WO2023207920A1 - 信道信息反馈方法、终端及网络侧设备 - Google Patents

信道信息反馈方法、终端及网络侧设备 Download PDF

Info

Publication number
WO2023207920A1
WO2023207920A1 PCT/CN2023/090381 CN2023090381W WO2023207920A1 WO 2023207920 A1 WO2023207920 A1 WO 2023207920A1 CN 2023090381 W CN2023090381 W CN 2023090381W WO 2023207920 A1 WO2023207920 A1 WO 2023207920A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
information
encoding
side device
target
Prior art date
Application number
PCT/CN2023/090381
Other languages
English (en)
French (fr)
Inventor
任千尧
谢天
Original Assignee
维沃移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2023207920A1 publication Critical patent/WO2023207920A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0417—Feedback systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0456—Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • 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
    • H04L1/0023—Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the signalling
    • H04L1/0026—Transmission of channel quality indication
    • 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
    • H04L1/02—Arrangements for detecting or preventing errors in the information received by diversity reception
    • H04L1/06—Arrangements for detecting or preventing errors in the information received by diversity reception using space diversity
    • 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
    • H04L1/02—Arrangements for detecting or preventing errors in the information received by diversity reception
    • H04L1/06—Arrangements for detecting or preventing errors in the information received by diversity reception using space diversity
    • H04L1/0618—Space-time coding
    • H04L1/0675—Space-time coding characterised by the signaling
    • H04L1/0693—Partial feedback, e.g. partial channel state information [CSI]

Definitions

  • This application belongs to the field of communication technology, and specifically relates to a channel information feedback method, terminal and network side equipment.
  • the terminal When the terminal communicates with the network side device, it usually needs to feed back channel information to the network side device so that the network side device can perform precoding and user scheduling based on the channel information. Generally, when feeding back channel information, the terminal can feed back the precoding matrix of the channel to the network side device. At the same time, in order to reduce resource overhead, the terminal can also compress the precoding matrix and feed back the compressed precoding information to the network. side equipment.
  • a network model compression method can be used.
  • an artificial intelligence (Artificial Intelligence, AI) model can be used to compress the precoding matrix.
  • AI Artificial Intelligence
  • Embodiments of the present application provide a channel information feedback method, a terminal and a network-side device, which can solve the problem that the terminal cannot predict the channel information based on the model when the precoding matrix of the channel corresponds to multiple layers and the channel characteristics between the multiple layers are significantly different.
  • the problem of effective compression of the channel's precoding matrix and accurate feedback of channel information is significantly different.
  • a channel information feedback method which method includes:
  • the terminal determines the information value of the precoding matrix of each layer
  • For each layer determine the target coding model corresponding to the target interval according to the target interval in which the information value of the precoding matrix of the layer is located, and different information value intervals correspond to different coding models;
  • the precoding feature information and first information are sent to the network side device, where the first information includes the model identifier of the target coding model and/or the target interval.
  • a channel information feedback device which device includes:
  • the first determination module is used to determine the information value of the precoding matrix of each layer
  • the second determination module is used to determine, for each layer, the target coding model corresponding to the target interval according to the target interval in which the information value of the precoding matrix of the layer is located. Different information value intervals correspond to different coding models. ;
  • a coding module used to process the precoding matrix of the layer according to the target coding model to obtain precoding feature information
  • a sending module configured to send the precoding feature information and first information to a network side device, where the first information includes a model identifier of the target coding model and/or the target interval.
  • a channel information feedback method which method includes:
  • the network side device receives the precoding characteristic information and the first information
  • the precoding feature information is obtained by processing the precoding matrix of the layer based on a target coding model.
  • the target coding model is a coding model corresponding to the target interval where the information value of the precoding matrix of the layer is located. Different The information value intervals correspond to different encoding models, and the first information includes the model identifier of the target encoding model and/or the target interval.
  • a channel information feedback device which device includes:
  • a receiving module configured to receive precoding feature information and first information
  • a determining module configured to determine a target decoding model according to the first information
  • a decoding module configured to process the precoding feature information according to the target decoding model to obtain the corresponding precoding matrix
  • the precoding feature information is obtained by processing the precoding matrix of the layer based on a target coding model.
  • the target coding model is a coding model corresponding to the target interval where the information value of the precoding matrix of the layer is located. Different The information value intervals correspond to different encoding models, and the first information includes the model identifier of the target encoding model and/or the target interval.
  • a terminal in a fifth aspect, includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor.
  • the program or instructions are executed by the processor, the following implementations are implemented: The steps of the method described in one aspect.
  • a terminal including a processor and a communication interface, wherein the processor is used to determine the information value of the precoding matrix of each layer; for each layer, according to the precoding matrix of the layer The target interval where the information value is located is determined, and the target coding model corresponding to the target interval is determined. Different information value intervals correspond to different coding models; the precoding matrix of the layer is processed according to the target coding model to obtain precoding. Characteristic information, the communication interface is used to send the precoding characteristic information and first information to the network side device, where the first information includes the model identifier of the target coding model and/or the target interval.
  • a network side device in a seventh aspect, includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor.
  • the program or instructions are executed by the processor.
  • a network side device including a processor and a communication interface, wherein the communication interface is used to receive precoding characteristic information and first information; the processor is used to determine a target according to the first information Decoding model; process the precoding feature information according to the target decoding model to obtain the corresponding precoding matrix; wherein the precoding feature information is obtained by processing the precoding matrix of the layer based on the target coding model.
  • the target coding model is the coding model corresponding to the target interval where the information value of the precoding matrix of the layer is located. Different information value intervals correspond to different coding models.
  • the first information includes the model identifier of the target coding model and /or the target interval.
  • a ninth aspect provides a channel information feedback system, including: a terminal and a network side device.
  • the terminal can be used to perform the steps of the channel information feedback method described in the first aspect.
  • the network side device can be used to perform the steps of the channel information feedback method as described in the first aspect. The steps of the channel information feedback method described in the third aspect.
  • a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method are implemented as described in the first aspect. The steps of the method described in the third aspect.
  • a chip in an eleventh aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method described in the first aspect. method, or implement a method as described in the third aspect.
  • a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement as described in the first aspect
  • the terminal when the terminal encodes the precoding matrix of the channel and feeds back the channel information to the network side device, the corresponding coding model can be selected for each layer according to the interval where the information value of the precoding matrix of each layer is located.
  • the precoding matrix of each layer is encoded, and different information value intervals can correspond to different coding models. Therefore, for each layer, a coding model that better matches the channel characteristics of the layer can be used to precode the layer's precoding matrix. Encoding can be performed to improve the coding accuracy and accurately feedback channel information; when the terminal feeds back the encoded precoding feature information to the network side device, it can also use the model identification and/or target of the target coding model used by the terminal.
  • the target interval corresponding to the encoding model is fed back to the network side device, so the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • different encoding models can be trained according to different intervals of the information value of the precoding matrix of the layer, there is no need to train the corresponding encoding model for each layer or multiple layers, so the complexity of model training can also be reduced.
  • Figure 1 is a schematic diagram of a wireless communication system according to an embodiment of the present application.
  • Figure 2 is a schematic flow chart of a channel information feedback method according to an embodiment of the present application.
  • Figure 3 is a schematic flow chart of a channel information feedback method according to an embodiment of the present application.
  • Figure 4 is a schematic structural diagram of a channel information feedback device according to an embodiment of the present application.
  • Figure 5 is a schematic structural diagram of a channel information feedback device according to an embodiment of the present application.
  • Figure 6 is a schematic structural diagram of a communication device according to an embodiment of the present application.
  • Figure 7 is a schematic structural diagram of a terminal according to an embodiment of the present application.
  • Figure 8 is a schematic structural diagram of a network side device according to an embodiment of the present application.
  • a model can be used to compress the precoding matrix of the channel to obtain precoding feature information, and then the precoding feature information is fed back to the network-side device.
  • the network side device can decode the precoding feature information using a decoding model corresponding to the coding model, and thereby obtain channel information.
  • the terminal when the terminal feeds back the channel information to the network side device, it needs to feed back the channel information of each layer. That is, the terminal needs to use a model to encode the precoding matrix of each layer, and Feed back the precoding feature information obtained by encoding to the network side device.
  • the model is used to encode the precoding matrices of multiple layers. At this time, the training of the model is difficult, resulting in the inability to use the model to effectively encode the precoding matrix of each layer.
  • the maximum eigenvalue of each sampled data in multiple sampling data sets may be different, that is, the data in the training set represent different proportions of channel information, and the eigenvalue span will lead to channel regularity.
  • the changes are large, so a model with a larger parameter scale is required for processing, making model training more difficult.
  • a large model scale will also bring about the problem of reduced model performance.
  • the terminal in the related technology cannot effectively compress the precoding matrix of the channel based on the model. Furthermore, if the terminal cannot use a model to effectively encode the precoding matrix of the channel, it cannot accurately feedback channel information to the network side device.
  • the channel information feedback method includes: the terminal determines the information value of the precoding matrix of each layer; for each layer, According to the target interval where the information value of the precoding matrix of the layer is located, determine the target coding model corresponding to the target interval. Different information value intervals correspond to different coding models; process the precoding matrix of the layer according to the target coding model to obtain the precoding Feature information; send precoding feature information and first information to the network side device, where the first information includes The model identifier and/or target interval for the target encoding model.
  • the terminal when the terminal encodes the precoding matrix of the channel and feeds back the channel information to the network side device, the corresponding coding model can be selected for each layer according to the interval where the information value of the precoding matrix of each layer is located.
  • the precoding matrix of each layer is encoded, and different information value intervals can correspond to different coding models. Therefore, for each layer, a coding model that better matches the channel characteristics of the layer can be used to precode the layer's precoding matrix. Encoding can be performed to improve the coding accuracy and accurately feedback channel information; when the terminal feeds back the encoded precoding feature information to the network side device, it can also use the model identification and/or target of the target coding model used by the terminal.
  • the target interval corresponding to the encoding model is fed back to the network side device, so the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • different encoding models can be trained according to different intervals of the information value of the precoding matrix of the layer, there is no need to train the corresponding encoding model for each layer or multiple layers, so the complexity of model training can also be reduced.
  • first, second, etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and “second” are distinguished objects It is usually one type, and the number of objects is not limited.
  • the first object can be one or multiple.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects are in an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced, LTE-A Long Term Evolution
  • LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • NR New Radio
  • FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable.
  • the wireless communication system includes a terminal 11 and a network side device 12.
  • the terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palmtop computer, a netbook, or a super mobile personal computer.
  • Tablet Personal Computer Tablet Personal Computer
  • laptop computer laptop computer
  • PDA Personal Digital Assistant
  • PDA Personal Digital Assistant
  • UMPC ultra-mobile personal computer
  • UMPC mobile Internet device
  • MID mobile Internet Device
  • AR augmented reality
  • VR virtual reality
  • robots wearable devices
  • VUE vehicle-mounted equipment
  • PUE pedestrian terminals
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture etc.
  • PCs personal computers
  • teller machines or self-service machines and other terminal-side devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart Bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc.
  • the network side device 12 may include an access network device or a core network device, where the access network device 12 may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or Wireless access network unit.
  • the access network device 12 may include a base station, a Wireless Local Area Network (Wireless Local Area Network, WLAN) access point or a Wireless Fidelity (Wireless Fidelity, WiFi) node, etc.
  • the base station may be called a Node B, an Evolved Node B (Evolved NodeB, eNB), access point, Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home B Node, home evolved B node, transmitting receiving point (Transmitting Receiving Point, TRP) or some other suitable term in the field.
  • BTS Base Transceiver Station
  • BSS Basic Service Set
  • ESS Extended Service Set
  • Home B Node home evolved B node
  • TRP Transmitting Receiving Point
  • the base station is not limited to specific technical terms. It needs to be explained that , in the embodiment of this application, only the base station in the NR system is taken as an example for introduction, and the specific type of the base station is not limited.
  • this embodiment of the present application provides a channel information feedback method 200.
  • the channel information feedback method can be executed by a terminal.
  • the channel information feedback method can be executed by software or hardware installed on the terminal.
  • the channel information feedback method can be executed by software or hardware installed on the terminal.
  • the information feedback method includes the following steps.
  • S202 The terminal determines the information value of the precoding matrix of each layer.
  • the terminal may determine the information value of the precoding matrix of each layer of the channel.
  • the information value of the precoding matrix of the layer can describe the precoding characteristics of the layer.
  • the above information value may be a normalized value, and the specific method of normalization may be specified by the protocol.
  • the terminal determines the information value of the precoding matrix of each layer, which may include at least one of the following methods 1 and 2.
  • Method 1 The terminal determines the precoding matrix of the channel and the rank of the channel according to the channel information; determines the precoding matrix of one or more layers corresponding to the rank; determines the first information value of the precoding matrix of each layer.
  • Method 2 The terminal determines the precoding matrix and codeword corresponding to the rank according to the channel information and the rank of the channel.
  • One codeword corresponds to one or more layers; determines the second information value corresponding to the codeword.
  • the terminal can determine the precoding matrix and rank of the channel based on the channel information to be fed back.
  • Channel information can be measured or predicted by the terminal.
  • the precoding matrix can be the complete precoder of Singular Value Decomposition (SVD), or the precoder of SVD projected on some specified orthogonal The coefficient on the basis, or the precoder obtained by weighting the coefficient and the orthogonal basis.
  • the specific implementation of the terminal determining the precoding matrix of the channel can be found in the specific implementation in related technologies, and will not be described in detail here.
  • the rank of the channel can be a certain value or a value selected by the terminal.
  • the value selected by the terminal may be a value that is cycled during the terminal adaptation process. For example, the terminal may traverse multiple rank values and select an appropriate rank value as the rank of the channel.
  • the terminal may further determine the precoding matrices of one or more layers corresponding to the rank of the channel.
  • Each layer may correspond to a precoding matrix.
  • the terminal may determine the first information value of the precoding matrix of each layer based on the precoding matrix of each layer.
  • the first information value may include at least one of channel quality indicator (Channel Quality Indicator, CQI), eigenvalue, singular value, channel capacity, and peak number of discrete Fourier Transform (Discrete Fourier Transform, DFT) projection.
  • CQI Channel Quality Indicator
  • eigenvalue Eigenvalue
  • singular value singular value
  • channel capacity channel capacity
  • peak number of discrete Fourier Transform Discrete Fourier Transform
  • the terminal can determine the precoding matrix and codeword corresponding to the rank based on the channel information to be fed back and the rank of the channel.
  • the channel information can be measured or predicted by the terminal, and the rank of the channel can be a certain value or a value selected by the terminal.
  • the precoding matrix corresponding to rank can be the complete SVD precoder, or the coefficient of the SVD precoder projected on some specified orthogonal basis, or the precoder obtained by weighting the coefficient and the orthogonal basis.
  • the codeword corresponding to rank can be one or more codewords, and one codeword can correspond to one or more layers.
  • the terminal may further determine the second information value corresponding to the codeword based on the precoding matrix and codeword.
  • the second information value includes at least one of CQI, Modulation and Coding Scheme (Modulation and Coding Scheme, MCS) and Transport Block size (TB size), and one codeword can correspond to one CQI.
  • MCS Modulation and Coding Scheme
  • TB size Transport Block size
  • the terminal when determining the information value of the precoding matrix of each layer, can determine the first information value of the precoding matrix of each layer (different layers can correspond to different first information values), or The second information value of the precoding matrix of each layer can be determined (different layers can correspond to the same second information value), and the first information value and the second information value of the precoding matrix of each layer can also be determined. That is to say, the information value determined by the terminal may include at least one of the first information value and the second information value.
  • S204 For each layer, determine the target encoding model corresponding to the target interval according to the target interval in which the information value of the precoding matrix of the layer is located. Different information value intervals correspond to different encoding models.
  • the information value can be divided into intervals in advance to obtain multiple different information value intervals, and corresponding encoding models can be determined for different information value intervals.
  • the terminal can store encoding models corresponding to different information value intervals. In this way, the terminal After determining the information value of the precoding matrix of each layer, for each layer, the target interval in which the information value of the precoding matrix of the layer is located can be determined, and then determined from the existing coding model of the terminal based on the target interval. Obtain the target encoding model corresponding to the target interval.
  • the target coding model is used to encode the precoding matrix of the layer to obtain the corresponding precoding feature information.
  • different information value intervals and coding models corresponding to different information value intervals may include at least one of the following:
  • Different information value intervals correspond to different model output lengths.
  • the model output lengths corresponding to different information value intervals are the same, and the encoding models corresponding to different information value intervals are trained based on different training sets;
  • One information value interval corresponds to multiple coding models.
  • the model output lengths of multiple coding models are different, and multiple coding models are suitable for different application scenarios.
  • the model output length is the length (payload) of the encoding result output by the encoding model, which is also the model input length of the decoding model.
  • one information value interval may correspond to one or more coding models.
  • the model output lengths of coding models corresponding to different information value intervals can be the same or different.
  • the model output lengths are different, the greater the information value of the information value interval, the better the model. The longer the output length is, when the model output length is the same, coding models corresponding to different information value intervals can be trained based on different training sets to ensure that the model output length is the same.
  • the model output lengths of the multiple coding models may be different, and the multiple coding models may be used in different application scenarios.
  • the information value interval [0.7,1] corresponds to encoding model 1 and encoding model 2.
  • Encoding model 1 is used indoors, the model output length is 200
  • encoding model 2 is used outdoors, the model output length is 160
  • the information value interval [0.4 ,0.7] corresponds to coding model 3 and coding model 4.
  • Coding model 3 is used indoors, the model output length is 160
  • coding model 4 is used outdoors, the model output length is 100
  • the information value interval [0,0.4] corresponds to coding model 6 , the model output length is 100.
  • different information value intervals and coding models corresponding to different information value intervals may include any of the following:
  • the division and number of information value intervals are stipulated by the agreement, and the model structure and model parameters of the encoding model are stipulated by the agreement, that is, the division method of the information value interval and the specific structure and parameters of the encoding model are stipulated by the agreement;
  • the division and number of information value intervals are specified by the protocol.
  • the model structure of the encoding model is specified by the agreement.
  • the model parameters of the encoding model are trained by the network side device. That is, the protocol and the network side device jointly determine the division method of the information value interval and the encoding model.
  • the specific structure and parameters of the protocol stipulate the division method of the information value interval and the specific structure of the encoding model.
  • the network side device performs model training based on the sample data to obtain the model parameters;
  • the division and number of information value intervals are specified by the protocol, or indicated by the core network, or determined by the network side device.
  • the model structure of the encoding model is determined by the network side device.
  • the model parameters of the encoding model are trained by the network side device, that is, There are many ways to divide the information value interval, which can be specified by the protocol, instructed by the core network, or determined by the network side device.
  • the specific structure and specific parameters of the encoding model need to be determined by the network side device;
  • the division and number of information value intervals and the encoding model are determined by the terminal-assisted network-side device. That is, the terminal-assisted network-side device determines the division method of the information value interval and the specific structure and parameters of the encoding model.
  • the information value can be divided into N information value intervals.
  • the N information value intervals correspond to N or N groups of models.
  • N is an integer greater than or equal to 1, that is, each information value interval can correspond to one or N groups of models.
  • a set of models that can include multiple models suitable for different application scenarios.
  • the network side device can also train the encoding model and obtain the model parameters of the encoding model.
  • the model parameters of the encoding model, or the model parameters of the encoding model and the model parameters of the decoding model corresponding to the encoding model are sent to the terminal.
  • the terminal can receive the model parameters of the encoding model sent by the network side device, or the model parameters of the encoding model and the corresponding encoding model.
  • the model parameters of the decoding model so that the encoding model can be determined based on the model structure of the encoding model specified by the protocol and the model parameters of the encoding model sent by the network side device.
  • the target encoding can be selected from the encoding model
  • the model is encoded.
  • the network side device When the model structure of the encoding model is determined by the network side device and the model parameters of the encoding model are trained by the network side device, after the network side device obtains the encoding model through training, it can associate the encoding model or the encoding model with the encoding model.
  • the decoding model is sent to the terminal, and the terminal can receive the encoding model sent by the network side device, or the encoding model and the decoding model corresponding to the encoding model, so that when encoding the precoding matrix, the target encoding model can be selected from the encoding model for encoding. .
  • the network side device can, after determining the division, number and encoding model of the information value interval, divide the divided information value intervals and different The encoding model corresponding to the information value interval is indicated to the terminal.
  • the network side device may send first indication information to the terminal.
  • the first indication information is used to indicate the division, quantity, and encoding model of the information value interval.
  • the terminal may receive the first indication information sent by the network side device, and based on the first indication information Indicating information, the terminal can know the information value intervals divided by the network side device and the encoding models corresponding to different information value intervals.
  • the terminal can select a target encoding model from the encoding models for encoding.
  • the specific implementation may include:
  • Receive second instruction information the second instruction information is used to instruct to start the auxiliary function
  • For different information values traverse multiple candidate models in the candidate model set, and determine target candidate models corresponding to different information values from the multiple candidate models;
  • Different information values and the model identification of the target candidate model are sent to the network side device.
  • the auxiliary capabilities reported by the terminal can represent what assistance the terminal can provide and the capability information for these assistances.
  • the auxiliary capabilities reported by the terminal can be the computing assistance the terminal can provide and the amount of calculations it can assist.
  • the network side device may send second instruction information to the terminal if the terminal needs assistance.
  • the second instruction information is used to instruct the start of the auxiliary function.
  • the second indication information may be sent by the network side device through signaling, and the second indication information may include the end time or the number of durations after the terminal starts the auxiliary function, or the second indication information may be a week Sent periodically. That is to say, the network side device can instruct the terminal to perform the auxiliary function in a certain period of time, or instruct the terminal to perform the auxiliary function periodically.
  • the terminal can start the auxiliary function. After starting the auxiliary function, the terminal can traverse multiple candidate models in the candidate model set for different information values, and determine the target candidate model corresponding to the different information values from the multiple candidate models.
  • the candidate model set can be configured by the network side device or specified by the protocol. In the case where the candidate model set is configured by the network side device, the network side device can configure the candidate model set through the control channel, or configure the candidate model set through the broadcast channel, that is, The terminal may receive a set of candidate models configured by the network side device through a control channel configuration or a broadcast channel.
  • the network side device when the network side device configures the candidate model set through the control channel, the network side device may independently configure the candidate model set for each terminal, and the candidate model set configured for each terminal may be different.
  • the network side device configures the candidate model set through the broadcast channel.
  • the network side device When configuring the candidate model set, the network side device may uniformly configure the candidate model set for multiple terminals, and the candidate model set configured for each terminal is the same. In this embodiment, only one terminal that assists the network side device is used as an example for explanation.
  • the traversal process may be a process of sequentially using each candidate model to encode the precoding matrices corresponding to different information values and using the decoding model to decode.
  • each candidate model encodes the precoding matrix corresponding to the information value and uses the decoding model to decode.
  • the decoded precoding matrix is the same as the precoding matrix before encoding.
  • the difference in the encoding matrix is then selected from multiple candidate models to select the candidate model with the smallest difference as the target candidate model corresponding to the information value. In this way, for different information values, after traversing multiple candidate models in the candidate model set, target candidate models corresponding to different information values can be obtained.
  • the terminal After determining the target candidate models corresponding to different information values, the terminal can send the different information values and the model identifiers of the target candidate models corresponding to the different information values to the network side device, and the network side device can perform information value calculation based on this information. Division of intervals and determination of coding models corresponding to different information value intervals.
  • the network side device divides the information value interval and determines the corresponding encoding model, it can divide 0.3 and 0.4 into an interval, and the corresponding model of this interval is model 1, and divide 0.5, 0.6, and 0.7 into an interval, and the corresponding interval
  • the model is Model 3.
  • the network side device can also combine the auxiliary results of other terminals or other factors to jointly divide the information value interval and determine the corresponding encoding model. I will not give examples one by one here.
  • the network-side device can determine the division and number of information value intervals based on the different information values and target candidate models. After encoding the model, or after configuring the terminal to cancel reporting the information value of the precoding matrix and the model identifier of the candidate model, or when the end time indicated by the second indication information is reached, the division of the indication information value interval may be sent to the terminal. and quantity, the first indication information of the encoding model. Based on the first indication information, the terminal can know the information value intervals divided by the network side device and the encoding models corresponding to different information value intervals. When encoding the precoding matrix, the terminal can select a target encoding model from the encoding models for encoding.
  • the terminal After obtaining different information value intervals through the content division described above and determining the coding models corresponding to the different information value intervals, the terminal determines the information value of the precoding matrix for each layer when feeding back the channel information.
  • the target coding model corresponding to the target interval can be determined from the terminal's existing (that is, previously determined) coding model according to the target interval in which the information value of the precoding matrix of the layer is located.
  • the information value of the precoding matrix of each layer may include at least one of the first information value and the second information value. Therefore, for each layer, the precoding matrix according to the layer When determining the target interval where the information value of the coding matrix is located and the target encoding model corresponding to the target interval, at least one of the following first to third implementation methods may be included.
  • the first implementation method determine the target coding model corresponding to the target interval according to the target interval where the first information value of the precoding matrix of the layer is located.
  • the target interval corresponding to the target interval can be determined according to the target interval in which the first information value of the precoding matrix of the layer is located.
  • the target coding model can be used to encode the precoding matrix of a layer to obtain the corresponding precoding feature information, that is, the input of the target coding model is the precoding matrix of a single layer.
  • the target coding models corresponding to different layers are the same.
  • the first information values of the precoding matrices of different layers are in the same interval, If the information value is in different intervals, different layers correspond to different target encoding models.
  • the second implementation method determine the target coding model corresponding to the target interval according to the target interval where the second information value of the precoding matrix of the layer is located.
  • the target interval corresponding to the target interval can be determined according to the target interval in which the second information value of the precoding matrix of the layer is located.
  • the target coding model can be used to encode the precoding matrix of at least one layer to obtain the corresponding precoding feature information.
  • the at least one layer corresponds to the same codeword, that is, the input of the target coding model corresponds to the same The precoding matrix of at least one layer of the codeword.
  • the target coding models corresponding to different layers are the same. If they are different The codewords corresponding to the layers are different, and the intervals where the different codewords are located are different, so the target encoding models corresponding to different layers are different.
  • the third implementation method determine the coding model set corresponding to the first target interval according to the first target interval where the second information value of the precoding matrix of the layer is located; according to the first information value of the precoding matrix of the layer
  • the second target interval determines the target encoding model corresponding to the second target interval in the encoding model set.
  • different encoding model sets may be determined in advance for different intervals of the second information value, and each encoding model set includes multiple encoding models, and the multiple encoding models correspond to different intervals of the first information value.
  • the first target where the second information value of the precoding matrix of the layer is located can be first interval, determine the coding model set corresponding to the first target interval from multiple predetermined coding model sets, and then determine the coding model set corresponding to the first target interval according to the precoding matrix of the layer.
  • a target encoding model corresponding to the second target interval is determined from the encoding model set. That is to say, for each layer, the corresponding coding model set can first be determined based on the interval where the second information value of the codeword corresponding to the layer is located, and then the coding model set is determined based on the interval where the first information value of the precoding matrix of the layer is located. Determine the corresponding target encoding model in the set.
  • the finalized target coding model can be used to encode the precoding matrix of a layer to obtain the corresponding precoding feature information, that is, the input of the target coding model is the precoding matrix of a single layer.
  • the CQI of the codeword corresponding to layer1 can be determined.
  • the interval 1 in which it is located determines the coding model set 1 corresponding to interval 1 (including several alternative models, different alternative models correspond to different eigenvalue intervals), and then, based on the eigenvalues of the precoding matrix of layer 1, Interval 2, determine the coding model corresponding to interval 2 in the coding model set 1.
  • This coding model is the target coding model, and the target coding model is used to code the precoding matrix of layer 1.
  • the target encoding model when the information value only includes the first information value, for each layer, the target encoding model can be determined according to the above-mentioned first implementation manner.
  • the target encoding model can be determined according to the second implementation method mentioned above.
  • the target encoding model can be determined according to any of the above three implementation methods.
  • the target encoding model is determined in one or more ways, and there are no specific limitations here.
  • S206 Process the precoding matrix of the layer according to the target coding model to obtain precoding feature information.
  • the precoding matrix of the layer can be coded according to the target coding model. After the coding process, the corresponding precoding feature information can be obtained. .
  • the terminal determines the target coding model in the above-mentioned S204, for each layer: if the corresponding target coding model is determined according to the target interval where the first information value of the precoding matrix of the layer is located (i.e., the first information value in S204 ( That is, the second implementation method in S204), the precoding feature information obtained based on the target coding model processing is the precoding feature information corresponding to one or more layers, and the one or more layers correspond to the same codeword, for example, For multiple layers corresponding to the same codeword, the target coding model determined for each layer is the same.
  • the precoding matrices of the multiple layers need to be input into the target coding model at the same time. , thereby obtaining the precoding feature information corresponding to the multiple layers; if the corresponding target coding model is determined according to the first target interval where the second information value of the precoding matrix of the layer is located and the second target interval where the first information value is located ( That is, the third implementation method in S204), the precoding feature information obtained based on the target coding model processing is the precoding feature information corresponding to a layer.
  • the first information includes the model identifier and/or the target interval of the target coding model.
  • the terminal can send the precoding feature information to the network side device, thereby achieving the purpose of feeding back the channel information to the network side device.
  • the terminal can also use the target coding model Inform the network side device so that the network side device can determine the corresponding decoding model according to the target coding model used by the terminal and correctly decode the precoding feature information based on the decoding model.
  • the terminal may send first information to the network side device.
  • the first information may be a model identifier of the target encoding model and/or a target interval corresponding to the target encoding model.
  • the terminal when the terminal determines the target coding model, if the target coding model corresponding to the target interval is not found in the existing coding models, that is, the target coding model is not found in the existing coding models. If there is a target encoding model corresponding to the target interval, the terminal can perform any of the following:
  • the instructions of the network side device or the provisions of the protocol determine the encoding model corresponding to the first interval as the target encoding model
  • the existing coding model corresponding to the first interval can be used as the target coding model of the layer.
  • the first interval or the coding model corresponding to the first interval is indicated by the network side device or specified by the protocol, or the terminal can also
  • the existing coding model corresponding to the second interval is used as the target coding model, that is, you can choose an existing coding model as the target coding model of the layer.
  • the terminal when the terminal self-selects the encoding model corresponding to the second interval as the target encoding model, it also needs to inform the network side device of the self-selected encoding model, that is, the second interval and/or the encoding model corresponding to the second interval.
  • the model identifier is sent to the network side device, so that the network side device can select the corresponding decoding model to decode correctly when decoding the precoding feature information reported by the terminal.
  • the terminal when the terminal encodes the precoding matrix of the channel and feeds back the channel information to the network side device, the corresponding coding model can be selected for each layer according to the interval where the information value of the precoding matrix of each layer is located.
  • the precoding matrix of each layer is encoded, and different information value intervals can correspond to different coding models. Therefore, for each layer, a coding model that better matches the channel characteristics of the layer can be used to precode the layer's precoding matrix. Encoding can be performed to improve the coding accuracy and accurately feedback channel information; when the terminal feeds back the encoded precoding feature information to the network side device, it can also use the model identification and/or target of the target coding model used by the terminal.
  • the target interval corresponding to the encoding model is fed back to the network side device, so the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • different encoding models can be trained according to different intervals of the information value of the precoding matrix of the layer, there is no need to train the corresponding encoding model for each layer or multiple layers, so the complexity of model training can also be reduced.
  • this embodiment of the present application provides a channel information feedback method 300.
  • the channel information feedback method can be executed by a network side device.
  • the channel information feedback method can be implemented by software or hardware installed on the network side device. Execution, the channel information feedback method includes the following steps.
  • the network side device receives the precoding feature information and the first information.
  • S304 Determine the target decoding model according to the first information.
  • S306 Process the precoding feature information according to the target decoding model to obtain the corresponding precoding matrix.
  • the precoding feature information is obtained by processing the precoding matrix of the layer based on the target coding model.
  • the target coding model is the precoding matrix of the layer.
  • the encoding model corresponding to the target interval where the information value of the encoding matrix is located. Different information value intervals correspond to different encoding models.
  • the first information includes the model identifier and/or the target interval of the target encoding model.
  • the network side device can receive the precoding characteristic information and the first information.
  • the precoding characteristic information is encoded by the terminal using the target code.
  • the model is obtained after encoding the precoding matrices of one or more layers.
  • the target coding model is a coding model determined by the terminal for each layer according to the target interval where the information value of the precoding matrix of the layer is located.
  • the first information is used to indicate For the target encoding model used by the terminal, the first information may specifically be the model identifier of the target encoding model and/or the target interval.
  • the specific implementation method of the terminal feeding back the channel information to the network side device including determining the information value of the precoding matrix of each layer, determining the target coding model, processing the precoding matrix of the layer according to the target coding model to obtain the precoding feature information
  • the specific implementation method of the terminal feeding back the channel information to the network side device including determining the information value of the precoding matrix of each layer, determining the target coding model, processing the precoding matrix of the layer according to the target coding model to obtain the precoding feature information
  • the information value can be divided into intervals in advance to obtain multiple different information value intervals, and the corresponding encoding model and decoding model can be determined for different information value intervals.
  • the terminal can store the encoding corresponding to the different information value intervals.
  • Model the network side device can store decoding models corresponding to different encoding models. In this way, when the terminal uses the coding model to code the precoding matrix of each layer that needs feedback, it can determine the target coding model corresponding to each layer from the stored coding model and code the precoding matrix.
  • the network side After receiving the channel information (ie, precoding feature information) fed back by the terminal, the device can determine the corresponding decoding model from the stored decoding model and decode the precoding feature information.
  • different information value intervals and coding models corresponding to different information value intervals may include at least one of the following:
  • Information value intervals and encoding models include any of the following:
  • the division and number of information value intervals are stipulated by the agreement, and the model structure and model parameters of the encoding model are stipulated by the agreement, that is, the division method of the information value interval and the specific structure and parameters of the encoding model are stipulated by the agreement;
  • the division and number of information value intervals are specified by the protocol.
  • the model structure of the encoding model is specified by the agreement.
  • the model parameters of the encoding model are trained by the network side device. That is, the protocol and the network side device jointly determine the division method of the information value interval and the encoding model.
  • the specific structure and parameters of the protocol stipulate the division method of the information value interval and the specific structure of the encoding model.
  • the network side device performs model training based on the sample data to obtain the model parameters;
  • the division and number of information value intervals are specified by the protocol, or indicated by the core network, or determined by the network side device.
  • the model structure of the encoding model is determined by the network side device.
  • the model parameters of the encoding model are trained by the network side device, that is, There are many ways to divide the information value interval, which can be specified by the protocol, instructed by the core network, or determined by the network side device.
  • the specific structure and specific parameters of the encoding model need to be determined by the network side device;
  • the division, quantity, and encoding model of information value intervals are determined by the terminal-assisted network side device, that is, the terminal-assisted network
  • the side device determines the division method of the information value interval and the specific structure and parameters of the encoding model.
  • the information value can be divided into N information value intervals.
  • the N information value intervals correspond to N or N groups of models.
  • N is an integer greater than or equal to 1, that is, each information value interval can correspond to one or N groups of models.
  • a set of models that can include multiple models suitable for different application scenarios.
  • the network side device when the model structure of the encoding model is specified by the protocol and the model parameters of the encoding model are trained by the network side device, the network side device obtains the encoding model after training. After the model parameters, you can also send the model parameters of the encoding model, or the model parameters of the encoding model and the model parameters of the decoding model corresponding to the encoding model to the terminal, so that the terminal can determine the encoding model. When encoding the precoding matrix, you can Select the target encoding model from the encoding models for encoding.
  • the network side device When the model structure of the encoding model is determined by the network side device and the model parameters of the encoding model are trained by the network side device, after the network side device obtains the encoding model through training, it can associate the encoding model or the encoding model with the encoding model.
  • the decoding model is sent to the terminal, so that when the terminal encodes the precoding matrix, it can select the target encoding model from the encoding model sent by the network side device for encoding.
  • the network side device can, after determining the division, number and encoding model of the information value interval, divide the divided information value intervals and different
  • the coding model corresponding to the information value interval is indicated to the terminal through the first indication information, so that the terminal can know the information value interval divided by the network side device and the encoding model corresponding to the different information value interval.
  • it can be In the encoding model, select the target encoding model for encoding.
  • the specific implementation may include:
  • the target candidate model is selected by the terminal from multiple candidate models after traversing multiple candidate models in the candidate model set for different information values. Get it for sure;
  • the auxiliary capabilities reported by the terminal can represent what assistance the terminal can provide and the capability information for these assistances. When the terminal reports the auxiliary capabilities, it can be reported periodically.
  • the network side device may send second instruction information to the terminal if assistance from the terminal is required.
  • the second instruction information is used to instruct the terminal to start the auxiliary function.
  • the second indication information may be sent by the network side device through signaling, and the second indication information may include the end time or the number of durations after the terminal starts the auxiliary function, or the second indication information may be sent periodically. That is to say, the network side device can instruct the terminal to perform the auxiliary function in a certain period of time, or instruct the terminal to perform the auxiliary function periodically.
  • the terminal After receiving the second instruction information, the terminal can start an auxiliary function, that is, it can traverse multiple candidate models in the candidate model set for different information values, and determine target candidates corresponding to different information values from the multiple candidate models.
  • Model The candidate model set may be configured by the network side device or specified by the protocol. In the case where the candidate model set is configured by the network side device, the network side device may configure the candidate model set through the control channel, or configure the candidate model set through the broadcast channel. Wherein, when the network side device configures the candidate model set through the control channel, the network side device may independently configure the candidate model set for each terminal, and the candidate model set configured for each terminal may be different. The network side device configures the candidate model set through the broadcast channel.
  • the network side device may uniformly configure the candidate model set for multiple terminals, and the candidate model set configured for each terminal is the same.
  • the specific implementation method may refer to the specific implementation of the corresponding steps in the embodiment shown in Figure 2, and the description will not be repeated here.
  • the terminal After the terminal determines the target candidate models corresponding to different information values, the terminal can send the different information values and the model identifiers of the target candidate models corresponding to the different information values to the network side device, and the network side device can receive this information and use it according to these Information is divided into information value intervals and coding models corresponding to different information value intervals are determined. Afterwards, the network side device may send the first indication information to the terminal.
  • the network side device sends the first indication information to the terminal, which may include any of the following:
  • the first indication information is sent to the terminal.
  • first indication information indicating the division, number, and encoding model of the information value intervals may be sent to the terminal, so that the terminal can know the information value intervals divided by the network side device and the information values associated with different information values based on the first indication information. Coding model corresponding to the interval.
  • the target encoding model can be selected from the encoding model for encoding.
  • the network side device can use the same model structure specified by the protocol, and use the four feature value proportion intervals specified by the protocol [0,0.3], [0.3,0.5], [0.5,0.7] and [0.7,1 ] precoding matrices to train the corresponding model 1, model 2, model 3, and model 4 respectively, that is, the network side device divides the data into four groups according to the characteristic values of the collected data, and each group is trained independently, using
  • the weight parameters obtained by training and the public model structure form a model, and the weight parameters are sent to the terminal at the same time.
  • the terminal obtains a complete encoding model, or an encoding model and a decoding model, based on the known model structure and the received weight parameters.
  • the terminal When the terminal feeds back channel information, it can obtain the channel matrix by estimating CSI-RS. After SVD calculation, the eigenvalues of the four layers are 0.6, 0.35, 0.03, 0.02.
  • the network side device can use the corresponding decoding networks of 3 and 2 to precode the precoding reported by the terminal. Feature information is decoded to obtain channel information.
  • the ratio of the above eigenvalues is calculated based on the eigenvalues of the four layers. If the ratio is calculated only for the eigenvalues of the two layers that need to be reported, then the ratio of the two layers that need to be reported will be calculated.
  • the terminal when the terminal encodes the precoding matrix of the channel and feeds back the channel information to the network side device, the corresponding coding model can be selected for each layer according to the interval where the information value of the precoding matrix of each layer is located.
  • the precoding matrix of each layer is encoded, and different information value intervals can correspond to different coding models. Therefore, for each layer, a coding model that better matches the channel characteristics of the layer can be used to precode the layer's precoding matrix. Encoding can be performed to improve the coding accuracy and accurately feedback channel information; when the terminal feeds back the encoded precoding feature information to the network side device, it can also use the model identification and/or target of the target coding model used by the terminal.
  • the target interval corresponding to the encoding model is fed back to the network side device, so the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • different encoding models can be trained according to different intervals of the information value of the precoding matrix of the layer, there is no need to train the corresponding encoding model for each layer or multiple layers, so the complexity of model training can also be reduced.
  • the execution subject may be a channel information feedback device.
  • the channel information feedback device performed by the channel information feedback device is used as an example to illustrate the channel information feedback device provided by the embodiment of the present application.
  • FIG 4 is a schematic structural diagram of a channel information feedback device according to an embodiment of the present application. This device may correspond to a terminal in other embodiments. As shown in Figure 4, the device 400 includes the following modules.
  • the first determination module 401 is used to determine the information value of the precoding matrix of each layer
  • the second determination module 402 is configured to determine, for each layer, the target coding model corresponding to the target interval according to the target interval in which the information value of the precoding matrix of the layer is located. Different information value intervals correspond to different encodings. Model;
  • Coding module 403 configured to process the precoding matrix of the layer according to the target coding model to obtain precoding feature information
  • the sending module 404 is configured to send the precoding feature information and first information to the network side device, where the first information includes the model identifier of the target coding model and/or the target interval.
  • the first determination module 401 is also used for at least one of the following:
  • One codeword corresponds to one or more layers; determine the second information value corresponding to the codeword.
  • the rank is a determined value or a value selected by the terminal
  • the precoding matrix is the precoder of the complete singular value decomposition SVD, or the coefficient of the SVD precoder projected on the specified orthogonal basis, or the precoder obtained by weighting the coefficient and the orthogonal basis. ;
  • the first information value includes at least one of channel quality indicator CQI, eigenvalue, singular value, channel capacity, and peak number of discrete Fourier transform DFT projection;
  • the second information value includes at least one of CQI, modulation and coding scheme MCS, and transport block size TB size, and one codeword corresponds to one CQI.
  • the second determination module 402 is also used for at least one of the following:
  • a target coding model corresponding to the target interval is determined; wherein the precoding feature information obtained based on the target coding model processing is a layer Corresponding precoding feature information;
  • a target coding model corresponding to the target interval is determined; wherein the precoding feature information obtained based on the target coding model processing is at least one Precoding feature information corresponding to the layer, and the at least one layer corresponds to the same codeword;
  • the second target interval where the value is located determines the target coding model corresponding to the second target interval in the coding model set; wherein, the precoding feature information obtained based on the target coding model processing is the precoding corresponding to a layer. Feature information.
  • the information value interval and the encoding model include at least one of the following:
  • Different information value intervals correspond to different model output lengths.
  • the model output lengths corresponding to different information value intervals are the same, and the encoding models corresponding to different information value intervals are trained based on different training sets;
  • One information value interval corresponds to multiple coding models, the model output lengths of the multiple coding models are different, and the multiple coding models are suitable for different application scenarios.
  • the information value interval and the coding model include any of the following:
  • the division and number of the information value intervals are specified by the agreement, and the model structure and model parameters of the encoding model are specified by the agreement;
  • the division and number of the information value intervals are specified by the protocol, the model structure of the encoding model is specified by the protocol, and the model parameters of the encoding model are obtained by training with the network side device;
  • the division and number of the information value intervals are specified by the protocol, or indicated by the core network, or determined by the network side device.
  • the model structure of the encoding model is determined by the network side device.
  • the model parameters are trained by the network side device;
  • the division and number of the information value intervals and the encoding model are determined by the terminal assisting the network side device;
  • N information value intervals correspond to N or N groups of models, and N is an integer greater than or equal to 1.
  • the device 400 also includes a first receiving module 405 (not shown in Figure 4), The first receiving module 405 is used for:
  • the model structure of the encoding model is specified by the protocol and the model parameters of the encoding model are trained by the network side device, receive the model parameters of the encoding model, or the model parameters of the encoding model and the The model parameters of the decoding model corresponding to the encoding model;
  • the model structure of the encoding model is determined by the network side device and the model parameters of the encoding model are trained by the network side device, receive the encoding model, or the encoding model and the The decoding model corresponding to the encoding model;
  • the terminal assisting the network side device When the division and number of the information value intervals and the encoding model are determined by the terminal assisting the network side device, receive first indication information, where the first indication information is used to indicate the information value interval.
  • the divisions and quantities, the coding model are determined by the terminal assisting the network side device.
  • the device 400 also includes a second receiving module 406 (not shown in Figure 4);
  • the sending module 404 is also used to report auxiliary capabilities
  • the second receiving module 406 is used to receive second indication information, and the second indication information is used to indicate starting the auxiliary function;
  • the second determination module 402 is also configured to traverse multiple candidate models in the candidate model set for different information values, and determine target candidate models corresponding to different information values from the multiple candidate models;
  • the sending module 404 is also configured to send the different information values and the model identification of the target candidate model to the network side device.
  • the candidate model set is configured by the network side device or specified by a protocol
  • the second receiving module 406 is also used for any of the following:
  • the second indication information is sent through signaling, and the second indication information includes the end time or the number of durations after the terminal starts the auxiliary function, or the second indication information Sent periodically.
  • the second determination module 402 is also used for any of the following:
  • the encoding model corresponding to the second interval is selected as the target encoding model.
  • the sending module 404 is also used for at least one of the following:
  • the information value is a normalized value, and the specific method of normalization is specified by the protocol.
  • the device 400 according to the embodiment of the present application can refer to the process of the method 200 corresponding to the embodiment of the present application, and each unit/module and the above-mentioned other operations and/or functions in the device 400 are respectively to implement the corresponding process in the method 200, And can achieve the same or equivalent technical effects. For the sake of simplicity, they will not be described again here.
  • FIG. 5 is a schematic structural diagram of a channel information feedback device according to an embodiment of the present application. This device may correspond to network-side equipment in other embodiments. As shown in Figure 5, the device 500 includes the following modules.
  • Receiving module 501 configured to receive precoding feature information and first information
  • Determining module 502 configured to determine a target decoding model according to the first information
  • Decoding module 503 configured to process the precoding feature information according to the target decoding model to obtain the corresponding precoding matrix
  • the precoding feature information is obtained by processing the precoding matrix of the layer based on a target coding model.
  • the target coding model is a coding model corresponding to the target interval where the information value of the precoding matrix of the layer is located. Different The information value intervals correspond to different encoding models, and the first information includes the model identifier of the target encoding model and/or the target interval.
  • the information value interval and the coding model include any of the following:
  • the division and number of the information value intervals are specified by the agreement, and the model structure and model parameters of the encoding model are specified by the agreement;
  • the division and number of the information value intervals are specified by the protocol, the model structure of the encoding model is specified by the protocol, and the model parameters of the encoding model are obtained by training with the network side device;
  • the division and number of the information value intervals are specified by the protocol, or indicated by the core network, or determined by the network side device.
  • the model structure of the encoding model is determined by the network side device.
  • the model parameters are trained by the network side device;
  • the division and number of the information value intervals and the encoding model are determined by the terminal assisting the network side device;
  • N information value intervals correspond to N or N groups of models, and N is an integer greater than or equal to 1.
  • the device 500 also includes a sending module 504 (not shown in Figure 5), the sending module 504 is used to:
  • model parameters and model parameters of the decoding model corresponding to the encoding model In the case where the model structure of the encoding model is specified by the protocol and the model parameters of the encoding model are trained by the network side device, send the model parameters of the encoding model, or the model parameters of the encoding model to the terminal.
  • Model parameters and model parameters of the decoding model corresponding to the encoding model In the case where the model structure of the encoding model is specified by the protocol and the model parameters of the encoding model are trained by the network side device, send the model parameters of the encoding model, or the model parameters of the encoding model to the terminal.
  • the encoding model is sent to the terminal, or the encoding model and The decoding model corresponding to the encoding model;
  • first indication information is sent to the terminal, and the first indication information is used to indicate the information The division and number of value intervals, the encoding model.
  • the receiving module 501 is configured to receive the auxiliary capabilities reported by the terminal;
  • the sending module 504 is configured to send second indication information to the terminal, where the second indication information is used to indicate starting the auxiliary function;
  • the receiving module 501 is configured to receive different information values sent by the terminal and model identifiers of target candidate models corresponding to the different information values.
  • the target candidate models are generated by the terminal for the different information.
  • the value is determined from the multiple candidate models after traversing multiple candidate models in the candidate model set;
  • the sending module 504 is configured to send the first indication information to the terminal.
  • the candidate model set is configured by the network side device or specified by a protocol
  • the sending module 504 is also used for any of the following:
  • the set of candidate models is configured through a broadcast channel.
  • the second indication information is sent through signaling, and the second indication information includes the end time or the number of durations after the terminal starts the auxiliary function, or the second indication information Sent periodically.
  • the sending module 504 is also used for any of the following:
  • the first indication information is sent to the terminal.
  • the device 500 according to the embodiment of the present application can refer to the process corresponding to the method 300 of the embodiment of the present application, and each unit/module and the above-mentioned other operations and/or functions in the device 500 are respectively intended to implement the corresponding process in the method 300. And can achieve the same or equivalent technical effects. For the sake of simplicity, they will not be described again here.
  • the channel information feedback device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip.
  • the electronic device may be a terminal or other devices other than the terminal.
  • terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.
  • NAS Network Attached Storage
  • the channel information feedback device provided by the embodiments of the present application can implement each process implemented by the method embodiments in Figures 2 to 3, and achieve the same technical effect. To avoid duplication, details will not be described here.
  • this embodiment of the present application also provides a communication device 600, which includes a processor 601 and a memory 602.
  • the memory 602 stores programs or instructions that can be run on the processor 601, for example.
  • the communication device 600 is a terminal, when the program or instruction is executed by the processor 601, each step of the above channel information feedback method embodiment is implemented, and the same technical effect can be achieved.
  • the communication device 600 is a network-side device, when the program or instruction is executed by the processor 601, each step of the above channel information feedback method embodiment is implemented, and the same technical effect can be achieved. To avoid duplication, the details are not repeated here.
  • Embodiments of the present application also provide a terminal, including a processor and a communication interface.
  • the processor is used to determine the information value of the precoding matrix of each layer; for each layer, according to the location of the information value of the precoding matrix of the layer target interval, determine the target coding model corresponding to the target interval, and different information value intervals correspond to different coding models; process the precoding matrix of the layer according to the target coding model to obtain precoding feature information, and communicate
  • the interface is used to send the precoding feature information and first information to a network side device, where the first information includes a model identifier of the target coding model and/or the target interval.
  • This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect.
  • FIG. 7 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
  • the terminal 700 includes but is not limited to: a radio frequency unit 701, a network module 702, an audio output unit 703, an input unit 704, a sensor 705, a display unit 706, a user input unit 707, an interface unit 708, a memory 709, a processor 710, etc. At least some parts.
  • the terminal 700 may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the processor 710 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions.
  • the terminal structure shown in FIG. 7 does not constitute a limitation on the terminal.
  • the terminal may include more or fewer components than shown in the figure, or some components may be combined or arranged differently, which will not be described again here.
  • the input unit 704 may include a graphics processor (Graphics Processing Unit, GPU) 7041 and a microphone 7042.
  • the graphics processor 7041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
  • the display unit 706 may include a display panel 7061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 707 includes a touch panel 7071 and at least one of other input devices 7072 .
  • Touch panel 7071 also called touch screen.
  • the touch panel 7071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 7072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
  • the radio frequency unit 701 after receiving downlink data from the network side device, can transmit it to the processor 710 for processing; in addition, the radio frequency unit 701 can send uplink data to the network side device.
  • the radio frequency unit 701 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
  • Memory 709 may be used to store software programs or instructions as well as various data.
  • the memory 709 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc.
  • memory 709 may include volatile memory or non-volatile memory, or memory 709 may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (Synch link DRAM, SLDRAM) and Direct Rambus RAM (DRRAM).
  • RAM Random Access Memory
  • Static RAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM, SLDRAM synchronous connection dynamic random access memory
  • DRRAM Direct Rambus RAM
  • the processor 710 may include one or more processing units; optionally, the processor 710 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above-mentioned modem processor may not be integrated into the processor 710.
  • the processor 710 is used to determine the information value of the precoding matrix of each layer; for each layer, determine the target interval corresponding to the target interval according to the target interval where the information value of the precoding matrix of the layer is located.
  • Target coding model different information value intervals correspond to different coding models; process the precoding matrix of the layer according to the target coding model to obtain precoding feature information;
  • the radio frequency unit 701 is configured to send the precoding feature information and first information to the network side device, where the first information includes the model identifier of the target coding model and/or the target interval.
  • the terminal when the terminal encodes the precoding matrix of the channel and feeds back the channel information to the network side device, the corresponding coding model can be selected for each layer according to the interval where the information value of the precoding matrix of each layer is located.
  • the precoding matrix of each layer is encoded, and different information value intervals can correspond to different coding models. Therefore, for each layer, a coding model that better matches the channel characteristics of the layer can be used to precode the layer's precoding matrix. Encoding can be performed to improve the coding accuracy and accurately feedback channel information; when the terminal feeds back the encoded precoding feature information to the network side device, it can also use the model identification and/or target of the target coding model used by the terminal.
  • the target interval corresponding to the encoding model is fed back to the network side device, so the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • the network side device can easily determine the corresponding decoding model, and then accurately decode the precoding feature information.
  • different encoding models can be trained according to different intervals of the information value of the precoding matrix of the layer, there is no need to train the corresponding encoding model for each layer or multiple layers, so the complexity of model training can also be reduced.
  • processor 710 is also used for at least one of the following:
  • the terminal determines the precoding matrix of the channel and the rank of the channel according to the channel information; determines the precoding matrix of one or more layers corresponding to the rank; determines the first information value of the precoding matrix of each layer;
  • the terminal determines the precoding matrix and codeword corresponding to the rank according to the channel information and the rank of the channel.
  • One codeword corresponds to one or more layers; and determines the second information value corresponding to the codeword.
  • processor 710 is also used for at least one of the following:
  • a target coding model corresponding to the target interval is determined; wherein the precoding feature information obtained based on the target coding model processing is a layer Corresponding precoding feature information;
  • the target interval where the second information value of the precoding matrix of the layer is located determine the target interval The corresponding target coding model; wherein the precoding feature information obtained by processing based on the target coding model is precoding feature information corresponding to at least one layer, and the at least one layer corresponds to the same codeword;
  • the second target interval where the value is located determines the target coding model corresponding to the second target interval in the coding model set; wherein, the precoding feature information obtained based on the target coding model processing is the precoding corresponding to a layer. Feature information.
  • radio frequency unit 701 is also used for:
  • the model structure of the encoding model is specified by the protocol and the model parameters of the encoding model are trained by the network side device, receive the model parameters of the encoding model, or the model parameters of the encoding model and the The model parameters of the decoding model corresponding to the encoding model;
  • the model structure of the encoding model is determined by the network side device and the model parameters of the encoding model are trained by the network side device, receive the encoding model, or the encoding model and the The decoding model corresponding to the encoding model;
  • the terminal assisting the network side device When the division and number of the information value intervals and the encoding model are determined by the terminal assisting the network side device, receive first indication information, where the first indication information is used to indicate the information value interval.
  • the divisions and quantities, the coding model are determined by the terminal assisting the network side device.
  • the radio frequency unit 701 is also used to report auxiliary capabilities; receive second indication information, the second indication information is used to indicate starting the auxiliary function;
  • the processor 710 is also configured to traverse multiple candidate models in the candidate model set for different information values, and determine target candidate models corresponding to different information values from the multiple candidate models;
  • the radio frequency unit 701 is also configured to send the different information values and the model identification of the target candidate model to the network side device.
  • radio frequency unit 701 is also used for any of the following:
  • processor 710 is also used for any of the following:
  • the encoding model corresponding to the second interval is selected as the target encoding model.
  • processor 710 is also used for at least one of the following:
  • the terminal 700 provided by the embodiment of the present application can also implement each process of the above channel information feedback method embodiment, and can achieve the same technical effect. To avoid duplication, the details will not be described here.
  • An embodiment of the present application also provides a network side device, including a processor and a communication interface.
  • the communication interface is used to receive precoding characteristic information and first information.
  • the processor is used to determine a target decoding model according to the first information; according to the The target decoding model processes the precoding feature information to obtain the corresponding precoding matrix; wherein the precoding feature information is obtained by processing the precoding matrix of the layer based on the target coding model, and the target coding model is the The coding model corresponding to the target interval where the information value of the precoding matrix of the layer is located. Different information value intervals correspond to different coding models.
  • the first information includes the model identifier of the target coding model and/or the target interval. .
  • This network-side device embodiment corresponds to the above-mentioned network-side device method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this network-side device embodiment, and can achieve the same technical effect.
  • the embodiment of the present application also provides a network side device.
  • the network side device 800 includes: an antenna 81 , a radio frequency device 82 , a baseband device 83 , a processor 84 and a memory 85 .
  • the antenna 81 is connected to the radio frequency device 82 .
  • the radio frequency device 82 receives information through the antenna 81 and sends the received information to the baseband device 83 for processing.
  • the baseband device 83 processes the information to be sent and sends it to the radio frequency device 82.
  • the radio frequency device 82 processes the received information and then sends it out through the antenna 81.
  • the method performed by the network side device in the above embodiment can be implemented in the baseband device 83, which includes a baseband processor.
  • the baseband device 83 may include, for example, at least one baseband board on which multiple chips are disposed, as shown in FIG. Program to perform the network device operations shown in the above method embodiments.
  • the network side device may also include a network interface 86, which is, for example, a common public radio interface (CPRI).
  • a network interface 86 which is, for example, a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the network side device 800 in this embodiment of the present invention also includes: instructions or programs stored in the memory 85 and executable on the processor 84.
  • the processor 84 calls the instructions or programs in the memory 85 to execute the various operations shown in Figure 6. The method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.
  • Embodiments of the present application also provide a readable storage medium.
  • Programs or instructions are stored on the readable storage medium.
  • the program or instructions are executed by a processor, each process of the above channel information feedback method embodiment is implemented, and can achieve The same technical effects are not repeated here to avoid repetition.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
  • An embodiment of the present application further provides a chip.
  • the chip includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the above embodiment of the channel information feedback method. Each process can achieve the same technical effect. To avoid repetition, we will not go into details here.
  • chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
  • the embodiment of the present application further provides a computer program/program product, the computer program/program product is stored in In the storage medium, the computer program/program product is executed by at least one processor to implement each process of the above channel information feedback method embodiment, and can achieve the same technical effect. To avoid repetition, the details will not be described here.
  • Embodiments of the present application also provide a channel information feedback system, including: a terminal and a network side device.
  • the terminal can be used to perform the steps of the channel information feedback method as described above.
  • the network side device can be used to perform the steps of the channel information feedback method as described above. The steps of the channel information feedback method.
  • the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
  • the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

Landscapes

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

Abstract

本申请公开了一种信道信息反馈方法、终端及网络侧设备，属于通信技术领域，本申请实施例的信道信息反馈方法包括：终端确定每个层layer的预编码矩阵的信息值；针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。

Description

信道信息反馈方法、终端及网络侧设备
交叉引用
本发明要求在2022年04月28日提交中国专利局、申请号为202210461819.2、发明名称为“信道信息反馈方法、终端及网络侧设备”的中国专利申请的优先权，该申请的全部内容通过引用结合在本发明中。
技术领域
本申请属于通信技术领域，具体涉及一种信道信息反馈方法、终端及网络侧设备。
背景技术
终端在与网络侧设备通信时，通常需要向网络侧设备反馈信道信息，以便网络侧设备可以基于信道信息进行预编码和用户调度等。一般地，终端在反馈信道信息时，可以将信道的预编码矩阵反馈给网络侧设备，同时为了减少资源开销，终端还可以对预编码矩阵进行压缩，并将压缩后的预编码信息反馈给网络侧设备。
目前，为了更好地对预编码矩阵进行压缩，可以采用网络模型压缩的方法，比如，可以采用人工智能(Artificial Intelligence，AI)模型对预编码矩阵进行压缩。然而，在实际的应用场景中，在信道的预编码矩阵对应多个层(layer)，且多个layer之间的信道特征差异较为明显的情况下，在对信道的预编码矩阵进行压缩时，需要考虑多个layer之间的信道特征的差异性，进而准确地向网络侧设备反馈信道信息，而在相关技术中，还缺少一种有效地方法能够实现这个目的。
发明内容
本申请实施例提供一种信道信息反馈方法、终端及网络侧设备，能够解决在信道的预编码矩阵对应多个layer且多个layer之间的信道特征差异明显的情况下，终端无法基于模型对信道的预编码矩阵进行有效压缩进而准确反馈信道信息的问题。
第一方面，提供了一种信道信息反馈方法，该方法包括：
终端确定每个层layer的预编码矩阵的信息值；
针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；
根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；
将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
第二方面，提供了一种信道信息反馈装置，该装置包括：
第一确定模块，用于确定每个层layer的预编码矩阵的信息值；
第二确定模块，用于针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；
编码模块，用于根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；
发送模块，用于将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
第三方面，提供了一种信道信息反馈方法，该方法包括：
网络侧设备接收预编码特征信息以及第一信息；
根据所述第一信息确定目标解码模型；
根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；
其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
第四方面，提供了一种信道信息反馈装置，该装置包括：
接收模块，用于接收预编码特征信息以及第一信息；
确定模块，用于根据所述第一信息确定目标解码模型；
解码模块，用于根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；
其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
第五方面，提供了一种终端，该终端包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面，提供了一种终端，包括处理器及通信接口，其中，所述处理器用于确定每个层layer的预编码矩阵的信息值；针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息，所述通信接口用于将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
第七方面，提供了一种网络侧设备，该网络侧设备包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第三方面所述的方法的步骤。
第八方面，提供了一种网络侧设备，包括处理器及通信接口，其中，所述通信接口用于接收预编码特征信息以及第一信息；所述处理器用于根据所述第一信息确定目标解码模型；根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
第九方面，提供了一种信道信息反馈系统，包括：终端及网络侧设备，所述终端可用于执行如第一方面所述的信道信息反馈方法的步骤，所述网络侧设备可用于执行如第三方面所述的信道信息反馈方法的步骤。
第十方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤，或者实现如第三方面所述的方法的步骤。
第十一方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如第一方面所述的方法，或实现如第三方面所述的方法。
第十二方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面所述的方法的步骤，或者实现如第三方面所述的方法的步骤。
在本申请实施例中，终端在对信道的预编码矩阵进行编码并向网络侧设备反馈信道信息时，由于可以根据每个layer的预编码矩阵的信息值所在的区间选择对应的编码模型对每个layer的预编码矩阵进行编码，且不同的信息值区间可以对应不同的编码模型，因此针对每个layer而言，可以使用与该layer的信道特征更加匹配的编码模型对该layer的预编码矩阵进行编码，从而可以提高编码准确度，进而准确反馈信道信息；由于终端在将编码后的预编码特征信息反馈给网络侧设备时，还可以将终端使用的目标编码模型的模型标识和/或目标编码模型对应的目标区间反馈给网络侧设备，因此可以便于网络侧设备确定相应的解码模型，进而对预编码特征信息进行准确解码。此外，由于可以根据layer的预编码矩阵的信息值的不同区间训练不同的编码模型，无需针对每个layer或多个layer训练对应的编码模型，因此还可以降低模型训练的复杂度。
附图说明
图1是根据本申请实施例的无线通信系统的示意图；
图2是根据本申请实施例的信道信息反馈方法的示意性流程图；
图3是根据本申请实施例的信道信息反馈方法的示意性流程图；
图4是根据本申请实施例的信道信息反馈装置的结构示意图；
图5是根据本申请实施例的信道信息反馈装置的结构示意图；
图6是根据本申请实施例的通信设备的结构示意图；
图7是根据本申请实施例的终端的结构示意图；
图8是根据本申请实施例的网络侧设备的结构示意图。
具体实施方式
在相关技术中，终端在向网络侧设备反馈信道信息时，可以采用模型对信道的预编码矩阵进行压缩得到预编码特征信息，然后将预编码特征信息反馈给网络侧设备。网络侧设备在接收到终端反馈的预编码特征信息后，可以采用与编码模型相应的解码模型对预编码特征信息进行解码，进而得到信道信息。
针对信道的预编码矩阵对应多个layer的场景，终端在向网络侧设备反馈信道信息时，需要反馈每个layer的信道信息，即终端需要采用模型对每个layer的预编码矩阵进行编码，并将编码得到预编码特征信息反馈给网络侧设备。然而，在多个layer之间的信道特征差异较为明显的场景下，比如，在多rank信道中，多个layer之间的信道差异较大，在采用模型对多个layer的预编码矩阵进行编码时，模型的训练难度较大，导致无法采用模型对每个layer的预编码矩阵进行有效编码。
比如，针对单个layer而言，在模型训练时，多个采样数据集中每个采样数据的最大特征值可能不相同，即训练集中的数据表征信道信息的比重不同，而特征值跨度大会导致信道规律变化大，因此就需要更大参数规模的模型进行处理，导致模型训练难度较大，此外，模型规模大还会带来模型性能下降的问题。
再比如，对于多layer而言，在每个layer独立采用一个模型进行编码的情况下，如果每个layer采用的模型不同，那么会导致模型开销大，如果每个layer采用的模型相同，那么由于每个layer之间的特征差异较大，对应的学习权重也存在差异，因此，会存在这个唯一的模型难以训练的问题，且模型的性能也不好。在多个layer采用同一个模型进行编码的情况下，由于不同layer之间的特征差异较大，分配的学习权重也不同，因此也会存在模型训练难度较大的问题。
由此可见，在多rank信道的场景下，相关技术中终端还无法基于模型对信道的预编码矩阵进行有效压缩。进一步地，终端在无法采用模型对信道的预编码矩阵进行有效编码的情况下，也无法准确地向网络侧设备反馈信道信息。
为了解决上述技术问题，本申请实施例提供一种信道信息反馈方法、终端及网络侧设备，该信道信息反馈方法包括：终端确定每个层layer的预编码矩阵的信息值；针对每个layer，根据layer的预编码矩阵的信息值所在的目标区间，确定与目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；根据目标编码模型对layer的预编码矩阵进行处理得到预编码特征信息；将预编码特征信息以及第一信息发送给网络侧设备，第一信息包括 目标编码模型的模型标识和/或目标区间。
在本申请实施例中，终端在对信道的预编码矩阵进行编码并向网络侧设备反馈信道信息时，由于可以根据每个layer的预编码矩阵的信息值所在的区间选择对应的编码模型对每个layer的预编码矩阵进行编码，且不同的信息值区间可以对应不同的编码模型，因此针对每个layer而言，可以使用与该layer的信道特征更加匹配的编码模型对该layer的预编码矩阵进行编码，从而可以提高编码准确度，进而准确反馈信道信息；由于终端在将编码后的预编码特征信息反馈给网络侧设备时，还可以将终端使用的目标编码模型的模型标识和/或目标编码模型对应的目标区间反馈给网络侧设备，因此可以便于网络侧设备确定相应的解码模型，进而对预编码特征信息进行准确解码。此外，由于可以根据layer的预编码矩阵的信息值的不同区间训练不同的编码模型，无需针对每个layer或多个layer训练对应的编码模型，因此还可以降低模型训练的复杂度。
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请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，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是，本申请实施例所描述的技术不限于长期演进型(Long Term Evolution，LTE)/LTE的演进(LTE-Advanced，LTE-A)系统，还可用于其他无线通信系统，诸如码分多址(Code Division Multiple Access，CDMA)、时分多址(Time Division Multiple Access，TDMA)、频分多址(Frequency Division Multiple Access，FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access，OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access，SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用，所描述的技术既可用于以上提及的系统和无线电技术，也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio，NR)系统，并且在以下大部分描述中使用NR术语，但是这些技术也可应用于NR系统应用以外的应用，如第6代(6th Generation，6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。其中，终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant，PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer，UMPC)、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实 (virtual reality，VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备，如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer，PC)、柜员机或者自助机等终端侧设备，可穿戴式设备包括：智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是，在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备，其中，接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、无线局域网(Wireless Local Area Network，WLAN)接入点或无线保真(Wireless Fidelity，WiFi)节点等，基站可被称为节点B、演进节点B(Evolved NodeB，eNB)、接入点、基收发机站(Base Transceiver Station，BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set，BSS)、扩展服务集(Extended Service Set，ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point，TRP)或所述领域中其他某个合适的术语，只要达到相同的技术效果，所述基站不限于特定技术词汇，需要说明的是，在本申请实施例中仅以NR系统中的基站为例进行介绍，并不限定基站的具体类型。
下面结合附图，通过一些实施例及其应用场景对本申请实施例提供的信道信息反馈方法、终端及网络侧设备进行详细地说明。
如图2所示，本申请实施例提供一种信道信息反馈方法200，该信道信息反馈方法可以由终端执行，换言之，该信道信息反馈方法可以由安装在终端的软件或硬件来执行，该信道信息反馈方法包括如下步骤。
S202：终端确定每个层layer的预编码矩阵的信息值。
在S202中，终端在向网络侧设备反馈信道信息时，可以确定信道的每个layer的预编码矩阵的信息值。其中，对于每个layer而言，layer的预编码矩阵的信息值可以描述layer的预编码特征。
可选地，作为一个实施例，上述信息值可以是归一化后的值，归一化的具体方式可以由协议规定。
可选地，作为一个实施例，终端确定每个层layer的预编码矩阵的信息值，可以包括以下方式1和方式2中的至少一种方式。
方式1：终端根据信道信息确定信道的预编码矩阵和信道的秩rank；确定与rank对应的一个或多个layer的预编码矩阵；确定每个layer的预编码矩阵的第一信息值。
方式2：终端根据信道信息和信道的rank确定与rank对应的预编码矩阵和码字，一个码字对应一个或多个layer；确定与码字对应的第二信息值。
在方式1中，终端可以根据待反馈的信道信息确定信道的预编码矩阵和信道的rank。信道信息可以由终端测量或预测得到。预编码矩阵可以是完整的奇异值分解(Singular Value Decomposition，SVD)的预编码器(precoder)，或SVD的precoder投影在指定的某些正交 基上的系数，或根据该系数和正交基加权得到的precoder。其中，终端确定信道的预编码矩阵的具体实现可以参见相关技术中的具体实现，这里不再详细说明。信道的rank可以是某个确定的值，也可以是终端自行选择的值。该终端自行选择的值可以是终端自适应的过程中循环的值，比如，终端可以对多个rank值进行遍历，选择其中一个合适的rank值作为信道的rank。
终端在确定信道的预编码矩阵和信道的rank后，可以进一步确定与信道的rank对应的一个或多个layer的预编码矩阵。其中，每个layer可以对应一个预编码矩阵，终端确定与信道的rank对应的每个layer的预编码矩阵的具体实现可以参见相关技术中的具体实现，这里不再详细说明。
终端在确定每个layer的预编码矩阵后，可以根据每个layer的预编码矩阵确定每个layer的预编码矩阵的第一信息值。其中，第一信息值可以包括信道质量指示(Channel Quality Indicator，CQI)、特征值、奇异值、信道容量以及离散傅里叶变换(Discrete Fourier Transform，DFT)投影的峰值数中的至少一项。
在方式2中，终端可以根据待反馈的信道信息和信道的rank确定与该rank对应的预编码矩阵和码字。其中，信道信息可以由终端测量或预测得到，信道的rank可以是某个确定值或终端自行选择的值。与rank对应的预编码矩阵可以是完整的SVD的precoder，或SVD的precoder投影在指定的某些正交基上的系数，或根据该系数和正交基加权得到的precoder。与rank对应的码字可以是一个或多个码字，一个码字可以对应一个或多个layer。终端根据信道信息和信道的rank确定与该rank对应的预编码矩阵和码字的具体实现可以参见相关技术中的具体实现，这里不再详细说明。
终端在确定与rank对应的预编码矩阵和码字后，可以根据预编码矩阵和码字进一步确定与码字对应的第二信息值。其中，第二信息值包括CQI、调制与编码方案(Modulation and Coding Scheme，MCS)以及传输块大小(Transport Block size，TB size)中的至少一项，一个码字可以对应一个CQI。
在本实施例中，终端在确定每个layer的预编码矩阵的信息值时，可以确定每个layer的预编码矩阵的第一信息值(不同的layer可以对应不同的第一信息值)，也可以确定每个layer的预编码矩阵的第二信息值(不同的layer可以对应同一个第二信息值)，还可以确定每个layer的预编码矩阵的第一信息值以及第二信息值。也就是说，终端确定的信息值可以包括第一信息值和第二信息值中的至少一项。
S204：针对每个layer，根据layer的预编码矩阵的信息值所在的目标区间，确定与目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型。
在S204中，可以预先针对信息值进行区间划分得到多个不同的信息值区间，并针对不同的信息值区间确定对应的编码模型，终端可以存储与不同信息值区间对应的编码模型，这样，终端在确定每个layer的预编码矩阵的信息值后，针对每个layer，可以确定layer的预编码矩阵的信息值所在的目标区间，然后根据该目标区间从终端已有的编码模型中确定 出与目标区间对应的目标编码模型。该目标编码模型用于对layer的预编码矩阵进行编码得到对应的预编码特征信息。
可选地，作为一个实施例，不同的信息值区间以及与不同信息值区间对应的编码模型可以包括以下至少一项：
不同信息值区间对应的模型输出长度不同，信息值区间的信息值越大，对应的模型输出长度越大；
不同信息值区间对应的模型输出长度相同，不同信息值区间对应的编码模型基于不同的训练集训练得到；
一个信息值区间对应多个编码模型，多个编码模型的模型输出长度不同，且多个编码模型适用于不同的应用场景。
模型输出长度即为编码模型输出的编码结果的长度(payload)，也即解码模型的模型输入长度。本实施例中，一个信息值区间可以对应一个或多个编码模型。在一个信息值区间对应一个编码模型的情况下，不同信息值区间对应的编码模型的模型输出长度可以相同也可以不同，在模型输出长度不同的情况下，信息值区间的信息值越大，模型输出长度越长，在模型输出长度相同的情况下，不同信息值区间对应的编码模型可以基于不同的训练集训练得到，以保证模型输出长度相同。在一个信息值区间对应多个编码模型的情况下，该多个编码模型的模型输出长度可以不同，且该多个编码模型可以用于不同的应用场景。比如，信息值区间[0.7,1]对应编码模型1和编码模型2，编码模型1用于室内，模型输出长度为200，编码模型2用于室外，模型输出长度为160，信息值区间[0.4,0.7]对应编码模型3和编码模型4，编码模型3用于室内，模型输出长度为160，编码模型4用于室外，模型输出长度为100，信息值区间[0,0.4]对应编码模型6，模型输出长度为100。
可选地，作为一个实施例，不同的信息值区间以及与不同信息值区间对应的编码模型可以包括以下任一项：
信息值区间的划分和数量由协议规定，编码模型的模型结构和模型参数由协议规定，即信息值区间的划分方式和编码模型的具体结构和参数都由协议规定；
信息值区间的划分和数量由协议规定，编码模型的模型结构由协议规定，编码模型的模型参数由网络侧设备训练得到，即由协议和网络侧设备共同决定信息值区间的划分方式和编码模型的具体结构和参数，其中，协议规定信息值区间的划分方式和编码模型的具体结构，由网络侧设备根据样本数据进行模型训练得到模型参数；
信息值区间的划分和数量由协议规定，或由核心网指示，或由网络侧设备自行确定，编码模型的模型结构由网络侧设备自行确定，编码模型的模型参数由网络侧设备训练得到，即信息值区间的划分方式有多种，具体可以由协议规定、或由核心网指示、或由网络侧设备自行确定，而编码模型的具体结构和具体参数需要由网络侧设备确定；
信息值区间的划分和数量、编码模型由终端辅助网络侧设备确定，即由终端辅助网络侧设备确定信息值区间的划分方式和编码模型的具体结构和参数。
在本实施例中，可以将信息值划分为N个信息值区间，N个信息值区间对应N个或N组模型，N为大于或等于1的整数，即每个信息值区间可以对应一个或一组模型，该一组模型中可以包括多个适用于不同应用场景的模型。
可选地，作为一个实施例，在编码模型的模型结构由协议规定、编码模型的模型参数由网络侧设备训练得到的情况下，网络侧设备在训练得到编码模型的模型参数后，还可以将编码模型的模型参数，或编码模型的模型参数以及编码模型对应的解码模型的模型参数发送给终端，终端可以接收网络侧设备发送的编码模型的模型参数，或编码模型的模型参数以及编码模型对应的解码模型的模型参数，从而可以基于协议规定的编码模型的模型结构以及网络侧设备发送的编码模型的模型参数确定编码模型，在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
在编码模型的模型结构由网络侧设备自行确定，编码模型的模型参数由网络侧设备训练得到的情况下，网络侧设备在训练得到编码模型后，可以将编码模型，或编码模型以及编码模型对应的解码模型发送给终端，终端可以接收网络侧设备发送的编码模型，或编码模型以及编码模型对应的解码模型，从而在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
在信息值区间的划分和数量、编码模型由终端辅助网络侧设备确定的情况下，网络侧设备在确定信息值区间的划分和数量，以及编码模型后，可以将划分的信息值区间以及与不同信息值区间对应的编码模型指示给终端。具体地，网络侧设备可以向终端发送第一指示信息，第一指示信息用于指示信息值区间的划分和数量、编码模型，终端可以接收网络侧设备发送的第一指示信息，基于该第一指示信息，终端可以知晓网络侧设备划分的信息值区间以及与不同信息值区间对应的编码模型，在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
可选地，作为一个实施例，在信息值区间的划分和数量、编码模型由终端辅助网络侧设备确定的情况下，具体实现方式可以包括：
终端上报辅助能力；
接收第二指示信息，第二指示信息用于指示开始辅助功能；
针对不同的信息值，遍历候选模型集合中的多个候选模型，并从多个候选模型中确定与不同的信息值对应的目标候选模型；
将不同的信息值以及目标候选模型的模型标识发送给网络侧设备。
终端上报的辅助能力可以表征终端能够提供哪些辅助以及针对这些辅助的能力信息，比如，终端上报的辅助能力可以是终端能够提供计算辅助以及能够辅助的计算量等。其中，考虑到实际的应用场景，终端上报辅助能力时，可以是周期性的上报。终端在上报辅助能力后，网络侧设备在需要终端辅助的情况下，可以向终端发送第二指示信息，第二指示信息用于指示开始辅助功能。其中，第二指示信息可以由网络侧设备通过信令发送，第二指示信息中可以包括终端开始辅助功能后的结束时间或持续次数，或者，第二指示信息为周 期性发送。也就是说，网络侧设备可以指示终端在某个时间段进行辅助功能，或指示终端周期性地进行辅助功能。
终端在接收到第二指示信息时，可以开始辅助功能。终端在开始辅助功能后，可以针对不同的信息值，遍历候选模型集合中的多个候选模型，并从多个候选模型中确定与不同的信息值对应的目标候选模型。候选模型集合由可以网络侧设备配置或由协议规定，在由网络侧设备配置候选模型集合的情况下，网络侧设备可以通过控制信道配置候选模型集合，或通过广播信道配置的候选模型集合，即终端可以通过控制信道配置或广播信道接收网络侧设备配置的候选模型集合。其中，在网络侧设备通过控制信道配置候选模型集合的情况下，可以是网络侧设备针对每个终端独立配置候选模型集合，每个终端配置的候选模型集合可以不同，在网络侧设备通过广播信道配置候选模型集合的情况下，可以是网络侧设备针对多个终端统一配置候选模型集合，每个终端配置的候选模型集合相同。本实施例中仅以辅助网络侧设备的其中一个终端为例进行说明。
终端在遍历候选模型集合中的多个候选模型时，遍历的过程可以是依次使用每个候选模型对不同信息值对应的预编码矩阵进行编码并使用解码模型进行解码的过程。在遍历多个候选模型后，针对每个信息值，可以确定每个候选模型对该信息值对应的预编码矩阵进行编码并使用解码模型进行解码后，解码得到的预编码矩阵与编码前的预编码矩阵的差异，然后从多个候选模型中选择差异最小的候选模型作为与该信息值对应的目标候选模型。这样，针对不同的信息值，在遍历候选模型集合中的多个候选模型后，可以得到与不同信息值对应的目标候选模型。
终端在确定与不同的信息值对应的目标候选模型后，可以将不同的信息值以及与不同信息值对应的目标候选模型的模型标识发送给网络侧设备，网络侧设备可以根据这些信息进行信息值区间的划分并确定与不同信息值区间对应的编码模型。
比如，假设终端发送给网络侧设备的信息值包括0.3、0.4、0.5、0.6和0.7，这些信息值对应的目标候选模型分别为模型1、模型1、模型3、模型3、模型3，那么，网络侧设备在划分信息值区间和确定对应的编码模型时，可以将0.3和0.4划分为一个区间，该区间对应的模型为模型1，将0.5、0.6和0.7划分为一个区间，该区间对应的模型为模型3。当然，在实际的应用场景中，网络侧设备还可以结合其他终端的辅助结果或其他因素共同划分信息值区间和确定对应的编码模型，这里不再一一举例说明。
终端在将不同的信息值以及与不同信息值对应的目标候选模型发送给网络侧设备后，针对网络侧设备而言，可以在根据不同的信息值以及目标候选模型确定信息值区间的划分和数量、编码模型后，或在配置终端取消上报预编码矩阵的信息值和候选模型的模型标识后，或在到达第二指示信息指示的结束时间时，可以向终端发送用于指示信息值区间的划分和数量、编码模型的第一指示信息。终端基于该第一指示信息可以知晓网络侧设备划分的信息值区间以及与不同信息值区间对应的编码模型，在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
在通过上述记载的内容划分得到不同的信息值区间以及确定与不同信息值区间对应的编码模型后，终端在反馈信道信息时，在确定每个layer的预编码矩阵的信息值的情况下，针对每个layer，可以根据layer的预编码矩阵的信息值所在的目标区间，从终端已有的(即之前确定的)编码模型中确定与目标区间对应的目标编码模型。
可选地，作为一个实施例，考虑到每个layer的预编码矩阵的信息值可以包括第一信息值和第二信息值中的至少一项，因此，针对每个layer，在根据layer的预编码矩阵的信息值所在的目标区间，确定与目标区间对应的目标编码模型时，可以包括以下第一实现方式至第三实现方式中的至少一种实现方式。
第一种实现方式：根据layer的预编码矩阵的第一信息值所在的目标区间，确定与目标区间对应的目标编码模型。
具体地，在每个layer的预编码矩阵的信息值包括第一信息值的情况下，针对每个layer，可以根据layer的预编码矩阵的第一信息值所在的目标区间，确定与目标区间对应的目标编码模型，该目标编码模型可以用于对一个layer的预编码矩阵进行编码得到对应的预编码特征信息，即目标编码模型的输入是单个layer的预编码矩阵。其中，针对不同的layer而言，若不同layer的预编码矩阵的第一信息值在同一个区间内，则不同的layer对应的目标编码模型相同，反之，若不同layer的预编码矩阵的第一信息值在不同的区间内，则不同的layer对应的目标编码模型不同。
第二种实现方式：根据layer的预编码矩阵的第二信息值所在的目标区间，确定与目标区间对应的目标编码模型。
具体地，在每个layer的预编码矩阵的信息值包括第二信息值的情况下，针对每个layer，可以根据layer的预编码矩阵的第二信息值所在的目标区间，确定与目标区间对应的目标编码模型，该目标编码模型可以用于对至少一个layer的预编码矩阵进行编码得到对应的预编码特征信息，该至少一个layer对应同一个码字，即目标编码模型的输入是对应同一个码字的至少一个layer的预编码矩阵。其中，针对不同的layer而言，若不同layer对应的码字相同，或，不同layer对应的码字不同但该不同码字位于同一个区间，则不同的layer对应的目标编码模型相同，若不同layer对应的码字不同，且不同码字所在的区间不同，则不同的layer对应的目标编码模型不同。
第三种实现方式：根据layer的预编码矩阵的第二信息值所在的第一目标区间，确定与第一目标区间对应的编码模型集合；根据layer的预编码矩阵的第一信息值所在的第二目标区间，确定编码模型集合中与第二目标区间对应的目标编码模型。
具体地，可以预先针对第二信息值的不同区间确定不同的编码模型集合，每个编码模型集合中包括多个编码模型，该多个编码模型与第一信息值的不同区间对应。这样，在每个layer的预编码矩阵的信息值包括第一信息值和第二信息值的情况下，针对每个layer，可以先根据layer的预编码矩阵的第二信息值所在的第一目标区间，从预先确定的多个编码模型集合中确定与第一目标区间对应的编码模型集合，之后，再根据layer的预编码矩阵的第 一信息值所在的第二目标区间，从编码模型集合中确定出与第二目标区间对应的目标编码模型。也就是说，针对每个layer，可以先根据layer对应的码字的第二信息值所在的区间确定对应的编码模型集合，再根据layer的预编码矩阵的第一信息值所在的区间从编码模型集合中确定对应的目标编码模型。其中，针对每个layer而言，最终确定的目标编码模型可以用于对一个layer的预编码矩阵进行编码得到对应的预编码特征信息，即目标编码模型的输入是单个layer的预编码矩阵。
比如，以第二信息值为码字对应的CQI、第一信息值为layer的预编码矩阵的特征值为例，在对layer1确定目标编码模型时，首先，可以根据layer1对应的码字的CQI所在的区间1确定与区间1对应的编码模型集合1(包含若干个备选的模型，不同的备选模型对应不同的特征值区间)，然后，再根据layer1的预编码矩阵的特征值所在的区间2，确定编码模型集合1中与区间2对应的编码模型，该编码模型即为目标编码模型，该目标编码模型用于对layer1的预编码矩阵进行编码。
在本实施例中，在信息值仅包括第一信息值的情况下，针对每个layer，可以根据上述第一种实现方式确定目标编码模型，在信息值仅包括第二信息值的情况下，针对每个layer，可以根据上述第二种实现方式确定目标编码模型，信息值包括第一信息值和第二信息值的情况下，针对每个layer，可以根据上述三种实现方式中的任一种或多种方式确定目标编码模型，这里不做具体限定。
S206：根据目标编码模型对layer的预编码矩阵进行处理得到预编码特征信息。
在S206中，在确定与每个layer对应的目标编码模型后，针对每个layer，可以根据目标编码模型对layer的预编码矩阵进行编码处理，在编码处理后，可以得到对应的预编码特征信息。
需要说明的是，终端在上述S204中确定目标编码模型时，针对每个layer：若根据layer的预编码矩阵的第一信息值所在的目标区间确定对应的目标编码模型(即S204中的第一种实现方式)，则基于目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息；若根据layer的预编码矩阵的第二信息值所在的目标区间确定对应的目标编码模型(即S204中的第二种实现方式)，则基于目标编码模型处理得到的预编码特征信息为一个或多个layer对应的预编码特征信息，该一个或多个layer对应同一个码字，比如，对于对应同一个码字的多个layer，针对每个layer确定得到的目标编码模型相同，在基于该目标预编码矩阵进行编码时，需要将该多个layer的预编码矩阵同时输入目标编码模型中，从而得到该多个layer对应的预编码特征信息；若根据layer的预编码矩阵的第二信息值所在的第一目标区间以及第一信息值所在的第二目标区间确定对应的目标编码模型(即S204中的第三种实现方式)，则基于目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息。
S208：将预编码特征信息以及第一信息发送给网络侧设备，第一信息包括目标编码模型的模型标识和/或目标区间。
在S208中，终端在得到每个layer的预编码特征信息后，可以将预编码特征信息发送给网络侧设备，从而实现向网络侧设备反馈信道信息的目的，同时还可以将使用的目标编码模型告知网络侧设备，以便网络侧设备可以根据终端使用的目标编码模型确定对应的解码模型并基于解码模型对预编码特征信息进行正确解码。其中，终端在将使用的目标编码模型告知网络侧设备时，可以向网络侧设备发送第一信息，该第一信息可以是目标编码模型的模型标识和/或该目标编码模型对应的目标区间。
可选地，作为一个实施例，在上述S204中，终端在确定目标编码模型时，若在已有的编码模型中没有找到与目标区间对应的目标编码模型，即在已有的编码模型中不存在与目标区间对应的目标编码模型，则，终端可以执行以下任一项：
取消上报layer的预编码矩阵；
根据网络侧设备的指示或协议的规定，将第一区间对应的编码模型确定为目标编码模型；
自行选择第二区间对应的编码模型为目标编码模型。
具体地，针对某个layer，若在已有的编码模型中不存在与该layer的预编码矩阵的信息值所在的目标区间对应的目标编码模型，则可以取消上报该layer的预编码矩阵，或者，可以将已有的与第一区间对应的编码模型作为layer的目标编码模型，该第一区间或与第一区间对应的编码模型由网络侧设备指示或由协议规定，或者，终端还可以将已有的与第二区间对应的编码模型作为目标编码模型，即可以自行选择已有的某个编码模型作为layer的目标编码模型。其中，终端在自行选择与第二区间对应的编码模型作为目标编码模型的情况下，还需要将自行选择的编码模型告知网络侧设备，即将第二区间和/或该第二区间对应的编码模型的模型标识发送给网络侧设备，以便网络侧设备在对终端上报的预编码特征信息进行解码时，可以选择对应的解码模型正确解码。
在本申请实施例中，终端在对信道的预编码矩阵进行编码并向网络侧设备反馈信道信息时，由于可以根据每个layer的预编码矩阵的信息值所在的区间选择对应的编码模型对每个layer的预编码矩阵进行编码，且不同的信息值区间可以对应不同的编码模型，因此针对每个layer而言，可以使用与该layer的信道特征更加匹配的编码模型对该layer的预编码矩阵进行编码，从而可以提高编码准确度，进而准确反馈信道信息；由于终端在将编码后的预编码特征信息反馈给网络侧设备时，还可以将终端使用的目标编码模型的模型标识和/或目标编码模型对应的目标区间反馈给网络侧设备，因此可以便于网络侧设备确定相应的解码模型，进而对预编码特征信息进行准确解码。此外，由于可以根据layer的预编码矩阵的信息值的不同区间训练不同的编码模型，无需针对每个layer或多个layer训练对应的编码模型，因此还可以降低模型训练的复杂度。
如图3所示，本申请实施例提供一种信道信息反馈方法300，该信道信息反馈方法可以由网络侧设备执行，换言之，该信道信息反馈方法可以由安装在网络侧设备的软件或硬件来执行，该信道信息反馈方法包括如下步骤。
S302：网络侧设备接收预编码特征信息以及第一信息。
S304：根据第一信息确定目标解码模型。
S306：根据所目标解码模型对预编码特征信息进行处理，得到对应的预编码矩阵，其中，预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，目标编码模型为layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，第一信息包括目标编码模型的模型标识和/或目标区间。
本申请实施例中，终端在基于图2所示的实施例向网络侧设备反馈信道信息后，网络侧设备可以接收到预编码特征信息和第一信息，该预编码特征信息由终端使用目标编码模型对一个或多个layer的预编码矩阵进行编码后得到，目标编码模型是终端针对每个layer，根据layer的预编码矩阵的信息值所在的目标区间确定的编码模型，第一信息用于指示终端使用的目标编码模型，第一信息具体可以是目标编码模型的模型标识和/或所述目标区间。其中，终端向网络侧设备反馈信道信息的具体实现方式(包括确定每个layer的预编码矩阵的信息值、确定目标编码模型、根据目标编码模型对layer的预编码矩阵进行处理得到预编码特征信息的具体实现方式)可以参见图2所示的实施例，这里不再重复说明。
本申请实施例中，可以预先针对信息值进行区间划分得到多个不同的信息值区间，并针对不同的信息值区间确定对应的编码模型和解码模型，终端可以存储与不同信息值区间对应的编码模型，网络侧设备可以存储与不同的编码模型对应的解码模型。这样，终端在使用编码模型对需要反馈的每个layer的预编码矩阵进行编码时，可以从已存储的编码模型中确定与每个layer对应的目标编码模型并对预编码矩阵进行编码，网络侧设备在接收到终端反馈的信道信息(即预编码特征信息)后，可以从已存储的解码模型中确定对应的解码模型并对预编码特征信息进行解码。
可选地，作为一个实施例，不同的信息值区间以及与不同信息值区间对应的编码模型可以包括以下至少一项：
信息值区间和编码模型包括以下任一项：
信息值区间的划分和数量由协议规定，编码模型的模型结构和模型参数由协议规定，即信息值区间的划分方式和编码模型的具体结构和参数都由协议规定；
信息值区间的划分和数量由协议规定，编码模型的模型结构由协议规定，编码模型的模型参数由网络侧设备训练得到，即由协议和网络侧设备共同决定信息值区间的划分方式和编码模型的具体结构和参数，其中，协议规定信息值区间的划分方式和编码模型的具体结构，由网络侧设备根据样本数据进行模型训练得到模型参数；
信息值区间的划分和数量由协议规定，或由核心网指示，或由网络侧设备自行确定，编码模型的模型结构由网络侧设备自行确定，编码模型的模型参数由网络侧设备训练得到，即信息值区间的划分方式有多种，具体可以由协议规定、或由核心网指示、或由网络侧设备自行确定，而编码模型的具体结构和具体参数需要由网络侧设备确定；
信息值区间的划分和数量、编码模型由终端辅助网络侧设备确定，即由终端辅助网络 侧设备确定信息值区间的划分方式和编码模型的具体结构和参数。
在本实施例中，可以将信息值划分为N个信息值区间，N个信息值区间对应N个或N组模型，N为大于或等于1的整数，即每个信息值区间可以对应一个或一组模型，该一组模型中可以包括多个适用于不同应用场景的模型。
针对网络侧设备而言，可选地，作为一个实施例，在编码模型的模型结构由协议规定、编码模型的模型参数由网络侧设备训练得到的情况下，网络侧设备在训练得到编码模型的模型参数后，还可以将编码模型的模型参数，或编码模型的模型参数以及编码模型对应的解码模型的模型参数发送给终端，以便终端可以确定编码模型，在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
在编码模型的模型结构由网络侧设备自行确定，编码模型的模型参数由网络侧设备训练得到的情况下，网络侧设备在训练得到编码模型后，可以将编码模型，或编码模型以及编码模型对应的解码模型发送给终端，以便终端在对预编码矩阵进行编码时，可以从网络侧设备发送的编码模型中选择目标编码模型进行编码。
在信息值区间的划分和数量、编码模型由终端辅助网络侧设备确定的情况下，网络侧设备在确定信息值区间的划分和数量，以及编码模型后，可以将划分的信息值区间以及与不同信息值区间对应的编码模型通过第一指示信息指示给终端，以便终端可以知晓网络侧设备划分的信息值区间以及与不同信息值区间对应的编码模型，在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
可选地，作为一个实施例，在信息值区间的划分和数量、编码模型由终端辅助网络侧设备确定的情况下，具体实现方式可以包括：
接收终端上报的辅助能力；
向终端发送第二指示信息，第二指示信息用于指示开始辅助功能；
接收终端发送的不同的信息值以及与不同的信息值对应的目标候选模型的模型标识，目标候选模型由终端针对不同的信息值，遍历候选模型集合中的多个候选模型后从多个候选模型中确定得到；
向终端发送第一指示信息。
终端上报的辅助能力可以表征终端能够提供哪些辅助以及针对这些辅助的能力信息，终端上报辅助能力时，可以是周期性的上报。网络侧设备在接收到终端上报的辅助能力后，在需要终端辅助的情况下，可以向终端发送第二指示信息，第二指示信息用于指示终端开始辅助功能。其中，第二指示信息可以由网络侧设备通过信令发送，第二指示信息中可以包括终端开始辅助功能后的结束时间或持续次数，或者，第二指示信息为周期性发送。也就是说，网络侧设备可以指示终端在某个时间段进行辅助功能，或指示终端周期性地进行辅助功能。
终端在接收到第二指示信息后，可以开始辅助功能即可以针对不同的信息值，遍历候选模型集合中的多个候选模型，并从多个候选模型中确定与不同的信息值对应的目标候选 模型。候选模型集合由可以网络侧设备配置或由协议规定，在由网络侧设备配置候选模型集合的情况下，网络侧设备可以通过控制信道配置候选模型集合，或通过广播信道配置的候选模型集合。其中，在网络侧设备通过控制信道配置候选模型集合的情况下，可以是网络侧设备针对每个终端独立配置候选模型集合，每个终端配置的候选模型集合可以不同，在网络侧设备通过广播信道配置候选模型集合的情况下，可以是网络侧设备针对多个终端统一配置候选模型集合，每个终端配置的候选模型集合相同。终端在从候选模型集合中确定目标候选模型时，具体实现方式可以参见图2所示实施例中相应步骤的具体实现，这里不再重复说明。
终端在确定与不同的信息值对应的目标候选模型后，可以将不同的信息值以及与不同信息值对应的目标候选模型的模型标识发送给网络侧设备，网络侧设备可以接收这些信息并根据这些信息进行信息值区间的划分以及确定与不同信息值区间对应的编码模型。之后，网络侧设备可以向终端发送第一指示信息。
可选地，作为一个实施例，网络侧设备向终端发送第一指示信息，可以包括以下任一项：
在根据不同的信息值以及目标候选模型确定信息值区间的划分和数量、编码模型后，向终端发送第一指示信息；
在配置终端取消上报预编码矩阵的信息值和候选模型的模型标识后，向终端发送所述第一指示信息；
在到达第二指示信息指示的结束时间时，向终端发送第一指示信息。
也就是说，网络侧设备在确定信息值区间的划分和数量、编码模型后，或在配置终端取消上报预编码矩阵的信息值和候选模型的模型标识后，或在到达第二指示信息指示的结束时间时，可以向终端发送用于指示信息值区间的划分和数量、编码模型的第一指示信息，以便终端可以基于该第一指示信息知晓网络侧设备划分的信息值区间以及与不同信息值区间对应的编码模型，在对预编码矩阵进行编码时，可以从编码模型中选择目标编码模型进行编码。
为了便于理解本申请实施例提供的技术方案，以下可以以一种更为具体的实现方式为例进行说明。
网络侧设备在模型训练时，可以使用协议规定的同一个模型结构，对协议规定的4个特征值比例区间[0,0.3]、[0.3,0.5]、[0.5,0.7]和[0.7,1]的预编码矩阵分别训练对应的模型1、模型2、模型3、模型4，即网络侧设备根据收集到的数据的特征值，按照特征值将数据分为四组，每组独立训练，使用训练得到的权重参数和公共的模型结构组成模型，同时将权重参数发送给终端，终端根据已知的模型结构和接收到的权重参数获取完整的编码模型，或编码模型和解码模型。
终端在反馈信道信息时，可以通过估计CSI-RS获取信道矩阵，经过SVD计算得到4个layer的特征值为0.6,0.35,0.03,0.02，终端根据rank自适应选择rank＝2，即上报特征值 为0.6和0.35，使用对应网络模型3和2对这两个layer分别进行编码处理并上报编码得到的预编码特征信息，网络侧设备可以使用对应的3和2的解码网络对终端上报的预编码特征信息进行解码，从而可以得到信道信息。
需要说明的是，上述特征值的比例是针对4个layer的特征值进行比例计算后得到的，如果是仅针对需要上报的两个layer的特征值进行比例计算，则需要上报的两个layer的特征值可以分别是0.6/0.95＝0.632和0.35/0.95＝0.368。
在本申请实施例中，终端在对信道的预编码矩阵进行编码并向网络侧设备反馈信道信息时，由于可以根据每个layer的预编码矩阵的信息值所在的区间选择对应的编码模型对每个layer的预编码矩阵进行编码，且不同的信息值区间可以对应不同的编码模型，因此针对每个layer而言，可以使用与该layer的信道特征更加匹配的编码模型对该layer的预编码矩阵进行编码，从而可以提高编码准确度，进而准确反馈信道信息；由于终端在将编码后的预编码特征信息反馈给网络侧设备时，还可以将终端使用的目标编码模型的模型标识和/或目标编码模型对应的目标区间反馈给网络侧设备，因此可以便于网络侧设备确定相应的解码模型，进而对预编码特征信息进行准确解码。此外，由于可以根据layer的预编码矩阵的信息值的不同区间训练不同的编码模型，无需针对每个layer或多个layer训练对应的编码模型，因此还可以降低模型训练的复杂度。
本申请实施例提供的信道信息反馈方法，执行主体可以为信道信息反馈装置。本申请实施例中以信道信息反馈装置执行信道信息反馈方法为例，说明本申请实施例提供的信道信息反馈装置。
图4是根据本申请实施例的信道信息反馈装置的结构示意图，该装置可以对应于其他实施例中的终端。如图4所示，装置400包括如下模块。
第一确定模块401，用于确定每个层layer的预编码矩阵的信息值；
第二确定模块402，用于针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；
编码模块403，用于根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；
发送模块404，用于将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
可选的，作为一个实施例，所述第一确定模块401，还用于以下至少一项：
根据信道信息确定信道的预编码矩阵和信道的秩rank；确定与所述rank对应的一个或多个layer的预编码矩阵；确定每个layer的预编码矩阵的第一信息值；
根据信道信息和信道的rank确定与所述rank对应的预编码矩阵和码字，一个码字对应一个或多个layer；确定与所述码字对应的第二信息值。
可选的，作为一个实施例，所述rank为确定值或为所述终端自行选择的值；
所述预编码矩阵为完整的奇异值分解SVD的预编码器precoder，或所述SVD的precoder投影在指定的正交基上的系数，或根据所述系数和所述正交基加权得到的precoder；
所述第一信息值包括信道质量指示CQI、特征值、奇异值、信道容量以及离散傅里叶变换DFT投影的峰值数中的至少一项；
所述第二信息值包括CQI、调制与编码方案MCS以及传输块大小TB size中的至少一项，一个码字对应一个CQI。
可选的，作为一个实施例，所述第二确定模块402，还用于以下至少一项：
根据所述layer的预编码矩阵的所述第一信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息；
根据所述layer的预编码矩阵的所述第二信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为至少一个layer对应的预编码特征信息，所述至少一个layer对应同一个码字；
根据所述layer的预编码矩阵的所述第二信息值所在的第一目标区间，确定与所述第一目标区间对应的编码模型集合；根据所述layer的预编码矩阵的所述第一信息值所在的第二目标区间，确定所述编码模型集合中与所述第二目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息。
可选的，作为一个实施例，所述信息值区间和所述编码模型包括以下至少一项：
不同信息值区间对应的模型输出长度不同，信息值区间的信息值越大，对应的模型输出长度越大；
不同信息值区间对应的模型输出长度相同，不同信息值区间对应的编码模型基于不同的训练集训练得到；
一个信息值区间对应多个编码模型，所述多个编码模型的模型输出长度不同，且所述多个编码模型适用于不同的应用场景。
可选的，作为一个实施例，所述信息值区间和所述编码模型包括以下任一项：
所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构和模型参数由协议规定；
所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到；
所述信息值区间的划分和数量由协议规定，或由核心网指示，或由所述网络侧设备自行确定，所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到；
所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定；
其中，N个信息值区间对应N个或N组模型，N为大于或等于1的整数。
可选的，作为一个实施例，所述装置400还包括第一接收模块405(图4并未示出)， 所述第一接收模块405，用于：
在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，接收所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，接收所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
在所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定的情况下，接收第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
可选的，作为一个实施例，所述装置400还包括第二接收模块406(图4并未示出)；
所述发送模块404，还用于上报辅助能力；
所述第二接收模块406，用于接收第二指示信息，所述第二指示信息用于指示开始辅助功能；
所述第二确定模块402，还用于针对不同的信息值，遍历候选模型集合中的多个候选模型，并从所述多个候选模型中确定与不同的信息值对应的目标候选模型；
所述发送模块404，还用于将所述不同的信息值以及所述目标候选模型的模型标识发送给所述网络侧设备。
可选的，作为一个实施例，所述候选模型集合由所述网络侧设备配置或由协议规定；
其中，所述第二接收模块406，还用于以下任一项：
接收所述网络侧设备通过控制信道配置的所述候选模型集合；
接收所述网络侧设备通过广播信道配置的所述候选模型集合。
可选的，作为一个实施例，所述第二指示信息通过信令发送，所述第二指示信息中包括所述终端开始辅助功能后的结束时间或持续次数，或者，所述第二指示信息为周期性发送。
可选的，作为一个实施例，所述第二确定模块402，还用于以下任一项：
取消上报所述layer的预编码矩阵；
根据所述网络侧设备的指示或协议的规定，将第一区间对应的编码模型确定为所述目标编码模型；
自行选择第二区间对应的编码模型为所述目标编码模型。
可选的，作为一个实施例，所述发送模块404，还用于以下至少一项：
将所述第二区间发送给所述网络侧设备；
将所述第二区间对应的编码模型的模型标识发送给所述网络侧设备。
可选的，作为一个实施例，所述信息值是归一化后的值，所述归一化的具体方式由协议规定。
根据本申请实施例的装置400可以参照对应本申请实施例的方法200的流程，并且，该装置400中的各个单元/模块和上述其他操作和/或功能分别为了实现方法200中的相应流程，并且能够达到相同或等同的技术效果，为了简洁，在此不再赘述。
图5是根据本申请实施例的信道信息反馈装置的结构示意图，该装置可以对应于其他实施例中的网络侧设备。如图5所示，装置500包括如下模块。
接收模块501，用于接收预编码特征信息以及第一信息；
确定模块502，用于根据所述第一信息确定目标解码模型；
解码模块503，用于根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；
其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
可选的，作为一个实施例，所述信息值区间和所述编码模型包括以下任一项：
所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构和模型参数由协议规定；
所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到；
所述信息值区间的划分和数量由协议规定，或由核心网指示，或由所述网络侧设备自行确定，所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到；
所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定；
其中，N个信息值区间对应N个或N组模型，N为大于或等于1的整数。
可选的，作为一个实施例，所述装置500还包括发送模块504(图5并未示出)，所述发送模块504，用于：
在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，向所述终端发送所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，向终端发送所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
在所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定的情况下，向所述终端发送第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
可选的，作为一个实施例，所述接收模块501，用于接收所述终端上报的辅助能力；
所述发送模块504，用于向所述终端发送第二指示信息，所述第二指示信息用于指示开始辅助功能；
所述接收模块501，用于接收所述终端发送的不同的信息值以及与所述不同的信息值对应的目标候选模型的模型标识，所述目标候选模型由所述终端针对所述不同的信息值，遍历候选模型集合中的多个候选模型后从所述多个候选模型中确定得到；
所述发送模块504，用于向所述终端发送所述第一指示信息。
可选的，作为一个实施例，所述候选模型集合由所述网络侧设备配置或由协议规定；
其中，所述发送模块504，还用于以下任一项：
通过控制信道配置所述候选模型集合；
通过广播信道配置所述候选模型集合。
可选的，作为一个实施例，所述第二指示信息通过信令发送，所述第二指示信息中包括所述终端开始辅助功能后的结束时间或持续次数，或者，所述第二指示信息为周期性发送。
可选的，作为一个实施例，所述发送模块504，还用于以下任一项：
在根据所述不同的信息值以及所述目标候选模型确定所述信息值区间的划分和数量、所述编码模型后，向所述终端发送所述第一指示信息；
在配置所述终端取消上报预编码矩阵的信息值和候选模型的模型标识后，向所述终端发送所述第一指示信息；
在到达所述第二指示信息指示的结束时间时，向所述终端发送所述第一指示信息。
根据本申请实施例的装置500可以参照对应本申请实施例的方法300的流程，并且，该装置500中的各个单元/模块和上述其他操作和/或功能分别为了实现方法300中的相应流程，并且能够达到相同或等同的技术效果，为了简洁，在此不再赘述。
本申请实施例中的信道信息反馈装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于上述所列举的终端11的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
本申请实施例提供的信道信息反馈装置能够实现图2至图3的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
可选的，如图6所示，本申请实施例还提供一种通信设备600，包括处理器601和存储器602，存储器602上存储有可在所述处理器601上运行的程序或指令，例如，该通信设备600为终端时，该程序或指令被处理器601执行时实现上述信道信息反馈方法实施例的各个步骤，且能达到相同的技术效果。该通信设备600为网络侧设备时，该程序或指令被处理器601执行时实现上述信道信息反馈方法实施例的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种终端，包括处理器和通信接口，处理器用于确定每个层layer的预编码矩阵的信息值；针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息，通信接口用于将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。该终端实施例与上述终端侧方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中，且能达到相同的技术效果。具体地，图7为实现本申请实施例的一种终端的硬件结构示意图。
该终端700包括但不限于：射频单元701、网络模块702、音频输出单元703、输入单元704、传感器705、显示单元706、用户输入单元707、接口单元708、存储器709以及处理器710等中的至少部分部件。
本领域技术人员可以理解，终端700还可以包括给各个部件供电的电源(比如电池)，电源可以通过电源管理系统与处理器710逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图7中示出的终端结构并不构成对终端的限定，终端可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置，在此不再赘述。
应理解的是，本申请实施例中，输入单元704可以包括图形处理器(Graphics Processing Unit，GPU)7041和麦克风7042，图形处理器7041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元706可包括显示面板7061，可以采用液晶显示器、有机发光二极管等形式来配置显示面板7061。用户输入单元707包括触控面板7071以及其他输入设备7072中的至少一种。触控面板7071，也称为触摸屏。触控面板7071可包括触摸检测装置和触摸控制器两个部分。其他输入设备7072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆，在此不再赘述。
本申请实施例中，射频单元701接收来自网络侧设备的下行数据后，可以传输给处理器710进行处理；另外，射频单元701可以向网络侧设备发送上行数据。通常，射频单元701包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器709可用于存储软件程序或指令以及各种数据。存储器709可主要包括存储程序或指令的第一存储区和存储数据的第二存储区，其中，第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器709可以包括易失性存储器或非易失性存储器，或者，存储器709可以包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory，RAM)，静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic RAM，DRAM)、同步动态随机存取存储器 (Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DRRAM)。本申请实施例中的存储器709包括但不限于这些和任意其它适合类型的存储器。
处理器710可包括一个或多个处理单元；可选的，处理器710集成应用处理器和调制解调处理器，其中，应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作，调制解调处理器主要处理无线通信信号，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器710中。
其中，处理器710，用于确定每个层layer的预编码矩阵的信息值；针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；
射频单元701，用于将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
在本申请实施例中，终端在对信道的预编码矩阵进行编码并向网络侧设备反馈信道信息时，由于可以根据每个layer的预编码矩阵的信息值所在的区间选择对应的编码模型对每个layer的预编码矩阵进行编码，且不同的信息值区间可以对应不同的编码模型，因此针对每个layer而言，可以使用与该layer的信道特征更加匹配的编码模型对该layer的预编码矩阵进行编码，从而可以提高编码准确度，进而准确反馈信道信息；由于终端在将编码后的预编码特征信息反馈给网络侧设备时，还可以将终端使用的目标编码模型的模型标识和/或目标编码模型对应的目标区间反馈给网络侧设备，因此可以便于网络侧设备确定相应的解码模型，进而对预编码特征信息进行准确解码。此外，由于可以根据layer的预编码矩阵的信息值的不同区间训练不同的编码模型，无需针对每个layer或多个layer训练对应的编码模型，因此还可以降低模型训练的复杂度。
可选的，处理器710，还用于以下至少一项：
所述终端根据信道信息确定信道的预编码矩阵和信道的秩rank；确定与所述rank对应的一个或多个layer的预编码矩阵；确定每个layer的预编码矩阵的第一信息值；
所述终端根据信道信息和信道的rank确定与所述rank对应的预编码矩阵和码字，一个码字对应一个或多个layer；确定与所述码字对应的第二信息值。
可选的，处理器710，还用于以下至少一项：
根据所述layer的预编码矩阵的所述第一信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息；
根据所述layer的预编码矩阵的所述第二信息值所在的目标区间，确定与所述目标区间 对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为至少一个layer对应的预编码特征信息，所述至少一个layer对应同一个码字；
根据所述layer的预编码矩阵的所述第二信息值所在的第一目标区间，确定与所述第一目标区间对应的编码模型集合；根据所述layer的预编码矩阵的所述第一信息值所在的第二目标区间，确定所述编码模型集合中与所述第二目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息。
可选的，射频单元701，还用于：
在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，接收所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，接收所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
在所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定的情况下，接收第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
可选的，射频单元701，还用于上报辅助能力；接收第二指示信息，所述第二指示信息用于指示开始辅助功能；
处理器710，还用于针对不同的信息值，遍历候选模型集合中的多个候选模型，并从所述多个候选模型中确定与不同的信息值对应的目标候选模型；
射频单元701，还用于将所述不同的信息值以及所述目标候选模型的模型标识发送给所述网络侧设备。
可选的，射频单元701，还用于以下任一项：
接收所述网络侧设备通过控制信道配置的所述候选模型集合；
接收所述网络侧设备通过广播信道配置的所述候选模型集合。
可选的，处理器710，还用于以下任一项：
取消上报所述layer的预编码矩阵；
根据所述网络侧设备的指示或协议的规定，将第一区间对应的编码模型确定为所述目标编码模型；
自行选择第二区间对应的编码模型为所述目标编码模型。
可选的，处理器710，还用于以下至少一项：
将所述第二区间发送给所述网络侧设备；
将所述第二区间对应的编码模型的模型标识发送给所述网络侧设备。
本申请实施例提供的终端700还可以实现上述信道信息反馈方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种网络侧设备，包括处理器和通信接口，通信接口用于接收预编码特征信息以及第一信息，处理器用于根据所述第一信息确定目标解码模型；根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。该网络侧设备实施例与上述网络侧设备方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该网络侧设备实施例中，且能达到相同的技术效果。
具体地，本申请实施例还提供了一种网络侧设备。如图8所示，该网络侧设备800包括：天线81、射频装置82、基带装置83、处理器84和存储器85。天线81与射频装置82连接。在上行方向上，射频装置82通过天线81接收信息，将接收的信息发送给基带装置83进行处理。在下行方向上，基带装置83对要发送的信息进行处理，并发送给射频装置82，射频装置82对收到的信息进行处理后经过天线81发送出去。
以上实施例中网络侧设备执行的方法可以在基带装置83中实现，该基带装置83包括基带处理器。
基带装置83例如可以包括至少一个基带板，该基带板上设置有多个芯片，如图8所示，其中一个芯片例如为基带处理器，通过总线接口与存储器85连接，以调用存储器85中的程序，执行以上方法实施例中所示的网络设备操作。
该网络侧设备还可以包括网络接口86，该接口例如为通用公共无线接口(common public radio interface，CPRI)。
具体地，本发明实施例的网络侧设备800还包括：存储在存储器85上并可在处理器84上运行的指令或程序，处理器84调用存储器85中的指令或程序执行图6所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
本申请实施例还提供一种可读存储介质，所述可读存储介质上存储有程序或指令，该程序或指令被处理器执行时实现上述信道信息反馈方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现上述信道信息反馈方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在 存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现上述信道信息反馈方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供了一种信道信息反馈系统，包括：终端及网络侧设备，所述终端可用于执行如上所述的信道信息反馈方法的步骤，所述网络侧设备可用于执行如上所述的信道信息反馈方法的步骤。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机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(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (40)

  1. 一种信道信息反馈方法，包括：
    终端确定每个层layer的预编码矩阵的信息值；
    针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；
    根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；
    将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
  2. 根据权利要求1所述的方法，其中，所述终端确定每个层layer的预编码矩阵的信息值，包括以下至少一项：
    所述终端根据信道信息确定信道的预编码矩阵和信道的秩rank；确定与所述rank对应的一个或多个layer的预编码矩阵；确定每个layer的预编码矩阵的第一信息值；
    所述终端根据信道信息和信道的rank确定与所述rank对应的预编码矩阵和码字，一个码字对应一个或多个layer；确定与所述码字对应的第二信息值。
  3. 根据权利要求2所述的方法，其中，
    所述rank为确定值或为所述终端自行选择的值；
    所述预编码矩阵为完整的奇异值分解SVD的预编码器precoder，或所述SVD的precoder投影在指定的正交基上的系数，或根据所述系数和所述正交基加权得到的precoder；
    所述第一信息值包括信道质量指示CQI、特征值、奇异值、信道容量以及离散傅里叶变换DFT投影的峰值数中的至少一项；
    所述第二信息值包括CQI、调制与编码方案MCS以及传输块大小TB size中的至少一项，一个码字对应一个CQI。
  4. 根据权利要求2所述的方法，其中，所述根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，包括以下至少一项：
    根据所述layer的预编码矩阵的所述第一信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息；
    根据所述layer的预编码矩阵的所述第二信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为至少一个layer对应的预编码特征信息，所述至少一个layer对应同一个码字；
    根据所述layer的预编码矩阵的所述第二信息值所在的第一目标区间，确定与所述第一目标区间对应的编码模型集合；根据所述layer的预编码矩阵的所述第一信息值所在的第二目标区间，确定所述编码模型集合中与所述第二目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息。
  5. 根据权利要求1所述的方法，其中，所述信息值区间和所述编码模型包括以下至少 一项：
    不同信息值区间对应的模型输出长度不同，信息值区间的信息值越大，对应的模型输出长度越大；
    不同信息值区间对应的模型输出长度相同，不同信息值区间对应的编码模型基于不同的训练集训练得到；
    一个信息值区间对应多个编码模型，所述多个编码模型的模型输出长度不同，且所述多个编码模型适用于不同的应用场景。
  6. 根据权利要求1所述的方法，其中，所述信息值区间和所述编码模型包括以下任一项：
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构和模型参数由协议规定；
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量由协议规定，或由核心网指示，或由所述网络侧设备自行确定，所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定；
    其中，N个信息值区间对应N个或N组模型，N为大于或等于1的整数。
  7. 根据权利要求6所述的方法，其中，所述方法还包括：
    在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，所述终端接收所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
    在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，所述终端接收所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
    在所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定的情况下，所述终端接收第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
  8. 根据权利要求6所述的方法，其中，在所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定的情况下，所述方法还包括：
    所述终端上报辅助能力；
    接收第二指示信息，所述第二指示信息用于指示开始辅助功能；
    针对不同的信息值，遍历候选模型集合中的多个候选模型，并从所述多个候选模型中确定与不同的信息值对应的目标候选模型；
    将所述不同的信息值以及所述目标候选模型的模型标识发送给所述网络侧设备。
  9. 根据权利要求8所述的方法，其中，所述候选模型集合由所述网络侧设备配置或由协议规定；
    其中，在所述候选模型集合由所述网络侧设备配置的情况下，所述方法还包括以下任一项：
    接收所述网络侧设备通过控制信道配置的所述候选模型集合；
    接收所述网络侧设备通过广播信道配置的所述候选模型集合。
  10. 根据权利要求8所述的方法，其中，
    所述第二指示信息通过信令发送，所述第二指示信息中包括所述终端开始辅助功能后的结束时间或持续次数，或者，所述第二指示信息为周期性发送。
  11. 根据权利要求1所述的方法，其中，在已有的编码模型中不存在与所述目标区间对应的目标编码模型的情况下，所述方法还包括以下任一项：
    取消上报所述layer的预编码矩阵；
    根据所述网络侧设备的指示或协议的规定，将第一区间对应的编码模型确定为所述目标编码模型；
    自行选择第二区间对应的编码模型为所述目标编码模型。
  12. 根据权利要求11所述的方法，其中，在自行选择第二区间对应的编码模型为所述目标编码模型的情况下，所述方法还包括以下至少一项：
    将所述第二区间发送给所述网络侧设备；
    将所述第二区间对应的编码模型的模型标识发送给所述网络侧设备。
  13. 根据权利要求1所述的方法，其中，
    所述信息值是归一化后的值，所述归一化的具体方式由协议规定。
  14. 一种信道信息反馈方法，包括：
    网络侧设备接收预编码特征信息以及第一信息；
    根据所述第一信息确定目标解码模型；
    根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；
    其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
  15. 根据权利要求14所述的方法，其中，所述信息值区间和所述编码模型包括以下任一项：
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构和模型参数由协议规定；
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量由协议规定，或由核心网指示，或由所述网络侧设备自行确定，所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定；
    其中，N个信息值区间对应N个或N组模型，N为大于或等于1的整数。
  16. 根据权利要求15所述的方法，其中，所述方法还包括：
    在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，所述网络侧设备向所述终端发送所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
    在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，所述网络侧设备向终端发送所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
    在所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定的情况下，网络侧设备向所述终端发送第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
  17. 根据权利要求16所述的方法，其中，在所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定的情况下，所述方法还包括：
    接收所述终端上报的辅助能力；
    向所述终端发送第二指示信息，所述第二指示信息用于指示开始辅助功能；
    接收所述终端发送的不同的信息值以及与所述不同的信息值对应的目标候选模型的模型标识，所述目标候选模型由所述终端针对所述不同的信息值，遍历候选模型集合中的多个候选模型后从所述多个候选模型中确定得到；
    向所述终端发送所述第一指示信息。
  18. 根据权利要求17所述的方法，其中，所述候选模型集合由所述网络侧设备配置或由协议规定；
    其中，在所述候选模型集合由所述网络侧设备配置的情况下，所述方法还包括以下任一项：
    通过控制信道配置所述候选模型集合；
    通过广播信道配置所述候选模型集合。
  19. 根据权利要求17所述的方法，其中，
    所述第二指示信息通过信令发送，所述第二指示信息中包括所述终端开始辅助功能后的结束时间或持续次数，或者，所述第二指示信息为周期性发送。
  20. 根据权利要求17所述的方法，其中，向所述终端发送所述第一指示信息，包括以下任一项：
    在根据所述不同的信息值以及所述目标候选模型确定所述信息值区间的划分和数量、 所述编码模型后，向所述终端发送所述第一指示信息；
    在配置所述终端取消上报预编码矩阵的信息值和候选模型的模型标识后，向所述终端发送所述第一指示信息；
    在到达所述第二指示信息指示的结束时间时，向所述终端发送所述第一指示信息。
  21. 一种信道信息反馈装置，包括：
    第一确定模块，用于确定每个层layer的预编码矩阵的信息值；
    第二确定模块，用于针对每个layer，根据所述layer的预编码矩阵的信息值所在的目标区间，确定与所述目标区间对应的目标编码模型，不同的信息值区间对应不同的编码模型；
    编码模块，用于根据所述目标编码模型对所述layer的预编码矩阵进行处理得到预编码特征信息；
    发送模块，用于将所述预编码特征信息以及第一信息发送给网络侧设备，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
  22. 根据权利要求21所述的装置，其中，所述第一确定模块，还用于以下至少一项：
    根据信道信息确定信道的预编码矩阵和信道的秩rank；确定与所述rank对应的一个或多个layer的预编码矩阵；确定每个layer的预编码矩阵的第一信息值；
    根据信道信息和信道的rank确定与所述rank对应的预编码矩阵和码字，一个码字对应一个或多个layer；确定与所述码字对应的第二信息值。
  23. 根据权利要求22所述的装置，其中，
    所述rank为确定值或为所述终端自行选择的值；
    所述预编码矩阵为完整的奇异值分解SVD的预编码器precoder，或所述SVD的precoder投影在指定的正交基上的系数，或根据所述系数和所述正交基加权得到的precoder；
    所述第一信息值包括信道质量指示CQI、特征值、奇异值、信道容量以及离散傅里叶变换DFT投影的峰值数中的至少一项；
    所述第二信息值包括CQI、调制与编码方案MCS以及传输块大小TB size中的至少一项，一个码字对应一个CQI。
  24. 根据权利要求22所述的装置，其中，所述第二确定模块，还用于以下至少一项：
    根据所述layer的预编码矩阵的所述第一信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息；
    根据所述layer的预编码矩阵的所述第二信息值所在的目标区间，确定与所述目标区间对应的目标编码模型；其中，基于所述目标编码模型处理得到的预编码特征信息为至少一个layer对应的预编码特征信息，所述至少一个layer对应同一个码字；
    根据所述layer的预编码矩阵的所述第二信息值所在的第一目标区间，确定与所述第一目标区间对应的编码模型集合；根据所述layer的预编码矩阵的所述第一信息值所在的第二目标区间，确定所述编码模型集合中与所述第二目标区间对应的目标编码模型；其中，基 于所述目标编码模型处理得到的预编码特征信息为一个layer对应的预编码特征信息。
  25. 根据权利要求21所述的装置，其中，所述信息值区间和所述编码模型包括以下至少一项：
    不同信息值区间对应的模型输出长度不同，信息值区间的信息值越大，对应的模型输出长度越大；
    不同信息值区间对应的模型输出长度相同，不同信息值区间对应的编码模型基于不同的训练集训练得到；
    一个信息值区间对应多个编码模型，所述多个编码模型的模型输出长度不同，且所述多个编码模型适用于不同的应用场景。
  26. 根据权利要求21所述的装置，其中，所述信息值区间和所述编码模型包括以下任一项：
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构和模型参数由协议规定；
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量由协议规定，或由核心网指示，或由所述网络侧设备自行确定，所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定；
    其中，N个信息值区间对应N个或N组模型，N为大于或等于1的整数。
  27. 根据权利要求26所述的装置，其中，所述装置还包括第一接收模块，所述第一接收模块，用于：
    在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，接收所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
    在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，接收所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
    在所述信息值区间的划分和数量、所述编码模型由所述终端辅助所述网络侧设备确定的情况下，接收第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
  28. 根据权利要求26所述的装置，其中，所述装置还包括第二接收模块；
    所述发送模块，还用于上报辅助能力；
    所述第二接收模块，用于接收第二指示信息，所述第二指示信息用于指示开始辅助功能，所述第二指示信息通过信令发送，所述第二指示信息中包括所述终端开始辅助功能后 的结束时间或持续次数，或者，所述第二指示信息为周期性发送；
    所述第二确定模块，还用于针对不同的信息值，遍历候选模型集合中的多个候选模型，并从所述多个候选模型中确定与不同的信息值对应的目标候选模型；
    所述发送模块，还用于将所述不同的信息值以及所述目标候选模型的模型标识发送给所述网络侧设备。
  29. 根据权利要求28所述的装置，其中，所述候选模型集合由所述网络侧设备配置或由协议规定；
    其中，所述第二接收模块，还用于以下任一项：
    接收所述网络侧设备通过控制信道配置的所述候选模型集合；
    接收所述网络侧设备通过广播信道配置的所述候选模型集合。
  30. 根据权利要求21所述的装置，其中，所述第二确定模块，还用于以下任一项：
    取消上报所述layer的预编码矩阵；
    根据所述网络侧设备的指示或协议的规定，将第一区间对应的编码模型确定为所述目标编码模型；
    自行选择第二区间对应的编码模型为所述目标编码模型。
  31. 根据权利要求30所述的装置，其中，所述发送模块，还用于以下至少一项：
    将所述第二区间发送给所述网络侧设备；
    将所述第二区间对应的编码模型的模型标识发送给所述网络侧设备。
  32. 一种信道信息反馈装置，包括：
    接收模块，用于接收预编码特征信息以及第一信息；
    确定模块，用于根据所述第一信息确定目标解码模型；
    解码模块，用于根据所述目标解码模型对所述预编码特征信息进行处理，得到对应的预编码矩阵；
    其中，所述预编码特征信息基于目标编码模型对layer的预编码矩阵进行处理后得到，所述目标编码模型为所述layer的预编码矩阵的信息值所在的目标区间对应的编码模型，不同的信息值区间对应不同的编码模型，所述第一信息包括所述目标编码模型的模型标识和/或所述目标区间。
  33. 根据权利要求32所述的装置，其中，所述信息值区间和所述编码模型包括以下任一项：
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构和模型参数由协议规定；
    所述信息值区间的划分和数量由协议规定，所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到；
    所述信息值区间的划分和数量由协议规定，或由核心网指示，或由所述网络侧设备自行确定，所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数 由所述网络侧设备训练得到；
    所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定；
    其中，N个信息值区间对应N个或N组模型，N为大于或等于1的整数。
  34. 根据权利要求33所述的装置，其中，所述装置还包括发送模块，所述发送模块，用于：
    在所述编码模型的模型结构由协议规定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，向所述终端发送所述编码模型的模型参数，或所述编码模型的模型参数以及所述编码模型对应的解码模型的模型参数；
    在所述编码模型的模型结构由所述网络侧设备自行确定，所述编码模型的模型参数由所述网络侧设备训练得到的情况下，向终端发送所述编码模型，或所述编码模型以及所述编码模型对应的解码模型；
    在所述信息值区间的划分和数量、所述编码模型由终端辅助所述网络侧设备确定的情况下，向所述终端发送第一指示信息，所述第一指示信息用于指示所述信息值区间的划分和数量、所述编码模型。
  35. 根据权利要求34所述的装置，其中，
    所述接收模块，用于接收所述终端上报的辅助能力；
    所述发送模块，用于向所述终端发送第二指示信息，所述第二指示信息用于指示开始辅助功能，所述第二指示信息通过信令发送，所述第二指示信息中包括所述终端开始辅助功能后的结束时间或持续次数，或者，所述第二指示信息为周期性发送；
    所述接收模块，用于接收所述终端发送的不同的信息值以及与所述不同的信息值对应的目标候选模型的模型标识，所述目标候选模型由所述终端针对所述不同的信息值，遍历候选模型集合中的多个候选模型后从所述多个候选模型中确定得到；
    所述发送模块，用于向所述终端发送所述第一指示信息。
  36. 根据权利要求35所述的装置，其中，所述候选模型集合由所述网络侧设备配置或由协议规定；
    其中，所述发送模块，还用于以下任一项：
    通过控制信道配置所述候选模型集合；
    通过广播信道配置所述候选模型集合。
  37. 根据权利要求35所述的装置，其中，所述发送模块，还用于以下任一项：
    在根据所述不同的信息值以及所述目标候选模型确定所述信息值区间的划分和数量、所述编码模型后，向所述终端发送所述第一指示信息；
    在配置所述终端取消上报预编码矩阵的信息值和候选模型的模型标识后，向所述终端发送所述第一指示信息；
    在到达所述第二指示信息指示的结束时间时，向所述终端发送所述第一指示信息。
  38. 一种终端，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或 指令，所述程序或指令被所述处理器执行时实现如权利要求1至13任一项所述的信道信息反馈方法的步骤。
  39. 一种网络侧设备，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求14至20任一项所述的信道信息反馈方法的步骤。
  40. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如权利要求1至13任一项所述的信道信息反馈方法，或者实现如权利要求14至20任一项所述的信道信息反馈方法的步骤。
PCT/CN2023/090381 2022-04-28 2023-04-24 信道信息反馈方法、终端及网络侧设备 WO2023207920A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210461819.2A CN117014044A (zh) 2022-04-28 2022-04-28 信道信息反馈方法、终端及网络侧设备
CN202210461819.2 2022-04-28

Publications (1)

Publication Number Publication Date
WO2023207920A1 true WO2023207920A1 (zh) 2023-11-02

Family

ID=88517820

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/090381 WO2023207920A1 (zh) 2022-04-28 2023-04-24 信道信息反馈方法、终端及网络侧设备

Country Status (2)

Country Link
CN (1) CN117014044A (zh)
WO (1) WO2023207920A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109873665A (zh) * 2017-12-01 2019-06-11 华为技术有限公司 数据传输的方法和设备
WO2019114690A1 (zh) * 2017-12-15 2019-06-20 华为技术有限公司 多天线系统中预编码的方法及装置
CN111327352A (zh) * 2018-12-17 2020-06-23 华为技术有限公司 一种通信方法及设备
WO2020186871A1 (zh) * 2019-03-21 2020-09-24 电信科学技术研究院有限公司 一种信道状态信息上报的方法和设备
US20220038159A1 (en) * 2018-12-11 2022-02-03 Qualcomm Incorporated Compressed csi feedback for non-contiguous frequency resources

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109873665A (zh) * 2017-12-01 2019-06-11 华为技术有限公司 数据传输的方法和设备
WO2019114690A1 (zh) * 2017-12-15 2019-06-20 华为技术有限公司 多天线系统中预编码的方法及装置
US20220038159A1 (en) * 2018-12-11 2022-02-03 Qualcomm Incorporated Compressed csi feedback for non-contiguous frequency resources
CN111327352A (zh) * 2018-12-17 2020-06-23 华为技术有限公司 一种通信方法及设备
WO2020186871A1 (zh) * 2019-03-21 2020-09-24 电信科学技术研究院有限公司 一种信道状态信息上报的方法和设备

Also Published As

Publication number Publication date
CN117014044A (zh) 2023-11-07

Similar Documents

Publication Publication Date Title
US20230246759A1 (en) Method and apparatus for processing channel state information, and terminal
CN114337953B (zh) 上行信道参数的确定和配置方法及装置
US20220038154A1 (en) Csi calculation method, user terminal and computer-readable storage medium
WO2023246618A1 (zh) 信道矩阵处理方法、装置、终端及网络侧设备
WO2023207920A1 (zh) 信道信息反馈方法、终端及网络侧设备
WO2024055974A1 (zh) Cqi传输方法、装置、终端及网络侧设备
WO2023179460A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2024007949A1 (zh) Ai模型处理方法、装置、终端及网络侧设备
WO2023185980A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2024055993A1 (zh) Cqi传输方法、装置、终端及网络侧设备
WO2024051564A1 (zh) 信息传输方法、ai网络模型训练方法、装置和通信设备
WO2023179570A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2024037380A1 (zh) 信道信息处理方法、装置、通信设备及存储介质
WO2023185995A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2024104126A1 (zh) 更新ai网络模型的方法、装置和通信设备
WO2024051594A1 (zh) 信息传输方法、ai网络模型训练方法、装置和通信设备
WO2024093999A1 (zh) 信道信息的上报和接收方法、终端及网络侧设备
WO2023179476A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
WO2023241643A1 (zh) 联合传输的pmi参数反馈方法、终端及网络侧设备
WO2024104070A1 (zh) 波束报告的发送方法、接收方法、装置及通信设备
WO2023016339A1 (zh) Csi上报方法、接收方法、装置、终端和网络侧设备
WO2023143361A1 (zh) 能力信息上报方法、装置及终端
WO2024007918A1 (zh) 预编码矩阵指示、确定方法、装置、网络侧设备及终端
WO2023179651A1 (zh) 波束处理方法、装置及设备
WO2024041420A1 (zh) 测量反馈处理方法、装置、终端及网络侧设备

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: 23795353

Country of ref document: EP

Kind code of ref document: A1