CSI reporting and receiving methods, and apparatus, device, readable storage medium and computer program product

Abstract

The present application belongs to the technical field of communications, and specifically relates to CSI reporting and receiving methods, and an apparatus, a device, a readable storage medium and a computer program product. The CSI reporting method comprises: a terminal receiving CSI reporting indication information from a network-side device; the terminal acquiring a target Rank; the terminal determining CSI reporting information on the basis of the target Rank and the CSI reporting indication information; and sending the CSI reporting information to the network-side device, wherein the CSI reporting information comprises CSI feedback information, the CSI feedback information being acquired by means of a target AI model, and the target AI model comprising one or more AI model units, and the CSI reporting indication information is used for indicating at least one of a payload of the CSI feedback information, a dataset associated with the AI model units, and the AI model units.

Description

CSI reporting, receiving method, device, equipment, readable storage medium and computer program product

[0001] The present application claims priority to the Chinese patent application No. 202411073425.5, filed on August 6, 2024, and entitled "CSI reporting, receiving method, device, equipment, readable storage medium and computer program product", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application belongs to the field of communication technology, and specifically relates to a CSI reporting, receiving method, device, equipment, readable storage medium and computer program product. BACKGROUND

[0003] For channel state information (CSI) reporting, the rank indication (RI) corresponding to each reporting may be different, resulting in different payloads. In the existing method, the corresponding payload can be determined under each different codebook under each rank, so that the CSI reporting information can be correctly parsed.

[0004] However, for artificial intelligence (AI) based CSI reporting information, in addition to the rank information of the terminal, it also depends on the model information used by the terminal, and there is currently no clear method for how to perform AI based CSI reporting. SUMMARY

[0005] The embodiments of the present application provide a CSI reporting, receiving method, device, equipment, readable storage medium and computer program product, which can solve the problem of how to perform AI based CSI reporting.

[0006] In a first aspect, a CSI reporting method is provided, comprising:

[0007] The terminal receives CSI reporting indication information from a network side device;

[0008] The terminal acquires a target rank;

[0009] The terminal determines CSI reporting information according to the target rank and the CSI reporting indication information;

[0010] The terminal sends the CSI reporting information to the network side device;

[0011] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used for indicating at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0012] In a second aspect, a CSI receiving method is provided, including:

[0013] A network side device sends CSI reporting indication information to a terminal;

[0014] The network side device receives CSI reporting information from the terminal;

[0015] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used for indicating at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0016] In a third aspect, a CSI reporting device is provided, including:

[0017] A first receiving module is configured to receive, by a terminal, CSI reporting indication information from a network side device;

[0018] A first processing module is configured to obtain, by the terminal, a target Rank;

[0019] A second processing module is configured to determine, by the terminal, CSI reporting information according to the target Rank and the CSI reporting indication information;

[0020] A first sending module is configured to send, by the terminal, the CSI reporting information to the network side device;

[0021] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used for indicating at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0022] In a fourth aspect, a CSI receiving device is provided, including:

[0023] A third sending module is configured to send, by a network side device, CSI reporting indication information to a terminal;

[0024] a third receiving module, configured to receive, by the network-side device, CSI reporting information from the terminal;

[0025] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used to indicate at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0026] In a fifth aspect, a CSI reporting device is provided, which is configured to perform the steps of the method according to the first aspect. In a sixth aspect, a CSI receiving device is provided, which is configured to perform the steps of the method according to the second aspect.

[0027] In a sixth aspect, a terminal is provided, which includes a processor and a memory, the memory stores programs or instructions executable on the processor, and the programs or instructions, when executed by the processor, implement the steps of the method according to the first aspect.

[0028] In a seventh aspect, a terminal is provided, which includes a processor and a communication interface, and the processor is configured to:

[0029] receiving, by the terminal, CSI reporting indication information from a network-side device;

[0030] obtaining, by the terminal, a target Rank;

[0031] determining, by the terminal, CSI reporting information according to the target Rank and the CSI reporting indication information;

[0032] sending, by the terminal, the CSI reporting information to the network-side device;

[0033] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used to indicate at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0034] In an eighth aspect, a network-side device is provided, which includes a processor and a memory, the memory stores programs or instructions executable on the processor, and the programs or instructions, when executed by the processor, implement the steps of the method according to the first aspect.

[0035] In a ninth aspect, a network-side device is provided, which includes a processor and a communication interface, and the processor is configured to:

[0036] Network-side devices send CSI reporting indication information to the terminal;

[0037] The network-side device receives CSI reporting information from the terminal;

[0038] The CSI reporting information includes CSI feedback information, which is obtained through a target AI model. The target AI model includes one or more AI model units. The CSI reporting indication information is used to indicate the payload of the CSI feedback information. The dataset associated with the AI ​​model unit and at least one of the AI ​​model units are also included.

[0039] In a tenth aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or implement the steps of the method described in the second aspect.

[0040] Eleventhly, a wireless communication system is provided, comprising: a terminal and a network-side device, wherein the terminal can be used to perform the steps of the method as described in the first aspect, and the network-side device can be used to perform the steps of the method as described in the second aspect.

[0041] In a twelfth aspect, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being configured to run programs or instructions to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0042] In a thirteenth aspect, a computer program / program product is provided, which is stored in a storage medium and is executed by at least one processor to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0043] In this embodiment, the terminal receives a payload indicating CSI feedback information, a dataset associated with an AI model unit, and CSI reporting indication information for at least one of the AI ​​model units from a network-side device. It then obtains a target Rank, determines CSI reporting information based on the target Rank and the CSI reporting indication information, and sends the CSI reporting information to the network-side device. The CSI feedback information is obtained through a target AI model that includes one or more AI model units. In this way, the terminal can determine the target AI model to use based on the CSI reporting indication information sent by the network side, and determine the CSI reporting information based on the obtained target Rank and target AI model, enabling the terminal to use the correct AI model and the network side to correctly parse the CSI reporting information. Attached Figure Description

[0044] Fig. 1a is a block diagram of a wireless communication system to which embodiments of the present application can be applied;

[0045] Fig. 1b is a schematic diagram of a conventional neural network;

[0046] Fig. 1c is a schematic diagram of a conventional neuron;

[0047] Fig. 1d is a schematic diagram of an application scenario of embodiments of the present application;

[0048] Fig. 1e is a schematic diagram of one of architectures of a target AI model according to embodiments of the present application;

[0049] Fig. 1f is a schematic diagram of another of architectures of a target AI model according to embodiments of the present application;

[0050] Fig. 2 is a flow diagram of a CSI reporting method according to embodiments of the present application;

[0051] Fig. 3 is a flow diagram of a CSI receiving method according to embodiments of the present application;

[0052] Fig. 4 is a schematic diagram of a CSI reporting apparatus according to embodiments of the present application;

[0053] Fig. 5 is a schematic diagram of a CSI receiving apparatus according to embodiments of the present application;

[0054] Fig. 6 is a schematic diagram of a communication device according to embodiments of the present application;

[0055] Fig. 7 is a schematic diagram of a terminal according to embodiments of the present application;

[0056] Fig. 8 is a schematic diagram of a network-side device according to embodiments of the present application;

[0057] Fig. 9 is a schematic diagram of a network-side device according to embodiments of the present application. DETAILED DESCRIPTION

[0058] The technical solutions in embodiments of the present application will be described clearly below with reference to the accompanying drawings in embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0059] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, the scope of protection for "A or B" covers at least three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. In addition, 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 scenarios. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0060] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc., in the instruction sent. An indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0061] It is worth noting that the technology described in the embodiments of the present application is not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, 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 described technology can be used in the above-mentioned systems and radio technologies, as well as in other systems and radio technologies. The following description describes a New Radio (NR) system for example 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) communication systems. th

[0062] ​FIG. 1a shows a block diagram of a wireless communication system to which embodiments of the present application can be applied. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can be a terminal-side device such as a mobile phone, a Tablet Personal Computer, a Laptop Computer, a notebook computer, a Personal Digital Assistant (PDA), a palm computer, a netbook, an Ultra-mobile Personal Computer (UMPC), a Mobile Internet Device (MID), an Augmented Reality (AR) device, a Virtual Reality (VR) device, a robot, a wearable device, a flight vehicle, a Vehicle User Equipment (VUE), a shipboard device, a Pedestrian User Equipment (PUE), a smart home (a home device with a wireless communication function such as a refrigerator, a television, a washing machine, or furniture, etc.), a game console, a Personal Computer (PC), a kiosk, or a self-service machine, etc. The wearable device includes a smart watch, a smart bracelet, a smart earphone, smart glasses, smart jewelry (a smart bracelet, a smart necklace, a smart ring, a smart necklace, a smart anklet, a smart necklace, etc.), a smart wristband, smart clothes, etc. The vehicle-mounted device can also be referred to as 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 embodiments of the present application. The network-side device 12 can include an access network device or a core network device, wherein the access network device can also be referred to as a Radio Access Network (RAN) device, a radio access network function, or a radio access network unit. The access network device can include a base station, a Wireless Local Area Network (WLAN) Access Point (AP), or a Wireless Fidelity (WiFi) node, etc.The base station can be referred to as a Node B (NB), an evolved Node B (eNB), a next generation Node B (gNB), a New Radio Node B (NR Node B), an access point, a relay 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), or some other suitable terminology in the art, and is not limited to a particular technical terminology, provided that the same technical effect is achieved. It should be noted that in the embodiments of the present application, only the base station in the NR system is taken as an example for introduction, and the specific type of the base station is not limited.

[0063] The core network device can also be referred to as a core network node, a core network function, or a core network network element, etc., which includes but is not limited to at least one of the following: a mobility management entity (MME), an access and mobility management function (AMF), a session management function (SMF), a user plane function (UPF), a policy control function (PCF), a policy and charging rules function (PCRF), an edge application server discovery function (EASDF), a unified data management (UDM), a unified data repository (UDR), a home subscriber server (HSS), a centralized network configuration (CNC), a network repository function (NRF), a network exposure function (NEF), a local NEF (or L-NEF), a binding support function (BSF), an application function (AF), a location management function (LMF), a gateway mobile location center (GMLC), a network data analytics function (NWDAF), etc. It should be noted that only the core network device in the NR system is taken as an example for introduction in the embodiments of the present application, and the specific type of the core network device is not limited. If the name of the core network device mentioned in the embodiments of the present application changes in the subsequent protocol version (for example, 6G), it is also within the protection scope of the present application.

[0064] Optionally, the core network device can be implemented by one or more function modules in one device, or can be implemented by multiple devices together, and the embodiments of the present application do not make a specific limitation hereon. It can be understood that the above function modules can be network elements in a hardware device, can be software function modules running on a special hardware, or can be virtualized function modules instantiated on a platform (for example, a cloud platform).

[0065] In order to better understand the technical solutions of the present application, the following is introduced first:

[0066] Artificial intelligence

[0067] Artificial intelligence has been widely applied in various fields at present. There are various implementation manners of the AI module, for example, neural network, decision tree, support vector machine, Bayesian classifier, etc. The present application takes the neural network as an example for illustration, but is not limited to the specific type of the AI module.

[0068] A schematic diagram of a neural network is shown in FIG. 1b.

[0069] Among them, the neural network is composed of neurons, and a schematic diagram of a neuron is shown in FIG. 1c. Among them, a1, a2, … aK are inputs, w is a weight (multiplicative coefficient), b is a bias (additive coefficient), and σ(.) is an activation function. Common activation functions include Sigmoid, tanh, Rectified Linear Unit (ReLU), etc.

[0070] The parameters of the neural network are optimized by an optimization algorithm. The optimization algorithm is a kind of algorithm that can help us minimize or maximize the objective function (sometimes also called the loss function). And the objective function is often a mathematical combination of model parameters and data. For example, given data X and its corresponding label Y, we construct a neural network model f(.), and after having the model, we can get the predicted output f(x) according to the input x, and can calculate the gap between the predicted value and the true value (f(x)-Y), which is the loss function. Our purpose is to find appropriate W, b to make the value of the above loss function reach the minimum, and the smaller the loss value is, the closer our model is to the true situation.

[0071] The common optimization algorithm is basically based on error back propagation (BP) algorithm. The basic idea of BP algorithm is that the learning process consists of two processes of forward propagation of signals and backward propagation of errors. When forward propagation, the input sample is transmitted from the input layer, processed by each hidden layer layer by layer, and transmitted to the output layer. If the actual output of the output layer does not match the expected output, the backward propagation of error is entered. Error back propagation is to transmit the output error to the input layer through the hidden layer in a certain form, and allocate the error to all units of each layer, so as to obtain the error signal of each layer unit, which is used as the basis for correcting the weight of each unit. The weight adjustment process of each layer of the signal forward propagation and the error backward propagation is repeated. The process of continuously adjusting the weight is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until the preset learning times are reached.

[0072] The common optimization algorithm includes gradient descent, stochastic gradient descent (SGD), mini-batch gradient descent, momentum method, Nesterov (the name of the inventor, specifically stochastic gradient descent with momentum), adaptive gradient descent (Adagrad), Adadelta, root mean square prop (RMSprop), adaptive moment estimation (Adam), etc.

[0073] These optimization algorithms, when error back propagation, are based on the error / loss obtained from the loss function, to derive the current neuron / partial derivative, plus the learning rate, the previous gradient / derivative / partial derivative, etc. Influence, get gradient, pass the gradient to the previous layer.

[0074] Life cycle of AI model

[0075] The life cycle management (LCM) of AI / machine learning (ML) model (for example: model training, model deployment, model inference, model monitoring, model updating) and its functional characteristics.

[0076] The life cycle of the model includes at least one or more of the following: model training, model deployment, model inference, model monitoring, and model updating.

[0077] Rank: the number of spatial multiplexing streams. Simply put, the same time-frequency resource is divided into several parts for simultaneous transmission in space. In the case of constant time-frequency resources, the higher the RANK, the higher the actual throughput rate.

[0078] The RI reported by the UE is based on channel state information reference signal (CSI-RS) measurement. The relationship between the number of CSI-RS ports and the RI reported by the UE is: the upper limit value of RI ≤ the number of CSI-RS ports. That is, if the number of CSI-RS ports is configured as 2, the maximum RI reported by the UE does not exceed 2.

[0079] The CSI reporting method and the CSI receiving method provided by the embodiments of the present application will be described in detail below in combination with the accompanying drawings, some embodiments and their application scenarios.

[0080] Referring to FIG. 1d, it is a schematic diagram of a layer specific, layer common, rank specific, or rank specific and layer specific model to which the embodiments of the present application can be applied. The layer specific means that the model is applicable to a specific layer, the rank specific means that the model is applicable to a specific rank, and the layer common means that the model is applicable to any layer or a layer within a specified rank range.

[0081] Referring to FIG. 1e, it is an architecture of a target AI model to which the embodiments of the present application can be applied. For example, the terminal side includes four models, each of which corresponds to a different payload. Alternatively, each model corresponds to a different layer.

[0082] Referring to FIG. 1f, it is an architecture of a target AI model to which the embodiments of the present application can be applied. For example, the terminal side includes four models, each of which corresponds to a different rank. The terminal can determine the target model based on the target rank.

[0083] Referring to FIG. 2, the present application provides a CSI reporting method, and the execution subject of the method is a terminal, which includes:

[0084] Step 201: The terminal receives CSI reporting indication information from the network side device;

[0085] Step 202: The terminal obtains a target Rank;

[0086] Step 203: The terminal determines CSI reporting information according to the target Rank and the CSI reporting indication information;

[0087] Step 204: The terminal sends the CSI reporting information to the network side device;

[0088] The CSI reporting information includes CSI feedback information, which refers to specific CSI related information fed back by the terminal side to the network side, for example, can include channel information or codebook information, etc. The specific content of the CSI feedback information is not limited in the embodiments of the present application; the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, and the CSI reporting indication information is used to indicate at least one of the payload of the CSI feedback information, the dataset associated with the AI model unit, and the AI model unit.

[0089] It should be noted that for the terminal side, a plurality of AI model units can be pre-set, each AI model unit is pre-trained to cope with different Ranks, layers or payloads;

[0090] The above CSI reporting indication information is used to indicate the payload of the CSI feedback information, which refers to the payload requirement of the CSI reporting indicated by the network side. In this way, the terminal side can select the corresponding AI model unit according to the payload requirement, and determine the target AI model according to the AI model unit, to obtain the CSI feedback information. Optionally, the above CSI reporting indication information is used to indicate the payload of the CSI feedback information corresponding to the candidate Rank, which refers to the payload of the CSI feedback information corresponding to each Rank indicated by the network side respectively. It can be understood that the payload of the CSI feedback information when the Rank = 1 is different from the payload of the CSI feedback information when the Rank = 2. Or, the payload of the CSI feedback information when the Rank is less than or equal to X is different from the payload of the CSI feedback information when the Rank is greater than X

[0091] Optionally, the CSI reporting indication information is used to indicate the payload of the CSI feedback information corresponding to different layers, which means that the network side respectively indicates the payload of the CSI feedback information corresponding to each layer. It can be understood that the payload of the CSI feedback information corresponding to the layer 0 is different from the payload of the CSI feedback information corresponding to the layer 1. Alternatively, the payload of the CSI feedback information corresponding to the first X layers is different from the payload of the CSI feedback information corresponding to the last Y layers.

[0092] The CSI reporting indication information is used to indicate the payload of the CSI feedback information, which means that the network side indicates the payload requirement related to the CSI reporting. In this way, the terminal side can select the corresponding AI model unit according to the payload requirement, and determine the target AI model according to the AI model unit, so as to obtain the CSI feedback information. Optionally, the payload related to the CSI reporting may include the payload of the Rank information, the CQI information, and the CSI feedback information. By discarding the payload of the information obtained by the AI, the payload of the CSI feedback information can be further determined.

[0093] The CSI reporting indication information is used to indicate the payload of the CSI feedback information, which means that the network side indicates the payload requirement related to the CSI reporting. The network side also needs to indicate the CSI reporting of the terminal, such as cri-RI-PMI-CQI, cri-RI-LI-PMI-CQI, AI-RI-PMI-CQI, or indicate that the CQI is wideband information or subband information.

[0094] The CSI reporting indication information is used to indicate the dataset associated with the AI model unit, which means that the network side indicates the specified dataset, which is used to obtain or determine the AI model unit. For example, the AI model unit can be trained according to the dataset. In this way, the terminal side can find the corresponding AI model unit in the plurality of AI model units pre-set according to the dataset, or the terminal can directly train the AI model unit according to the dataset, and determine the target AI model according to the AI model unit, so as to obtain the CSI feedback information. It can be understood that the information corresponding to the dataset can implicitly or explicitly include the payload or format of the CSI feedback information.

[0095] The CSI reporting indication information is used to indicate the AI model unit, which means that the network side directly indicates the AI model unit, the terminal side selects the corresponding AI model unit according to the network side indication, and determines the target AI model according to the AI model unit, so as to obtain the CSI feedback information, thereby determining the size of the CSI reporting information.

[0096] The target Rank refers to the Rank obtained by the terminal when performing the current CSI reporting. The target Rank can be obtained by CSI-RS measurement, and the application embodiment does not make specific limitation to the obtaining method of the target Rank. The candidate Rank refers to the Rank candidate range that can be associated when performing optional CSI reporting. Optionally, the target Rank belongs to the candidate Rank, and the terminal determines the information corresponding to the target Rank, such as the model or the CSI reporting information corresponding to the target Rank, from the information associated with the plurality of candidate Ranks based on the target Rank.

[0097] In the embodiment of the application, the terminal receives, from the network side device, the payload used to indicate the CSI feedback information, the CSI reporting indication information of at least one of the dataset associated with the AI model unit and the AI model unit, and obtains the target Rank. According to the target Rank and the CSI reporting indication information, the CSI reporting information is determined and sent to the network side device, wherein the CSI feedback information is obtained by the target AI model corresponding to the target Rank, and the target AI model includes a plurality of different AI model units. In this way, the terminal side can obtain the target AI model including a plurality of AI model units corresponding to the target Rank by splicing a plurality of AI model units. The terminal side can use the correct AI model, and the network side can correctly parse the CSI reporting information.

[0098] In an optional embodiment, the CSI reporting indication information includes at least one of the following:

[0099] (1) The first payload information is used to indicate the payload of the CSI feedback information corresponding to the Rank information.

[0100] For example, it can be the payload associated with the CSI feedback information of Rank=M. The indication information can be applied to the scenario shown in FIG. 1f. In an embodiment, the CSI reporting indication information includes the payload corresponding to different Ranks, for example: Rank=1, payload 64; Rank=2, payload 100, …, so as to determine the corresponding model or reporting payload of the terminal when Rank=1, 2, 3 or 4, etc.

[0101] For example, the payload associated with CSI feedback information of Rank<=M, the payload associated with CSI feedback information of Rank>M, and / or the payload associated with CSI feedback information of MY>=Rank>M. In one embodiment, the CSI reporting indication information includes the payload associated with CSI feedback information of Rank<=M and the payload associated with CSI feedback information of Rank>M. In one embodiment, when Rank<=M is associated with one model, when Rank is equal to 1, 2 or M, it corresponds to one model, and when Rank>M, it is associated with another model.

[0102] (2) Second payload information for indicating the payload of CSI feedback information corresponding to layer information.

[0103] The indication information can be applied to the scenario shown in FIG. 1e. The second Payload information is used to indicate the payload information associated with the layer ID, for example, to indicate the payload associated with at least one layer of the M layers, so as to find the corresponding AI model unit through the payload of each layer;

[0104] In one embodiment, the second Payload information is used to indicate the payload information associated with a plurality of layer IDs, for example, to indicate that the payload information associated with a plurality of layers is the same;

[0105] In one embodiment, the second Payload information is used to indicate the payload information associated with the first M layers, for example, to indicate that the payload information associated with the first M layers is the same;

[0106] For example: indicating that the layer 0 payload=Z bit, and the total bit when Rank=M is Z1 bit, wherein Z1 is greater than or equal to Z.

[0107] For another example, the CSI reporting indication information indicates the payload of each layer.

[0108] For another example, the CSI reporting indication information indicates payload0 and payload1, that is, when Rank=1, the payload corresponding to the CSI reporting information is payload0, and when Rank=2, the payload corresponding to the CSI reporting information is payload0 and payload1.

[0109] For example, the CSI reporting indication information indicates payload0, payload1, and Rank threshold 1. When Rank<=Rank threshold 1, the CSI reporting information corresponds to payload0, and when Rank>Rank threshold 1, the CSI reporting information corresponds to payload1.

[0110] (3) A first dataset ID (dataset ID) for indicating a dataset corresponding to Rank information;

[0111] For example, it can be a dataset ID corresponding to Rank=M. Based on the dataset ID, an AI model unit (abbreviated as model) based on the dataset can be determined, or a payload of CSI feedback information can be determined. The indication information can be applicable to the scenario shown in FIG. 1f. In an embodiment, the CSI reporting indication information includes dataset IDs corresponding to different Ranks, so that AI model units corresponding to the terminal when Rank=1, 2, 3, or 4, or the reporting payload can be determined (it can be understood that determining the AI model unit is equivalent to indirectly determining the payload, because the payload of the information that can be output by the AI model unit is known).

[0112] In an embodiment, the CSI reporting indication information includes a dataset ID 0 corresponding to Rank<=M, and / or a dataset ID 1 corresponding to Rank>M, so as to determine an AI model unit (abbreviated as model) based on the dataset, or to determine a payload of CSI feedback information, or to determine CSI reporting information.

[0113] (4) A second dataset ID for indicating a dataset corresponding to layer information;

[0114] For example, the CSI reporting indication information indicates M1 dataset IDs, and M layers are associated with models or payloads determined by the dataset IDs, respectively. M1 can be less than M, that is, one or more layers can be associated with a model or a payload determined by one dataset ID.

[0115] For example, the CSI reporting indication information indicates M1 dataset IDs, and the model or payload associated with each of the M layers is determined by the dataset ID. M1 can be equal to M, that is, one dataset ID can determine the model or payload associated with one layer.

[0116] For another example, the CSI reporting indication information indicates dataset 0+dataset 1, that is, when Rank=1, the model corresponding to dataset 0 is used, when Rank=2, the model corresponding to dataset 0 is used to process layer 0, and the model corresponding to dataset 1 is used to process layer 1.

[0117] For another example, the CSI reporting indication information indicates dataset 0+dataset 1+dataset 2+dataset 3, that is, when Rank=1, the model corresponding to dataset 0 is used, when Rank=2, the model corresponding to dataset 0 is used to process layer 0, and the model corresponding to dataset 1 is used to process layer 1, when Rank=4, the model corresponding to dataset 0 is used to process layer 0, the model corresponding to dataset 1 is used to process layer 1, the model corresponding to dataset 2 is used to process layer 2, and the model corresponding to dataset 3 is used to process layer 3.

[0118] For another example, the CSI reporting indication information indicates M1 dataset IDs and payload, and the dataset ID and the corresponding model are determined based on the payload.

[0119] For another example, layer 0 is associated with dataset id X0, layer 1 is associated with dataset X1, and layer 2 is associated with dataset X2.

[0120] For another example, layer 0 and 1 are associated with dataset id X0, and layer 1 / 2 is associated with dataset X1.

[0121] (5) First model identifier (model ID), used to indicate the AI model unit corresponding to the Rank information.

[0122] For example, it can be a model ID corresponding to Rank=M, based on which the AI model unit can be determined. In an embodiment, the CSI reporting indication information includes model IDs corresponding to different Ranks, for example: Rank=1, model ID x0; Rank=2, model ID x1, …, so as to determine the model or CSI reporting information corresponding to the terminal when Rank=1, 2, 3, or 4.

[0123] For example, it can be a model ID associated with CSI feedback information of Rank<=M, a model ID associated with CSI feedback information of Rank>M, and / or a model ID associated with CSI feedback information of MY>=Rank>M. In an embodiment, the CSI reporting indication information includes a model ID associated with CSI feedback information of Rank<=M and a model ID associated with CSI feedback information of Rank>M. In an embodiment, when Rank<=M is associated with one model, when Rank is equal to 1, 2, or M, it corresponds to one model, and when Rank>M, it is associated with another model.

[0124] (6) A second model ID for indicating an AI model unit corresponding to layer information.

[0125] For example: Indicate M1 model IDs, determine the model associated with M layers through the model IDs respectively, where M1 can be less than or equal to M, that is, one or more layers associated with a model can be determined by one model ID;

[0126] For example: Indicate model 0 and model 1, when Rank=1, use the model corresponding to model 0, when Rank=2, use the model corresponding to model 0 to process layer 0 and the model corresponding to model 1 to process layer 1;

[0127] For example: Indicate model 0, model 1, model 2, and model 3, when Rank=1, use the model corresponding to model 0, when Rank=2, use the model corresponding to model 0 to process layer 0 and the model corresponding to model 1 to process layer 1, when Rank=4, use the model corresponding to model 0 to process layer 0, the model corresponding to model 1 to process layer 1, the model corresponding to model 2 to process layer 2, and the model corresponding to model 3 to process layer 3.

[0128] For example, indicating M1 model IDs and a payload, and determining the model ID and its corresponding model based on the payload.

[0129] For example, the model ID list associated with the CSI feedback information of Rank<=M, the model ID list associated with the CSI feedback information of Rank>M, and / or the model ID list associated with the CSI feedback information of MY>=Rank>M. In an embodiment, the CSI reporting indication information includes the model ID list associated with the CSI feedback information of Rank<=M and the model ID list associated with the CSI feedback information of Rank>M.

[0130] In the embodiments of the present application, the CSI reporting indication information (1)-(6) above realizes the manner in which the network side indicates the AI model unit for the terminal side.

[0131] In an optional implementation, the layer information includes at least one of a layer ID and layer grouping information, i.e., the layer information can refer to a single layer or a layer group containing multiple layers or layer list information; in this way, the CSI reporting indication information can indicate at least one of the payload dataset and the AI model unit for a single layer, or the CSI reporting indication information can also indicate at least one of the payload dataset and the AI model unit for the layer group or the layer list.

[0132] The Rank information includes at least one of a Rank value and Rank grouping information, i.e., the Rank information can refer to a single Rank value or a Rank group containing multiple Ranks or Rank list information. In this way, the CSI reporting indication information can indicate at least one of the payload dataset and the AI model unit for a single Rank, or the CSI reporting indication information can also indicate at least one of the payload dataset and the AI model unit for the Rank group or the Rank list.

[0133] In an optional implementation, the CSI reporting indication information includes X first payload information, wherein the X first payload information is associated with the Rank information.

[0134] Or,

[0135] The CSI reporting indication information includes X first dataset IDs, wherein the X first dataset IDs are associated with Rank information.

[0136] Alternatively,

[0137] The CSI reporting indication information includes X first model IDs. The X first dataset IDs are associated with Rank information.

[0138] X is the number of Rank values or the number of Rank groups or the number of Rank lists.

[0139] In an optional implementation, the CSI reporting indication information includes Y second payload information, wherein the Y second payload information is associated with layer information; or

[0140] The CSI reporting indication information includes Y second dataset IDs, wherein the Y second dataset IDs are associated with layer information; or

[0141] The CSI reporting indication information includes Y second model IDs, wherein the Y second model IDs are associated with layer information; wherein Y is the number of layers or the number of layer groups or the number of layer lists.

[0142] In an optional implementation, the CSI feedback information includes N pieces of CSI feedback information corresponding to layers, and the target AI model includes K AI model units, and K is less than or equal to N.

[0143] In the embodiments of the present application, for the case where K is equal to N, it means that for the N pieces of CSI feedback information corresponding to layers, the CSI feedback information of each layer is obtained by different AI model units, that is, the AI model unit can be a layer specific AI model unit; for the case where K is less than N, it means that for the N pieces of CSI feedback information corresponding to layers, the CSI feedback information of multiple layers can be obtained by the same AI model unit, that is, the AI model unit can be a layer common AI model unit.

[0144] The K AI model units are respectively used for processing CSI information corresponding to different layers, and the payloads of the CSI feedback information corresponding to different layers are different.

[0145] The CSI information is channel information or codebook information obtained based on a measurement reference signal (such as a CSI-RS, a non-zero CSI-RS, or a CRS).

[0146] In an optional implementation, the CSI reporting indication information includes at least one of second payload information, a second dataset ID, and a second model ID.

[0147] In an optional implementation, the AI model unit includes any one of the following:

[0148] (1) a first AI model unit used by the terminal and the network-side device;

[0149] That is, the terminal side and the network side need to be adapted to ensure that the network side can recover the correct CSI information.

[0150] (2) a reference AI model unit of a second AI model unit used by the terminal and the network-side device;

[0151] Optionally, the terminal side and the network side can further optimize the model based on the reference AI model unit to obtain the finally used AI model unit.

[0152] (3) a third AI model unit used by the terminal, the network-side device, and a test device in testing;

[0153] The test device refers to a device used in AI testing with the terminal and the network-side device in advance.

[0154] (4) a reference AI model unit of a fourth AI model unit used by the terminal, the network-side device, and the test device in testing.

[0155] In an optional implementation, the K AI model units satisfy any one of the following:

[0156] (1) The K AI model units are K layer corresponding AI model units, and the payload of the CSI reporting indication information is used to select the AI model unit of each layer.

[0157] Optionally, the K layer AI model units can be layer specific models or layer common models, but correspond to different payloads

[0158] (2) The K AI model units are target Rank corresponding AI model units.

[0159] (3) The K AI model units are one layer group corresponding AI model units.

[0160] (4) The K AI model units are AI model units corresponding to a layer list. Optionally, the layer list can be pre-configured or determined based on a Rank threshold.

[0161] Optionally, the K layer AI model units can be layer-common models, which are selected and combined based on payload. For example, Enc 0 is 120 bits, Enc 1 is 80 bits, and Enc 2W is 64 bits. Based on the CSI reporting indication information, the payload selects the AI model units of each layer.

[0162] In an optional implementation, the CSI reporting information further includes at least one of the following:

[0163] (1) Target Rank;

[0164] That is, the terminal provides the Rank corresponding to the current CSI reporting to the network side device.

[0165] (2) Channel quality information (Channel Quality Indicator, CQI);

[0166] That is, the terminal provides the CQI to the network side device.

[0167] (3) Payload corresponding to at least one layer;

[0168] That is, the terminal provides the payload corresponding to at least one layer to the network side device.

[0169] (4) Model ID corresponding to the target AI model;

[0170] That is, the terminal provides the model ID corresponding to the AI model used by itself to the network side device.

[0171] (5) Dataset ID corresponding to the target AI model.

[0172] That is, the terminal provides the dataset ID corresponding to the AI model used by itself to the network side device.

[0173] It can be understood that the terminal receives the CSI reporting indication information, and then the terminal sends the CSI reporting information. Therefore, the terminal and the network side device determine the payload or content of the CSI reporting information in combination with the CSI reporting information and the CSI reporting indication information.

[0174] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0175] (1) The terminal determines the target AI model corresponding to the target Rank according to the target Rank and the first payload information;

[0176] (2) The terminal determines the CSI reporting information according to the target AI model;

[0177] Wherein, the CSI reporting information is obtained through the target AI model, and the target AI model includes an AI model unit, and the AI model unit is applicable to the target Rank.

[0178] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0179] (1) The terminal determines the payload of the CSI feedback information corresponding to N layers corresponding to the target Rank according to the target Rank and the first payload information;

[0180] (2) The terminal determines the CSI reporting information according to the payload of the CSI feedback information corresponding to N layers;

[0181] Wherein, the CSI reporting information is obtained through the target AI model, and the target AI model includes K AI model units corresponding to the payload of the CSI feedback information corresponding to N layers.

[0182] In the embodiments of the present application, for the case that the CSI reporting indication information is the first payload information, the terminal determines the payload of the CSI feedback information corresponding to N layers corresponding to the target Rank, and further determines K AI model units, to determine the target AI model, and obtains the CSI reporting information based on the target AI model.

[0183] Wherein, the payload of the CSI feedback information corresponding to N layers satisfies one or more of the following:

[0184] (1) The sum of the payload of the CSI feedback information corresponding to N layers is less than or equal to the payload indicated by the first payload information;

[0185] That is, the sum of the payload of the CSI feedback information corresponding to N layers corresponding to the target Rank cannot exceed the payload indicated by the first payload information;

[0186] (2) In the N layers, the payload of the CSI feedback information corresponding to the first layer ID is greater than or equal to the payload of the CSI feedback information corresponding to the second layer ID, wherein the first layer ID is less than the second layer ID;

[0187] That is, the payload of the low layer needs to be greater than or equal to the payload of the high layer, so that the low layer can carry more information, the information can be concentrated upwards, the information energy can be enhanced, and the basic information transmission can be ensured.

[0188] (3) In the N layers, the difference between the payload of the CSI feedback information corresponding to the first layer ID and the payload of the CSI feedback information corresponding to the second layer ID is less than or equal to a first threshold, wherein the first layer ID is less than the second layer ID;

[0189] That is, the difference between the payload of the low layer and the payload of the high layer cannot be greater than the first threshold. Optionally, specifically, the difference between the payloads of the adjacent two layers cannot be greater than the first threshold, or the difference between the lowest layer and the highest layer cannot be greater than a second threshold.

[0190] The combination of the CSI feedback information satisfies at least one of the following conditions through the condition of one or more of the above (1) to (3),

[0191] a) The payload corresponding to the CSI feedback information is closest to the payload indicated by the first payload information;

[0192] b) The payload corresponding to layer 0 in the N layers is the largest.

[0193] In an embodiment, wherein the payload of the CSI feedback information corresponding to the N layers satisfies one or more of the following:

[0194] a) The payload corresponding to the CSI feedback information is closest to the payload indicated by the first payload information;

[0195] b) The payload corresponding to layer 0 in the N layers is the largest.

[0196] Optionally, there are at least two layers using one model (i.e., using the same AI model unit) (for example, the target AI model obtained by splicing multiple model units has at least two layers using the same model unit, and the two layers are adjacent to each other).

[0197] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0198] The terminal determines the CSI feedback information corresponding to the N layers corresponding to the target Rank according to the second payload information.

[0199] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0200] (1) The terminal determines the payload of the CSI feedback information corresponding to the N layers corresponding to the target Rank according to the second payload information.

[0201] (2) The terminal determines the CSI reporting information according to the payload of the CSI feedback information corresponding to the N layers.

[0202] That is, according to the payload of the CSI feedback information corresponding to the N layers indicated by the second payload information, the terminal determines K AI model units to determine the target AI model, and obtains the CSI reporting information based on the target AI model.

[0203] That is, according to the payload of the CSI feedback information corresponding to the N layers indicated by the second payload information, the terminal determines K AI model units to determine the target AI model, and obtains the CSI reporting information based on the target AI model.

[0204] For example, according to the payload of the CSI feedback information corresponding to the N layers indicated by the second payload information, the terminal determines to determine the CSI feedback information corresponding to the N layers according to the AI model units respectively, so as to obtain the CSI reporting information.

[0205] For example, according to the payload of the CSI feedback information corresponding to the N layers indicated by the second payload information, the terminal determines to determine the CSI feedback information corresponding to the N layers according to the N AI model units respectively, so as to obtain the CSI reporting information.

[0206] That is, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, the CSI reporting indication information includes Y second payloads, and / or the second payload and layer information association relationship, the layer information includes at least one of layer ID, layer list information and layer grouping information.

[0207] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0208] (1) The terminal determines the dataset corresponding to the N layers corresponding to the target Rank according to the second dataset ID;

[0209] (2) The terminal determines the CSI reporting information according to the dataset corresponding to the N layers;

[0210] That is, according to the dataset corresponding to the N layers indicated by the second dataset ID, the terminal determines K AI model units to determine the target AI model acquisition and obtain the CSI reporting information based on the target AI model.

[0211] For example, the terminal determines N AI model units according to N second dataset IDs, respectively, to determine the target AI model acquisition and obtain the CSI reporting information based on the target AI model.

[0212] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0213] (1) The terminal determines the CSI feedback information corresponding to the N layers corresponding to the target Rank according to the second model ID;

[0214] For example, the CSI reporting indication information includes a second model ID, and the terminal determines the AI model corresponding to the N layers corresponding to the target Rank, thereby determining the CSI feedback information corresponding to the N layers corresponding to the target Rank.

[0215] That is, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, the CSI reporting indication information includes Y second dataset IDs, and / or the second dataset ID and the layer information association relationship, the layer information includes at least one of the layer ID, the layer list information and the layer grouping information.

[0216] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0217] (1) The terminal determines K AI model units corresponding to the N layers corresponding to the target Rank according to the second model ID;

[0218] (2) The terminal determines the CSI reporting information according to the K AI model units corresponding to the N layers;

[0219] That is, according to the K AI model units corresponding to the N layers indicated by the second model ID, the terminal obtains a target AI model, and obtains the CSI reporting information based on the target AI model.

[0220] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0221] (1) The terminal determines the N AI model units corresponding to the N layers corresponding to the target Rank according to the N second model IDs;

[0222] (2) The terminal determines the CSI reporting information according to the N AI model units corresponding to the N layers;

[0223] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0224] (1) The terminal determines the N AI model units corresponding to the N layers corresponding to the target Rank according to the Y second model IDs;

[0225] (2) The terminal determines the CSI reporting information according to the N AI model units corresponding to the N layers;

[0226] Wherein, the Y second model IDs and the N layers can determine which models are selected for the N layers according to the order of the model IDs, or can be determined according to the association between the second model IDs and the layer information included in the CSI reporting indication information.

[0227] That is, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, the CSI reporting indication information includes Y second model IDs, and / or the association between the second model IDs and the layer information, and the layer information includes at least one of the layer ID, the layer list information and the layer grouping information.

[0228] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0229] (1) The terminal determines a first set according to the target Rank, the first payload information, and the second dataset ID, and the first set contains datasets corresponding to N layers corresponding to the target Rank;

[0230] (2) The terminal determines the CSI reporting information according to the first set;

[0231] The CSI reporting information is obtained through a target AI model, and the target AI model includes K AI model units corresponding to the datasets corresponding to the N layers.

[0232] The N layers satisfy:

[0233] The sum of the payloads of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0234] In the embodiments of the present application, the payloads of the CSI feedback information corresponding to the K layers in the first set are less than or equal to the payload indicated in the first payload information, and the AI model units corresponding to the K layers in the first set can be determined in combination with the second dataset ID.

[0235] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and CSI reporting indication information, including:

[0236] (1) The terminal determines a second set according to the target Rank, the first payload information, and the second model ID, and the second set contains K AI model units corresponding to N layers corresponding to the target Rank;

[0237] (2) The terminal determines the CSI reporting information according to the second set;

[0238] The CSI reporting information is obtained through a target AI model, and the target AI model includes K AI model units.

[0239] The N layers satisfy:

[0240] The sum of the payloads of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0241] In the embodiments of the present application, the payload of the CSI feedback information corresponding to the K layers in the second set is less than or equal to the payload indicated in the first payload information, and the AI model units corresponding to the K layers in the first set can be determined in combination with the second model ID.

[0242] In an optional implementation, the terminal determines the CSI reporting information according to the target Rank and the CSI reporting indication information, including:

[0243] (1) The terminal determines a third set according to the target Rank, the second payload information and the second model ID, and the third set includes K AI model units corresponding to N layers corresponding to the target Rank;

[0244] (2) The terminal determines the CSI reporting information according to the third set;

[0245] Wherein, the CSI reporting information is obtained through a target AI model, and the target AI model includes K AI model units corresponding to N layers;

[0246] Wherein, N layers satisfy:

[0247] The payload of the CSI feedback information corresponding to the N layers is the payload indicated by the second payload information.

[0248] In the embodiments of the present application, the K AI model units corresponding to the layers in the third set can be determined through the second model ID, which can be an AI model unit common to one layer, and the payload corresponding to the K layers in the third set can be determined in combination with the second payload information. In this way, the output information of the target AI model can be post-processed, such as truncation, quantization, etc.

[0249] In an optional implementation, the above N layers further satisfy:

[0250] Among the N layers, there are at least two layers corresponding to the same AI model unit, or K < N.

[0251] In an optional implementation, the method further includes:

[0252] In the case where the payload of the CSI feedback information is less than the payload indicated by the first payload information, the terminal performs any one of the following:

[0253] (1) performing padding processing on the CSI feedback information; for example, performing 0 padding on the remaining bits.

[0254] (2) The terminal sends first indication information to the network side device, and the first indication information is used to indicate the payload of the CSI feedback information.

[0255] In an optional implementation, the method further includes:

[0256] In a case where the payload of the CSI feedback information corresponding to the layer is less than the payload indicated by the second payload information, the terminal performs any one of the following:

[0257] (1) performs padding processing on the CSI feedback information; for example, performing 0 padding on the remaining bits.

[0258] (2) The terminal sends first indication information to the network side device, and the first indication information is used to indicate the payload of the CSI feedback information corresponding to the layer.

[0259] In an optional implementation, the CSI reporting indication information further includes at least one of the following:

[0260] (1) a report configuration identifier (Report config ID);

[0261] (2) a monitoring identifier (Monitoring ID);

[0262] (3) related information of the target AI model; for example, a model ID and a dataset ID, which are used to assist the terminal in determining an AI model unit;

[0263] (4) time domain information of CSI reporting; for example, at least one of a period and a starting position offset, for example, at least one of a period, a semi-static, and a non-periodic, which is used to indicate a time domain position at which the terminal sends the CSI reporting information;

[0264] (5) a type of the target AI model; for example, time-spatial-frequency compression (TSF), such as prediction-comprising compression, such as spatial-frequency compression (SF) of the present embodiment 0.

[0265] In an optional implementation, the CSI reporting information is sent to the network side device, including:

[0266] In a case where the target Rank is greater than a first value, the terminal sends, to the network side device, only CSI feedback information corresponding to a Rank less than or equal to the first value;

[0267] The first value is a value of the Rank information, or a value specified by a protocol, or a value configured by the network side device.

[0268] In the embodiments of the present application, if the target Rank obtained by the terminal is greater than the Rank indicated in the CSI reporting indication information, or greater than a value specified by a protocol (for example, Rank=4), or greater than a value configured by the network side device (for example, Rank=4), the terminal only reports the CSI feedback information corresponding to the Rank that is less than or equal to the Rank indicated in the CSI reporting indication information, or greater than the value specified by the protocol (for example, Rank=4), or greater than the value configured by the network side device (for example, Rank=4).

[0269] In an optional implementation, before the terminal receives the CSI reporting indication information from the network side device, the method further comprises:

[0270] The terminal receives first information from the network side device, and the first information comprises at least one of the following:

[0271] (1) first data set information, used to obtain a candidate AI model unit;

[0272] (2) first model indication information, used to obtain a candidate AI model unit.

[0273] The terminal sends second information to the network side device, and the second information comprises:

[0274] The first model indication information is used to indicate a candidate AI model unit.

[0275] In the embodiments of the present application, before the network side device sends the CSI reporting indication information to the terminal, the network side device provides at least one of the first data set information and the first model indication information for the terminal, which is used to obtain a candidate AI model unit (corresponding to the pre-set AI model unit described above), and then after the terminal receives the CSI reporting indication information, one or more AI model units are selected from the candidate AI model units to determine a target AI model.

[0276] In an optional implementation, the first data set information comprises at least one of the following:

[0277] (1) a plurality of first data, the first data comprising at least two of CSI information, CSI feedback information and reconstructed CSI information;

[0278] The first data described above can be used to train the obtained candidate AI model unit, that is, the first data described above can be used as training data for training the candidate AI model unit;

[0279] (2) the number of first data;

[0280] (3) dataset ID;

[0281] If the candidate AI model unit is determined according to the dataset, when the candidate AI model unit corresponding to the dataset ID is used, the input of the AI can be in a format corresponding to the CSI information in the dataset, the payload can be the payload of the CSI feedback information in the dataset, and the payload can be determined through the dataset ID;

[0282] (4) dataset format or data format;

[0283] For example, the payload of the CSI feedback information;

[0284] For example, the format of the CSI information, such as the number of antennas, the number of subbands, such as float 32, float 16

[0285] a) Wherein, it can be understood that the dataset format can be understood as all data in the entire dataset using the same format

[0286] b) Wherein, it can be understood that the data format can be understood as indicating the format of each data respectively.

[0287] (5) Second indication information, used to indicate that the first data corresponds to the layer, or the first data corresponds to the layer group, or the first data corresponds to the Rank, or the first data corresponds to the Rank and the layer; That is, used to indicate that the first data can be indicated as layer common, layer specific, Rank specific, or Rank specific and layer specific.

[0288] (6) Layer information; For example, layer ID, which can correspond to the case that the first data is layer specific;

[0289] (7) Rank information; Which can correspond to the case that the first data is Rank specific;

[0290] (8) The number of time domains corresponding to each first data. For example, the number of time domains is Y, if Y = 5, then it includes first data in 5 different time domain positions.

[0291] In an optional implementation, the first model indication information includes at least one of the following:

[0292] (1) model ID of the candidate AI model unit;

[0293] (2) third indication information for indicating that the candidate AI model unit corresponds to a layer, or the candidate AI model unit corresponds to a layer group, or the candidate AI model unit corresponds to a Rank, or the candidate AI model unit corresponds to a Rank and a layer, and the first data includes at least two of CSI information, CSI feedback information, and reconstructed CSI information; that is, indicating that the candidate AI model unit can be layer common, layer specific, Rank specific, or Rank specific and layer specific.

[0294] (3) layer information; for example, layer ID, which can correspond to the case that the first data is layer specific, and for example, layer group information or layer list information;

[0295] (4) Rank information; which can correspond to the case that the first data is Rank specific, or Rank;

[0296] (5) model structure indication information of the candidate AI model unit, for selecting one from a plurality of model structures;

[0297] (6) model parameter information of the candidate AI model unit;

[0298] The model parameter information includes weights (multiplicative coefficients, which can be understood as first parameters) and biases (additive coefficients, which can be understood as second parameters) of subunits or layers of the candidate AI model unit, or activation functions, convolution kernel related parameters of subunits or layers of the candidate AI model unit.

[0299] (7) model structure information of the candidate AI model unit, for obtaining the model structure.

[0300] In an optional implementation, after the terminal receives the first information from the network side device, the method further includes:

[0301] The terminal sends first response information to the network side device, and the first response information includes at least one of the following:

[0302] (1) model ID of the candidate AI model unit;

[0303] (2) dataset ID of the candidate AI model unit;

[0304] (3) dataset format or data format of the candidate AI model unit.

[0305] payload of CSI feedback information, for example;

[0306] format of CSI information, such as the number of antennas, the number of subbands, such as float 32, float 16

[0307] a) Wherein, it can be understood that the dataset format can be understood as all data in the entire dataset using the same format

[0308] b) Wherein, it can be understood that the data format can be understood as indicating the format of each data respectively.

[0309] In the embodiments of the present application, after the terminal obtains the candidate AI model unit according to the first information, the terminal sends the first response information to the network side device, and provides the network side with the related information of the candidate AI model unit obtained by the terminal, so as to ensure that the terminal side and the network side have the same understanding of the candidate AI model unit. In this way, when the network side device indicates the payload, the dataset and the AI model unit through the CSI reporting indication information in the subsequent terminal, the terminal can select the correct AI model unit from the candidate AI model unit.

[0310] In an optional implementation, the CSI reporting indication information further includes:

[0311] Fourth indication information, used to indicate a padding mode of the CSI feedback information, so as to be consistent with the input format of the AI model unit;

[0312] The padding mode can include, for example, zero padding, difference value, etc.

[0313] For example, the input of the AI model unit is determined according to the Rank indicated by the network, for example, the input vector is always equal to the Rank indicated by the network, and may not be equal to the actual Rank number reported by the terminal.

[0314] It should be noted that the input or output of the AI model in the embodiments of the present application includes at least one of the following: reference signal, channel-borne signal, channel state information, beam information, channel prediction information, interference information, positioning information, trajectory information, codebook information, original channel information, specified prediction information and management information, and control signaling.

[0315] Optionally, the function of the AI model includes at least one of the following:

[0316] a) Reference signal processing, including signal detection, filtering, equalization, etc. Including DMRS (Demodulation Reference Signal), SRS (Sounding Reference Signal), SSB (Synchronization Signal Block), CSI-RS (Channel State Information Reference Signal), TRS (Tracking reference signal), PRS (Positioning Reference Signals), PTRS (Phase-tracking reference signal), etc.

[0317] b) Channel signal transmission / reception / demodulation / sending. Including channel PDCCH (Physical Downlink Control Channel), PDSCH (Physical Downlink Shared Channel), PUCCH (Physical Uplink Control Channel), PUSCH (Physical Uplink Shared Channel), PRACH (Physical Random Access Channel), PBCH (Physical Broadcast Channel), etc.

[0318] c) Channel state information acquisition. Including:

[0319] i. Channel state information feedback. Including channel related information, channel matrix related information, channel feature information, channel matrix feature information, PMI (Precoding Matrix Indicator), RI (Rank Indication), CRI (CSI-RS Resource Indicator), CQI (Channel Quality Indicator), LI (Layer Indicator), etc.

[0320] ii. FDD (Frequency Division Duplexing) uplink and downlink partial reciprocity. For FDD systems, according to the partial reciprocity, the base station obtains the angle and time delay information according to the uplink channel, and can inform the UE of the angle information and the time delay information through the CSI-RS precoding or direct indication method. The UE reports according to the indication of the base station or selects and reports within the indication range of the base station, thereby reducing the calculation amount of the UE and the overhead of CSI reporting.

[0321] d) Beam management. Including beam measurement, beam reporting, beam prediction, beam failure detection, beam failure recovery, new beam indication in beam failure recovery.

[0322] e) Channel prediction. Including prediction of channel state information, beam prediction.

[0323] f) Channel / source coding. Such as channel encoding, channel decoding, source encoding, source decoding, joint source and channel encoding, joint source and channel decoding.

[0324] g) Interference suppression. Including intra-cell interference, inter-cell interference, out-of-band interference, intermodulation interference, etc.

[0325] h) Positioning. The specific position (including horizontal position and / or vertical position) of the UE estimated by the reference signal (such as SRS) or the future possible trajectory, or the information assisting the position estimation or trajectory estimation.

[0326] i) Prediction and management of high-level services and parameters. Including throughput, required packet size, service demand, moving speed, noise information, etc.

[0327] j) Analysis of control signaling. Such as power control related signaling, beam management related signaling.

[0328] In summary, in the embodiments of the present application, the matrix representation AI model is used to realize the interaction and configuration between different devices, so that the understanding of the AI model by different devices is consistent, and then the cooperation problem of AI model training can be solved.

[0329] In order to more clearly describe the method for determining the AI model provided in the embodiments of the present application, the following will be described in conjunction with several examples.

[0330] Example one, implementation scene of AI model application method

[0331] Scene 1: In the traditional space-frequency domain compression, the CSI channel information (space-frequency domain channel) of time slot X belongs to the encoder, which generates the CSI reporting information. The network device receives the CSI reporting information for decoding, and obtains the recovered CSI information through the decoder.

[0332] Scenario 2: In the space-time-frequency domain compression, the encoder will use the CSI information at time X and use the intermediate state information (a kind of buffer information) of the encoder-1, that is, the previous historical information, to assist the encoding. The intermediate state information of the decoder is the same.

[0333] Scenario 3: The space-time-frequency domain compression uses the CSI information (such as PMI3) at time X and the CSI information (such as PMI0, PMI1, PMI2) at times X-N, X-2N, …, X+M*N. It can be understood that the CSI information output by the decoder can be the CSI information at the current time X, or the CSI information at future times X+L, X+L+D, …, X+L+(K)*D.

[0334] Referring to FIG. 3, the embodiment of the present application provides a CSI receiving method, and the execution subject of the method is a network side device, which comprises the following steps:

[0335] Step 301: The network side device sends CSI reporting indication information to the terminal.

[0336] Step 302: The network side device receives CSI reporting information from the terminal.

[0337] The CSI reporting information comprises CSI feedback information, and the CSI feedback information is obtained through a target AI model. The target AI model comprises one or more AI model units. The CSI reporting indication information is used to indicate at least one of the payload of the CSI feedback information, the dataset associated with the AI model unit, and the AI model unit.

[0338] It should be noted that the network side device is the opposite device interacting with the terminal, and the information interaction steps and information content should be the same as those of the terminal. Therefore, the related content of the network side method provided by the embodiment of the present application can be understood by referring to the terminal side method content, and will not be repeated here.

[0339] In an optional implementation, the CSI reporting indication information comprises at least one of the following:

[0340] First payload information, used to indicate the payload of the CSI feedback information corresponding to the Rank information.

[0341] Second payload information, used to indicate the payload of the CSI feedback information corresponding to the layer information.

[0342] The first dataset ID is used to indicate a dataset corresponding to the Rank information.

[0343] The second dataset ID is used to indicate a dataset corresponding to the layer information.

[0344] The first model ID is used to indicate an AI model unit corresponding to the Rank information.

[0345] The second model ID is used to indicate an AI model unit corresponding to the layer information.

[0346] In an optional implementation, the layer information includes at least one of a layer ID and layer grouping information.

[0347] The Rank information includes at least one of a Rank value and Rank grouping information.

[0348] In an optional implementation, the CSI feedback information includes CSI feedback information corresponding to N layers, and the target AI model includes K AI model units, and K is less than or equal to N.

[0349] In an optional implementation, the CSI reporting indication information includes at least one of second payload information, a second dataset ID, and a second model ID.

[0350] The K AI model units are respectively used to process CSI information corresponding to different layers, and payloads of the CSI feedback information corresponding to different layers are different.

[0351] In an optional implementation, the AI model unit includes any one of the following:

[0352] A first AI model unit used by a terminal and a network side device;

[0353] A reference AI model unit of a second AI model unit used by the terminal and the network side device

[0354] A third AI model unit used by a terminal, a network side device, and a test device in testing;

[0355] A reference AI model unit of a fourth AI model unit used by the terminal, the network side device, and the test device in testing.

[0356] In an optional implementation, the K AI model units satisfy any one of the following:

[0357] The K AI model units are K AI model units corresponding to K layers;

[0358] K AI model units are AI model units corresponding to a target Rank;

[0359] K AI model units are AI model units corresponding to a layer group.

[0360] In an optional implementation, the CSI reporting information further includes at least one of the following:

[0361] a target Rank;

[0362] a CQI;

[0363] a payload corresponding to at least one layer;

[0364] a model ID corresponding to a target AI model;

[0365] a dataset ID corresponding to the target AI model.

[0366] In an optional implementation, the method further includes:

[0367] The network-side device receives first indication information from the terminal, and the first indication information is used to indicate a payload of CSI feedback information.

[0368] In an optional implementation, the CSI reporting indication information further includes at least one of the following:

[0369] a report configuration identifier;

[0370] a monitoring identifier;

[0371] related information of a target AI model;

[0372] time domain information of CSI reporting;

[0373] a type of the target AI model.

[0374] In an optional implementation, before the network-side device sends the CSI reporting indication information to the terminal, the method further includes:

[0375] The network-side device sends first information to the terminal, and the first information includes at least one of the following:

[0376] first dataset information, used to obtain candidate AI model units;

[0377] first model indication information, used to obtain candidate AI model units.

[0378] In an optional implementation, the first dataset information includes at least one of the following:

[0379] a plurality of first data, the first data comprising at least two of CSI information, CSI feedback information, and reconstructed CSI information;

[0380] a number of the first data;

[0381] a dataset ID of the candidate AI model unit;

[0382] a dataset format or a data format of the candidate AI model unit;

[0383] second indication information, used for indicating that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer;

[0384] layer information;

[0385] Rank information;

[0386] a time domain number corresponding to each of the first data.

[0387] In an optional implementation, the first model indication information comprises at least one of:

[0388] a model ID of the candidate AI model unit;

[0389] third indication information, used for indicating that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer, the first data comprising at least two of CSI information, CSI feedback information, and reconstructed CSI information;

[0390] layer information;

[0391] Rank information;

[0392] model structure indication information of the candidate AI model unit;

[0393] model parameter information of the candidate AI model unit;

[0394] model structure information of the candidate AI model unit.

[0395] In an optional implementation, after the network side device sends the first information to the terminal, the method further comprises:

[0396] the network side device receives first response information from the terminal, the first response information comprising at least one of:

[0397] a model ID of the candidate AI model unit;

[0398] a dataset ID of the candidate AI model unit;

[0399] a dataset format or a data format of the candidate AI model unit.

[0400] In an optional implementation, the CSI reporting indication information further includes:

[0401] fourth indication information used for indicating a padding manner of the CSI feedback information.

[0402] The CSI reporting / receiving method provided in the embodiments of the present application can be executed by a CSI reporting / receiving device. In the embodiments of the present application, the CSI reporting / receiving method executed by the CSI reporting / receiving device is taken as an example to illustrate the CSI reporting / receiving device provided in the embodiments of the present application.

[0403] The CSI reporting / receiving device provided in the embodiments of the present application can be a communication device or a component in a communication device, for example, a chip. The communication device can be a terminal, a network side device or a server, etc. For example, the terminal can include but is not limited to the types of the terminal 11 listed above, the network side device can include but is not limited to the types of the network side device 12 listed above, and the embodiments of the present application are not limited in this regard.

[0404] The CSI reporting / receiving device includes a receiving module, a sending module and a processing module. The receiving module, the sending module and the processing module can be implemented by software or hardware. When implemented by hardware, the processing module can be implemented by a processor, for example, a general processor, a special purpose processor, etc., such as a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), an artificial intelligent (AI) processor, a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a network processor (NP), a field programmable gate array (FPGA) or other programmable logic device, a gate circuit, a transistor, a discrete hardware component, etc. The receiving module and the sending module can be implemented by a communication interface, which can include one or more of a transceiver, a pin, a circuit, a bus, a radio frequency unit, etc.

[0405] Specifically, referring to FIG. 4, when the CSI reporting device is a terminal or a component in the terminal, the CSI reporting device 400 includes:

[0406] A first receiving module 401, configured to receive, by a terminal, CSI reporting indication information from a network side device;

[0407] A first processing module 402, configured to acquire, by the terminal, target Rank;

[0408] A second processing module 403, configured to determine, by the terminal, CSI reporting information according to the target Rank and the CSI reporting indication information;

[0409] A first sending module 404, configured to send, to the network side device, the CSI reporting information;

[0410] The CSI reporting information includes CSI feedback information, the CSI feedback information is acquired through a target AI model, the target AI model includes one or more AI model units, and the CSI reporting indication information is used to indicate at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0411] In an optional implementation, the CSI reporting indication information includes at least one of the following:

[0412] First payload information, used to indicate a payload of CSI feedback information corresponding to Rank information;

[0413] Second payload information, used to indicate a payload of CSI feedback information corresponding to layer information.

[0414] First dataset identification dataset ID, used to indicate a dataset corresponding to Rank information;

[0415] Second dataset ID, used to indicate a dataset corresponding to layer information;

[0416] First model identification model ID, used to indicate an AI model unit corresponding to Rank information;

[0417] Second model ID, used to indicate an AI model unit corresponding to layer information.

[0418] In an optional implementation, the layer information includes at least one of a layer ID and layer grouping information.

[0419] The Rank information includes at least one of a Rank value and Rank grouping information.

[0420] In an optional implementation, the CSI feedback information includes CSI feedback information corresponding to N layers, and the target AI model includes K AI model units, and the K is less than or equal to the N.

[0421] In an optional implementation, the CSI reporting indication information includes at least one of the second payload information, the second dataset ID and the second model ID.

[0422] The K AI model units are respectively used for processing CSI information corresponding to different layers, and payload of the CSI feedback information corresponding to the different layers is different.

[0423] In an optional implementation, the AI model unit includes any one of the following:

[0424] A first AI model unit used by the terminal and the network-side device;

[0425] A reference AI model unit of a second AI model unit used by the terminal and the network-side device

[0426] A third AI model unit used by the terminal, the network-side device and a test device in testing;

[0427] A reference AI model unit of a fourth AI model unit used by the terminal, the network-side device and the test device in testing.

[0428] In an optional implementation, the K AI model units satisfy any one of the following:

[0429] The K AI model units are K AI model units corresponding to K layers;

[0430] The K AI model units are K AI model units corresponding to the target Rank;

[0431] The K AI model units are K AI model units corresponding to one layer grouping.

[0432] In an optional implementation, the CSI reporting information further includes at least one of the following:

[0433] The target Rank;

[0434] Channel quality information CQI;

[0435] payload corresponding to the at least one layer;

[0436] a model ID corresponding to the target AI model;

[0437] a dataset ID corresponding to the target AI model.

[0438] In an optional implementation, the second processing module is configured to:

[0439] The terminal determines, according to the target Rank and the first payload information, payload of CSI feedback information corresponding to N layers corresponding to the target Rank.

[0440] The terminal determines the CSI reporting information according to the payload of the CSI feedback information corresponding to the N layers.

[0441] The CSI reporting information is obtained through the target AI model, and the target AI model includes K AI model units corresponding to the payload of the CSI feedback information corresponding to the N layers.

[0442] The payload of the CSI feedback information corresponding to the N layers satisfies:

[0443] The sum of the payload of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0444] In the N layers, the payload of the CSI feedback information corresponding to a first layer ID is greater than or equal to the payload of the CSI feedback information corresponding to a second layer ID.

[0445] In the N layers, the difference between the payload of the CSI feedback information corresponding to the first layer ID and the payload of the CSI feedback information corresponding to the second layer ID is less than or equal to a first threshold.

[0446] The first layer ID is less than the second layer ID.

[0447] In an optional implementation, the second processing module is configured to:

[0448] The terminal determines a first set according to the target Rank, the first payload information, and the second dataset ID, the first set including datasets corresponding to N layers corresponding to the target Rank;

[0449] The terminal determines the CSI reporting information according to the first set;

[0450] The CSI reporting information is obtained through the target AI model, and the target AI model includes K AI model units corresponding to datasets corresponding to the N layers.

[0451] The N layers satisfy:

[0452] The sum of payloads of CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0453] In an optional implementation, the second processing module is configured to:

[0454] The terminal determines a second set according to the target Rank, the first payload information, and the second model ID, the second set including K AI model units corresponding to N layers corresponding to the target Rank;

[0455] The terminal determines the CSI reporting information according to the second set;

[0456] The CSI reporting information is obtained through the target AI model, and the target AI model includes the K AI model units.

[0457] The N layers satisfy:

[0458] The sum of payloads of CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0459] In an optional implementation, the second processing module is configured to:

[0460] The terminal determines a third set according to the target Rank, the second payload information, and the second model ID, the third set including K AI model units corresponding to N layers corresponding to the target Rank;

[0461] The terminal determines the CSI reporting information according to the third set;

[0462] The CSI reporting information is obtained through the target AI model, and the target AI model includes K AI model units corresponding to the N layers.

[0463] The N layers satisfy:

[0464] The payload of the CSI feedback information corresponding to the N layers is the payload indicated by the second payload information.

[0465] In an optional implementation, the N layers further satisfy:

[0466] In the N layers, at least two layers correspond to the same AI model unit.

[0467] Or;

[0468] The K is less than N.

[0469] In an optional implementation, the apparatus further includes:

[0470] The third processing module is configured to, in a case where the payload of the CSI feedback information is less than the payload indicated by the first payload information, perform any one of the following by the terminal:

[0471] Padding processing is performed on the CSI feedback information;

[0472] The terminal sends first indication information to the network side device, and the first indication information is used to indicate the payload of the CSI feedback information.

[0473] In an optional implementation, the CSI reporting indication information further includes at least one of the following:

[0474] Reporting configuration identification;

[0475] Monitoring identification;

[0476] Related information of the target AI model;

[0477] Time domain information of CSI reporting;

[0478] Type of the target AI model.

[0479] In an optional implementation, the first sending module is configured to:

[0480] In a case where the target Rank is greater than a first value, the terminal only sends, to the network side device, CSI feedback information corresponding to a Rank less than or equal to the first value.

[0481] The first value is a value of the Rank information, or a value specified by a protocol, or a value configured by the network side device.

[0482] In an optional implementation, the apparatus further includes:

[0483] The second receiving module is configured to receive, before the terminal receives the CSI reporting indication information from the network side device, first information from the network side device, the first information including at least one of the following:

[0484] The first dataset information is used to obtain a candidate AI model unit.

[0485] The first model indication information is used to obtain the candidate AI model unit.

[0486] In an optional implementation, the first dataset information includes at least one of the following:

[0487] A plurality of first data, the first data including at least two of the following: CSI information, CSI feedback information, and reconstructed CSI information.

[0488] The number of the first data.

[0489] The dataset ID of the candidate AI model unit.

[0490] The dataset format dataset format or data format data format of the candidate AI model unit.

[0491] Second indication information is used to indicate that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer.

[0492] The layer information.

[0493] The Rank information.

[0494] The number of time domains corresponding to each of the first data.

[0495] In an optional implementation, the first model indication information includes at least one of the following:

[0496] The model ID of the candidate AI model unit.

[0497] The third indication information is used for indicating that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer, and the first data includes at least two of CSI information, CSI feedback information and reconstructed CSI information.

[0498] Layer information;

[0499] Rank information;

[0500] Model structure indication information of the candidate AI model unit;

[0501] Model parameter information of the candidate AI model unit;

[0502] Model structure information of the candidate AI model unit.

[0503] In an optional implementation, the apparatus further includes:

[0504] The second sending module is configured to, after the terminal receives the first information from the network side device, send first response information to the network side device, and the first response information includes at least one of the following:

[0505] The model ID of the candidate AI model unit;

[0506] The dataset ID of the candidate AI model unit;

[0507] The dataset format or data format of the candidate AI model unit.

[0508] In an optional implementation, the CSI reporting indication information further includes:

[0509] The fourth indication information is used for indicating a filling mode of the CSI feedback information.

[0510] Specifically, referring to FIG. 5, when the CSI receiving apparatus is a network side device or a component in the network side device, the CSI receiving apparatus 500 includes:

[0511] The third sending module 501 is configured to send, by the network side device, CSI reporting indication information to a terminal;

[0512] The third receiving module 502 is configured to receive, by the network side device, CSI reporting information from the terminal;

[0513] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used for indicating at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0514] In an optional implementation, the CSI reporting indication information includes at least one of the following:

[0515] First payload information, used for indicating a payload of CSI feedback information corresponding to Rank information;

[0516] Second payload information, used for indicating a payload of CSI feedback information corresponding to layer information.

[0517] First dataset ID, used for indicating a dataset corresponding to Rank information;

[0518] Second dataset ID, used for indicating a dataset corresponding to layer information;

[0519] First model ID, used for indicating an AI model unit corresponding to Rank information;

[0520] Second model ID, used for indicating an AI model unit corresponding to layer information.

[0521] In an optional implementation, the layer information includes at least one of a layer ID and layer grouping information.

[0522] The Rank information includes at least one of a Rank value and Rank grouping information.

[0523] In an optional implementation, the CSI feedback information includes CSI feedback information corresponding to N layers, and the target AI model includes K AI model units, where K is less than or equal to N.

[0524] In an optional implementation, the CSI reporting indication information includes at least one of the second payload information, the second dataset ID, and the second model ID.

[0525] The K AI model units are respectively used for processing CSI information corresponding to different layers, and payloads of CSI feedback information corresponding to the different layers are different.

[0526] In an optional implementation, the AI model unit includes any of the following:

[0527] a first AI model unit used by the terminal and the network-side device;

[0528] a second AI model unit used by the terminal and the network-side device, and a reference AI model unit of the second AI model unit;

[0529] a third AI model unit used by the terminal, the network-side device, and a test device in testing;

[0530] a fourth AI model unit used by the terminal, the network-side device, and a test device in testing, and a reference AI model unit of the fourth AI model unit.

[0531] In an optional implementation, the K AI model units satisfy any of the following:

[0532] The K AI model units are K AI model units corresponding to K layers;

[0533] The K AI model units are AI model units corresponding to a target Rank;

[0534] The K AI model units are AI model units corresponding to a layer group.

[0535] In an optional implementation, the CSI reporting information further includes at least one of the following:

[0536] the target Rank;

[0537] CQI;

[0538] a payload corresponding to at least one layer;

[0539] a model ID corresponding to the target AI model;

[0540] a dataset ID corresponding to the target AI model.

[0541] In an optional implementation, the apparatus further includes:

[0542] a fifth receiving module configured to receive, by the network-side device, first indication information from the terminal, the first indication information being used to indicate a payload of the CSI feedback information.

[0543] In an optional implementation, the CSI reporting indication information further includes at least one of the following:

[0544] reporting configuration identification;

[0545] monitoring identifier;

[0546] related information of the target AI model;

[0547] time domain information of CSI reporting;

[0548] type of the target AI model.

[0549] In an optional implementation, the apparatus further includes:

[0550] A fourth sending module, configured to, before the network side device sends CSI reporting indication information to a terminal, the network side device sends first information to the terminal, the first information including at least one of the following:

[0551] first dataset information, used to obtain a candidate AI model unit;

[0552] first model indication information, used to obtain the candidate AI model unit.

[0553] In an optional implementation, the first dataset information includes at least one of the following:

[0554] a plurality of first data, the first data including at least two of the following: CSI information, CSI feedback information and reconstructed CSI information;

[0555] number of the first data;

[0556] dataset ID of the candidate AI model unit;

[0557] dataset format or data format of the candidate AI model unit;

[0558] second indication information, used to indicate that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer;

[0559] the layer information;

[0560] the Rank information;

[0561] a number of time domains corresponding to each of the first data.

[0562] In an optional implementation, the first model indication information includes at least one of the following:

[0563] model ID of the candidate AI model unit;

[0564] The third indication information is used for indicating that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer, and the first data includes at least two of CSI information, CSI feedback information and reconstructed CSI information.

[0565] Layer information;

[0566] Rank information;

[0567] Model structure indication information of the candidate AI model unit;

[0568] Model parameter information of the candidate AI model unit;

[0569] Model structure information of the candidate AI model unit.

[0570] In an optional implementation, the apparatus further includes:

[0571] The fourth receiving module is configured to receive, by the network side device, first response information from the terminal after the network side device sends the first information to the terminal, and the first response information includes at least one of the following:

[0572] The model ID of the candidate AI model unit;

[0573] The dataset ID of the candidate AI model unit;

[0574] The dataset format or the data format of the candidate AI model unit.

[0575] In an optional implementation, the CSI reporting indication information further includes:

[0576] The fourth indication information is used for indicating a filling mode of the CSI feedback information.

[0577] The apparatus provided by the embodiments of the present application can implement each process implemented by the method embodiments of FIGS. 2 to 3 and achieve the same technical effects. To avoid repetition, details are not described herein.

[0578] As shown in FIG. 6, the embodiment of the present application further provides a communication device 600, comprising a processor 601 and a memory 602, wherein the memory 602 stores programs or instructions executable by the processor 601. For example, when the communication device 600 is a terminal, the programs or instructions are executed by the processor 601 to implement each step of the above method embodiments and achieve the same technical effects. When the communication device 600 is a network side device, the programs or instructions are executed by the processor 601 to implement each step of the above method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0579] The embodiment of the present application further provides a terminal, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to run programs or instructions to implement steps in the method embodiment shown in FIG. 2. The terminal embodiment corresponds to the above terminal side method embodiment, and each implementation process and implementation manner of the above method embodiment can be applied to the terminal embodiment and achieve the same technical effects. The terminal can be the CSI reporting apparatus shown in FIG. 4. Specifically, FIG. 7 is a schematic diagram of a hardware structure of a terminal for implementing the embodiment of the present application.

[0580] The terminal 700 includes, but is not limited to, at least part of the following components: a radio frequency unit 701, a network module 702, an audio output unit 703, an input unit 704, a sensor 705, a display unit 706, a user input unit 707, an interface unit 708, a memory 709, and a processor 710.

[0581] Those skilled in the art can understand that the terminal 700 can further include a power supply (such as a battery) for supplying power to each component, and the power supply can be logically connected to the processor 710 through a power management system, so as to realize functions such as power management, discharge management and power consumption management through the power management system. The terminal structure shown in FIG. 7 does not constitute a limitation on the terminal, and the terminal can include more or fewer components than those shown, or combine certain components, or different component arrangements, which are not described herein.

[0582] It should be understood that in the embodiments of the present application, the input unit 704 can include a graphics processor 7041 and a microphone 7042, and the graphics processor 7041 processes image data of a still picture or a video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 706 can include a display panel 7061, which can be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 707 includes at least one of a touch panel 7071 and other input devices 7072. The touch panel 7071 is also called a touch screen. The touch panel 7071 can include two parts of a touch detection device and a touch controller. The other input devices 7072 can include, but are not limited to, a physical keyboard, function keys (such as volume control keys, on-off keys, etc.), a trackball, a mouse, a joystick, and the like, which will not be described here.

[0583] In the embodiments of the present application, after the radio frequency unit 701 receives the downlink data from the network side device, it can be transmitted to the processor 710 for processing. In addition, the radio frequency unit 701 can send uplink data to the network side device. Generally, the radio frequency unit 701 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.

[0584] The memory 709 can be used to store software programs or instructions and various data. The memory 709 can mainly include a first storage area storing programs or instructions and a second storage area storing data, wherein the first storage area can store an operating system, application programs or instructions required by at least one function (such as a sound playing function, an image playing function, etc.), and the like. In addition, the memory 709 can include a volatile memory or a non-volatile memory. The non-volatile memory can be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable Programmable ROM (EPROM), an Electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a Random Access Memory (RAM), a Static RAM (SRAM), a Dynamic RAM (DRAM), a Synchronous DRAM (SDRAM), a Double Data Rate SDRAM (DDR SDRAM), an Enhanced SDRAM (ESDRAM), a Synchlink DRAM (SLDRAM), and a Direct Rambus RAM (DRRAM). The memory 709 in the embodiments of the present application includes but is not limited to these and any other suitable types of memory.

[0585] The processor 710 can include one or more processing units; optionally, the processor 710 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and an application program, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 710.

[0586] The processor 710 is configured to:

[0587] The terminal receives CSI reporting indication information from a network side device;

[0588] The terminal acquires a target Rank;

[0589] The terminal determines CSI reporting information according to the target Rank and the CSI reporting indication information;

[0590] sending the CSI reporting information to the network side device;

[0591] The CSI reporting information includes CSI feedback information, the CSI feedback information is obtained through a target AI model, the target AI model includes one or more AI model units, the CSI reporting indication information is used for indicating at least one of a payload of the CSI feedback information, a dataset associated with the AI model unit, and the AI model unit.

[0592] In an optional implementation, the CSI reporting indication information includes at least one of the following:

[0593] First payload information, used for indicating a payload of CSI feedback information corresponding to Rank information;

[0594] Second payload information, used for indicating a payload of CSI feedback information corresponding to layer information.

[0595] First dataset identification dataset ID, used for indicating a dataset corresponding to Rank information;

[0596] Second dataset ID, used for indicating a dataset corresponding to layer information;

[0597] First model identification model ID, used for indicating an AI model unit corresponding to Rank information;

[0598] Second model ID, used for indicating an AI model unit corresponding to layer information.

[0599] In an optional implementation, the layer information includes at least one of a layer ID and layer grouping information.

[0600] The Rank information includes at least one of a Rank value and Rank grouping information.

[0601] In an optional implementation, the CSI feedback information includes CSI feedback information corresponding to N layers, and the target AI model includes K AI model units, and the K is less than or equal to the N.

[0602] In an optional implementation, the CSI reporting indication information includes at least one of the second payload information, the second dataset ID, and the second model ID.

[0603] The K AI model units are respectively used for processing CSI information corresponding to different layers, and payload of the CSI feedback information corresponding to the different layers is different.

[0604] In an optional implementation, the AI model unit includes any one of the following:

[0605] The first AI model unit used by the terminal and the network-side device;

[0606] The reference AI model unit of the second AI model unit used by the terminal and the network-side device

[0607] The third AI model unit used by the terminal, the network-side device, and the test device in testing;

[0608] The reference AI model unit of the fourth AI model unit used by the terminal, the network-side device, and the test device in testing.

[0609] In an optional implementation, the K AI model units satisfy any one of the following:

[0610] The K AI model units are AI model units corresponding to K layers;

[0611] The K AI model units are AI model units corresponding to the target Rank;

[0612] The K AI model units are AI model units corresponding to one layer group.

[0613] In an optional implementation, the CSI reporting information further includes at least one of the following:

[0614] The target Rank;

[0615] Channel quality information CQI;

[0616] The payload corresponding to at least one layer;

[0617] The model ID corresponding to the target AI model;

[0618] The dataset ID corresponding to the target AI model.

[0619] In an optional implementation, the processor 710 is configured to:

[0620] The terminal determines, according to the target Rank and the first payload information, the payload of the CSI feedback information corresponding to N layers corresponding to the target Rank.

[0621] The terminal determines the CSI reporting information according to the payloads of the CSI feedback information corresponding to the N layers;

[0622] The CSI reporting information is obtained through the target AI model, and the target AI model includes K AI model units corresponding to the payloads of the CSI feedback information corresponding to the N layers;

[0623] The payloads of the CSI feedback information corresponding to the N layers satisfy one or more of the following conditions:

[0624] The sum of the payloads of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information;

[0625] In the N layers, the payload of the CSI feedback information corresponding to the first layer ID is greater than or equal to the payload of the CSI feedback information corresponding to the second layer ID;

[0626] In the N layers, the difference between the payload of the CSI feedback information corresponding to the first layer ID and the payload of the CSI feedback information corresponding to the second layer ID is less than or equal to a first threshold value;

[0627] The first layer ID is less than the second layer ID.

[0628] In an optional implementation, the processor 710 is configured to:

[0629] The terminal determines a first set according to a target rank, the first payload information, and the second dataset ID, and the first set includes datasets corresponding to the N layers corresponding to the target rank;

[0630] The terminal determines the CSI reporting information according to the first set;

[0631] The CSI reporting information is obtained through the target AI model, and the target AI model includes K AI model units corresponding to the datasets corresponding to the N layers;

[0632] The N layers satisfy:

[0633] A sum of payloads of CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0634] In an optional implementation, the processor 710 is configured to:

[0635] The terminal determines a second set according to the target Rank, the first payload information, and the second model ID, the second set containing K AI model units corresponding to N layers corresponding to the target Rank;

[0636] The terminal determines the CSI reporting information according to the second set;

[0637] The CSI reporting information is obtained through the target AI model, and the target AI model includes the K AI model units.

[0638] The N layers satisfy:

[0639] A sum of payloads of CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0640] In an optional implementation, the processor 710 is configured to:

[0641] The terminal determines a third set according to the target Rank, the second payload information, and the second model ID, the third set containing K AI model units corresponding to N layers corresponding to the target Rank;

[0642] The terminal determines the CSI reporting information according to the third set;

[0643] The CSI reporting information is obtained through the target AI model, and the target AI model includes the K AI model units corresponding to the N layers.

[0644] The N layers satisfy:

[0645] A sum of payloads of CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

[0646] In an optional implementation, the N layers further satisfy:

[0647] In the N layers, there are at least two layers corresponding to the same AI model unit.

[0648] or;

[0649] The K < N.

[0650] In an optional implementation, the processor 710 is configured to, in a case where a payload of the CSI feedback information is less than a payload indicated by the first payload information, perform any one of the following by the terminal:

[0651] perform padding processing on the CSI feedback information;

[0652] The terminal sends first indication information to the network side device, and the first indication information is used to indicate the payload of the CSI feedback information.

[0653] In an optional implementation, the CSI reporting indication information further includes at least one of the following:

[0654] reporting configuration identification;

[0655] monitoring identification;

[0656] related information of the target AI model;

[0657] time domain information of CSI reporting;

[0658] type of the target AI model.

[0659] In an optional implementation, the processor 710 is configured to:

[0660] In a case where the target Rank is greater than a first value, the terminal sends, to the network side device, only CSI feedback information corresponding to a Rank less than or equal to the first value;

[0661] The first value is a value of the Rank information, or a value specified by a protocol, or a value configured by the network side device.

[0662] In an optional implementation, the processor 710 is configured to, before the terminal receives the CSI reporting indication information from the network side device, receive, by the terminal, first information from the network side device, and the first information includes at least one of the following:

[0663] first data set information used to obtain a candidate AI model unit;

[0664] first model indication information used to obtain the candidate AI model unit.

[0665] In an optional implementation, the first data set information includes at least one of the following:

[0666] a plurality of first data, the first data comprising at least two of CSI information, CSI feedback information, and reconstructed CSI information;

[0667] a number of the first data;

[0668] a dataset ID of the candidate AI model unit;

[0669] a dataset format or a data format of the candidate AI model unit;

[0670] second indication information, used for indicating that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer;

[0671] the layer information;

[0672] the Rank information;

[0673] a time domain number corresponding to each of the first data.

[0674] In an optional implementation, the first model indication information comprises at least one of:

[0675] a model ID of the candidate AI model unit;

[0676] third indication information, used for indicating that the first data corresponds to a layer, or the first data corresponds to a layer group, or the first data corresponds to a Rank, or the first data corresponds to a Rank and a layer, the first data comprising at least two of CSI information, CSI feedback information, and reconstructed CSI information;

[0677] layer information;

[0678] Rank information;

[0679] model structure indication information of the candidate AI model unit;

[0680] model parameter information of the candidate AI model unit;

[0681] model structure information of the candidate AI model unit.

[0682] In an optional implementation, the processor 710 is configured to, after the terminal receives the first information from the network side device, send first response information to the network side device, the first response information comprising at least one of:

[0683] a model ID of the candidate AI model unit;

[0684] a dataset ID of the candidate AI model unit;

[0685] a dataset format or a data format of the candidate AI model unit.

[0686] In an optional implementation, the CSI reporting indication information further includes:

[0687] fourth indication information for indicating a padding mode of the CSI feedback information.

[0688] It can be understood that the implementation processes of the implementation manners mentioned in the embodiments can refer to the related descriptions of the method embodiments and achieve the same or corresponding technical effects. To avoid repetition, they will not be described here again.

[0689] The embodiments of the present application also provide a network side device, which includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to realize the steps of the method embodiments shown in FIG. 3. The network side device embodiments correspond to the network side device method embodiments described above. The various implementation processes and implementation manners of the method embodiments described above can be applied to the network side device embodiments, and the same technical effects can be achieved.

[0690] Specifically, the embodiments of the present application also provide a network side device, which can be the CSI receiving apparatus shown in FIG. 5. As shown in FIG. 8, the network side device 800 includes an antenna 81, a radio frequency device 82, a baseband device 83, a processor 84 and a memory 85. The antenna 81 is connected with the radio frequency device 82. In the uplink direction, the radio frequency device 82 receives information through the antenna 81, and sends the received information to the baseband device 83 for processing. In the downlink direction, the baseband device 83 processes the information to be sent and sends it to the radio frequency device 82. The radio frequency device 82 processes the received information and sends it out through the antenna 81.

[0691] The method performed by the network side device in the above embodiments can be implemented in the baseband device 83, which includes a baseband processor.

[0692] The baseband device 83 may, for example, include at least one baseband board, which is provided with a plurality of chips, as shown in FIG. 8. One of the chips is, for example, a baseband processor, which is connected with the memory 85 through a bus interface to call the programs in the memory 85 and perform the network device operations shown in the above method embodiments.

[0693] The network-side device can further include a network interface 86, for example, a Common Public Radio Interface (CPRI).

[0694] Specifically, the network-side device 800 of the embodiments of the present application further includes instructions or programs stored on the memory 85 and executable on the processor 84, the processor 84 invoking the instructions or programs in the memory 85 to perform the method performed by the modules shown in FIG. 5 and achieve the same technical effects. To avoid repetition, details are not described herein.

[0695] Specifically, the embodiments of the present application further provide a network-side device. As shown in FIG. 9, the network-side device 900 includes a processor 901, a network interface 902, and a memory 903. The network-side device can be the CSI receiving apparatus shown in FIG. 5. The network interface 902 is, for example, a common public radio interface (CPRI).

[0696] Specifically, the network-side device 900 of the embodiments of the present application further includes instructions or programs stored on the memory 903 and executable on the processor 901, the processor 901 invoking the instructions or programs in the memory 903 to perform the method performed by the modules shown in FIG. 5 and achieve the same technical effects. To avoid repetition, details are not described herein.

[0697] The embodiments of the present application further provide a readable storage medium, the readable storage medium storing programs or instructions, the programs or instructions being executed by a processor to implement various processes of the above method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0698] The processor is the processor in the terminal in the above embodiments. The readable storage medium includes a computer readable storage medium, such as a computer readable only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc. In some examples, the readable storage medium can be a non-transitory readable storage medium.

[0699] The embodiments of the present application further provide a chip, the chip including a processor and a communication interface, the communication interface being coupled with the processor, the processor being configured to run programs or instructions to implement various processes of the above method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0700] It should be understood that the chip mentioned in the embodiments of the present application can also be referred to as a system chip, a system on chip, a chip system, or a system on chip, etc.

[0701] The embodiment of the present application further provides a computer program / program product stored in a storage medium, which is executed by at least one processor to implement the processes of the above method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0702] The embodiment of the present application further provides a wireless communication system, which comprises a terminal and a network side device. The terminal can be used to execute the steps of the terminal side method described above, and the network side device can be used to execute the steps of the network side method described above.

[0703] It should be noted that, in this document, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that processes, methods, articles, or devices that include a series of elements not only include those elements, but also include other elements not explicitly listed, or further include elements inherent to such processes, methods, articles, or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article, or device that includes the element. In addition, it should be pointed out 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, but can also include performing functions in a substantially simultaneous manner or in reverse order, for example, the described method can be performed in an order different from that described, and various steps can also be added, omitted, or combined. In addition, features described with reference to certain examples can be combined in other examples.

[0704] From the above description of the embodiments, those skilled in the art can clearly understand that the above method embodiments can be realized by means of computer software product and general hardware platform, of course, also can be realized by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disc, optical disc, etc.), including a plurality of instructions, used to make the terminal or network side device execute the method described in each embodiment of the present application.

[0705] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above specific embodiments, and the above specific embodiments are only illustrative, not restrictive. Those skilled in the art can make many forms of embodiments under the inspiration of the present application without departing from the scope of the present application and the protection scope of the claims, and these embodiments all belong to the protection scope of the present application.

Claims

1. A method for reporting Channel State Information (CSI), comprising: The terminal receives CSI reporting indication information from the network-side equipment; The terminal acquires the target Rank; The terminal determines the CSI reporting information based on the target Rank and the CSI reporting indication information; Send the CSI reporting information to the network-side device; The CSI reporting information includes CSI feedback information, which is obtained through a target AI model. The target AI model includes one or more AI model units. The CSI reporting indication information is used to indicate the payload of the CSI feedback information. The dataset associated with the AI ​​model unit and at least one of the AI ​​model units are also included.

2. The method according to claim 1, wherein, The CSI reporting indication information includes at least one of the following: The first payload information is used to indicate the payload of the CSI feedback information corresponding to the Rank information; The second payload information is used to indicate the payload of CSI feedback information corresponding to the layer information; The first dataset identifier, dataset ID, is used to indicate the dataset corresponding to the Rank information; The second dataset ID is used to indicate the dataset corresponding to the layer information; The first model identifier, model ID, is used to indicate the AI ​​model unit corresponding to the Rank information; The second model ID is used to indicate the AI ​​model unit corresponding to the layer information.

3. The method according to claim 2, wherein, The layer information includes at least one of layer ID and layer grouping information; The Rank information includes at least one of Rank value and Rank grouping information.

4. The method according to claim 2, wherein, The CSI feedback information includes CSI feedback information corresponding to N layers, and the target AI model includes K AI model units, where K is less than or equal to N.

5. The method according to any one of claims 1-4, wherein, The AI ​​model unit includes any one of the following: The first AI model unit used by the terminal and the network-side device; The reference AI model unit of the second AI model unit used by the terminal and the network-side device The third AI model unit used in the test by the terminal, the network-side device, and the test equipment; The reference AI model unit of the fourth AI model unit used in the test by the terminal, the network-side device, and the test equipment.

6. The method according to claim 5, wherein, The K AI model units satisfy any one of the following: The K AI model units are AI model units corresponding to K layers; The K AI model units are the AI ​​model units corresponding to the target Rank; The K AI model units are AI model units corresponding to a layer group.

7. The method according to any one of claims 1-6, wherein, The CSI reporting information also includes at least one of the following: The target Rank; Channel Quality Information (CQI); At least one layer's corresponding payload; The model ID corresponding to the target AI model; The dataset ID corresponding to the target AI model.

8. The method according to claim 2, wherein, The terminal determines the CSI reporting information based on the target Rank and the CSI reporting indication information, including: The terminal determines the payload of CSI feedback information corresponding to the N layers corresponding to the target Rank based on the target Rank and the first payload information; The terminal determines the CSI reporting information based on the payload of the CSI feedback information corresponding to the N layers; The CSI reporting information is obtained through the target AI model, which includes K AI model units corresponding to the payload of the CSI feedback information of the N layers. The payload of the CSI feedback information corresponding to the N layers satisfies one or more of the following: The sum of the payloads of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information; In the N layers, the payload of the CSI feedback information corresponding to the first layer ID is greater than or equal to the payload of the CSI feedback information corresponding to the second layer ID; In the N layers, the difference between the payload of the CSI feedback information corresponding to the first layer ID and the payload of the CSI feedback information corresponding to the second layer ID is less than or equal to the first threshold. Wherein, the first layer ID is less than the second layer ID.

9. The method according to claim 2, wherein, The terminal determines the CSI reporting information based on the target Rank and the CSI reporting indication information, including: The terminal determines a first set based on the target Rank, the first payload information, and the second dataset ID. The first set contains datasets corresponding to N layers corresponding to the target Rank. The terminal determines the CSI reporting information based on the first set; The CSI reporting information is obtained through the target AI model, which includes K AI model units corresponding to the datasets of the N layers. Wherein, the N layers satisfy: The sum of the payloads of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

10. The method according to claim 2, wherein, The terminal determines the CSI reporting information based on the target Rank and the CSI reporting indication information, including: The terminal determines a second set based on the target Rank, the first payload information, and the second model ID. The second set contains K AI model units corresponding to the N layers corresponding to the target Rank. The terminal determines the CSI reporting information based on the second set; The CSI reporting information is obtained through the target AI model, which includes the K AI model units. Wherein, the N layers satisfy: The sum of the payloads of the CSI feedback information corresponding to the N layers is less than or equal to the payload indicated by the first payload information.

11. The method according to claim 2, wherein, The terminal determines the CSI reporting information based on the target Rank and the CSI reporting indication information, including: The terminal determines a third set based on the target Rank, the second payload information, and the second model ID. The third set contains K AI model units corresponding to the N layers corresponding to the target Rank. The terminal determines the CSI reporting information based on the third set; The CSI reporting information is obtained through the target AI model, which includes K AI model units corresponding to the N layers. Wherein, the N layers satisfy: The payload of the CSI feedback information corresponding to the N layers is the payload indicated by the second payload information.

12. The method according to any one of claims 9 to 11, wherein, The N layers also satisfy: Among the N layers, at least two layers correspond to the same AI model unit; or; K < N.

13. The method according to claim 2, wherein, The method further includes: If the payload of the CSI feedback information is less than the payload indicated by the first payload information, the terminal performs any of the following: The CSI feedback information is populated. The terminal sends a first indication message to the network-side device, the first indication message being used to indicate the payload of the CSI feedback information.

14. The method according to any one of claims 1-13, wherein, The CSI reporting instruction information also includes at least one of the following: Report configuration identifier; Monitoring signage; Relevant information about the target AI model; Time-domain information reported by CSI; The type of the target AI model.

15. The method according to claim 2, wherein, Sending CSI reporting information to the network-side device includes: If the target Rank is greater than the first value, the terminal only sends CSI feedback information to the network-side device corresponding to a Rank less than or equal to the first value; Wherein, the first value is the value of the Rank information, or the value specified by the protocol, or the value configured by the network-side device.

16. The method according to any one of claims 1-15, wherein, Before the terminal receives CSI reporting indication information from the network-side device, the method further includes: The terminal receives first information from the network-side device, the first information including at least one of the following: The first dataset information is used to obtain candidate AI model units; The first model indication information is used to obtain the candidate AI model unit.

17. The method according to claim 16, wherein, The first dataset information includes at least one of the following: Multiple first data, wherein the first data includes at least two of the following: CSI information, CSI feedback information, and reconstructed CSI information; The number of the first data; The dataset ID of the candidate AI model unit; The dataset format or data format of the candidate AI model unit; The second indication information is used to indicate that the first data corresponds to a layer, or that the first data corresponds to a layer group, or that the first data corresponds to a Rank, or that the first data corresponds to both Rank and layer; layer information; Rank information; The number of time domains corresponding to each of the first data.

18. The method according to claim 16, wherein, The first model indication information includes at least one of the following: The model ID of the candidate AI model unit; The third indication information is used to indicate that the candidate AI model unit corresponds to a layer, or that the candidate AI model unit corresponds to a layer group, or that the candidate AI model unit corresponds to a Rank, or that the candidate AI model unit corresponds to both Rank and layer; layer information; Rank information; The model structure indication information of the candidate AI model unit; The model parameter information of the candidate AI model unit; The model structure information of the candidate AI model unit.

19. The method of claim 16, wherein, After the terminal receives the first information from the network-side device, the method further includes: The terminal sends a first response message to the network-side device, the first response message including at least one of the following: The model ID of the candidate AI model unit; The dataset ID of the candidate AI model unit; The dataset format or data format of the candidate AI model unit.

20. The method according to claim 13, wherein, The CSI reporting instruction information also includes: The fourth indication information is used to indicate how the CSI feedback information is filled.

21. A CSI receiving method, comprising: Network-side devices send CSI reporting indication information to the terminal; The network-side device receives CSI reporting information from the terminal; The CSI reporting information includes CSI feedback information, which is obtained through a target AI model. The target AI model includes one or more AI model units. The CSI reporting indication information is used to indicate the payload of the CSI feedback information. The dataset associated with the AI ​​model unit and at least one of the AI ​​model units are also included.

22. The method according to claim 21, wherein, The CSI reporting indication information includes at least one of the following: The first payload information is used to indicate the payload of the CSI feedback information corresponding to the Rank information; The second payload information is used to indicate the payload of CSI feedback information corresponding to the layer information; The first dataset ID is used to indicate the dataset corresponding to the Rank information; The second dataset ID is used to indicate the dataset corresponding to the layer information; The first model ID is used to indicate the AI ​​model unit corresponding to the Rank information; The second model ID is used to indicate the AI ​​model unit corresponding to the layer information.

23. The method according to claim 22, wherein, The layer information includes at least one of layer ID and layer grouping information; The Rank information includes at least one of Rank value and Rank grouping information.

24. The method according to claim 22, wherein, The CSI feedback information includes CSI feedback information corresponding to N layers, and the target AI model includes K AI model units, where K is less than or equal to N.

25. The method according to any one of claims 21-24, wherein, The AI ​​model unit includes any one of the following: The first AI model unit used by the terminal and the network-side device; The reference AI model unit of the second AI model unit used by the terminal and the network-side device The third AI model unit used in the test by the terminal, the network-side device, and the test equipment; The reference AI model unit of the fourth AI model unit used in the test by the terminal, the network-side device, and the test equipment.

26. The method of claim 25, wherein, The K AI model units satisfy any one of the following: The K AI model units are AI model units corresponding to K layers; The K AI model units are the AI ​​model units corresponding to the target Rank; The K AI model units are AI model units corresponding to a layer group.

27. The method according to any one of claims 21-26, wherein, The CSI reporting information also includes at least one of the following: The target Rank; CQI; At least one layer's corresponding payload; The model ID corresponding to the target AI model; The dataset ID corresponding to the target AI model.

28. The method according to any one of claims 21-27, wherein, The method further includes: The network-side device receives first indication information from the terminal, the first indication information being used to indicate the payload of the CSI feedback information.

29. The method according to any one of claims 21-28, wherein, The CSI reporting instruction information also includes at least one of the following: Report configuration identifier; Monitoring signage; Relevant information about the target AI model; Time-domain information reported by CSI; The type of the target AI model.

30. The method according to any one of claims 21-29, wherein, Before the network-side device sends CSI reporting indication information to the terminal, the method further includes: The network-side device sends first information to the terminal, the first information including at least one of the following: The first dataset information is used to obtain candidate AI model units; The first model indication information is used to obtain the candidate AI model unit.

31. The method according to claim 30, wherein, The first dataset information includes at least one of the following: Multiple first data, wherein the first data includes at least two of the following: CSI information, CSI feedback information, and reconstructed CSI information; The number of the first data; The dataset ID of the candidate AI model unit; The dataset format or data format of the candidate AI model unit; The second indication information is used to indicate that the first data corresponds to a layer, or that the first data corresponds to a layer group, or that the first data corresponds to a Rank, or that the first data corresponds to both Rank and layer; The layer information; The Rank information; The number of time domains corresponding to each of the first data.

32. The method according to claim 30, wherein, The first model indication information includes at least one of the following: The model ID of the candidate AI model unit; The third indication information is used to indicate that the first data corresponds to a layer, or that the first data corresponds to a layer group, or that the first data corresponds to a Rank, or that the first data corresponds to both Rank and layer. The first data includes at least two of the following: CSI information, CSI feedback information, and reconstructed CSI information. layer information; Rank information; The model structure indication information of the candidate AI model unit; The model parameter information of the candidate AI model unit; The model structure information of the candidate AI model unit.

33. The method according to claim 30, wherein, After the network-side device sends the first information to the terminal, the method further includes: The network-side device receives first response information from the terminal, the first response information including at least one of the following: The model ID of the candidate AI model unit; The dataset ID of the candidate AI model unit; The dataset format or data format of the candidate AI model unit.

34. The method according to claim 28, wherein, The CSI reporting instruction information also includes: The fourth indication information is used to indicate how the CSI feedback information is filled.

35. A CSI reporting device, comprising: The first receiving module is used for the terminal to receive CSI reporting indication information from the network-side device; The first processing module is used for the terminal to obtain the target Rank; The second processing module is used by the terminal to determine the CSI reporting information based on the target Rank and the CSI reporting indication information; The first sending module is used to send the CSI reporting information to the network-side device; The CSI reporting information includes CSI feedback information, which is obtained through a target AI model. The target AI model includes one or more AI model units. The CSI reporting indication information is used to indicate the payload of the CSI feedback information. The dataset associated with the AI ​​model unit and at least one of the AI ​​model units are also included.

36. The apparatus according to claim 35, wherein, The CSI reporting indication information includes at least one of the following: The first payload information is used to indicate the payload of the CSI feedback information corresponding to the Rank information; The second payload information is used to indicate the payload of CSI feedback information corresponding to the layer information; The first dataset ID is used to indicate the dataset corresponding to the Rank information; The second dataset ID is used to indicate the dataset corresponding to the layer information; The first model ID is used to indicate the AI ​​model unit corresponding to the Rank information; The second model ID is used to indicate the AI ​​model unit corresponding to the layer information.

37. The apparatus according to claim 35, wherein, The device further includes: The second receiving module is configured to, before the terminal receives CSI reporting indication information from the network-side device, receive first information from the network-side device, wherein the first information includes at least one of the following: The first dataset information is used to obtain the AI ​​model unit; The first model indication information is used to obtain the AI ​​model unit.

38. The apparatus according to claim 37, wherein, The device further includes: The second sending module is configured to, after the terminal receives the first information from the network-side device, send the first response information to the network-side device, wherein the first response information includes at least one of the following: The model ID of the AI ​​model unit; The dataset ID of the AI ​​model unit; The dataset format or data format of the AI ​​model unit.

39. A CSI receiving device, comprising: The third sending module is used for network-side devices to send CSI reporting indication information to terminals; The third receiving module is used for the network-side device to receive CSI reporting information from the terminal; The CSI reporting information includes CSI feedback information, which is obtained through a target AI model. The target AI model includes one or more AI model units. The CSI reporting indication information is used to indicate the payload of the CSI feedback information. The dataset associated with the AI ​​model unit and at least one of the AI ​​model units are also included.

40. The apparatus according to claim 39, wherein, The CSI reporting indication information includes at least one of the following: The first payload information is used to indicate the payload of the CSI feedback information corresponding to the Rank information; The second payload information is used to indicate the payload of CSI feedback information corresponding to the layer information; The first dataset ID is used to indicate the dataset corresponding to the Rank information; The second dataset ID is used to indicate the dataset corresponding to the layer information; The first model ID is used to indicate the AI ​​model unit corresponding to the Rank information; The second model ID is used to indicate the AI ​​model unit corresponding to the layer information.

41. The apparatus according to claim 39, wherein, The device further includes: The fourth sending module is configured to send first information to the terminal before the network-side device sends CSI reporting indication information to the terminal, wherein the first information includes at least one of the following: The first dataset information is used to obtain candidate AI model units; The first model indication information is used to obtain the candidate AI model unit.

42. The apparatus according to claim 41, wherein, The device further includes: The fourth receiving module is configured to receive first response information from the terminal after the network-side device sends first information to the terminal, wherein the first response information includes at least one of the following: The model ID of the candidate AI model unit; The dataset ID of the candidate AI model unit; The dataset format or data format of the candidate AI model unit.

43. A terminal comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the CSI reporting method as claimed in any one of claims 1 to 20.

44. A network-side device comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the CSI receiving method as claimed in any one of claims 21 to 34.

45. A readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the CSI reporting method as claimed in any one of claims 1 to 20, or the steps of the CSI receiving method as claimed in any one of claims 21 to 34.

46. ​​A computer program product comprising computer instructions that, when executed by a processor, implement the steps of the CSI reporting method as claimed in any one of claims 1 to 20, or implement the steps of the CSI receiving method as claimed in any one of claims 21 to 34.

Images