WO2023116407A1 - 信息处理方法、装置、终端及网络设备 - Google Patents
信息处理方法、装置、终端及网络设备 Download PDFInfo
- Publication number
- WO2023116407A1 WO2023116407A1 PCT/CN2022/136615 CN2022136615W WO2023116407A1 WO 2023116407 A1 WO2023116407 A1 WO 2023116407A1 CN 2022136615 W CN2022136615 W CN 2022136615W WO 2023116407 A1 WO2023116407 A1 WO 2023116407A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- channel information
- network model
- data
- information
- terminal
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 96
- 238000003672 processing method Methods 0.000 title claims abstract description 59
- 238000012545 processing Methods 0.000 claims abstract description 153
- 238000000034 method Methods 0.000 claims abstract description 93
- 239000013598 vector Substances 0.000 claims description 136
- 238000004590 computer program Methods 0.000 claims description 84
- 230000008569 process Effects 0.000 claims description 50
- 238000013139 quantization Methods 0.000 claims description 30
- 239000011159 matrix material Substances 0.000 claims description 26
- 238000000354 decomposition reaction Methods 0.000 claims description 24
- 238000003860 storage Methods 0.000 claims description 24
- 101150071746 Pbsn gene Proteins 0.000 claims description 7
- 230000000717 retained effect Effects 0.000 claims description 7
- 238000004891 communication Methods 0.000 abstract description 4
- 230000005540 biological transmission Effects 0.000 description 27
- 230000006835 compression Effects 0.000 description 21
- 238000007906 compression Methods 0.000 description 21
- 230000003044 adaptive effect Effects 0.000 description 20
- 238000010586 diagram Methods 0.000 description 20
- 230000006870 function Effects 0.000 description 12
- 238000012549 training Methods 0.000 description 8
- 230000009286 beneficial effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000007774 longterm Effects 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000000873 masking effect Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 3
- 238000010295 mobile communication Methods 0.000 description 3
- 239000007787 solid Substances 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 230000011664 signaling Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0009—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the channel coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
Definitions
- the present disclosure relates to the field of communication technologies, and in particular, to an information processing method, device, terminal and network equipment.
- artificial intelligence-based neural networks can be used for compressed feedback of channel information, such as compressed feedback of Channel State Information (CSI).
- CSI Channel State Information
- the sparsity of channel information can be utilized, and the encoder module can be used to compress the channel information to reduce the feedback overhead, and then the decoder module can be used to restore the channel information similar to the original.
- it is considered to deploy multiple network models including encoders and/or decoders on the terminal side and/or network device side.
- multiple network models including encoders and/or decoders on the terminal side and/or network device side.
- the present disclosure provides an information processing method, device, terminal, and network equipment, which solves the problem of how to deploy multiple network models on the terminal side and/or network equipment side.
- An embodiment of the present disclosure provides an information processing method, including:
- the terminal receives the downlink signal sent by the network device
- the terminal determines a first network model according to the downlink signal; wherein, the first network model is at least one of 1 network models pre-deployed or downloaded from the network device side, and I is a positive integer;
- the terminal encodes the first channel information through the first network model to obtain encoded data to be transmitted;
- the terminal sends the processed data to the network device.
- the downlink signal is a downlink reference signal; the terminal determines a first network model according to the downlink signal, including:
- the terminal determines the first channel information according to the downlink reference signal
- the terminal determines the first network model according to the first channel information.
- the terminal determines the first network model according to the first channel information, it further includes:
- the terminal sends a first uplink signal to the network device; wherein the first uplink signal carries first indication information used to indicate the first network model.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the first indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the first indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the terminal determines the first network model according to the first channel information, including:
- the terminal respectively performs encoding and/or decoding processing through the one network model according to the first channel information, to obtain second channel information corresponding to each network model;
- the terminal calculates and obtains target parameters corresponding to each network model according to the first channel information and the second channel information corresponding to each network model;
- the terminal selects the network with the smallest number of elements in the encoded data obtained after encoding from the M network models a model, determined as the first network model;
- the terminal selects from the one network model the number of elements in the encoded data obtained after encoding processing is the largest
- the network model is determined as the first network model.
- the terminal calculates target parameters corresponding to each network model according to the first channel information and the second channel information corresponding to each network model, including:
- the terminal calculates and obtains target parameters corresponding to each network model through target criteria according to the first channel information and the second channel information corresponding to each network model;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the downlink signal carries second indication information for indicating the first network model; before the terminal receives the downlink signal sent by the network device, it further includes:
- the terminal sends a second uplink signal to the network device; wherein the second uplink signal is used by the network device to determine the first network model.
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indicator (Precoding matrix indicator, PMI).
- Precoding matrix indicator PMI
- the first channel information includes L pieces of channel information, where L is a positive integer; wherein, the first channel information in the L pieces of channel information includes the following item:
- the downlink channel information H l corresponding to the lth receiving antenna of the first frequency domain unit;
- N t is the number of transmitting antenna ports of the network device
- N c is the number of frequency domain units in the first frequency domain unit
- P1 is the number of receiving antennas of the terminal;
- the downlink reference signal received by the lth receiving antenna of the first frequency domain unit passes through the received signal Y l of the channel;
- the lth eigenvector U l obtained by singular value decomposition of the downlink channel information of the first frequency domain unit; where, P2 is the number of feature vectors;
- the precoding vector V l of the lth data stream of the first frequency domain unit where, P3 is the number of data streams;
- a combination of downlink channel information corresponding to the R receiving antennas of the first frequency domain unit; wherein, L 1;
- a combination of the precoding vectors of the R data streams of the first frequency domain unit and (v1-R) zero vectors 0j ; wherein, L 1; v1 is the maximum number of data streams that the terminal can transmit;
- a combination of downlink channel information corresponding to the R receiving antennas of the first frequency domain unit and (v2-R) zero vectors; wherein, L 1; v2 is the maximum number of receiving antennas of the terminal;
- the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in is the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in
- the first frequency domain unit is all frequency domain units or one frequency domain unit.
- the frequency domain unit is a subcarrier, or the frequency domain unit is a physical resource block (Physical Resource Block, PRB), or the frequency domain unit is a subband including multiple PRBs.
- PRB Physical Resource Block
- the terminal sends the encoded data processed data to the network device, including:
- the terminal performs quantization processing on the first element in the coded data, and transmits the quantized data to the network device; where the first element is part or all of the elements in the coded data.
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the first channel information includes L pieces of channel information, and L ⁇ 1; the method further includes the following item:
- the terminal sends third indication information to the network device; wherein, one piece of channel information corresponds to one third indication information, and the third indication information is used to indicate the number of the first elements and /or the position of the first element in the encoded data;
- the terminal for the The first part of channel information is to send common fourth indication information to the network device; where the fourth indication information is used to indicate the number of the first elements and/or the number of the first elements in the encoded data position in ; the first part of channel information is part or all of the channel information in the L pieces of channel information;
- the terminal sends fifth indication information for the joint indication of the L pieces of channel information to the network device; wherein, the fifth indication information is used to indicate the coded data encoded by each channel information The number of first elements in and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the number of the first element in the L encoded data after the L channel information is encoded The total number and/or the position of the first element in the encoded data.
- the first element includes K1 non-zero elements in the coded data:
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements and the number K 1 of the K 1 non-zero elements in the encoded data Location;
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements.
- the third indication information or the fourth indication information includes N3 bits, and the N3 bits are used to indicate the number K 1 of the first elements; wherein,
- the first element includes K1 elements in the coded data determined according to the mask vector, or in the case of K1 consecutive elements starting from the first position in the coded data:
- the third indication information or the fourth indication information includes N4 bits, and the N4 bits are used to indicate the number K 2 of non-zero elements in the K 1 elements, and the third indication information or the The fourth indication information is also used to indicate the positions of K 2 non-zero elements in the encoded data; wherein,
- the fifth indication information includes N5 bits, and the N5 bits are used to indicate the number of first elements in the encoded data after each channel information is encoded; wherein, The maximum number of elements reported for each encoded data;
- the fifth indication information includes N6 bits, and the N6 bits are used to indicate the total number of first elements in the L coded data; wherein, It is the maximum total number of reported elements in the L coded data.
- the terminal determines the first network model according to the downlink signal, it further includes:
- the terminal obtains the recovered second channel information through the first network model based on the processed data of the encoded data
- the terminal calculates and obtains a channel quality indicator (Channel quality indicator, CQI) according to the second channel information;
- the terminal sends the CQI to the network device.
- the terminal determines the first network model according to the downlink signal, it further includes:
- the terminal receives the beamformed channel information reference signal sent by the network device; wherein the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the terminal determines effective channel information according to the channel information reference signal
- the terminal calculates and obtains the CQI according to the effective channel information
- the terminal sends the CQI to the network device.
- the information processing method further includes:
- the terminal sends the interference and noise related information of the downlink channel to the network device; wherein the interference and noise related information of the downlink channel is used for the network device to calculate the CQI.
- An embodiment of the present disclosure provides an information processing method, including:
- the network device sends a downlink signal to the terminal; wherein the downlink signal is used by the terminal to determine a first network model, where the first network model is at least one of the pre-deployed I network models, and I is a positive integer;
- the network device receives data after the terminal processes the coded data; wherein, the coded data is obtained by the terminal after coding the first channel information according to the first network model;
- the network device determines the data to be decoded according to the data after the terminal processes the encoded data
- the network device performs decoding processing on the data to be decoded according to the first network model to obtain restored channel information.
- the downlink signal is a downlink reference signal
- the network device decodes the data to be decoded according to the first network model, and before obtaining the restored channel information, further includes:
- the network device receives the first uplink signal sent by the terminal
- the network device determines the first network model according to the first uplink signal.
- the downlink signal carries second indication information for indicating the first network model; before the network device sends the downlink signal to the terminal, it further includes:
- the network device receives the second uplink signal sent by the terminal
- the network device determines third channel information according to the second uplink signal
- the network device determines the first network model according to the third channel information.
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the network device determining the first network model according to the third channel information includes:
- the network device respectively performs encoding and/or decoding processing through the one network model according to the third channel information, to obtain fourth channel information corresponding to each network model;
- the network device calculates and obtains target parameters corresponding to each network model according to the third channel information and the fourth channel information corresponding to each network model;
- the network device selects the one with the smallest number of elements in the encoded data obtained after encoding from the M network models A network model, determined as the first network model;
- the network device selects the number of elements in the encoded data obtained after encoding from the one network model The largest network model is determined as the first network model.
- the network device calculates target parameters corresponding to each network model according to the third channel information and fourth channel information corresponding to each network model, including:
- the network device calculates and obtains target parameters corresponding to each network model through target criteria according to the third channel information and the fourth channel information corresponding to each network model;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the second indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the second indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the first channel information includes L pieces of channel information, and L ⁇ 1; the network device further includes the following item before determining the data to be decoded according to the data after the terminal processes the encoded data:
- the network device receives third indication information sent by the terminal; wherein, one piece of channel information corresponds to one third indication information, and the third indication information is used to indicate the number of first elements and/or or the position of the first element in the encoded data, where the first element is part or all of the elements in the encoded data;
- the network device receives common fourth indication information sent by the terminal for the first part of channel information; wherein the first part of channel information is part or all of the L pieces of channel information , and the number of the first element and/or the position of the first element in the encoded data in the encoded data after the encoding of the first partial channel information is the same; the fourth indication information is used to indicate the The number of the first elements and/or the position of the first elements in the encoded data;
- the network device receives the fifth indication information sent by the terminal for the joint indication of the L channel information; wherein the fifth indication information is used to indicate the encoded code of each channel information The number of first elements in the data and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the first element in the L encoded data after the L channel information is encoded and/or the position of the first element in the encoded data.
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the network device determines the data to be decoded according to the data after the terminal processes the encoded data, including:
- the network device performs dequantization processing on the encoded data processed by the terminal to obtain dequantized data
- the network device determines the data to be decoded by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data.
- the network device determines the data to be decoded by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data, including the following: item:
- the network device determines the data to be decoded through a mask vector according to the dequantized data
- the network device determines the data to be decoded by the number of the first elements according to the dequantized data
- the network device determines the data to be decoded by the position of the first element in the coded data according to the dequantized data
- the network device determines the data to be decoded by the number of non-zero elements and the position of the non-zero elements in the coded data according to the dequantized data;
- the network device passes the number of the first elements, the number of non-zero elements in the first elements, and the position of the non-zero elements in the encoded data according to the data after dequantization processing , to determine the data to be decoded.
- the information processing method further includes:
- the network device receives the channel quality indicator CQI sent by the terminal.
- the information processing method further includes:
- the network device sends a beamformed channel information reference signal to the terminal; wherein the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the network device receives the CQI sent by the terminal; wherein, the CQI is calculated by the terminal according to the channel information reference signal.
- the information processing method further includes:
- the network device receives the interference and noise related information of the downlink channel sent by the terminal;
- the network device calculates and obtains the CQI according to the interference and noise related information of the downlink channel and the recovered channel information.
- An embodiment of the present disclosure provides an information processing device, including a memory, a transceiver, and a processor;
- the memory is used to store computer programs;
- the transceiver is used to send and receive data under the control of the processor;
- the processor is used to read the computer programs in the memory and perform the following operations:
- the downlink signal determine a first network model; wherein, the first network model is at least one of 1 network models pre-deployed or downloaded from the network device side, and I is a positive integer;
- the downlink signal is a downlink reference signal
- the processor is configured to read a computer program in the memory and perform the following operations:
- the processor is configured to read a computer program in the memory and perform the following operations:
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the first indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the first indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the processor is configured to read a computer program in the memory and perform the following operations:
- the first channel information respectively perform encoding and/or decoding processing through the one network model to obtain second channel information corresponding to each network model;
- the one network model includes M ⁇ 1 network models whose target parameters meet the preset conditions, then select the network model with the smallest number of elements in the encoded data obtained after the encoding process from the M network models, and determine is the first network model;
- the processor is configured to read a computer program in the memory and perform the following operations:
- the first channel information and the second channel information corresponding to each network model calculate and obtain the target parameters corresponding to each network model through target criteria;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the downlink signal carries second indication information used to indicate the first network model;
- the processor is configured to read a computer program in the memory and perform the following operations:
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the first channel information includes L pieces of channel information, where L is a positive integer; wherein, the first channel information in the L pieces of channel information includes the following item:
- the downlink channel information H l corresponding to the lth receiving antenna of the first frequency domain unit;
- N t is the number of transmitting antenna ports of the network device
- N c is the number of frequency domain units in the first frequency domain unit
- P1 is the number of receiving antennas of the terminal;
- the downlink reference signal received by the lth receiving antenna of the first frequency domain unit passes through the received signal Y l of the channel;
- the lth eigenvector U l obtained by singular value decomposition of the downlink channel information of the first frequency domain unit; where, P2 is the number of feature vectors;
- the precoding vector V l of the lth data stream of the first frequency domain unit where, P3 is the number of data streams;
- a combination of downlink channel information respectively corresponding to the R receiving antennas of the first frequency domain unit; wherein, L 1;
- a combination of the precoding vectors of the R data streams of the first frequency domain unit and (v1-R) zero vectors 0j ; wherein, L 1; v1 is the maximum number of data streams that the terminal can transmit;
- a combination of downlink channel information corresponding to the R receiving antennas of the first frequency domain unit and (v2-R) zero vectors; wherein, L 1; v2 is the maximum number of receiving antennas of the terminal;
- the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in is the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in
- the first frequency domain unit is all frequency domain units or one frequency domain unit.
- the frequency domain unit is a subcarrier, or the frequency domain unit is a PRB, or the frequency domain unit is a subband including multiple PRBs.
- the processor is configured to read a computer program in the memory and perform the following operations:
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the first channel information includes L pieces of channel information, and L ⁇ 1;
- the processor is configured to read a computer program in the memory and perform one of the following operations:
- the fourth indication information is used to indicate the number of the first elements and/or the position of the first elements in the encoded data ;
- the first part of channel information is part or all of the channel information in the L pieces of channel information;
- the fifth indication information is used to indicate the first in the coded data encoded by each channel information.
- the number of elements and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the total number of first elements in the L encoded data after encoding the L channel information And/or the position of the first element in the encoded data.
- the first element includes K1 non-zero elements in the coded data:
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements and the number K 1 of the K 1 non-zero elements in the encoded data Location;
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements.
- the third indication information or the fourth indication information includes N3 bits, and the N3 bits are used to indicate the number K 1 of the first elements; wherein,
- the first element includes K1 elements in the coded data determined according to the mask vector, or in the case of K1 consecutive elements starting from the first position in the coded data:
- the third indication information or the fourth indication information includes N4 bits, and the N4 bits are used to indicate the number K 2 of non-zero elements in the K 1 elements, and the third indication information or the The fourth indication information is also used to indicate the positions of K 2 non-zero elements in the encoded data; wherein,
- the fifth indication information includes N5 bits, and the N5 bits are used to indicate the number of first elements in the encoded data after each channel information is encoded; wherein, The maximum number of elements reported for each encoded data;
- the fifth indication information includes N6 bits, and the N6 bits are used to indicate the total number of first elements in the L coded data; wherein, It is the maximum total number of reported elements in the L coded data.
- the processor is configured to read a computer program in the memory and perform the following operations:
- the processor is configured to read a computer program in the memory and perform the following operations:
- the processor is configured to read a computer program in the memory and perform the following operations:
- An embodiment of the present disclosure provides a terminal, including:
- a first receiving unit configured to receive a downlink signal sent by a network device
- the first processing unit is configured to determine a first network model according to the downlink signal; wherein, the first network model is at least one of I network models pre-deployed or downloaded from the network device side, and I is a positive integer ;
- An encoding unit configured to encode the first channel information through the first network model to obtain encoded data to be transmitted;
- a first sending unit configured to send the processed data of the encoded data to the network device.
- An embodiment of the present disclosure provides an information processing device, including a memory, a transceiver, and a processor;
- the memory is used to store computer programs;
- the transceiver is used to send and receive data under the control of the processor;
- the processor is used to read the computer programs in the memory and perform the following operations:
- the downlink signal is used by the terminal to determine a first network model, where the first network model is at least one of the pre-deployed I network models, where I is a positive integer;
- the downlink signal is a downlink reference signal
- the processor is configured to read a computer program in the memory and perform the following operations:
- the downlink signal carries second indication information used to indicate the first network model;
- the processor is configured to read a computer program in the memory and perform the following operations:
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the processor is configured to read a computer program in the memory and perform the following operations:
- the third channel information respectively perform encoding and/or decoding processing through the one network model to obtain fourth channel information corresponding to each network model;
- the one network model includes M ⁇ 1 network models whose target parameters meet the preset conditions, then select the network model with the smallest number of elements in the encoded data obtained after the encoding process from the M network models, and determine is the first network model;
- the processor is configured to read a computer program in the memory and perform the following operations:
- the third channel information and the fourth channel information corresponding to each network model calculate and obtain the target parameters corresponding to each network model through target criteria;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the second indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the second indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the first channel information includes L pieces of channel information, and L ⁇ 1;
- the processor is configured to read a computer program in the memory and perform one of the following operations:
- receiving common fourth indication information sent by the terminal for the first part of channel information wherein the first part of channel information is part or all of the L pieces of channel information, and the The number of the first element and/or the position of the first element in the encoded data in the encoded data after encoding the first part of channel information is the same; the fourth indication information is used to indicate the first element and/or the position of the first element in the coded data;
- the fifth indication information is used to indicate the first in the coded data encoded by each channel information The number of elements and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the total number of first elements in the L encoded data after encoding the L channel information And/or the position of the first element in the encoded data.
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the processor is configured to read a computer program in the memory and perform the following operations:
- the data to be decoded is determined by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data.
- the processor is configured to read a computer program in the memory and perform one of the following operations:
- the number of the first elements, the number of non-zero elements in the first elements and the position of the non-zero elements in the coded data are used to determine the Data to be decoded.
- the processor is configured to read a computer program in the memory and perform the following operations:
- the processor is configured to read a computer program in the memory and perform the following operations:
- the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the processor is configured to read a computer program in the memory and perform the following operations:
- An embodiment of the present disclosure provides a network device, including:
- the first sending unit is configured to send a downlink signal to the terminal; wherein the downlink signal is used by the terminal to determine a first network model, and the first network model is at least one of pre-deployed I network models, I is a positive integer;
- the first receiving unit is configured to receive data obtained by the terminal after processing the encoded data; wherein the encoded data is obtained by the terminal after encoding the first channel information according to the first network model;
- the first processing unit is configured to determine the data to be decoded according to the data after the terminal processes the encoded data
- the decoding unit is configured to perform decoding processing on the data to be decoded according to the first network model to obtain restored channel information.
- An embodiment of the present disclosure provides a processor-readable storage medium, the processor-readable storage medium stores a computer program, and the computer program is used to enable the processor to execute the information described above on the terminal side or the network device side Steps in a processing method.
- the terminal receives the downlink signal sent by the network device, and according to the downlink signal, determines the first network model used to encode the channel information corresponding to the downlink reference information from one pre-deployed network model, And use the first network model to encode the first channel information to obtain encoded data to be transmitted, and send the processed data to the network device, so as to ensure that the terminal side and/or When multiple network models are deployed on the network device side, the terminal side and/or the network device side can determine the network model used for channel information transmission from the pre-deployed multiple network models, and can reduce the feedback overhead of channel information transmission and improve the system performance. performance.
- Figure 1 shows a schematic diagram of the structure of AE
- Fig. 2 represents the network training flowchart of the CSI compression feedback based on artificial intelligence
- FIG. 3 shows a flowchart of an information processing method on a terminal side according to an embodiment of the present disclosure
- FIG. 4 shows a block diagram of a channel information compression feedback model of the i-th network model in an embodiment of the present disclosure
- Fig. 5a shows a block diagram of an adaptive encoder model training process according to an embodiment of the present disclosure
- FIG. 5b shows a block diagram of an adaptive decoder model training process according to an embodiment of the present disclosure
- FIG. 6 shows a block diagram of a CSI compression feedback model of a feedback information length adaptive variable network according to an embodiment of the present disclosure
- FIG. 7 shows a block diagram of a terminal according to an embodiment of the present disclosure
- FIG. 8 shows a block diagram of an information processing device on the terminal side of an embodiment of the present disclosure
- FIG. 9 shows a flowchart of an information processing method on the network device side according to an embodiment of the present disclosure.
- Fig. 10 represents the block diagram of the network device of the disclosed embodiment
- FIG. 11 shows a block diagram of an information processing device on the network device side according to an embodiment of the present disclosure
- FIG. 12 shows one of the block diagrams of the CSI compression feedback processing flow of multiple network models in an embodiment of the present disclosure
- Fig. 13 shows the second block diagram of the CSI compression feedback processing flow of multiple network models in the embodiment of the present disclosure
- Fig. 14 is a schematic diagram showing position indications of non-zero elements of a truncated codeword according to an embodiment of the present disclosure.
- sequence numbers of the following processes do not mean the order of execution, and the execution order of each process should be determined by its functions and internal logic, and should not be implemented in the present disclosure.
- the implementation of the examples constitutes no limitation.
- system and “network” are often used interchangeably herein.
- the applicable system may be a global system of mobile communication (GSM) system, a code division multiple access (CDMA) system, a wideband code division multiple access (WCDMA) general packet Wireless business (general packet radio service, GPRS) system, long term evolution (long term evolution, LTE) system, LTE frequency division duplex (frequency division duplex, FDD) system, LTE time division duplex (time division duplex, TDD) system, Long term evolution advanced (LTE-A) system, universal mobile telecommunications system (UMTS), worldwide interoperability for microwave access (WiMAX) system, fifth generation mobile communication (5th generation -Generation, 5G) new air interface (New Radio, NR) system, etc.
- GSM global system of mobile communication
- CDMA code division multiple access
- WCDMA wideband code division multiple access
- GPRS general packet Wireless business
- long term evolution long term evolution
- LTE long term evolution
- LTE frequency division duplex frequency division duplex
- FDD frequency division duplex
- TDD time division du
- MIMO transmission can be Single User MIMO (Single User MIMO, SU-MIMO) or Multi-User MIMO (Multiple User MIMO, MU-MIMO).
- MIMO transmission can be two-dimensional MIMO (2 Dimission MIMO, 2D-MIMO), three-dimensional MIMO (3 Dimission MIMO, 3D-MIMO), full-dimensional MIMO (Full Dimension MIMO, FD-MIMO) Or Massive MIMO, it can also be diversity transmission or precoding transmission or beamforming transmission, etc.
- the auto-encoder (Auto-encoder, AE) involved in the embodiment of the present disclosure is a neural network with the same input and output, consisting of two parts: an encoder and a decoder.
- the encoder compresses the input data into a latent space representation, from which the decoder reconstructs the input.
- the structure of AE is shown in Figure 1.
- the encoder module and the decoder module can adopt convolutional neural network, fully connected neural network, recurrent neural network, etc. and their combination structures.
- the dimension of the encoder output Z is smaller than its input X, the smaller the dimension of the output Z, the higher the compression rate, and the corresponding error of the decoder recovering CSI is also larger.
- the network training process of artificial intelligence-based CSI compression feedback is shown in Figure 2.
- the terminal side inputs the obtained channel information V to the encoder to obtain the compressed codeword C (the information length of the codeword C is much smaller than the length of the input V), and then inputs the codeword C to the quantizer, and the quantized
- the binary bit stream is input to the dequantizer to obtain the code word C' with quantization error, and finally C' is input to the decoder module to obtain the restored channel information V'.
- an embodiment of the present disclosure provides an information processing method, including the following steps:
- Step 31 The terminal receives the downlink signal sent by the network device.
- the downlink signal may be a downlink reference signal, and the downlink reference signal may be used by the terminal to determine a channel corresponding to the downlink reference information from one network model pre-deployed or downloaded from the network device side
- the first network model for encoding information or, the downlink signal may carry first indication information for indicating the first network model.
- Step 32 The terminal determines a first network model according to the downlink signal.
- the first network model is at least one of I network models pre-deployed or downloaded from the network device side, where I is a positive integer.
- the first network model may be selected and determined by the terminal side from one network model, if the downlink signal is a downlink reference signal sent by a network device, then the terminal may select from one network model according to the downlink reference signal Select at least one network model as the first network model.
- the first network model may be configured by the network device to the terminal, for example, the downlink signal carries first indication information for indicating the first network model among the one network models.
- Step 33 The terminal encodes the first channel information through the first network model to obtain encoded data to be transmitted.
- the first channel information may include L pieces of channel information, the channel information may be CSI, and L is a positive integer; where, when L is greater than 1, different channel information may be encoded using the same network model ( or called compression) processing, or different channel information can also be encoded using different network models. For example, after encoding the first channel information through the network model, the codeword corresponding to each channel information can be obtained (that is, to be transmitted coded data) to reduce the feedback overhead of the terminal.
- Step 34 The terminal sends the processed data of the encoded data to the network device.
- the terminal may send some or all elements in the coded data to the network device after quantization processing.
- the terminal receives the downlink signal sent by the network device, and according to the downlink signal, determines the first network model used to encode the channel information corresponding to the downlink reference information from one pre-deployed network model, And use the first network model to encode the first channel information to obtain encoded data to be transmitted, and send the processed data to the network device, so as to ensure that the terminal side and/or When multiple network models are deployed on the network device side, the terminal side and/or the network device side can determine the network model used for channel information transmission from the pre-deployed multiple network models, and can reduce the feedback overhead of channel information transmission and improve the system performance. performance.
- the network model includes an encoder and/or a decoder
- the network model may be a network model including an encoder and a decoder
- the network model may be a network model including an encoder
- the The network model described above may be a network model including a decoder.
- At least one encoder for channel information encoding can be selected from multiple encoders based on the above-mentioned solution of the present disclosure
- at least one encoder can be selected from multiple decoders based on the above solution of the present disclosure.
- the codeword C i output by the encoder of the i-th network model has a length of k i .
- the ki corresponding to different network models may be equal or unequal.
- the block diagram of the channel information compression feedback model based on the i-th network model is shown in Figure 4.
- a network model including an adaptive encoder and an adaptive decoder can be used. If one network model is pre-trained, the one network model can all be a network model including an adaptive encoder and an adaptive decoder, or some network models in this one network model include an adaptive encoder and an adaptive decoder.
- Adaptive Decoder Network Model The adaptive encoder and adaptive decoder are based on adding a random mask vector module after the encoder module, and adding a corresponding mask vector filling module before the decoder, which can be obtained through network training, such as Figures 5a and 5b.
- the codeword C is obtained from the channel information V through the encoder, and the length of the vector is k max .
- the codeword C is obtained after passing through a mask vector module with a length of k max
- the vector M is a mask vector
- k ⁇ [0,k max ] is a random value that can obey the random distribution of equal probability
- the operation ⁇ means bitwise multiplication. It is equivalent to truncating the information length k max output after encoding to k as effective information for feedback; the adaptive decoder introduces a mask vector filling module to complement the received information with 0, and then converts the information length Revert to k max as input to the decoder.
- a block diagram of a CSI compression feedback model based on adaptive variable feedback information length network is given.
- the channel information V is first input to the adaptive encoder, and the output codeword C is truncated by the mask vector module and then input to the quantizer module for quantization to realize the compression of the channel information V, and the quantized binary bit stream is then input in sequence Go to the inverse quantizer module and the mask vector filling module to obtain the codeword C'. Then input the code word C' to the adaptive decoder to obtain the restored channel information V'.
- the introduction of the mask vector, quantizer, inverse quantizer and mask vector filling module in the embodiments of the present disclosure is to better assist in explaining the process of CSI compression feedback. Some of these modules may not exist, or may be replaced by other modules with similar functionality.
- the mask vector module is used to determine all or part of the element information in the codeword fed back by the terminal
- the quantizer is used to quantize some discrete values of each element value in the codeword by N bits (N bits)
- the inverse quantizer is used to quantize some discrete values according to N
- the bits represent the determined value as the element value in the codeword
- the mask vector filling module is used to determine the corresponding element in the codeword and fill the remaining positions with 0s according to the number of elements and/or the indication information of the element position reported by the terminal.
- the downlink signal is a downlink reference signal; the terminal determines a first network model according to the downlink signal, including:
- the terminal determines the first channel information according to the downlink reference signal
- the terminal determines the first network model according to the first channel information.
- downlink channel information that is, first channel information
- the terminal determines the first network model according to the first channel information, it further includes:
- the terminal sends a first uplink signal to the network device; wherein the first uplink signal carries first indication information used to indicate the first network model.
- the terminal side selects the network model (namely the first network model) used for channel information encoding and/or decoding processing from one network model
- the terminal determines that the first channel information encoding and/or after decoding the first network model
- the terminal sends the first indication information to the network device, so that the network device learns the first channel information encoding and/or decoding process according to the first indication information.
- network model namely the first network model
- the first channel information includes L pieces of channel information, where L is a positive integer;
- the first indication information when different channel information is encoded and/or decoded through different network models, the first indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information; ,
- N1 may be determined according to the number L of channel information and the number I of network models. For example: in Indicates that L network models are selected from I network models, and L represents the number of transmission layers (that is, the number of channel information).
- the first indication information when different channel information is encoded and/or decoded through the same network model, the first indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- N2 can be determined according to the number I of network models. For example:
- the terminal may select an appropriate first network model from one network model in a traversal manner; specifically, the terminal determines the first network model according to the first channel information, including:
- the terminal respectively performs encoding and/or decoding processing through the one network model according to the first channel information, to obtain second channel information corresponding to each network model;
- the terminal calculates and obtains target parameters corresponding to each network model according to the first channel information and the second channel information corresponding to each network model;
- the terminal selects the network with the smallest number of elements in the encoded data obtained after encoding from the M network models a model, determined as the first network model;
- the terminal selects from the one network model the number of elements in the encoded data obtained after encoding processing is the largest
- the network model is determined as the first network model.
- the terminal respectively performs encoding and/or decoding processing through the one network model according to the first channel information, and obtains the information corresponding to each network model.
- the second channel information may include: the terminal uses the first channel information as the input of the encoder in the first network model to obtain coded data after encoding and compression; the terminal may also perform the first processing on the coded data (for example, through the mask vector truncation and/or quantization processing, etc.), and the data after the first processing is subjected to the second processing (for example, the second processing can be understood as the restoration of the first processing, such as through inverse quantization processing and/or masking vector filling, etc.); the terminal takes the data after the second processing as input to the decoder in the first network model, and decodes to obtain restored channel information (that is, the second channel information).
- the terminal respectively performs encoding and/or decoding processing through the one network model to obtain the second channel information corresponding to each network model, which is not limited to the fact that the terminal needs to target All network models obtain corresponding second channel information. That is to say, the terminal can obtain the recovered second channel information through the first channel information for all network models, and then select the first network model according to the target parameters determined by the first channel information and the second channel information; or, the terminal can also Only for some network models, the recovered second channel information is obtained through the first channel information. For example, the terminal performs encoding and/or decoding processing through the I network models in sequence according to the first channel information, and obtains the information corresponding to each network model.
- the second channel information of the i-th network model is used, if the target parameter determined by the i-th network model is judged, and the first network model can be selected from the i-th network model, the following (I-i) network models can no longer The recovered second channel information is obtained through the first channel information, so that the processing efficiency can be improved.
- the terminal calculates target parameters corresponding to each network model according to the first channel information and the second channel information corresponding to each network model, including:
- the terminal calculates and obtains target parameters corresponding to each network model through target criteria according to the first channel information and the second channel information corresponding to each network model;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- Step A1 According to the first channel information, the terminal performs encoding and/or decoding processing on the i-th network model among the I network models to obtain the i-th second channel information;
- Step A2 Calculate and obtain the i-th target parameter according to the first channel information and the i-th second channel information;
- Step A3 If i ⁇ I and there are M network models whose target parameters are less than or equal to the preset threshold, then minimize the number of elements in the output coded data (that is, the length of the coded data) in the M network models
- the network model of is determined as the first network model; wherein, M is a positive integer;
- the corresponding broadband signal-to-interference-noise ratio ⁇ WB,i ⁇ T is calculated according to the i-th network model, it means that the selected i-th network model cannot meet the requirements, and then continue to calculate the i+1th network model corresponding to The wideband signal-to-interference-noise ratio ⁇ WB,i+1 until the calculated ⁇ WB,i ⁇ T .
- the network model that satisfies the requirement of ⁇ WB,i ⁇ ⁇ T and the code word length output by the encoder (that is, the number of elements in the encoded data) is the minimum is determined as the first network model.
- ⁇ T represents a preset threshold value.
- ⁇ WB,i ⁇ ⁇ T calculated according to the i-th network model, although the currently selected network model can meet the requirements, the output codeword length of the network model may be too large.
- the wideband SINR ⁇ WB,i +1 corresponding to the i+1th network model can be continuously calculated until the calculated ⁇ WB,i ⁇ T .
- select the network model with the smallest output codeword length among all the models satisfying ⁇ WB,i ⁇ ⁇ T and determine it as the first network model.
- the terminal may also determine the selected network model according to the uplink transmission resources.
- the downlink signal carries second indication information for indicating the first network model; before the terminal receives the downlink signal sent by the network device, it further includes:
- the terminal sends a second uplink signal to the network device; wherein the second uplink signal is used by the network device to determine the first network model.
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the first network model may be determined by the network device side, and the network device may configure the determined network model to the terminal through signaling, for example, the network device sends network model indication information (ie, second indication information) to the terminal for
- the first network model is used to indicate the encoding and/or decoding processing of the first channel information, so that the terminal can perform compressed feedback of the first channel information according to the first network model determined by the second indication information.
- the terminal before the terminal receives the network model indication information sent by the network device side (that is, the terminal receives the downlink signal sent by the network device), the terminal sends an uplink signal to the network device side at least once.
- the uplink signal can be an uplink reference signal, or carry a terminal
- the calculated PMI (such as Type I (Type I) or Type II (Type II) codebook) is used for selecting the first network model on the network device side.
- the first channel information includes L pieces of channel information, where L is a positive integer; wherein, the first channel information in the L pieces of channel information includes the following item:
- the downlink channel information H l corresponding to the lth receiving antenna of the first frequency domain unit;
- N t is the number of transmitting antenna ports of the network device
- N c is the number of frequency domain units in the first frequency domain unit
- P1 is the number of receiving antennas of the terminal;
- the downlink reference signal received by the lth receiving antenna of the first frequency domain unit passes through the received signal Y l of the channel; wherein, P1 is the number of receiving antennas of the terminal;
- the lth eigenvector U l obtained by singular value decomposition of the downlink channel information of the first frequency domain unit; where, P2 is the number of feature vectors;
- the precoding vector V l of the lth data stream of the first frequency domain unit where, P3 is the number of data streams;
- a combination of R eigenvectors obtained by singular value decomposition of the downlink channel information of the first frequency domain unit; wherein, L 1; for example, the combined matrix is [U 1 ; U 2 ;...; U R ];
- a combination of the precoding vectors of the R data streams of the first frequency domain unit and (v1-R) zero vectors 0j ; wherein, L 1; v1 is the maximum number of data streams that the terminal can transmit; for example, the combined matrix is [V 1 ; V 2 ; ...; V R ; 0 R+1 ; ...; 0 v ];
- a combination of downlink channel information corresponding to the R receiving antennas of the first frequency domain unit and (v2-R) zero vectors; wherein, L 1; v2 is the maximum number of receiving antennas of the terminal; such as combination
- the matrix after is [H 1 ; H 2 ; ...; H R ; 0 R+1 ; ...; 0 v ];
- the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in is the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in
- the first frequency domain unit is all frequency domain units or one frequency domain unit, Indicates that the dimension of the matrix is N c ⁇ N t , where the elements are complex numbers.
- the frequency domain unit is a subcarrier, or the frequency domain unit is a PRB, or the frequency domain unit is a subband including multiple PRBs.
- the terminal sends the encoded data processed data to the network device, including:
- the terminal performs quantization processing on the first element in the coded data, and transmits the quantized data to the network device; where the first element is part or all of the elements in the coded data.
- the terminal inputs it to the first network model corresponding to the l-th channel information, and obtains a code word C (ie encoded data) through the output of the encoder therein.
- the code word C can be expressed as a vector, and the number of elements is K 0 .
- the terminal feeds some or all elements in the codeword C, that is, K 1 ⁇ K 0 elements, after quantization processing, to the network device through the uplink channel (for example, quantization processing is performed through a quantizer module).
- the terminal determines a manner of encoding some or all elements in the data, that is, the first element includes the following one:
- K 1 non-zero elements in the encoded data wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data; for example: if the terminal passes through the lth channel
- the first network model corresponding to the information compresses the l-th channel information (such as CSI) to obtain codewords C i , i ⁇ 1,...,I ⁇ (i.e. coded data), and C i indicates that the selected Codewords compressed by i network models.
- the terminal quantizes K 1 ⁇ 1 non-zero elements in the codeword C i and reports it to the network device side.
- the K1 elements in the coded data determined according to the mask vector for example, the terminal quantizes the elements in the coded data corresponding to the non-zero elements of the mask vector and reports it to the network device.
- the mask vector corresponds to the network model that is, the corresponding mask vector can also be determined when the network model is determined, that is, the mask vector can also be selected by the terminal itself, or configured to the terminal by the network.
- K 1 consecutive elements starting from the first position in the encoded data.
- the terminal quantizes K1 consecutive elements starting from the first position in the coded data and reports it to the network device.
- K 1 may be selected by the terminal itself, or configured to the terminal by the network; the first position may be determined according to a pre-definition, or be reported and indicated by the terminal, or configured to the terminal by the network.
- the first channel information includes L pieces of channel information, and L ⁇ 1; the method further includes the following item:
- the terminal When L ⁇ 1, the terminal sends third indication information to the network device; wherein, one piece of channel information corresponds to one third indication information, and the third indication information is used to indicate the number of the first elements and /or the position of the first element in the encoded data; that is, to independently determine the first element among the L pieces of channel information and independently report the number of the first element and/or the first element The position of an element in the encoded data.
- the terminal for the The first part of channel information is to send common fourth indication information to the network device; where the fourth indication information is used to indicate the number of the first elements and/or the number of the first elements in the encoded data
- the position in ; the first part of the channel information is part or all of the channel information in the L pieces of channel information; that is, for the first part of the channel information in the L pieces of channel information, because the coded data described in The number of the first element and/or the position of the first element in the encoded data are the same, so the number of the first element reported and/or the position of the first element in the coded data can be reported through a common indication information
- the location in the coded data is reported to the network device.
- the terminal sends fifth indication information for the joint indication of the L pieces of channel information to the network device; wherein, the fifth indication information is used to indicate the coded data encoded by each channel information The number of first elements in and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the number of the first element in the L encoded data after the L channel information is encoded The total number and/or the position of the first element in the encoded data.
- the first element includes K1 non-zero elements in the coded data:
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements and the number K 1 of the K 1 non-zero elements in the encoded data Location;
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements.
- the terminal when K 1 ⁇ K 0 , that is, when some elements in the encoded data are non-zero elements, the terminal needs to indicate to the network device the number K 1 of the non-zero elements and the number of K 1 non-zero elements in the The position in the coded data, for example, the terminal can indicate the position of the non-zero element in the code word through a bitmap (bitmap) or combination number, and report it to the network side.
- bitmap bitmap
- the terminal can indicate the number K 1 of the non-zero elements to the network device. At this time, there is no need to indicate that the non-zero elements are The position in the encoded data.
- the third indication information or the fourth indication information includes N3 bits, and the N3 bits are used to indicate the number K 1 of the first elements; wherein,
- the terminal may use N3 bits to indicate the number K 1 of the first elements.
- N3 can be determined according to the number K of all elements in the encoded data, such as
- the terminal also indicates the positions of the K 1 first elements in the encoded data through a bitmap or a combination number.
- the first element includes K 1 elements in the coded data determined according to the mask vector, or in the case of K 1 consecutive elements starting from the first position in the coded data: the The third indication information or the fourth indication information includes N4 bits, and the N4 bits are used to indicate the number K 2 of non-zero elements in the K 1 elements, and the third indication information or the fourth indication information The indication information is also used to indicate the positions of K 2 non-zero elements in the encoded data; where,
- the terminal may use N4 bits to indicate the number K 2 of non-zero elements among the K 1 first elements.
- N4 can be determined according to the number K1 of the first element, such as
- the terminal also indicates the positions of the K 2 non-zero elements to the network device through a bitmap or a combined number. At this time, the terminal quantizes K 2 non-zero elements and reports them to the network device, while other elements are not reported.
- the fifth indication information includes N5 bits, and the N5 bits are used to indicate the number of first elements in the encoded data after each channel information is encoded; wherein, The maximum number of elements reported for each encoded data;
- the fifth indication information includes N6 bits, and the N6 bits are used to indicate the total number of first elements in the L coded data; wherein, It is the maximum total number of reported elements in the L coded data.
- the terminal may jointly determine the number of first elements to be reported in the L pieces of coded data and/or the position of the first element in the coded data. Wherein, the number of first elements reported in different encoded data and/or the position of the first element in the encoded data may be different.
- the number of first elements reported in each coded data and/or the position of the first elements in the coded data may be indicated to the network device in the above-mentioned independent reporting manner; or jointly reported to the network device, such as
- the terminal can use the N5 bit to indicate the number of first elements in the coded data encoded by each channel information, and N5 can be based on the maximum number of elements reported in each coded data OK, as in in, It can be configured to the terminal by the network or pre-determined.
- the terminal can also use the N6 bit to indicate the total number of the first elements in the L coded data, N6 can be based on the maximum total number of elements reported in the L coded data OK, as in in, It can be configured to the terminal by the network or pre-determined. At this time, the terminal may also indicate the position of the first element reported in the different encoded data through the bitmap or combination number corresponding to the different encoded data.
- the above quantization process may be to quantize the value of the first element to a discrete value in a certain value range, and the quantized discrete value of each element is represented by N bits, and a binary bit stream is output.
- the inverse quantization processing required by the network device refers to determining the value of the first element in the encoded data according to the received binary bit stream, and the value of each element is determined by the value represented by N bits.
- the terminal determines the first network model according to the downlink signal, it further includes:
- the terminal obtains the recovered second channel information through the first network model based on the processed data of the encoded data
- the terminal calculates and obtains a channel quality indicator CQI according to the second channel information
- the terminal sends the CQI to the network device.
- the terminal may perform quantization processing on the first element in the encoded data through a quantizer module to obtain quantized data, and then input the quantized data to a corresponding inverse quantizer module for inverse quantization processing, and
- the dequantized data is input to the decoder in the first network model, and the recovered second channel information is obtained by decoding, and then the CQI is calculated according to the recovered second channel information.
- the terminal may process the encoded data through a mask vector module to determine the first element, and perform quantization processing on the first element through a quantizer module to obtain quantized data, and then input the quantized data to
- the corresponding inverse quantizer module performs inverse quantization processing, and inputs the quantized data to the mask vector filling module for processing, and inputs the processed data to the decoder in the first network model, and decodes to obtain the restored first second channel information, and then calculate the CQI according to the restored second channel information.
- the terminal determines the first network model according to the downlink signal, it further includes:
- the terminal receives the beamformed channel information reference signal sent by the network device; wherein the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the terminal determines effective channel information according to the channel information reference signal
- the terminal calculates and obtains the CQI according to the effective channel information
- the terminal sends the CQI to the network device.
- the channel information reference signal can be a CSI Reference Signal (CSI Reference Signal, CSI-RS)
- the terminal estimates downlink effective channel information according to the received beamforming CSI-RS, and then calculates CQI according to the estimated effective channel information, where CSI -
- the beam used by the RS is determined by the channel information recovered by the first network model at the network device side.
- the method also includes:
- the terminal sends the interference and noise related information of the downlink channel to the network device; wherein the interference and noise related information of the downlink channel is used for the network device to calculate the CQI.
- the terminal reports the interference and noise related information of the downlink channel to the network device, so that the network device side calculates the CQI according to the interference and noise related information of the downlink channel.
- codeword involved in the embodiments of the present disclosure is encoded data obtained by encoding channel information, and the channel information may be SCI.
- the method of determining the channel information, the method of determining the first element, the method of reporting the number and/or position of the first element, the CQI calculation method, etc. in the above-mentioned embodiments of the present disclosure can also be applied to the terminal side and the network
- the scenario where the network model used for encoding and/or decoding processing has been determined between the parties for example, the terminal and/or network device side may not perform the step of determining the first network model), in order to avoid repetition, details are not described here.
- the terminal involved in the embodiments of the present disclosure may be a device that provides voice and/or data connectivity to a user, a handheld device with a wireless connection function, or other processing devices connected to a wireless modem.
- the name of the terminal equipment may be different.
- the terminal equipment may be called User Equipment (User Equipment, UE).
- the wireless terminal equipment can communicate with one or more core networks (Core Network, CN) via the radio access network (Radio Access Network, RAN), and the wireless terminal equipment can be a mobile terminal equipment, such as a mobile phone (or called a "cellular "telephones) and computers with mobile terminal equipment, such as portable, pocket, hand-held, computer built-in or vehicle-mounted mobile devices, which exchange language and/or data with the radio access network.
- a mobile terminal equipment such as a mobile phone (or called a "cellular "telephones) and computers with mobile terminal equipment, such as portable, pocket, hand-held, computer built-in or vehicle-mounted mobile devices, which exchange language and/or data with the radio access network.
- PCS Personal Communication Service
- SIP Session Initiated Protocol
- WLL Wireless Local Loop
- PDA Personal Digital Assistant
- Wireless terminal equipment can also be called system, subscriber unit, subscriber station, mobile station, mobile station, remote station, access point , remote terminal (remote terminal), access terminal (access terminal), user terminal (user terminal), user agent (user agent), and user device (user device), which are not limited in the embodiments of the present disclosure.
- the network device involved in the embodiments of the present disclosure may be a base station, and the base station may include multiple cells that provide services for terminals.
- the base station can also be called an access point, or it can be a device in the access network that communicates with the wireless terminal device through one or more sectors on the air interface, or other names.
- the network device can be used to interchange received over-the-air frames with Internet Protocol (IP) packets and act as a router between the wireless terminal device and the rest of the access network, which can include the Internet Protocol (IP) communication network.
- IP Internet Protocol
- Network devices may also coordinate attribute management for the air interface.
- the network equipment involved in the embodiments of the present disclosure may be a network equipment (Base Transceiver Station, BTS) in Global System for Mobile communications (GSM) or Code Division Multiple Access (Code Division Multiple Access, CDMA) ), it can also be a network device (NodeB) in Wide-band Code Division Multiple Access (WCDMA), or it can be an evolved network device in a long-term evolution (long term evolution, LTE) system (evolutional Node B, eNB or e-NodeB), 5G base station (gNB) in the 5G network architecture (next generation system), can also be a home evolved base station (Home evolved Node B, HeNB), relay node (relay node) , a home base station (femto), a pico base station (pico), etc., are not limited in this embodiment of the present disclosure.
- a network device may include a centralized unit (centralized unit, CU) node and a distributed unit (distributed unit, DU) node
- an embodiment of the present disclosure provides a terminal 700, including:
- the first receiving unit 710 is configured to receive a downlink signal sent by the network device
- the first processing unit 720 is configured to determine a first network model according to the downlink signal; wherein, the first network model is at least one of I network models pre-deployed or downloaded from the network device side, and I is positive integer;
- An encoding unit 730 configured to encode the first channel information through the first network model to obtain encoded data to be transmitted;
- the first sending unit 740 is configured to send the processed data of the encoded data to the network device.
- the downlink signal is a downlink reference signal; the first processing unit 720 is further configured to:
- the terminal 700 further includes:
- the second sending unit is configured to send a first uplink signal to the network device; wherein the first uplink signal carries first indication information used to indicate the first network model.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the first indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the first indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the first processing unit 720 is further configured to:
- the first channel information respectively perform encoding and/or decoding processing through the one network model to obtain second channel information corresponding to each network model;
- the terminal selects the network with the smallest number of elements in the encoded data obtained after encoding from the M network models a model, determined as the first network model;
- the terminal selects from the one network model the number of elements in the encoded data obtained after encoding processing is the largest
- the network model is determined as the first network model.
- the first processing unit 720 is further configured to:
- the first channel information and the second channel information corresponding to each network model calculate and obtain the target parameters corresponding to each network model through target criteria;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the terminal 700 further includes:
- a third sending unit configured to send a second uplink signal to the network device; wherein the second uplink signal is used by the network device to determine the first network model.
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the first channel information includes L channel information, and L is a positive integer; wherein, the lth channel information in the L channel information includes the following item:
- the downlink channel information H l corresponding to the lth receiving antenna of the first frequency domain unit;
- N t is the number of transmitting antenna ports of the network device
- N c is the number of frequency domain units in the first frequency domain unit
- P1 is the number of receiving antennas of the terminal;
- the downlink reference signal received by the lth receiving antenna of the first frequency domain unit passes through the received signal Y l of the channel;
- the lth eigenvector U l obtained by singular value decomposition of the downlink channel information of the first frequency domain unit; where, P2 is the number of feature vectors;
- the precoding vector V l of the lth data stream of the first frequency domain unit where, P3 is the number of data streams;
- a combination of downlink channel information respectively corresponding to the R receiving antennas of the first frequency domain unit; wherein, L 1;
- a combination of the precoding vectors of the R data streams of the first frequency domain unit and (v1-R) zero vectors 0j ; wherein, L 1; v1 is the maximum number of data streams that the terminal can transmit;
- a combination of downlink channel information corresponding to the R receiving antennas of the first frequency domain unit and (v2-R) zero vectors; wherein, L 1; v2 is the maximum number of receiving antennas of the terminal;
- the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in is the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in
- the first frequency domain unit is all frequency domain units or one frequency domain unit.
- the frequency domain unit is a subcarrier, or the frequency domain unit is a PRB, or the frequency domain unit is a subband including multiple PRBs.
- the first sending unit 740 is further configured to:
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the first channel information includes L pieces of channel information, and L ⁇ 1; the terminal 700 further includes the following item:
- a fourth sending unit configured to send third indication information to the network device when L ⁇ 1; wherein, one piece of channel information corresponds to one third indication information, and the third indication information is used to indicate the first element and/or the position of the first element in the coded data;
- the fifth sending unit is configured to, when L>1, if the number of the first elements in the coded data after encoding the first part of channel information and/or the position of the first elements in the coded data is the same, For the first part of channel information, send common fourth indication information to the network device; where the fourth indication information is used to indicate the number of the first elements and/or the number of the first elements in The position in the coded data; the first part of channel information is part or all of the channel information in the L pieces of channel information;
- a sixth sending unit configured to send fifth indication information for the joint indication of the L channel information to the network device when L>1; wherein the fifth indication information is used to indicate that each channel information code The number of first elements in the encoded data and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate that in the L encoded data after the L channel information is encoded The total number of first elements and/or the position of the first elements in the encoded data.
- the first element includes K1 non-zero elements in the coded data:
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements and the number K 1 of the K 1 non-zero elements in the encoded data Location;
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements.
- the third indication information or the fourth indication information includes N3 bits, and the N3 bits are used to indicate the number K 1 of the first elements; wherein,
- the first element includes K1 elements in the coded data determined according to the mask vector, or in the case of K1 consecutive elements starting from the first position in the coded data:
- the third indication information or the fourth indication information includes N4 bits, and the N4 bits are used to indicate the number K 2 of non-zero elements in the K 1 elements, and the third indication information or the The fourth indication information is also used to indicate the positions of K 2 non-zero elements in the encoded data; wherein,
- the fifth indication information includes N5 bits, and the N5 bits are used to indicate the number of first elements in the encoded data after each channel information is encoded; wherein, The maximum number of elements reported for each encoded data;
- the fifth indication information includes N6 bits, and the N6 bits are used to indicate the total number of first elements in the L coded data; wherein, It is the maximum total number of reported elements in the L coded data.
- the terminal 700 further includes:
- the second processing unit is configured to obtain the recovered second channel information through the first network model from the processed data according to the encoded data;
- a first calculation unit configured to calculate and obtain a channel quality indicator CQI according to the second channel information
- a seventh sending unit configured to send the CQI to the network device.
- the terminal 700 further includes:
- the second receiving unit is configured to receive the beamformed channel information reference signal sent by the network device; wherein the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- a third processing unit configured to determine effective channel information according to the channel information reference signal
- a second calculation unit configured to calculate and obtain the CQI according to the effective channel information
- An eighth sending unit configured to send the CQI to the network device.
- the terminal 700 further includes:
- a ninth sending unit configured to send the interference and noise related information of the downlink channel to the network device; wherein the interference and noise related information of the downlink channel is used for the network device to calculate the CQI.
- each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
- the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.
- the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a processor-readable storage medium.
- the essence of the technical solution of the present disclosure or the part that contributes to the related technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium.
- a computer device which may be a personal computer, a server, or a network device, etc.
- a processor processor
- the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .
- an information processing device including a memory 801, a transceiver 802, and a processor 803; wherein, the memory 807 is used to store computer programs; Send and receive data under the control of the processor 803; for example, the transceiver 802 is used to receive and send data under the control of the processor 803; the processor 803 is used to read the computer program in the memory 801 and perform the following operations:
- the downlink signal determine a first network model; wherein, the first network model is at least one of 1 network models pre-deployed or downloaded from the network device side, and I is a positive integer;
- the downlink signal is a downlink reference signal
- the processor 803 is configured to read a computer program in the memory 801 and perform the following operations:
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the first indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the first indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the first channel information respectively perform encoding and/or decoding processing through the one network model to obtain second channel information corresponding to each network model;
- the one network model includes M ⁇ 1 network models whose target parameters meet the preset conditions, then select the network model with the smallest number of elements in the encoded data obtained after the encoding process from the M network models, and determine is the first network model;
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the first channel information and the second channel information corresponding to each network model calculate and obtain the target parameters corresponding to each network model through target criteria;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the downlink signal carries second indication information for indicating the first network model
- the processor 803 is configured to read a computer program in the memory 801 and perform the following operations:
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the first channel information includes L pieces of channel information, where L is a positive integer; wherein, the first channel information in the L pieces of channel information includes the following item:
- the downlink channel information H l corresponding to the lth receiving antenna of the first frequency domain unit;
- N t is the number of transmitting antenna ports of the network device
- N c is the number of frequency domain units in the first frequency domain unit
- P1 is the number of receiving antennas of the terminal;
- the downlink reference signal received by the lth receiving antenna of the first frequency domain unit passes through the received signal Y l of the channel;
- the lth eigenvector U l obtained by singular value decomposition of the downlink channel information of the first frequency domain unit; where, P2 is the number of feature vectors;
- the precoding vector V l of the lth data stream of the first frequency domain unit where, P3 is the number of data streams;
- a combination of downlink channel information respectively corresponding to the R receiving antennas of the first frequency domain unit; wherein, L 1;
- a combination of the precoding vectors of the R data streams of the first frequency domain unit and (v1-R) zero vectors 0j ; wherein, L 1; v1 is the maximum number of data streams that the terminal can transmit;
- a combination of downlink channel information corresponding to the R receiving antennas of the first frequency domain unit and (v2-R) zero vectors; wherein, L 1; v2 is the maximum number of receiving antennas of the terminal;
- the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in is the channel information after performing two-dimensional Fourier transform on the channel information of the first frequency domain unit in
- the first frequency domain unit is all frequency domain units or one frequency domain unit.
- the frequency domain unit is a subcarrier, or the frequency domain unit is a PRB, or the frequency domain unit is a subband including multiple PRBs.
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the first channel information includes L pieces of channel information, and L ⁇ 1;
- the processor 803 is configured to read a computer program in the memory 801 and perform one of the following operations:
- the fourth indication information is used to indicate the number of the first elements and/or the position of the first elements in the encoded data ;
- the first part of channel information is part or all of the channel information in the L pieces of channel information;
- the fifth indication information is used to indicate the first in the coded data encoded by each channel information.
- the number of elements and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the total number of first elements in the L encoded data after encoding the L channel information And/or the position of the first element in the encoded data.
- the first element includes K1 non-zero elements in the coded data:
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements and the number K 1 of the K 1 non-zero elements in the encoded data Location;
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements.
- the third indication information or the fourth indication information includes N3 bits, and the N3 bits are used to indicate the number K 1 of the first elements; wherein,
- the first element includes K1 elements in the coded data determined according to the mask vector, or in the case of K1 consecutive elements starting from the first position in the coded data:
- the third indication information or the fourth indication information includes N4 bits, and the N4 bits are used to indicate the number K 2 of non-zero elements in the K 1 elements, and the third indication information or the The fourth indication information is also used to indicate the positions of K 2 non-zero elements in the encoded data; wherein,
- the fifth indication information includes N5 bits, and the N5 bits are used to indicate the number of first elements in the encoded data after each channel information is encoded; wherein, The maximum number of elements reported for each encoded data;
- the fifth indication information includes N6 bits, and the N6 bits are used to indicate the total number of first elements in the L coded data; wherein, It is the maximum total number of reported elements in the L coded data.
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the processor 803 is configured to read the computer program in the memory 801 and perform the following operations:
- the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by the processor 803 and various circuits of the memory represented by the memory 801 are linked together.
- the bus architecture can also link together various other circuits such as peripherals, voltage regulators, and power management circuits, etc., which are well known in the art and therefore will not be further described herein.
- the bus interface provides the interface.
- Transceiver 802 may be a plurality of elements, including a transmitter and a receiver, providing a unit for communicating with various other devices over transmission media, including wireless channels, wired channels, optical cables, and other transmission media.
- the user interface 804 may also be an interface capable of connecting externally and internally to required equipment, and the connected equipment includes but not limited to a keypad, a display, a speaker, a microphone, a joystick, and the like.
- the processor 803 is responsible for managing the bus architecture and general processing, and the memory 801 can store data used by the processor 803 when performing operations.
- the processor 803 can be a central processing unit (Central Processing Unit, CPU), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or a complex programmable logic device (Complex Programmable Logic Device, CPLD), the processor can also adopt a multi-core architecture.
- CPU Central Processing Unit
- ASIC Application Specific Integrated Circuit
- FPGA Field-Programmable Gate Array
- CPLD Complex Programmable Logic Device
- An embodiment of the present disclosure also provides a processor-readable storage medium, the processor-readable storage medium stores a computer program, and the computer program is used to enable the processor to execute the steps in the above information processing method, and can To achieve the same technical effect, the same parts and beneficial effects in this embodiment as in the method embodiment will not be described in detail here.
- the processor-readable storage medium may be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (such as floppy disk, hard disk, magnetic tape, magneto-optical disk (Magneto-Optical Disk, MO) etc.) , optical storage (such as compact disc (Compact Disk, CD), digital video disc (Digital Versatile Disc, DVD), Blu-ray Disc (Blu-ray Disc, BD), high-definition universal disc (High-Definition Versatile Disc, HVD), etc.), And semiconductor memory (such as read-only memory (Read-Only Memory, ROM), erasable programmable read-only memory (Erasable Programmable ROM, EPROM), charged erasable programmable read-only memory (Electrically EPROM, EEPROM), non-volatile Non-volatile memory (NAND FLASH), solid state hard disk (Solid State Disk or Solid State Drive, SSD)), etc.
- magnetic storage such as floppy disk, hard disk
- an embodiment of the present disclosure provides an information processing method, including the following steps:
- Step 91 The network device sends a downlink signal to the terminal.
- the downlink signal is used by the terminal to determine a first network model, where the first network model is at least one of I pre-deployed network models, where I is a positive integer.
- the downlink signal may be a downlink reference signal, and the downlink reference signal may be used by the terminal to determine a channel corresponding to the downlink reference information from one network model pre-deployed or downloaded from the network device side
- the first network model for encoding information or, the downlink signal may carry first indication information for indicating the first network model.
- Step 92 The network device receives the data after the terminal processes the coded data.
- the encoded data is obtained by the terminal after encoding the first channel information according to the first network model.
- the first channel information may include L pieces of channel information, the channel information may be CSI, and L is a positive integer; wherein, when L is greater than 1, the terminal side may adopt the same network model for different channel information Perform encoding (or compression) processing, or use different network models for encoding processing for different channel information. For example, after encoding the first channel information through the network model, the code corresponding to each channel information can be obtained Words (that is, encoded data to be transmitted) to reduce the feedback overhead of the terminal. Further, the terminal processes the L encoded data obtained by encoding the L pieces of channel information and reports it to the network device, so that the network device receives the data obtained by the terminal processing the encoded data.
- the code corresponding to each channel information can be obtained Words (that is, encoded data to be transmitted) to reduce the feedback overhead of the terminal.
- the terminal processes the L encoded data obtained by encoding the L pieces of channel information and reports it to the network device, so that the network device receives the data obtained by the terminal processing the encoded data
- Step 93 The network device determines the data to be decoded according to the data after the terminal processes the encoded data.
- the terminal may quantize part or all of the elements in the coded data and send them to the network device.
- the network device needs to perform inverse quantization processing on the received data to determine the corresponding coded data.
- Step 94 The network device performs decoding processing on the data to be decoded according to the first network model to obtain restored channel information.
- the network device sends to the terminal a downlink signal for the terminal to determine the first network model, and when receiving the data processed by the terminal on the coded data encoded by the first network model, according to the The terminal processes the encoded data, determines the data to be decoded, and performs decoding processing on the data to be decoded according to the first network model to obtain restored channel information, thereby ensuring that the terminal side and/or network device side
- the terminal side and/or the network device side can determine the network model used for channel information transmission from multiple pre-deployed network models, and can reduce the feedback overhead of channel information transmission and improve system performance.
- the downlink signal is a downlink reference signal
- the network device decodes the data to be decoded according to the first network model, and before obtaining the restored channel information, further includes:
- the network device receives the first uplink signal sent by the terminal
- the network device determines the first network model according to the first uplink signal.
- the terminal side selects the network model (namely the first network model) used for channel information encoding and/or decoding processing from one network model
- the terminal determines that the first channel information encoding and/or after decoding the processed first network model
- the terminal sends to the network device first indication information for indicating the first network model, where the first indication information is carried in the first uplink signal, so that The network device may learn the first network model for encoding and/or decoding processing of the first channel information according to the first uplink signal.
- the first channel information includes L pieces of channel information, where L is a positive integer;
- the first indication information when different channel information is encoded and/or decoded through different network models, the first indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information; ,
- N1 may be determined according to the number L of channel information and the number I of network models. For example: in Indicates that L network models are selected from I network models, and L represents the number of transmission layers (that is, the number of channel information), so that the network device can determine the first network model through the N1 bits.
- the first indication information when different channel information is encoded and/or decoded through the same network model, the first indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the terminal when the terminal sends the first network model for instructing it to encode and/or decode the first channel information to the network device, it may specifically use N2 bits to indicate the same network model corresponding to different channel information.
- N2 can be determined according to the number I of network models. For example: Therefore, the network device can determine the first network model through the N1 bits.
- the downlink signal carries second indication information for indicating the first network model; before the network device sends the downlink signal to the terminal, it further includes:
- the network device receives the second uplink signal sent by the terminal
- the network device determines third channel information according to the second uplink signal
- the network device determines the first network model according to the third channel information.
- the network model i.e. the first network model
- the network device determines that the network model used for the first channel After the first network model of information encoding and/or decoding processing
- the network device sends second indication information to the terminal, so that the terminal learns the first channel information encoding and/or decoding processing according to the second indication information.
- a network model i.e. the first network model
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the terminal before the terminal receives the network model indication information (that is, the second indication information) sent by the network equipment side, the terminal sends an uplink signal to the network equipment side at least once, and the uplink signal may be an uplink reference signal, or carry the PMI calculated by the terminal (such as Type I or Type II codebook).
- the network model indication information that is, the second indication information
- the terminal sends an uplink signal to the network equipment side at least once, and the uplink signal may be an uplink reference signal, or carry the PMI calculated by the terminal (such as Type I or Type II codebook).
- the network device side estimates the uplink channel information (that is, determines the third channel information) according to the uplink reference signal. If the uplink and downlink channels are completely reciprocal, the network device can determine the corresponding first network model according to the estimated uplink channel information. Alternatively, the network device side determines the corresponding first network model according to the received PMI, and configures the selected first network model to the terminal through signaling, such as the network device sending network model indication information (ie, second indication information) to the terminal It is used to indicate the first network model for encoding and/or decoding processing of the first channel information, so that the terminal can perform compressed feedback of the first channel information according to the first network model determined according to the second indication information.
- network model indication information ie, second indication information
- the network device may select an appropriate first network model from one network model in a traversal manner; specifically, the network device determines the first network model according to the third channel information, including:
- the network device respectively performs encoding and/or decoding processing through the one network model according to the third channel information, to obtain fourth channel information corresponding to each network model;
- the network device calculates and obtains target parameters corresponding to each network model according to the third channel information and the fourth channel information corresponding to each network model;
- the network device selects the one with the smallest number of elements in the encoded data obtained after encoding from the M network models A network model, determined as the first network model;
- the network device selects the number of elements in the encoded data obtained after encoding from the one network model The largest network model is determined as the first network model.
- the network device respectively performs encoding and/or decoding processing through the one network model according to the third channel information to obtain each network model
- the corresponding fourth channel information may include: the network device uses the third channel information as the input of the encoder in the first network model to obtain coded data after encoding and compression; the network device may also perform the first processing on the coded data (such as through Mask vector truncation and/or quantization processing, etc.), and the data after the first processing is subjected to the second processing (for example, the second processing can be understood as the restoration of the first processing, such as through inverse quantization processing and/or or mask vector filling, etc.); the network device takes the data after the second processing as input to the decoder in the first network model, and decodes to obtain restored channel information (ie, fourth channel information).
- the network device respectively performs encoding and/or decoding processing through the one network model to obtain the fourth channel information corresponding to each network model, which is not limited to the network device Corresponding fourth channel information needs to be obtained for all network models.
- the network device can respectively obtain the restored fourth channel information through the third channel information for all network models, and then select the first network model according to the target parameters determined by the third channel information and the fourth channel information; or, the network device It is also possible to obtain the restored fourth channel information through the third channel information only for some network models, for example: the network device performs encoding and/or decoding processing through the one network model in turn according to the third channel information, and obtains each For the fourth channel information corresponding to the network model, if the target parameter determined by the i-th network model can be judged and the first network model can be selected from the i network model, the following (I-i) networks can be The model no longer obtains the restored fourth channel information through the third channel information, so that the processing efficiency can be improved.
- the network device calculates target parameters corresponding to each network model according to the third channel information and the fourth channel information corresponding to each network model, including:
- the network device calculates and obtains target parameters corresponding to each network model through target criteria according to the third channel information and the fourth channel information corresponding to each network model;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- Step A1 The network device performs encoding and/or decoding processing on the i-th network model among the I network models according to the third channel information, to obtain the i-th fourth channel information;
- Step A2 Calculate and obtain the i-th target parameter according to the third channel information and the i-th fourth channel information;
- Step A3 If i ⁇ I and there are M network models whose target parameters are less than or equal to the preset threshold, then minimize the number of elements in the output coded data (that is, the length of the coded data) in the M network models
- the network model of is determined as the first network model; wherein, M is a positive integer;
- the corresponding broadband signal-to-interference-noise ratio ⁇ WB,i ⁇ T is calculated according to the i-th network model, it means that the selected i-th network model cannot meet the requirements, and then continue to calculate the i+1th network model corresponding to The wideband signal-to-interference-noise ratio ⁇ WB,i+1 until the calculated ⁇ WB,i ⁇ T .
- the network model that satisfies the requirement of ⁇ WB,i ⁇ ⁇ T and the code word length output by the encoder (that is, the number of elements in the encoded data) is the minimum is determined as the first network model.
- ⁇ T represents a preset threshold value.
- ⁇ WB,i ⁇ ⁇ T calculated according to the i-th network model, although the currently selected network model can meet the requirements, the output codeword length of the network model may be too large.
- the wideband SINR ⁇ WB,i +1 corresponding to the i+1th network model can be continuously calculated until the calculated ⁇ WB,i ⁇ T .
- select the network model with the smallest output codeword length among all the models satisfying ⁇ WB,i ⁇ ⁇ T and determine it as the first network model.
- the network device may also determine the selected network model according to the uplink transmission resources of the terminal.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the second indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- N1 may be determined according to the number L of channel information and the number I of network models. For example: in Indicates that L network models are selected from I network models, and L represents the number of transmission layers (that is, the number of channel information).
- the second indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- N2 can be determined according to the number I of network models. For example:
- the first channel information includes L pieces of channel information, and L ⁇ 1; the network device further includes the following item before determining the data to be decoded according to the data after the terminal processes the encoded data:
- the network device receives third indication information sent by the terminal; wherein, one piece of channel information corresponds to one third indication information, and the third indication information is used to indicate the number of first elements and/or or the position of the first element in the coded data, where the first element is part or all of the elements in the coded data; that is, the terminal independently determines the first element and Independently report the number of the first elements and/or the position of the first elements in the encoded data.
- the network device receives common fourth indication information sent by the terminal for the first part of channel information; wherein the first part of channel information is part or all of the L pieces of channel information , and the number of the first element and/or the position of the first element in the encoded data in the encoded data after the encoding of the first partial channel information is the same; the fourth indication information is used to indicate the The number of the first elements and/or the position of the first elements in the coded data; that is, for the first part of channel information in the L pieces of channel information, because the first part of the coded data in the coded data The number of elements and/or the position of the first element in the encoded data are the same, so the terminal can use a common indication information to report the number of first elements and/or the position of the first element in The location in the coded data is reported to the network device.
- the network device receives the fifth indication information sent by the terminal for the joint indication of the L channel information; wherein the fifth indication information is used to indicate the encoded code of each channel information The number of first elements in the data and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the first element in the L encoded data after the L channel information is encoded and/or the position of the first element in the encoded data.
- the first element includes K1 non-zero elements in the coded data:
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements and the number K 1 of the K 1 non-zero elements in the encoded data Location;
- the third indication information or the fourth indication information is used to indicate the number K 1 of the non-zero elements.
- the terminal when K 1 ⁇ K 0 , that is, when some elements in the encoded data are non-zero elements, the terminal needs to indicate to the network device the number K 1 of the non-zero elements and the number of K 1 non-zero elements in the The position in the coded data, for example, the terminal can indicate the position of the non-zero element in the code word through a bitmap (bitmap) or combination number, and report it to the network side.
- bitmap bitmap
- the terminal can indicate the number K 1 of the non-zero elements to the network device. At this time, there is no need to indicate that the non-zero elements are The position in the encoded data.
- the third indication information or the fourth indication information includes N3 bits, and the N3 bits are used to indicate the number K 1 of the first elements; wherein,
- the terminal may use N3 bits to indicate the number K 1 of the first elements.
- N3 can be determined according to the number K of all elements in the encoded data, such as
- the terminal also indicates the positions of the K 1 first elements in the encoded data through a bitmap or a combination number.
- the first element includes K 1 elements in the coded data determined according to the mask vector, or in the case of K 1 consecutive elements starting from the first position in the coded data: the The third indication information or the fourth indication information includes N4 bits, and the N4 bits are used to indicate the number K 2 of non-zero elements in the K 1 elements, and the third indication information or the fourth indication information The indication information is also used to indicate the positions of K 2 non-zero elements in the encoded data; where,
- the terminal may use N4 bits to indicate the number K 2 of non-zero elements among the K 1 first elements.
- N4 can be determined according to the number K1 of the first element, such as
- the terminal also indicates the positions of the K 2 non-zero elements to the network device through a bitmap or a combined number. At this time, the terminal quantizes K 2 non-zero elements and reports them to the network device, while other elements are not reported.
- the fifth indication information includes N5 bits, and the N5 bits are used to indicate the number of first elements in the encoded data after each channel information is encoded; wherein, The maximum number of elements reported for each encoded data;
- the fifth indication information includes N6 bits, and the N6 bits are used to indicate the total number of first elements in the L coded data; wherein, It is the maximum total number of reported elements in the L coded data.
- the terminal may jointly determine the number of first elements to be reported in the L pieces of coded data and/or the position of the first element in the coded data. Wherein, the number of first elements reported in different encoded data and/or the position of the first element in the encoded data may be different.
- the number of first elements reported in each coded data and/or the position of the first elements in the coded data may be indicated to the network device in the above-mentioned independent reporting manner; or jointly reported to the network device, such as
- the terminal can use the N5 bit to indicate the number of first elements in the coded data encoded by each channel information, and N5 can be based on the maximum number of elements reported in each coded data OK, as in in, It can be configured to the terminal by the network or pre-determined.
- the terminal can also use the N6 bit to indicate the total number of the first elements in the L coded data, N6 can be based on the maximum total number of elements reported in the L coded data OK, as in in, It can be configured to the terminal by the network or pre-determined. At this time, the terminal may also indicate the position of the first element reported in the different encoded data through the bitmap or combination number corresponding to the different encoded data.
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the network device determines the data to be decoded according to the data after the terminal processes the encoded data, including:
- the network device performs dequantization processing on the encoded data processed by the terminal to obtain dequantized data
- the network device determines the data to be decoded by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data.
- the network device side uses the quantized data reported by the terminal as the input of the dequantizer module to obtain the dequantized data, and then according to the dequantized data and the used to indicate the number of first elements And/or the third indication information or the fourth indication information or the fifth indication information of the position to obtain the restored encoded data (that is, the data to be decoded), and decode the data to be decoded by the decoder of the corresponding first network model , and finally decoded to obtain the recovered channel information.
- the network device side uses the quantized data reported by the terminal as the input of the dequantizer module to obtain the dequantized data, and then inputs the dequantized data to the mask vector filling module to obtain the restored
- the coded data (that is, the data to be decoded) is decoded by the corresponding decoder of the first network model, and the recovered channel information is finally decoded.
- the network device determines the data to be decoded by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data, including the following: item:
- the network device determines the data to be decoded through a mask vector according to the dequantized data; for example, the network device side may determine the data to be decoded according to a mask vector, wherein the mask vector corresponds to a network model, That is, when the network model is determined, the corresponding mask vector can also be determined, that is, the mask vector can also be selected by the terminal itself, or configured to the terminal by the network; for example, the network device can process the encoded data according to the terminal Data, the data to be decoded is output through the mask vector filling module.
- the data to be decoded can be the first K 1 or the last K 1 elements after dequantization processing, and the remaining K 0 -K 1 elements are filled with 0.
- the network device determines the data to be decoded by the number of the first elements according to the dequantized data;
- the information determines the number K 1 of the first elements reported by the terminal.
- the network device can output the data to be decoded through the mask vector filling module according to the data processed by the terminal on the encoded data.
- the data to be decoded can be the first K 1 or the last K 1 elements are dequantized data, and the remaining K 0 -K 1 elements are filled with 0s.
- the network device determines the data to be decoded by the position of the first element in the coded data according to the dequantized data; for example, the terminal reports a bitmap indicating the coded channel information The position of the non-zero element in the data.
- the network device determines the data to be decoded according to the position of the non-zero element indicated by the bitmap and the data after dequantization processing. For example, the position of the non-zero element in the data to be decoded is the dequantized data. Data, the remaining position is 0.
- the network device determines the data to be decoded according to the number of non-zero elements and the position of the non-zero elements in the encoded data according to the dequantized data; for example, the network device side can according to The third indication information or the fourth indication information or the fifth indication information determines the number of non-zero elements K 2 ⁇ K 0 and the position of the non-zero elements reported by the terminal, such as the network device can process the encoded data according to the terminal
- the data to be decoded is output through the mask vector filling module.
- the data to be decoded can be dequantized data at K 2 non-zero element positions, and 0 is added to K 0 -K 2 element positions.
- the network device uses the number of the first elements, the number of non-zero elements in the first elements, and the number of non-zero elements in the encoded data to determine the data to be decoded; for example, the network device side can determine the number K 1 and K 1 of the first elements reported by the terminal according to the third indication information or the fourth indication information or the fifth indication information
- the number K of non-zero elements in the element and the position of the non-zero element such as the data that the network device can process the encoded data according to the terminal, output the data to be decoded through the mask vector filling module, and the data to be decoded can be It is the dequantized data at the positions of K 2 non-zero elements, and the remaining K 0 -K 1 elements and K 1 -K 2 element positions are filled with 0.
- the method also includes:
- the network device receives the channel quality indicator CQI sent by the terminal.
- the terminal obtains the recovered second channel information through the first network model according to the data obtained by processing the encoded data after channel information encoding; calculates and obtains the CQI according to the second channel information, And send the CQI to the network device.
- the terminal may perform quantization processing on the first element in the encoded data through a quantizer module to obtain quantized data, and then input the quantized data to a corresponding inverse quantizer module for inverse quantization processing, and
- the dequantized data is input to the decoder in the first network model, and the recovered second channel information is obtained by decoding, and then the CQI is calculated according to the recovered second channel information.
- the terminal may process the encoded data through a mask vector module to determine the first element, and perform quantization processing on the first element through a quantizer module to obtain quantized data, and then input the quantized data to
- the corresponding inverse quantizer module performs inverse quantization processing, and inputs the quantized data to the mask vector filling module for processing, and inputs the processed data to the decoder in the first network model, and decodes to obtain the restored first second channel information, and then calculate the CQI according to the restored second channel information.
- the method also includes:
- the network device sends a beamformed channel information reference signal to the terminal; wherein the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the network device receives the CQI sent by the terminal; wherein, the CQI is calculated by the terminal according to the channel information reference signal.
- the terminal receives a channel information reference signal for beamforming; determines effective channel information according to the channel information reference signal; calculates a CQI according to the effective channel information, and sends the CQI to the network device.
- the channel information reference signal can be a CSI Reference Signal (CSI Reference Signal, CSI-RS)
- the terminal estimates downlink effective channel information according to the received beamforming CSI-RS, and then calculates CQI according to the estimated effective channel information, where CSI -
- the beam used by the RS is determined by the channel information recovered by the first network model at the network device side.
- the method also includes:
- the network device receives the interference and noise related information of the downlink channel sent by the terminal;
- the network device calculates and obtains the CQI according to the interference and noise related information of the downlink channel and the recovered channel information.
- the information processing methods on the terminal side and the network device side are corresponding, and the two embodiments may refer to each other, and repeated descriptions will not be repeated.
- an embodiment of the present disclosure provides a network device 1000, including:
- the first sending unit 1010 is configured to send a downlink signal to the terminal; wherein the downlink signal is used by the terminal to determine a first network model, where the first network model is at least one of the pre-deployed 1 network models, I is a positive integer;
- the first receiving unit 1020 is configured to receive data obtained by the terminal after processing the encoded data; wherein the encoded data is obtained by the terminal after encoding the first channel information according to the first network model;
- the first processing unit 1030 is configured to determine the data to be decoded according to the data processed by the terminal on the encoded data;
- the decoding unit 1040 is configured to perform decoding processing on the data to be decoded according to the first network model to obtain restored channel information.
- the downlink signal is a downlink reference signal; the network device 1000 further includes:
- a second receiving unit configured to receive the first uplink signal sent by the terminal
- the second processing unit is configured to determine the first network model according to the first uplink signal.
- the downlink signal carries second indication information for indicating the first network model; the network device 1000 further includes:
- a third receiving unit configured to receive a second uplink signal sent by the terminal
- a third processing unit configured to determine third channel information according to the second uplink signal
- a fourth processing unit configured to determine the first network model according to the third channel information.
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the fourth processing unit is also used for:
- the third channel information respectively perform encoding and/or decoding processing through the one network model to obtain fourth channel information corresponding to each network model;
- the network device selects the one with the smallest number of elements in the encoded data obtained after encoding from the M network models A network model, determined as the first network model;
- the network device selects the number of elements in the encoded data obtained after encoding from the one network model The largest network model is determined as the first network model.
- the fourth processing unit is also used for:
- the third channel information and the fourth channel information corresponding to each network model calculate and obtain the target parameters corresponding to each network model through target criteria;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the second indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the second indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the first channel information includes L pieces of channel information, and L ⁇ 1; the network device 1000 further includes the following item:
- the fourth receiving unit is configured to receive the third indication information sent by the terminal when L ⁇ 1; wherein, one channel information corresponds to one third indication information, and the third indication information is used to indicate the number of the first element number and/or the position of the first element in the coded data, the first element being part or all of the elements in the coded data;
- the fifth receiving unit is configured to receive common fourth indication information sent by the terminal for the first part of channel information when L>1; wherein the first part of channel information is a part or part of the L pieces of channel information All channel information, and the number of the first elements in the encoded data encoded by the first partial channel information and/or the position of the first element in the encoded data are the same; the fourth indication information uses to indicate the number of the first elements and/or the position of the first elements in the encoded data;
- the sixth receiving unit is configured to receive fifth indication information sent by the terminal for the joint indication of the L channel information when L>1; wherein the fifth indication information is used to indicate that each channel information code The number of first elements in the encoded data and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate that in the L encoded data after the L channel information is encoded The total number of first elements and/or the position of the first elements in the encoded data.
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the first processing unit 1030 is further configured to:
- the data to be decoded is determined by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data.
- the first processing unit 1030 is also used for one of the following:
- the data to be decoded is determined by the number of non-zero elements and the position of the non-zero elements in the encoded data;
- the number of the first elements, the number of non-zero elements in the first elements and the position of the non-zero elements in the coded data are used to determine the Data to be decoded.
- the network device 1000 also includes:
- the seventh receiving unit is configured to receive the channel quality indicator CQI sent by the terminal.
- the network device 1000 also includes:
- the second sending unit is configured to send a beamformed channel information reference signal to the terminal; wherein, the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the eighth receiving unit is configured to receive the CQI sent by the terminal; wherein the CQI is calculated by the terminal according to the channel information reference signal.
- the network device 1000 also includes:
- a ninth receiving unit configured to receive interference and noise-related information of a downlink channel sent by the terminal
- a calculation unit configured to calculate and obtain the CQI according to the interference and noise related information of the downlink channel and the recovered channel information.
- each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
- the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.
- the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a processor-readable storage medium.
- the essence of the technical solution of the present disclosure or the part that contributes to the related technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium.
- a computer device which may be a personal computer, a server, or a network device, etc.
- a processor processor
- the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .
- this embodiment provides a network device, including a memory 1101, a transceiver 1102, and a processor 1103; wherein, the memory 1101 is used to store computer programs; the transceiver 1102 is used to control the processor 1103 Send and receive data; For example, the transceiver 1102 is used to receive and send data under the control of the processor 1103; the processor 1103 is used to read the computer program in the memory 1101 and perform the following operations:
- the downlink signal is used by the terminal to determine a first network model, where the first network model is at least one of the pre-deployed I network models, where I is a positive integer;
- the downlink signal is a downlink reference signal
- the processor 1103 is configured to read a computer program in the memory 1101 and perform the following operations:
- the downlink signal carries second indication information for indicating the first network model
- the processor 1103 is configured to read a computer program in the memory 1101 and perform the following operations:
- the second uplink signal is an uplink reference signal; or, the second uplink signal carries a precoding matrix indication PMI.
- the processor 1103 is configured to read the computer program in the memory 1101 and perform the following operations:
- the third channel information respectively perform encoding and/or decoding processing through the one network model to obtain fourth channel information corresponding to each network model;
- the one network model includes M ⁇ 1 network models whose target parameters meet the preset conditions, then select the network model with the smallest number of elements in the encoded data obtained after the encoding process from the M network models, and determine is the first network model;
- the processor 1103 is configured to read the computer program in the memory 1101 and perform the following operations:
- the third channel information and the fourth channel information corresponding to each network model calculate and obtain the target parameters corresponding to each network model through target criteria;
- the target criterion is a cosine similarity criterion, or the target criterion is a normalized mean square error criterion, or the target criterion is a broadband signal-to-interference-noise ratio criterion.
- the first channel information includes L pieces of channel information, where L is a positive integer; where,
- the second indication information includes N1 bits, and the N1 bits are used to indicate network models corresponding to different channel information;
- the second indication information includes N2 bits, and the N2 bits are used to indicate the same network model; wherein,
- the first channel information includes L pieces of channel information, and L ⁇ 1;
- the processor 1103 is configured to read a computer program in the memory 1101 and perform one of the following operations:
- receiving common fourth indication information sent by the terminal for the first part of channel information wherein the first part of channel information is part or all of the L pieces of channel information, and the The number of the first element and/or the position of the first element in the encoded data in the encoded data after encoding the first part of channel information is the same; the fourth indication information is used to indicate the first element and/or the position of the first element in the coded data;
- the fifth indication information is used to indicate the first in the coded data encoded by each channel information The number of elements and/or the position of the first element in the encoded data; or, the fifth indication information is used to indicate the total number of first elements in the L encoded data after encoding the L channel information And/or the position of the first element in the encoded data.
- the first element includes one of the following:
- K 1 non-zero elements in the encoded data; wherein, K 1 is a positive integer, K 1 ⁇ K 0 , and K 0 is the number of all elements in the encoded data;
- the processor 1103 is configured to read the computer program in the memory 1101 and perform the following operations:
- the data to be decoded is determined by using the third indication information, the fourth indication information, or the fifth indication information according to the dequantized data.
- the processor 1103 is configured to read the computer program in the memory 1101 and perform one of the following operations:
- the number of the first elements, the number of non-zero elements in the first elements and the position of the non-zero elements in the coded data are used to determine the Data to be decoded.
- the processor 1103 is configured to read the computer program in the memory 1101 and perform the following operations:
- the processor 1103 is configured to read the computer program in the memory 1101 and perform the following operations:
- the beam used by the channel information reference signal is determined according to the channel information recovered by the network device;
- the processor 1103 is configured to read the computer program in the memory 1101 and perform the following operations:
- the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by the processor 1103 and various circuits of the memory represented by the memory 1102 are linked together.
- the bus architecture can also link together various other circuits such as peripherals, voltage regulators, and power management circuits, etc., which are well known in the art and therefore will not be further described herein.
- the bus interface provides the interface.
- Transceiver 1102 may be a plurality of elements, including a transmitter and a receiver, providing means for communicating with various other devices over transmission media, including wireless channels, wired channels, fiber optic cables, etc. Transmission medium.
- the processor 1103 is responsible for managing the bus architecture and general processing, and the memory 1101 can store data used by the processor 1103 when performing operations.
- the processor 1103 can be a central processing unit (Central Processing Unit, CPU), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or a complex programmable Logic device (Complex Programmable Logic Device, CPLD), the processor can also adopt a multi-core architecture.
- CPU Central Processing Unit
- ASIC Application Specific Integrated Circuit
- FPGA Field-Programmable Gate Array
- CPLD Complex Programmable Logic Device
- the processor is used to execute any one of the methods provided by the embodiments of the present disclosure according to the obtained executable instructions by calling the computer program stored in the memory.
- the processor and memory may also be physically separated.
- An embodiment of the present disclosure also provides a processor-readable storage medium, the processor-readable storage medium stores a computer program, and the computer program is used to enable the processor to execute the steps in the above information processing method, and can To achieve the same technical effect, the same parts and beneficial effects in this embodiment as in the method embodiment will not be described in detail here.
- the processor-readable storage medium can be any available medium or data storage device that can be accessed by a processor, including but not limited to magnetic storage (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical storage (e.g., CD, DVD, BD, HVD, etc.), and semiconductor memory (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)), etc.
- magnetic storage e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.
- optical storage e.g., CD, DVD, BD, HVD, etc.
- semiconductor memory such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)
- Embodiment 1 Deploying a network model including an encoder and a decoder on the network side or the terminal side
- the terminal side determines the first network model, for example, the UE (that is, the terminal) downloads one network model from the gNB (that is, the network device) for reasoning and implementing CSI compression feedback.
- the UE estimates downlink channel state information h f on different frequency domain units according to the received CSI-RS, and each frequency domain unit may be a subband, PRB or subcarrier, etc.
- N t is the number of transmit antenna ports on the network side
- N c is the number of frequency domain units.
- the quantized codeword is input to an inverse quantizer module, and the corresponding non-quantized information is output. Finally, the output non-quantized information is used as the input of the i-th decoder, and the output of the encoder is the restored downlink channel, as shown in FIG. 12 .
- the UE uses the recovered downlink channel h′ f,i to calculate the SINR ⁇ of each frequency domain unit and the corresponding wideband SINR ⁇ WB,i . It is assumed that the minimum value of the wideband SINR preset or configured by the gNB to the UE is ⁇ T . If ⁇ WB,i ⁇ T calculated by the UE indicates that the i-th model selected by the UE cannot meet the requirements, then the UE selects the i+1th network model and inputs the calculated downlink channel h f,i to the i+1th network model In one network model, the restored downlink channel h′ f,i+1 output by the decoder corresponding to the i+1th network model is finally obtained.
- the UE calculates the wideband ⁇ WB, i+1 according to h′ f, i+1 . If ⁇ WB,i+1 ⁇ ⁇ T , the network model meets the requirements, otherwise UE will select the i+2th network model until the calculated wideband SINR is not less than ⁇ T according to the selected network model , and send the selected optimal network model to gNB through the uplink channel. If the wideband SINRs calculated according to all I network models do not meet the requirements, the UE selects a network model corresponding to the largest wideband SINR for CSI compression feedback.
- the gNB determines the network model selected by the UE according to the received codeword information with a length of ki , thereby determining the decoder used to recover the downlink channel, and then uses the selected decoding
- the device deduces the feature vector h′ f , which is used for precoding of downlink data transmission.
- Embodiment 2 CSI compression feedback of the first layer based on AI
- the UE estimates the downlink channel state information H f on different frequency domain units according to the received CSI-RS, and then performs singular value decomposition on the channel state information of the fth frequency domain unit to obtain the corresponding eigenvector v f .
- Each frequency domain unit may also be a subband, PRB or subcarrier, etc.
- a matrix composed of eigenvectors corresponding to the largest eigenvalues of all frequency domain unit downlink channels As the input of the adaptive encoder, the output of the encoder is a codeword C containing K 0 elements, as shown in Figure 13.
- Mode 1 Assuming that gNB configures a mask vector M 1 for the UE with the first K bits as 1 and the last K 0 -K 1 bits as 0, then input the codeword into the mask vector module, and finally output A codeword of length K1 . That is, the codeword C is obtained after passing through the mask vector module
- Z masking is equivalent to 0 for the last K 0 -K 1 element position in codeword C, and only retains the content of the first K 1 bit elements in codeword C, that is, the truncated information length is K 1 ;
- the UE then inputs the truncated codeword information to the quantization module.
- the UE reports the quantized binary bit stream information to the gNB.
- gNB inputs the bit stream information to the dequantizer module to obtain the quantized value of K 1 elements, and then adds K 0 -K 1 elements that are all 0 to the quantized value of the K 1 elements and K 1 elements , get the code word C'. Finally, gNB inputs the code word C' to the adaptive decoder, and uses the reasoning of the adaptive decoder network model to obtain the restored feature vector v′ f,i , which is used for precoding of downlink data transmission.
- Mode 2 Assuming that the above mask vector is selected and determined by the UE, the method for the UE to select the mask vector is the same as the method for the UE to select the network model. UE needs to pass The bit indicates that the number of all preceding 1s in the mask vector is K 1 , and the indication information is sent to the gNB through the uplink channel. The gNB determines the mask vector used by the UE according to the indication information, and also determines that the output information filled with the mask vector is that the first K 1 elements are the output information of the inverse quantizer module, and the last K 0 -K 1 elements are all 0 element. Other UE and gNB processing procedures are the same as above.
- the gNB determines the number of non-zero elements according to the indication information of the number of non-zero elements, and then determines the position of the non-zero elements according to the indication information of the position of the non-zero elements. gNB determines the quantized value according to the number of non-zero elements, and the quantized value is used as the input of the mask vector filling module. gNB determines that the output information of the mask vector filling module is that the reported non-zero element position is the quantized value and then fills all positions other than the reported non-zero element with 0.
- Embodiment 3 AI-based CSI compression feedback of L>1 layer
- the UE estimates the downlink channel state information H f on different frequency domain units according to the received CSI-RS, and then performs singular value decomposition on the channel state information of the fth frequency domain unit to obtain the corresponding eigenvector v f .
- the gNB configures the maximum number of reported elements in each codeword as According to the estimated downlink channel, the UE determines the number of reporting elements of the 4 codewords to be 10, 10, 8 and 6 respectively, and then quantizes the first 10, 10, 8 and 6 of the 4 codewords respectively and reports them to the gNB.
- the UE also passed 4 Indicates the number of elements reported by layers 1 to 4 respectively.
- the UE passes a Indicate the number of elements reported by layers 1 and 2, and then pass 2 Indicates the number of elements reported by layer 3 and layer 4 respectively.
- the gNB determines the elements reported in each codeword according to the indication information of the number of elements reported by the UE.
- the embodiment of the present invention provides a network model or mask vector selection method that can select an appropriate network model according to the channel differences of different users, so as to reduce the feedback overhead of the terminal. And the embodiment of the present disclosure also provides the network model input information form of L-layer compressed feedback and the reporting elements of L codewords and the method for determining the number and position of reporting elements, and can also realize the L-layer channel information based on artificial intelligence. Effective compression feedback.
- the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Accordingly, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) having computer-usable program code embodied therein.
- processor-executable instructions may also be stored in a processor-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the processor-readable memory produce a manufacturing product, the instruction device realizes the functions specified in one or more procedures of the flow chart and/or one or more blocks of the block diagram.
- processor-executable instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented
- the executed instructions provide steps for implementing the functions specified in the procedure or procedures of the flowchart and/or the block or blocks of the block diagrams.
- each component or each step can be decomposed and/or reassembled. These decompositions and/or recombinations should be considered equivalents of the present disclosure. Also, the steps for executing the above series of processes can naturally be executed in chronological order according to the illustrated order, but it is not necessary to be executed in chronological order, and some steps can be executed in parallel or independently of each other.
- modules can all be implemented in the form of calling software through processing elements; they can also be implemented in the form of hardware; some modules can also be implemented in the form of calling software through processing elements, and some modules can be implemented in the form of hardware.
- a certain module can be a separate processing element, or it can be integrated into a chip of the above-mentioned device.
- it can also be stored in the memory of the above-mentioned device in the form of program code, and processed by one of the above-mentioned devices.
- the component invokes and executes the functions of the modules identified above.
- each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or an instruction in the form of software.
- each module, unit, subunit or submodule may be one or more integrated circuits configured to implement the above method, for example: one or more specific integrated circuits (Application Specific Integrated Circuit, ASIC), or, one or Multiple microprocessors (digital signal processor, DSP), or, one or more field programmable gate arrays (Field Programmable Gate Array, FPGA), etc.
- ASIC Application Specific Integrated Circuit
- DSP digital signal processor
- FPGA Field Programmable Gate Array
- the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, CPU) or other processors that can call program codes.
- these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).
- SOC system-on-a-chip
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本公开提供了一种信息处理方法、装置、终端及网络设备,涉及通信技术领域。其中该方法包括:终端接收网络设备发送的下行信号;所述终端根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;所述终端通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;所述终端将所述编码数据进行处理后的数据发送至所述网络设备。
Description
相关申请的交叉引用
本申请主张在2021年12月25日在中国提交的中国专利申请号No.202111605575.2的优先权,其全部内容通过引用包含于此。
本公开涉及通信技术领域,尤其涉及一种信息处理方法、装置、终端及网络设备。
在大规模多输入多输出(Massive Multi Input Multi Output,Massive MIMO)系统中,基于人工智能的神经网络可用于信道信息的压缩反馈,如信道状态信息(Channel State Information,CSI)的压缩反馈。如可以利用信道信息的稀疏性,采用编码器模块实现对信道信息的压缩以减少反馈开销,再通过解码器模块恢复出与原来近似的信道信息。为了适应不同用户信道信息变化,减少终端的反馈开销或者提升系统性能,考虑在终端侧和/或网络设备侧部署包含编码器和/或解码器的多个网络模型。但是在终端侧和/或网络设备侧部署多个网络模型时,目前还没有给出终端侧和/或网络设备侧如何通过其中的网络模型进行信道信息传输的方案。
发明内容
本公开提供一种信息处理方法、装置、终端及网络设备,解决了在终端侧和/或网络设备侧部署多个网络模型时,目前还没有给出终端侧和/或网络设备侧如何通过其中的网络模型进行信道信息传输的方案。
本公开的实施例提供一种信息处理方法,包括:
终端接收网络设备发送的下行信号;
所述终端根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整 数;
所述终端通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;
所述终端将所述编码数据进行处理后的数据发送至所述网络设备。
可选地,所述下行信号为下行参考信号;所述终端根据所述下行信号,确定第一网络模型,包括:
所述终端根据所述下行参考信号,确定所述第一信道信息;
所述终端根据所述第一信道信息,确定所述第一网络模型。
可选地,所述终端根据所述第一信道信息,确定所述第一网络模型之后,还包括:
所述终端向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述终端根据所述第一信道信息,确定所述第一网络模型,包括:
所述终端根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息;
所述终端根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述终端从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模 型,则所述终端从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述终端根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数,包括:
所述终端根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述终端接收网络设备发送的下行信号之前,还包括:
所述终端向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示(Precoding matrix indicator,PMI)。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,所述L个信道信息中的第l个信道信息包括以下一项:
第一频域单元的R个数据流的预编码向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中, L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;
其中,所述第一频域单元为所有频域单元或一个频域单元。
可选地,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块(Physical Resource Block,PRB),或者所述频域单元为一个包含多个PRB的子带。
可选地,所述终端将所述编码数据进行处理后的数据发送至所述网络设备,包括:
所述终端对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述方法还包括以下一项:
在L≥1时,所述终端向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则所述终端针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;
在L>1时,所述终端向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,在所述第一元素包括所述编码数据中的K
1个非零元素的情况下:
若K
1<K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置;
若K
1=K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1。
可选地,所述第一元素包括根据掩模向量确定的所述编码数据中的K
1个元素,或所述编码数据中从第一位置开始的连续K
1个元素的情况下:
所述第三指示信息或所述第四指示信息包括N4比特,所述N4比特用于指示所述K
1个元素中非零元素的个数K
2,且所述第三指示信息或所述第四指示信息还用于指示K
2个非零元素在所述编码数据中的位置;其中,
或者,
可选地,所述终端根据所述下行信号,确定第一网络模型之后,还包括:
所述终端根据所述编码数据进行处理后的数据,通过所述第一网络模型得到恢复的第二信道信息;
所述终端根据所述第二信道信息,计算得到信道质量指示(Channel quality indicator,CQI);
所述终端向所述网络设备发送所述CQI。
可选地,所述终端根据所述下行信号,确定第一网络模型之后,还包括:
所述终端接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;
所述终端根据所述信道信息参考信号,确定有效信道信息;
所述终端根据所述有效信道信息,计算得到CQI;
所述终端向所述网络设备发送所述CQI。
可选地,所述信息处理方法还包括:
所述终端向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
本公开实施例提供一种信息处理方法,包括:
网络设备向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;
所述网络设备接收所述终端对编码数据进行处理后的数据;其中,所述 编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;
所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据;
所述网络设备根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
可选地,所述下行信号为下行参考信号;所述网络设备根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息之前,还包括:
所述网络设备接收所述终端发送的第一上行信号;
所述网络设备根据所述第一上行信号,确定所述第一网络模型。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述网络设备向终端发送下行信号之前,还包括:
所述网络设备接收所述终端发送的第二上行信号;
所述网络设备根据所述第二上行信号,确定第三信道信息;
所述网络设备根据所述第三信道信息,确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
可选地,所述网络设备根据所述第三信道信息,确定所述第一网络模型,包括:
所述网络设备根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;
所述网络设备根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述网络设备从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述网络设备从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述网络设备根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数,包括:
所述网络设备根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据之前,还包括以下一项:
在L≥1时,所述网络设备接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述第一元素在所述编码数据中的位置,所述第一元素为所述编码数据中的部分或全部元素;
在L>1时,所述网络设备接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
在L>1时,所述网络设备接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的 位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据,包括:
所述网络设备根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;
所述网络设备根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
可选地,所述网络设备根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据,包括以下一项:
所述网络设备根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;
所述网络设备根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;
所述网络设备根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的位置确定所述待解码数据;
所述网络设备根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;
所述网络设备根据所述反量化处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据。
可选地,所述信息处理方法还包括:
所述网络设备接收所述终端发送的信道质量指示CQI。
可选地,所述信息处理方法还包括:
所述网络设备向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;
所述网络设备接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信息参考信号计算得到的。
可选地,所述信息处理方法还包括:
所述网络设备接收所述终端发送的下行信道的干扰与噪声相关信息;
所述网络设备根据所述下行信道的干扰与噪声相关信息和所述恢复得到的信道信息,计算得到CQI。
本公开实施例提供一种信息处理装置,包括存储器,收发机,处理器;
其中,存储器用于存储计算机程序;收发机用于在所述处理器的控制下收发数据;处理器用于读取所述存储器中的计算机程序并执行以下操作:
接收网络设备发送的下行信号;
根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;
通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;
将所述编码数据进行处理后的数据发送至所述网络设备。
可选地,所述下行信号为下行参考信号;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
根据所述下行参考信号,确定所述第一信道信息;
根据所述第一信道信息,确定所述第一网络模型。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息;
根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,所述L个信道信息中的第l个信道信息包括以下一项:
第一频域单元的R个数据流的预编码向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;
其中,所述第一频域单元为所有频域单元或一个频域单元。
可选地,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块PRB,或者所述频域单元为一个包含多个PRB的子带。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述处理器用于读取所述存储器中的计算机程序并执行以下操作中的一项:
在L≥1时,向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;
在L>1时,向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的 总个数和/或所述第一元素在所述编码数据中的位置。
可选地,在所述第一元素包括所述编码数据中的K
1个非零元素的情况下:
若K
1<K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置;
若K
1=K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1。
可选地,所述第一元素包括根据掩模向量确定的所述编码数据中的K
1个元素,或所述编码数据中从第一位置开始的连续K
1个元素的情况下:
所述第三指示信息或所述第四指示信息包括N4比特,所述N4比特用于指示所述K
1个元素中非零元素的个数K
2,且所述第三指示信息或所述第四指示信息还用于指示K
2个非零元素在所述编码数据中的位置;其中,
或者,
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
根据所述编码数据进行处理后的数据,通过所述第一网络模型得到恢复的第二信道信息;
根据所述第二信道信息,计算得到信道质量指示CQI;
向所述网络设备发送所述CQI。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;
根据所述信道信息参考信号,确定有效信道信息;
根据所述有效信道信息,计算得到CQI;
向所述网络设备发送所述CQI。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
本公开实施例提供一种终端,包括:
第一接收单元,用于接收网络设备发送的下行信号;
第一处理单元,用于根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;
编码单元,用于通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;
第一发送单元,用于将所述编码数据进行处理后的数据发送至所述网络设备。
本公开实施例提供一种信息处理装置,包括存储器,收发机,处理器;
其中,存储器用于存储计算机程序;收发机用于在所述处理器的控制下收发数据;处理器用于读取所述存储器中的计算机程序并执行以下操作:
向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;
接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;
根据所述终端对编码数据进行处理后的数据,确定待解码数据;
根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
可选地,所述下行信号为下行参考信号;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
接收所述终端发送的第一上行信号;
根据所述第一上行信号,确定所述第一网络模型。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
接收所述终端发送的第二上行信号;
根据所述第二上行信号,确定第三信道信息;
根据所述第三信道信息,确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;
根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述处理器用于读取所述存储器中的计算机程序并执行以下操作中的一项:
在L≥1时,接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述第一元素在所述编码数据中的位置,所述第一元素为所述编码数据中的部分或全部元素;
在L>1时,接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
在L>1时,接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操 作:
根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;
根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作中的一项:
根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的位置确定所述待解码数据;
根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
接收所述终端发送的信道质量指示CQI。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;
接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信息参考信号计算得到的。
可选地,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:
接收所述终端发送的下行信道的干扰与噪声相关信息;
根据所述下行信道的干扰与噪声相关信息和所述恢复得到的信道信息,计算得到CQI。
本公开实施例提供一种网络设备,包括:
第一发送单元,用于向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;
第一接收单元,用于接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;
第一处理单元,用于根据所述终端对编码数据进行处理后的数据,确定待解码数据;
解码单元,用于根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
本公开实施例提供一种处理器可读存储介质,所述处理器可读存储介质存储有计算机程序,所述计算机程序用于使所述处理器执行如上终端侧或网络设备侧所述的信息处理方法中的步骤。
本公开的上述技术方案的有益效果是:
该方案中,终端接收网络设备发送的下行信号,并根据所述下行信号,从预先部署的I个网络模型中确定用于对所述下行参考信息对应的信道信息进行编码的第一网络模型,并通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据,以及将所述编码数据进行处理后的数据发送至所述网络设备,从而保证在终端侧和/或网络设备侧部署多个网络模型时,终端侧和/或网络设备侧可以从预先部署的多个网络模型确定用于信道信息传输的网络模型,并且可以减小信道信息传输的反馈开销以及提高系统性能。
图1表示AE的结构示意图;
图2表示基于人工智能的CSI压缩反馈的网络训练流程图;
图3表示本公开实施例的终端侧的信息处理方法的流程图;
图4表示本公开实施例的第i个网络模型的信道信息压缩反馈模型框图;
图5a表示本公开实施例的自适应编码器模型训练流程框图;
图5b表示本公开实施例的自适应解码器模型训练流程框图;
图6表示本公开实施例的反馈信息长度自适应可变网络的CSI压缩反馈模型框图;
图7表示本公开实施例的终端的框图;
图8表示本公开实施例的终端侧的信息处理装置的框图;
图9表示本公开实施例的网络设备侧的信息处理方法的流程图;
图10表示本公开实施例的网络设备的框图;
图11表示本公开实施例的网络设备侧的信息处理装置的框图;
图12表示本公开实施例的多个网络模型的CSI压缩反馈处理流程的框图之一;
图13表示本公开实施例的多个网络模型的CSI压缩反馈处理流程的框图之二;
图14表示本公开实施例的码字截短后的非零元素的位置指示示意图。
为使本公开要解决的技术问题、技术方案和优点更加清楚,下面将结合附图及具体实施例进行详细描述。在下面的描述中,提供诸如具体的配置和组件的特定细节仅仅是为了帮助全面理解本公开的实施例。因此,本领域技术人员应该清楚,可以对这里描述的实施例进行各种改变和修改而不脱离本公开的范围和精神。另外,为了清楚和简洁,省略了对已知功能和构造的描述。
应理解,说明书通篇中提到的“一个实施例”或“一实施例”意味着与实施例有关的特定特征、结构或特性包括在本公开的至少一个实施例中。因此,在整个说明书各处出现的“在一个实施例中”或“在一实施例中”未必一定指相同的实施例。此外,这些特定的特征、结构或特性可以任意适合的方式结合在一个或多个实施例中。
在本公开的各种实施例中,应理解,下述各过程的序号的大小并不意味 着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本公开实施例的实施过程构成任何限定。
另外,本文中术语“系统”和“网络”在本文中常可互换使用。
本公开实施例提供的技术方案可以适用于多种系统,尤其是5G系统。例如适用的系统可以是全球移动通讯(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)通用分组无线业务(general packet radio service,GPRS)系统、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)系统、高级长期演进(long term evolution advanced,LTE-A)系统、通用移动系统(universal mobile telecommunication system,UMTS)、全球互联微波接入(worldwide interoperability for microwave access,WiMAX)系统、第五代移动通信(5th-Generation,5G)新空口(New Radio,NR)系统等。这多种系统中均包括终端设备和网络设备。系统中还可以包括核心网部分,例如演进的分组系统(Evloved Packet System,EPS)、5G系统(5G System,5GS)等。
网络设备与终端设备之间可以各自使用一或多根天线进行多输入多输出(Multi Input Multi Output,MIMO)传输,MIMO传输可以是单用户MIMO(Single User MIMO,SU-MIMO)或多用户MIMO(Multiple User MIMO,MU-MIMO)。根据根天线组合的形态和数量,MIMO传输可以是二维MIMO(2 Dimission MIMO,2D-MIMO)、三维MIMO(3 Dimission MIMO,3D-MIMO)、全维度MIMO(Full Dimension MIMO,FD-MIMO)或Massive MIMO,也可以是分集传输或预编码传输或波束赋形传输等。
本公开实施例中术语“和/或”,描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。
本公开实施例中术语“多个”是指两个或两个以上,其它量词与之类似。
下面将结合本公开实施例中的附图,对本公开实施例中的技术方案进行 清楚、完整地描述,显然,所描述的实施例仅仅是本公开一部分实施例,并不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
本公开实施例涉及的自编码器(Auto-encoder,AE)是一种输入与输出相同的神经网络,由编码器和解码器两部分组成,编码器把输入数据压缩成潜在空间(latent space)表征,解码器根据这一表征重构输入。AE的结构如图1所示,输入数据X首先经过编码器被压缩为Z=f(X),然后编码特征Z经过解码器得到
目的是重构输入数据X。AE通过反向传播的训练方式求解f(·)和g(·)的映射关系,最小化重构误差
编码器模块和解码器模块可以采用卷积神经网络、全连接神经网络、递归神经网络等及其组合构造。
当AE应用于CSI压缩反馈时,编码器输出Z的维度小于其输入X,输出Z的维度越小则压缩率越高,相应地解码器恢复CSI的误差也越大。
基于人工智能的CSI压缩反馈的网络训练流程,如图2所示。终端侧把获得的信道信息V输入到编码器获得压缩压缩后的码字C(码字C的信息长度远小于输入V的长度),然后把该码字C输入量化器,并把量化后的二进制比特流输入到反量化器得到具有量化误差的码字C’,最后把C’输入到解码器模块得到恢复的信道信息V’。
如图3所示,本公开实施例提供一种信息处理方法,包括以下步骤:
步骤31:终端接收网络设备发送的下行信号。
可选地,所述下行信号可以是下行参考信号,所述下行参考信号可以用于终端从预先部署或从网络设备侧下载的I个网络模型中确定用于对所述下行参考信息对应的信道信息进行编码的第一网络模型;或者,所述下行信号可以携带用于指示所述第一网络模型的第一指示信息。
步骤32:所述终端根据所述下行信号,确定第一网络模型。
其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数。
可选地,所述第一网络模型可以由终端侧从I个网络模型选择确定,如所述下行信号为网络设备发送的下行参考信号,则终端可以根据所述下行参 考信号从I个网络模型选择至少一个网络模型作为所述第一网络模型。或者,所述第一网络模型可以由网络设备配置给终端,如所述下行信号携带用于指示所述I个网络模型中的第一网络模型的第一指示信息。
步骤33:所述终端通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据。
可选地,所述第一信道信息可以包括L个信道信息,该信道信息可以是CSI,L为正整数;其中,当L大于1时,不同的信道信息可以采用相同的网络模型进行编码(或称为压缩)处理,或者不同的信道信息也可以采用不同的网络模型进行编码处理,如通过网络模型对第一信道信息进行编码处理后,可以得到每个信道信息对应的码字(即待传输的编码数据),以减小终端的反馈开销。
步骤34:所述终端将所述编码数据进行处理后的数据发送至所述网络设备。
可选地,所述终端可以将所述编码数据中的部分或全部元素通过量化处理后发送至所述网络设备。
上述方案中,终端接收网络设备发送的下行信号,并根据所述下行信号,从预先部署的I个网络模型中确定用于对所述下行参考信息对应的信道信息进行编码的第一网络模型,并通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据,以及将所述编码数据进行处理后的数据发送至所述网络设备,从而保证在终端侧和/或网络设备侧部署多个网络模型时,终端侧和/或网络设备侧可以从预先部署的多个网络模型确定用于信道信息传输的网络模型,并且可以减小信道信息传输的反馈开销以及提高系统性能。
可选地,所述网络模型包括编码器和/或解码器,如所述网络模型可以是包括编码器和解码器的网络模型,或者所述网络模型可以是包括编码器的网络模型,或者所述网络模型可以是包括解码器的网络模型。例如:对于终端和网络设备之间采用多个编码器对应一个解码器进行信道信息传输的情况,可以基于本公开的上述方案从多个编码器中选择至少一个用于进行信道信息编码的编码器;又例如:对于终端和网络设备之间采用一个编码器对应多个解码器进行信道信息传输的情况,可以基于本公开的上述方案从多个解码器 中选择至少一个编码器。
以下结合预先训练了I个网络模型,每个网络模型包含一个编码器和一个解码器为例进行说明,如第i个网络模型的编码器输出的码字C
i,其长度为k
i。不同网络模型对应的k
i可以相等,也可以不等。基于第i个网络模型的信道信息压缩反馈模型框图,如图4所示。
为了使得反馈信息长度可自适应改变,可以将采用包含自适应编码器和自适应的解码器的网络模型。如预先训练了I个网络模型,这I个网络模型可以均是包含自适应编码器和自适应的解码器的网络模型,或者这I个网络模型中的部分网络模型是包含自适应编码器和自适应的解码器的网络模型。自适应编码器和自适应解码器是根据在编码器模块后加入了一个随机的掩模向量模块,在解码器之前加入了一个相应的掩模向量的填充模块,其可以通过网络训练得到,如图5a和5b所示。如信道信息V经过编码器得到码字C,向量长度为k
max。码字C经过一个长度为k
max的掩模向量模块后得到
Z
masking=C⊙M
其中,向量M为掩模向量,M是一个前k位为1后(k
max-k)位为0的向量,即M=[1,1,…,1,0,0,…,0]。网络训练时,k∈[0,k
max]为随机值可服从等概率的随机分布,运算⊙表示按位乘法。它相当于把原来编码后输出的信息长度k
max截短为k作为有效信息进行反馈;自适应解码器是通过引入掩模向量填充模块对所接收的信息以补0的方式,再把信息长度恢复为k
max作为解码器的输入。
如图6所示,给出了一种基于反馈信息长度自适应可变网络的CSI压缩反馈模型框图。信道信息V首先输入到自适应编码器后所输出的码字C通过掩模向量模块截短后输入到量化器模块进行量化,实现对信道信息V的压缩,量化后的二进制比特流再依次输入到反量化器模块和掩模向量填充模块,获得码字C’。然后再把码字C’输入到自适应解码器得到恢复的信道信息V’。
需要说明的是,本公开实施例中引入掩模向量、量化器、反量化器和掩模向量填充模块是为了更好地辅助说明CSI压缩反馈的过程。这些模块中有些模块可能不存在,或者可以通过具有相似功能的其他模块替代。例如,掩模向量模块用于确定终端反馈码字中的全部或部分元素信息,量化器用于将 码字中每个元素值通过N比特(N bits)量化一些离散数值,反量化器用于根据N bits表示所确定值作为码字中元素值,掩模向量填充模块用于根据终端上报的元素个数和/或元素位置的指示信息确定码字中的相应元素以及其余位置补0。
可选地,所述下行信号为下行参考信号;所述终端根据所述下行信号,确定第一网络模型,包括:
所述终端根据所述下行参考信号,确定所述第一信道信息;
所述终端根据所述第一信道信息,确定所述第一网络模型。
具体的,终端侧可以根据接收的下行参考信号估计下行信道信息(即第一信道信息),并根据估计的下行信道信息从I个网络模型中选择一个或至少两个网络模型。例如:当存在L=1个信道信息需传输时,终端可以从I个网络模型中选择一个网络模型;当存在L>1个信道信息需传输,且L个信道信息采用相同的网络模型进行编码和/或解码处理时,终端可以从I个网络模型中选择一个网络模型;当存在L>1个信道信息需传输,且L个信道信息采用不同的网络模型进行编码/解码处理时,终端可以从I个网络模型中选择至少两个/L个网络模型等,本公开实施例不以此为限。
可选地,所述终端根据所述第一信道信息,确定所述第一网络模型之后,还包括:
所述终端向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
该实施例中,在由终端侧从I个网络模型中选择用于信道信息编码和/或解码处理的网络模型(即第一网络模型)的情况下,在终端确定用于第一信道信息编码和/或解码处理的第一网络模型后,由终端向网络设备发送第一指示信息,以使得网络设备根据所述第一指示信息,获知对第一信道信息编码和/或解码处理的第一网络模型。
可选地,所述第一信道信息包括L个信道信息,L为正整数;
换言之,终端在向网络设备发送用于指示其对第一信道信息进行编码和/或解码处理的第一网络模型时,具体可以通过N1比特来指示不同信道信息对应的不同网络模型。其中,N1可以根据信道信息的个数L、网络模型的个数I来确定。例如:
其中
表示从I个网络模型中选择L个网络模型,L表示传输的层数(即信道信息的个数)。
可选地,终端可以通过遍历方式从I个网络模型中选择合适的第一网络模型;具体的,所述终端根据所述第一信道信息,确定所述第一网络模型,包括:
所述终端根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息;
所述终端根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述终端从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述终端从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
具体的,以第一网络模型包括编码器和解码器为例,所述终端根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息可以包括:终端将第一信道信息作为第一网络模型中编码器的输入,得到编码压缩后的编码数据;终端还可以对该编码 数据进行第一处理(如通过掩模向量截短和/或量化处理等),并将进行第一处理后的数据进行第二处理(如该第二处理可理解为是对第一处理的还原,如通过反量化处理和/或掩模向量填充等);终端将进行第二处理后的数据作为所述第一网络模型中的解码器输入,解码得到恢复的信道信息(即第二信道信息)。
需要说明的是,这里终端根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息,并不限定为终端需要针对所有网络模型均得到对应的第二信道信息。也就是说,终端可以针对所有网络模型分别通过第一信道信息得到恢复的第二信道信息,进而根据第一信道信息和第二信道信息确定的目标参数选择第一网络模型;或者,终端也可以仅针对部分网络模型分别通过第一信道信息得到恢复的第二信道信息,例如:终端根据所述第一信道信息依次通过这I个网络模型进行编码和/或解码处理,得到每个网络模型对应的第二信道信息时,如果在通过第i个网络模型确定的目标参数进行判断,能够从这i个网络模型中选择到第一网络模型时,可以针对之后的(I-i)个网络模型不再通过第一信道信息得到恢复的第二信道信息,从而可以提高处理效率。
可选地,所述终端根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数,包括:
所述终端根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
具体的,以宽带信干噪比准则为例,对本公开实施例确定第一网络模型的一种可选实施例进行说明:
步骤A1:所述终端根据所述第一信道信息,通过所述I个网络模型中第i个网络模型进行编码和/或解码处理,得到第i个第二信道信息;
步骤A2:根据所述第一信道信息和所述第i个第二信道信息,计算得到第i个目标参数;
若所述第i个目标参数大于预设门限,则令i=i+1并重复所述步骤A1;
若所述第i个目标参数小于或等于所述预设门限,则令i=i+1并重复所述步骤A1,直至所述第i个目标参数大于所述预设门限;
步骤A3:若i≤I且存在所述目标参数小于或等于预设门限的M个网络模型,则将所述M个网络模型中输出编码数据中元素个数(也即编码数据的长度)最小的网络模型确定为所述第一网络模型;其中,M为正整数;
若i=I且所述I个网络模型对应的目标参数均大于所述预设门限,则将所述I个网络模型中输出编码数据中元素个数最大的网络模型确定为所述第一网络模型。
例如:若根据第i个网络模型计算相应的宽带的信干噪比γ
WB,i<γ
T,表示所选的第i个网络模型不能满足要求,再继续计算第i+1个网络模型对应的宽带的信干噪比γ
WB,i+1,直至所计算的γ
WB,i≥γ
T。最后从满足γ
WB,i≥γ
T要求且编码器输出的码字长度(即编码数据中的元素个数)为最小所对应的网络模型,确定为第一网络模型。其中,γ
T表示预设的门限值。
若根据第i个网络模型计算的γ
WB,i≥γ
T,虽然当前所选的网络模型能够满足要求,但是该网络模型输出码字长度可能偏大。可继续再计算第i+1个网络模型对应的宽带的信干噪比γ
WB,i+1,直至所计算的γ
WB,i<γ
T。最后选择所有满足γ
WB,i≥γ
T的模型中输出码字长度最小的网络模型,确定为所述第一网络模型。
若所计算出所有网络模型对应的宽带信噪比均不能满足要求,则从中选择码字长度为最大所对应的网络模型,确定为所述第一网络模型。
可选地,终端还可以根据上行传输资源确定所选的网络模型。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述终端接收网络设备发送的下行信号之前,还包括:
所述终端向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
其中,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
该实施例中,该第一网络模型可以由网络设备侧确定,网络设备可以将确定的网络模型通过信令配置给终端,如网络设备向终端发送网络模型指示 信息(即第二指示信息)用于指示对第一信道信息进行编码和/或解码处理的第一网络模型,从而终端可以根据该第二指示信息确定的第一网络模型进行第一信道信息的压缩反馈。
例如:终端在接收网络设备侧发送的网络模型指示信息(即终端接收网络设备发送的下行信号)之前,终端向网络设备侧至少发送一次上行信号,该上行信号可以是上行参考信号,或携带终端计算的PMI(如类型I(Type I)或者类型II(Type II)码本),用于网络设备侧选择第一网络模型。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,所述L个信道信息中的第l个信道信息包括以下一项:
第一频域单元的R个数据流的预编码向量的组合;其中,L=1;如组合后的矩阵为[V
1;V
2;…;V
R];
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;如组合后的矩阵为[Y
1;Y
2;…;Y
R];
第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中,L=1;如组合后的矩阵为[H
1;H
2;…;H
R];
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;如组合后的矩阵为[U
1;U
2;…;U
R];
第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;如组合后的矩阵为[H
1;H
2;…;H
R;0
R+1;…;0
v];
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;如组合后的矩阵为[Y
1;Y
2;…;Y
R;0
R+1;…;0
v];
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;如组合后的矩阵为[U
1;U
2;…;U
R;0
R+1;…;0
v];
可选地,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块PRB,或者所述频域单元为一个包含多个PRB的子带。
可选地,所述终端将所述编码数据进行处理后的数据发送至所述网络设备,包括:
所述终端对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
例如:针对L个信道信息的第l个信道信息,终端将其输入到第l个信道信息对应的第一网络模型,通过其中的编码器输出获得一个码字C(即编码数据)。码字C可以表示为向量,其元素个数为K
0。终端将码字C中的部分或者全部元素,即K
1≤K
0个元素经过量化处理后通过上行信道反馈给网络设备(如通过量化器模块进行量化处理)。
可选地,终端确定编码数据中的部分或全部元素的方式,即所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;例如:如终端通过第l个信道信息对应的第一网络模型对第l个信道信息(如CSI)压缩后获得码字C
i,i∈{1,...,I}(即编码数据),C
i表示采用所选的第i个网络模型压缩后的码字。终端把码字C
i中K
1≥1个非零元素量化后上报给网络设备侧。
根据掩模向量确定的所述编码数据中的K
1个元素;例如:终端将对应于掩模向量非零元素的编码数据中的元素量化后上报给网络设备。其中,掩模向量与网络模型对应,即网络模型确定的情况下其对应的掩模向量也可以确定,即掩模向量也可以由终端自行选择,或者由网络配置给终端。
所述编码数据中从第一位置开始的连续K
1个元素。例如:终端将编码数据中从第一位置开始的连续K
1个元素量化后上报给网络设备。其中,K
1可以由终端自行选择,或者由网络配置给终端;该第一位置可以根据预定义确定,或者由终端上报指示,或者由网络配置给终端。
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述方法还包括以下一项:
在L≥1时,所述终端向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;也即是针对L个信道信息之间独立地确定第一元素以及独立地上报所述第一元素的个数和/或所述第一元素在编码数据中的位置。
在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则所述终端针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;也即是针对L个信道信息中的第一部分信道信息,由于其编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,故可通过一个共同的指示信息将上报的第一元素的个数和/或所述第一元素在所述编码数据中的位置上报给网络设备。
在L>1时,所述终端向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,在所述第一元素包括所述编码数据中的K
1个非零元素的情况下:
若K
1<K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置;
若K
1=K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1。
例如:在K
1<K
0,即编码数据中的部分元素为非零元素的情况下,终端需要向网络设备指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置,如终端可以通过一个位图(bitmap)或者组合数指示码字中非零元素的位置,并上报给网络侧。在K
1=K
0,即编码数据中的全部元素均为非零元素的情况下,终端可以向网络设备指示所述非零元素的个数K
1,此时可以无需再指示非零元素在所述编码数据中的位置。
换言之,针对上述任意确定编码数据中的部分或全部元素的方式,所述终端可以通过N3比特指示所述第一元素的个数K
1。其中,N3可以根据编码数据中所有元素个数K
0确定,如
同时终端还通过位图或者组合数指示所述K
1个第一元素在编码数据中的位置。
可选地,所述第一元素包括根据掩模向量确定的所述编码数据中的K
1个元素,或所述编码数据中从第一位置开始的连续K
1个元素的情况下:所述第三指示信息或所述第四指示信息包括N4比特,所述N4比特用于指示所述K
1个元素中非零元素的个数K
2,且所述第三指示信息或所述第四指示信息还用于指示K
2个非零元素在所述编码数据中的位置;其中,
换言之,针对上述任意确定编码数据中的部分或全部元素的方式,所述 终端可以通过N4比特指示所述K
1个第一元素中非零元素的个数K
2。其中,N4可以根据第一元素的个数K
1确定,如
同时终端还通过位图或者组合数向网络设备指示这K
2个非零元素的位置。此时终端将K
2个非零元素量化之后上报给网络设备,而其他的元素不再上报。
或者,
例如:终端可以联合确定L个编码数据中待上报的第一元素的个数和/或所述第一元素在所述编码数据中的位置。其中,不同编码数据中上报的第一元素个数和/或第一元素在所述编码数据中的位置可以不同。可选地,每个编码数据中上报的第一元素的个数和/或第一元素在所述编码数据中位置可以按照上述独立上报的方式指示给网络设备;或者联合上报给网络设备,如终端可以通过N5比特指示每个信道信息编码后的编码数据中第一元素个数,N5可以根据每个编码数据中上报元素的最大个数
确定,如
其中,
可以由网络配置给终端或者预定义确定。如终端还可以通过N6比特指示L个编码数据中第一元素的总个数,N6可以根据L个编码数据中上报元素的最大总个数
确定,如
其中,
可以由网络配置给终端或者预定义确定。此时终端还可以通过不同编码数据对应的位图或组合数指示不同编码数据中上报的第一元素的位置。
可选地,上述量化处理可以是将第一元素的值量化为一定取值范围的离散值,每个元素量化后的离散值通过N bits表示,输出二进制比特流。相应地,网络设备需进行反量化处理是指根据接收的二进制比特流确定编码数据中该第一元素的值,每个元素的值通过N bits表示的值所确定。
可选地,所述终端根据所述下行信号,确定第一网络模型之后,还包括:
所述终端根据所述编码数据进行处理后的数据,通过所述第一网络模型 得到恢复的第二信道信息;
所述终端根据所述第二信道信息,计算得到信道质量指示CQI;
所述终端向所述网络设备发送所述CQI。
例如:终端可以对所述编码数据中的第一元素通过量化器模块进行量化处理,得到量化处理后的数据,再将量化处理后的数据输入到对应的反量化器模块进行反量化处理,并将反量化处理后的数据输入到第一网络模型中的解码器,解码得到恢复的第二信道信息,再根据恢复的第二信道信息计算CQI。
或者,终端可以将所述编码数据通过掩模向量模块进行处理确定第一元素,以及将第一元素通过量化器模块进行量化处理,得到量化处理后的数据,再将量化处理后的数据输入到对应的反量化器模块进行反量化处理,以及将量化处理后的数据输入到掩模向量填充模块进行处理,并将处理后的数据输入到第一网络模型中的解码器,解码得到恢复的第二信道信息,再根据恢复的第二信道信息计算CQI。
可选地,所述终端根据所述下行信号,确定第一网络模型之后,还包括:
所述终端接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;
所述终端根据所述信道信息参考信号,确定有效信道信息;
所述终端根据所述有效信道信息,计算得到CQI;
所述终端向所述网络设备发送所述CQI。
例如:该信道信息参考信号可以是CSI参考信号(CSI Reference Signal,CSI-RS),终端根据接收的波束赋形CSI-RS估计下行有效信道信息,然后根据估计的有效信道信息计算CQI,其中CSI-RS所采用的波束由位于网络设备侧的第一网络模型恢复的信道信息确定。
可选地,所述方法还包括:
所述终端向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
该实施例中,终端上报下行信道的干扰与噪声相关信息给网络设备,以使网络设备侧根据该下行信道的干扰与噪声相关信息来计算CQI。
需要说明的是,本公开实施例中涉及的“码字”即是对信道信息编码后得到的编码数据,信道信息可以是SCI。
可选地,本公开上述实施例中的信道信息的确定方式、确定第一元素的方式、第一元素个数和/或位置的上报方式、CQI计算方法等,还可以应用于终端侧和网络侧之间已确定用于编码和/或解码处理的网络模型的场景(如终端和/或网络设备侧可以不执行确定第一网络模型的步骤),为避免重复,这里不再赘述。
本公开实施例涉及的终端,可以是指向用户提供语音和/或数据连通性的设备,具有无线连接功能的手持式设备、或连接到无线调制解调器的其他处理设备等。在不同的系统中,终端设备的名称可能也不相同,例如在5G系统中,终端设备可以称为用户设备(User Equipment,UE)。无线终端设备可以经无线接入网(Radio Access Network,RAN)与一个或多个核心网(Core Network,CN)进行通信,无线终端设备可以是移动终端设备,如移动电话(或称为“蜂窝”电话)和具有移动终端设备的计算机,例如,可以是便携式、袖珍式、手持式、计算机内置的或者车载的移动装置,它们与无线接入网交换语言和/或数据。例如,个人通信业务(Personal Communication Service,PCS)电话、无绳电话、会话发起协议(Session Initiated Protocol,SIP)话机、无线本地环路(Wireless Local Loop,WLL)站、个人数字助理(Personal Digital Assistant,PDA)等设备。无线终端设备也可以称为系统、订户单元(subscriber unit)、订户站(subscriber station),移动站(mobile station)、移动台(mobile)、远程站(remote station)、接入点(access point)、远程终端设备(remote terminal)、接入终端设备(access terminal)、用户终端设备(user terminal)、用户代理(user agent)、用户装置(user device),本公开实施例中并不限定。
本公开实施例涉及的网络设备,可以是基站,该基站可以包括多个为终端提供服务的小区。根据具体应用场合不同,基站又可以称为接入点,或者可以是接入网中在空中接口上通过一个或多个扇区与无线终端设备通信的设备,或者其它名称。网络设备可用于将收到的空中帧与网际协议(Internet Protocol,IP)分组进行相互更换,作为无线终端设备与接入网的其余部分之间的路由器,其中接入网的其余部分可包括网际协议(IP)通信网络。网络 设备还可协调对空中接口的属性管理。例如,本公开实施例涉及的网络设备可以是全球移动通信系统(Global System for Mobile communications,GSM)或码分多址接入(Code Division Multiple Access,CDMA)中的网络设备(Base Transceiver Station,BTS),也可以是带宽码分多址接入(Wide-band Code Division Multiple Access,WCDMA)中的网络设备(NodeB),还可以是长期演进(long term evolution,LTE)系统中的演进型网络设备(evolutional Node B,eNB或e-NodeB)、5G网络架构(next generation system)中的5G基站(gNB),也可以是家庭演进基站(Home evolved Node B,HeNB)、中继节点(relay node)、家庭基站(femto)、微微基站(pico)等,本公开实施例中并不限定。在一些网络结构中,网络设备可以包括集中单元(centralized unit,CU)节点和分布单元(distributed unit,DU)节点,集中单元和分布单元也可以地理上分开布置。
以上实施例就本公开的终端侧的信息处理方法做出介绍,下面本实施例将结合附图对其对应的装置、终端做进一步说明。
具体地,如图7所示,本公开实施例提供一种终端700,包括:
第一接收单元710,用于接收网络设备发送的下行信号;
第一处理单元720,用于根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;
编码单元730,用于通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;
第一发送单元740,用于将所述编码数据进行处理后的数据发送至所述网络设备。
可选地,所述下行信号为下行参考信号;所述第一处理单元720还用于:
根据所述下行参考信号,确定所述第一信道信息;
根据所述第一信道信息,确定所述第一网络模型。
可选地,所述终端700还包括:
第二发送单元,用于向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述第一处理单元720还用于:
根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息;
根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述终端从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述终端从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述第一处理单元720还用于:
根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述终端700还包括:
第三发送单元,用于向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,所 述L个信道信息中的第l个信道信息包括以下一项:
第一频域单元的R个数据流的预编码向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;
其中,所述第一频域单元为所有频域单元或一个频域单元。
可选地,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块PRB,或者所述频域单元为一个包含多个PRB的子带。
可选地,所述第一发送单元740还用于:
对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述终端700还包括以下一项:
第四发送单元,用于在L≥1时,向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
第五发送单元,用于在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;
第六发送单元,用于在L>1时,向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,在所述第一元素包括所述编码数据中的K
1个非零元素的情况下:
若K
1<K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置;
若K
1=K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1。
可选地,所述第一元素包括根据掩模向量确定的所述编码数据中的K
1个元素,或所述编码数据中从第一位置开始的连续K
1个元素的情况下:
所述第三指示信息或所述第四指示信息包括N4比特,所述N4比特用于指示所述K
1个元素中非零元素的个数K
2,且所述第三指示信息或所述第四指示信息还用于指示K
2个非零元素在所述编码数据中的位置;其中,
或者,
可选地,所述终端700还包括:
第二处理单元,用于根据所述编码数据进行处理后的数据,通过所述第一网络模型得到恢复的第二信道信息;
第一计算单元,用于根据所述第二信道信息,计算得到信道质量指示CQI;
第七发送单元,用于向所述网络设备发送所述CQI。
可选地,所述终端700还包括:
第二接收单元,用于接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;
第三处理单元,用于根据所述信道信息参考信号,确定有效信道信息;
第二计算单元,用于根据所述有效信道信息,计算得到CQI;
第八发送单元,用于向所述网络设备发送所述CQI。
可选地,所述终端700还包括:
第九发送单元,用于向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
需要说明的是,本公开实施例中对单元的划分是示意性的,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。另外,在本公开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。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元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,可以存储在一个处理器可读取存储介质中。基于这样的理解,本公开的技术方案本质上或者说对相关技术做出贡献的部分或者该技术方案的全部或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)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
在此需要说明的是,本公开实施例提供的上述终端,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
为了更好的实现上述目的,如图8所示,提供一种信息处理装置,包括存储器801,收发机802,处理器803;其中,存储器807用于存储计算机程序;收发机802用于在所述处理器803的控制下收发数据;如收发机802用于在处理器803的控制下接收和发送数据;处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
接收网络设备发送的下行信号;
根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;
通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;
将所述编码数据进行处理后的数据发送至所述网络设备。
可选地,所述下行信号为下行参考信号;所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
根据所述下行参考信号,确定所述第一信道信息;
根据所述第一信道信息,确定所述第一网络模型。
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第二信道信息;
根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,所述L个信道信息中的第l个信道信息包括以下一项:
第一频域单元的R个数据流的预编码向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;
第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;
第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;
第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;
其中,所述第一频域单元为所有频域单元或一个频域单元。
可选地,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块PRB,或者所述频域单元为一个包含多个PRB的子带。
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作中的一项:
在L≥1时,向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;
在L>1时,向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,在所述第一元素包括所述编码数据中的K
1个非零元素的情况下:
若K
1<K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置;
若K
1=K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1。
可选地,所述第一元素包括根据掩模向量确定的所述编码数据中的K
1个元素,或所述编码数据中从第一位置开始的连续K
1个元素的情况下:
所述第三指示信息或所述第四指示信息包括N4比特,所述N4比特用于指示所述K
1个元素中非零元素的个数K
2,且所述第三指示信息或所述第四指示信息还用于指示K
2个非零元素在所述编码数据中的位置;其中,
或者,
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
根据所述编码数据进行处理后的数据,通过所述第一网络模型得到恢复的第二信道信息;
根据所述第二信道信息,计算得到信道质量指示CQI;
向所述网络设备发送所述CQI。
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;
根据所述信道信息参考信号,确定有效信道信息;
根据所述有效信道信息,计算得到CQI;
向所述网络设备发送所述CQI。
可选地,所述处理器803用于读取所述存储器801中的计算机程序并执行以下操作:
向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
其中,在图8中,总线架构可以包括任意数量的互联的总线和桥,具体由处理器803代表的一个或多个处理器和存储器801代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起,这些都是本领域所公知的,因此,本文不再对其进行进一步描述。总线接口提供接口。收发机802可以是多个元件, 即包括发送机和接收机,提供用于在传输介质上与各种其他装置通信的单元,这些传输介质包括无线信道、有线信道、光缆等传输介质。针对不同的用户设备,用户接口804还可以是能够外接内接需要设备的接口,连接的设备包括但不限于小键盘、显示器、扬声器、麦克风、操纵杆等。
处理器803负责管理总线架构和通常的处理,存储器801可以存储处理器803在执行操作时所使用的数据。
处理器803可以是中央处理器(Central Processing Unit,CPU)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或复杂可编程逻辑器件(Complex Programmable Logic Device,CPLD),处理器也可以采用多核架构。
在此需要说明的是,本公开实施例提供的上述装置,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
本公开实施例还提供一种处理器可读存储介质,所述处理器可读存储介质存储有计算机程序,所述计算机程序用于使所述处理器执行上述信息处理方法中的步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
所述处理器可读存储介质可以是处理器能够存取的任何可用介质或数据存储设备,包括但不限于磁性存储器(例如软盘、硬盘、磁带、磁光盘(Magneto-Optical Disk,MO)等)、光学存储器(例如光盘(Compact Disk,CD)、数字视频光盘(Digital Versatile Disc,DVD)、蓝光光碟(Blu-ray Disc,BD)、高清通用光盘(High-Definition Versatile Disc,HVD)等)、以及半导体存储器(例如只读存储器(Read-Only Memory,ROM)、可擦除可编程只读存储器(Erasable Programmable ROM,EPROM)、带电可擦可编程只读存储器(Electrically EPROM,EEPROM)、非易失性存储器(NAND FLASH)、固态硬盘(Solid State Disk或Solid State Drive,SSD))等。
以上从终端侧介绍了本公开实施例的信息处理方法,下面将结合附图对网络侧的信息处理方法做进一步说明。
如图9所示,本公开实施例提供了一种信息处理方法,包括以下步骤:
步骤91:网络设备向终端发送下行信号。
其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数。
可选地,所述下行信号可以是下行参考信号,所述下行参考信号可以用于终端从预先部署或从网络设备侧下载的I个网络模型中确定用于对所述下行参考信息对应的信道信息进行编码的第一网络模型;或者,所述下行信号可以携带用于指示所述第一网络模型的第一指示信息。
步骤92:所述网络设备接收所述终端对编码数据进行处理后的数据。
其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的。
可选地,所述第一信道信息可以包括L个信道信息,该信道信息可以是CSI,L为正整数;其中,当L大于1时,终端侧针对不同的信道信息可以采用相同的网络模型进行编码(或称为压缩)处理,或者针对不同的信道信息也可以采用不同的网络模型进行编码处理,如通过网络模型对第一信道信息进行编码处理后,可以得到每个信道信息对应的码字(即待传输的编码数据),以减小终端的反馈开销。进一步地,终端对L个信道信息编码得到的L编码数据进行处理后上报给网络设备,从而网络设备接收所述终端对编码数据进行处理后的数据。
步骤93:所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据。
可选地,所述终端可以将所述编码数据中的部分或全部元素通过量化处理后发送至所述网络设备,相应地网络设备需要根据接收的数据进行反量化处理确定相应的编码数据。
步骤94:所述网络设备根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
上述方案中,网络设备向终端发送用于所述终端确定第一网络模型的下行信号,并在接收所述终端对通过所述第一网络模型编码后编码数据进行处理后的数据时,根据所述终端对编码数据进行处理后的数据,确定待解码数据,根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信 道信息,从而保证在终端侧和/或网络设备侧部署多个网络模型时,终端侧和/或网络设备侧可以从预先部署的多个网络模型确定用于信道信息传输的网络模型,并且可以减小信道信息传输的反馈开销以及提高系统性能。
可选地,所述下行信号为下行参考信号;所述网络设备根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息之前,还包括:
所述网络设备接收所述终端发送的第一上行信号;
所述网络设备根据所述第一上行信号,确定所述第一网络模型。
该实施例中,在由终端侧从I个网络模型中选择用于信道信息编码和/或解码处理的网络模型(即第一网络模型)的情况下,在终端确定用于第一信道信息编码和/或解码处理的第一网络模型后,由终端向网络设备发送用于指示所述第一网络模型的第一指示信息,所述第一指示信息携带在所述第一上行信号中,从而网络设备可以根据所述第一上行信号,获知对第一信道信息编码和/或解码处理的第一网络模型。
可选地,所述第一信道信息包括L个信道信息,L为正整数;
换言之,终端在向网络设备发送用于指示其对第一信道信息进行编码和/或解码处理的第一网络模型时,具体可以通过N1比特来指示不同信道信息对应的不同网络模型。其中,N1可以根据信道信息的个数L、网络模型的个数I来确定。例如:
其中
表示从I个网络模型中选择L个网络模型,L表示传输的层数(即信道信息的个数),从而网络设备可以通过这N1比特来确定所述第一网络模型。
换言之,终端在向网络设备发送用于指示其对第一信道信息进行编码和/或解码处理的第一网络模型时,具体可以通过N2比特来指示不同信道信息对应的相同的网络模型。其中,N2可以根据网络模型的个数I确定。例如:
从而网络设备可以通过这N1比特来确定所述第一网络模型。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述网络设备向终端发送下行信号之前,还包括:
所述网络设备接收所述终端发送的第二上行信号;
所述网络设备根据所述第二上行信号,确定第三信道信息;
所述网络设备根据所述第三信道信息,确定所述第一网络模型。
该实施例中,在由网络设备侧从I个网络模型中选择用于信道信息编码和/或解码处理的网络模型(即第一网络模型)的情况下,在网络设备确定用于第一信道信息编码和/或解码处理的第一网络模型后,由网络设备向终端发送第二指示信息,以使得终端根据所述第二指示信息,获知对第一信道信息编码和/或解码处理的第一网络模型。
其中,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
例如:终端在接收网络设备侧发送的网络模型指示信息(即第二指示信息)之前,终端向网络设备侧至少发送一次上行信号,该上行信号可以是上行参考信号,或携带终端计算的PMI(如Type I或者Type II码本)。
网络设备侧根据上行参考信号估计上行信道信息(即确定第三信道信息),如上下行信道是完全互易的,则网络设备可以根据估计的上行信道信息确定对应的第一网络模型。或者,网络设备侧根据接收的PMI确定对应的第一网络模型,并把所选择的第一网络模型通过信令配置给终端,如网络设备向终端发送网络模型指示信息(即第二指示信息)用于指示对第一信道信息进行编码和/或解码处理的第一网络模型,从而终端可以根据该第二指示信息确定的第一网络模型进行第一信道信息的压缩反馈。
可选地,网络设备可以通过遍历方式从I个网络模型中选择合适的第一网络模型;具体的,所述网络设备根据所述第三信道信息,确定所述第一网络模型,包括:
所述网络设备根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;
所述网络设备根据所述第三信道信息和每个网络模型对应的第四信道信 息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述网络设备从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述网络设备从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
具体的,以第一网络模型包括编码器和解码器为例,所述网络设备根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息可以包括:网络设备将第三信道信息作为第一网络模型中编码器的输入,得到编码压缩后的编码数据;网络设备还可以对该编码数据进行第一处理(如通过掩模向量截短和/或量化处理等),并将进行第一处理后的数据进行第二处理(如该第二处理可理解为是对第一处理的还原,如通过反量化处理和/或掩模向量填充等);网络设备将进行第二处理后的数据作为所述第一网络模型中的解码器输入,解码得到恢复的信道信息(即第四信道信息)。
需要说明的是,这里网络设备根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息,并不限定为网络设备需要针对所有网络模型均得到对应的第四信道信息。也就是说,网络设备可以针对所有网络模型分别通过第三信道信息得到恢复的第四信道信息,进而根据第三信道信息和第四信道信息确定的目标参数选择第一网络模型;或者,网络设备也可以仅针对部分网络模型分别通过第三信道信息得到恢复的第四信道信息,例如:网络设备根据所述第三信道信息依次通过这I个网络模型进行编码和/或解码处理,得到每个网络模型对应的第四信道信息时,如果在通过第i个网络模型确定的目标参数进行判断,能够从这i个网络模型中选择到第一网络模型时,可以针对之后的(I-i)个网络模型不再通过第三信道信息得到恢复的第四信道信息,从而可以提高处理效率。
可选地,所述网络设备根据所述第三信道信息和每个网络模型对应的第 四信道信息,计算得到每个网络模型对应的目标参数,包括:
所述网络设备根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
具体的,以宽带信干噪比准则为例,对本公开实施例确定第一网络模型的一种可选实施例进行说明:
步骤A1:所述网络设备根据所述第三信道信息,通过所述I个网络模型中第i个网络模型进行编码和/或解码处理,得到第i个第四信道信息;
步骤A2:根据所述第三信道信息和所述第i个第四信道信息,计算得到第i个目标参数;
若所述第i个目标参数大于预设门限,则令i=i+1并重复所述步骤A1;
若所述第i个目标参数小于或等于所述预设门限,则令i=i+1并重复所述步骤A1,直至所述第i个目标参数大于所述预设门限;
步骤A3:若i≤I且存在所述目标参数小于或等于预设门限的M个网络模型,则将所述M个网络模型中输出编码数据中元素个数(也即编码数据的长度)最小的网络模型确定为所述第一网络模型;其中,M为正整数;
若i=I且所述I个网络模型对应的目标参数均大于所述预设门限,则将所述I个网络模型中输出编码数据中元素个数最大的网络模型确定为所述第一网络模型。
例如:若根据第i个网络模型计算相应的宽带的信干噪比γ
WB,i<γ
T,表示所选的第i个网络模型不能满足要求,再继续计算第i+1个网络模型对应的宽带的信干噪比γ
WB,i+1,直至所计算的γ
WB,i≥γ
T。最后从满足γ
WB,i≥γ
T要求且编码器输出的码字长度(即编码数据中的元素个数)为最小所对应的网络模型,确定为第一网络模型。其中,γ
T表示预设的门限值。
若根据第i个网络模型计算的γ
WB,i≥γ
T,虽然当前所选的网络模型能够满足要求,但是该网络模型输出码字长度可能偏大。可继续再计算第i+1个网络模型对应的宽带的信干噪比γ
WB,i+1,直至所计算的γ
WB,i<γ
T。最后选择所有满足γ
WB,i≥γ
T的模型中输出码字长度最小的网络模型,确定为所述第一 网络模型。
若所计算出所有网络模型对应的宽带信噪比均不能满足要求,则从中选择码字长度为最大所对应的网络模型,确定为所述第一网络模型。
可选地,网络设备还可以根据终端的上行传输资源确定所选的网络模型。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
换言之,网络设备在向终端发送用于指示终端对第一信道信息进行编码和/或解码处理的第一网络模型时,具体可以通过N1比特来指示不同信道信息对应的不同网络模型。其中,N1可以根据信道信息的个数L、网络模型的个数I来确定。例如:
其中
表示从I个网络模型中选择L个网络模型,L表示传输的层数(即信道信息的个数)。
换言之,网络设备在向终端发送用于指示终端对第一信道信息进行编码和/或解码处理的第一网络模型时,具体可以通过N2比特来指示不同信道信息对应的相同的网络模型。其中,N2可以根据网络模型的个数I确定。例如:
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据之前,还包括以下一项:
在L≥1时,所述网络设备接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述第一元素在所述编码数据中的位置,所述第一元素为所述编码数据中的部分或全部元素;也即是针对L个信道信息之间终端独立地确定第一元素以及独立地上报所述第一元素的个数和/或所述第一元素在编码数据中的位置。
在L>1时,所述网络设备接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;也即是针对L个信道信息中的第一部分信道信息,由于其编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,故终端可通过一个共同的指示信息将上报的第一元素的个数和/或所述第一元素在所述编码数据中的位置上报给网络设备。
在L>1时,所述网络设备接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,在所述第一元素包括所述编码数据中的K
1个非零元素的情况下:
若K
1<K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置;
若K
1=K
0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K
1。
例如:在K
1<K
0,即编码数据中的部分元素为非零元素的情况下,终端需要向网络设备指示所述非零元素的个数K
1以及所述K
1个非零元素在所述编码数据中的位置,如终端可以通过一个位图(bitmap)或者组合数指示码字中非零元素的位置,并上报给网络侧。在K
1=K
0,即编码数据中的全部元素均为非零元素的情况下,终端可以向网络设备指示所述非零元素的个数K
1,此时可以无需再指示非零元素在所述编码数据中的位置。
换言之,针对上述任意确定编码数据中的部分或全部元素的方式,所述终端可以通过N3比特指示所述第一元素的个数K
1。其中,N3可以根据编码 数据中所有元素个数K
0确定,如
同时终端还通过位图或者组合数指示所述K
1个第一元素在编码数据中的位置。
可选地,所述第一元素包括根据掩模向量确定的所述编码数据中的K
1个元素,或所述编码数据中从第一位置开始的连续K
1个元素的情况下:所述第三指示信息或所述第四指示信息包括N4比特,所述N4比特用于指示所述K
1个元素中非零元素的个数K
2,且所述第三指示信息或所述第四指示信息还用于指示K
2个非零元素在所述编码数据中的位置;其中,
换言之,针对上述任意确定编码数据中的部分或全部元素的方式,所述终端可以通过N4比特指示所述K
1个第一元素中非零元素的个数K
2。其中,N4可以根据第一元素的个数K
1确定,如
同时终端还通过位图或者组合数向网络设备指示这K
2个非零元素的位置。此时终端将K
2个非零元素量化之后上报给网络设备,而其他的元素不再上报。
或者,
例如:终端可以联合确定L个编码数据中待上报的第一元素的个数和/或所述第一元素在所述编码数据中的位置。其中,不同编码数据中上报的第一元素个数和/或第一元素在所述编码数据中的位置可以不同。可选地,每个编码数据中上报的第一元素的个数和/或第一元素在所述编码数据中位置可以按照上述独立上报的方式指示给网络设备;或者联合上报给网络设备,如终端可以通过N5比特指示每个信道信息编码后的编码数据中第一元素个数,N5可以根据每个编码数据中上报元素的最大个数
确定,如
其中,
可以由网络配置给终端或者预定义确定。如终端还可以通过N6比特指示L个编码数据中第一元素的总个数,N6可以根据L个编码数据中上报元 素的最大总个数
确定,如
其中,
可以由网络配置给终端或者预定义确定。此时终端还可以通过不同编码数据对应的位图或组合数指示不同编码数据中上报的第一元素的位置。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据,包括:
所述网络设备根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;
所述网络设备根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
例如:网络设备侧根据终端上报的量化处理后的数据作为反量化器模块的输入,得到反量化处理后的数据,然后再根据反量化处理后的数据以及所述用于指示第一元素个数和/或位置的第三指示信息或第四指示信息或第五指示信息,得到恢复的编码数据(即待解码数据),将该待解码数据通过对应的第一网络模型的解码器进行解码处理,最终解码得到恢复的信道信息。
或者,网络设备侧根据终端上报的量化处理后的数据作为反量化器模块的输入,得到反量化处理后的数据,然后再把反量化处理后的数据输入到掩模向量填充模块,得到恢复的编码数据(即待解码数据),将该待解码数据通过对应的第一网络模型的解码器进行解码处理,最终解码得到恢复的信道信息。
可选地,所述网络设备根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据,包括以下一项:
所述网络设备根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;例如:网络设备侧可以根据掩模向量确定待解码数据,其中,掩 模向量与网络模型对应,即网络模型确定的情况下其对应的掩模向量也可以确定,即掩模向量也可以由终端自行选择,或者由网络配置给终端;如网络设备可以根据所述终端对编码数据进行处理后的数据,通过掩模向量填充模块输出待解码数据,该待解码数据可以是前K
1个或后K
1个元素为反量化处理后数据,剩余的K
0-K
1个元素补0。
所述网络设备根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;例如:网络设备侧可以根据第三指示信息或第四指示信息或第五指示信息确定终端上报的第一元素的个数K
1,如网络设备可以根据所述终端对编码数据进行处理后的数据,通过掩模向量填充模块输出待解码数据,该待解码数据可以是前K
1个或后K
1个元素为反量化处理后数据,剩余的K
0-K
1个元素补0。
所述网络设备根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的位置确定所述待解码数据;例如:终端上报了一个位图指示信道信息编码后的编码数据中非零元素的位置,网络设备根据该位图指示的非零元素的位置和反量化处理后的数据,确定待解码数据,如待解码数据中非零元素的位置为反量化处理后的数据,剩余位置为0。
所述网络设备根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;例如:网络设备侧可以根据第三指示信息或第四指示信息或第五指示信息确定终端上报的非零元素的个数K
2<K
0和非零元素的位置,如网络设备可以根据所述终端对编码数据进行处理后的数据,通过掩模向量填充模块输出待解码数据,该待解码数据可以是K
2个非零元素位置处为反量化处理后的数据,K
0-K
2个元素位置处补0。
所述网络设备根据所述终端对编码数据进行处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;例如:网络设备侧可以根据第三指示信息或第四指示信息或第五指示信息确定终端上报的第一元素的个数K
1、K
1个第一元素中非零元素的个数K
2以及非零元素的位置,如网络设备可以根据所述终端对编码数据进行处理后的数据,通过掩模向量填充模块输出待解码 数据,该待解码数据可以是为K
2个非零元素位置处为反量化处理后的数据,剩余的K
0-K
1个元素和K
1-K
2个元素位置处补0。
可选地,所述方法还包括:
所述网络设备接收所述终端发送的信道质量指示CQI。
具体的,所述终端根据其对信道信息编码后的编码数据进行处理后得到的数据,通过所述第一网络模型得到恢复的第二信道信息;根据所述第二信道信息,计算得到CQI,并向所述网络设备发送所述CQI。
例如:终端可以对所述编码数据中的第一元素通过量化器模块进行量化处理,得到量化处理后的数据,再将量化处理后的数据输入到对应的反量化器模块进行反量化处理,并将反量化处理后的数据输入到第一网络模型中的解码器,解码得到恢复的第二信道信息,再根据恢复的第二信道信息计算CQI。
或者,终端可以将所述编码数据通过掩模向量模块进行处理确定第一元素,以及将第一元素通过量化器模块进行量化处理,得到量化处理后的数据,再将量化处理后的数据输入到对应的反量化器模块进行反量化处理,以及将量化处理后的数据输入到掩模向量填充模块进行处理,并将处理后的数据输入到第一网络模型中的解码器,解码得到恢复的第二信道信息,再根据恢复的第二信道信息计算CQI。
可选地,所述方法还包括:
所述网络设备向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;
所述网络设备接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信息参考信号计算得到的。
具体的,终端接收波束赋形的信道信息参考信号;根据所述信道信息参考信号,确定有效信道信息;根据所述有效信道信息,计算得到CQI,并向所述网络设备发送所述CQI。
例如:该信道信息参考信号可以是CSI参考信号(CSI Reference Signal,CSI-RS),终端根据接收的波束赋形CSI-RS估计下行有效信道信息,然后根据估计的有效信道信息计算CQI,其中CSI-RS所采用的波束由位于网络设备 侧的第一网络模型恢复的信道信息确定。
可选地,所述方法还包括:
所述网络设备接收所述终端发送的下行信道的干扰与噪声相关信息;
所述网络设备根据所述下行信道的干扰与噪声相关信息和所述恢复得到的信道信息,计算得到CQI。
需要说明的是,本公开实施例中终端侧和网络设备侧的信息处理方法是对应的,两者实施例可以互相参见,重复之处不再赘述。
以上实施例就本公开的网络设备侧的信息处理方法做出介绍,下面本实施例将结合附图对其对应的网络设备做进一步说明。
具体地,如图10所示,本公开实施例提供一种网络设备1000,包括:
第一发送单元1010,用于向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;
第一接收单元1020,用于接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;
第一处理单元1030,用于根据所述终端对编码数据进行处理后的数据,确定待解码数据;
解码单元1040,用于根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
可选地,所述下行信号为下行参考信号;所述网络设备1000还包括:
第二接收单元,用于接收所述终端发送的第一上行信号;
第二处理单元,用于根据所述第一上行信号,确定所述第一网络模型。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述网络设备1000还包括:
第三接收单元,用于接收所述终端发送的第二上行信号;
第三处理单元,用于根据所述第二上行信号,确定第三信道信息;
第四处理单元,用于根据所述第三信道信息,确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号 携带预编码矩阵指示PMI。
可选地,所述第四处理单元还用于:
根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;
根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述网络设备从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述网络设备从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述第四处理单元还用于:
根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述网络设备1000还包括以下一项:
第四接收单元,用于在L≥1时,接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述第一元素在所述编码数据中的位置,所述第一元素为所 述编码数据中的部分或全部元素;
第五接收单元,用于在L>1时,接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
第六接收单元,用于在L>1时,接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述第一处理单元1030还用于:
根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;
根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
可选地,所述所述第一处理单元1030还用于以下一项:
根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的位置确定所述待解码数据;
根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在 所述编码数据中的位置,确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据。
可选地,所述网络设备1000还包括:
第七接收单元,用于接收所述终端发送的信道质量指示CQI。
可选地,所述网络设备1000还包括:
第二发送单元,用于向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;
第八接收单元,用于接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信息参考信号计算得到的。
可选地,所述网络设备1000还包括:
第九接收单元,用于接收所述终端发送的下行信道的干扰与噪声相关信息;
计算单元,用于根据所述下行信道的干扰与噪声相关信息和所述恢复得到的信道信息,计算得到CQI。
需要说明的是,本公开实施例中对单元的划分是示意性的,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。另外,在本公开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。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元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,可以存储在一个处理器可读取存储介质中。基于这样的理解,本公开的技术方案本质上或者说对相关技术做出贡献的部分或者该技术方案的全部或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)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读 存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
在此需要说明的是,本公开实施例提供的上述网络设备,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
如图11所示,本实施例提供一种网络设备,包括存储器1101,收发机1102,处理器1103;其中,存储器1101用于存储计算机程序;收发机1102用于在所述处理器1103的控制下收发数据;如收发机1102用于在处理器1103的控制下接收和发送数据;处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;
接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;
根据所述终端对编码数据进行处理后的数据,确定待解码数据;
根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
可选地,所述下行信号为下行参考信号;所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
接收所述终端发送的第一上行信号;
根据所述第一上行信号,确定所述第一网络模型。
可选地,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
接收所述终端发送的第二上行信号;
根据所述第二上行信号,确定第三信道信息;
根据所述第三信道信息,确定所述第一网络模型。
可选地,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;
根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数;
若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;
若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;
其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
可选地,所述第一信道信息包括L个信道信息,L为正整数;其中,
可选地,所述第一信道信息包括L个信道信息,且L≥1;所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作中的一项:
在L≥1时,接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述 第一元素在所述编码数据中的位置,所述第一元素为所述编码数据中的部分或全部元素;
在L>1时,接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;
在L>1时,接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
可选地,所述第一元素包括以下一项:
所述编码数据中的K
1个非零元素;其中,K
1为正整数,K
1≤K
0,K
0为所述编码数据中所有元素的个数;
根据掩模向量确定的所述编码数据中的K
1个元素;
所述编码数据中从第一位置开始的连续K
1个元素。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;
根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作中的一项:
根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的 位置确定所述待解码数据;
根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;
根据所述反量化处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
接收所述终端发送的信道质量指示CQI。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;
接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信息参考信号计算得到的。
可选地,所述处理器1103用于读取所述存储器1101中的计算机程序并执行以下操作:
接收所述终端发送的下行信道的干扰与噪声相关信息;
根据所述下行信道的干扰与噪声相关信息和所述恢复得到的信道信息,计算得到CQI。
其中,在图11中,总线架构可以包括任意数量的互联的总线和桥,具体由处理器1103代表的一个或多个处理器和存储器1102代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起,这些都是本领域所公知的,因此,本文不再对其进行进一步描述。总线接口提供接口。收发机1102可以是多个元件,即包括发送机和接收机,提供用于在传输介质上与各种其他装置通信的单元,这些传输介质包括,这些传输介质包括无线信道、有线信道、光缆等传输介质。处理器1103负责管理总线架构和通常的处理,存储器1101可以存储处理器1103在执行操作时所使用的数据。
可选的,处理器1103可以是中央处理器(Central Processing Unit,CPU)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或复杂可编程逻辑器件(Complex Programmable Logic Device,CPLD),处理器也可以采用多核架构。
处理器通过调用存储器存储的计算机程序,用于按照获得的可执行指令执行本公开实施例提供的任一所述方法。处理器与存储器也可以物理上分开布置。
在此需要说明的是,本公开实施例提供的上述装置,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
本公开实施例还提供一种处理器可读存储介质,所述处理器可读存储介质存储有计算机程序,所述计算机程序用于使所述处理器执行上述信息处理方法中的步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
所述处理器可读存储介质可以是处理器能够存取的任何可用介质或数据存储设备,包括但不限于磁性存储器(例如软盘、硬盘、磁带、磁光盘(MO)等)、光学存储器(例如CD、DVD、BD、HVD等)、以及半导体存储器(例如ROM、EPROM、EEPROM、非易失性存储器(NAND FLASH)、固态硬盘(SSD))等。
以下结合具体实施例,对本公开实施例的处理方法进行说明:
实施例一:网络侧或终端侧部署包含编码器和解码器的I个网络模型
该实施例中,终端侧确定第一网络模型,如UE(即终端)从gNB(即网络设备)下载了I个网络模型用于推理实现CSI的压缩反馈。UE根据接收的CSI-RS估计出不同频域单元上的下行信道状态信息h
f,每个频域单元可以是子带、PRB或子载波等。令所有频域单元的第一根接收天线对应的下行信道的构成的矩阵
作为第i个编码器的输入,N
t是网络侧的发送天线端口数,N
c是频域单元的个数。该编码器的输出为一个长度为k
i的码字,然后把该码字输入到一个量化器模块使得码字的每个元素采用Q=2bits进行量化。把量化后的码字再输入到一个反量化器模块,输出相应的非量化信息。 最后把输出的非量化信息作为第i个解码器的输入,该编码器的输出为恢复的下行信道,如图12所示。
UE采用恢复的下行信道h′
f,i计算各频域单元的信干噪比γ和相应的宽带的信干噪比γ
WB,i。假设预设的或gNB配置给UE的宽带信干噪比的最小值为γ
T。若UE计算的γ
WB,i<γ
T,这表明UE选择的第i个模型不能满足要求,则UE选择第i+1个网络模型,把计算的下行信道h
f,i输入到第i+1个网络模型中,最终得到第i+1网络模型对应的解码器所输出的恢复的下行信道h′
f,i+1。UE在根据h′
f,i+1计算出宽带的γ
WB,i+1。若γ
WB,i+1≥γ
T,则该网络模型满足要求,否则UE会选择第i+2个网络模型,直至根据所选的网络模型能够使计算的宽带信干噪比不小于γ
T,并把所选的最优网络模型通过上行信道发送给gNB。若根据所有I个网络模型所计算的宽带信干噪比均不满足要求,则UE选择一个宽带信干噪比最大所对应的网络模型用于CSI的压缩反馈。
gNB根据接收的一个长度为k
i的码字信息确定UE所选的网络模型,从而确定用于恢复下行信道的解码器,再根据所接收UE发送的码字量化后信息,利用所选的解码器推理出特征向量h′
f,用于下行数据传输的预编码。
实施例二:基于AI的第一层的CSI压缩反馈
UE根据接收的CSI-RS估计出不同频域单元上的下行信道状态信息H
f,再对第f个频域单元的信道状态信息做奇异值分解得到对应的特征向量v
f。每个频域单元也可以是子带、PRB或子载波等。所有频域单元下行信道的最大特征值对应的特征向量构成的矩阵
作为自适应编码器的输入,该编码器的输出为一个包含K
0个元素的码字C,如图13所示。
以下给出了掩模向量分别由gNB配置或UE选择两种方式。
方式1):假设gNB给UE配置了一个前K位为1,后K
0-K
1位为0的向量的掩模向量M
1,然后把该码字输入到该掩模向量模块,最后输出一个长度为K
1码字。即码字C经过掩膜向量模块后得到
Z
masking=C⊙M
i
根据上式计算,Z
masking等价于对码字C中的后K
0-K
1元素位置0,只保留了码字C中前K
1位元素内容,即截短后的信息长度为K
1;UE再把截短后的码字信息输入到量化模块。其中,量化器模块采用Q=2bits量化每个元 素。例如,把第i个元素a
i量化为A,其中B<a
i≤A,A的值两个比特对应的值表示。UE把量化后的二进制比特流信息上报给gNB。gNB把比特流信息输入到反量化器模块得到K
1个元素量化后的值,然后把该K
1个元素量化后的值和K
1个元素后面添加K
0-K
1个全为0的元素,得到码字C’。最后gNB把码字C’输入到自适应解码器,利用自适解码器网络模型的推理得到恢复的特征向量v′
f,i,用于下行数据传输的预编码。
方式2):假设上述的掩模向量为UE选择确定,UE选择掩模向量的方法同UE选择网络模型的方法。UE需要通过
比特指示掩模向量中前面全为1的个数为K
1,并通过上行信道把该指示信息发送给gNB。gNB根据此指示信息确定UE所采用的掩膜向量,还可确定掩模向量填充的输出信息为前K
1个元素为反量化器模块的输出信息,后K
0-K
1个全为0的元素。其它的UE和gNB处理流程同上。
UE还可上报非零元素的个数和非零元素的位置信息一指示上报的元素。例如:图14中a)所示码字C中元素的个数K
0=16,UE通过掩模向量截短只上报了上述的K
1=12个元素中K
2=6个非零元素,并通过一个位图指示6个非零元素的位置。如图14中b)所示,只选择了前12个元素,即元素编号0~11。但是UE只上报了6个元素,这6个元素的位置如图14中c)所示,对应位置包含黑点的表示需上报的元素。UE通过
指示上报元素的个数。UE还通过一个如图14中d)所示位图指示,位图中对应比特值为1表示上报的元素,为0表示该元素不上报。
gNB根据非零元素个数指示信息确定非零元素的个数,再根据非零元素位置指示信息确定非零元素的位置。gNB根据非零元素的个数确定量化后的值,量化后的值作为掩膜向量填充模块的输入。gNB确定掩模向量填充模块的输出信息为上报的非零元素位置为量化后的值与再把上报非零元素之外的位置处全补为0。
实施例三:基于AI的L>1层的CSI压缩反馈
假设UE端有R=4根天线接收数据。UE根据接收的CSI-RS估计出不同频域单元上的下行信道状态信息H
f,再对第f个频域单元的信道状态信息做 奇异值分解得到对应的特征向量v
f。所有频域单元下行信道的L=4个特征值对应的特征向量构成4个矩阵
分别作为同4个自适应编码器的输入,可输出4个码字C,假设4个码字的长度相等,即K
0,1=K
0,2=K
0,3=K
0,4=16。gNB配置每个码字中上报元素的最大个数为
UE根据估计的下行信道确定4个码字上报元素的个数分别为10,10,8和6,再分别对4个码字中前10,10,8和6量化后上报给gNB。
UE还通过4个
分别指示第1~4层上报元素的个数。可选地,UE通过1个
指示第1和2层上报元素的个数,再通过2个
分别指示第3层和4层上报元素的个数。gNB根据UE上报元素个数指示信息确定各码字中上报的元素。
本方实施例给出了网络模型或掩膜向量选择方法能够根据不同用户的信道差异选择合适的网络模型,以减少终端的反馈开销。并且本公开实施例还给出了L层压缩反馈的网络模型输入信息形式和L个码字的上报元素及上报元素个数和位置的确定方法,还可以实现基于人工智能的L层信道信息的有效压缩反馈。
本领域内的技术人员应明白,本公开的实施例可提供为方法、系统、或计算机程序产品。因此,本公开可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本公开可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本公开是参照根据本公开实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机可执行指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机可执行指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些处理器可执行指令也可存储在能引导计算机或其他可编程数据处理 设备以特定方式工作的处理器可读存储器中,使得存储在该处理器可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些处理器可执行指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
此外,需要指出的是,在本公开的装置和方法中,显然,各部件或各步骤是可以分解和/或重新组合的。这些分解和/或重新组合应视为本公开的等效方案。并且,执行上述系列处理的步骤可以自然地按照说明的顺序按时间顺序执行,但是并不需要一定按照时间顺序执行,某些步骤可以并行或彼此独立地执行。对本领域的普通技术人员而言,能够理解本公开的方法和装置的全部或者任何步骤或者部件,可以在任何计算装置(包括处理器、存储介质等)或者计算装置的网络中,以硬件、固件、软件或者它们的组合加以实现,这是本领域普通技术人员在阅读了本公开的说明的情况下运用他们的基本编程技能就能实现的。
需要说明的是,应理解以上各个模块的划分仅仅是一种逻辑功能的划分,实际实现时可以全部或部分集成到一个物理实体上,也可以物理上分开。且这些模块可以全部以软件通过处理元件调用的形式实现;也可以全部以硬件的形式实现;还可以部分模块通过处理元件调用软件的形式实现,部分模块通过硬件的形式实现。例如,某个模块可以为单独设立的处理元件,也可以集成在上述装置的某一个芯片中实现,此外,也可以以程序代码的形式存储于上述装置的存储器中,由上述装置的某一个处理元件调用并执行以上确定模块的功能。其它模块的实现与之类似。此外这些模块全部或部分可以集成在一起,也可以独立实现。这里所述的处理元件可以是一种集成电路,具有信号的处理能力。在实现过程中,上述方法的各步骤或以上各个模块可以通过处理器元件中的硬件的集成逻辑电路或者软件形式的指令完成。
例如,各个模块、单元、子单元或子模块可以是被配置成实施以上方法的一个或多个集成电路,例如:一个或多个特定集成电路(Application Specific Integrated Circuit,ASIC),或,一个或多个微处理器(digital signal processor,DSP),或,一个或者多个现场可编程门阵列(Field Programmable Gate Array,FPGA)等。再如,当以上某个模块通过处理元件调度程序代码的形式实现时,该处理元件可以是通用处理器,例如中央处理器(Central Processing Unit,CPU)或其它可以调用程序代码的处理器。再如,这些模块可以集成在一起,以片上系统(system-on-a-chip,SOC)的形式实现。
本公开的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本公开的实施例,例如除了在这里图示或描述的那些以外的顺序实施。此外,术语“包括”和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。此外,说明书以及权利要求中使用“和/或”表示所连接对象的至少其中之一,例如A和/或B和/或C,表示包含单独A,单独B,单独C,以及A和B都存在,B和C都存在,A和C都存在,以及A、B和C都存在的7种情况。类似地,本说明书以及权利要求中使用“A和B中的至少一个”应理解为“单独A,单独B,或A和B都存在”。
显然,本领域的技术人员可以对本公开进行各种改动和变型而不脱离本公开的精神和范围。这样,倘若本公开的这些修改和变型属于本公开权利要求及其等同技术的范围之内,则本公开也意图包含这些改动和变型在内。
Claims (71)
- 一种信息处理方法,包括:终端接收网络设备发送的下行信号;所述终端根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;所述终端通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;所述终端将所述编码数据进行处理后的数据发送至所述网络设备。
- 根据权利要求1所述的信息处理方法,其中,所述下行信号为下行参考信号;所述终端根据所述下行信号,确定第一网络模型,包括:所述终端根据所述下行参考信号,确定所述第一信道信息;所述终端根据所述第一信道信息,确定所述第一网络模型。
- 根据权利要求2所述的信息处理方法,其中,所述终端根据所述第一信道信息,确定所述第一网络模型之后,还包括:所述终端向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
- 根据权利要求2所述的信息处理方法,其中,所述终端根据所述第一信道信息,确定所述第一网络模型,包括:所述终端根据所述第一信道信息,通过所述I个网络模型分别进行编码 和/或解码处理,得到每个网络模型对应的第二信道信息;所述终端根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述终端从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述终端从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
- 根据权利要求5所述的信息处理方法,其中,所述终端根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数,包括:所述终端根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
- 根据权利要求1所述的信息处理方法,其中,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述终端接收网络设备发送的下行信号之前,还包括:所述终端向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
- 根据权利要求7所述的信息处理方法,其中,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
- 根据权利要求1所述的信息处理方法,其中,所述第一信道信息包括L个信道信息,L为正整数;其中,所述L个信道信息中的第l个信道信息包括以下一项:第一频域单元的R个数据流的预编码向量的组合;其中,L=1;第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中,L=1;第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;其中,所述第一频域单元为所有频域单元或一个频域单元。
- 根据权利要求9所述的信息处理方法,其中,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块PRB,或者所述频域单元为一 个包含多个PRB的子带。
- 根据权利要求1所述的信息处理方法,其中,所述终端将所述编码数据进行处理后的数据发送至所述网络设备,包括:所述终端对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
- 根据权利要求11所述的信息处理方法,其中,所述第一元素包括以下一项:所述编码数据中的K 1个非零元素;其中,K 1为正整数,K 1≤K 0,K 0为所述编码数据中所有元素的个数;根据掩模向量确定的所述编码数据中的K 1个元素;所述编码数据中从第一位置开始的连续K 1个元素。
- 根据权利要求12所述的信息处理方法,其中,所述第一信道信息包括L个信道信息,且L≥1;所述方法还包括以下一项:在L≥1时,所述终端向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则所述终端针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;在L>1时,所述终端向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
- 根据权利要求13所述的信息处理方法,其中,在所述第一元素包括 所述编码数据中的K 1个非零元素的情况下:若K 1<K 0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K 1以及所述K 1个非零元素在所述编码数据中的位置;若K 1=K 0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K 1。
- 根据权利要求1所述的信息处理方法,其中,所述终端根据所述下行信号,确定第一网络模型之后,还包括:所述终端根据所述编码数据进行处理后的数据,通过所述第一网络模型得到恢复的第二信道信息;所述终端根据所述第二信道信息,计算得到信道质量指示CQI;所述终端向所述网络设备发送所述CQI。
- 根据权利要求1所述的信息处理方法,其中,所述终端根据所述下 行信号,确定第一网络模型之后,还包括:所述终端接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;所述终端根据所述信道信息参考信号,确定有效信道信息;所述终端根据所述有效信道信息,计算得到CQI;所述终端向所述网络设备发送所述CQI。
- 根据权利要求1所述的信息处理方法,还包括:所述终端向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
- 一种信息处理方法,包括:网络设备向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;所述网络设备接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据;所述网络设备根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
- 根据权利要求21所述的信息处理方法,其中,所述下行信号为下行参考信号;所述网络设备根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息之前,还包括:所述网络设备接收所述终端发送的第一上行信号;所述网络设备根据所述第一上行信号,确定所述第一网络模型。
- 根据权利要求21所述的信息处理方法,其中,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述网络设备向终端发送下行信号之前,还包括:所述网络设备接收所述终端发送的第二上行信号;所述网络设备根据所述第二上行信号,确定第三信道信息;所述网络设备根据所述第三信道信息,确定所述第一网络模型。
- 根据权利要求23所述的信息处理方法,其中,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
- 根据权利要求23所述的信息处理方法,其中,所述网络设备根据所述第三信道信息,确定所述第一网络模型,包括:所述网络设备根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;所述网络设备根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数;若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则所述网络设备从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则所述网络设备从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
- 根据权利要求25所述的信息处理方法,其中,所述网络设备根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数,包括:所述网络设备根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
- 根据权利要求21所述的信息处理方法,其中,所述第一信道信息包括L个信道信息,且L≥1;所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据之前,还包括以下一项:在L≥1时,所述网络设备接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述第一元素在所述编码数据中的位置,所述第一元素为所述编码数据中的部分或全部元素;在L>1时,所述网络设备接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;在L>1时,所述网络设备接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
- 根据权利要求28所述的信息处理方法,其中,所述第一元素包括以下一项:所述编码数据中的K 1个非零元素;其中,K 1为正整数,K 1≤K 0,K 0为所述编码数据中所有元素的个数;根据掩模向量确定的所述编码数据中的K 1个元素;所述编码数据中从第一位置开始的连续K 1个元素。
- 根据权利要求28或29所述的信息处理方法,其中,所述网络设备根据所述终端对编码数据进行处理后的数据,确定待解码数据,包括:所述网络设备根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;所述网络设备根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
- 根据权利要求30所述的信息处理方法,其中,所述网络设备根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据,包括以下一项:所述网络设备根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;所述网络设备根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;所述网络设备根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的位置确定所述待解码数据;所述网络设备根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;所述网络设备根据所述反量化处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据。
- 根据权利要求21所述的信息处理方法,还包括:所述网络设备接收所述终端发送的信道质量指示CQI。
- 根据权利要求21所述的信息处理方法,还包括:所述网络设备向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;所述网络设备接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信息参考信号计算得到的。
- 根据权利要求21所述的信息处理方法,还包括:所述网络设备接收所述终端发送的下行信道的干扰与噪声相关信息;所述网络设备根据所述下行信道的干扰与噪声相关信息和所述恢复得到 的信道信息,计算得到CQI。
- 一种信息处理装置,包括存储器,收发机,处理器;其中,存储器用于存储计算机程序;收发机用于在所述处理器的控制下收发数据;处理器用于读取所述存储器中的计算机程序并执行以下操作:接收网络设备发送的下行信号;根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;将所述编码数据进行处理后的数据发送至所述网络设备。
- 根据权利要求35所述的信息处理装置,其中,所述下行信号为下行参考信号;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述下行参考信号,确定所述第一信道信息;根据所述第一信道信息,确定所述第一网络模型。
- 根据权利要求36所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:向所述网络设备发送第一上行信号;其中,所述第一上行信号携带用于指示所述第一网络模型的第一指示信息。
- 根据权利要求36所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述第一信道信息,通过所述I个网络模型分别进行编码和/或解码 处理,得到每个网络模型对应的第二信道信息;根据所述第一信道信息和每个网络模型对应的第二信道信息,计算得到每个网络模型对应的目标参数;若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
- 根据权利要求39所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述第一信道信息和每个网络模型对应的所述第二信道信息,通过目标准则计算得到每个网络模型对应的目标参数;其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均方差准则,或者所述目标准则为宽带信干噪比准则。
- 根据权利要求35所述的信息处理装置,其中,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:向所述网络设备发送第二上行信号;其中,所述第二上行信号用于所述网络设备确定所述第一网络模型。
- 根据权利要求41所述的信息处理装置,其中,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
- 根据权利要求35所述的信息处理装置,其中,所述第一信道信息包括L个信道信息,L为正整数;其中,所述L个信道信息中的第l个信道信息包括以下一项:第一频域单元的R个数据流的预编码向量的组合;其中,L=1;第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号的组合;其中,L=1;第一频域单元的R根接收天线分别对应的下行信道信息的组合;其中,L=1;第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量的组合;其中,L=1;第一频域单元的R根接收天线分别对应的下行信道信息与(v2-R)个所述零向量的组合;其中,L=1;v2是所述终端的最大接收天线个数;第一频域单元的R根接收天线分别对应的下行参考信号经过信道的接收信号与(v2-R)个所述零向量的组合;其中,L=1;第一频域单元的下行信道信息通过奇异值分解得到的R个特征向量与(v1-R)个所述零向量的组合;其中,L=1;其中,所述第一频域单元为所有频域单元或一个频域单元。
- 根据权利要求43所述的信息处理装置,其中,所述频域单元为一个子载波,或者所述频域单元为一个物理资源块PRB,或者所述频域单元为一个包含多个PRB的子带。
- 根据权利要求35所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:对所述编码数据中的第一元素进行量化处理,并将量化处理后的数据传输至所述网络设备;其中,所述第一元素为所述编码数据中的部分或全部元素。
- 根据权利要求45所述的信息处理装置,其中,所述第一元素包括以下一项:所述编码数据中的K 1个非零元素;其中,K 1为正整数,K 1≤K 0,K 0为所述编码数据中所有元素的个数;根据掩模向量确定的所述编码数据中的K 1个元素;所述编码数据中从第一位置开始的连续K 1个元素。
- 根据权利要求46所述的信息处理装置,其中,所述第一信道信息包括L个信道信息,且L≥1;所述处理器用于读取所述存储器中的计算机程序并执行以下操作中的一项:在L≥1时,向所述网络设备发送第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;在L>1时,若第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同,则针对所述第一部分信道信息,向所述网络设备发送共同的第四指示信息;其中,所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息;在L>1时,向所述网络设备发送针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
- 根据权利要求47所述的信息处理装置,其中,在所述第一元素包括所述编码数据中的K 1个非零元素的情况下:若K 1<K 0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K 1以及所述K 1个非零元素在所述编码数据中的位置;若K 1=K 0,则所述第三指示信息或所述第四指示信息用于指示所述非零元素的个数K 1。
- 根据权利要求35所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述编码数据进行处理后的数据,通过所述第一网络模型得到恢复的第二信道信息;根据所述第二信道信息,计算得到信道质量指示CQI;向所述网络设备发送所述CQI。
- 根据权利要求35所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:接收所述网络设备发送的波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备恢复得到的信道信息确定;根据所述信道信息参考信号,确定有效信道信息;根据所述有效信道信息,计算得到CQI;向所述网络设备发送所述CQI。
- 根据权利要求35所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:向所述网络设备发送下行信道的干扰与噪声相关信息;其中,所述下行信道的干扰与噪声相关信息用于所述网络设备计算CQI。
- 一种终端,包括:第一接收单元,用于接收网络设备发送的下行信号;第一处理单元,用于根据所述下行信号,确定第一网络模型;其中,所述第一网络模型为预先部署或从网络设备侧下载的I个网络模型中的至少一个,I为正整数;编码单元,用于通过所述第一网络模型对第一信道信息进行编码处理,得到待传输的编码数据;第一发送单元,用于将所述编码数据进行处理后的数据发送至所述网络设备。
- 一种信息处理装置,包括存储器,收发机,处理器;其中,存储器用于存储计算机程序;收发机用于在所述处理器的控制下收发数据;处理器用于读取所述存储器中的计算机程序并执行以下操作:向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;根据所述终端对编码数据进行处理后的数据,确定待解码数据;根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
- 根据权利要求56所述的信息处理装置,其中,所述下行信号为下行参考信号;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:接收所述终端发送的第一上行信号;根据所述第一上行信号,确定所述第一网络模型。
- 根据权利要求56所述的信息处理装置,其中,所述下行信号携带用于指示所述第一网络模型的第二指示信息;所述处理器用于读取所述存储器中的计算机程序并执行以下操作:接收所述终端发送的第二上行信号;根据所述第二上行信号,确定第三信道信息;根据所述第三信道信息,确定所述第一网络模型。
- 根据权利要求58所述的信息处理装置,其中,所述第二上行信号为上行参考信号;或者,所述第二上行信号携带预编码矩阵指示PMI。
- 根据权利要求58所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述第三信道信息,通过所述I个网络模型分别进行编码和/或解码处理,得到每个网络模型对应的第四信道信息;根据所述第三信道信息和每个网络模型对应的第四信道信息,计算得到每个网络模型对应的目标参数;若所述I个网络模型中包括所述目标参数满足预设条件的M≥1个网络模型,则从M个网络模型中选择编码处理后得到的编码数据中元素个数最小的网络模型,确定为所述第一网络模型;若所述I个网络模型中包括所述目标参数满足预设条件的M=0个网络模型,则从所述I个网络模型中选择编码处理后得到的编码数据中元素个数最大的网络模型,确定为所述第一网络模型。
- 根据权利要求60所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述第三信道信息和每个网络模型对应的所述第四信道信息,通过目标准则计算得到每个网络模型对应的目标参数;其中,所述目标准则为余弦相似度准则,或者所述目标准则为归一化均 方差准则,或者所述目标准则为宽带信干噪比准则。
- 根据权利要求56所述的信息处理装置,其中,所述第一信道信息包括L个信道信息,且L≥1;所述处理器用于读取所述存储器中的计算机程序并执行以下操作中的一项:在L≥1时,接收所述终端发送的第三指示信息;其中,一个信道信息对应一个第三指示信息,所述第三指示信息用于指示第一元素的个数和/或所述第一元素在所述编码数据中的位置,所述第一元素为所述编码数据中的部分或全部元素;在L>1时,接收所述终端针对第一部分信道信息发送的共同的第四指示信息;其中,所述第一部分信道信息为所述L个信道信息中的部分或全部信道信息,且所述第一部分信道信息编码后的编码数据中所述第一元素的个数和/或所述第一元素在所述编码数据中的位置相同;所述第四指示信息用于指示所述第一元素的个数和/或所述第一元素在所述编码数据中的位置;在L>1时,接收所述终端发送的针对所述L个信道信息联合指示的第五指示信息;其中,所述第五指示信息用于指示每个信道信息编码后的编码数据中第一元素个数和/或所述第一元素在所述编码数据中的位置;或者,所述第五指示信息用于指示L个信道信息编码后的L个编码数据中第一元素的总个数和/或所述第一元素在所述编码数据中的位置。
- 根据权利要求63所述的信息处理装置,其中,所述第一元素包括以下一项:所述编码数据中的K 1个非零元素;其中,K 1为正整数,K 1≤K 0,K 0为所 述编码数据中所有元素的个数;根据掩模向量确定的所述编码数据中的K 1个元素;所述编码数据中从第一位置开始的连续K 1个元素。
- 根据权利要求63或64所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:根据所述终端对编码数据进行处理后的数据进行反量化处理,得到反量化处理后的数据;根据所述反量化处理后的数据,通过所述第三指示信息或所述第四指示信息或所述第五指示信息,确定所述待解码数据。
- 根据权利要求65所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作中的一项:根据所述反量化处理后的数据,通过掩模向量确定所述待解码数据;根据所述反量化处理后的数据,通过所述第一元素的个数确定所述待解码数据;根据所述反量化处理后的数据,通过所述第一元素在所述编码数据中的位置确定所述待解码数据;根据所述反量化处理后的数据,通过非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据;根据所述反量化处理后的数据,通过所述第一元素的个数、所述第一元素中非零元素的个数和所述非零元素在所述编码数据中的位置,确定所述待解码数据。
- 根据权利要求56所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:接收所述终端发送的信道质量指示CQI。
- 根据权利要求56所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:向所述终端发送波束赋形的信道信息参考信号;其中,所述信道信息参考信号所采用的波束根据所述网络设备的恢复得到的信道信息确定;接收所述终端发送的CQI;其中,所述CQI是所述终端根据所述信道信 息参考信号计算得到的。
- 根据权利要求56所述的信息处理装置,其中,所述处理器用于读取所述存储器中的计算机程序并执行以下操作:接收所述终端发送的下行信道的干扰与噪声相关信息;根据所述下行信道的干扰与噪声相关信息和所述恢复得到的信道信息,计算得到CQI。
- 一种网络设备,包括:第一发送单元,用于向终端发送下行信号;其中,所述下行信号用于所述终端确定第一网络模型,所述第一网络模型为预先部署的I个网络模型中的至少一个,I为正整数;第一接收单元,用于接收所述终端对编码数据进行处理后的数据;其中,所述编码数据是所述终端根据所述第一网络模型对第一信道信息进行编码处理后得到的;第一处理单元,用于根据所述终端对编码数据进行处理后的数据,确定待解码数据;解码单元,用于根据所述第一网络模型对所述待解码数据进行解码处理,得到恢复的信道信息。
- 一种处理器可读存储介质,所述处理器可读存储介质存储有计算机程序,所述计算机程序用于使所述处理器执行权利要求1至20中任一项所述的信息处理方法中的步骤,或权利要求21至34中任一项所述的信息处理方法中的步骤。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111605575.2 | 2021-12-25 | ||
CN202111605575.2A CN116346279A (zh) | 2021-12-25 | 2021-12-25 | 信息处理方法、装置、终端及网络设备 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023116407A1 true WO2023116407A1 (zh) | 2023-06-29 |
Family
ID=86875172
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2022/136615 WO2023116407A1 (zh) | 2021-12-25 | 2022-12-05 | 信息处理方法、装置、终端及网络设备 |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN116346279A (zh) |
TW (2) | TWI830543B (zh) |
WO (1) | WO2023116407A1 (zh) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021108940A1 (en) * | 2019-12-01 | 2021-06-10 | Nokia Shanghai Bell Co., Ltd. | Channel state information feedback |
US20210273707A1 (en) * | 2020-02-28 | 2021-09-02 | Qualcomm Incorporated | Neural network based channel state information feedback |
CN113381950A (zh) * | 2021-04-25 | 2021-09-10 | 清华大学 | 基于网络聚合策略的高效mimo信道反馈方法及装置 |
CN113660020A (zh) * | 2021-06-25 | 2021-11-16 | 陕西尚品信息科技有限公司 | 一种无线通信信道信息传输方法、系统和解码器 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10572830B2 (en) * | 2017-04-24 | 2020-02-25 | Virginia Tech Intellectual Properties, Inc. | Learning and deploying compression of radio signals |
CN113283571A (zh) * | 2017-06-19 | 2021-08-20 | 弗吉尼亚科技知识产权有限公司 | 使用多天线收发器无线传输的信息的编码和解码 |
CN110350958B (zh) * | 2019-06-13 | 2021-03-16 | 东南大学 | 一种基于神经网络的大规模mimo的csi多倍率压缩反馈方法 |
-
2021
- 2021-12-25 CN CN202111605575.2A patent/CN116346279A/zh active Pending
-
2022
- 2022-12-05 WO PCT/CN2022/136615 patent/WO2023116407A1/zh unknown
- 2022-12-19 TW TW111148663A patent/TWI830543B/zh active
- 2022-12-19 TW TW112149594A patent/TW202418794A/zh unknown
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021108940A1 (en) * | 2019-12-01 | 2021-06-10 | Nokia Shanghai Bell Co., Ltd. | Channel state information feedback |
US20210273707A1 (en) * | 2020-02-28 | 2021-09-02 | Qualcomm Incorporated | Neural network based channel state information feedback |
CN113381950A (zh) * | 2021-04-25 | 2021-09-10 | 清华大学 | 基于网络聚合策略的高效mimo信道反馈方法及装置 |
CN113660020A (zh) * | 2021-06-25 | 2021-11-16 | 陕西尚品信息科技有限公司 | 一种无线通信信道信息传输方法、系统和解码器 |
Also Published As
Publication number | Publication date |
---|---|
TWI830543B (zh) | 2024-01-21 |
TW202418794A (zh) | 2024-05-01 |
CN116346279A (zh) | 2023-06-27 |
TW202327323A (zh) | 2023-07-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111314034B (zh) | 用于csi报告和使用的增强频率压缩的方法和装置 | |
CN106899379B (zh) | 用于处理极化码的方法和通信设备 | |
CN109039406B (zh) | 一种信道状态信息发送、接收方法、存储介质及设备 | |
WO2018202071A1 (zh) | 数据传输方法、终端设备和网络设备 | |
WO2020083057A1 (zh) | 指示和确定预编码向量的方法以及通信装置 | |
KR20100081333A (ko) | 자가적응 코드북의 처리방법 | |
CN114270723A (zh) | 一种信道状态信息处理方法、电子设备及存储介质 | |
WO2023116407A1 (zh) | 信息处理方法、装置、终端及网络设备 | |
WO2022237360A1 (zh) | 码本指示方法、装置及存储介质 | |
US20240146582A1 (en) | Information encoding control method and related apparatus | |
WO2019222882A1 (en) | Channel state information feedback | |
WO2022236785A1 (zh) | 信道信息的反馈方法、收端设备和发端设备 | |
CN112205049B (zh) | 信道状态信息反馈 | |
CN110875767B (zh) | 指示和确定预编码向量的方法和通信装置 | |
WO2024026793A1 (zh) | 数据传输方法、装置、设备、存储介质及系统 | |
CN114866202A (zh) | Csi反馈方法及装置、存储介质、终端、网络设备 | |
WO2024032701A1 (zh) | 信道状态信息处理方法及装置 | |
WO2023202385A1 (zh) | 信息传输方法、装置及存储介质 | |
WO2024164903A1 (zh) | 性能评估方法及装置 | |
TWI856775B (zh) | 通道狀態資訊處理方法及裝置 | |
WO2024164858A1 (zh) | 一种性能评估方法、设备及可读存储介质 | |
WO2024026792A1 (zh) | 通信方法、装置、设备、存储介质、芯片及程序产品 | |
WO2024032387A1 (zh) | 码本参数传输方法、装置及存储介质 | |
WO2024169659A1 (zh) | Ptrs传输方法、装置及存储介质 | |
WO2023231933A1 (zh) | 一种通信方法及装置 |
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: 22909722 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022909722 Country of ref document: EP Effective date: 20240725 |