CSI processing method and apparatus, device, and medium

Abstract

The present application relates to the technical field of communications, and discloses a channel state information (CSI) processing method and apparatus, a device, and a medium. The CSI processing method provided by embodiments of the present application comprises: a user equipment (UE) receives first report configuration information sent by a network side device, the first report configuration information comprising at least one of the following: compression configuration information of a CSI and prediction configuration information of the CSI; and on the basis of the first report configuration information, the UE uses an artificial intelligence (AI) model to process first CSI information of the UE to obtain second CSI information, wherein the second CSI information at least comprises third CSI information, and the third CSI information is obtained by predicting CSI information of the UE at a future time point and compressing the predicted CSI information.

Description

CSI processing method, device, equipment and medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on February 19, 2024, with application number 202410185594.1 and titled “CSI processing method, device, equipment and medium,” the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present application belongs to the field of communication technology, and specifically relates to a CSI processing method, apparatus, device, and medium. Background Art

[0004] Current channel state information (CSI) processing methods are mainly designed for situations where only CSI compression or CSI prediction is required, and do not consider how to implement CSI processing based on artificial intelligence (AI) when CSI prediction and compression are required. Summary of the Invention

[0005] The embodiments of the present application provide a CSI processing method, apparatus, device, and medium that can solve the problem of how to implement AI-based predictive compression CSI processing.

[0006] In a first aspect, a CSI processing method is provided, which is performed by a user equipment. The method includes:

[0007] The user equipment UE receives first reporting configuration information sent by the network side device, where the first reporting configuration information includes at least one of the following: compression configuration information of the channel state information CSI, and prediction configuration information of the CSI;

[0008] The UE processes the first CSI information of the UE using an artificial intelligence (AI) model according to the first reporting configuration information to obtain second CSI information;

[0009] The second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

[0010] In a second aspect, a CSI processing method is provided, which is performed by a network-side device. The method includes:

[0011] The network side device sends first reporting configuration information to the user equipment UE, where the first reporting configuration information includes at least one of the following: compression configuration information of the channel state information CSI, and prediction configuration information of the CSI;

[0012] The first reporting configuration information is used for the UE to process the first CSI information of the UE using an artificial intelligence AI model to obtain second CSI information;

[0013] The second CSI information includes at least third CSI information, where the third CSI information is: CSI information obtained by predicting CSI information of the UE at a future time point and compressing the predicted CSI information.

[0014] In a third aspect, a CSI processing device is provided, applied to a user equipment (UE), the device including:

[0015] A first receiving module is configured to receive first reporting configuration information sent by a network-side device, where the first reporting configuration information includes at least one of the following: compression configuration information of channel state information (CSI) and prediction configuration information of the CSI;

[0016] A first processing module, configured to process the first CSI information of the UE using an artificial intelligence (AI) model according to the first reporting configuration information to obtain second CSI information;

[0017] The second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

[0018] In a fourth aspect, a CSI processing device is provided, applied to a network-side device, the device comprising:

[0019] A configuration information sending module, configured to send first reporting configuration information to a user equipment UE, where the first reporting configuration information includes at least one of the following: compression configuration information of channel state information CSI, and prediction configuration information of the CSI;

[0020] The first reporting configuration information is used for the UE to process the first CSI information of the UE using an artificial intelligence AI model to obtain second CSI information;

[0021] The second CSI information includes at least third CSI information, where the third CSI information is: CSI information obtained by predicting CSI information of the UE at a future time point and compressing the predicted CSI information.

[0022] In a fifth aspect, a user equipment is provided, which communication device includes a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the CSI processing method described in the first aspect are implemented.

[0023] In the sixth aspect, a network side device is provided, which communication device includes a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the CSI processing method described in the second aspect are implemented.

[0024] In a seventh aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the CSI processing method described in the first aspect or the second aspect are implemented.

[0025] In an eighth aspect, a wireless communication system is provided, comprising: a user device and a network side device, wherein the user device can be used to execute the steps of the CSI processing method as described in the first aspect, and the network side device can be used to execute the steps of the CSI processing method as described in the second aspect.

[0026] In a ninth aspect, a chip is provided, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the steps of the CSI processing method described in the first aspect or the second aspect.

[0027] In the tenth aspect, a computer program / program product is provided, which is stored in a storage medium and executed by at least one processor to implement the steps of the CSI processing method as described in the first aspect or the second aspect.

[0028] In an embodiment of the present application, CSI processing logic combined with the first reporting configuration information is introduced into the user equipment, thereby providing a solution for how to implement AI-based prediction and compression CSI processing and how to monitor the performance of AI-based prediction and compression CSI processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] FIG1 is a block diagram of a wireless communication system to which embodiments of the present application may be applied;

[0030] FIG2 is a schematic diagram of a neural network in an embodiment of the present application;

[0031] FIG3 is a schematic diagram of a neuron in an embodiment of the present application;

[0032] FIG4 is a schematic diagram of a compression and decompression process in an embodiment of the present application;

[0033] FIG5 is a schematic diagram of a prediction process in an embodiment of the present application;

[0034] FIG6 is a flowchart of an implementation method of a CSI processing method in an embodiment of the present application;

[0035] FIG7 is a schematic diagram of a CSI-related time point in an embodiment of the present application;

[0036] FIG8 is a schematic diagram of time domain compression and decompression in an embodiment of the present application;

[0037] FIG9 is a flowchart of another CSI processing method according to an embodiment of the present application;

[0038] FIG10 is a structural block diagram of a CSI processing device according to an embodiment of the present application;

[0039] FIG11 is a structural block diagram of another CSI processing device in an embodiment of the present application;

[0040] FIG12 is a structural block diagram of a communication device in an embodiment of the present application;

[0041] FIG13 is a schematic diagram of the hardware structure of a user equipment in an embodiment of the present application;

[0042] FIG14 is a schematic diagram of the hardware structure of a network-side device in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The following will be combined with the accompanying drawings in the embodiments of this application to clearly describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of this application.

[0044] The terms "first," "second," and the like in this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable, where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein. Furthermore, the objects distinguished by "first" and "second" generally refer to a class and do not limit the number of objects. For example, the first object can be one or more. Furthermore, "or" in this application represents at least one of the connected objects. For example, the scope of protection for "A or B" covers at least three options: Option 1: includes A but not B; Option 2: includes B but not A; and Option 3: includes both A and B. Furthermore, the terms "A and / or B," "at least one of A and B," and "at least one of A or B" also cover at least the above three options, respectively. The character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0045] The term "indication" in this application can be either a direct indication (or explicit indication) or an indirect indication (or implicit indication). A direct indication can be understood as the sender explicitly informing the receiver of specific information, the operation to be performed, or the result of the request in the instruction sent; an indirect indication can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the operation to be performed or the result of the request based on the judgment result.

[0046] It is worth noting that the technology described in the embodiments of the present application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA) or other systems. The terms "system" and "network" in the embodiments of the present application are often used interchangeably, and the technology described can be used for the systems and radio technologies mentioned above, as well as for other systems and radio technologies. The following description describes a New Radio (NR) system for illustrative purposes, and NR terminology is used in most of the following description, but these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) systems. th Generation, 6G) communication system.

[0047] FIG1 shows a block diagram of a wireless communication system applicable to an embodiment of the present application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 may also be referred to as a user equipment (UE), and may be a mobile phone, a tablet personal computer, a laptop computer, a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR), a virtual reality (VR) device, a robot, a wearable device, an aircraft, a vehicle user equipment (VUE), a ship-borne device, a pedestrian user equipment (PUE), a smart home (a home appliance with wireless communication capabilities, such as a refrigerator, a television, a washing machine, or furniture), a game console, a personal computer (PC), a teller machine, or a self-service machine, or other terminal-side device. Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle-mounted device can also be called a vehicle-mounted terminal, a vehicle-mounted controller, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network side device 12 may include an access network device or a core network device, wherein the access network device may also be called a radio access network (Radio Access Network, RAN) device, a radio access network function or a radio access network unit. The access network device may include a base station, a wireless local area network (WLAN) access point (AP) or a wireless fidelity (WiFi) node, etc.Among them, a base station may be referred to as a Node B (NB), an evolved Node B (eNB), the next generation Node B (gNB), a New Radio Node B (NR Node B), an access point, a Relay Base Station (RBS), a Serving Base Station (SBS), a Base Transceiver Station (BTS), a radio base station, a radio transceiver, a Basic Service Set (BSS), an Extended Service Set (ESS), a home Node B (HNB), a home evolved Node B, a Transmit / Receive Point (TRP), a Non-Terrestrial Network (NTN) device (such as a satellite or a high altitude platform station), or a radio base station. station) or other appropriate terms in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical vocabulary. It should be noted that in the embodiment of the present application, only the base station in the NR system is introduced as an example, and the specific type of the base station is not limited.

[0048] The core network device may also be referred to as a core network node, a core network function or a core network element, etc., which includes but is not limited to at least one of the following: Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized Network Configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (L-NEF), Binding Support Function (Binding Support Function) Function (BSF), Application Function (AF), Location Management Function (LMF), Gateway Mobile Location Centre (GMLC), Network Data Analytics Function (NWDAF), Non-Terrestrial Network (NTN) equipment (such as satellite or high altitude platform station, etc.), etc.It should be noted that in the embodiments of this application, only the core network equipment in the NR system is introduced as an example, and the specific type of the core network equipment is not limited. If the name of the core network equipment mentioned in the embodiments of this application changes in subsequent protocol versions (such as 6G), it is also within the scope of protection of this application.

[0049] Optionally, the core network device can be implemented by one or more functional modules in a single device, or by multiple devices together, and this is not specifically limited in the embodiments of the present application. It is understood that the above-mentioned functional modules can be network elements in hardware devices, or software functional modules running on dedicated hardware, or virtualized functional modules instantiated on a platform (e.g., a cloud platform).

[0050] To facilitate understanding of the technical solutions provided by this application, the main technical concepts involved in the embodiments of this application are briefly described below.

[0051] 1. Artificial Intelligence

[0052] AI is currently being widely used in various fields. For example, in the communications sector, integrating AI into wireless communication networks to significantly improve technical indicators such as throughput, latency, and user capacity is a key task for future wireless communication networks.

[0053] In specific applications, the AI ​​module can be implemented based on a variety of methods such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc. This application mainly uses neural networks as an example to illustrate the AI ​​module, but it does not limit the implementation method of the AI ​​module.

[0054] Referring to the neural network diagram shown in Figure 2, a neural network primarily comprises layers such as an input layer, a hidden layer, and an output layer. A neural network layer may contain one or more neurons. Here, X1, X2, …, Xn represent the inputs of the neural network, and Y represents the outputs of the neural network.

[0055] Referring to the schematic diagram of a neuron as shown in FIG3 , the output of the neuron can be determined by the following formula: z=a1w1+…+a k w k +…+a K w K +b

[0056] Among them, a1,…,a k ,…,a K represents the input of the neuron; w1,…,w k ,…,w Krepresents the weight (multiplicative coefficient); b represents the bias (additive coefficient); z represents the output of the neuron, which will be further nonlinearly processed by the activation function σ(.). The activation function can be an S-shaped growth curve (Sigmoid), a hyperbolic tangent function (tanh), a linear rectification function (Rectified Linear Unit, ReLU, also known as a rectified linear unit), etc.

[0057] The parameters of neural networks are mainly optimized using gradient optimization algorithms, which are a type of algorithm that minimizes or maximizes an objective function (sometimes called a loss function), and the objective function is often a mathematical combination of model parameters and data.

[0058] For example, given data X and its corresponding label Y, a neural network model f(.) is constructed that can predict the output f(x) based on the input x. When the input x is data X, the loss function is calculated by calculating the difference (f(x) - Y) between the predicted output f(x) and the true value (i.e., label Y). The goal of parameter optimization is to find the appropriate model parameters W and b to minimize the value of this loss function. It is understandable that the smaller the loss value, the closer the model's predicted output is to the true situation.

[0059] Neural network parameter optimization algorithms include gradient descent, stochastic gradient descent (SGD), mini-batch gradient descent, momentum, stochastic gradient descent with momentum (also known as Nesterov (named after its inventor)), adaptive gradient descent (Adagrad), adaptive learning rate optimization algorithm based on gradient descent (Adaptive delta, Adadelta), root mean square error propagation (RMSprop), and adaptive moment estimation (Adam). During error backpropagation, these parameter optimization algorithms all calculate the derivative (or partial derivative) of the current neuron based on the error (or loss) obtained from the loss function, and then calculate the gradient based on influencing parameters such as the learning rate, the previous gradient or derivative (or partial derivative), and pass the gradient to the neurons in the previous layer.

[0060] 2. Non-AI CSI Compression

[0061] The non-AI CSI compression in the protocol mainly includes type I, type II and enhanced type II.

[0062] 1. Type I CSI

[0063] Type I mainly needs to report the index of the two-dimensional discrete Fourier transform (DFT) vector and its phase rotation. When there is no way to report the complete channel or precoder, it reports the precoding matrix indicator (PMI) of the wideband or subband, that is, a two-dimensional DFT vector and its phase rotation on the wideband or subband.

[0064] Type I reporting format:

[0065] For broadband CSI:

[0066] Rank Indication-Precoding Matrix Indicator-Channel Quality Indicator (RI-PMI-CQI), where RI>4 corresponds to 2 transport blocks (TBs) and two CQIs need to be reported accordingly; otherwise, one CQI needs to be reported.

[0067] For sub-band CSI:

[0068] Part 1: CSI: RI and CQI of the first TB;

[0069] Part 2 CSI: wideband CQI-wideband PMI-CQI and PMI of even subbands-CQI and PMI of odd subbands.

[0070] Among them, the reporting format of type I has the following omission principle: part2CSI is omitted based on priority, that is, the reporting of odd subbands can be omitted first.

[0071] 2. Type IICSI

[0072] Type II requires reporting the basis vector index and its projection (amplitude and phase) on the basis vector. Compared to simply reporting a two-dimensional DFT vector and its phase rotation, type II represents the PMI as a linear weighted sum of a set of basis vectors.

[0073] Type II reporting format:

[0074] Part 1: RI-CQI - the number of non-zero wideband amplitude coefficients (encoded separately for each layer);

[0075] Part 2: Broadband PMI {L vector (basis vector)}-PMI-LI {i 1,4,l (Broadband Amplitude 1), i 2,1,l (Phase), i 2，2， l (amplitude 2)}; where amplitude 1 occupies 3 bits in scalar mode, and amplitude 2 occupies 1 bit.

[0076] 3. Enhanced type IICSI

[0077] Because Type II has a high overhead of hundreds or even thousands of bits, Enhanced Type II was designed to further compress Type II. Enhanced Type II compresses the vector of weighting coefficients for different subbands into a vector consisting of a set of frequency-domain basis vectors.

[0078] Reporting format for enhanced type II:

[0079] Part 1: RI-CQI - the number of non-zero wideband amplitude coefficients (encoded separately for each layer);

[0080] Part 2: PMI:i 2,4,l Amplitude, i 2,5,l Phase, i 1,7,l {report bitmap}.

[0081] The priority value Pri(l, i, f) reported by the CSI information can be determined by the following formula: Pri(l, i, f) = 2·L·υ·π(f)+υ·i+l

[0082] Where l is the layer index; i is the spatial basis vector index, i = 0, 1, ..., 2L-1; f is the frequency domain basis vector index, f = 0, 1, ..., M v -1,M v is the maximum number of basis vectors in the frequency domain corresponding to the number of layers v; L is the number of basis vectors in the spatial domain; v is the number of layers; π(f) and f are the frequency domain basis vectors, and π(f) has the highest priority when it is the strongest, that is, the 0th frequency domain vector, π(f) = {…3, 1, 0, 2, 4}, π(0) = 0, π(N3-1) = 1, π(1) = 2.

[0083] Among them, the frequency domain basis vector π(f) can be determined by the following formula:

[0084] Where l = 1, 2, ..., υ, v is the number of layers; N3 is the number of frequency domain basis vectors, is the codebook index.

[0085] It should be noted that Part 2 above uses a feedback method that uniformly compresses all subbands. The information represented by each bit position in the compressed information stream is as follows:

[0086] 0: L spatial basis vectors i 1,1 ,i 1,2 And the strongest coefficient information of each layer {log2 2L bit}i 1,8,l (l=1,…,υ);

[0087] 1: M frequency domain basis vectors (i 1,5 (if reported), i 1,6,l (if reported), reference amplitude information i 2,3,l , the highest priority v2LM-[KNZ / 2]bit in the non-zero coefficient position, i 1,7,l , v is the rank, v2LM-[KNZ / 2] coefficients with the highest priority {i 2,4,l ,i 2,5,l};

[0088] 2: The coefficient with the lowest priority [KNZ / 2] among the non-zero coefficient positions.

[0089] 3. AI-based CSI or PMI compression

[0090] 4 shows a schematic diagram of the compression and decompression process, W N*B Indicates the expected or target CSI (or codebook) of the user equipment (UE), N indicates the number of CSI ports, and B indicates the number of subbands. The user equipment generates W through the CSI generation model, that is, through the AI-based compression model. N*B Compress to obtain the codebook value, and then report the codebook value to the network side device. The network side device decompresses the codebook value through the CSI reconstruction model, that is, through the AI-based decompression model, to obtain W′ N*B .

[0091] 4. AI-based CSI prediction

[0092] Referring to the schematic diagram of the prediction process shown in FIG5 , W M-D , W M-2D ,…,W M-nDrepresents the actual channel or codebook at n time points before the reference time point M, where the interval between the n time points before the reference time point M is D; W X , W X+K ,…,W X+nK represents the channel or codebook predicted for n time points after the reference time point X, where the interval between the n time points after the reference time point X is K. By replacing W M-D , W M-2D ,…,W M-nD Input into the CSI prediction model, that is, W M-D , W M-2D ,…,W M-nD Input into the AI-based prediction model, and then W can be obtained based on AI prediction. X , W X+K ,…,W X+nK .

[0093] Since the current CSI processing method is mainly designed for situations where only CSI compression or only CSI prediction is required, it does not consider how to implement AI-based prediction and compression CSI processing when CSI prediction and compression are required, nor does it consider how to implement AI performance monitoring in such mixed situations.

[0094] The present application provides a CSI processing method, apparatus, device, and medium, which provide solutions to the problem of how to implement AI-based predictive compression CSI processing and how to implement AI performance monitoring in such mixed situations.

[0095] In the following, in conjunction with the accompanying drawings, a CSI processing method, apparatus, device, and medium provided in an embodiment of the present application are described in detail through some embodiments and their application scenarios.

[0096] For ease of description, the following unified description of the various terms involved in the embodiments of this application is first provided. (Unless otherwise specified, the following unified description of the various terms applies to all embodiments of this application. Due to the fact that some individual terms first appear later, they will not be uniformly described here. Below, they will be described when they first appear.)

[0097] Network side device: It can be the network side device 12 in Figure 1. For examples of the network side device 12, please refer to the previous text and will not be repeated here.

[0098] User equipment: can be the terminal 11 in FIG1 . For an example of the terminal 11 , please refer to the above text and will not be described in detail here.

[0099] AI model: It may also be referred to as an AI unit, a machine learning (ML) model, an ML unit, an AI structure, an AI function, an AI feature, a neural network, a neural network function, or a neural network function. An AI model may refer to a processing unit that can implement specific algorithms, formulas, processing procedures, or capabilities related to AI, or may refer to a processing method, algorithm, function, module, or unit for a specific data set (such as a data set that includes at least one of the input and output of an AI model), or may refer to a processing method, algorithm, function, module, or unit running on AI or ML-related hardware such as a graphics processing unit (GPU), a neural processing unit (NPU), a tensor processing unit (TPU), or an application-specific integrated circuit (ASIC). This application does not specifically limit this.

[0100] Identification of the AI ​​model: It can be the identification of the AI ​​model (or AI structure, AI algorithm, etc.) itself, or the identification of a specific data set associated with the AI ​​model, or the identification of a specific scenario, environment, channel characteristics or device related to the AI ​​model, or the identification of a function, feature, capability or module related to the AI ​​model. This application does not make any specific restrictions on this.

[0101] CSI information: It includes at least one of the expected codebook (precoding matrix) and channel information, such as CQI, RI, layer indication (LI), call request with identification (CRI), synchronization signal block resource indication (SSBRI) and layer 1 reference signal receiving power (L1-RSRP). It may also include PMI (which is the expected codebook indication information and can be calculated based on a known base sequence) and other information.

[0102] In a first aspect of an embodiment of the present application, a CSI processing method is provided. The method is performed by a user equipment. FIG6 is a flowchart of an implementation of a CSI processing method provided in an embodiment of the present application. The method may include the following steps:

[0103] Step S101: User equipment UE receives first reporting configuration information sent by a network side device.

[0104] The first reporting configuration information includes at least one of the following: compression configuration information of channel state information CSI, and prediction configuration information of the CSI.

[0105] During specific implementation, the network side device instructs the user equipment on how to perform prediction or compression by sending the first reporting configuration information, so as to obtain the CSI information that the network side device expects the user equipment to report.

[0106] For example, the first reporting configuration information includes at least CSI compression configuration information, and the CSI compression configuration information can be used to indicate the specific method of performing compression (such as using time domain compression), whether to perform AI-based compression on certain CSI information, and other information related to compression execution.

[0107] For another example, the first reporting configuration information includes at least CSI prediction configuration information, and the CSI prediction configuration information can be used to indicate how many time points or which future time points need to be predicted, the interval between the predicted time points, whether to predict the CSI information to be compressed or the compressed CSI information (that is, omitting the compression step and directly predicting the compressed CSI information at future time points), and other information related to the prediction execution.

[0108] In one embodiment, the network side device can indicate (or agree by protocol) the association relationship between the compression configuration information and the prediction configuration information of the CSI to the user equipment, so that the user equipment can determine the associated CSI prediction configuration information and at least one of the prediction configuration information based on the compression configuration information and the prediction configuration information of the CSI sent by the network side device, so as to subsequently perform prediction and compression CSI processing.

[0109] Step S102: The UE processes the first CSI information of the UE using an artificial intelligence (AI) model according to the first reporting configuration information to obtain second CSI information.

[0110] The second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

[0111] In a specific implementation, the UE obtains the first CSI information required for prediction based on the first reporting configuration information, and uses the AI ​​model to predict and compress all or part of the CSI information in the first CSI information to obtain second CSI information that includes at least third CSI information. After obtaining the second CSI information, the user equipment reports the second CSI information to the network-side device, which then decompresses the second CSI information to obtain relevant wireless channel characteristics.

[0112] In an optional embodiment, the first reporting configuration information includes at least one of the following:

[0113] The type of the reported second CSI information, such as third CSI information, fourth CSI information, and fifth CSI information;

[0114] The information of the AI ​​model used to obtain the second CSI information, such as parameters of the AI ​​model, the type of the AI ​​model, or monitoring information of the AI ​​model.

[0115] Among them, the type of the AI ​​model may include an AI compression model, an AI prediction model, or an AI compression and prediction model, etc.; the monitoring information of the AI ​​model may include a monitoring method, a monitoring type, or monitoring auxiliary information, etc.

[0116] It can be seen from the above steps that the present application introduces CSI processing logic combined with the first reporting configuration information to the user equipment, thereby providing a solution for how to implement AI-based prediction and compression CSI processing and how to monitor the performance of AI-based prediction and compression CSI processing.

[0117] In some embodiments, the second CSI information further includes CSI information at at least two time points that has been compressed in the time domain;

[0118] The CSI information at the at least two time points satisfies any one of the following:

[0119] The CSI information at the at least two time points is all CSI information of the UE at at least two second time points predicted by the AI ​​model, that is, the second CSI information includes CSI information predicted for the at least two second time points after time domain compression;

[0120] At least one of the CSI information at the at least two time points is CSI information of the UE at at least one second time point predicted by the AI ​​model, that is, the second CSI information includes CSI information predicted for the at least one second time point after time domain compression;

[0121] At least one of the CSI information at the at least two time points is the first CSI information of the UE at at least one first time point obtained by the UE, that is, the second CSI information includes actual CSI information or predicted CSI information measured at at least one first time point after time domain compression.

[0122] In an optional embodiment, the first reporting configuration information includes information of at least one time point among the at least two time points mentioned above.

[0123] In some embodiments, the at least two time points include at least one of the following:

[0124] The Xth time point, wherein the Xth time point is the first time point, or the third time point;

[0125] The X+Kth time point, wherein the X+Kth time point is a time point K away from the Xth time point;

[0126] The X+NKth time point, where the X+NKth time point is a time point that is NK away from the Xth time point, and N is an integer greater than 1.

[0127] For example, referring to the schematic diagram of CSI-related time points shown in FIG7 , the Xth time point may be the time point corresponding to the CSI reference resource (corresponding to the case where the Xth time point is the first time point); or the Xth time point may be the time point corresponding to the first predicted or reported CSI information indicated by the network device or agreed upon by the protocol (corresponding to the case where the Xth time point is the third time point), such as the starting time point of the prediction window. When the Xth time point is the starting time point of the prediction window, the X+Kth to X+NKth time points may be the time points after the starting time point in the prediction window at which CSI information prediction is required.

[0128] Optionally, K satisfies at least one of the following:

[0129] K is the time interval between two time points;

[0130] K is 1 time slot, 2 time slots, 4 time slots or 5 time slots;

[0131] The value of K is determined according to the value of D, D is determined according to the instruction of the network side device, or D is the interval between two adjacent first time points.

[0132] In an optional embodiment, the first reporting configuration information includes at least one of X or K.

[0133] In some implementations, the CSI information at at least two time points after time domain compression satisfies at least one of the following:

[0134] The CSI information at the Xth time point is CSI information that has not been time-domain compressed, or the CSI information at the Xth time point is CSI information obtained by performing time-domain compression on CSI information reported by the UE to the network-side device before the Xth time point;

[0135] The CSI information at the X+Kth time point is CSI information obtained by performing time domain compression on the CSI information at the Xth time point, where the X+Kth time point is K time points away from the Xth time point.

[0136] The CSI information at the X+NKth time point is CSI information obtained by time domain compression using the CSI information of at least one time point from the Xth time point to the X+(N-1)Kth time point, where the X+NKth time point is a time point NK away from the Xth time point, where N is an integer greater than 1.

[0137] For example, referring to the schematic diagram of time domain compression and decompression as shown in Figure 8, the CSI information at SlotX (i.e., the Xth time point) can be directly compressed and decompressed; for the CSI information at SlotX+K (i.e., the X+Kth time point) and subsequent time points, it is necessary to combine the CSI information at the previous time point (such as SlotX) for compression and decompression.

[0138] In an optional embodiment, the first reporting configuration information may also be used to instruct the user equipment to perform compression based on the CSI information at the previous time point (such as SlotX).

[0139] In some implementations, the UE reports at least one of the following to the network-side device:

[0140] CSI information of the UE at at least one time point between the X+K time point and the X+NK time point predicted by the UE through the AI ​​model;

[0141] CSI information at the Mth time point or the M-1th time point obtained by compressing the CSI information at M time points by the UE using the AI ​​model;

[0142] The UE predicts CSI information at at least one time point from the X+Kth time point to the X+NKth time point through the AI ​​model, and compresses the predicted CSI information to obtain CSI information.

[0143] In some implementations, the CSI information at at least two time points after time domain compression satisfies at least one of the following:

[0144] The CSI information at the at least two time points has different payloads;

[0145] The CSI information at the at least two time points has different coding and decoding methods, such as using time domain compression and corresponding decompression methods for some time points, and using other coding and decoding methods other than time domain compression and corresponding decompression methods for some time points;

[0146] There is a sequence for compression and decompression of the CSI information at the at least two time points. For example, for time sequence compression, the CSI information at the Xth time point needs to be compressed and decompressed first, and then combined with the CSI information at the Xth time point, the CSI information at the X+Kth time point needs to be further compressed and decompressed.

[0147] In an optional embodiment, the first reporting configuration information further includes at least one of the following: a payload of the second CSI information; a coding and decoding method of the second CSI information; in some embodiments, the second CSI information includes the third CSI information; in step S102, the UE processes the first CSI information of the UE using the AI ​​model according to the first reporting configuration information to obtain the second CSI information, including:

[0148] Step S1021: The UE obtains first CSI information of the UE at a first time point according to the first reporting configuration information.

[0149] The first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource.

[0150] In specific implementation, the user equipment can measure or predict one or more CSI information of the current time point, the historical time point or the time point of the CSI reference resource based on the information of the first time point indicated by the first reporting configuration information or related resource information to obtain the first CSI information of the first time point.

[0151] Step S1022: The UE predicts the CSI information of the UE at the second time point using a first prediction compression model based on the first CSI information of the UE at the first time point, and compresses the predicted CSI information using the first prediction compression model to obtain the third CSI information.

[0152] The second time point is a time point after the first time point.

[0153] In specific implementation, the network side device can inform the user equipment (or the user equipment itself determines) the identifier of the prediction compression model (which can be an independent model with prediction and compression functions, or a cascade model of the prediction model and the compression model) used to process the first CSI information at the first time point through the first reporting configuration information.

[0154] For example, the network side device informs the user device of the identifier of the prediction compression model used to process the first CSI information at the first time point through the first reporting configuration information. The user device can thereby determine the first prediction compression model, and use the first CSI information at the first time point as the input of the first prediction compression model, and perform prediction and compression CSI processing on the CSI information at the second time point after the first time point to obtain third CSI information.

[0155] For another example, the network-side device informs the user equipment of the structure or parameter information of the prediction compression model used to process the first CSI information at the first time point through the first reporting configuration information.

[0156] In some embodiments, the second CSI information further includes fifth CSI information; in step S102, the UE processes the first CSI information of the UE using an AI model according to the first reporting configuration information to obtain the second CSI information, including:

[0157] Step S1023: The UE obtains first CSI information of the UE at a first time point according to the first reporting configuration information.

[0158] The first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource.

[0159] In an optional embodiment, the first reporting configuration information further includes at least one of the following: configuration information of the CSI-RS for obtaining the first CSI information at the first time point.

[0160] In specific implementation, the user equipment can measure or predict one or more CSI information of the current time point, the historical time point or the time point of the CSI reference resource based on the information of the first time point indicated by the first reporting configuration information or related resource information to obtain the first CSI information of the first time point.

[0161] Step S1024: The UE predicts the CSI information of the UE at the second time point using a first prediction model based on the first CSI information of the UE at the first time point to obtain fifth CSI information in the second CSI information.

[0162] The second time point is a time point after the first time point.

[0163] In specific implementation, the network side device can inform the user equipment (or the user equipment determines it itself) of the identifier of the prediction model used to process the first CSI information at the first time point through the first reporting configuration information (which can be a prediction model independently deployed on the user equipment, or can be a prediction model deployed on the user equipment in cascade with the compression model).

[0164] For example, the network side device informs the user device of the identifier of the prediction model used to process the first CSI information at the first time point through the first reporting configuration information. The user device can thereby determine the first prediction model, and use the first CSI information at the first time point as the input of the first prediction compression model to predict the CSI information at the second time point after the first time point to obtain the fifth CSI information.

[0165] In some embodiments, the second CSI information further includes sixth CSI information; in step S102, the UE processes the first CSI information of the UE using an AI model according to the first reporting configuration information to obtain the second CSI information, including:

[0166] Step S1025: The UE obtains first CSI information of the UE at the first time point according to the first reporting configuration information.

[0167] The first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource.

[0168] In specific implementation, the user equipment can measure or predict one or more CSI information of the current time point, the historical time point or the time point of the CSI reference resource based on the information of the first time point indicated by the first reporting configuration information or related resource information to obtain the first CSI information of the first time point.

[0169] Step S1026: The UE compresses the first CSI information of the UE at the first time point using the first compression model to obtain sixth CSI information in the second CSI information.

[0170] In specific implementation, the network side device can inform the user device (or the user device determines it itself) of the identifier of the compression model used to process the first CSI information at the first time point through the first reporting configuration information (which can be a compression model independently deployed on the user device, or can be a compression model deployed on the user device in cascade with the prediction model).

[0171] For example, the network-side device informs the user device of the identifier of the compression model used to process the first CSI information at the first time point through the first reporting configuration information. The user device can thereby determine the first compression model, and use the first CSI information at the first time point as the input of the first compression model to compress the CSI information at the first time point, or use the first CSI information at the first time point as the input of a cascaded AI model (such as a cascaded first prediction model and a first compression model), predict and compress the CSI information at the first time point, and obtain the sixth CSI information.

[0172] In some implementations, obtaining, by the UE according to the first reporting configuration information, first CSI information of the UE at a first time point includes:

[0173] The UE obtains W based on the measurement results of multiple channel state information reference signals CSI-RS M-D ,W M-2D ,……W M-nD ;

[0174] Wherein, the W M-D represents the codebook information or channel information obtained by the UE at the MDth time point, where MD, M-2D, ..., M-nD represent different first time points.

[0175] Optionally, the M is determined based on the CSI-RS most recently before the Xth time point, D is the period or repetition interval of the CSI-RS, or D is the interval between the multiple CSI-RSs.

[0176] For example, as shown in FIG5 , the user equipment measures the CSI-RS at multiple first time points and obtains the measurement result W M-D ,W M-2D ,……W M-nD Then, based on the multiple measurement results (ie, codebooks), the CSI information at the time point where the CSI reference resource is located (ie, the Xth time point) and the time points thereafter is predicted to obtain the prediction result W X ,W X+K ,……W X+nK Wherein, D can be the time point distance between the two input codebooks, and K can be the time point distance between the two output codebooks.

[0177] In some embodiments, the UE monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, such as inputting at least two items of the first to sixth CSI information into a pre-built AI model to obtain corresponding performance monitoring results.

[0178] In some embodiments, the first reporting configuration information includes configuration information monitored by the AI ​​model, or, in some embodiments, the user equipment receives the first reporting configuration information and the configuration information monitored by the AI ​​model.

[0179] Optionally, there is an association between the first reporting configuration information and the configuration information monitored by the AI ​​model.

[0180] The fourth CSI information is: CSI information obtained by decompressing the second CSI information, the fifth CSI information is: predicted CSI information of the UE at a future time point, and the sixth CSI information is CSI information obtained by compressing the first CSI information.

[0181] In specific implementation, the UE may compare the actual CSI information with the relevant predicted and decompressed CSI information, such as comparing the actual CSI information at a certain time point in the first CSI information with the CSI information obtained by decompressing the third information associated with the time point in the fourth CSI information, to monitor the overall performance of the prediction and compression of the AI ​​model; or may compare the decompressed CSI information or the compressed CSI information with the CSI information before compression, such as comparing the CSI information to be compressed in the first CSI information (such as the actual CSI or the fifth CSI information) with the corresponding CSI information obtained by decompression in the fourth CSI information, or comparing the first CSI information with the sixth CSI information, to monitor the compression performance of the AI ​​model; or may compare the actual CSI information at a certain time point in the first CSI information with the fifth CSI information associated with the time point, to monitor the prediction performance of the AI ​​model.

[0182] In some implementations, the user equipment may obtain the fourth CSI information through the following method 1 or 2.

[0183] Mode 1: The UE receives the fourth CSI information sent by the network-side device, where the fourth CSI information is obtained by the network-side device decompressing the second CSI information using the first decompression model.

[0184] In method 1, after the network side device completes decompression of the second CSI information reported by the user equipment using the first decompression model configured by itself, the network side device can send the obtained decompression result as the fourth CSI information to the user equipment, so that the user equipment can monitor the AI ​​model.

[0185] Mode 2: The UE decompresses the second CSI information using the first reference model of the first decompression model to obtain the fourth CSI information.

[0186] In this method 2, the UE can use the first reference model configured by itself to simulate the process of the network side device using the first decompression model to decompress the second CSI information, and use the obtained decompression result as the fourth CSI information for subsequent AI model performance monitoring, thereby reducing the communication overhead between the network side device and the user device; or the UE can use the first reference model sent by the network side device to simulate the process of the network side device using the first decompression model to decompress the second CSI information, and use the obtained decompression result as the fourth CSI information for subsequent AI model performance monitoring, thereby unifying the understanding of the model between the network side device and the user device.

[0187] In some embodiments, the UE reports at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information to the network side device so that the network side device can perform performance monitoring. The specific monitoring process can be referred to the following embodiments one to three, which are not repeated here.

[0188] The above performance monitoring solution is further described below in conjunction with implementation modes one to three.

[0189] Implementation Method 1

[0190] Network-side devices or user devices monitor the overall performance of AI model prediction and compression.

[0191] In this embodiment, the network side device or the user device compares the actual CSI information at one or more time points (i.e., the first CSI information at the third time point) with the CSI information obtained after sequentially predicting, compressing and decompressing the associated time points (i.e., the fourth CSI information, which can be obtained by decompressing the CSI information at the second time point after performing prediction and compression on the first prediction compression model) to obtain an overall monitoring result of the prediction performance and compression performance of the AI ​​model (i.e., the first prediction compression model).

[0192] In some embodiments, the first CSI information at the third time point may be obtained by the user equipment by measuring the CSI-RS at the third time point.

[0193] The third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining the monitoring results.

[0194] Specifically, the user equipment can measure the CSI-RS at the time point indicated by the network side device for obtaining the monitoring result, and compare the measurement result as the first CSI information of the third time point with the fourth CSI information, so as to obtain the monitoring result that the network side device expects the user equipment to report; or the user equipment can measure the CSI-RS at the time point predicted by the fourth CSI information (i.e., the second time point), and compare the measurement result as the first CSI information of the third time point with the fourth CSI information to ensure the accuracy of the obtained monitoring result; or the user equipment can measure the CSI-RS associated (such as adjacent) with the time point predicted by the fourth CSI information (i.e., the second time point), and compare the measurement result as the first CSI information of the third time point with the fourth CSI information to ensure the normal acquisition of the measurement result when there is no CSI-RS at the second time point.

[0195] For example, the overall performance parameter SGCS1 characterizing the prediction and compression performance of the first prediction compression model and the information based thereon can be expressed as SGCS1{W N*B , W' N*B}, where W N*B represents the actual precoding information measured by the user equipment (ie, the first CSI information at the third time point), W' N*B Indicates the relevant decompression result of the network side device (ie, the fourth CSI information).

[0196] It should be noted that the overall performance parameter SGCS1, the prediction performance parameter SGCS3 or the compression performance parameter SGCS2 in the embodiment of the present application can adopt parameters such as cosine similarity that can measure model performance. For example, the above SGCS1 can be W N*B and W' N*B The cosine similarity between N*B , W' N*B}), similarly, SGCS2 and SGCS3 can also be obtained based on the cosine similarity between the information they are based on; if the network side device performs the above overall performance monitoring, the user equipment needs to report W to the network side device. N*B If the user equipment performs the above overall performance monitoring, the network side device may send W 'to the user equipment N*B , or the user equipment will simulate the decompression performed by the network side device to obtain the decompression result W" N*B As W' N*B .

[0197] In some embodiments, if multiple first CSI information and fourth CSI information are obtained, statistical characteristic values ​​such as weighted mean and median of multiple overall performance parameters determined based on the multiple first CSI information and fourth CSI information can be further obtained, and the obtained statistical characteristic values ​​can be used as the overall monitoring results of the prediction performance and compression performance of the first prediction compression model.

[0198] For example, the above statistical characteristic values ​​can be determined by the following formula:

[0199] Among them, SGCS{W i , W i '} represents the first CSI information W according to the i-th i and the fourth CSI information W associated therewith i ' The overall performance parameter determined, N represents the number of first CSI information (or fourth CSI information) obtained, and sta represents the statistical characteristic value (in this case, the mean value) of the overall performance parameter.

[0200] In some embodiments, for a cascaded AI model (such as a second prediction compression model, which is formed by cascading a second compression model after a second prediction model), the network side device or the user device can obtain the monitoring results of the compression performance of the second compression model (the acquisition method can refer to the following implementation method two), and obtain the monitoring results of the prediction performance of the second prediction model (the acquisition method can refer to the following implementation method three); then, based on the monitoring results of the compression performance of the second compression model and the monitoring results of the prediction performance of the second prediction model, the overall monitoring results of the prediction performance and compression performance of the second prediction compression model are obtained.

[0201] For example, for the second predictive compression model, the user device performs predictive performance monitoring to obtain the predicted performance parameter SGCS3 (i.e., the monitoring result of the predicted performance), and reports the predicted performance parameter SGCS3 to the network-side device so that the network-side device can calculate the overall performance parameter SGCS1 (i.e., the monitoring result of the overall performance of compression and prediction) based on the product relationship between the performance of the cascaded AI models. After the network-side device completes the decompression of the CSI information and obtains the associated predicted CSI information, it can calculate the compression performance parameter SGCS2 (i.e., the monitoring result of the compression performance), and then further calculate SGCS1 based on SGCS2 and SGCS3, such as by calculating SGCS1 using the formula SGCS1=SGCS2*SGCS3.

[0202] Implementation Method 2

[0203] Network-side devices or user devices monitor the compression performance of AI models in the following two situations.

[0204] Case 1: A monitoring result of the compression performance of the first compression model is obtained according to the fourth CSI information and the sixth CSI information in the second CSI information.

[0205] In case 1, the network side device or user equipment monitors the compression performance of the first compression model (which can be deployed independently or cascaded with other models) based on the decompressed CSI information (i.e., the fourth CSI information) and the compressed CSI information (i.e., the sixth CSI information).

[0206] Case 2: A monitoring result of the compression performance of the first compression model is obtained according to the fourth CSI information and the fifth CSI information in the second CSI information.

[0207] In case 2, the first compression model is included in the first prediction compression model and is cascaded after the first prediction model in the first prediction compression model. When the compression performance of the cascaded AI model (i.e., the first prediction compression model) is monitored, the network side device or the user device will compare the CSI information predicted for one or more time points (i.e., the fifth CSI information) with the associated decompressed CSI information (i.e., the fourth CSI information) to obtain the monitoring results of the compression performance of the cascaded AI model (i.e., the compression performance of the first compression model in the first prediction compression model).

[0208] For example, the compression performance parameter SGCS2 characterizing the compression performance of the first compression model and the information based thereon can be expressed as in, represents the CSI information (i.e., the fifth CSI information) predicted by the user equipment through the first prediction model, Indicates the relevant decompression result of the network side device (ie, the fourth CSI information).

[0209] It should be noted that if the network side device performs the above compression performance monitoring, the user device needs to report to the network side device Or the network side device uses the prediction result obtained by simulating the user equipment to perform the prediction as If the user equipment performs the above compression performance monitoring, the network side device can send a Or the user equipment uses the decompression result obtained by simulating the decompression performed by the network side equipment as

[0210] In some embodiments, if multiple fourth CSI information and fifth CSI information are obtained, statistical characteristic values ​​such as weighted mean and median of multiple compression performance parameters determined based on the multiple fourth CSI information and fifth CSI information can be further obtained, and the obtained statistical characteristic values ​​can be used as monitoring results of the compression performance of the first compression model.

[0211] For example, the above statistical characteristic values ​​can be determined by the following formula:

[0212] in, Indicates the lth fourth CSI information and the fifth CSI information associated therewith The compression performance parameter is determined, N represents the number of fifth CSI information (or fourth CSI information) obtained, and stb represents the statistical characteristic value (in this case, the mean value).

[0213] In some embodiments, for a cascaded AI model (such as a second prediction compression model, which includes a second prediction model and a second compression model cascaded after the second prediction model), the network side device or the user device can obtain the monitoring results of the prediction performance of the second prediction model (the acquisition method can refer to the following implementation method three), and obtain the overall monitoring results of the prediction performance and compression performance of the second prediction compression model (the acquisition method can refer to the above implementation method one); then, based on the monitoring results of the prediction performance of the second prediction model and the overall monitoring results of the prediction performance and compression performance of the second prediction compression model, the monitoring results of the compression performance of the second compression model are obtained.

[0214] For example, for the second predictive compression model, the user device performs predictive performance monitoring to obtain the predicted performance parameter SGCS3 (i.e., the monitoring result of the predicted performance), and reports the predicted performance parameter SGCS3 to the network-side device so that the network-side device can calculate the compression performance parameter SGCS2 (i.e., the monitoring result of the compression performance) based on the product relationship between the performance of the cascaded AI models. After the network-side device completes the decompression of the CSI information and obtains the associated actual CSI information, it can calculate the overall performance parameter SGCS1 (i.e., the monitoring result of the overall performance of compression and prediction), and then further calculate SGCS2 based on SGCS1 and SGCS3, such as by calculating SGCS2 using the formula SGCS2 = SGCS1 / SGCS3.

[0215] For another example, for the second predictive compression model, the user equipment performs overall performance monitoring and predictive performance monitoring. When receiving the decompressed CSI information sent by the network-side device (or simulating the decompression performed by the network-side device), SGCS1 is calculated in combination with the actual CSI information, and SGCS3 is calculated based on the actual CSI information and the predicted CSI information measured by the user equipment. Subsequently, SGCS2 can be further calculated based on SGCS1 and SGCS3, such as SGCS2 calculated by the formula SGCS2 = SGCS1 / SGCS3.

[0216] Implementation Method 3

[0217] Network-side devices or user devices monitor the prediction performance of the AI ​​model.

[0218] In this embodiment, the network side device or the user device compares the actual CSI information at one or more time points (i.e., the first CSI information at the third time point) with the CSI information obtained by predicting the associated time point (i.e., the fifth CSI information, which can be the CSI information at the second time point obtained after the first prediction model executes the prediction) to obtain the monitoring results of the prediction performance of the AI ​​model (i.e., the first prediction model).

[0219] In some embodiments, the first CSI information at the third time point may be obtained by the user equipment by measuring the CSI-RS at the third time point.

[0220] The third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining the monitoring results.

[0221] Specifically, the user equipment can measure the CSI-RS at the time point indicated by the network side device for obtaining the monitoring result, and compare the measurement result as the first CSI information of the third time point with the fifth CSI information, so as to obtain the monitoring result that the network side device expects the user equipment to report; or the user equipment can measure the CSI-RS at the time point predicted by the fourth CSI information (i.e., the second time point), and compare the measurement result as the first CSI information of the third time point with the fifth CSI information to ensure the accuracy of the obtained monitoring result; or the user equipment can measure the CSI-RS associated (such as adjacent) with the time point predicted by the fourth CSI information (i.e., the second time point), and compare the measurement result as the first CSI information of the third time point with the fifth CSI information to ensure the normal acquisition of the measurement result when there is no CSI-RS at the second time point.

[0222] For example, the prediction performance parameter SGCS3 characterizing the prediction performance of the first prediction model and the information based thereon can be expressed as Among them, W N*B represents the actual CSI measured by the user equipment (i.e., the first CSI information at the third time point), Indicates the CSI information predicted by the user equipment (ie, the fifth CSI information).

[0223] It should be noted that if the network side device performs the above-mentioned prediction performance monitoring, the user equipment needs to report W to the network side device. N*B The user equipment can report to the network side equipment Or the network side device uses the prediction result obtained by simulating the user equipment to perform the prediction as

[0224] In some embodiments, for a cascaded AI model (such as a second prediction compression model, which includes a second compression model and a second prediction model cascaded before the second compression model), the network side device or the user device can obtain the monitoring results of the compression performance of the second compression model (the acquisition method can refer to the above-mentioned embodiment 2), and obtain the overall monitoring results of the prediction performance and compression performance of the second prediction compression model (the acquisition method can refer to the above-mentioned embodiment 1); then, based on the monitoring results of the compression performance of the second compression model and the overall monitoring results of the prediction performance and compression performance of the second prediction compression model, the monitoring results of the prediction performance of the second prediction model are obtained.

[0225] For example, for the second predictive compression model, the network side device performs compression performance monitoring and overall performance monitoring to obtain the compression performance parameter SGCS2 (i.e., the monitoring result of the compression performance) and the overall performance parameter SGCS1 (i.e., the monitoring result of the overall performance of compression and prediction). Subsequently, SGCS3 can be further calculated based on SGCS1 and SGCS2, such as SGCS3 calculated by the formula SGCS3 = SGCS1 / SGCS2.

[0226] In a second aspect of an embodiment of the present application, another CSI processing method is provided. The method is performed by a network-side device. FIG9 is a flowchart illustrating an implementation of another CSI processing method provided in an embodiment of the present application. The method may include the following steps:

[0227] Step S201: The network side device sends first reporting configuration information to the user equipment UE, where the first reporting configuration information includes at least one of the following: compression configuration information of the channel state information CSI, and prediction configuration information of the CSI.

[0228] Among them, the first reporting configuration information is used to request the UE to report the first CSI information of the UE processed by the artificial intelligence AI model to obtain second CSI information; the second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

[0229] It can be seen from the above steps that the present application introduces CSI processing logic combined with the first reporting configuration information to the user equipment, thereby providing a solution for how to implement AI-based prediction and compression CSI processing and how to monitor the performance of AI-based prediction and compression CSI processing.

[0230] In some embodiments, the method further comprises:

[0231] The network side device receives at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information reported by the UE;

[0232] The fifth CSI information is: CSI information of the UE predicted by the UE at a future time point.

[0233] In some embodiments, the method further comprises at least one of the following:

[0234] The network-side device decompresses the second CSI information using the first decompression model to obtain the fourth CSI information;

[0235] The network side device sends the fourth CSI information to the UE.

[0236] In some embodiments, the method further comprises:

[0237] The network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information;

[0238] Among them, the fourth CSI information is: the CSI information obtained by the network side device or the UE decompressing the second CSI information, the fifth CSI information is: the CSI information of the UE predicted by the UE at a future time point, and the sixth CSI information is the CSI information obtained by the UE compressing the first CSI information.

[0239] In some embodiments, the network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, including:

[0240] The network-side device obtains, based on the first CSI information and the fourth CSI information of the UE at the third time point, an overall monitoring result of the prediction performance and the compression performance of the first prediction compression model;

[0241] Among them, the third time point is at least one of the following: the second time point, a time point associated with the second time point, and the time point indicated by the network side device to obtain the monitoring results; the second time point is a time point after the first time point.

[0242] In some implementations, the information reported by the UE and received by the network-side device includes at least one of the following:

[0243] first CSI information of the UE at a third time point;

[0244] The third CSI information is obtained by processing the first CSI information of the UE at the first time point using a first prediction compression model.

[0245] In some embodiments, the network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, including:

[0246] The network-side device obtains a monitoring result of compression performance of the first compression model based on the fourth CSI information and fifth CSI information in the second CSI information;

[0247] The fifth CSI information in the second CSI information is obtained by the UE predicting the CSI information of the UE at the second time point based on the first CSI information of the UE at the first time point using a first prediction model, and the second time point is a time point after the first time point; the first time point is at least one of the following: the current time point, a historical time point or a time point of a CSI reference resource.

[0248] In some embodiments, the network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, including:

[0249] The network-side device obtains a monitoring result of compression performance of the first compression model based on the fourth CSI information and sixth CSI information in the second CSI information;

[0250] Among them, the sixth CSI information in the second CSI information is obtained by the UE compressing the CSI information of the UE at the first time point using the first compression model based on the first CSI information of the UE at the first time point, and the first time point is at least one of the following: the current time point, the historical time point or the time point of the CSI reference resource.

[0251] In some embodiments, the method further comprises:

[0252] The network-side device receives the monitoring result of the prediction performance of the second prediction model reported by the UE, and obtains the monitoring result of the compression performance of the second compression model;

[0253] The network side device obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model based on the monitoring result of the compression performance of the second compression model and the monitoring result of the prediction performance of the second prediction model.

[0254] In some embodiments, the method further comprises:

[0255] The network-side device receives the monitoring result of the prediction performance of the second prediction model reported by the UE, and obtains the overall monitoring result of the prediction performance and compression performance of the second prediction compression model;

[0256] The network side device obtains the monitoring result of the compression performance of the second compression model based on the monitoring result of the prediction performance of the second prediction model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

[0257] In some embodiments, the method further comprises:

[0258] The network-side device obtains a monitoring result of the compression performance of the second compression model, and obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model;

[0259] The network side device obtains the monitoring result of the prediction performance of the second prediction model based on the monitoring result of the compression performance of the second compression model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

[0260] It should be noted that the CSI processing method embodiment provided in the embodiment of the present application is a network side device side method embodiment corresponding to the CSI processing method embodiment described in the first aspect, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0261] The CSI processing method provided in the embodiment of the present application may be executed by a CSI processing device. In the embodiment of the present application, the CSI processing device provided in the embodiment of the present application is described by taking the CSI processing method performed by the CSI processing device as an example.

[0262] In a third aspect, an embodiment of the present application provides a CSI processing device, which can be applied to a user equipment. As shown in FIG10 , the CSI processing device 100 includes:

[0263] The first receiving module 101 is configured to receive first reporting configuration information sent by a network-side device, where the first reporting configuration information includes at least one of the following: compression configuration information of channel state information (CSI) and prediction configuration information of the CSI;

[0264] A first processing module 102 is configured to process the first CSI information of the UE using an artificial intelligence (AI) model according to the first reporting configuration information to obtain second CSI information;

[0265] The second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

[0266] Optionally, the device further comprises:

[0267] a first monitoring module, configured to monitor the performance of the AI ​​model based on at least two items of the first to sixth CSI information;

[0268] The fourth CSI information is: CSI information obtained by decompressing the second CSI information, the fifth CSI information is: predicted CSI information of the UE at a future time point, and the sixth CSI information is CSI information obtained by compressing the first CSI information.

[0269] Optionally, the second CSI information includes the third CSI information; and the first processing module 102 includes:

[0270] a first processing submodule, configured to obtain, according to the first reporting configuration information, first CSI information of the UE at a first time point, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource;

[0271] The second processing submodule is used to predict the CSI information of the UE at the second time point based on the first CSI information of the UE at the first time point using a first prediction compression model and compress the predicted CSI information using the first prediction compression model to obtain the third CSI information, where the second time point is a time point after the first time point.

[0272] Optionally, the first monitoring module includes:

[0273] a first monitoring submodule, configured to obtain first CSI information of the UE at a third time point, where the third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining a monitoring result;

[0274] The second monitoring submodule is used to obtain an overall monitoring result of the prediction performance and compression performance of the first prediction compression model based on the first CSI information and the fourth CSI information of the UE at a third time point.

[0275] Optionally, the device further comprises at least one of the following:

[0276] a second receiving module, configured to receive the fourth CSI information sent by the network-side device, where the fourth CSI information is obtained by the network-side device decompressing the second CSI information using the first decompression model;

[0277] The second processing module is used to decompress the second CIS information by using the first reference model of the first decompression model to obtain the fourth CSI information.

[0278] Optionally, the second CSI information further includes fifth CSI information; and the first processing module 102 includes:

[0279] a third processing submodule, configured to obtain, according to the first reporting configuration information, first CSI information of the UE at a first time point, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource;

[0280] The fourth processing submodule is used to predict the CSI information of the UE at a second time point using a first prediction model based on the first CSI information of the UE at the first time point, to obtain the fifth CSI information in the second CSI information, where the second time point is a time point after the first time point.

[0281] Optionally, the first monitoring module includes:

[0282] a third monitoring submodule, configured to obtain first CSI information of the UE at a third time point, where the third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining a monitoring result;

[0283] The fourth monitoring submodule is used to obtain a monitoring result of the prediction performance of the first prediction model based on the first CSI information of the UE at the third time point and the fifth CSI information in the second CSI information.

[0284] Optionally, the first monitoring module includes:

[0285] The fifth monitoring submodule is used to obtain a monitoring result of the compression performance of the first compression model based on the fourth CSI information and the fifth CSI information in the second CSI information.

[0286] Optionally, the second CSI information further includes sixth CSI information; and the first processing module 102 includes:

[0287] a fifth processing submodule, configured to obtain, according to the first reporting configuration information, first CSI information of the UE at a first time point, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource;

[0288] The sixth processing submodule is configured to compress the first CSI information of the UE at the first time point by using the first compression model to obtain sixth CSI information in the second CSI information.

[0289] Optionally, the first monitoring module includes:

[0290] The sixth monitoring submodule is used to obtain a monitoring result of the compression performance of the first compression model based on the fourth CSI information and the sixth CSI information in the second CSI information.

[0291] Optionally, the device further comprises:

[0292] A first acquisition module is used to obtain a monitoring result of the prediction performance of the second prediction model and obtain an overall monitoring result of the prediction performance and compression performance of the second prediction compression model;

[0293] The second monitoring module is used to obtain the monitoring result of the compression performance of the second compression model based on the monitoring result of the prediction performance of the second prediction model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

[0294] Optionally, the device further comprises:

[0295] A second acquisition module is used to obtain a monitoring result of the compression performance of the second compression model, and obtain an overall monitoring result of the prediction performance and compression performance of the second prediction compression model;

[0296] The third monitoring module is used to obtain the monitoring result of the prediction performance of the second prediction model based on the monitoring result of the compression performance of the second compression model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

[0297] Optionally, the device further comprises:

[0298] a third acquisition module, configured to obtain a monitoring result of the compression performance of the second compression model and a monitoring result of the prediction performance of the second prediction model;

[0299] The fourth monitoring module is used to obtain the overall monitoring results of the prediction performance and compression performance of the second prediction compression model based on the monitoring results of the compression performance of the second compression model and the monitoring results of the prediction performance of the second prediction model.

[0300] Optionally, the second CSI information further includes CSI information at at least two time points compressed in the time domain;

[0301] The CSI information at the at least two time points satisfies any one of the following:

[0302] The CSI information at the at least two time points is CSI information of the UE at at least two second time points predicted by the AI ​​model;

[0303] At least one of the CSI information at the at least two time points is CSI information of the UE at at least one second time point predicted by the AI ​​model;

[0304] At least one of the CSI information at the at least two time points is first CSI information of the UE at at least one first time point acquired by the UE.

[0305] Optionally, the at least two time points include at least one of the following:

[0306] The Xth time point, wherein the Xth time point is the first time point, or the third time point;

[0307] The X+Kth time point, wherein the X+Kth time point is a time point K away from the Xth time point;

[0308] The X+NKth time point, where the X+NKth time point is a time point that is NK away from the Xth time point, and N is an integer greater than 1.

[0309] Optionally, K satisfies at least one of the following:

[0310] K is the time interval between two time points;

[0311] K is 1 time slot, 2 time slots, 4 time slots or 5 time slots;

[0312] The value of K is determined according to the value of D, D is determined according to the instruction of the network side device, or D is the interval between two adjacent first time points.

[0313] Optionally, the CSI information at at least two time points after time domain compression satisfies at least one of the following:

[0314] The CSI information at the Xth time point is CSI information that has not been time-domain compressed, or the CSI information at the Xth time point is CSI information obtained by performing time-domain compression on CSI information reported by the UE to the network-side device before the Xth time point;

[0315] The CSI information at the X+Kth time point is CSI information obtained by performing time domain compression on the CSI information at the Xth time point, where the X+Kth time point is K time points away from the Xth time point.

[0316] The CSI information at the X+NKth time point is CSI information obtained by time domain compression using the CSI information of at least one time point from the Xth time point to the X+(N-1)Kth time point, where the X+NKth time point is a time point NK away from the Xth time point, where N is an integer greater than 1.

[0317] Optionally, the CSI information at at least two time points after time domain compression satisfies at least one of the following:

[0318] The CSI information at the at least two time points has different payloads;

[0319] The CSI information at the at least two time points has different encoding and decoding modes;

[0320] There is a sequence for compression and decompression of the CSI information at the at least two time points.

[0321] Optionally, the device further comprises:

[0322] The first reporting module is configured to report at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information to the network side device.

[0323] Optionally, the device further comprises:

[0324] The second reporting module is configured to report at least one of the following to the network-side device:

[0325] CSI information of the UE at at least one time point between the X+K time point and the X+NK time point predicted by the UE through the AI ​​model;

[0326] CSI information at the Mth time point or the M-1th time point obtained by compressing the CSI information at M time points by the UE using the AI ​​model;

[0327] The UE predicts CSI information at at least one time point from the X+Kth time point to the X+NKth time point through the AI ​​model, and compresses the predicted CSI information to obtain CSI information.

[0328] Optionally, the first processing submodule or the third processing submodule includes:

[0329] a seventh processing submodule, configured to obtain WM-D, WM-2D, ..., WM-nD based on measurement results of multiple channel state information reference signals (CSI-RS);

[0330] The WM-D represents the codebook information or channel information obtained by the UE at the MDth time point, and MD, M-2D, ...M-nD represent different first time points.

[0331] Optionally, the M is determined based on the CSI-RS most recently before the Xth time point, D is the period or repetition interval of the CSI-RS, or D is the interval between the multiple CSI-RSs.

[0332] The CSI processing device provided in the embodiment of the present application can implement the various processes implemented in the embodiment of the CSI processing method described in the first aspect and achieve the same technical effects. To avoid repetition, they are not described here.

[0333] In a fourth aspect, an embodiment of the present application provides another CSI processing device, which can be applied to a network-side device. As shown in FIG11 , the CSI processing device 200 includes:

[0334] The configuration information sending module 201 is configured to send first reporting configuration information to a user equipment UE, where the first reporting configuration information includes at least one of the following: compression configuration information of channel state information (CSI) and prediction configuration information of the CSI;

[0335] Among them, the first reporting configuration information is used for the UE to process the first CSI information of the UE using an artificial intelligence AI model to obtain second CSI information; the second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

[0336] Optionally, the device further comprises:

[0337] A CSI information receiving module, configured to receive at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information reported by the UE;

[0338] The fifth CSI information is: CSI information of the UE predicted by the UE at a future time point.

[0339] Optionally, the device further comprises at least one of the following:

[0340] a decompression module, configured to decompress the second CSI information using a first decompression model to obtain the fourth CSI information;

[0341] The CSI information sending module is configured to send the fourth CSI information to the UE.

[0342] Optionally, the device further comprises:

[0343] a performance monitoring module, configured to monitor the performance of the AI ​​model based on at least two items of the first to sixth CSI information;

[0344] Among them, the fourth CSI information is: the CSI information obtained by the network side device or the UE decompressing the second CSI information, the fifth CSI information is: the CSI information of the UE predicted by the UE at a future time point, and the sixth CSI information is the CSI information obtained by the UE compressing the first CSI information.

[0345] Optionally, the performance monitoring module includes:

[0346] A first performance monitoring submodule, configured to obtain an overall monitoring result of prediction performance and compression performance of the first prediction compression model based on the first CSI information and the fourth CSI information of the UE at a third time point;

[0347] Among them, the third time point is at least one of the following: the second time point, a time point associated with the second time point, and the time point indicated by the network side device to obtain the monitoring results; the second time point is a time point after the first time point.

[0348] Optionally, the information reported by the UE and received by the network side device includes at least one of the following:

[0349] first CSI information of the UE at a third time point;

[0350] The third CSI information is obtained by processing the first CSI information of the UE at the first time point using a first prediction compression model.

[0351] Optionally, the performance monitoring module includes:

[0352] a second performance monitoring submodule, configured to obtain a monitoring result of the compression performance of the first compression model according to the fourth CSI information and fifth CSI information in the second CSI information;

[0353] The fifth CSI information in the second CSI information is obtained by the UE predicting the CSI information of the UE at the second time point based on the first CSI information of the UE at the first time point using a first prediction model, and the second time point is a time point after the first time point; the first time point is at least one of the following: the current time point, a historical time point or a time point of a CSI reference resource.

[0354] Optionally, the performance monitoring module includes:

[0355] a third performance monitoring submodule, configured to obtain a monitoring result of the compression performance of the first compression model according to the fourth CSI information and sixth CSI information in the second CSI information;

[0356] Among them, the sixth CSI information in the second CSI information is obtained by the UE compressing the CSI information of the UE at the first time point using the first compression model based on the first CSI information of the UE at the first time point, and the first time point is at least one of the following: the current time point, the historical time point or the time point of the CSI reference resource.

[0357] Optionally, the device further comprises:

[0358] a first result acquisition module, configured to receive a monitoring result of the prediction performance of the second prediction model reported by the UE, and obtain a monitoring result of the compression performance of the second compression model;

[0359] The first result processing module is used to obtain the overall monitoring results of the prediction performance and compression performance of the second prediction compression model based on the monitoring results of the compression performance of the second compression model and the monitoring results of the prediction performance of the second prediction model.

[0360] Optionally, the device further comprises:

[0361] a second result acquisition module, configured to receive the monitoring result of the prediction performance of the second prediction model reported by the UE, and obtain the overall monitoring result of the prediction performance and compression performance of the second prediction compression model;

[0362] The second result processing module is used to obtain the monitoring result of the compression performance of the second compression model based on the monitoring result of the prediction performance of the second prediction model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

[0363] Optionally, the device further comprises:

[0364] a third result acquisition module, configured to obtain a monitoring result of the compression performance of the second compression model, and obtain an overall monitoring result of the prediction performance and compression performance of the second prediction compression model;

[0365] The third result processing module is used to obtain the monitoring result of the prediction performance of the second prediction model based on the monitoring result of the compression performance of the second compression model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

[0366] It should be noted that the CSI processing device provided in the embodiment of the present application can implement the various processes implemented in the CSI processing method embodiment described in the second aspect and achieve the same technical effect. To avoid repetition, it will not be described here.

[0367] The CSI processing device in the embodiments of the present application can be an electronic device, such as an electronic device with an operating system, or a component of an electronic device, such as an integrated circuit or chip. The electronic device can be a terminal or other device other than a terminal. For example, the terminal can include but is not limited to the types of terminal 11 listed above, and other devices can include servers, network attached storage (NAS), etc., which are not specifically limited in the embodiments of the present application.

[0368] As shown in Figure 12, an embodiment of the present application further provides a communication device 1200, including a processor 1201 and a memory 1202. The memory 1202 stores a program or instruction executable on the processor 1201. For example, when the communication device 1200 is a terminal, the program or instruction, when executed by the processor 1201, implements the various steps of the embodiment of the CSI processing method described in the first aspect above, and can achieve the same technical effects. When the communication device 1200 is a network-side device, the program or instruction, when executed by the processor 1201, implements the various steps of the embodiment of the CSI processing method described in the second aspect above, and can achieve the same technical effects. To avoid repetition, these steps are not further described here.

[0369] An embodiment of the present application also provides a user equipment (UE), including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the CSI processing method embodiment described in the first aspect. This UE embodiment corresponds to the aforementioned UE-side method embodiment, and each implementation process and implementation method of the aforementioned UE-side method embodiment is applicable to this UE embodiment and can achieve the same technical effects. Specifically, Figure 13 is a schematic diagram of the hardware structure of a UE implementing an embodiment of the present application.

[0370] The user equipment 1300 includes but is not limited to: a radio frequency unit 1301, a network module 1302, an audio output unit 1303, an input unit 1304, a sensor 1305, a display unit 1306, a user input unit 1307, an interface unit 1308, a memory 1309 and at least some of the components of the processor 1310.

[0371] Those skilled in the art will appreciate that the user equipment 1300 may also include a power supply (such as a battery) to power various components. The power supply may be logically connected to the processor 1310 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system. The user equipment structure shown in Figure 13 does not constitute a limitation of the user equipment. The user equipment may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be described in detail here.

[0372] It should be understood that in an embodiment of the present application, the input unit 1304 may include a graphics processor 13041 and a microphone 13042, and the graphics processor 13041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 1306 may include a display panel 13061, and the display panel 13061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1307 includes a touch panel 13071 and at least one of other input devices 13072. The touch panel 13071 is also called a touch screen. The touch panel 13071 may include two parts: a touch detection device and a touch controller. Other input devices 13072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0373] In the embodiment of the present application, after receiving downlink data from a network-side device, the RF unit 1301 may transmit the data to the processor 1310 for processing. Furthermore, the RF unit 1301 may send uplink data to the network-side device. Typically, the RF unit 1301 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like.

[0374] The memory 1309 can be used to store software programs or instructions and various data. The memory 1309 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 1309 may include a volatile memory or a non-volatile memory. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM). The memory 1309 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0375] Processor 1310 may include one or more processing units. Optionally, processor 1310 integrates an application processor and a modem processor. The application processor primarily handles operations related to the operating system, user interface, and application programs, while the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into processor 1310.

[0376] It can be understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the CSI processing method embodiment described in the first aspect, and achieve the same or corresponding technical effects. To avoid repetition, it will not be repeated here.

[0377] An embodiment of the present application further provides a network-side device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the CSI processing method embodiment described in the second aspect. This network-side device embodiment corresponds to the aforementioned network-side device method embodiment, and each implementation process and implementation method of the aforementioned network-side device method embodiment is applicable to this network-side device embodiment and can achieve the same technical effects.

[0378] Specifically, embodiments of the present application also provide a network-side device. As shown in Figure 14, network-side device 1400 includes an antenna 141, a radio frequency device 142, a baseband device 143, a processor 144, and a memory 145. Antenna 141 is connected to radio frequency device 142. In the uplink direction, radio frequency device 142 receives information via antenna 141 and sends the received information to baseband device 143 for processing. In the downlink direction, baseband device 143 processes the information to be transmitted and sends it to radio frequency device 142. Radio frequency device 142 processes the received information and then sends it through antenna 141.

[0379] The method executed by the network-side device in the above embodiment may be implemented in the baseband device 143 , which includes a baseband processor.

[0380] The baseband device 143 may include, for example, at least one baseband board, on which multiple chips are arranged, as shown in Figure 14, one of the chips is, for example, a baseband processor, which is connected to the memory 145 through a bus interface to call the program or instructions in the memory 145 to execute the network device operations shown in the above method embodiment.

[0381] The network side device may further include a network interface 146 , which is, for example, a Common Public Radio Interface (CPRI).

[0382] In addition, the network-side device 1400 of an embodiment of the present invention also includes: a program or instruction stored in the memory 145 and executable on the processor 144. The processor 144 calls the program or instruction in the memory 145 to execute the method executed by each module in the CSI processing device described in the fourth aspect and achieves the same technical effect. To avoid repetition, it will not be described here.

[0383] An embodiment of the present application further provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the CSI processing method embodiment described in the first or second aspect above are implemented, and the same technical effects can be achieved. To avoid repetition, they are not described here.

[0384] The processor is the processor in the terminal or the processor in the network-side device described in the above embodiments. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk. In some examples, the readable storage medium may be a non-transitory readable storage medium.

[0385] An embodiment of the present application further provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the various processes of the CSI processing method embodiment described in the first or second aspect above, and to achieve the same technical effects. To avoid repetition, these processes are not described herein.

[0386] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0387] An embodiment of the present application further provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the CSI processing method embodiment described in the first or second aspect above, and can achieve the same technical effects. To avoid repetition, they are not further described here.

[0388] An embodiment of the present application also provides a wireless communication system, including: a user device and a network side device, wherein the user device can be used to perform the steps of the CSI processing method described in the first aspect above, and the network side device can be used to perform the steps of the CSI processing method described in the second aspect above.

[0389] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0390] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of a computer software product plus a necessary general-purpose hardware platform, or of course, by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes a number of instructions for enabling a terminal or network-side device to execute the methods described in each embodiment of the present application.

[0391] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms of implementation methods without departing from the purpose of this application and the scope of protection of the claims. These implementation methods are all within the protection of this application.

Claims

1. A CSI processing method, wherein: include: The user equipment UE receives first reporting configuration information sent by the network side device, where the first reporting configuration information includes at least one of the following: compression configuration information of the channel state information CSI, and prediction configuration information of the CSI; The UE processes the first CSI information of the UE using an artificial intelligence (AI) model according to the first reporting configuration information to obtain second CSI information; The second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

2. The method according to claim 1, wherein The method further comprises: The UE monitors the performance of the AI ​​model according to at least two items of the first to sixth CSI information; The fourth CSI information is: CSI information obtained by decompressing the second CSI information, the fifth CSI information is: predicted CSI information of the UE at a future time point, and the sixth CSI information is CSI information obtained by compressing the first CSI information.

3. The method according to claim 2, wherein: The second CSI information includes the third CSI information; and the UE processes the first CSI information of the UE using the AI ​​model according to the first reporting configuration information to obtain the second CSI information, including: Acquiring, by the UE, first CSI information of the UE at a first time point according to the first reporting configuration information, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource; The UE predicts the CSI information of the UE at a second time point based on the first CSI information of the UE at a first time point using a first prediction compression model and compresses the predicted CSI information using the first prediction compression model to obtain the third CSI information, where the second time point is a time point after the first time point.

4. The method according to claim 3, wherein: The UE monitors the performance of the AI ​​model according to at least two items of the first to sixth CSI information, including: The UE obtains, by the UE, first CSI information of the UE at a third time point, where the third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining a monitoring result; The UE obtains an overall monitoring result of the prediction performance and compression performance of the first prediction compression model based on the first CSI information and the fourth CSI information of the UE at the third time point.

5. The method according to any one of claims 2 to 4, wherein: The method further comprises at least one of the following: The UE receives the fourth CSI information sent by the network side device, where the fourth CSI information is obtained by the network side device decompressing the second CSI information using the first decompression model; The UE decompresses the second CIS information using the first reference model of the first decompression model to obtain the fourth CSI information.

6. The method according to claim 2, wherein: The second CSI information also includes fifth CSI information; the UE processes the first CSI information of the UE using the AI ​​model according to the first reporting configuration information to obtain the second CSI information, including: Acquiring, by the UE, first CSI information of the UE at a first time point according to the first reporting configuration information, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource; The UE predicts the CSI information of the UE at a second time point using a first prediction model based on the first CSI information of the UE at the first time point to obtain fifth CSI information in the second CSI information, where the second time point is a time point after the first time point.

7. The method according to claim 6, wherein: The UE monitors the performance of the AI ​​model according to at least two items of the first to sixth CSI information, including: The UE obtains, by the UE, first CSI information of the UE at a third time point, where the third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining a monitoring result; The UE obtains a monitoring result of the prediction performance of the first prediction model based on the first CSI information of the UE at a third time point and the fifth CSI information in the second CSI information.

8. The method according to claim 6, wherein: The UE monitors the performance of the AI ​​model according to at least two items of the first to sixth CSI information, including: The UE obtains a monitoring result of the compression performance of the first compression model based on the fourth CSI information and the fifth CSI information in the second CSI information.

9. The method according to claim 2, wherein: The second CSI information also includes sixth CSI information; The UE processes the first CSI information of the UE using the AI ​​model according to the first reporting configuration information to obtain second CSI information, including: Acquiring, by the UE, first CSI information of the UE at a first time point according to the first reporting configuration information, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource; The UE compresses the first CSI information of the UE at the first time point by using the first compression model to obtain sixth CSI information in the second CSI information.

10. The method according to claim 9, wherein: The UE monitors the performance of the AI ​​model according to at least two items of the first to sixth CSI information, including: The UE obtains a monitoring result of the compression performance of the first compression model based on the fourth CSI information and the sixth CSI information in the second CSI information.

11. The method according to claim 1, wherein The method further comprises: The UE obtains a monitoring result of the prediction performance of the second prediction model, and obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model; The UE obtains the monitoring result of the compression performance of the second compression model based on the monitoring result of the prediction performance of the second prediction model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

12. The method according to claim 1, wherein The method further comprises: The UE obtains a monitoring result of the compression performance of the second compression model, and obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model; The UE obtains the monitoring result of the prediction performance of the second prediction model based on the monitoring result of the compression performance of the second compression model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

13. The method according to claim 1, wherein The method further comprises: The UE obtains a monitoring result of compression performance of the second compression model and obtains a monitoring result of prediction performance of the second prediction model; The UE obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model based on the monitoring result of the compression performance of the second compression model and the monitoring result of the prediction performance of the second prediction model.

14. The method according to any one of claims 1 to 13, wherein: The second CSI information further includes CSI information at at least two time points compressed in the time domain; The CSI information at the at least two time points satisfies any one of the following: The CSI information at the at least two time points is CSI information of the UE at at least two second time points predicted by the AI ​​model; At least one of the CSI information at the at least two time points is CSI information of the UE at at least one second time point predicted by the AI ​​model; At least one of the CSI information at the at least two time points is first CSI information of the UE at at least one first time point acquired by the UE.

15. The method according to claim 14, wherein The at least two time points include at least one of the following: The Xth time point, wherein the Xth time point is the first time point, or the third time point; The X+Kth time point, wherein the X+Kth time point is a time point K away from the Xth time point; The X+NKth time point, where the X+NKth time point is a time point that is NK away from the Xth time point, and N is an integer greater than 1.

16. The method according to claim 15, wherein K satisfies at least one of the following: K is the time interval between two time points; K is 1 time slot, 2 time slots, 4 time slots or 5 time slots; The value of K is determined according to the value of D, D is determined according to the instruction of the network side device, or D is the interval between two adjacent first time points.

17. The method according to claim 15 or 16, wherein The CSI information at at least two time points after time domain compression satisfies at least one of the following: The CSI information at the Xth time point is CSI information that has not been time-domain compressed, or the CSI information at the Xth time point is CSI information obtained by performing time-domain compression on CSI information reported by the UE to the network-side device before the Xth time point; The CSI information at the X+Kth time point is CSI information obtained by performing time domain compression on the CSI information at the Xth time point, where the X+Kth time point is K time points away from the Xth time point. The CSI information at the X+NKth time point is CSI information obtained by time domain compression using the CSI information of at least one time point from the Xth time point to the X+(N-1)Kth time point, where the X+NKth time point is a time point NK away from the Xth time point, where N is an integer greater than 1.

18. The method according to any one of claims 14 to 17, wherein: The CSI information at at least two time points after time domain compression satisfies at least one of the following: The CSI information at the at least two time points has different payloads; The CSI information at the at least two time points has different encoding and decoding modes; There is a sequence for compression and decompression of the CSI information at the at least two time points.

19. The method according to any one of claims 1 to 18, wherein: The method further comprises: The UE reports at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information to the network side device.

20. The method according to any one of claims 1 to 19, wherein: The UE reports at least one of the following to the network-side device: CSI information of the UE at at least one time point between the X+K time point and the X+NK time point predicted by the UE through the AI ​​model; CSI information at the Mth time point or the M-1th time point obtained by compressing the CSI information at M time points by the UE using the AI ​​model; The UE predicts CSI information at at least one time point from the X+Kth time point to the X+NKth time point through the AI ​​model, and compresses the predicted CSI information to obtain CSI information.

21. The method according to claim 3 or 6, wherein The UE acquiring, according to the first reporting configuration information, first CSI information of the UE at a first time point, including: The UE obtains W based on the measurement results of multiple channel state information reference signals CSI-RS M-D ,W M-2D ,……W M-nD ; Wherein, the W M-D represents the codebook information or channel information obtained by the UE at the MDth time point, where MD, M-2D, ..., M-nD represent different first time points.

22. The method according to claim 21, wherein The M is determined according to the CSI-RS most recently before the Xth time point, D is the period or repetition interval of the CSI-RS, or D is the interval between the multiple CSI-RSs.

23. A CSI processing method, wherein: include: The network side device sends first reporting configuration information to the user equipment UE, where the first reporting configuration information includes at least one of the following: compression configuration information of the channel state information CSI, and prediction configuration information of the CSI; The first reporting configuration information is used to request the UE to report processing of the first CSI information of the UE using an artificial intelligence AI model to obtain second CSI information; The second CSI information includes at least third CSI information, where the third CSI information is: CSI information obtained by predicting CSI information of the UE at a future time point and compressing the predicted CSI information.

24. The method according to claim 23, wherein The method further comprises: The network side device receives at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information reported by the UE; The fifth CSI information is: CSI information of the UE predicted by the UE at a future time point.

25. The method according to claim 23 or 24, wherein The method further comprises at least one of the following: The network-side device decompresses the second CSI information using the first decompression model to obtain the fourth CSI information; The network side device sends the fourth CSI information to the UE.

26. The method according to any one of claims 23 to 25, wherein: The method further comprises: The network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information; Among them, the fourth CSI information is: the CSI information obtained by the network side device or the UE decompressing the second CSI information, the fifth CSI information is: the CSI information of the UE predicted by the UE at a future time point, and the sixth CSI information is the CSI information obtained by the UE compressing the first CSI information.

27. The method according to claim 26, wherein The network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, including: The network-side device obtains, based on the first CSI information and the fourth CSI information of the UE at the third time point, an overall monitoring result of the prediction performance and the compression performance of the first prediction compression model; Among them, the third time point is at least one of the following: the second time point, a time point associated with the second time point, and the time point indicated by the network side device to obtain the monitoring results; the second time point is a time point after the first time point.

28. The method according to claim 27, wherein The information reported by the UE and received by the network side device includes at least one of the following: first CSI information of the UE at a third time point; The third CSI information is obtained by processing the first CSI information of the UE at the first time point using a first prediction compression model.

29. The method according to claim 26, wherein The network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, including: The network-side device obtains a monitoring result of compression performance of the first compression model based on the fourth CSI information and fifth CSI information in the second CSI information; The fifth CSI information in the second CSI information is obtained by the UE predicting the CSI information of the UE at the second time point based on the first CSI information of the UE at the first time point using a first prediction model, and the second time point is a time point after the first time point; the first time point is at least one of the following: the current time point, a historical time point or a time point of a CSI reference resource.

30. The method of claim 26, wherein: The network-side device monitors the performance of the AI ​​model based on at least two items of the first to sixth CSI information, including: The network-side device obtains a monitoring result of compression performance of the first compression model based on the fourth CSI information and sixth CSI information in the second CSI information; Among them, the sixth CSI information in the second CSI information is obtained by the UE compressing the CSI information of the UE at the first time point using the first compression model based on the first CSI information of the UE at the first time point, and the first time point is at least one of the following: the current time point, a historical time point or a time point of a CSI reference resource.

31. The method according to any one of claims 23 to 25, wherein: The method further comprises: The network-side device receives the monitoring result of the prediction performance of the second prediction model reported by the UE, and obtains the monitoring result of the compression performance of the second compression model; The network side device obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model based on the monitoring result of the compression performance of the second compression model and the monitoring result of the prediction performance of the second prediction model.

32. The method according to any one of claims 23 to 25, wherein: The method further comprises: The network side device receives the monitoring results of the prediction performance of the second prediction model reported by the UE, and obtains the overall monitoring results of the prediction performance and compression performance of the second prediction compression model; the network side device obtains the monitoring results of the compression performance of the second compression model based on the monitoring results of the prediction performance of the second prediction model and the overall monitoring results of the prediction performance and compression performance of the second prediction compression model.

33. The method according to any one of claims 23 to 25, wherein: The method further comprises: The network-side device obtains a monitoring result of the compression performance of the second compression model, and obtains an overall monitoring result of the prediction performance and compression performance of the second prediction compression model; The network side device obtains the monitoring result of the prediction performance of the second prediction model based on the monitoring result of the compression performance of the second compression model and the overall monitoring result of the prediction performance and compression performance of the second prediction compression model.

34. A CSI processing device, wherein: Applied to user equipment UE, the apparatus includes: A first receiving module is configured to receive first reporting configuration information sent by a network-side device, where the first reporting configuration information includes at least one of the following: compression configuration information of channel state information (CSI) and prediction configuration information of the CSI; A first processing module, configured to process the first CSI information of the UE using an artificial intelligence (AI) model according to the first reporting configuration information to obtain second CSI information; The second CSI information includes at least third CSI information, and the third CSI information is: CSI information obtained by predicting the CSI information of the UE at a future time point and compressing the predicted CSI information.

35. The apparatus of claim 34, wherein: The device further comprises: a first monitoring module, configured to monitor the performance of the AI ​​model based on at least two items of the first to sixth CSI information; The fourth CSI information is: CSI information obtained by decompressing the second CSI information, the fifth CSI information is: predicted CSI information of the UE at a future time point, and the sixth CSI information is CSI information obtained by compressing the first CSI information.

36. The apparatus of claim 35, wherein: The second CSI information includes the third CSI information; The first processing module includes: a first processing submodule, configured to obtain, according to the first reporting configuration information, first CSI information of the UE at a first time point, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource; The second processing submodule is used to predict the CSI information of the UE at the second time point based on the first CSI information of the UE at the first time point using a first prediction compression model and compress the predicted CSI information using the first prediction compression model to obtain the third CSI information, where the second time point is a time point after the first time point.

37. The apparatus according to claim 36, wherein The first monitoring module includes: a first monitoring submodule, configured to obtain first CSI information of the UE at a third time point, where the third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining a monitoring result; The second monitoring submodule is used to obtain an overall monitoring result of the prediction performance and compression performance of the first prediction compression model based on the first CSI information and the fourth CSI information of the UE at a third time point.

38. The device according to any one of claims 35 to 37, wherein: The device further comprises at least one of the following: a second receiving module, configured to receive the fourth CSI information sent by the network-side device, where the fourth CSI information is obtained by the network-side device decompressing the second CSI information using the first decompression model; The second processing module is used to decompress the second CIS information by using the first reference model of the first decompression model to obtain the fourth CSI information.

39. The apparatus according to claim 35, wherein The second CSI information also includes fifth CSI information; and the first processing module includes: a third processing submodule, configured to obtain, according to the first reporting configuration information, first CSI information of the UE at a first time point, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource; The fourth processing submodule is used to predict the CSI information of the UE at a second time point using a first prediction model based on the first CSI information of the UE at the first time point, to obtain the fifth CSI information in the second CSI information, where the second time point is a time point after the first time point.

40. The apparatus of claim 39, wherein The first monitoring module includes: a third monitoring submodule, configured to obtain first CSI information of the UE at a third time point, where the third time point is at least one of the following: the second time point, a time point associated with the second time point, and a time point indicated by the network side device for obtaining a monitoring result; The fourth monitoring submodule is used to obtain a monitoring result of the prediction performance of the first prediction model based on the first CSI information of the UE at the third time point and the fifth CSI information in the second CSI information.

41. The apparatus of claim 39, wherein: The first monitoring module includes: The fifth monitoring submodule is used to obtain a monitoring result of the compression performance of the first compression model based on the fourth CSI information and the fifth CSI information in the second CSI information.

42. The apparatus of claim 35, wherein: The second CSI information also includes sixth CSI information; The first processing module includes: a fifth processing submodule, configured to obtain, according to the first reporting configuration information, first CSI information of the UE at a first time point, where the first time point is at least one of the following: a current time point, a historical time point, or a time point of a CSI reference resource; The sixth processing submodule is configured to compress the first CSI information of the UE at the first time point by using the first compression model to obtain sixth CSI information in the second CSI information.

43. The apparatus according to claim 42, wherein The first monitoring module includes: The sixth monitoring submodule is used to obtain a monitoring result of the compression performance of the first compression model based on the fourth CSI information and the sixth CSI information in the second CSI information.

44. A CSI processing device, wherein: Applied to network-side equipment, the device includes: A configuration information sending module, configured to send first reporting configuration information to a user equipment UE, where the first reporting configuration information includes at least one of the following: compression configuration information of channel state information CSI, and prediction configuration information of the CSI; The first reporting configuration information is used for the UE to process the first CSI information of the UE using an artificial intelligence AI model to obtain second CSI information; The second CSI information includes at least third CSI information, where the third CSI information is: CSI information obtained by predicting CSI information of the UE at a future time point and compressing the predicted CSI information.

45. The apparatus of claim 44, wherein: The device further comprises: A CSI information receiving module, configured to receive at least one of the first CSI information, the second CSI information, the third CSI information, and the fifth CSI information reported by the UE; The fifth CSI information is: CSI information of the UE predicted by the UE at a future time point.

46. ​​The apparatus according to claim 44 or 45, wherein The device further comprises at least one of the following: a decompression module, configured to decompress the second CSI information using a first decompression model to obtain the fourth CSI information; The CSI information sending module is configured to send the fourth CSI information to the UE.

47. The device according to any one of claims 44 to 46, wherein: The device further comprises: a performance monitoring module, configured to monitor the performance of the AI ​​model based on at least two items of the first to sixth CSI information; Among them, the fourth CSI information is: the CSI information obtained by the network side device or the UE decompressing the second CSI information, the fifth CSI information is: the CSI information of the UE predicted by the UE at a future time point, and the sixth CSI information is the CSI information obtained by the UE compressing the first CSI information.

48. A user equipment, wherein The method comprises a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the CSI processing method according to any one of claims 1 to 22 are implemented.

49. A network side device, wherein: It includes a processor and a memory, the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the CSI processing method according to any one of claims 23 to 33 are implemented.

50. A readable storage medium, wherein: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the program or instruction implements the steps of the CSI processing method according to any one of claims 1 to 22, or implements the steps of the CSI processing method according to any one of claims 23 to 33.

Images