WO2023246618A1 - 信道矩阵处理方法、装置、终端及网络侧设备 - Google Patents

信道矩阵处理方法、装置、终端及网络侧设备 Download PDF

Info

Publication number
WO2023246618A1
WO2023246618A1 PCT/CN2023/100426 CN2023100426W WO2023246618A1 WO 2023246618 A1 WO2023246618 A1 WO 2023246618A1 CN 2023100426 W CN2023100426 W CN 2023100426W WO 2023246618 A1 WO2023246618 A1 WO 2023246618A1
Authority
WO
WIPO (PCT)
Prior art keywords
terminal
orthogonal bases
orthogonal
model
network side
Prior art date
Application number
PCT/CN2023/100426
Other languages
English (en)
French (fr)
Inventor
任千尧
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 WO2023246618A1 publication Critical patent/WO2023246618A1/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
    • 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
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H04L25/0242—Channel estimation channel estimation algorithms using matrix methods

Definitions

  • This application belongs to the field of communication technology, and specifically relates to a channel matrix processing method, device, terminal and network side equipment.
  • the base station sends a channel state information reference signal (Channel State Information Reference Signal, CSI-RS) on certain time-frequency resources in a certain time slot.
  • CSI-RS Channel State Information Reference Signal
  • the terminal performs channel estimation based on the CSI-RS and calculates this
  • the channel information on the time slot feeds back the Precoding Matrix Indicator (PMI) to the base station through the codebook.
  • PMI Precoding Matrix Indicator
  • the base station combines the channel information based on the codebook information fed back by the terminal.
  • the base station uses this to process data before the next CSI report. Precoding and multi-user scheduling.
  • the codebook content fed back by the terminal is the characteristic matrix of the channel matrix, that is, only the precoding matrix of the transmitter is fed back.
  • the base station only predicts channel information based on the precoding matrix of the transmitter, resulting in The accuracy of channel information prediction is low.
  • Embodiments of the present application provide a channel matrix processing method, device, terminal and network-side equipment, which can solve the problem in related technologies that network-side equipment predicts channel information with low accuracy.
  • a channel matrix processing method including:
  • the terminal determines N first orthogonal bases and M second orthogonal bases, and preprocesses the channel matrix of each subband based on the N first orthogonal bases and the M second orthogonal bases,
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna
  • N and M are both positive integers
  • the terminal inputs the preprocessed channel matrix into the first artificial intelligence AI model, and obtains the channel characteristic information output by the first AI model;
  • the terminal reports the channel characteristic information to the network side device.
  • a channel matrix processing method including:
  • the network side device receives the channel characteristic information reported by the terminal
  • the network side device inputs the channel characteristic information into the second AI model, and obtains the second AI model output Output channel matrix;
  • the channel characteristic information is the output of the first AI model of the terminal
  • the input of the first AI model is that the terminal uses N first orthogonal bases and M second orthogonal bases to pair each sub
  • the channel matrix of the band is preprocessed.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • N and M are both positive integers.
  • a channel matrix processing device including:
  • a processing module configured to determine N first orthogonal bases and M second orthogonal bases, and generate a channel matrix for each subband based on the N first orthogonal bases and the M second orthogonal bases. Perform preprocessing, the first orthogonal basis corresponds to the airspace information of the transmitting end antenna, the second orthogonal base corresponds to the airspace information of the receiving end antenna, N and M are both positive integers;
  • the first acquisition module is used to input the preprocessed channel matrix into the first artificial intelligence AI model and obtain the channel characteristic information output by the first AI model;
  • a reporting module is used to report the channel characteristic information to the network side device.
  • a channel matrix processing device including:
  • the receiving module is used to receive the channel characteristic information reported by the terminal;
  • a second acquisition module configured to input the channel characteristic information into the second AI model and obtain the channel matrix output by the second AI model
  • the channel characteristic information is the output of the first AI model of the terminal
  • the input of the first AI model is that the terminal uses N first orthogonal bases and M second orthogonal bases to pair each sub
  • the channel matrix of the band is preprocessed.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • N and M are both positive integers.
  • 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 channel matrix processing method described in one aspect.
  • a terminal including a processor and a communication interface, wherein the processor is configured to determine N first orthogonal bases and M second orthogonal bases, and determine N first orthogonal bases based on the N first orthogonal bases.
  • the orthogonal base and the M second orthogonal bases preprocess the channel matrix of each subband.
  • the first orthogonal base corresponds to the air domain information of the transmitting end antenna, and the second orthogonal base corresponds to the receiving end antenna.
  • Spatial information, N and M are both positive integers; and used to input the preprocessed channel matrix into the first artificial intelligence AI model to obtain the channel characteristic information output by the first AI model; the communication interface is used to provide The network side device reports the channel characteristic information.
  • 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.
  • the communication interface is used to receive channel characteristic information reported by a terminal;
  • the processor is used to input the channel characteristic information into a second AI model. , obtain the channel matrix output by the second AI model;
  • the channel characteristic information is the output of the first AI model of the terminal
  • the input of the first AI model is that the terminal uses N first orthogonal bases and M second orthogonal bases to pair each sub
  • the channel matrix of the band is preprocessed.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • N and M are both positive integers.
  • a ninth aspect provides a communication system, including: a terminal and a network side device.
  • the terminal can be used to perform the steps of the channel matrix processing method as described in the first aspect.
  • the network side device can be used to perform the steps of the second aspect. The steps of the channel matrix processing method described in this aspect.
  • a readable storage medium In a tenth 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 channel matrix processing method as described in the first aspect are implemented. Or implement the steps of the channel matrix processing method described in the second 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. Channel matrix processing method, or implementing the channel matrix processing method as described in the second 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
  • terminals and network-side devices can process and recover the channel matrix through the AI model, which can effectively save transmission resources.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • the channel characteristic information obtained by the terminal through the first AI model processing is also considered at the same time.
  • the network side equipment performs channel matrix recovery through the second AI model, and thus can obtain the full channel information of the receiving end and transmitting end. In this way, the network side device can not only obtain the channel information of the sending end, but also the channel information of the receiving end, which further helps the network side device to predict the channel information to improve the accuracy of channel information prediction.
  • Figure 1 is a block diagram of a wireless communication system applicable to the embodiment of the present application.
  • Figure 2 is a flow chart of a channel matrix processing method provided by an embodiment of the present application.
  • Figure 3 is a flow chart of another channel matrix processing method provided by an embodiment of the present application.
  • Figure 4 is a structural diagram of a channel matrix processing device provided by an embodiment of the present application.
  • Figure 5 is a structural diagram of another channel matrix processing device provided by an embodiment of the present application.
  • Figure 6 is a structural diagram of a communication device provided by an embodiment of the present application.
  • Figure 7 is a structural diagram of a terminal provided by an embodiment of the present application.
  • Figure 8 is a structural diagram of a network side device provided by an embodiment of the present application.
  • 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
  • WUE Vehicle User Equipment
  • PUE Pedestrian User Equipment
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • game consoles personal computers (personal computer, PC), teller machine or self-service machine and other terminal-side devices.
  • Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets) bracelets, 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 may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or a wireless device.
  • Access network equipment may include base stations, Wireless Local Area Network (WLAN) access points, Wireless Fidelity (WiFi) nodes, etc.
  • the base station may be called Node B, evolved node B (eNB), access point, Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (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.
  • eNB evolved node B
  • BTS Base Transceiver Station
  • BSS Basic Service Set
  • Extended Service Set Extended Service Set
  • ESS Extended Service Set
  • home B-node home evolved B-node
  • transmitting receiving point Transmitting Receiving Point
  • Channel State Information (Channel State Information, CSI) is crucial to channel capacity.
  • the transmitter can optimize signal transmission based on CSI to better match the channel status.
  • CQI Channel Quality Indicator
  • MCS Modulation and Coding Scheme
  • PMI Precoding Matrix Indicator
  • Eigen beamforming Eigen beamforming
  • the base station sends a channel state information reference signal (CSI-RS) on certain time-frequency resources in a certain time slot (slot).
  • CSI-RS channel state information reference signal
  • the terminal performs channel estimation based on the CSI-RS and calculates the The channel information is fed back to the base station through the codebook.
  • the base station combines the channel information based on the codebook information fed back by the terminal. The base station uses this to perform data precoding and multi-user scheduling before the next CSI report.
  • the terminal can change the PMI reported on each subband to report PMI based on delay. Since the channels in the delay domain are more concentrated, PMI with fewer delays can approximately represent the PMI of all subbands. That is, the delay field information will be compressed before reporting.
  • the base station can precode the CSI-RS in advance and send the coded CSI-RS to the terminal. What the terminal sees is the channel corresponding to the coded CSI-RS. The terminal only needs to Just select several ports with greater strength among the indicated ports and report the coefficients corresponding to these ports.
  • neural network or machine learning methods can be used.
  • AI Artificial intelligence
  • neural networks such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc.
  • This application uses neural network as an example for explanation, but does not limit the specific type of AI model.
  • the terminal uses an AI model to compress and encode the channel information, and the base station decodes the compressed content through the AI model to restore the channel information.
  • the base station's AI model for decoding and the terminal's AI model for encoding AI models need to be jointly trained to achieve a reasonable matching degree.
  • the terminal's AI model for encoding and the base station's AI model for decoding form a joint neural network model, which is jointly trained by the network side.
  • the base station sends the AI model for encoding to the terminal.
  • the terminal estimates CSI-RS, calculates channel information, uses the calculated channel information or original estimated channel information through the AI model used for encoding to obtain the encoding result, and sends the encoding result to the base station.
  • the base station receives the encoded result. Input it into the decoding AI model to restore the channel information.
  • the terminal will obtain an estimated channel matrix on each subband, such as a 4 ⁇ 32 channel matrix, representing 4 receiving antennas and 32 CSI-RS ports.
  • the channel matrix on each subband is different. Due to frequency selective fading, the two built-in channel matrices may be very different.
  • the traditional R15 codebook feeds back the channel information of each subband separately.
  • the traditional R16 codebook uses this relationship to convert the frequency domain channel into the delay domain, thereby obtaining a channel matrix with concentrated energy, and only reports the part with strong energy. By delaying the corresponding channel information, more complete channel information can be obtained.
  • Codebook-based CSI feedback schemes mostly feed back the channel precoding matrix, that is, the characteristic matrix of the channel matrix. This is mainly to save overhead.
  • the reported CSI is mainly used by the base station for precoding, so a complete channel is not required.
  • Information only requires the characteristic information of the transmitter of the channel, so only reporting the channel characteristic matrix can reduce overhead without affecting performance.
  • base stations need to predict channel information at a certain time in the future in order to better schedule users, so base stations need complete channel information.
  • Figure 2 is a flow chart of a channel matrix processing method provided by an embodiment of the present application. As shown in Figure 2, the method includes the following steps:
  • Step 201 The terminal determines N first orthogonal bases and M second orthogonal bases, and performs the channel matrix calculation on each subband based on the N first orthogonal bases and the M second orthogonal bases.
  • the first orthogonal basis corresponds to the spatial domain information of the transmitting end antenna
  • the second orthogonal basis corresponds to the spatial domain information of the receiving end antenna.
  • N and M are both positive integers.
  • the sending end may refer to a network side device (also called a base station), and the receiving end may refer to a terminal; or, in other communication scenarios, the sending end may refers to a terminal, and the receiving end may refer to a network-side device.
  • a network side device also called a base station
  • the receiving end may refer to a terminal
  • the sending end may refers to a terminal
  • the receiving end may refer to a network-side device.
  • the terminal may select among candidate orthogonal bases to determine N first orthogonal bases and M second orthogonal bases, and correspond to the N first orthogonal bases to send The spatial domain information of the terminal antenna is determined, and the M second orthogonal bases are corresponding to the spatial domain information of the receiving terminal antenna.
  • the terminal can determine corresponding orthogonal bases for both the transmitting end and the receiving end, so as to process the airspace information of both the transmitting end and the receiving end through the orthogonal bases.
  • the candidate orthogonal base may be obtained by referring to related technologies, which will not be described in detail in this application.
  • preprocessing the channel matrix of each subband based on the N first orthogonal bases and the M second orthogonal bases includes any of the following:
  • the channel matrix of each subband based on the N first orthogonal bases and the M second orthogonal bases, and perform weighting processing on the projected channel matrix, so that the weighted channel
  • the dimension of the matrix is consistent with the channel matrix before projection processing, and the orthogonal bases used in the weighting process are the N first orthogonal bases and the M second orthogonal bases.
  • the terminal obtains the estimated channel matrix of each subband, and after determining N first orthogonal bases and M second orthogonal bases, projects the estimated channel matrix of each subband.
  • N first orthogonal bases and M second orthogonal bases the channel matrix with dimension N ⁇ M or M ⁇ N after projection processing of each subband is obtained, and the projection of the channel matrix after projection processing can be calculated. coefficient.
  • the N first orthogonal bases are based on the corresponding
  • the projection coefficients are sorted according to the first preset order, and the M second orthogonal bases are sorted according to the second preset order based on the corresponding projection coefficients.
  • the N first orthogonal bases may be sorted from large to small based on the corresponding projection coefficients
  • the M second orthogonal bases may also be sorted from large to small based on the corresponding projection coefficients.
  • the projection coefficients in the first row and first column of the obtained projected channel matrix are the projection coefficients corresponding to the strongest first orthogonal basis and the strongest second orthogonal basis.
  • the terminal does not need to report the sorting method of the first orthogonal basis and the second orthogonal basis to the network side device.
  • the terminal obtains the estimated channel matrix of each subband, and after determining the N first orthogonal bases and M second orthogonal bases, projects the estimated channel matrix of each subband to Based on N first orthogonal bases and M second orthogonal bases, a channel matrix with a dimension of N ⁇ M or M ⁇ N after projection processing of each subband is obtained, and then based on the N first orthogonal bases and The M second orthogonal bases perform weighting processing on a channel matrix with a dimension of N ⁇ M or M ⁇ N, so that the dimension of the channel matrix after the weighting processing is consistent with the dimension of the channel matrix before the projection processing.
  • Step 202 The terminal inputs the preprocessed channel matrix into the first AI model, and obtains the channel characteristic information output by the first AI model.
  • the terminal inputs the channel matrix after projection processing of each subband into the first AI model, or may input the channel matrix after projection processing and weighting processing for each subband into the first AI model.
  • the first AI model performs encoding processing on the input channel matrix to obtain channel characteristic information corresponding to each subband output by the first AI model.
  • the first AI model is an AI model obtained after pre-training.
  • the trained first AI model can process the input channel matrix based on its own network structure and network parameters, and output channel characteristic information.
  • the training method of the first AI model may refer to the training method of the network model in the related art, which will not be described in detail in this embodiment.
  • Step 203 The terminal reports the channel characteristic information to the network side device.
  • the terminal after obtaining the channel characteristic information output by the first AI model, the terminal reports the channel characteristic information to the network side device.
  • the network side device includes a second AI model that matches the first AI model.
  • the network side device inputs the channel characteristic information into the second AI model, and uses the second AI model to compare the input channel
  • the feature information is decoded and the complete channel matrix is restored and output.
  • the first AI model and the second AI model match each other, the output of the first AI model is the input of the second AI model, and the output of the second AI model can be as close as possible to the input of the first AI model. , thereby enabling the network side device to recover the channel matrix through the second AI model.
  • the first AI model and the second AI model may be jointly trained by a network side device, and the network side device sends the trained first AI model to the terminal, so that the terminal can pass the first AI model.
  • An AI model processes the input channel matrix into channel characteristic information and outputs it.
  • the terminal determines the N first orthogonal bases and the M second orthogonal bases, it determines the N first orthogonal bases and the M second orthogonal bases for each subband.
  • the channel matrix is preprocessed, and the preprocessed channel matrix of each subband is input into the first AI model to obtain the channel characteristic information output by the first AI model.
  • the terminal reports the channel characteristic information to the network side device.
  • the network side device processes the channel characteristic information through the second AI model and outputs the channel matrix, so that the network side device restores the channel matrix through the second AI model. In this way, terminals and network-side devices can process and restore the channel matrix through the AI model, which can effectively save transmission resources.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • the channel characteristic information obtained by the terminal through the first AI model processing is also considered at the same time.
  • the network side equipment performs channel matrix recovery through the second AI model, and thus can obtain the full channel information of the receiving end and transmitting end. In this way, the network side device can not only obtain the channel information of the sending end, but also the channel information of the receiving end, which further helps the network side device to predict the channel information to improve the accuracy of channel information prediction.
  • the terminal determines N first orthogonal bases and M second orthogonal bases, including any one of the following:
  • the terminal determines N first orthogonal bases and M second orthogonal bases based on the first AI model, and the positions of the N first orthogonal bases and the M second orthogonal bases. Match with the first AI model;
  • the terminal receives the first indication information sent by the network side device, and determines N first orthogonal bases and M second orthogonal bases based on the first indication information, where the first indication information is used to indicate the The value of N and the value of M.
  • the network side device may specify or pre-configure the positions of the N first orthogonal bases and the M second orthogonal bases to match the first AI model, The matching relationship is indicated to the terminal, and the terminal determines N first orthogonal bases and M second orthogonal bases based on the matching relationship indicated by the network side device.
  • the network side device may also send the matching relationship to the terminal when sending the first AI model to the terminal.
  • the network side device may also, when training the first AI model, synchronously train the positions of the first AI model and the N first orthogonal bases and the M second orthogonal bases. Matching relationship between locations.
  • the terminal determines the first AI model, that is, determines the corresponding N first orthogonal bases and M second orthogonal bases; or, the terminal determines the N first orthogonal bases.
  • Cross basis and M second orthonormal basis also That is, the corresponding first AI model is determined. Since the first AI model is trained by the network side device and sent to the terminal, in this case, the terminal does not need to report the index of the first orthogonal base and the index of the second orthogonal base to the network side device.
  • the network side device is based on the terminal
  • the first AI model used can know the first orthogonal basis and the second orthogonal basis selected by the terminal. In this way, the terminal's transmission resources can be saved, and the terminal does not need to select the first orthogonal base and the second orthogonal base, thereby simplifying the terminal's orthogonal base selection process.
  • the network side device indicates the values of N and M through first indication information, and then the terminal determines N first orthogonal bases and M second orthogonal bases based on the first indication information. base. It should be noted that the network side device only indicates the number of orthogonal bases, and does not indicate the specific positions of the orthogonal bases.
  • the method also includes:
  • the terminal reports to the network side device the indexes of the N first orthogonal bases, the indexes of the M second orthogonal bases, and the index of the oversampling group, where the N first orthogonal bases and The M second orthogonal bases are orthogonal bases determined by the terminal from the oversampling group after oversampling.
  • the terminal may oversample the candidate orthogonal bases to obtain Candidate orthogonal base oversampling group, the terminal selects N first orthogonal bases and M second orthogonal bases from the oversampling group, and reports the index and M of the N first orthogonal bases to the network side device the respective indexes of the second orthogonal basis, and the index of the oversampling group reported for selection.
  • the network side device can accurately know the oversampling group and orthogonal basis selected by the terminal, which helps the network side device accurately restore the channel matrix.
  • the terminal performs oversampling processing on the candidate orthogonal bases and determines N first orthogonal bases and M second orthogonal bases from the oversampling group.
  • the terminal may include the following steps:
  • the terminal determines the first candidate orthogonal base and the second candidate orthogonal base, and performs oversampling processing on the first candidate orthogonal base and the second candidate orthogonal base respectively to obtain n groups of first candidate orthogonal bases.
  • Intersection basis and m group of second candidate orthogonal basis, n and m are both positive integers;
  • the terminal selects among j groups of first candidate orthogonal bases to obtain N first orthogonal bases, and selects among k groups of second candidate orthogonal bases to obtain M second orthogonal bases;
  • the j group of first candidate orthogonal bases is any one of the n groups of first candidate orthogonal bases
  • the terminal can oversample the candidate orthogonal basis and select from the oversampling group obtained by the oversampling process.
  • a group is used to select the first orthogonal basis and the second orthogonal basis to enhance the richness and range of orthogonal basis selection.
  • the oversampling processing of the orthogonal basis may refer to related technologies, and will not be described in detail in this embodiment.
  • the value of N and the value of M may also be the same as The first AI model matches.
  • the network side device may pre-configure the mapping relationship between the first AI model and the value of N and the value of M, and send the mapping relationship to the terminal. For example, it may be sent synchronously with the first AI model, and then the terminal Based on the first indication information and the mapping relationship, the first AI model to be used can be determined.
  • the network side device may also train the mapping relationship and the first AI model together to include the mapping relationship in the first AI model, and then the terminal obtains the first AI model.
  • the mapping relationship can be obtained based on the first AI model, and the value of N and the value of M can be obtained, thereby selecting a corresponding number of first orthogonal bases. and the second orthonormal basis.
  • the terminal determines the number of the first orthogonal base and the second orthogonal base based on the first instruction information of the network side device, and selects the orthogonal base based on the number of instructions indicated by the network side device. This can improve the terminal's accuracy of orthogonal bases. Flexibility in cross-base selection.
  • the method further includes:
  • the terminal reports the positions of the N first orthogonal bases and/or the positions of the M second orthogonal bases to the network side device.
  • the network side device does not indicate to the terminal the matching relationship between the first AI model and the positions of the N first orthogonal bases and the M second orthogonal bases, or that the N first The orthogonal base and the M second orthogonal bases are not determined by the terminal based on instructions from the network side device, but are determined by the terminal itself.
  • the terminal also needs to report the N first orthogonal bases to the network side device. and/or the positions of the M second orthogonal bases.
  • the terminal may report the indexes of the N first orthogonal bases and/or the indexes of the M second orthogonal bases to the network side device, where the indexes can represent the positions of the corresponding orthogonal bases. In this way, the network side device can accurately know which first orthogonal base and second orthogonal base are selected by the terminal, which is more helpful for the network side device to recover the channel matrix.
  • the terminal reports the positions of the N first orthogonal bases and/or the positions of the M second orthogonal bases to the network side device, including:
  • the terminal reports the positions of the N first orthogonal bases and/or the positions of the M second orthogonal bases to the network side device through CSI.
  • the terminal may report the indexes of the N first orthogonal bases and/or the indexes of the M second orthogonal bases to the network side device through the first part of the CSI (CSI part 1), thereby saving money. Terminal reporting resources.
  • the terminal does not need to report the selected value to the network side device.
  • the positions of the N first orthogonal bases and the positions of the M second orthogonal bases match the first AI model, the terminal does not need to report the selected value to the network side device.
  • N first positive Cross bases and M second orthonormal bases can also include:
  • the terminal obtains the airspace orthogonal bases reported through CSI, and determines N first orthogonal bases from the airspace orthogonal bases reported through CSI;
  • the terminal determines M second orthogonal bases by itself.
  • the first orthogonal basis is selected from the air domain orthogonal basis reported by the terminal through CSI, and the second orthogonal basis may be determined by the terminal itself.
  • the selection range of the first orthogonal basis is limited, but the selection range of the second orthogonal basis is not limited, making the terminal more flexible and autonomous in selecting the second orthogonal basis.
  • the airspace orthogonal base may adopt the reporting method in the R16 codebook.
  • the airspace orthogonal base may be included in i1,1 and i1,2 of the CSI reporting content.
  • the terminal may also perform projection processing and weighting processing on the channel matrix of each subband based on the M second orthogonal bases, so that the dimension of the channel matrix after weighting processing is consistent with the channel matrix before projection processing.
  • the method when the terminal determines M second orthogonal bases by itself, the method further includes:
  • the terminal indicates the M second orthogonal bases to the network side device.
  • the terminal may report the indexes of the M second orthogonal bases to the network side device, thereby indicating the positions of the M second orthogonal bases.
  • the network side device can accurately know the positions of the M second orthogonal bases to better recover the channel matrix.
  • the N first orthogonal bases are selected from the air domain orthogonal bases reported by the terminal through CSI, and the network side device can determine the first orthogonal base based on the received CSI, that is, the terminal may not The positions of the N first orthogonal bases need to be reported to save reporting resources of the terminal.
  • the terminal inputs the preprocessed channel matrix into the first AI model, and obtains the channel characteristic information output by the first AI model, including:
  • the terminal inputs the preprocessed channel matrix into the target first AI model and obtains the channel characteristic information output by the first AI model.
  • the target first AI model matches the subband corresponding to the input channel matrix.
  • each subband corresponds to a target first AI model, and each subband independently processes the preprocessed channel matrix through the corresponding target first AI model to obtain corresponding channel characteristic information.
  • this method can avoid confusion between the processing of channel matrices of different subbands, thereby ensuring the accuracy of the output channel characteristic information.
  • the target first AI model corresponding to each subband is the same or different.
  • the target first AI model corresponding to each subband is the same.
  • the network side device trains the target first AI model, it can mix the channel matrices of all subbands in the same training set for training, that is, the channel matrices of all subbands are used as the target first during training.
  • the input of the AI model, and then the trained target first AI model can process all sub-band channel matrices. In this way, the training cost and time of the network-side device for the target first AI model can be effectively saved.
  • the target first AI model corresponding to each subband is the same, which does not mean that all subbands use the same target first AI model, but that each subband uses the same target first AI model.
  • the AI model processes the input channel matrix separately.
  • the target first AI model corresponding to each subband is different.
  • the channel matrices of certain subbands can be mixed in a training set for training, so as to save the network side device's training cost for the target first AI model. and time.
  • the terminal reports the channel characteristic information to the network side device, including any of the following:
  • the terminal reports the channel characteristic information to the network side device through CSI;
  • the terminal receives the signaling sent by the network side device, and reports the channel characteristic information to the network side device based on the signaling;
  • the terminal reports the channel characteristic information to the network side device based on the configuration information in the CSI report.
  • the terminal may report the channel characteristic information output by the first AI model processing to the network side device through CSI.
  • the channel characteristic information may be included in the first part of the CSI (CSI part 1).
  • the network side device may configure the channel characteristic information in the CSI report in a reporting manner, and the terminal reports the channel characteristic information to the network side device based on the configuration information in the CSI report. For example, the terminal may report the channel characteristic information through CSI.
  • the network side device triggers the terminal to report channel characteristic information through signaling, and the terminal reports the channel characteristic information based on the signaling sent by the network side device.
  • the signaling sent by the network side device includes at least one of the following:
  • DCI Downlink Control Information
  • MAC CE Media Access Control Element
  • Radio Resource Control (RRC).
  • the terminal reports the channel characteristic information to the network side device, including:
  • the terminal performs quantization processing on the channel characteristic information, and reports the quantized channel characteristic information to the network side device.
  • the terminal may perform quantification processing on the channel characteristic information through a quantization table, and report the quantized channel characteristic information to the network side.
  • the network side device inputs the quantized channel characteristic information into the second AI model, and restores the channel matrix through the second AI model.
  • the quantization table and the first AI model may be independent, that is, the quantization table does not participate in the training of the first AI model, or the quantization table may also participate in the training of the first AI model, but will not be used by the first AI model. Affected by gradient information in AI model training.
  • the terminal inputs the preprocessed channel matrix into the first AI model, and obtains the channel characteristic information output by the first AI model, including:
  • the terminal inputs the preprocessed channel matrix into the first AI model, and obtains the channel characteristic information output by the first AI model after quantization processing.
  • the terminal pairs the channel of each subband based on N first orthogonal bases and M second orthogonal bases.
  • the terminal inputs the preprocessed channel matrix to the first AI model, and obtains the channel characteristic information output by the first AI model after quantization processing.
  • the first AI model includes a quantization structure, and the quantization processing is not independent of the first AI model.
  • the first AI model outputs channel characteristic information in binary bits
  • the network side device inputs the channel characteristic information in binary bits into the second AI model, and the channel matrix is restored through the second AI model.
  • the first AI model includes a quantization method, such as a quantization table, and the quantization table is jointly trained with the first AI model to ensure that the trained first AI model can implement quantization processing of the input channel matrix.
  • a quantization method such as a quantization table
  • the terminal reports CSI according to the R16 codebook, estimates the channel matrix on the N SB subband, selects the spatial domain orthogonal basis W1 and the frequency domain orthogonal basis Wf, calculates the corresponding coefficient W2 and reports it.
  • the network side device triggers full channel information reporting.
  • the terminal uses W1 as the first orthogonal basis to obtain the equivalent channel matrix.
  • the equivalent channel matrix in the i-th subband is:
  • the terminal constructs the second orthogonal basis matrix Where Nr is the number of terminal antennas, and k is the number of oversampling groups.
  • the terminal calculates the projection coefficient corresponding to each orthogonal basis in each sub-band.
  • the calculation formula is as follows:
  • the terminal selects larger M orthogonal bases in the same orthogonal group to form an orthogonal matrix.
  • the equivalent channel matrix is weighted and the formula is as follows:
  • the weighted channel matrix is the input of the first AI model, and the terminal will Input it into the first AI model, obtain the channel characteristic information output by the first AI model, and report it to the network side device.
  • the network side device After receiving the channel characteristic information of each subband, the network side device uses the corresponding second AI model to obtain the restored channel matrix.
  • the network side device restores the complete channel matrix of each subband based on W1 reported by the R16 codebook:
  • the network side device can recover the full channel matrix of each subband through the second AI model, that is, The channel matrix of the receiving end and the transmitting end can be known, so as to better predict the channel information at a certain time in the future and improve the accuracy of channel information prediction.
  • Figure 3 is a flow chart of another channel matrix processing method provided by an embodiment of the present application. As shown in Figure 3, the method includes the following steps:
  • Step 301 The network side device receives the channel characteristic information reported by the terminal;
  • Step 302 The network side device inputs the channel characteristic information into the second AI model, and obtains the channel matrix output by the second AI model.
  • the channel characteristic information is the output of the first AI model of the terminal
  • the input of the first AI model is that the terminal uses N first orthogonal bases and M second orthogonal bases to pair each sub
  • the channel matrix of the band is preprocessed.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • N and M are both positive integers.
  • the terminal determines the N first orthogonal bases and the M second orthogonal bases, it determines the N first orthogonal bases and the M second orthogonal bases for each subband.
  • the channel matrix is preprocessed, and the preprocessed channel matrix of each subband is input into the first AI model to obtain the channel characteristic information output by the first AI model.
  • the terminal reports the channel characteristic information to the network side device.
  • the network side device processes the channel characteristic information through the second AI model and outputs the channel matrix, so that the network side device restores the channel matrix through the second AI model. In this way, terminals and network-side devices can process and restore the channel matrix through the AI model, which can effectively save transmission resources.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • the channel characteristic information obtained by the terminal through the first AI model processing is also considered at the same time.
  • the network side equipment performs channel matrix recovery through the second AI model, and thus can obtain the full channel information of the receiving end and transmitting end. In this way, the network side device can not only obtain the channel information of the sending end, but also the channel information of the receiving end, which further helps the network side device to predict the channel information to improve the accuracy of channel information prediction.
  • the method also includes:
  • the network side device sends first indication information to the terminal, where the first indication information is used to indicate the value of N and the value of M.
  • the method also includes:
  • the network side device receives the indexes of the N first orthogonal bases, the indexes of the M second orthogonal bases, and the index of the oversampling group reported by the terminal, where the N first orthogonal bases and the M second orthogonal bases are orthogonal bases determined by the terminal from the oversampling group after oversampling processing.
  • the value of N and the value of M match the first AI model.
  • the method further includes:
  • the network side device receives the positions of the N first orthogonal bases and/or the M second orthogonal bases reported by the terminal. The position of the orthogonal basis.
  • the network side device receives the positions of the N first orthogonal bases and/or the positions of the M second orthogonal bases reported by the terminal, including:
  • the network side device receives the positions of the N first orthogonal bases and/or the positions of the M second orthogonal bases reported by the terminal through CSI.
  • the method further includes:
  • the network side device receives second indication information reported by the terminal, where the second indication information is used to indicate the positions of the M second orthogonal bases.
  • the network side device receives channel characteristic information reported by the terminal, including any of the following:
  • the network side device receives the channel characteristic information reported by the terminal through CSI;
  • the network side device sends signaling to the terminal, and receives the channel characteristic information reported by the terminal based on the signaling;
  • the network side device receives the channel characteristic information reported by the terminal through the configuration information in the CSI report.
  • the signaling includes at least one of the following:
  • the network side device receives channel characteristic information reported by the terminal, including:
  • the network side device receives the quantized channel characteristic information of the terminal.
  • channel matrix processing method provided by the embodiments of the present application is executed by a network-side device and corresponds to the channel matrix processing method executed by the terminal.
  • the relevant concepts and specific implementation processes involved in the embodiments of the present application may be: Referring to the description in the method embodiment described above in Figure 2, to avoid repetition, the details will not be described again here.
  • the execution subject may be a channel matrix processing device.
  • the channel matrix processing device performing the channel matrix processing method is taken as an example to illustrate the channel matrix processing device provided by the embodiment of the present application.
  • Figure 4 is a structural diagram of a channel matrix processing device provided by an embodiment of the present application.
  • the channel matrix processing device 400 includes:
  • Processing module 401 configured to determine N first orthogonal bases and M second orthogonal bases, and calculate the channel of each subband based on the N first orthogonal bases and the M second orthogonal bases.
  • the matrix is preprocessed, the first orthogonal basis corresponds to the airspace information of the transmitting end antenna, the second orthogonal base corresponds to the airspace information of the receiving end antenna, N and M are both positive integers;
  • the first acquisition module 402 is used to input the preprocessed channel matrix into the first artificial intelligence AI model and obtain the channel characteristic information output by the first AI model;
  • the reporting module 403 is used to report the channel characteristic information to the network side device.
  • processing module 401 is also used to perform any of the following:
  • N first orthogonal bases and M second orthogonal bases are determined based on the first AI model.
  • the N first orthogonal bases The positions of the basis and the positions of the M second orthogonal basis match the first AI model;
  • reporting module 403 is also used to:
  • the second orthogonal basis is an orthogonal basis determined from the oversampling group by the device after oversampling.
  • the value of N and the value of M match the first AI model.
  • the reporting module 403 is also used to:
  • reporting module 403 is also used to:
  • the positions of the N first orthogonal bases and/or the positions of the M second orthogonal bases are reported to the network side device through CSI.
  • processing module 401 is also used to:
  • the device is also used for:
  • the first acquisition module 402 is also used to:
  • the preprocessed channel matrix is input into the target first AI model, and the channel characteristic information output by the target first AI model is obtained.
  • the target first AI model matches the subband corresponding to the input channel matrix.
  • the target first AI model corresponding to each subband is the same or different.
  • processing module 401 is also used to perform any of the following:
  • the N first orthogonal bases are based on corresponding
  • the projection coefficients are sorted according to the first preset order, and the M second orthogonal bases are sorted according to the second preset order based on the corresponding projection coefficients.
  • reporting module 403 is also used to perform any of the following:
  • the signaling includes at least one of the following:
  • reporting module 403 is also used to:
  • the channel characteristic information is quantized, and the quantized channel characteristic information is reported to the network side device.
  • the first acquisition module 402 is also used to:
  • the terminal inputs the preprocessed channel matrix into the first AI model, and obtains the channel characteristic information output by the first AI model after quantization processing.
  • the device can process and recover the channel matrix through the AI model with the network side equipment, which can effectively save transmission resources.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • the channel characteristic information obtained by the device through the first AI model processing is also simultaneously Taking into account the receiving end and the transmitting end, the network side device performs channel matrix recovery through the second AI model, so that the full channel information of the receiving end and the transmitting end can be obtained. In this way, the network side device can not only obtain the channel information of the sending end, but also the channel information of the receiving end, which further helps the network side device to predict the channel information to improve the accuracy of channel information prediction.
  • the channel matrix processing device 400 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 matrix processing device 400 provided by the embodiment of the present application can implement each process implemented by the method embodiment described in Figure 2 and achieve the same technical effect. To avoid duplication, the details will not be described here.
  • Figure 5 is a structural diagram of another channel matrix processing device provided by an embodiment of the present application. As shown in Figure 5, the channel matrix processing device 500 includes:
  • the receiving module 501 is used to receive channel characteristic information reported by the terminal;
  • the second acquisition module 502 is used to input the channel characteristic information into the second AI model and obtain the channel matrix output by the second AI model;
  • the channel characteristic information is the output of the first AI model of the terminal
  • the input of the first AI model is that the terminal uses N first orthogonal bases and M second orthogonal bases to pair each sub
  • the channel matrix of the band is preprocessed.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • N and M are both positive integers.
  • the device also includes:
  • a sending module configured to send first indication information to the terminal, where the first indication information is used to indicate the value of N and the value of M.
  • the receiving module 501 is also used to:
  • the second orthogonal basis is an orthogonal basis determined by the terminal from the oversampling group after oversampling.
  • the value of N and the value of M match the first AI model.
  • the receiving module 501 is also used to:
  • the receiving module 501 is also used to:
  • the receiving module 501 is also used to:
  • the receiving module 501 is also used to perform any of the following:
  • the signaling includes at least one of the following:
  • the receiving module 501 is also used to:
  • the network side device receives the quantized channel characteristic information of the terminal.
  • the device processes the channel characteristic information through the second AI model and outputs the channel matrix, so that the device restores the channel matrix through the second AI model.
  • the terminal and the device can process and recover the channel matrix through the AI model, which can effectively save transmission resources.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • the channel characteristic information obtained by the terminal through the first AI model processing is also considered at the same time.
  • the device performs channel matrix recovery through the second AI model, so that the full channel information of the receiving end and the transmitting end can be obtained. In this way, the device can not only obtain the channel information of the transmitting end, but also obtain the channel information of the receiving end. Helps the device predict channel information to improve the accuracy of channel information prediction.
  • the channel matrix processing device 500 provided by the embodiment of the present application can implement each process implemented by the method embodiment described in Figure 3 and achieve the same technical effect. To avoid duplication, the 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, such as , when the communication device 600 is a terminal, when the program or instruction is executed by the processor 601, each step of the method embodiment described in Figure 2 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 method embodiment described in FIG. 3 is implemented, and the same technical effect can be achieved. To avoid duplication, the details will not be described here.
  • Embodiments of the present application also provide a terminal, including a processor and a communication interface.
  • the processor is configured to determine N first orthogonal bases and M second orthogonal bases, and determine N first orthogonal bases based on the N first orthogonal bases.
  • the orthogonal base and the M second orthogonal bases preprocess the channel matrix of each subband.
  • the first orthogonal base corresponds to the spatial information of the transmitting end antenna
  • the second orthogonal base corresponds to the receiving end antenna.
  • airspace information, N and M are both positive integers; and for inputting the preprocessed channel matrix into the first artificial intelligence AI model to obtain the channel characteristic information output by the first AI model; the communication interface is used for Report the channel characteristic information to the network side device.
  • 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 processing unit (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.
  • 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 link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM 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
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM synchronous link dynamic random access memory
  • SLDRAM direct memory bus
  • 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.
  • processor 710 is used for:
  • N first orthogonal bases and M second orthogonal bases and preprocess the channel matrix of each subband based on the N first orthogonal bases and the M second orthogonal bases, so
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna
  • N and M are both positive integers
  • the radio frequency unit 701 is used to report the channel characteristic information to the network side device.
  • the terminal 700 and the network side device can process and restore the channel matrix through the AI model, which can effectively save transmission resources.
  • the first orthogonal basis corresponds to the airspace information of the transmitting end antenna
  • the second orthogonal base corresponds to the airspace information of the receiving end antenna.
  • the channel characteristic information obtained by the terminal 700 through the first AI model processing is also considered at the same time.
  • the network side device After identifying the receiving end and transmitting end, the network side device performs channel matrix recovery through the second AI model, and can obtain the full channel information of the receiving end and transmitting end. In this way, the network side device can not only obtain the channel information of the sending end, but also the channel information of the receiving end, which further helps the network side device to predict the channel information to improve the accuracy of channel information prediction.
  • 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 channel characteristic information reported by the terminal; the processor is used to input the channel characteristic information into the second AI model. , obtain the channel matrix output by the second AI model; wherein the channel characteristic information is the first AI of the terminal
  • the output of the model, the input of the first AI model is the channel matrix after the terminal preprocesses the channel matrix of each subband through N first orthogonal bases and M second orthogonal bases, and the first One orthogonal base corresponds to the air domain information of the transmitting end antenna, and the second orthogonal base corresponds to the air domain information of the receiving end antenna.
  • N and M are both positive integers.
  • 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 the embodiment of the present application 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 5. 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 method embodiment described in Figure 2 is implemented, or Each process of the method embodiment described in Figure 3 above can achieve the same technical effect. To avoid repetition, it will not be described again here.
  • 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 method described in Figure 2.
  • 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.
  • Embodiments of the present application further provide a computer program/program product.
  • the computer program/program product is stored in a storage medium.
  • the computer program/program product is executed by at least one processor to implement the method described in Figure 2 above.
  • Embodiments of the present application also provide a communication system, including: a terminal and a network side device.
  • the terminal can be used to perform the steps of the channel matrix processing method as shown in Figure 2.
  • the network side device can be used to perform the steps of the channel matrix processing method as shown in Figure 3. The steps of the channel matrix processing 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 related technologies.
  • 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)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种信道矩阵处理方法、装置、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矩阵处理方法包括：终端确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；所述终端将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；所述终端向网络侧设备上报所述信道特征信息。

Description

信道矩阵处理方法、装置、终端及网络侧设备
相关申请的交叉引用
本申请主张在2022年06月22日在中国提交的中国专利申请No.202210716334.3的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于通信技术领域，具体涉及一种信道矩阵处理方法、装置、终端及网络侧设备。
背景技术
相关技术中的通信系统中，基站在某个时隙的某些时频资源上发送信道状态信息参考信号(Channel State Information Reference Signal，CSI-RS)，终端根据CSI-RS进行信道估计，计算这个时隙上的信道信息，通过码本将预编码矩阵指示(Precoding Matrix Indicator，PMI)反馈给基站，基站根据终端反馈的码本信息组合出信道信息，在下一次CSI上报之前，基站以此进行数据预编码以及多用户调度。
目前，终端反馈的码本内容是信道矩阵的特征矩阵，也即仅反馈发送端的预编码矩阵，但是对于一些场景(例如高速场景)下，基站仅基于发送端的预编码矩阵进行信道信息预测，导致信道信息预测的准确度较低。
发明内容
本申请实施例提供一种信道矩阵处理方法、装置、终端及网络侧设备，能够解决相关技术中网络侧设备对信道信息预测的准确性较低的问题。
第一方面，提供了一种信道矩阵处理方法，包括：
终端确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；
所述终端将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；
所述终端向网络侧设备上报所述信道特征信息。
第二方面，提供了一种信道矩阵处理方法，包括：
网络侧设备接收终端上报的信道特征信息；
所述网络侧设备将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输 出的信道矩阵；
其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
第三方面，提供了一种信道矩阵处理装置，包括：
处理模块，用于确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；
第一获取模块，用于将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；
上报模块，用于向网络侧设备上报所述信道特征信息。
第四方面，提供了一种信道矩阵处理装置，包括：
接收模块，用于接收终端上报的信道特征信息；
第二获取模块，用于将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵；
其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
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信道矩阵处理方法的步骤。
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N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；以及用于将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；所述通信接口用于向网络侧设备上报所述信道特征信息。
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一方面所述的信道矩阵处理方法的步骤。
第八方面，提供了一种网络侧设备，包括处理器及通信接口，所述通信接口用于接收终端上报的信道特征信息；所述处理器用于将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵；
其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
第九方面，提供了一种通信系统，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矩阵处理方法的步骤，所述网络侧设备可用于执行如第二方面所述的信道矩阵处理方法的步骤。
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信道矩阵处理方法的步骤，或者实现如第二方面所述的信道矩阵处理方法的步骤。
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信道矩阵处理方法，或实现如第二方面所述的信道矩阵处理方法。
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信道矩阵处理方法，或实现如第二方面所述的信道矩阵处理方法。
在本申请实施例中，终端和网络侧设备能够通过AI模型来实现对信道矩阵的处理和恢复，能够有效节省传输资源。并且，第一正交基对应的是发送端天线的空域信息，第二正交基对应的是接收端天线的空域信息，进而终端通过第一AI模型处理得到的信道特征信息也就同时考虑了接收端和发送端，网络侧设备通过第二AI模型进行信道矩阵恢复，也就能够得到接收端和发送端的全信道信息。这样，也就使得网络侧设备不仅能够得到发送端的信道信息，还能够获得接收端的信道信息，更有助于网络侧设备对信道信息进行预测，以提升信道信息预测的准确性。
附图说明
图1是本申请实施例可应用的一种无线通信系统的框图；
图2是本申请实施例提供的一种信道矩阵处理方法的流程图；
图3是本申请实施例提供的另一种信道矩阵处理方法的流程图；
图4是本申请实施例提供的一种信道矩阵处理装置的结构图；
图5是本申请实施例提供的另一种信道矩阵处理装置的结构图；
图6是本申请实施例提供的一种通信设备的结构图；
图7是本申请实施例提供的一种终端的结构图；
图8是本申请实施例提供的一种网络侧设备的结构图。
具体实施方式
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(Vehicle User Equipment，VUE)、行人终端(Pedestrian User Equipment，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。接入网设备可以包括基站、无线局域网(Wireless Local Area Network，WLAN)接入点、无线保真(Wireless Fidelity，WiFi)节点等，基站可被称为 节点B、演进节点B(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。
为更好地理解本申请实施例提供的技术方案，以下对本申请实施例中可能涉及的相关概念及原理进行解释说明。
由信息论可知，准确的信道状态信息(Channel State Information，CSI)对信道容量的至关重要。尤其是对于多天线系统来讲，发送端可以根据CSI优化信号的发送，使其更加匹配信道的状态。例如：信道质量指示(Channel Quality Indicator，CQI)可以用来选择合适的调制编码方案(Modulation and Coding Scheme，MCS)实现链路自适应；预编码矩阵指示(Precoding Matrix Indicator，PMI)可以用来实现特征波束成形(eigen beamforming)从而最大化接收信号的强度，或者用来抑制干扰(如小区间干扰、多用户之间干扰等)。因此，自从多天线技术(multi-input multi-output，MIMO)被提出以来，CSI获取一直都是研究热点。
通常，基站在在某个时隙(slot)的某些时频资源上发送信道状态信息参考信号(Channel State Information Reference Signal，CSI-RS)，终端根据CSI-RS进行信道估计，计算这个slot上的信道信息，通过码本将PMI反馈给基站，基站根据终端反馈的码本信息组合出信道信息，在下一次CSI上报之前，基站以此进行数据预编码及多用户调度。
为了进一步减少CSI反馈开销，终端可以将每个子带上报PMI改成按照延迟(delay)上报PMI，由于delay域的信道更集中，用更少的delay的PMI就可以近似表示全部子带的PMI，即将delay域信息压缩之后再上报。
同样，为了减少开销，基站可以事先对CSI-RS进行预编码，将编码后的CSI-RS发送给终端，终端看到的是经过编码之后的CSI-RS对应的信道，终端只需要在网络侧指示的端口中选择若干个强度较大的端口，并上报这些端口对应的系数即可。
进一步地，为了更好地压缩信道信息，可以使用神经网络或机器学习的方法。
人工智能目前在各个领域获得了广泛的应用。人工智能(Artificial Intelligence，AI)模块有多种实现方式，例如神经网络、决策树、支持向量机、贝叶斯分类器等。本申请以神经网络为例进行说明，但是并不限定AI模型的具体类型。
具体地，在终端通过AI模型对信道信息进行压缩编码，在基站通过AI模型对压缩后的内容进行解码，从而恢复信道信息，此时基站的用于解码的AI模型和终端的用于编码的AI模型需要联合训练，达到合理的匹配度。通过终端的用于编码的AI模型和基站的用于解码的AI模型组成联合的神经网络模型，由网络侧进行联合训练，训练完成后，基站将用于编码的AI模型发送给终端。
例如，终端估计CSI-RS，计算信道信息，将计算的信道信息或者原始的估计到的信道信息通过用于编码的AI模型得到编码结果，将编码结果发送给基站，基站接收编码后的结果，输入到解码AI模型中，恢复信道信息。
信道估计的时候，终端在每个子带上会得到一个估计的信道矩阵，例如一个4×32的信道矩阵，表示4个接收天线和32个CSI-RS端口，每个子带上的信道矩阵不同，由于频率选择性衰落，两个自带的信道矩阵可能有很大的差别，传统的R15码本就是分别反馈每个子带的信道信息。
但是连续多个自带的信道矩阵之间具有一定的关系，传统的R16码本就是利用这个关系将频域信道转换到时延域，从而获得能量集中的信道矩阵，只上报能量强的部分时延对应的信道信息就可以获得较为完整的信道信息。
基于码本的CSI反馈方案多是针对信道预编码矩阵进行反馈的，即信道矩阵的特征矩阵，这主要是为了节约开销，上报的CSI主要用于基站做预编码使用，所以不需要完整的信道信息，只需要信道的发送端的特征信息即可，因此只上报信道特征矩阵可以减少开销且不影响性能。但是随着高速场景的日益普及，基站需要对未来某个时刻的信道信息进行预测，以便更好的调度用户，因此基站需要完整的信道信息。
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矩阵处理方法进行详细地说明。
请参照图2，图2是本申请实施例提供的一种信道矩阵处理方法的流程图，如图2所示，所述方法包括以下步骤：
步骤201、终端确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息。
其中，N和M均为正整数。
需要说明的是，在一些通信场景下，所述发送端可以是指网络侧设备(也称基站)，所述接收端可以是指终端；或者，在另一些通信场景下，所述发送端可以是指终端，所述接收端可以是指网络侧设备。
本申请实施例中，终端可以是在候选正交基中进行选择，以确定N个第一正交基和M个第二正交基，并将所述N个第一正交基对应于发送端天线的空域信息，将M个第二正交基对应于接收端天线的空域信息。这样，也就使得终端能够针对发送端和接收端都确定出对应的正交基，以通过正交基对发送端和接收端的空域信息都进行处理。其中，所述候选正交基的获取可以是参照相关技术，本申请对此不做赘述。
可选地，所述基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，包括如下任意一项：
基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理，投影处理后的信道矩阵的维度为N×M或者M×N；
基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理，并对投影处理后的信道矩阵进行加权处理，以使得加权处理后的信道矩阵与投影处理前的信道矩阵维度一致，所述加权处理使用的正交基为所述N个第一正交基和所述M个第二正交基。
例如，在一种实施方式中，终端获取每个子带预估的信道矩阵，在确定出N个第一正交基和M个第二正交基后，将每个子带预估的信道矩阵投影至N个第一正交基和M个第二正交基上，得到每个子带投影处理后的维度为N×M或M×N的信道矩阵，并可以计算投影处理后的信道矩阵的投影系数。
需要说明地，在基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理的情况下，所述N个第一正交基基于对应的投影系数按照第一预设顺序进行排序，所述M个第二正交基基于对应的投影系数按照第二预设顺序进行排序。
例如，所述N个第一正交基可以是基于对应的投影系数按照从大到小的顺序进行排序，所述M个第二正交基也可以是基于对应的投影系数按照从大到小的顺序进行排序，进而得到的投影后的信道矩阵中的第一行第一列的投影系数为最强的第一正交基和最强的第二正交基对应的投影系数。这种情况下，终端可以不需要向网络侧设备上报第一正交基和第二正交基的排序方式。
在另一种实施方式中，终端获取每个子带预估的信道矩阵，在确定出N个第一正交基和M个第二正交基后，将每个子带预估的信道矩阵投影至N个第一正交基和M个第二正交基上，得到每个子带投影处理后的维度为N×M或M×N的信道矩阵，然后基于所述N个第一正交基和所述M个第二正交基对维度为N×M或M×N的信道矩阵进行加权处理，使得加权处理后的信道矩阵与投影处理前的信道矩阵维度一致。
步骤202、所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型输出的信道特征信息。
例如，终端将经过每个子带投影处理后的信道矩阵输入至第一AI模型中，或者可以是将每个子带经过投影处理以及加权处理后的信道矩阵输入至第一AI模型中。进一步地，所述第一AI模型对输入的所述信道矩阵进行编码处理，以获取所述第一AI模型输出的每个子带对应的信道特征信息。
需要说明的是，所述第一AI模型为预先训练后得到的AI模型，该训练后的第一AI模型能够基于自身的网络结构及网络参数对输入的信道矩阵进行处理，并输出信道特征信息。所述第一AI模型的训练方法可以是参照相关技术中网络模型的训练方式，本实施例对此不做赘述。
步骤203、所述终端向网络侧设备上报所述信道特征信息。
本申请实施例中，终端在获得所述第一AI模型输出的信道特征信息后，将所述信道特征信息上报给网络侧设备。网络侧设备包括与所述第一AI模型匹配的第二AI模型，网络侧设备将所述信道特征信息输入所述第二AI模型，通过所述第二AI模型对输入的信道 特征信息进行解码处理，恢复出完整的信道矩阵并输出。
需要说明的是，所述第一AI模型和所述第二AI模型相互匹配，第一AI模型的输出为第二AI模型的输入，第二AI模型的输出能够尽量贴近第一AI模型的输入，进而以也就使得网络侧设备能够通过第二AI模型实现信道矩阵的恢复。其中，所述第一AI模型和所述第二AI模型可以是由网络侧设备联合训练得到，并由网络侧设备将训练好的第一AI模型发送给终端，以使得所述终端能够通过第一AI模型将输入的信道矩阵处理成信道特征信息而输出。
本申请实施例中，终端确定N个第一正交基和M个第二正交基后，基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，并将每个子带预处理后的信道矩阵输入至第一AI模型中，获取所述第一AI模型输出的信道特征信息，终端将所述信道特征信息上报给网络侧设备，网络侧设备通过第二AI模型对信道特征信息进行处理，输出信道矩阵，从而网络侧设备通过第二AI模型恢复信道矩阵。这样，也就使得终端和网络侧设备能够通过AI模型来实现对信道矩阵的处理和恢复，能够有效节省传输资源。
另外，第一正交基对应的是发送端天线的空域信息，第二正交基对应的是接收端天线的空域信息，进而终端通过第一AI模型处理得到的信道特征信息也就同时考虑了接收端和发送端，网络侧设备通过第二AI模型进行信道矩阵恢复，也就能够得到接收端和发送端的全信道信息。这样，也就使得网络侧设备不仅能够得到发送端的信道信息，还能够获得接收端的信道信息，更有助于网络侧设备对信道信息进行预测，以提升信道信息预测的准确性。
本申请实施例中，所述终端确定N个第一正交基和M个第二正交基，包括如下任意一项：
所述终端基于所述第一AI模型确定N个第一正交基和M个第二正交基，所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型匹配；
所述终端接收网络侧设备发送的第一指示信息，并基于所述第一指示信息确定N个第一正交基和M个第二正交基，所述第一指示信息用于指示所述N的取值和所述M的取值。
例如，在一种实施方式中，网络侧设备可以是指定或预先配置所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型匹配，并将该匹配关系指示给终端，进而终端基于网络侧设备指示的匹配关系来确定N个第一正交基和M个第二正交基。或者，网络侧设备也可以是在向终端发送第一AI模型的时候，一起将该匹配关系发送给终端。
可选地，网络侧设备还可以是在训练第一AI模型的时候，同步训练所述第一AI模型与所述N个第一正交基的位置和所述M个第二正交基的位置之间的匹配关系。
需要说明的是，该实施方式下，终端确定了第一AI模型，也即确定了对应的N个第一正交基和M个第二正交基；或者，终端确定了N个第一正交基和M个第二正交基，也 即确定了对应的第一AI模型。由于第一AI模型是网络侧设备训练并发送给终端的，这种情况下，终端不需要向网络侧设备上报第一正交基的索引和第二正交基的索引，网络侧设备基于终端所使用的第一AI模型既可获知终端所选择的第一正交基和第二正交基。这样，能够节省终端的传输资源，也使得终端无需对第一正交基和第二正交基进行选择，进而也能够简化终端的对正交基的选择流程。
在另一种实施方式中，网络侧设备通过第一指示信息来指示N和M的取值，进而终端基于所述第一指示信息来确定N个第一正交基和M个第二正交基。需要说明的是，网络侧设备仅是指示的正交基的数量，并没有指示具体是哪些位置的正交基。
可选地，这种情况下，所述方法还包括：
所述终端向网络侧设备上报所述N个第一正交基的索引和所述M个第二正交基的索引以及过采样组的索引，其中，所述N个第一正交基和M个第二正交基为所述终端经过过采样处理后从过采样组中确定的正交基。
可以理解地，在网络侧设备通过第一指示信息仅指示了第一正交基的数量和第二正交基的数量的情况下，终端可以是对候选正交基进行过采样处理，以得到候选正交基过采样组，终端从过采样组中选择N个第一正交基和M个第二正交基，并向网络侧设备上报所述N个第一正交基的索引和M个第二正交基各自的索引，以及上报进行选择的过采样组的索引。这样，也就使得网络侧设备能够准确获知终端所选择的过采样组和所选择的正交基具体是哪些，有助于网络侧设备准确地恢复信道矩阵。
可选地，终端对候选正交基进行过采样处理并从过采样组中确定出N个第一正交基和M个第二正交基，具体可以是包括如下步骤：
所述终端确定第一候选正交基和第二候选正交基，并对所述第一候选正交基和所述第二候选正交基分别进行过采样处理，得到n组第一候选正交基和m组第二候选正交基，n和m均为正整数；
所述终端在j组第一候选正交基中进行选择以得到N个第一正交基，以及在k组第二候选正交基中进行选择以得到M个第二正交基；
其中，所述j组第一候选正交基为所述n组第一候选正交基中的任一组，所述k组第二候选正交基为所述m组第二候选正交基中的任一组，也即j＝1，k＝1。
示例性地，终端可以是先获取第一候选正交基和第二候选正交基，例如可以是确定32个第一候选正交基和32个第二候选正交基，对这32个第一候选正交基和32个第二候选正交基进行过采样处理，例如得到4组32个第一候选正交基和4组32个第二候选正交基，也即4×32个第一候选正交基和4×32个第二候选正交基(也即n＝4，m＝4)；进一步地，终端可以是各自从中选择一组，从其中一组(也即j组)对应的32个第一候选正交基中选择出N个第一正交基，以及从其中一组(也即k组)对应的32个第二候选正交基中选择出M个第二正交基。
这样，终端能够对候选正交基进行过采样处理，从过采样处理得到的过采样组中选择 一组来进行第一正交基和第二正交基的选择，以提升正交基选择的丰富度和选择范围。其中，正交基的过采样处理可以是参照相关技术，本实施例不做具体赘述。
可选地，在网络侧设备通过第一指示信息向终端指示所述N的取值和所述M的取值的情况下，所述N的取值和所述M的取值还可以是与所述第一AI模型匹配。例如，网络侧设备可以是预先配置第一AI模型与N的取值及M的取值的映射关系，并将所述映射关系发送给终端，例如可以是与第一AI模型同步发送，进而终端基于所述第一指示信息以及该映射关系，也就能够确定所需要使用的第一AI模型。
可选地，网络侧设备也可以是将所述映射关系与所述第一AI模型一起训练，以将所述映射关系包括进第一AI模型中，进而终端在获取到所述第一AI模型的情况下，也就能够基于所述第一AI模型获取到所述映射关系，也就能够获取所述N的取值和所述M的取值，从而选择出对应数量的第一正交基和第二正交基。
该实施方式下，终端基于网络侧设备的第一指示信息确定了第一正交基和第二正交基的数量，基于网络侧设备指示的数量来选择正交基，这样能够提升终端对于正交基选择的灵活性。
可选地，在所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型不匹配的情况下，所述方法还包括：
所述终端向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。
例如，网络侧设备并没有向终端指示第一AI模型与N个第一正交基的位置和所述M个第二正交基的位置之间的匹配关系，或者说所述N个第一正交基和M个第二正交基并非是终端基于网络侧设备的指示来确定的，而是终端自行确定的情况下，终端还需向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。例如，终端可以是向网络侧设备上报所述N个第一正交基的索引和/或所述M个第二正交基的索引，所述索引能够表征对应的正交基的位置。这样，也就使得网络侧设备能够准确获知终端选择的第一正交基和第二正交基是哪些，更有助于网络侧设备恢复得到信道矩阵。
可选地，所述终端向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置，包括：
所述终端通过CSI向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。
例如，终端可以是通过CSI的第一部分(CSI part 1)来向网络侧设备上报所述N个第一正交基的索引和/或所述M个第二正交基的索引，进而以节省终端的上报资源。
需要说明地，若所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型是匹配的，则终端无需向网络侧设备上报所选择的N个第一正交基的位置和所述M个第二正交基的位置。
或者，上述上报过程也可以是独立于CSI之外。可选地，所述终端确定N个第一正 交基和M个第二正交基，还可以包括：
所述终端获取通过CSI上报的空域正交基，并从所述通过CSI上报的空域正交基中确定出N个第一正交基；
所述终端自行确定M个第二正交基。
在该实施方式中，第一正交基是从终端通过CSI上报的空域正交基中选择出来的，而第二正交基可以是终端自行确定的。这样，也就限定了第一正交基的选择范围，而对第二正交基的选择范围不做限定，使得终端对于第二正交基的选择更具灵活性和自主性。
需要说明的是，所述空域正交基可以采用R16码本中的上报方式，例如所述空域正交基可以是包括在CSI上报内容的i1,1和i1,2中。
可选地，终端还可以是基于所述M个第二正交基对每个子带的信道矩阵进行投影处理和加权处理，使得加权处理后的信道矩阵与投影处理前的信道矩阵维度一致。
本申请实施例，在所述终端自行确定M个第二正交基的情况下，所述方法还包括：
所述终端向网络侧设备指示所述M个第二正交基。
例如，终端可以是向网络侧设备上报所述M个第二正交基的索引，进而以指示所述M个第二正交基的位置。这样，也就使得网络侧设备能够准确获知M个第二正交基的位置，以更好地实现对信道矩阵的恢复。而所述N个第一正交基是从终端通过CSI上报的空域正交基中选择出来的，进而网络侧设备可以是基于接收到的CSI来确定第一正交基，也即终端可以不需要上报所述N个第一正交基的位置，以节省终端的上报资源。
本申请实施例中，所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型输出的信道特征信息，包括：
所述终端将预处理后的信道矩阵输入目标第一AI模型中，获取所述第一AI模型输出的信道特征信息，所述目标第一AI模型与输入的信道矩阵对应的子带匹配。
也就是说，每个子带都各自对应一个目标第一AI模型，每个子带独立通过对应的目标第一AI模型对预处理的信道矩阵进行处理，以获得对应的信道特征信息。相比于用一个AI模型来处理所有子带的信道矩阵，这样的方式能够避免不同子带的信道矩阵处理之间的混乱，进而以确保输出的信道特征信息准确性。
可选地，每个子带对应的所述目标第一AI模型相同或者不同。
例如，每个子带对应的所述目标第一AI模型相同。这种情况下，网络侧设备在训练目标第一AI模型的时候，可以是将所有子带的信道矩阵混合在同一个训练集中进行训练，也即将所有子带的信道矩阵作为训练时目标第一AI模型的输入，进而训练后的目标第一AI模型能够实现对所有子带信道矩阵的处理。这样，能够有效节省网络侧设备对目标第一AI模型的训练成本和时间。
需要说明的是，该实施方式中，每个子带对应的所述目标第一AI模型相同，不代表所有子带使用同一个目标第一AI模型，而是每个子带对应使用相同的目标第一AI模型分别对输入的信道矩阵进行处理。
或者，每个子带对应的所述目标第一AI模型不同。这种情况下，网络侧设备在训练目标第一AI模型的时候，可以是将某些子带的信道矩阵混合在一个训练集中进行训练，以节省网络侧设备对目标第一AI模型的训练成本和时间。
本申请实施例中，所述终端向网络侧设备上报所述信道特征信息，包括如下任意一项：
所述终端通过CSI向网络侧设备上报所述信道特征信息；
所述终端接收网络侧设备发送的信令，并基于所述信令向网络侧设备上报所述信道特征信息；
所述终端基于CSI报告中的配置信息向网络侧设备上报所述信道特征信息。
例如，终端可以是将第一AI模型处理输出的信道特征信息通过CSI上报给网络侧设备，例如所述信道特征信息可以是包括在CSI的第一部分(CSI part 1)中。
或者，网络侧设备可以是在CSI报告中配置所述信道特征信息的上报方式，终端基于所述CSI报告中的配置信息向网络侧设备上报所述信道特征信息。例如，终端可以是通过CSI上报所述信道特征信息。
又或者，网络侧设备通过信令触发所述终端上报信道特征信息，终端基于网络侧设备发送的信令上报所述信道特征信息。
这样，也就使得终端对于信道特征信息的上报方式更加灵活多样。
可选地，所述网络侧设备发送的所述信令包括如下至少一项：
下行控制信息(Downlink Control Information，DCI)；
媒体接入控制控制元素(Medium Access Control Control Element，MAC CE)；
无线资源控制(Radio Resource Control，RRC)。
可选地，所述终端向网络侧设备上报所述信道特征信息，包括：
所述终端对所述信道特征信息进行量化处理，并向网络侧设备上报量化处理后的信道特征信息。
本申请实施例中，终端在获取到第一AI模型输出的信道特征信息后，可以是对所述信道特征信息通过一个量化表进行量化处理，并将量化处理后的信道特征信息上报给网络侧设备，网络侧设备将量化处理后的信道特征信息输入第二AI模型，通过第二AI模型恢复信道矩阵。
其中，所述量化表与所述第一AI模型可以是独立的，也即量化表不参与第一AI模型的训练，或者量化表也可以参与第一AI模型的训练，但是不会被第一AI模型训练中的梯度信息所影响。
可选地，所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型输出的信道特征信息，包括：
所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型量化处理后输出的信道特征信息。
本申请实施例中，终端在基于N个第一正交基和M个第二正交基对每个子带的信道 矩阵预处理后，终端将预处理后的信道矩阵输入至第一AI模型，获取第一AI模型量化处理后输出的信道特征信息。也就是说，第一AI模型是包括量化结构的，量化处理不是独立于第一AI模型之外。该实施方式下，第一AI模型输出二进制比特(bit)的信道特征信息，网络侧设备将该二进制bit的信道特征信息输入第二AI模型，通过第二AI模型恢复信道矩阵。
其中，所述第一AI模型包括量化方式，例如量化表，所述量化表与第一AI模型联合训练，进而以确保训练后的第一AI模型能够实现对输入的信道矩阵的量化处理。
为更好地理解本申请实施例的技术方案，以下通过一个具体的实施例来进行解释说明。
实施例一
终端根据R16码本的方式上报CSI，在NSB个子带上估计信道矩阵，并且选择空域正交基W1和频域正交基Wf，计算对应的系数W2并上报。
某个时刻网络侧设备触发全信道信息上报，终端使用W1作为第一正交基，得到等效信道矩阵，在第i个子带的等效信道矩阵为：
其中，为第i个子带的等效信道矩阵，Hi为第i个子带的信道矩阵，W1为第一正交基。
终端构建第二正交基矩阵其中Nr是终端天线数量，k是过采样组数量。
终端在每个子带计算每个正交基对应的投影系数，计算公式如下：
其中，为第i个子带对应的投影系数，为对第二正交基矩阵中第j列的正交基进行共轭转置，为第i个子带的等效信道矩阵。
终端在同一个正交组中选择较大的M个正交基，组成正交矩阵对等效信道矩阵进行加权处理，公式如下：
其中，为加权处理后的信道矩阵，为第i个子带对应的投影系数，为对进行共轭转置，为第i个子带的等效信道矩阵。
加权处理后的信道矩阵即为第一AI模型的输入，终端在每个子带将输入到第一AI模型中，获得第一AI模型输出的信道特征信息并上报给网络侧设备。
网络侧设备接收每个子带的信道特征信息之后，使用对应的第二AI模型得到恢复的信道矩阵网络侧设备根据R16码本上报的W1，恢复每个子带的完整的信道矩阵：
其中，为第i个子带恢复的完整的信道矩阵，为第二AI模型恢复的第i个子带的信道矩阵，为对空域正交基进行共轭转置。
这样，网络侧设备也就能够通过第二AI模型恢复得到每个子带的全信道矩阵，也即 能够获知接收端和发送端的信道矩阵，进而以更好地对未来某时刻的信道信息进行预测，提升信道信息预测的准确性。
请参照图3，图3是本申请实施例提供的另一种信道矩阵处理方法的流程图，如图3所示，所述方法包括以下步骤：
步骤301、网络侧设备接收终端上报的信道特征信息；
步骤302、所述网络侧设备将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵。
其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
本申请实施例中，终端确定N个第一正交基和M个第二正交基后，基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，并将每个子带预处理后的信道矩阵输入至第一AI模型中，获取所述第一AI模型输出的信道特征信息，终端将所述信道特征信息上报给网络侧设备，网络侧设备通过第二AI模型对信道特征信息进行处理，输出信道矩阵，从而网络侧设备通过第二AI模型恢复信道矩阵。这样，也就使得终端和网络侧设备能够通过AI模型来实现对信道矩阵的处理和恢复，能够有效节省传输资源。
其中，第一正交基对应的是发送端天线的空域信息，第二正交基对应的是接收端天线的空域信息，进而终端通过第一AI模型处理得到的信道特征信息也就同时考虑了接收端和发送端，网络侧设备通过第二AI模型进行信道矩阵恢复，也就能够得到接收端和发送端的全信道信息。这样，也就使得网络侧设备不仅能够得到发送端的信道信息，还能够获得接收端的信道信息，更有助于网络侧设备对信道信息进行预测，以提升信道信息预测的准确性。
可选地，所述方法还包括：
所述网络侧设备向终端发送第一指示信息，所述第一指示信息用于指示所述N的取值和所述M的取值。
可选地，所述方法还包括：
所述网络侧设备接收终端上报的所述N个第一正交基的索引和所述M个第二正交基的索引以及过采样组的索引，其中，所述N个第一正交基和所述M个第二正交基为所述终端经过过采样处理后从过采样组中确定的正交基。
可选地，所述N的取值和所述M的取值与所述第一AI模型匹配。
可选地，在所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型不匹配的情况下，所述方法还包括：
所述网络侧设备接收所述终端上报的所述N个第一正交基的位置和/或所述M个第二 正交基的位置。
可选地，所述网络侧设备接收所述终端上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置，包括：
所述网络侧设备接收所述终端通过CSI上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置。
可选地，在所述终端自行确定所述M个第二正交基的情况下，所述方法还包括：
所述网络侧设备接收所述终端上报的第二指示信息，所述第二指示信息用于指示所述M个第二正交基的位置。
可选地，所述网络侧设备接收终端上报的信道特征信息，包括如下任意一项：
所述网络侧设备接收终端通过CSI上报的信道特征信息；
所述网络侧设备向终端发送信令，并接收所述终端基于所述信令上报的信道特征信息；
所述网络侧设备接收终端通过CSI报告中的配置信息上报的信道特征信息。
可选地，所述信令包括如下至少一项：
DCI；
MAC CE；
RRC。
可选地，所述网络侧设备接收终端上报的信道特征信息，包括：
所述网络侧设备接收终端量化处理后的信道特征信息。
需要说明地，本申请实施例所提供的信道矩阵处理方法，执行主体为网络侧设备，与上述终端执行的信道矩阵处理方法相对应，本申请实施例所涉及的相关概念及具体实现过程可以是参照上述图2所述方法实施例中的描述，为避免重复，此处不再赘述。
本申请实施例提供的信道矩阵处理方法，执行主体可以为信道矩阵处理装置。本申请实施例中以信道矩阵处理装置执行信道矩阵处理方法为例，说明本申请实施例提供的信道矩阵处理装置。
请参照图4，图4是本申请实施例提供的一种信道矩阵处理装置的结构图，如图4所示，所述信道矩阵处理装置400包括：
处理模块401，用于确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；
第一获取模块402，用于将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；
上报模块403，用于向网络侧设备上报所述信道特征信息。
可选地，所述处理模块401还用于执行如下任意一项：
基于所述第一AI模型确定N个第一正交基和M个第二正交基，所述N个第一正交 基的位置和所述M个第二正交基的位置与所述第一AI模型匹配；
接收网络侧设备发送的第一指示信息，并基于所述第一指示信息确定N个第一正交基和M个第二正交基，所述第一指示信息用于指示所述N的取值和所述M的取值。
可选地，所述上报模块403还用于：
向网络侧设备上报所述N个第一正交基的索引和所述M个第二正交基的索引以及过采样组的索引，其中，所述N个第一正交基和所述M个第二正交基为所述装置经过过采样处理后从过采样组中确定的正交基。
可选地，所述N的取值和所述M的取值与所述第一AI模型匹配。
可选地，在所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型不匹配的情况下，所述上报模块403还用于：
向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。
可选地，所述上报模块403还用于：
通过CSI向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。
可选地，所述处理模块401还用于：
获取通过CSI上报的空域正交基，并从所述通过CSI上报的空域正交基中确定出N个第一正交基；
自行确定M个第二正交基。
可选地，所述装置还用于：
向网络侧设备指示所述M个第二正交基。
可选地，所述第一获取模块402还用于：
将预处理后的信道矩阵输入目标第一AI模型中，获取所述目标第一AI模型输出的信道特征信息，所述目标第一AI模型与输入的信道矩阵对应的子带匹配。
可选地，每个子带对应的所述目标第一AI模型相同或者不同。
可选地，所述处理模块401还用于执行如下任意一项：
基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理，投影处理后的信道矩阵的维度为N×M或者M×N；
基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理，并对投影处理后的信道矩阵进行加权处理，以使得加权处理后的信道矩阵与投影处理前的信道矩阵维度一致，所述加权处理使用的正交基为所述N个第一正交基和所述M个第二正交基。
可选地，在基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理的情况下，所述N个第一正交基基于对应的投影系数按照第一预设顺序进行排序，所述M个第二正交基基于对应的投影系数按照第二预设顺序进行排序。
可选地，所述上报模块403还用于执行如下任意一项：
通过CSI向网络侧设备上报所述信道特征信息；
接收网络侧设备发送的信令，并基于所述信令向网络侧设备上报所述信道特征信息；
基于CSI报告中的配置信息向网络侧设备上报所述信道特征信息。
可选地，所述信令包括如下至少一项：
DCI；
MAC CE；
RRC。
可选地，所述上报模块403还用于：
对所述信道特征信息进行量化处理，并向网络侧设备上报量化处理后的信道特征信息。
可选地，所述第一获取模块402还用于：
所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型量化处理后输出的信道特征信息。
本申请实施例中，所述装置能够和网络侧设备通过AI模型来实现对信道矩阵的处理和恢复，能够有效节省传输资源。并且，第一正交基对应的是发送端天线的空域信息，第二正交基对应的是接收端天线的空域信息，进而所述装置通过第一AI模型处理得到的信道特征信息也就同时考虑了接收端和发送端，网络侧设备通过第二AI模型进行信道矩阵恢复，也就能够得到接收端和发送端的全信道信息。这样，也就使得网络侧设备不仅能够得到发送端的信道信息，还能够获得接收端的信道信息，更有助于网络侧设备对信道信息进行预测，以提升信道信息预测的准确性
本申请实施例中的信道矩阵处理装置400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矩阵处理装置400能够实现图2所述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。
请参照图5，图5是本申请实施例提供的另一种信道矩阵处理装置的结构图，如图5所示，所述信道矩阵处理装置500包括：
接收模块501，用于接收终端上报的信道特征信息；
第二获取模块502，用于将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵；
其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
可选地，所述装置还包括：
发送模块，用于向终端发送第一指示信息，所述第一指示信息用于指示所述N的取值和所述M的取值。
可选地，所述接收模块501还用于：
接收终端上报的所述N个第一正交基的索引和所述M个第二正交基的索引以及过采样组的索引，其中，所述N个第一正交基和所述M个第二正交基为所述终端经过过采样处理后从过采样组中确定的正交基。
可选地，所述N的取值和所述M的取值与所述第一AI模型匹配。
可选地，在所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型不匹配的情况下，所述接收模块501还用于：
接收所述终端上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置。
可选地，所述接收模块501还用于：
接收所述终端通过CSI上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置。
可选地，在所述终端自行确定所述M个第二正交基的情况下，所述接收模块501还用于：
接收所述终端上报的第二指示信息，所述第二指示信息用于指示所述M个第二正交基的位置。
可选地，所述接收模块501还用于执行如下任意一项：
接收终端通过CSI上报的信道特征信息；
向终端发送信令，并接收所述终端基于所述信令上报的信道特征信息；
接收终端通过CSI报告中的配置信息上报的信道特征信息。
可选地，所述信令包括如下至少一项：
DCI；
MAC CE；
RRC。
可选地，所述接收模块501还用于：
所述网络侧设备接收终端量化处理后的信道特征信息。
本申请实施例中，所述装置通过第二AI模型对信道特征信息进行处理，输出信道矩阵，从而所述装置通过第二AI模型恢复信道矩阵。这样，也就使得终端和所述装置能够通过AI模型来实现对信道矩阵的处理和恢复，能够有效节省传输资源。并且，第一正交基对应的是发送端天线的空域信息，第二正交基对应的是接收端天线的空域信息，进而终端通过第一AI模型处理得到的信道特征信息也就同时考虑了接收端和发送端，所述装置通过第二AI模型进行信道矩阵恢复，也就能够得到接收端和发送端的全信道信息。这样，也就使得所述装置不仅能够得到发送端的信道信息，还能够获得接收端的信道信息，更有 助于所述装置对信道信息进行预测，以提升信道信息预测的准确性。
本申请实施例提供的信道矩阵处理装置500能够实现图3所述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。
可选地，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执行时实现上述图2所述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执行时实现上述图3所述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，处理器用于所述处理器用于确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；以及用于将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；所述通信接口用于向网络侧设备上报所述信道特征信息。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可以包括图形处理单元(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；可选地，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N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；
以及用于将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息
射频单元701，用于向网络侧设备上报所述信道特征信息。
本申请实施例中，终端700能够和网络侧设备通过AI模型来实现对信道矩阵的处理和恢复，能够有效节省传输资源。并且，第一正交基对应的是发送端天线的空域信息，第二正交基对应的是接收端天线的空域信息，进而终端700通过第一AI模型处理得到的信道特征信息也就同时考虑了接收端和发送端，网络侧设备通过第二AI模型进行信道矩阵恢复，也就能够得到接收端和发送端的全信道信息。这样，也就使得网络侧设备不仅能够得到发送端的信道信息，还能够获得接收端的信道信息，更有助于网络侧设备对信道信息进行预测，以提升信道信息预测的准确性。
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，所述通信接口用于接收终端上报的信道特征信息；所述处理器用于将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵；其中，所述信道特征信息为所述终端的第一AI 模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
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，本申请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5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图2所述方法实施例的各个过程，或者实现上述图3所述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，实现上述图2所述方法实施例的各个过程，或者实现上述图3所述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图2所述方法实施例的各个过程，或者实现上述图3所述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通信系统，包括：终端及网络侧设备，所述终端可用于执行如图2所述的信道矩阵处理方法的步骤，所述网络侧设备可用于执行如上图3所述的信道矩阵处理方法的步骤。
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相关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 (31)

  1. 一种信道矩阵处理方法，包括：
    终端确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；
    所述终端将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；
    所述终端向网络侧设备上报所述信道特征信息。
  2. 根据权利要求1所述的方法，其中，所述终端确定N个第一正交基和M个第二正交基，包括如下任意一项：
    所述终端基于所述第一AI模型确定N个第一正交基和M个第二正交基，所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型匹配；
    所述终端接收网络侧设备发送的第一指示信息，并基于所述第一指示信息确定N个第一正交基和M个第二正交基，所述第一指示信息用于指示所述N的取值和所述M的取值。
  3. 根据权利要求2所述的方法，其中，在所述终端基于所述第一指示信息确定N个第一正交基和M个第二正交基的情况下，所述方法还包括：
    所述终端向网络侧设备上报所述N个第一正交基的索引和所述M个第二正交基的索引以及过采样组的索引，其中，所述N个第一正交基和所述M个第二正交基为所述终端经过过采样处理后从过采样组中确定的正交基。
  4. 根据权利要求2所述的方法，其中，所述N的取值和所述M的取值与所述第一AI模型匹配。
  5. 根据权利要求1所述的方法，其中，在所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型不匹配的情况下，所述方法还包括：
    所述终端向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。
  6. 根据权利要求5所述的方法，其中，所述终端向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置，包括：
    所述终端通过信道状态信息CSI向网络侧设备上报所述N个第一正交基的位置和/或所述M个第二正交基的位置。
  7. 根据权利要求1所述的方法，其中，所述终端确定N个第一正交基和M个第二正交基，包括：
    所述终端获取通过CSI上报的空域正交基，并从所述通过CSI上报的空域正交基中确定出N个第一正交基；
    所述终端自行确定M个第二正交基。
  8. 根据权利要求7所述的方法，其中，所述方法还包括：
    所述终端向网络侧设备指示所述M个第二正交基。
  9. 根据权利要求1所述的方法，其中，所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型输出的信道特征信息，包括：
    所述终端将预处理后的信道矩阵输入目标第一AI模型中，获取所述目标第一AI模型输出的信道特征信息，所述目标第一AI模型与输入的信道矩阵对应的子带匹配。
  10. 根据权利要求9所述的方法，其中，每个子带对应的所述目标第一AI模型相同或者不同。
  11. 根据权利要求1所述的方法，其中，所述基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，包括如下任意一项：
    基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理，投影处理后的信道矩阵的维度为N×M或者M×N；
    基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理，并对投影处理后的信道矩阵进行加权处理，以使得加权处理后的信道矩阵与投影处理前的信道矩阵维度一致，所述加权处理使用的正交基为所述N个第一正交基和所述M个第二正交基。
  12. 根据权利要求11所述的方法，其中，在基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行投影处理的情况下，所述N个第一正交基基于对应的投影系数按照第一预设顺序进行排序，所述M个第二正交基基于对应的投影系数按照第二预设顺序进行排序。
  13. 根据权利要求1所述的方法，其中，所述终端向网络侧设备上报所述信道特征信息，包括如下任意一项：
    所述终端通过CSI向网络侧设备上报所述信道特征信息；
    所述终端接收网络侧设备发送的信令，并基于所述信令向网络侧设备上报所述信道特征信息；
    所述终端基于CSI报告中的配置信息向网络侧设备上报所述信道特征信息。
  14. 根据权利要求13所述的方法，其中，所述信令包括如下至少一项：
    下行控制信息DCI；
    媒体接入控制控制元素MAC CE；
    无线资源控制RRC。
  15. 根据权利要求1所述的方法，其中，所述终端向网络侧设备上报所述信道特征信息，包括：
    所述终端对所述信道特征信息进行量化处理，并向网络侧设备上报量化处理后的信道特征信息。
  16. 根据权利要求1所述的方法，其中，所述终端将预处理后的信道矩阵输入第一AI 模型中，获取所述第一AI模型输出的信道特征信息，包括：
    所述终端将预处理后的信道矩阵输入第一AI模型中，获取所述第一AI模型量化处理后输出的信道特征信息。
  17. 一种信道矩阵处理方法，包括：
    网络侧设备接收终端上报的信道特征信息；
    所述网络侧设备将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵；
    其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
  18. 根据权利要求17所述的方法，其中，所述方法还包括：
    所述网络侧设备向终端发送第一指示信息，所述第一指示信息用于指示所述N的取值和所述M的取值。
  19. 根据权利要求18所述的方法，其中，所述方法还包括：
    所述网络侧设备接收终端上报的所述N个第一正交基的索引和所述M个第二正交基的索引以及过采样组的索引，其中，所述N个第一正交基和所述M个第二正交基为所述终端经过过采样处理后从过采样组中确定的正交基。
  20. 根据权利要求18所述的方法，其中，所述N的取值和所述M的取值与所述第一AI模型匹配。
  21. 根据权利要求17所述的方法，其中，在所述N个第一正交基的位置和所述M个第二正交基的位置与所述第一AI模型不匹配的情况下，所述方法还包括：
    所述网络侧设备接收所述终端上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置。
  22. 根据权利要求21所述的方法，其中，所述网络侧设备接收所述终端上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置，包括：
    所述网络侧设备接收所述终端通过CSI上报的所述N个第一正交基的位置和/或所述M个第二正交基的位置。
  23. 根据权利要求17所述的方法，其中，在所述终端自行确定所述M个第二正交基的情况下，所述方法还包括：
    所述网络侧设备接收所述终端上报的第二指示信息，所述第二指示信息用于指示所述M个第二正交基的位置。
  24. 根据权利要求17所述的方法，其中，所述网络侧设备接收终端上报的信道特征信息，包括如下任意一项：
    所述网络侧设备接收终端通过CSI上报的信道特征信息；
    所述网络侧设备向终端发送信令，并接收所述终端基于所述信令上报的信道特征信息；
    所述网络侧设备接收终端通过CSI报告中的配置信息上报的信道特征信息。
  25. 根据权利要求24所述的方法，其中，所述信令包括如下至少一项：
    DCI；
    MAC CE；
    RRC。
  26. 根据权利要求17所述的方法，其中，所述网络侧设备接收终端上报的信道特征信息，包括：
    所述网络侧设备接收终端量化处理后的信道特征信息。
  27. 一种信道矩阵处理装置，包括：
    处理模块，用于确定N个第一正交基和M个第二正交基，并基于所述N个第一正交基和所述M个第二正交基对每个子带的信道矩阵进行预处理，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数；
    第一获取模块，用于将预处理后的信道矩阵输入第一人工智能AI模型中，获取所述第一AI模型输出的信道特征信息；
    上报模块，用于向网络侧设备上报所述信道特征信息。
  28. 一种信道矩阵处理装置，包括：
    接收模块，用于接收终端上报的信道特征信息；
    第二获取模块，用于将所述信道特征信息输入至第二AI模型中，获取所述第二AI模型输出的信道矩阵；
    其中，所述信道特征信息为所述终端的第一AI模型的输出，所述第一AI模型的输入为所述终端通过N个第一正交基和M个第二正交基对每个子带的信道矩阵进行预处理后的信道矩阵，所述第一正交基对应发送端天线的空域信息，所述第二正交基对应接收端天线的空域信息，N和M均为正整数。
  29. 一种终端，包括处理器和存储器，其中，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1-16中任一项所述的信道矩阵处理方法的步骤。
  30. 一种网络侧设备，包括处理器和存储器，其中，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17-26中任一项所述的信道矩阵处理方法的步骤。
  31. 一种可读存储介质，所述可读存储介质上存储程序或指令，其中，所述程序或指令被处理器执行时实现如权利要求1-16中任一项所述的信道矩阵处理方法的步骤，或者实现如权利要求17-26中任一项所述的信道矩阵处理方法的步骤。
PCT/CN2023/100426 2022-06-22 2023-06-15 信道矩阵处理方法、装置、终端及网络侧设备 WO2023246618A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210716334.3 2022-06-22
CN202210716334.3A CN117318774A (zh) 2022-06-22 2022-06-22 信道矩阵处理方法、装置、终端及网络侧设备

Publications (1)

Publication Number Publication Date
WO2023246618A1 true WO2023246618A1 (zh) 2023-12-28

Family

ID=89241307

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/100426 WO2023246618A1 (zh) 2022-06-22 2023-06-15 信道矩阵处理方法、装置、终端及网络侧设备

Country Status (2)

Country Link
CN (1) CN117318774A (zh)
WO (1) WO2023246618A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118300656A (zh) * 2024-04-18 2024-07-05 浙江大学 一种基于液态神经网络的鲁棒性多模态波束赋形设计方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117527113B (zh) * 2024-01-08 2024-04-12 深圳市迈腾电子有限公司 一种基于ai模型的无线信道衰落的预估方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210344395A1 (en) * 2018-10-26 2021-11-04 Telefonaktiebolaget Lm Ericsson (Publ) Systems and methods for providing forced full orthogonality for beams in a mu/mimo radio system
CN113746514A (zh) * 2020-05-29 2021-12-03 华为技术有限公司 一种通信方法、装置及系统
WO2021253937A1 (zh) * 2020-06-19 2021-12-23 株式会社Ntt都科摩 无线通信系统的终端、基站以及由终端和基站执行的方法
CN113840324A (zh) * 2020-06-24 2021-12-24 华为技术有限公司 一种测量上报方法及装置
CN114070675A (zh) * 2020-08-05 2022-02-18 展讯半导体(南京)有限公司 Ai网络模型匹配方法及装置、存储介质、用户设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210344395A1 (en) * 2018-10-26 2021-11-04 Telefonaktiebolaget Lm Ericsson (Publ) Systems and methods for providing forced full orthogonality for beams in a mu/mimo radio system
CN113746514A (zh) * 2020-05-29 2021-12-03 华为技术有限公司 一种通信方法、装置及系统
WO2021253937A1 (zh) * 2020-06-19 2021-12-23 株式会社Ntt都科摩 无线通信系统的终端、基站以及由终端和基站执行的方法
CN113840324A (zh) * 2020-06-24 2021-12-24 华为技术有限公司 一种测量上报方法及装置
CN114070675A (zh) * 2020-08-05 2022-02-18 展讯半导体(南京)有限公司 Ai网络模型匹配方法及装置、存储介质、用户设备

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118300656A (zh) * 2024-04-18 2024-07-05 浙江大学 一种基于液态神经网络的鲁棒性多模态波束赋形设计方法

Also Published As

Publication number Publication date
CN117318774A (zh) 2023-12-29

Similar Documents

Publication Publication Date Title
WO2023246618A1 (zh) 信道矩阵处理方法、装置、终端及网络侧设备
JP2024512358A (ja) 情報報告方法、装置、第１機器及び第２機器
WO2023185978A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
WO2023185980A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2023185995A1 (zh) 信道特征信息传输方法、装置、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
WO2024007949A1 (zh) Ai模型处理方法、装置、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
WO2023179460A1 (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
WO2023179570A1 (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
WO2024164961A1 (zh) 信息处理方法、装置、终端及网络侧设备
WO2024149157A1 (zh) Csi传输方法、装置、终端及网络侧设备
WO2024032606A1 (zh) 信息传输方法、装置、设备、系统及存储介质
WO2024104126A1 (zh) 更新ai网络模型的方法、装置和通信设备
WO2023179473A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
WO2024140578A1 (zh) 基于ai模型的csi反馈方法、终端及网络侧设备
WO2024093999A1 (zh) 信道信息的上报和接收方法、终端及网络侧设备
WO2024088161A1 (zh) 信息传输方法、信息处理方法、装置和通信设备
WO2024051564A1 (zh) 信息传输方法、ai网络模型训练方法、装置和通信设备
WO2023160442A1 (zh) Csi上报方法、信道预测方法、终端和网络侧设备
CN117318773A (zh) 信道矩阵处理方法、装置、终端及网络侧设备
WO2024140422A1 (zh) Ai单元的性能监测方法、终端及网络侧设备
WO2023179474A1 (zh) 信道特征信息辅助上报及恢复方法、终端和网络侧设备

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: 23826267

Country of ref document: EP

Kind code of ref document: A1