, Inc. System and Method for Multi-Stage Zero Forcing Beamforming in a Wireless Communications System
CN102412941A (zh) * 2010-09-21 2012-04-11 华为技术有限公司 周期信道状态信息的传输配置和传输方法及装置
EP2665198A1 (en) * 2012-05-14 2013-11-20 BlackBerry Limited Codebook based downlink multi-user interference alignment scheme
CN108631837A (zh) * 2017-03-24 2018-10-09 华为技术有限公司 信息的传输方法和设备

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HUAWEI ET AL.: "Non-linear Precoding for Downlink Multiuser MIMO", 3GPP TSG RAN WG1 AD-HOC MEETING NR R1-1700053, 20 January 2017 (2017-01-20), XP051207595, DOI: 20200109152610A *
NTT DOCOMO: "Views on Single-Cell CSI Feedback Enhancement for DL MU-MIMO in LTE-Advanced", 3GPP TSG RAN WG1 MEETING #58BIS R1-094241, 16 October 2009 (2009-10-16), XP050388705, DOI: 20200109152509A *

Similar Documents

Publication Publication Date Title
JP7143506B2 (ja) 第２タイプフィードバック報告のためのコードブック制限のための方法および装置、ならびに無線通信ネットワークにおける線形結合コードブックのための上位層構成および報告
TWI795336B (zh) 一種基於通道互易性的預編碼矩陣配置方法及裝置
US11811431B2 (en) Enhanced frequency compression for overhead reduction for CSI reporting and usage
CN109845154B (zh) 通信方法、基站和终端设备
US9843366B1 (en) Method for transmitting 4-antenna precoding matrix, user equipment and base station
KR20260023109A (ko) 무선 통신 시스템에서 csi 보고를 위한 방법 및 장치
AU2015406856B2 (en) Precoding information sending and feedback method and apparatus
CN108631847A (zh) 传输信道状态信息的方法、终端设备和网络设备
US20200067588A1 (en) Data transmission method, terminal device, and network device
WO2019137445A1 (zh) 信道状态信息的测量方法和装置
CN104782071B (zh) 信道状态信息的测量和反馈方法、终端及基站
US10756801B2 (en) Method and apparatus for transmitting channel state information
CN108282254B (zh) 一种信道状态信息上报方法、基站和用户设备
WO2016183835A1 (zh) 传输信号的方法和设备
WO2019144418A1 (zh) 一种预编码矩阵索引上报方法、通信装置及介质
CN117811621A (zh) 一种多站协作的信道状态信息指示方法和通信装置
WO2018094709A1 (zh) 一种确定预编码矩阵的方法、装置及系统
CN110380767B (zh) 一种预编码矩阵确定方法及装置
WO2021087908A1 (zh) 无线通信的方法和终端设备
CN113131980B (zh) 信息上报、指示方法、终端设备和网络设备
WO2017075819A1 (zh) 一种预编码矩阵索引的传输方法及装置
WO2020227941A1 (zh) 上报信道状态信息的方法、装置及相关设备
WO2017088658A1 (zh) 获取信道信息的方法及装置
WO2020047774A1 (zh) 信道信息处理方法和装置
WO2016066036A1 (zh) 一种信道状态信息的反馈、获取方法及装置

Legal Events

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

Ref document number: 19929132

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19929132

Country of ref document: EP

Kind code of ref document: A1