WO2023179476A1

WO2023179476A1 - Channel feature information reporting and recovery methods, terminal and network side device

Info

Publication number: WO2023179476A1
Application number: PCT/CN2023/082131
Authority: WO
Inventors: 任千尧
Original assignee: 维沃移动通信有限公司
Priority date: 2022-03-21
Filing date: 2023-03-17
Publication date: 2023-09-28
Also published as: CN116828498A

Abstract

The present application relates to the technical field of communications. Disclosed are channel feature information reporting and recovery methods, a terminal, and a network side device. The channel feature information reporting method in the embodiments of the present application comprises: a terminal acquiring first channel information of a target channel; the terminal determining, from a pre-configured AI network model, a target AI network model corresponding to a first length, the first length being indicated by a network side device or determined by the terminal according to first information, and the first information comprising at least one of the following: the first channel information and an AI network model index indicated by the network side device; the terminal using the target AI network model to process the first channel information to obtain first channel feature information of the first length; and the terminal sending the first channel feature information to the network side device.

Description

Channel characteristic information reporting and recovery method, terminal and network side equipment

Cross-references to related applications

This application claims priority from Chinese Patent Application No. 202210283902.5 filed in China on March 21, 2022, the entire content of which is incorporated herein by reference.

Technical field

This application belongs to the field of communication technology, and specifically relates to a channel characteristic information reporting and recovery method, terminal and network side equipment.

Background technique

With the application of artificial intelligence (AI) in the communication field, AI network models can be used to encode and decode channel state information (CSI) information.

However, in different channel environments, the degree of compressibility of channel information is different, and the length of information after encoding is also different. For example, simple channel information only requires a short encoding length, but complex channel information requires longer encoding information. . The weight parameters and even network structures of the AI network models corresponding to different lengths of coding information are different. In related technologies, when the length of the currently acquired channel information does not match the coding length of the terminal's coding network, the terminal Using this encoding network to encode channel information will result in low accuracy of the encoding results, which will reduce the communication performance between the terminal and the network-side device when communicating based on the encoding results.

Contents of the invention

Embodiments of the present application provide a channel characteristic information reporting and recovery method, a terminal, and a network-side device, so that the terminal can adaptively use an AI network model corresponding to the length of the channel information for encoding, which can improve the accuracy of the encoding results. performance, and then when communicating based on the encoding result, the communication performance between the terminal and the network side device can be improved.

In the first aspect, a method for reporting channel characteristic information is provided, which method includes:

The terminal obtains the first channel information of the target channel;

The terminal determines the target AI network model corresponding to the first length from the preconfigured AI network model, the first length is indicated by the network side device or determined by the terminal according to the first information, wherein the first length The information includes at least one of the following: the first channel information and the AI network model index indicated by the network side device;

The terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The terminal sends the first channel characteristic information to the network side device.

In the second aspect, a device for reporting channel characteristic information is provided, which is applied to a terminal. The device includes:

The first acquisition module is used to acquire the first channel information of the target channel;

The first determination module is configured to determine the target AI network model corresponding to the first length from the preconfigured AI network model, the first length being indicated by the network side device or determined by the terminal according to the first information, wherein, The first information includes at least one of the following: the first channel information and the AI network model index indicated by the network side device;

A first processing module, configured to use the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The first sending module is configured to send the first channel characteristic information to the network side device.

In the third aspect, a channel characteristic information recovery method is provided, including:

The network side device receives the first channel characteristic information from the terminal, where the first channel characteristic information is the channel characteristic information of the first length obtained by processing the first channel information by the terminal using the target AI network model;

The network side device uses a fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the first channel information.

In the fourth aspect, a device for recovering channel characteristic information is provided, which is applied to network side equipment. The device includes:

The first receiving module is configured to receive the first channel characteristic information from the terminal, where the first channel characteristic information is the channel characteristic of the first length obtained by the terminal using the target AI network model to process the first channel information. information;

The second processing module is configured to use a fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the first channel information.

In a fifth aspect, a terminal is provided. The terminal includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, the following implementations are implemented: The steps of the method described in one aspect.

In a sixth aspect, a terminal is provided, including a processor and a communication interface, wherein the communication interface is used to obtain the first channel information of the target channel; the processor is used to determine the relationship with the first channel from a preconfigured AI network model. A target AI network model corresponding to a length, and using the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length, wherein the first length is determined by the network side The device indicates or is determined by the terminal according to the first information. The first information includes at least one of the following: the first channel information, the network side device The AI network model index indicated by the equipment; the communication interface is also used to send the first channel characteristic information to the network side device.

In a seventh aspect, a network side device is provided. The network side device includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. The program or instructions are executed by the processor. When implementing the steps of the method described in the third aspect.

In an eighth aspect, a network side device is provided, including a processor and a communication interface, wherein the communication interface is used to receive first channel characteristic information from a terminal, wherein the first channel characteristic information is the terminal Channel characteristic information of a first length obtained by processing the first channel information using a target AI network model; the processor is configured to process the first channel characteristic information using a fourth AI network model corresponding to the first length. Process to obtain the first channel information.

A ninth aspect provides a communication system, including: a terminal and a network side device. The terminal can be configured to perform the steps of the channel characteristic information reporting method described in the first aspect. The network side device can be configured to perform the steps of the channel characteristic information reporting method as described in the first aspect. The steps of the channel characteristic information recovery method described in the three aspects.

In a tenth aspect, a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method are implemented as described in the first aspect. The steps of the method described in the third aspect.

In an eleventh aspect, a chip is provided. The chip includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the method described in the first aspect. method, or implement a method as described in the third aspect.

In a twelfth aspect, a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement as described in the first aspect The steps of the channel characteristic information reporting method, or the computer program/program product is executed by at least one processor to implement the steps of the channel characteristic information recovery method as described in the third aspect.

In this embodiment of the present application, the terminal can determine the first length according to the instructions of the network side device and/or the first channel information, so as to use the target AI network model capable of outputting the channel characteristic information of the first length to convert the first channel The information is processed into the first channel characteristic information of the first length. In the application, for different channel information or different application environments, etc., an AI network model with a length corresponding to the channel information or application environment can be used to process the channel information. Encoding, so that the length of the output channel characteristic information is the minimum length that can reflect the channel information. In this way, the transmission overhead can be reduced on the basis of meeting the requirements for channel information reporting.

Description of the drawings

Figure 1 is a schematic structural diagram of a wireless communication system to which embodiments of the present application can be applied;

Figure 2 is a flow chart of a method for reporting channel characteristic information provided by an embodiment of the present application;

Figure 3 is a schematic diagram of the architecture of the neural network model;

Figure 4 is a schematic diagram of a neuron;

Figure 5 is a flow chart of a method for recovering channel characteristic information provided by an embodiment of the present application;

Figure 6 is a schematic structural diagram of a device for reporting channel characteristic information provided by an embodiment of the present application;

Figure 7 is a schematic structural diagram of a device for recovering channel characteristic information provided by an embodiment of the present application;

Figure 8 is a schematic structural diagram of a communication device provided by an embodiment of the present application;

Figure 9 is a schematic structural diagram of a terminal provided by an embodiment of the present application;

Figure 10 is a schematic structural diagram of a network side device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art fall within the scope of protection of this application.

The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and "second" are distinguished objects It is usually one type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and/or" in the description and claims indicates at least one of the connected objects, and the character "/" generally indicates that the related objects are in an "or" relationship.

It is worth pointing out that the technology described in the embodiments of this application is not limited to Long Term Evolution (LTE)/LTE Evolution (LTE-Advanced, LTE-A) systems, and can also be used in other wireless communication systems, such as code 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) and other systems. The terms "system" and "network" in the embodiments of this application are often used interchangeably, and the described technology can be used not only for the above-mentioned systems and radio technologies, but also for other systems and radio technologies. The following description describes a New Radio (NR) system for example purposes, and NR terminology is used in much of the following description, but these techniques can also be applied to applications other than NR system applications, such as 6th ^generation Generation, 6G) communication system.

Figure 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable. The wireless communication system includes a terminal 11 and a network side device 12. Among them, the terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), laptop computer (Laptop Computer), also known as notebook computer, personal digital assistant (Personal Digital Assistant, PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile Internet Device (Mobile Internet Device, MID), augmented reality (AR)/virtual reality (VR) equipment, robot, wearable device (Wearable Device), vehicle user equipment (VUE), pedestrian Terminal side (Pedestrian User Equipment, PUE), smart home (home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.), game consoles, personal computers (PC), teller machines or self-service machines, etc. Equipment, wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, Smart clothing, etc. It should be noted that the embodiment of the present application does not limit the specific type of the terminal 11. The network side device 12 may include an access network device or a core network device, where the access network device may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or a wireless access network unit. Access network equipment may include base stations, Wireless Local Area Networks (WLAN) access points or WiFi nodes, etc. The base stations may be called Node B, Evolved Node B (eNB), access point, base transceiver station ( Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), home B-node, home evolved B-node, transmitting and receiving point ( Transmitting Receiving Point (TRP) or some other appropriate terminology in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that in the embodiment of this application, only in the NR system The base station is introduced as an example, and the specific type of base station is not limited.

In wireless communication technology, accurate CSI feedback is crucial to channel capacity. Especially for multi-antenna systems, the transmitter can optimize signal transmission based on CSI to better match the channel status. For example: Channel Quality Indicator (CQI) can be used to select the appropriate modulation and coding scheme (Modulation and Coding Scheme, MCS) to achieve link adaptation; Precoding Matrix Indicator (PMI) can be used To achieve eigen beamforming (eigen beamforming), thereby maximizing the strength of the received signal, or to suppress interference (such as inter-cell interference, interference between multiple users, etc.). Therefore, since multi-antenna technology (such as Multi-Input Multi-Output (MIMO)) was proposed, the acquisition of CSI has always been a research hotspot.

Usually, the network-side device sends CSI-Reference Signals (CSI-RS) on certain time-frequency resources in a certain time slot. The terminal performs channel estimation based on the CSI-RS and calculates the channel on this slot. Information, the PMI is fed back to the base station through the codebook. The network side device combines the channel information based on the codebook information fed back by the terminal, and before the terminal reports the CSI next time, the network side device This channel information is used for data precoding and multi-user scheduling.

In order to further reduce CSI feedback overhead, the terminal can change the PMI reported on each subband to report PMI according to the delay (delay domain, that is, frequency domain). Since the channels in the delay domain are more concentrated, PMI with less delay can be approximated The PMI of all subbands can be regarded as reporting after compressing the delay field information.

Similarly, in order to reduce overhead, the network side device can precode the CSI-RS in advance and send the coded CSI-RS to the terminal. What the terminal sees is the channel corresponding to the coded CSI-RS. The terminal only needs to Just select several stronger ports from the ports indicated by the network-side device and report the coefficients corresponding to these ports.

In related technologies, the use of AI network models to compress channel information can improve the compression effect of channel feature information. Among them, AI network models have many implementation methods, such as: neural networks, decision trees, support vector machines, and Bayesian Classifier etc. For ease of explanation, in the embodiment of the present application, the AI network model is a neural network as an example, but the specific type of the AI network model is not limited.

In the embodiment of the present application, the terminal uses a target AI network model with encoding function (that is, the AI network model in the encoder, which can also be called the encoder network model or the encoding AI network model) to compress and encode the channel information. And report the encoded channel characteristic information to the network side equipment (for example: base station). On the base station side, the fourth AI network model with decoding function (that is, the AI network model in the decoder, which can also be called decoding The decoder network model or decoding AI network model) decodes the compressed channel characteristic information to restore the channel information. Among them, the fourth AI network model of the base station and the target AI network model of the terminal need to be jointly trained to achieve a reasonable matching degree. Specifically, the codec neural network model can be the encoder network model of the terminal and the decoder network model of the base station. The formed joint neural network model is jointly trained by network-side devices. After the training is completed, the base station sends the encoder network model to the terminal.

The terminal estimates the CSI Reference Signal (CSI-RS) or Tracking Reference Signal (TRS), performs calculations based on the estimated channel information, and obtains the calculated channel information; then, the calculated channel information or The original estimated channel information is encoded through the encoding network model to obtain the encoding result, and finally the encoding result is sent to the base station. On the base station side, after receiving the encoded result, the base station can input it into the decoding network model and use the decoding network model to restore the channel information.

However, under different channel environments, the degree of compressibility of channel information is different. Therefore, the length of the channel information after encoding is also different. For example, simple channel information only requires a short encoding length, but complex channel information requires a longer length. encoded information. In this way, the weight parameters and even the network structure of the AI network model corresponding to the encoding information of different lengths are different, which requires retraining the AI network model that matches the encoding length.

It can be seen that in related technologies, channel information of different lengths has different matching degrees with a certain AI network model. That is to say, as the channel quality changes, the matching degree between the AI network model and the channel state will decrease, thus As a result, the accuracy of the encoding and decoding results of the channel feature information by the AI network model is reduced. In addition, in related technologies, after the network side device delivers an AI network model to the terminal, the terminal directly uses the AI network model to encode any channel information and reports a fixed-length encoding result. It is assumed that the network side device is in During the subsequent communication process, if it is determined that the channel information recovered based on the encoding result is not accurate enough, the network side device needs to retrain and issue a new AI network model, and the terminal uses the new AI network model to encode and report the channel again. Until the network side device can obtain accurate channel information. During this process, the network side device may train and deliver the AI network model multiple times, which increases the amount of calculation, occupied resources, and delay caused by training and transmitting the AI network model between the terminal and the network side device.

In the embodiment of this application, the terminal can determine the target AI network model of the specified coding length from the pre-configured AI network model according to the instructions of the network side device or the current channel information, so as to use the target AI network model to process the channel information into the specified length of coding information (i.e., the first channel characteristic information), and reports the first channel characteristic information to the network side device. In this way, the matching degree between the coding length of the target AI network model and the channel status or application environment can be improved, thereby improving the network side The accuracy of the channel information restored by the device based on the channel state information of this coding length.

It should be noted that in the implementation, the above-mentioned terminal reports the first channel characteristic information to the network side device, and may use the CSI reporting method to carry the first channel characteristic information in the CSI report to report to the network side device, where, The channel characteristic information may specifically be PMI information. Of course, the above-mentioned first channel characteristic information can also be reported to the network side device in any other manner. For the convenience of explanation, in the embodiment of this application, the first channel characteristic information is reported using CSI reporting as an example. Herein Does not constitute a specific limitation.

In addition, the first length, the second length and the third length in the embodiment of the present application may be the number of bits of the corresponding channel characteristic information after quantization, or the number of coefficients included in the corresponding channel characteristic information before quantization. number. For ease of explanation, in the embodiment of the present application, the first length, the second length, and the third length are respectively the number of bits, as an example, and no specific limitation is constituted here.

Below in conjunction with the accompanying drawings, the channel characteristic information reporting method, channel characteristic information recovery method, channel characteristic information reporting device, channel characteristic information recovery device and communication equipment provided by the embodiments of the present application will be described in detail through some embodiments and application scenarios. .

Referring to Figure 2, an embodiment of the present application provides a method for reporting channel characteristic information. The execution subject may be a terminal. The terminal may be various types of terminals 11 listed in Figure 1, or other than those shown in Figure 1. Terminals other than the terminal types listed in the embodiment are not specifically limited here. As shown in Figure 2, the channel characteristic information reporting method may include the following steps:

Step 201: The terminal obtains the first channel information of the target channel.

In an implementation, the above-mentioned first channel information may be channel information obtained by the terminal through channel estimation of CSI-RS, TRS or other reference signals corresponding to the target channel, or the terminal may perform certain calculations on the estimated channel information. Or the channel information obtained by preprocessing, which is not specifically limited here.

Step 202: The terminal determines a target AI network model corresponding to a first length from a preconfigured AI network model, and the first length is indicated by the network side device or determined by the terminal based on the first information, wherein the first length is The first information includes at least one of the following: the first channel information and the AI network model index indicated by the network side device.

In implementation, the network side device can pre-configure at least two AI network models for the terminal, and the AI network model can include multiple types of AI algorithm modules, such as: neural networks, decision trees, support vector machines, and Bayesian classification. device, etc., no specific limitation is made here, and for the convenience of explanation, in the following embodiments, the AI algorithm model is a neural network model as an example for illustration, and no specific limitation is constituted here.

As shown in Figure 3, the neural network model includes an input layer, a hidden layer and an output layer, which can predict possible output results (Y) based on the entry and exit information (X ₁ ~ X _n ) obtained by the input layer. The neural network model consists of a large number of neurons, as shown in Figure 4. The parameters of the neurons include: input parameters a ₁ ~ a _K , weight w, bias b and activation function σ(z), and the parameters obtained with these parameters Output value a, among which, common activation functions include Sigmoid function, hyperbolic tangent (tanh) function, linear rectified function (Rectified Linear Unit, ReLU, also called modified linear unit) function, etc. , and z in the above function σ(z) can be calculated by the following formula:
z＝a ₁ w ₁ +…+a _k w _k +a _K w _K +b

Among them, K represents the total number of input parameters.

The parameters of the neural network are optimized through optimization algorithms. An optimization algorithm is a type of algorithm that can help us minimize or maximize an objective function (sometimes also called a loss function). The objective function is often a mathematical combination of model parameters and data. For example, given the data The difference (f(x)-Y) between it and the true value is the loss function. Our purpose is to find appropriate W and b to minimize the value of the above loss function. The smaller the loss value, the closer our model is to the real situation.

Currently, common optimization algorithms are basically based on error back propagation algorithm. The basic idea of the error back propagation algorithm is that the learning process consists of two processes: forward propagation of signals and back propagation of errors. During forward propagation, the input sample is passed in from the input layer, processed layer by layer by each hidden layer, and then transmitted to the output layer. If the actual output of the output layer does not match the expected output, it will enter the error backpropagation stage. Error backpropagation is to propagate the output error back to the input layer in some form through the hidden layer layer by layer, and allocate the error to all units in each layer, thereby obtaining the error signal of the unit in each layer. This error signal is used as the correction The basis for correcting the weight of each unit. This process of adjusting the weights of each layer in forward signal propagation and error back propagation is carried out over and over again. The process of continuous adjustment of weights 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 a preset number of learning times.

Common optimization algorithms include gradient descent (Gradient Descent), stochastic gradient descent (Stochastic Gradient Descent, SGD), mini-batch gradient descent (mini-batch gradient descent), momentum method (Momentum), Nesterov (which represents stochastic gradient with momentum). descent), adaptive gradient descent (Adaptive gradient descent, Adagrad), adaptive learning rate adjustment (Adadelta), root mean square error reduction (root mean square prop, RMSprop), adaptive momentum estimation (Adaptive Moment Estimation, Adam) wait.

When these optimization algorithms perform error backpropagation, they calculate the derivative/partial derivative of the current neuron based on the error/loss obtained by the loss function, plus the influence of the learning rate, previous gradient/derivative/partial derivative, etc., to obtain the gradient. Pass the gradient to the previous layer.

In implementation, the target AI network model can be used to encode channel information, which can encode channel information under various different channel environments into first channel characteristic information of a first length. In implementation, each preconfigured AI network model has its own corresponding length, which can be understood as the encoding length of the corresponding AI network model. That is, after inputting channel information to a certain AI network model, the AI network model can output Corresponding length of channel characteristic information.

In this step, the terminal can determine the first length according to the instruction of the network side device and/or the first information, so as to select the target AI network model from the preconfigured AI network models according to the first length.

Step 203: The terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length.

Step 204: The terminal sends the first channel characteristic information to the network side device.

In an implementation, the terminal may use the target AI network model to encode the first channel information to encode the first channel information into first channel characteristic information of a first length, and send the first channel characteristic information to the network side device, On the network side, a decoding AI network model with an input length of the first length can be used for decoding to restore the first channel characteristic information. In this way, the transmission of the first channel information can be reduced on the basis of realizing the transmission of the first channel information. resource overhead.

It should be noted that the encoding of channel information in the embodiment of the present application is different from the channel coding in related technologies. The encoding process of the channel information in the embodiment of the present application may include the following steps:

Step 1. The terminal detects CSI-RS or TRS at the time-frequency domain location specified by the network, and performs channel estimation to obtain the first channel information;

Step 2: The terminal encodes the first channel information into the first channel characteristic information through the target AI network model (i.e., encoding AI network model);

Step 3: The terminal transmits part or all of the first channel characteristic information and other control information Combine it into uplink control information (UCI), or use part or all of the first channel characteristic information as UCI;

Step 4. The terminal divides the UCI according to the length of the UCI and adds cyclic redundancy check (CRC) bits;

Step 5: The terminal performs channel coding on the UCI with CRC bits added;

Step 6: The terminal performs rate matching on UCI;

Step 7: The terminal performs code block association on UCI;

Step 8: The terminal maps the UCI to the Physical Uplink Control Channel (PUCCH) or the Physical Uplink Shared Channel (PUSCH) for reporting.

It should be noted that in the above-mentioned encoding process of channel information, the order of some steps can be adjusted or omitted, and there is no specific limitation here.

As an optional implementation, before the terminal determines the target AI network model corresponding to the first length from the preconfigured AI network model, the method further includes:

The terminal receives relevant information from the N first AI network models of the network side device, where the preconfigured AI network models include the N first AI network models, and the N first AI network models The network model has a one-to-one correspondence with N lengths, where N is an integer greater than or equal to 1.

In implementation, the relevant information of the above-mentioned first AI network model may be model parameters, model configuration, model identification information, etc., and the terminal can determine which first AI network model is configured by the network side device based on the relevant information. In this way, after the terminal obtains the N first AI network models, the terminal can select one from the obtained N first AI network models as the target AI network model.

Among them, the N first AI network models may have different coding lengths respectively. In addition, at least one of the model parameters such as weight parameters and structures of the N first AI network models may also be different from each other, or the N first AI network models may have different coding lengths. Each AI network model has different coding lengths and weight parameters, but the structures may be the same or partially the same, and are not specifically limited here.

In implementation, the above-mentioned N first AI network models can be trained by network side equipment. For example, N codec AI network models can be obtained by training by network side equipment. Each codec AI network model includes an encoding AI network. model (i.e., the first AI network model) and a decoded AI network model (i.e., the fourth AI network model), then the network side device can send the relevant information of the encoded AI network model to the terminal, so that the terminal can use the received The encoding AI network model encodes the channel information. Specifically, when the terminal accesses the network-side device, the network-side device configures N first AI network models for the terminal, or when the terminal accesses the network-side device, the network-side device configures the N-th AI network model for the terminal. One part of the AI network model, and the other part can be sent to the terminal during subsequent transmission.

For example: the terminal receives the phase information of N first AI network models from the network side device. Relevant information, including:

When the terminal accesses the network side device, it receives relevant information of the N first AI network models; or,

When the terminal accesses the network side device, it receives relevant information of a part of the N first AI network models, and after the terminal accesses the network side device, it receives the N first AI network models. Related information from another part of the first AI network model.

In implementation, the time-frequency domain location of the transmission resources of another part of the relevant information in the above-mentioned N first AI network models can be agreed by the protocol, or configured by the network side device, or by the network side device through instructions Triggered by information and other methods, which are not specifically limited here.

It should be noted that when the terminal accesses the network side device, it receives relevant information of a part of the N first AI network models, and after the terminal accesses the network side device , in the implementation of receiving relevant information of another part of the N first AI network models, it may be that when the terminal obtains the first channel information, the terminal temporarily only configures the N first AI network models. At this time, the terminal may select the target AI network model from a part of the N first AI network models that have been configured. Then, after the terminal receives the new first AI network model, the selection range of the target AI network model determined by the terminal is expanded.

As an optional implementation manner, before the terminal receives relevant information of the N first AI network models from the network side device, the channel characteristic information reporting method further includes:

The terminal sends target capability information to the network side device, where the target capability information is used to assist the network side device in determining the N first AI network models.

In implementation, the above target capability information may be used to indicate at least one of the following:

The identification of the first AI network model supported by the terminal;

The number of switching times of the first AI network model supported by the terminal;

The amount of data of the AI network model that the terminal supports transmission;

The terminal supports the calculated channel status.

Option one, for the identification of the first AI network model supported by the terminal, it can be determined according to the capabilities of the terminal, the identification information or index information of the first AI network model that the terminal can run, or the AI network model type, etc. In this way, the network side device can configure the first AI network model that it supports to the terminal, which can reduce the waste of resources caused by the network side device configuring the terminal with an AI network model that it does not support.

Option two, for the number of switching times of the first AI network model supported by the terminal, can be applied to the process of the terminal changing the selected target AI network model according to changes in the channel environment. The replacement target is limited by the terminal's capabilities. The number of AI network models is limited. For example, assuming that the number of switching times of the first AI network model supported by the terminal is L, the network side device can configure less than L or equal to L first AI network models for the terminal. In this way, the network side device can be configured to the first The number of switching times of the AI network model is less than or equal to the first AI network model it supports, which can reduce the waste of resources caused by the network side device configuring too many first AI network models for the terminal.

Option three, for the amount of data of the AI network model that the terminal supports transmission, it can be: limited by the transmission capability of the terminal, the amount of data of the AI network model that it supports transmission is limited, for example: assuming that the terminal supports the transmission of the AI network model If the data amount is X, then the network-side device can send the first AI network model with a data amount less than In this way, the network side device can configure the first AI network model to the terminal with a data amount less than or equal to the AI network model it supports, which can improve the transmission reliability of the first AI network model.

Option four, for the channel status that the above terminal supports calculation, the terminal can calculate the channel status of the target channel based on the first channel information, and then can determine the coding length of the target AI network model adopted based on the channel status. Correspondingly, The network side device can configure the first AI network model associated with each channel state according to the channel state that the terminal supports calculation. For example: assuming that the terminal supports calculation of whether the target channel is line of sight (Line of Sight, LOS) propagation or non-line of sight. (Non-Line of Sight, NLOS) propagation, the network side device can configure a first AI network model with a shorter coding length associated with LOS propagation and a first AI network model with a longer coding length associated with NLOS propagation. In this way, after the terminal calculates the channel state of the target channel, it can directly determine the first AI network model associated with the channel state as the target AI network model.

As an optional implementation, the terminal determines the target AI network model corresponding to the first length from the preconfigured AI network model, including:

The terminal receives first indication information from the network side device, the first indication information is used to indicate at least one of the second AI network model and the length corresponding to the second AI network model, and the Nth An AI network model includes the second AI network model;

The terminal determines that the target AI network model is the second AI network model indicated in the first indication information, and/or the terminal determines that the first length is the second AI network model indicated in the first indication information. The length corresponding to the second AI network model.

In an implementation, the above-mentioned first indication information may be indication signaling, which carries at least one of the index of the second AI network model and the corresponding length of the second AI network model. In this way, according to the first indication, The information clearly instructs the terminal to use the second AI network model to encode the first channel information.

Of course, the above-mentioned first indication information may also implicitly indicate the second AI network model and/or the corresponding length of the second AI network model, for example: the first channel information and the CSI-Reference signal (CSI-Reference) of the terminal. Signals (CSI-RS), the first indication information corresponds to the CSI resources used by the terminal.

Among them, taking the first indication information used to indicate a second AI network model as an example, the above-mentioned first indication The display information corresponds to the CSI resources used by the terminal, which may include: determining the second AI network model according to the quasi co-location (QCL) relationship of the CSI-RS, for example: adding the QCL relationship of the CSI-RS and The association relationship between the second AI network models, in this way, when the terminal determines the QCL relationship of CSI-RS, it can determine based on the QCL relationship that the associated second AI network model is the AI network model indicated by the network side device. In addition, the above-mentioned first indication information corresponds to the CSI resource used by the terminal, and may also include: when the CSI resource is a periodic CSI-RS resource or an aperiodic triggered CSI-RS resource, the CSI resource may be configured when the CSI resource is configured. , indicating the second AI network model.

In this embodiment, the terminal may use the second AI network model to encode the first channel information according to instructions from the network side device.

In addition to the above terminal determining the target AI network model according to the first instruction of the network side device, the terminal can also autonomously select the target AI network model from preconfigured AI network models.

The terminal determines the target AI network model from the N first AI network models according to at least one of channel characteristics and channel conditions corresponding to the first channel information, and/or, from the N The first length is determined among the lengths.

In this embodiment, the terminal can determine the target AI network model based on at least one of the channel characteristics and channel conditions of the target channel.

In implementation, after the terminal autonomously determines the target AI network model, the terminal may also report the selected target AI network model and/or the first length corresponding to the target AI network model to the network side device. For example: the channel characteristic information reporting method further includes:

The terminal sends second indication information to the network side device, where the second indication information is used to indicate at least one of the target AI network model and the first length.

In this embodiment, when determining at least one of the target AI network model and the first length, the terminal also reports at least one of the target AI network model and the first length to the network side device. item, so that the network side device directly uses the decoding AI network model with a decoding length equal to the first length to decode the first channel characteristic information of the first length obtained by the target AI network model.

The above-mentioned second indication information may be information carried in the CSI report, or information in any uplink signaling sent by the terminal to the network side device. There is no specific limitation here. For the convenience of explanation, the second indication information is Taking the information carried in the CSI report as an example, the second indication information may be carried in the same CSI report as the first channel characteristic information.

Optionally, the second indication information is carried in a fixed-length CSI part in the channel state information CSI report, and the first channel characteristic information is carried in a variable-length CSI part in the CSI report; or,

The second length part of the first channel characteristic information and the second indication information are carried in In the fixed-length CSI part, the part of the first channel characteristic information except the part of the second length is carried in the variable-length CSI part; or,

The second indication information and the first channel characteristic information of the first length are both carried in the variable-length CSI part.

In an implementation, the CSI report may include a fixed-length CSI part (eg, CSI Part1) and a variable-length CSI part (eg, CSI Part2). At this time, the above-mentioned second indication information may be carried in a fixed-length CSI part or a variable-length CSI part in the CSI report.

In addition, the first channel characteristic information of the first length may also be located in the variable-length CSI part, or part of it may be located in the fixed-length CSI part and the other part may be located in the variable-length CSI part.

Specifically, given that there are many possible values for the first length, and the fixed-length CSI part in the CSI report often only accommodates a smaller length, at this time, the first length of the first channel characteristic information can be The second length part is placed in the fixed-length CSI part in the CSI report, and the other part of the first channel characteristic information is placed in the variable-length CSI part in the CSI report. In implementation, the above-mentioned second length may be equal to the minimum value that the first length may take. For example, assuming that the input lengths of the N first AI network models are N ₁ to N _K arranged from short to long, then The second length may be equal to N ₁ . In the implementation, N ₂ to N _K all include the channel characteristic information corresponding to N ₁ , so that the fixed-length CSI part in the CSI report can fixedly carry the 0th to (N ₁ -1)th of the first channel characteristic information. bit content, and when the first length is greater than N ₁ , the content located after the (N ₁ -1)th bit in the first channel characteristic information may be placed in the variable-length CSI part of the CSI report.

Of course, in implementation, the second length may also be the minimum encoding length of all AI network models agreed in the protocol.

It is worth mentioning that the above examples are only examples. In practical applications, the position of the part of the second length in the first channel characteristic information can be agreed by the protocol, for example: the first channel characteristic information The second length portion of may include at least one of the following:

The first X bits in the first channel characteristic information, X is equal to the second length;

X bits in the first channel characteristic information whose importance level is greater than a preset level.

In a possible implementation, the terminal determines the target AI network model from the N first AI network models based on at least one of channel characteristics and channel conditions corresponding to the first channel information. , and/or, determining the first length from the N lengths includes:

The terminal determines that the first length is equal to the length associated with the value of the target channel parameter in the first channel information according to the first association relationship, and/or determines that the target AI network model is the target channel AI network models associated with parameter values, wherein the first association includes each value or each value range of the target channel parameter and the N first AI network models and/or the N lengths the relationship between; or,

The terminal determines that the first length is equal to the target channel parameter according to the second association relationship. The length corresponding to the coding identifier associated with the value, and/or, determining the target AI network model to be the AI network model corresponding to the coding identifier associated with the value of the target channel parameter, wherein the second association includes The association between each value or each value range of the target channel parameter and N coding identifiers, the N coding identifiers corresponding to the N first AI network models, and/or, The N coding identifiers correspond to the N lengths one-to-one.

Taking the second indication information used to indicate the target AI network model as an example, in this embodiment, the terminal can determine the first length associated with the target channel parameter based on the correlation between the value of the channel parameter and the length or coding identifier, or based on The correlation between the value of the channel parameter and the length or coding identifier is used to determine the coding identifier associated with the target channel parameter, and then the AI network model and/or length corresponding to the coding identifier is determined to be the target AI network model and/or the first length.

Wherein, the target channel parameter corresponding to the first channel information may include at least one of the following:

The target channel is line-of-sight propagation or non-line-of-sight propagation;

The number of effective delay paths of the target channel;

The delay interval between the two target paths of the target channel;

The number of effective beams of the target channel. The effective beams include beams corresponding to the orthogonal basis of Discrete Fourier Transform (DFT) whose power is greater than the first threshold.

Option 1. When the target channel is line-of-sight propagation, the channel quality is better than that of non-line-of-sight propagation. When the target channel is line-of-sight propagation, an AI network model with a shorter coding length can be used for the first channel. The information is encoded and the encoding result is reported. When the target channel is non-line-of-sight propagation, it is necessary to use an AI network model with a longer encoding length to encode the first channel information and report the encoding result. For example: assuming N=2, use in LOS The coding length is N ₀ , and the coding length used in NLOS is N ₁ , N ₀ <N ₁ , in particular, N ₁ =L, where N represents the number of values of the first length (that is, the preconfigured first The number of AI network models), L represents the maximum value among the coding lengths of the preconfigured N first AI network models;

Option 2: The greater the number of effective delay paths of the target channel, the longer the reported first channel characteristic information can be. The effective delay paths include at least one of the following: the corresponding power or amplitude is greater than the first channel characteristic information. The delay path of the threshold value, the corresponding delay path of the maximum power or amplitude.

Option 3: The greater the delay distance between the two target paths of the target channel, the longer the reported first channel characteristic information can be. Among them, the two target paths can be any two paths of the target channel, for example: paths corresponding to two maximum values. The time delay spacing of the two target paths can reflect the path included in the target signal in the frequency domain. Concentrated intensity.

Option 4: The greater the number of effective beams of the target channel, the longer the reported first channel characteristic information can be.

In this embodiment, the terminal can determine the first length and/or the target AI network model according to the detected value of the target channel parameter of the target channel, so as to use the target AI network model to convert the first Processing the channel information into the first channel characteristic information of the first length can make the length of the first channel characteristic information reported by the terminal match the channel state of the target channel.

In another optional implementation, the terminal can also receive relevant information of the decoded AI network model corresponding to the first AI network model, and the terminal can simulate and obtain the network-side device's response to the first AI based on the decoded AI network model. The decoding result of the encoding result of the network model is compared, and the decoding result of the decoded AI network model is compared with the channel information before encoding by the first AI network model to obtain the matching degree of the two, wherein the higher the matching degree of the two. , it means that the decoding AI network model can more accurately restore the channel characteristic information of the coding length corresponding to the first AI network model.

In this embodiment, in addition to sending relevant information of the N first AI network models to the terminal, the network side device may also send both the encoded AI network model and the corresponding decoded AI network model to the terminal.

Optionally, the terminal can receive the above-mentioned N fifth AI network models and the corresponding N first AI network models together, that is, the network side device sends the jointly trained codec AI network model as a whole to the terminal; or, Independently receive the above-mentioned N fifth AI network models and N first AI network models, that is, the network side device splits the codec AI network model obtained by joint training into the first AI network model and the fifth AI network model, and Use mutually independent transmission processes to send the first AI network model and the fifth AI network model; or, independently receive M sixth AI network models, that is, the network side device also sends a common decoding AI network model to the terminal.

Of course, in addition to receiving the N fifth AI network models corresponding to the above N first AI network models, the terminal may also only receive a simplified AI network model that decodes the AI network model, or may receive M sixth AI networks. model, where the sixth AI network model can be understood as a public decoding AI network model, that is, a public decoding AI network model can be used to simulate at least two decoding AI network models, and M is a positive integer less than or equal to N.

As an optional implementation, the channel characteristic information reporting method further includes:

The terminal receives relevant information of K third AI network models from the network side device, wherein the third AI network model is related to a fourth AI network model, and the fourth AI network model is the network The decoding network model of the side device, or the third AI network model is a public decoding network model, and K third AI network models correspond to N first AI network models, and K is greater than or equal to 1 integer;

The terminal determines the target AI network model from the N first AI network models according to the channel status of the target channel, including:

The terminal processes the first channel characteristic information obtained by processing the target first AI network model into second channel information through the target third AI network model, where the target first AI network model is the N first AI Any one of the network models, the K third AI network models include the target third AI network model, and the target third AI network model is the same as the target first AI network Model correspondence;

The terminal obtains the degree of matching between the second channel information corresponding to the first channel characteristic information processed by the N first AI network models and the first channel information respectively;

When the terminal determines that the matching degree between the target second channel information and the first channel information satisfies the preset conditions, the terminal determines that the first AI network model of the processed target first channel characteristic information is the target AI network. A model, wherein the target second channel information corresponds to the target first channel characteristic information.

In implementation, the above K third AI network models may include a decoding AI network model adopted by the network side device, a simplified model of the decoding AI network model, and a public decoding AI network used to simulate the decoding AI network model adopted by the network side device. Model.

For example: the K third AI network models include at least one of the following:

N fifth AI network models corresponding one-to-one to the N first AI network models, where the fifth AI network model is related to the fourth AI network model corresponding to the same first AI network model;

M sixth AI network models corresponding to the N first AI network models, each of the sixth AI network models corresponding to at least one first AI network model, and the sixth AI network model is used for simulation Corresponding to the fourth AI network model of the same first AI network model, M is a positive integer less than or equal to N.

In a possible implementation, the fifth AI network model is related to the fourth AI network model corresponding to the same first AI network model, which may be: the fifth AI network model is a model of the first AI network model. A decoding AI network model or a simplified model of the decoding AI network model. In this way, the terminal can use the fifth AI network model to decode the first channel characteristic information to simulate the recovery result of the first channel characteristic information by the network side device using the fourth AI network model.

In another possible implementation, each of the above-mentioned sixth AI network models corresponds to at least one first AI network model, which may be: the sixth AI network model is a public decoding AI network model, and the public decoding AI network model The model can simulate the recovery results of at least one fourth AI network model for respective corresponding first channel characteristic information. For example: assuming N is equal to 5, the terminal can receive a public decoding AI network model, and use the public decoding AI network model to respectively The encoding results of the five first AI network models are decoded to simulate the decoding process of the five fourth AI network models that correspond one-to-one to the five first AI network models.

It should be noted that, after the terminal obtains the restoration results (i.e., second channel information) of the first channel characteristic information of the K third AI network models, the terminal can restore the K third AI network models to their respective first channel characteristic information. The second channel information is compared with the first channel information respectively to determine the degree of matching between the K pieces of second channel information and the corresponding first channel characteristic information. The higher the matching degree, the better the second channel information is. The higher the accuracy. Then, according to the matching degree, the first AI network model corresponding to the third AI network model that meets the communication quality requirements is selected. Target AI network model.

In the case where the K third AI network models include N fifth AI network models, the N fifth AI network models are respectively decoding AI network models corresponding to the N first AI network models one-to-one. By using the The fifth AI network model decodes the encoding result of the corresponding first AI network model, and compares the decoding result with the channel information input by the first AI network model, so that the accuracy of the decoding result of the fifth AI network model can be obtained. The input length of the fifth AI network model is the same as the output length of the first AI network model corresponding to the fifth AI network model.

When the terminal determines that the matching degree between the target second channel information and the first channel information satisfies the preset conditions, the terminal determines that the first AI network model of the processed target first channel characteristic information is the target AI network. The model may be: traversing the decoding results of the K third AI network models and the input information of the respective corresponding encoding AI network models to determine the encoding AI network model that satisfies the preset conditions and has the smallest encoding length. The preset condition may be a matching degree threshold value determined based on communication quality requirements, business requirements, etc., or a matching degree threshold value agreed upon in the protocol. For example: the decoding AI network model corresponding to the known encoding AI network model of the terminal, or the simplified decoding AI network model, or the public decoding AI network model. The terminal uses the encoded result to obtain the hypothetical network side through its own known decoder. The device then compares the channel information recovered by the device with the first channel information calculated based on its own estimated original channel information. If the difference between the two is greater than a certain threshold, it is considered that the two do not match. Therefore, it is necessary to For a longer coding length, if the difference between the two is less than the threshold, the first AI network model with a smaller coding length can be traversed to finally find the first AI network model with the minimum coding length that satisfies the threshold.

Optionally, the degree of matching between the target second channel information and the first channel information satisfies a preset condition including at least one of the following:

The correlation between the target second channel information and the first channel information is greater than or equal to a preset correlation;

The channel capacity of the target second channel information is greater than or equal to a first preset value times the channel capacity of the first channel information, and the first preset value is greater than 0 and less than or equal to 1;

The target second channel information is the one in which a channel quality indicator (Channel quality indicator, CQI) among the K pieces of second channel information is the same as or closest to the CQI of the first channel information;

The target second channel information is one of the K second channel information whose modulation and coding scheme (Modulation and coding scheme, MCS) is the same as or closest to the MCS of the first channel information;

The target second channel information is the one with the shortest length among the K pieces of second channel information.

In implementation, the correlation between the above-mentioned target second channel information and the first channel information may be the similarity of the information content of the target second channel information and the first channel information, for example: target Mutual information between the second channel information and the first channel information.

In addition, if the channel characteristic information output by different first AI network models has different lengths, the channel information decoded based on the channel characteristic information of different lengths may contain different channel capacity, CQI, MCS, etc. In this case, it may be preferred to Select the second channel information whose channel capacity is greater than the channel capacity of the first channel information as the target second channel information, thereby determining the first AI network model that can encode the target first channel characteristic information corresponding to the target second channel information as the target AI network model, wherein the target first channel feature information corresponding to the target second channel information may be: the target second channel information is obtained by using a decoding AI network model corresponding to the length of the target first channel feature information to decode the target first channel information. The decoding result is obtained after decoding the channel characteristic information. In addition, the above-mentioned first preset value may be a value indicated by the network side device or agreed upon by the protocol. When the channel capacity meets the first preset value, the first channel characteristic information of the first length can be decoded to meet the business requirements. and/or channel information for channel quality requirements. Of course, in implementation, the second channel information that is closest to the CQI and/or MCS in the first channel information may also be used as the target second channel information.

As an optional implementation manner, the terminal receives first indication information from the network side device, and the first indication information is used to indicate the second AI network model and the length corresponding to the second AI network model. In the case of at least one of the above, the terminal determines the target AI network model based on the first information including:

The terminal determines a seventh AI network model that matches the channel state of the target channel and has the smallest corresponding length according to at least one of the channel characteristics and channel conditions corresponding to the first channel information. The N first AI The network model includes the seventh AI network model, and the first channel information is the channel information of the target channel;

When the length corresponding to the second AI network model is greater than the length corresponding to the seventh AI network model, the terminal determines that the target AI network model is the seventh AI network model.

In this embodiment, when the network side device instructs the terminal to use the second AI network model to encode channel information, if the terminal finds a seventh AI network model that meets the channel quality requirements, and the coding length of the seventh AI network model is less than The coding length of the second AI network model, the terminal can use the seventh AI network model with a shorter coding length to encode the channel information, that is, the terminal can use the coded AI network model that has obtained a length smaller than that indicated by the network side device, so that Simplify the coding process.

Optionally, the terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length, including:

The terminal uses the seventh AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The channel characteristic information reporting method also includes:

The terminal sends second indication information to the network side device, and the second indication information is used to Indicate at least one of the seventh AI network model and the first length;

or,

The terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length, including:

The terminal uses the seventh AI network model to process the first channel information to obtain channel characteristic information of a third length, and the terminal supplements the channel characteristic information of the third length to the first length to obtain the channel characteristic information of the third length. The first channel characteristic information, wherein the first length is the length corresponding to the second AI network model.

In a possible implementation, the terminal can directly use the seventh AI network model selected by itself to process the first channel information and obtain the first channel characteristic information of the first length. At this time, the first length is the seventh AI network model. The coding length of the AI network model. At this time, the coding length of the first channel characteristic information reported by the terminal is inconsistent with the coding length indicated by the network side device. The terminal can report the actual coding length of the first channel characteristic information or the actual coding length to the network side. The coding network corresponding to the coding length enables the network side device to use the decoding AI network model corresponding to the actual coding length to perform channel recovery on the first channel characteristic information reported by the terminal according to the instruction.

For example: assuming that the base station instructs the terminal to use a coded AI network model with a coding length of 200 bits, and the terminal independently chooses to use a coded AI network model with a coding length of 100 bits to code the first channel information, and obtains 100-bit first channel characteristic information, then The terminal sends the 100-bit first channel characteristic information to the network-side device, and tells the network-side device that the encoding length of the first channel characteristic information is 100 bits.

In another possible implementation, the terminal can use the seventh AI network model selected by itself to process the first channel information and obtain the first channel characteristic information of the third length, and then also use the third length of the first channel characteristic information. The first channel characteristic information is supplemented to the first length by any method such as replacement, and the first channel characteristic information of the first length is reported. At this time, the encoding length of the first channel characteristic information reported by the terminal is consistent with the network side device indication. The coding length is consistent, the terminal may not report the actual coding length of the first channel characteristic information or the coding network corresponding to the actual coding length to the network side, and the network side device may use the same coding length as indicated in the previously sent first indication information. The decoding AI network model corresponding to the second AI network model is used to perform channel recovery on the first channel characteristic information reported by the terminal.

For example: Suppose the base station instructs the terminal to use a coded AI network model with a coding length of 200 bits, and the terminal independently chooses to use a coded AI network model with a coding length of 100 bits to code the first channel information. After obtaining the 100-bit first channel characteristic information, The 100-bit first channel characteristic information is supplemented to 200 bits by zero-filling, and then the terminal sends the 200-bit first channel characteristic information to the network side device.

In the embodiment of the present application, the terminal can determine the first length according to the instructions of the network side device and/or the first channel information, so as to use the target AI network that can output the channel characteristic information of the first length. model to process the first channel information into first channel characteristic information of a first length. In applications, for different channel information or different application environments, an AI network with a length corresponding to the channel information or application environment can be used. The model is used to encode the channel information, so that the length of the output channel characteristic information is the minimum length that can reflect the channel information. In this way, the transmission overhead can be reduced on the basis of meeting the requirements for channel information reporting.

Referring to Figure 5, an embodiment of the present application provides a channel characteristic information recovery method. The execution subject may be a network side device. The terminal may be various types of network side devices 12 listed in Figure 1, or other than Network-side devices other than the network-side device types listed in the embodiment shown in FIG. 1 are not specifically limited here. As shown in Figure 5, the channel characteristic information recovery method may include the following steps:

Step 501: The network side device receives the first channel characteristic information from the terminal, where the first channel characteristic information is the channel characteristic information of the first length obtained by the terminal using the target AI network model to process the first channel information. .

Step 502: The network side device uses the fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the first channel information.

In implementation, the above-mentioned first channel characteristic information and first channel information have the same meaning respectively as the first channel characteristic information and the first channel information in the method embodiment as shown in Figure 2. The above-mentioned target AI network model may be coded AI network model, the first length is the length of the encoding result output by the encoding AI network model, and the fourth AI network model can be a decoding AI network model corresponding to the encoding AI network model, and the encoding of the decoding AI network model is input The length of the information is equal to the length of the encoding result output by the encoding AI network model, which will not be described again here.

Optionally, before the network side device receives the first channel characteristic information from the terminal, the channel characteristic information recovery method further includes:

The network side device sends relevant information of N first AI network models to the terminal, where the N first AI network models correspond to N lengths one-to-one, and the N first AI network models include For the target AI network model, the N lengths include the first length, and N is an integer greater than or equal to 1.

Optionally, the network side device sends relevant information of N first AI network models to the terminal, including:

When the terminal accesses the network side device, the network side device sends relevant information of the N first AI network models to the terminal; or,

When the terminal accesses the network side device, the network side device sends relevant information of a part of the N first AI network models to the terminal, and when the terminal accesses the network After the network side device is connected to the network side device, the network side device sends relevant information of another part of the N first AI network models to the terminal.

The network side device sends first indication information to the terminal, where the first indication information is used to indicate at least one of a second AI network model and a length corresponding to the second AI network model.

Optionally, before the network side device sends the first indication information to the terminal, the channel characteristic information recovery method further includes:

The network side device receives target capability information from the terminal, where the target capability information is used to assist the network side device in determining the N first AI network models.

Optionally, the target capability information is used to indicate at least one of the following:

The identification of the first AI network model supported by the terminal;

The terminal supports the calculated channel status.

Optionally, the first channel information is related to the channel estimation result of the channel state information reference signal CSI-RS by the terminal, and the first indication information corresponds to the CSI resources used by the terminal.

The network side device receives second indication information from the terminal, where the second indication information is used to indicate at least one of the target AI network model and the first length.

It should be noted that in implementation, the network side device may first send the first indication information to the terminal, and then receive the second indication information from the terminal, where the second indication information may indicate the same or different AI as the first indication information. network model.

The part of the second length in the first channel characteristic information and the second indication information are carried in the fixed length CSI part, and the part in the first channel characteristic information except the part of the second length is carried in the variable length CSI part; or,

Optionally, the second length is equal to the minimum length among the N lengths.

Optionally, before the network side device receives the second indication information from the terminal, the channel characteristic information recovery method further includes:

The network side device sends relevant information of K third AI network models to the terminal, where the third AI network model is related to the fourth AI network model, or the third AI The network model is a common decoding network model, and K third AI network models correspond to N first AI network models, and K is an integer greater than or equal to 1.

Optionally, the K third AI network models include at least one of the following:

In the embodiment of this application, the network-side device can indicate to the terminal the target AI network model that meets the minimum length of the channel status according to the scheduling situation and/or communication quality requirements, or the terminal determines the target AI network model based on the channel information and reports it The identification and/or first length of the target AI network model. Then, when receiving the first channel characteristic information reported by the terminal, the network side device can use the fourth AI network model corresponding to the target AI network model to calculate the third Using a length of first channel characteristic information for channel recovery can reduce transmission overhead on the basis of meeting the requirements for channel information reporting.

In order to facilitate understanding of the channel characteristic information reporting method and channel characteristic information recovery method provided by the embodiments of this application, the following application scenarios are taken as an example to illustrate the channel characteristic information reporting method and channel characteristic information recovery method provided by the embodiments of this application:

Assuming that there are a total of 6 encoding and decoding AI network models agreed in the protocol, the corresponding encoding lengths are: 16, 32, 64, 128, 256 and 512, and the corresponding encoding identifiers are: 0-5, then the embodiment of this application provides The channel characteristic information reporting method and channel characteristic information recovery method may include the following processes:

1. Assume that the base station has 6 already trained encoding and decoding AI network models. Among them, the encoding and decoding AI network models can be trained on the core network or the base station;

2. When the terminal accesses the cell, the base station can deliver the codec AI network model to the terminal. For example: the base station first delivers two codec AI network models with coding identifiers 0 and 1, then issues two codec AI network models with coding identifiers 2 and 3 after 50ms, and issues 4 and 5 after 200ms. Two codec AI network models, where the specific delivery time of the codec AI network model can be notified to the terminal when the terminal accesses the cell or when the codec AI network model is delivered for the first time, or when the codec AI network model is delivered for the first time. Before sending the codec AI network model, the terminal is notified of the time-frequency domain location of the resource for transmitting the codec AI network model through a signaling trigger.

3. The terminal detects CSI-RS, performs channel estimation, and selects an appropriate coding AI network model based on the channel estimation results. For example: If the terminal receives the encoding AI network model and the decoding AI network model with encoding identifiers 0 and 1 at this time, the terminal inputs the first channel information into the encoding AI network model with encoding identifier 1, and then passes the encoding identifier corresponding to 1. Decode the AI network model to obtain the restored second channel information. The terminal calculates the correlation between the first channel information and the second channel information. If If the correlation is greater than the threshold A, it is considered that the coding AI network model with coding ID 1 meets the conditions, and then the coding AI network model with coding ID 0 is tried. If the coding AI network model with coding ID 0 also meets the threshold, then the coding ID is used The coding AI network model with coding ID of 0 is used to encode the channel information, because the coding length of the coding AI network model with coding ID of 0 is shorter; if the coding AI network model with coding ID of 0 is not satisfied, the coding with coding ID of 1 is used. The AI network model encodes channel information.

It should be noted that if the subsequent terminal receives more encoding and decoding AI network models, the terminal can start traversing from the encoding and decoding AI network model with the longest encoding length, and at least find the smallest encoding length that satisfies the correlation degree greater than the threshold A. Coding AI network model.

In addition, in the implementation, there are cases where the terminal only receives the encoded AI network model sent by the network side device but does not receive the decoded AI network model. In this case, the terminal can calculate that the first channel information reflected is greater than a certain threshold B The number of paths, and the target AI network model is determined according to the first association between the number of paths greater than a certain threshold B configured by the protocol or the network side device and the corresponding coded AI network model, where, The more paths a channel information reflects that are greater than a certain threshold B, the longer the required coding length is. In addition, if the number of coded AI network models corresponding to the number of paths greater than a certain threshold B determined by the terminal according to the first association relationship is at least two, the terminal can select the default one among them (for example: select one of them by default The one with the shortest or longest coding length), or further select the target AI network model from the at least two coding AI network models based on other parameters reflected by the first channel information, for example: based on the channel characteristics reflected by the first channel information The first correlation between the number of values greater than the threshold C and the coding AI network model or the second correlation between the coding identification is used to determine the target AI network model, where the channel characteristic value can be a power value or an amplitude value, For example: the second association relationship is used to indicate: when the number of valid delay paths is 1 and the characteristic value is 1, the coding identifier of the associated coding AI network model is 0; the second association relationship is also used to indicate: when it is valid When the number of extension paths is 2 and the eigenvalue is 1, the coding identifier of the associated coding AI network model is 1.

4. The terminal obtains the corresponding encoding result (i.e., the first channel characteristic information) based on the determined target AI network model, and converts the identification of the target AI network model or the length, rank identification (Rank Index, RI), and CQI of the target AI network model into Waiting for mapping to CSI Part1, map the first channel characteristic information to CSI Part2 to feed back to the base station through CSI report.

In summary, in the embodiments of this application, the terminal can adaptively select the encoding AI network model according to the channel conditions, or the base station instructs the terminal to encode the AI network model according to the scheduling and channel conditions, so that the terminal uses the minimum CSI bit length that can reflect the channel information. The coding AI network model reduces transmission overhead while ensuring the channel information feedback effect.

For the channel characteristic information reporting method provided by the embodiments of the present application, the execution subject may be a channel characteristic information reporting device. In the embodiment of the present application, the channel characteristic information reporting device is used to execute the channel characteristic information The reporting method is taken as an example to describe the channel characteristic information reporting device provided by the embodiment of the present application.

Please refer to Figure 6. A device for reporting channel characteristic information provided by an embodiment of the present application may be a device within a terminal. As shown in Figure 6, the device 600 for reporting channel characteristic information may include the following modules:

The first acquisition module 601 is used to acquire the first channel information of the target channel;

The first determination module 602 is configured to determine the target AI network model corresponding to the first length from the preconfigured AI network model, the first length being indicated by the network side device or determined by the terminal according to the first information, wherein , the first information includes at least one of the following: the first channel information, the AI network model index indicated by the network side device;

The first processing module 603 is configured to use the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The first sending module 604 is configured to send the first channel characteristic information to the network side device.

Optionally, the channel characteristic information reporting device 600 also includes:

The second receiving module is configured to receive relevant information from the N first AI network models of the network side device, where the preconfigured AI network models include the N first AI network models, and the N Each first AI network model has a one-to-one correspondence with N lengths, where N is an integer greater than or equal to 1.

Optionally, the second receiving module is specifically used for:

When the terminal accesses the network side device, receive relevant information of the N first AI network models; or,

When the terminal accesses the network side device, relevant information of a part of the N first AI network models is received, and after the terminal accesses the network side device, it receives the N Related information from another part of the first AI network model.

The second sending module is configured to send target capability information to the network side device, where the target capability information is used to assist the network side device in determining the N first AI network models.

The identification of the first AI network model supported by the terminal;

The terminal supports the calculated channel status.

Optional, the first determination module 602 includes:

A receiving unit configured to receive first indication information from a network side device, where the first indication information is used to indicate at least one of the second AI network model and the length corresponding to the second AI network model, and the N A first AI network model includes the second AI network model;

A first determining unit configured to determine that the target AI network model is the second AI network model indicated in the first indication information, and/or determine that the first length is the second AI network model indicated in the first indication information. Indicates the corresponding length of the second AI network model.

Optional, the first determination module 602 is specifically used for:

Determine the target AI network model from the N first AI network models according to at least one of channel characteristics and channel conditions corresponding to the first channel information, and/or, from the N lengths Determine the first length.

A third sending module, configured to send second indication information to the network side device, where the second indication information is used to indicate at least one of the target AI network model and the first length.

Optional, the first determination module 602 is specifically used for:

According to the first association relationship, it is determined that the first length is equal to the length associated with the value of the target channel parameter in the first channel information, and/or the target AI network model is determined to be the value of the target channel parameter. Associated AI network models, wherein the first association relationship includes the relationship between each value or each value range of the target channel parameter and the N first AI network models and/or the N lengths. relationship; or,

According to the second association relationship, it is determined that the first length is equal to the length corresponding to the coding identifier associated with the value of the target channel parameter, and/or it is determined that the target AI network model is associated with the value of the target channel parameter. The AI network model corresponding to the encoding identifier, wherein the second association relationship includes an association relationship between each value or each value range of the target channel parameter and N encoding identifiers, and the N encoding identifiers and The N first AI network models have a one-to-one correspondence, and/or the N coding identifiers have a one-to-one correspondence with the N lengths.

Optionally, the target channel parameter corresponding to the first channel information includes at least one of the following:

The number of effective delay paths of the target channel;

The delay interval between the two target paths of the target channel;

The number of effective beams of the target channel, the effective beams include beams corresponding to the discrete Fourier transform DFT orthogonal basis with power greater than the first threshold.

A third receiving module, configured to receive relevant information of K third AI network models from the network side device, where the third AI network model is related to a fourth AI network model, and the fourth AI network model is the decoding network model of the network side device, or the third AI network model is a public decoding network model, and K third AI network models correspond to N first AI network models, and K is greater than or an integer equal to 1;

The first determination module 602 includes:

A processing unit configured to process the first channel characteristic information obtained by processing the target first AI network model into second channel information through the target third AI network model, wherein the target first AI network model is the Nth Any one of an AI network model, the K third AI network models include the target third AI network model, and the target third AI network model corresponds to the target first AI network model;

An acquisition unit, configured to acquire the degree of matching between the second channel information corresponding to the first channel characteristic information processed by the N first AI network models and the first channel information respectively;

The second determination unit is configured to determine that the first AI network model of the processed target first channel characteristic information is the first AI network model when it is determined that the matching degree between the target second channel information and the first channel information satisfies the preset conditions. The target AI network model, wherein the target second channel information corresponds to the target first channel characteristic information.

The target second channel information is the one in which the channel quality indicator CQI of the K second channel information is the same as or closest to the CQI of the first channel information;

The target second channel information is one of the K second channel information whose modulation and coding scheme MCS is the same as or closest to the MCS of the first channel information;

Optionally, the terminal receives first indication information from the network side device, the first indication information is used to indicate at least one of the second AI network model and the length corresponding to the second AI network model. In this case, the first determination module 602 includes:

A third determination unit configured to determine, based on at least one of the channel characteristics and channel conditions corresponding to the first channel information, the seventh AI network model that matches the channel state of the target channel and has the smallest corresponding length, the N A first AI network model includes the seventh AI network model, and the first channel information is the channel information of the target channel;

A fourth determination unit configured to determine that the target AI network model is the seventh AI network model when the length corresponding to the second AI network model is greater than the length corresponding to the seventh AI network model.

Optionally, the first processing module 603 is specifically used for:

Using the seventh AI network model to process the first channel information, obtain the first channel characteristic information of the first length;

The channel characteristic information reporting device 600 also includes:

A third sending module, configured to send second indication information to the network side device, where the second indication information is used to indicate at least one of the seventh AI network model and the first length;

or,

The first processing module 603 is specifically used for:

Use the seventh AI network model to process the first channel information to obtain channel characteristic information of a third length, and supplement the channel characteristic information of the third length to the first length to obtain the first channel characteristic information. , wherein the first length is the length corresponding to the second AI network model.

The channel characteristic information reporting device 600 in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip. The electronic device may be a terminal or other devices other than the terminal. For example, terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.

The channel characteristic information reporting device 600 provided by the embodiment of this application can implement each process implemented by the method embodiment shown in Figure 2 and achieve the same technical effect. To avoid duplication, the details will not be described here.

For the channel characteristic information recovery method provided by the embodiments of the present application, the execution subject may be a channel characteristic information recovery device. In the embodiments of the present application, the channel characteristic information restoration method performed by the channel characteristic information restoration apparatus is used as an example to illustrate the channel characteristic information restoration apparatus provided by the embodiments of the present application.

Please refer to Figure 7. A device for recovering channel characteristic information provided by an embodiment of the present application can be a device in a network-side device. As shown in Figure 7, the device for restoring channel characteristic information 700 can include the following modules:

The first receiving module 701 is used to receive the first channel characteristic information from the terminal, where the first channel characteristic information is a channel of the first length obtained by the terminal using the target AI network model to process the first channel information. feature information;

The second processing module 702 is configured to use the fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the first channel information.

Optionally, the channel characteristic information recovery device 700 also includes:

The fourth sending module is used to send relevant information of N first AI network models to the terminal, wherein the N first AI network models correspond to N lengths one-to-one, and the N first AI network models The model includes the target AI network model, the N lengths include the first length, and N is an integer greater than or equal to 1.

Optionally, the fourth sending module is specifically used for:

When the terminal accesses the network side device, send relevant information of the N first AI network models to the terminal; or,

When the terminal accesses the network side device, relevant information of a part of the N first AI network models is sent to the terminal, and after the terminal accesses the network side device, The terminal sends related information of another part of the N first AI network models.

The fifth sending module is configured to send first indication information to the terminal, where the first indication information is used to indicate at least one of the second AI network model and the length corresponding to the second AI network model.

The fourth receiving module is configured to receive target capability information from the terminal, where the target capability information is used to assist the network side device in determining the N first AI network models.

The identification of the first AI network model supported by the terminal;

The terminal supports the calculated channel status.

A sixth receiving module, configured to receive second indication information from the terminal, where the second indication information is used to indicate at least one of the target AI network model and the first length.

A sixth sending module, configured to send relevant information of K third AI network models to the terminal, where the third AI network model is related to the fourth AI network model, or the third AI network model is a common decoding network model, and K third AI network models correspond to N first AI network models, and K is an integer greater than or equal to 1.

The channel characteristic information recovery device 700 in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip. The electronic device may be a network-side device, or may be other devices besides the network-side device. For example, the terminal may include but is not limited to the types of network side devices 12 listed above. Other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.

The channel characteristic information recovery device 700 provided by the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 5 and achieve the same technical effect. To avoid duplication, the details will not be described here.

Optionally, as shown in Figure 8, this embodiment of the present application also provides a communication device 800, which includes a processor 801 and a memory 802. The memory 802 stores information that can run on the processor 801. Programs or instructions, for example, when the communication device 800 is a terminal, when the program or instructions are executed by the processor 801, each step of the above channel characteristic information reporting method embodiment is implemented, and the same technical effect can be achieved. When the communication device 800 is a network-side device, when the program or instruction is executed by the processor 801, the steps of the above channel characteristic information recovery method embodiment are implemented, and the same technical effect can be achieved. To avoid duplication, they will not be described again here.

Embodiments of the present application also provide a terminal, including a processor and a communication interface. The communication interface is used to obtain the first channel information of the target channel; the processor is used to determine the target AI corresponding to the first length from a preconfigured AI network model. network model, and use the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length, wherein the first length is indicated by the network side device or by the The terminal determines based on the first information, which includes at least one of the following: the first channel information and the AI network model index indicated by the network side device; the communication interface is also used to send a message to the network side device. Send the first channel characteristic information.

This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect. Specifically, FIG. 9 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.

The terminal 900 includes but is not limited to: a radio frequency unit 901, a network module 902, an audio output unit 903, an input unit 904, a sensor 905, a display unit 906, a user input unit 907, an interface unit 908, a memory 909, a processor 910, etc. At least some parts.

Those skilled in the art can understand that the terminal 900 may also include a power supply (such as a battery) that supplies power to various components. The power supply may be logically connected to the processor 910 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions. The terminal structure shown in FIG. 9 does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown in the figure, or may combine certain components, or arrange different components, which will not be described again here.

It should be understood that in the embodiment of the present application, the input unit 904 may include a graphics processing unit (Graphics Processing Unit, GPU) 9041 and a microphone 9042. The graphics processor 9041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras). The display unit 906 may include a display panel 9061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 907 includes a touch panel 9071 and at least one of other input devices 9072 . Touch panel 9071, also known as touch screen. The touch panel 9071 may include two parts: a touch detection device and a touch controller. Other input devices 9072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.

In this embodiment of the present application, after receiving downlink data from the network side device, the radio frequency unit 901 can transmit it to the processor 910 for processing; in addition, the radio frequency unit 901 can send data to the network side device. Upstream data. Generally, the radio frequency unit 901 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.

Memory 909 may be used to store software programs or instructions as well as various data. The memory 909 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc. Additionally, memory 909 may include volatile memory or nonvolatile memory, or memory 909 may include both volatile and nonvolatile memory. Among them, non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory. Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM). Memory 909 in embodiments of the present application includes, but is not limited to, these and any other suitable types of memory.

The processor 910 may include one or more processing units; optionally, the processor 910 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 910.

Among them, the radio frequency unit 901 is used to obtain the first channel information of the target channel;

Processor 910, configured to determine a target AI network model corresponding to a first length from a preconfigured AI network model, the first length being indicated by the network side device or determined by the terminal according to the first information, wherein the first length is The first information includes at least one of the following: the first channel information and the AI network model index indicated by the network side device;

The processor 910 is also configured to use the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The radio frequency unit 901 is also configured to send the first channel characteristic information to the network side device.

Optionally, before the processor 910 executes the step of determining the target AI network model corresponding to the first length from the preconfigured AI network models, the method further includes:

The radio frequency unit 901 is also configured to receive relevant information from the N first AI network models of the network side device, where the preconfigured AI network models include the N first AI network models, and the N The first AI network model has a one-to-one correspondence with N lengths, where N is greater than or equal to an integer of 1.

Optionally, the reception performed by the radio frequency unit 901 of the relevant information of the N first AI network models from the network side device includes:

Optionally, before performing the receiving of relevant information of the N first AI network models from the network side device, the radio frequency unit 901 is also configured to send target capability information to the network side device, wherein the target The capability information is used to assist the network side device in determining the N first AI network models.

The identification of the first AI network model supported by the terminal;

The terminal supports the calculated channel status.

Optionally, the step of determining the target AI network model corresponding to the first length from the preconfigured AI network models performed by the processor 910 includes:

The first indication information from the network side device is received through the radio frequency unit 901. The first indication information is used to indicate at least one of the second AI network model and the corresponding length of the second AI network model. The N The first AI network model includes the second AI network model;

The processor 910 determines that the target AI network model is the second AI network model indicated in the first indication information, and/or the terminal determines that the first length is the second AI network model indicated in the first indication information. The length corresponding to the second AI network model.

The processor 910 determines the target AI network model from the N first AI network models according to at least one of channel characteristics and channel conditions corresponding to the first channel information, and/or, from the N The first length is determined among the lengths.

Optionally, the radio frequency unit 901 is also configured to send second indication information to the network side device, where the second indication information is used to indicate at least one of the target AI network model and the first length.

Optionally, the processor 910 determines the target AI network model from the N first AI network models based on at least one of channel characteristics and channel conditions corresponding to the first channel information, And/or, determining the first length from the N lengths includes:

The processor 910 determines that the first length is equal to the length associated with the value of the target channel parameter in the first channel information, and/or determines that the target AI network model is the target channel according to the first association relationship. AI network models associated with parameter values, wherein the first association includes each value or each value range of the target channel parameter and the N first AI network models and/or the N lengths the relationship between; or,

The processor 910 determines that the first length is equal to the length corresponding to the coding identifier associated with the value of the target channel parameter according to the second association relationship, and/or determines that the target AI network model is the target channel parameter. The AI network model corresponding to the value-associated coding identifier, wherein the second association relationship includes an association relationship between each value or each value range of the target channel parameter and N coding identifiers, and the N The encoding identifiers correspond to the N first AI network models one-to-one, and/or the N encoding identifiers correspond to the N lengths one-to-one.

The number of effective delay paths of the target channel;

The delay interval between the two target paths of the target channel;

Optionally, the radio frequency unit 901 is also configured to receive relevant information of K third AI network models from the network side device, where the third AI network model is related to the fourth AI network model, and the third AI network model is related to the fourth AI network model. The four AI network models are decoding network models of the network side device, or the third AI network model is a public decoding network model, and K third AI network models correspond to N first AI network models , K is an integer greater than or equal to 1;

The processor 910 executes the step of obtaining the information from the N first AI networks according to the channel status of the target channel. Determine the target AI network model in the network model, including:

The processor 910 processes the first channel characteristic information obtained by processing the target first AI network model into second channel information through the target third AI network model, where the target first AI network model is the N first AI Any one of the network models, the K third AI network models include the target third AI network model, and the target third AI network model corresponds to the target first AI network model;

The processor 910 obtains the degree of matching between the second channel information corresponding to the first channel characteristic information processed by the N first AI network models and the first channel information respectively;

When the processor 910 determines that the matching degree between the target second channel information and the first channel information satisfies the preset conditions, the processor 910 determines that the processed first AI network model of the target first channel characteristic information is the target AI network. A model, wherein the target second channel information corresponds to the target first channel characteristic information.

Optionally, the radio frequency unit 901 receives first indication information from the network side device, where the first indication information is used to indicate at least one of the second AI network model and the length corresponding to the second AI network model. In this case, the determination of the target AI network model based on the first information performed by the processor 910 includes:

The processor 910 performs the processing according to at least one of the channel characteristics and channel conditions corresponding to the first channel information. One item, determining a seventh AI network model that matches the channel state of the target channel and has the smallest corresponding length, the N first AI network models include the seventh AI network model, and the first channel information is the Channel information of the target channel;

When the length corresponding to the second AI network model is greater than the length corresponding to the seventh AI network model, the processor 910 determines that the target AI network model is the seventh AI network model.

Optionally, the processing performed by the processor 910 using the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length includes:

The processor 910 uses the seventh AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

Send second indication information to the network side device through the radio frequency unit 901, where the second indication information is used to indicate at least one of the seventh AI network model and the first length;

or,

The processing performed by the processor 910 using the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length includes:

The processor 910 uses the seventh AI network model to process the first channel information to obtain the channel characteristic information of the third length, and the processor 910 supplements the channel characteristic information of the third length to the first length to obtain the channel characteristic information of the third length. The first channel characteristic information, wherein the first length is the length corresponding to the second AI network model.

The terminal 900 provided by the embodiment of the present application can perform each process performed by each module in the channel characteristic information reporting device 600 as shown in Figure 6, and can achieve the same beneficial effects. To avoid duplication, details will not be described here.

An embodiment of the present application also provides a network side device, including a processor and a communication interface. The communication interface is used to receive first channel characteristic information from a terminal, where the first channel characteristic information is a target AI network adopted by the terminal. The model processes the first channel characteristic information to obtain the first length of channel characteristic information; the processor is configured to use a fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the third channel characteristic information. One channel information.

This network-side device embodiment corresponds to the above-mentioned network-side device method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this network-side device embodiment, and can achieve the same technical effect.

Specifically, the embodiment of the present application also provides a network side device. As shown in Figure 10, the network side device 1000 includes: an antenna 1001, a radio frequency device 1002, a baseband device 1003, a processor 1004 and a memory 1005. Antenna 1001 is connected to radio frequency device 1002. In the uplink direction, the radio frequency device 1002 receives information through the antenna 1001 and sends the received information to the baseband device 1003 for processing. In the downlink direction, the baseband device 1003 processes the information to be sent and sends it to the radio frequency device 1002. The radio frequency device 1002 processes the received information and sends it out through the antenna 1001.

The method performed by the network side device in the above embodiment can be implemented in the baseband device 1003, which includes a baseband processor.

The baseband device 1003 may include, for example, at least one baseband board on which multiple chips are disposed, as shown in FIG. Program to perform the network device operations shown in the above method embodiments.

The network side device may also include a network interface 1006, which is, for example, a common public radio interface (CPRI).

Specifically, the network side device 1000 in this embodiment of the present invention also includes: instructions or programs stored in the memory 1005 and executable on the processor 1004. The processor 1004 calls the instructions or programs in the memory 1005 to execute each of the steps shown in Figure 7 The method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.

Embodiments of the present application also provide a readable storage medium. Programs or instructions are stored on the readable storage medium. When the program or instructions are executed by a processor, each process of the method embodiment shown in Figure 2 or Figure 5 is implemented. , and can achieve the same technical effect, so to avoid repetition, they will not be described again here.

Wherein, the processor is the processor in the terminal described in the above embodiment. The readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.

An embodiment of the present application further provides a chip. The chip includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions. The implementation is as shown in Figure 2 or Figure 5. Each process of the method embodiment is shown, and the same technical effect can be achieved. To avoid repetition, the details will not be described here.

It should be understood that the chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.

Embodiments of the present application further provide a computer program/program product, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement Figure 2 or Figure 5 Each process of the method embodiment shown can achieve the same technical effect. To avoid repetition, it will not be described again here.

An embodiment of the present application also provides a communication system, including: a terminal and a network side device. The terminal can be used to perform the steps of the channel characteristic information reporting method described in the first aspect, and the network side device can be used to perform the steps of the channel characteristic information reporting method as described in the first aspect. The steps of the channel characteristic information recovery method described in the third aspect.

It should be noted that, in this document, the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device that includes a series of elements not only includes those elements, It also includes other elements not expressly listed or inherent in the process, method, article or apparatus. without further restrictions In this case, an element defined by the statement "comprises a..." does not exclude the presence of other identical elements in the process, method, article or device including 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 may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions may be performed, for example, the methods described may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation. Based on this understanding, the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology. The computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Inspired by this application, many forms can be made without departing from the purpose of this application and the scope protected by the claims, all of which fall within the protection of this application.

Claims

A method for reporting channel characteristic information, including:

The terminal obtains the first channel information of the target channel;

The terminal determines the target AI network model corresponding to the first length from the preconfigured AI network model, the first length is indicated by the network side device or determined by the terminal according to the first information, wherein the first length The information includes at least one of the following: the first channel information and the AI network model index indicated by the network side device;

The terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The terminal sends the first channel characteristic information to the network side device.
The method according to claim 1, wherein before the terminal determines the target AI network model corresponding to the first length from the preconfigured AI network model, the method further includes:

The terminal receives relevant information from the N first AI network models of the network side device, where the preconfigured AI network models include the N first AI network models, and the N first AI network models The network model has a one-to-one correspondence with N lengths, where N is an integer greater than or equal to 1.
The method according to claim 2, wherein the terminal receives relevant information of N first AI network models from the network side device, including:

When the terminal accesses the network side device, it receives relevant information of the N first AI network models; or,

When the terminal accesses the network side device, it receives relevant information of a part of the N first AI network models, and after the terminal accesses the network side device, it receives the N first AI network models. Related information from another part of the first AI network model.
The method according to claim 2, wherein before the terminal receives relevant information of the N first AI network models from the network side device, the method further includes:

The terminal sends target capability information to the network side device, where the target capability information is used to assist the network side device in determining the N first AI network models.
The method of claim 4, wherein the target capability information is used to indicate at least one of the following:

The identification of the first AI network model supported by the terminal;

The number of switching times of the first AI network model supported by the terminal;

The amount of data of the AI network model that the terminal supports transmission;

The terminal supports the calculated channel status.
The method according to claim 2, wherein the terminal models from a preconfigured AI network Determine the target AI network model corresponding to the first length in the model, including:

The terminal receives first indication information from the network side device, the first indication information is used to indicate at least one of the second AI network model and the length corresponding to the second AI network model, and the Nth An AI network model includes the second AI network model;

The terminal determines that the target AI network model is the second AI network model indicated in the first indication information, and/or the terminal determines that the first length is the second AI network model indicated in the first indication information. The length corresponding to the second AI network model.
The method according to claim 6, wherein the first channel information is related to the terminal's channel estimation result of the channel state information reference signal CSI-RS, and the first indication information corresponds to the CSI resources used by the terminal. .
The method according to claim 2, wherein the terminal determines the target AI network model corresponding to the first length from the preconfigured AI network model, including:

The terminal determines the target AI network model from the N first AI network models according to at least one of channel characteristics and channel conditions corresponding to the first channel information, and/or, from the N The first length is determined among the lengths.
The method of claim 8, further comprising:

The terminal sends second indication information to the network side device, where the second indication information is used to indicate at least one of the target AI network model and the first length.
The method of claim 9, wherein the second indication information is carried in a fixed-length CSI part of the channel state information CSI report, and the first channel characteristic information is carried in a variable-length CSI part of the CSI report. CSI portion; or,

The part of the second length in the first channel characteristic information and the second indication information are carried in the fixed length CSI part, and the part in the first channel characteristic information except the part of the second length is carried in the variable length CSI part; or,

The second indication information and the first channel characteristic information of the first length are both carried in the variable-length CSI part.
The method of claim 10, wherein the second length is equal to a minimum length among the N lengths.
The method according to claim 8, wherein the terminal determines the target AI from the N first AI network models based on at least one of channel characteristics and channel conditions corresponding to the first channel information. a network model, and/or determining the first length from the N lengths, including:

The terminal determines that the first length is equal to the length associated with the value of the target channel parameter in the first channel information according to the first association relationship, and/or determines that the target AI network model is the target channel AI network model associated with parameter values, wherein the first association relationship includes all The association between each value or each value range of the target channel parameter and the N first AI network models and/or the N lengths; or,

The terminal determines that the first length is equal to the length corresponding to the coding identifier associated with the value of the target channel parameter according to the second association relationship, and/or determines that the target AI network model is the target channel parameter. The AI network model corresponding to the value-associated coding identifier, wherein the second association relationship includes an association relationship between each value or each value range of the target channel parameter and N coding identifiers, and the N The encoding identifiers correspond to the N first AI network models one-to-one, and/or the N encoding identifiers correspond to the N lengths one-to-one.
The method according to claim 12, wherein the target channel parameter corresponding to the first channel information includes at least one of the following:

The target channel is line-of-sight propagation or non-line-of-sight propagation;

The number of effective delay paths of the target channel;

The delay interval between the two target paths of the target channel;

The number of effective beams of the target channel, the effective beams include beams corresponding to the discrete Fourier transform DFT orthogonal basis with power greater than the first threshold.
The method of claim 8, further comprising:

The terminal receives relevant information of K third AI network models from the network side device, wherein the third AI network model is related to a fourth AI network model, and the fourth AI network model is the network The decoding network model of the side device, or the third AI network model is a public decoding network model, and K third AI network models correspond to N first AI network models, and K is greater than or equal to 1 integer;

The terminal determines the target AI network model from the N first AI network models according to the channel status of the target channel, including:

The terminal processes the first channel characteristic information obtained by processing the target first AI network model into second channel information through the target third AI network model, where the target first AI network model is the N first AI Any one of the network models, the K third AI network models include the target third AI network model, and the target third AI network model corresponds to the target first AI network model;

The terminal obtains the degree of matching between the second channel information corresponding to the first channel characteristic information processed by the N first AI network models and the first channel information respectively;

When the terminal determines that the matching degree between the target second channel information and the first channel information satisfies the preset conditions, the terminal determines that the first AI network model of the processed target first channel characteristic information is the target AI network. A model, wherein the target second channel information corresponds to the target first channel characteristic information.
The method of claim 14, wherein the K third AI network models include At least one of the following:

N fifth AI network models corresponding one-to-one to the N first AI network models, where the fifth AI network model is related to the fourth AI network model corresponding to the same first AI network model;

M sixth AI network models corresponding to the N first AI network models, each of the sixth AI network models corresponding to at least one first AI network model, and the sixth AI network model is used for simulation Corresponding to the fourth AI network model of the same first AI network model, M is a positive integer less than or equal to N.
The method according to claim 14, wherein the matching degree of the target second channel information and the first channel information satisfying a preset condition includes at least one of the following:

The correlation between the target second channel information and the first channel information is greater than or equal to a preset correlation;

The channel capacity of the target second channel information is greater than or equal to a first preset value times the channel capacity of the first channel information, and the first preset value is greater than 0 and less than or equal to 1;

The target second channel information is the one in which the channel quality indicator CQI of the K second channel information is the same as or closest to the CQI of the first channel information;

The target second channel information is one of the K second channel information whose modulation and coding scheme MCS is the same as or closest to the MCS of the first channel information;

The target second channel information is the one with the shortest length among the K pieces of second channel information.
The method according to any one of claims 2 to 16, wherein the terminal receives first indication information from a network side device, the first indication information is used to indicate the second AI network model and the third In the case of at least one of the lengths corresponding to the two AI network models, the terminal determines the target AI network model based on the first information including:

The terminal determines a seventh AI network model that matches the channel state of the target channel and has the smallest corresponding length according to at least one of the channel characteristics and channel conditions corresponding to the first channel information. The N first AI The network model includes the seventh AI network model, and the first channel information is the channel information of the target channel;

When the length corresponding to the second AI network model is greater than the length corresponding to the seventh AI network model, the terminal determines that the target AI network model is the seventh AI network model.
The method according to claim 17, wherein the terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length, including:

The terminal uses the seventh AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The method also includes:

The terminal sends second indication information to the network side device, and the second indication information is used to Indicate at least one of the seventh AI network model and the first length;

or,

The terminal uses the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length, including:

The terminal uses the seventh AI network model to process the first channel information to obtain channel characteristic information of a third length, and the terminal supplements the channel characteristic information of the third length to the first length to obtain the channel characteristic information of the third length. The first channel characteristic information, wherein the first length is the length corresponding to the second AI network model.
A device for reporting channel characteristic information, applied to terminals, the device includes:

The first acquisition module is used to acquire the first channel information of the target channel;

The first determination module is configured to determine the target AI network model corresponding to the first length from the preconfigured AI network model, the first length being indicated by the network side device or determined by the terminal according to the first information, wherein, The first information includes at least one of the following: the first channel information and the AI network model index indicated by the network side device;

A first processing module, configured to use the target AI network model to process the first channel information to obtain the first channel characteristic information of the first length;

The first sending module is configured to send the first channel characteristic information to the network side device.
A method for recovering channel characteristic information, including:

The network side device receives the first channel characteristic information from the terminal, where the first channel characteristic information is the channel characteristic information of the first length obtained by processing the first channel information by the terminal using the target AI network model;

The network side device uses a fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the first channel information.
The method according to claim 20, wherein before the network side device receives the first channel characteristic information from the terminal, the method further includes:

The network side device sends relevant information of N first AI network models to the terminal, where the N first AI network models correspond to N lengths one-to-one, and the N first AI network models include For the target AI network model, the N lengths include the first length, and N is an integer greater than or equal to 1.
The method according to claim 21, wherein the network side device sends relevant information of N first AI network models to the terminal, including:

When the terminal accesses the network side device, the network side device sends relevant information of the N first AI network models to the terminal; or,

When the terminal accesses the network side device, the network side device sends relevant information of a part of the N first AI network models to the terminal, and when the terminal accesses the After the network side device is connected to the network side device, the network side device sends relevant information of another part of the N first AI network models to the terminal.
The method according to claim 21, wherein before the network side device receives the first channel characteristic information from the terminal, the method further includes:

The network side device sends first indication information to the terminal, where the first indication information is used to indicate at least one of a second AI network model and a length corresponding to the second AI network model.
The method according to claim 23, wherein before the network side device sends the first indication information to the terminal, the method further includes:

The network side device receives target capability information from the terminal, where the target capability information is used to assist the network side device in determining the N first AI network models.
The method of claim 24, wherein the target capability information is used to indicate at least one of the following:

The identification of the first AI network model supported by the terminal;

The number of switching times of the first AI network model supported by the terminal;

The amount of data of the AI network model that the terminal supports transmission;

The terminal supports the calculated channel status.
The method of claim 23, wherein the first channel information is related to a channel estimation result of a channel state information reference signal CSI-RS by the terminal, and the first indication information corresponds to a CSI resource used by the terminal. .
The method according to claim 21, wherein before the network side device receives the first channel characteristic information from the terminal, the method further includes:

The network side device receives second indication information from the terminal, where the second indication information is used to indicate at least one of the target AI network model and the first length.
The method of claim 27, wherein the second indication information is carried in a fixed-length CSI part of the channel state information CSI report, and the first channel characteristic information is carried in a variable-length CSI part of the CSI report. CSI portion; or,

The part of the second length in the first channel characteristic information and the second indication information are carried in the fixed length CSI part, and the part in the first channel characteristic information except the part of the second length is carried in the variable length CSI part; or,

The second indication information and the first channel characteristic information of the first length are both carried in the variable-length CSI part.
The method of claim 28, wherein the second length is equal to a minimum length of the N lengths.
The method according to claim 27, wherein before the network side device receives the second indication information from the terminal, the method further includes:

The network side device sends relevant information of K third AI network models to the terminal, where the third AI network model is related to the fourth AI network model, or the third AI network model is a public Decode the network model, and K third AI network models correspond to N first AI network models, and K is an integer greater than or equal to 1.
The method of claim 30, wherein the K third AI network models include at least one of the following:

N fifth AI network models corresponding one-to-one to the N first AI network models, where the fifth AI network model is related to the fourth AI network model corresponding to the same first AI network model;

M sixth AI network models corresponding to the N first AI network models, each of the sixth AI network models corresponding to at least one first AI network model, and the sixth AI network model is used for simulation Corresponding to the fourth AI network model of the same first AI network model, M is a positive integer less than or equal to N.
A channel characteristic information recovery device, applied to network side equipment, the device includes:

The first receiving module is configured to receive the first channel characteristic information from the terminal, where the first channel characteristic information is the channel characteristic of the first length obtained by the terminal using the target AI network model to process the first channel information. information;

The second processing module is configured to use a fourth AI network model corresponding to the first length to process the first channel characteristic information to obtain the first channel information.
A terminal includes a processor and a memory, the memory stores programs or instructions that can be run on the processor, wherein when the program or instructions are executed by the processor, any one of claims 1 to 18 is implemented. The steps of the channel characteristic information reporting method described in one item.
A network-side device includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, the implementation of claims 20 to 31 is achieved. The steps of the channel characteristic information recovery method described in any one of the above.
A readable storage medium storing programs or instructions on the readable storage medium, wherein when the program or instructions are executed by a processor, the channel characteristic information reporting method as described in any one of claims 1 to 18 is implemented. , or implement the steps of the channel characteristic information recovery method as described in any one of claims 20 to 31.