WO2025209317A1 - Channel feedback information transmission method and apparatus - Google Patents

Abstract

A channel feedback information transmission method and apparatus, relating to the technical field of wireless communications. The method comprises: a second device receives a first signal from a first device; and the second device sends channel feedback information to the first device, the channel feedback information indicating non-zero coefficient quantization information and non-zero coefficient selection information in a coefficient matrix, and the coefficient matrix being obtained on the basis of a spatial domain basis and a frequency domain basis. The spatial domain basis and the frequency domain basis satisfy a first option, a second option, or a third option; the first option is that the spatial domain basis consists of all the vectors in a spatial domain basis matrix, and the frequency domain basis consists of all the vectors in a frequency domain basis matrix; the second option is that the spatial domain basis consists of a subset of vectors in the spatial domain basis matrix, and the frequency domain basis consists of all the vectors in the frequency domain basis matrix; and the third option is that the spatial domain basis consists of all the vectors in the spatial domain basis matrix, and the frequency domain basis consists of a subset of vectors in the frequency domain basis matrix.

Description

A method and device for transmitting channel feedback information

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of the People's Republic of China on April 3, 2024, with application number 202410417546.0 and invention name "A channel feedback information transmission method and device", the entire contents of which are incorporated by reference into this application.

Technical Field

The present application relates to the field of wireless communication technology, and in particular to a method and device for transmitting channel feedback information.

Background Art

In wireless communication systems, base stations need to obtain downlink channel state information (CSI) to determine the downlink data channel resources, modulation and coding scheme (MCS), precoding, and other configurations for scheduling terminals. Terminals can obtain CSI by measuring downlink reference signals and send it to the base station. To reduce CSI transmission overhead, terminals that support codebook features can use a codebook-based approach for CSI feedback.

In wireless communication systems that incorporate artificial intelligence (AI) technology, terminals can provide CSI feedback based on AI models. When AI is introduced into wireless communication networks, model monitoring is necessary to monitor the performance of the AI models. This requires terminals to send true CSI to the base station for model monitoring. To reduce CSI overhead, terminals that support codebook features can use a codebook-based approach to transmit true CSI.

For terminals that do not support the codebook feature, how to reduce the CSI transmission overhead while taking into account CSI accuracy is a problem that needs to be solved.

Summary of the Invention

The embodiments of the present application provide a method and apparatus for transmitting channel feedback information, which are used to reduce the transmission overhead of the channel feedback information while taking into account the accuracy of the channel feedback information. The embodiments of the present application can be applied to terminals that do not support codebook characteristics.

The embodiment of the present application can be applied to a process of transmitting channel feedback information between a first device and a second device. The first device and the second device have wireless communication capabilities, and the second device can send channel feedback information to the first device, and the first device can also send channel feedback information to the second device.

The channel feedback information can be obtained by the second device based on the first signal sent by the first device. The first signal is used for channel measurement, and the channel measurement result can be used for resource scheduling, etc. The first signal can be a reference signal, for example, a channel state information reference signal (CSI-RS) or a synchronization signal block (SSB). The channel feedback information is used to indicate the channel quality between the first device and the second device, or can be information used to characterize the channel, such as channel response, channel characteristics (composed of eigenvectors of channel response), precoding matrix, etc. An example of the channel feedback information is CSI.

The first device may be a network-side device, for example, the first device may be a base station, and the second device may be a terminal-side device. The network-side device may be a network device, or a module (such as a chip) in a network device, or software containing network device functions (such as a control subsystem), or other devices that communicate with the network device, such as an AI network element, which is a server, such as an OTT device or a cloud server. The terminal-side device may be a terminal device, or a module (such as a chip) in a terminal device, or software containing terminal device functions (such as a control subsystem), or other devices that communicate with the terminal device, such as an AI network element, which is a server, such as an OTT device or a cloud server.

In a first aspect, a channel feedback information transmission method is provided, which can be applied to a second device, the method comprising: receiving a first signal from a first device; sending channel feedback information to the first device, wherein the channel feedback information is obtained based on the first signal, the channel feedback information indicates quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, the non-zero coefficient selection information indicates the position of the non-zero coefficients, and the coefficient matrix is obtained based on a spatial basis and a frequency domain basis. The spatial basis and the frequency domain basis satisfy the first option, the second option, or the third option; the first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix.

In the above implementation, when a terminal transmits channel feedback information to a network device, the terminal may not need to perform one or more of the following operations: spatial basis selection and frequency basis selection. Accordingly, the terminal does not need to transmit the corresponding selection information to the network device. For terminals that do not support codebook features, the above implementation can achieve a balance between accuracy and transmission overhead when transmitting channel feedback information.

In one possible implementation, the spatial basis and the frequency domain basis satisfy the second option; the channel feedback information also includes spatial basis selection information, and the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix.

In one possible implementation, the spatial domain basis and the frequency domain basis satisfy the third option; the channel feedback information also includes frequency domain basis selection information, and the frequency domain basis selection information indicates partial vectors in the frequency domain basis matrix, and the partial vectors are used to determine the coefficient matrix.

In a possible implementation, the channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine a spatial basis.

In one possible implementation, the spatial basis is determined based on spatial oversampling factor selection information, the spatial oversampling factor selection information is based on a configuration from the first device, or the spatial oversampling factor selection information is preconfigured, or the spatial oversampling factor selection information is predefined.

In one possible implementation, the method further includes: sending first channel feedback information to the first device, where the first channel feedback information is obtained by a first AI model based on the first signal, and the first AI model is located on the second device; and sending channel feedback information to the first device includes: sending second channel feedback information to the first device, where the second channel feedback information is used to monitor the performance of the first AI model and/or the second AI model, the second AI model is located on the first device, and the second AI model is used to recover the first channel feedback information.

In a possible implementation, it also includes: receiving configuration information from the first device, the configuration information including first indication information, the first indication information being used to determine the first option, the second option or the third option; sending channel feedback information to the first device, including: determining the first option, the second option or the third option based on the first indication information; determining the channel feedback information based on the first option, the second option or the third option; and sending the channel feedback information to the first device.

In a possible implementation, the configuration information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis.

In one possible implementation, the first indication information is used to determine the second option, and the first indication information indicates the number of spatial bases, and the number of spatial bases is less than the number of all vectors in the spatial basis matrix; the channel feedback information also includes spatial basis selection information, and the spatial basis selection information indicates some vectors in the spatial basis matrix, and the some vectors are determined based on the number of spatial bases; the coefficient matrix is determined based on the spatial basis vectors indicated by the spatial basis selection information and all vectors in the frequency domain basis matrix.

In one possible implementation, the first indication information is used to determine the third option, the first indication information indicates the frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; the channel feedback information also includes frequency domain basis selection information, and the frequency domain basis selection information indicates part of the vectors in the frequency domain basis matrix, and the part of the vectors is determined according to the frequency domain basis selection ratio; the coefficient matrix is determined according to the frequency domain basis vector indicated by the frequency domain basis selection information and all vectors in the spatial domain basis matrix.

In one possible implementation, before receiving the configuration information from the first device, it also includes: sending terminal capability information to the first device, the terminal capability information indicating that the second device does not support the codebook characteristics, or the terminal capability information indicates that the second device has the ability to determine the coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix, or the terminal capability information indicates that the second device has the ability to send channel feedback information based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix.

By adopting the above implementation, an appropriate channel feedback information transmission mode can be configured for the second device (terminal) according to its capability.

In a second aspect, a channel feedback information transmission method is provided, which can be applied to a first device, the method comprising: sending a first signal to a second device; receiving channel feedback information from the second device, the channel feedback information being obtained based on the first signal, the channel feedback information indicating quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, the non-zero coefficient selection information indicating the position of the non-zero coefficient, the channel feedback information, the spatial basis, and the frequency domain basis being used to determine a precoding matrix. The spatial basis and the frequency domain basis satisfy the first option, the second option, or the third option; the first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix.

In one possible implementation, the spatial basis and the frequency domain basis satisfy the second option; the channel feedback information also includes spatial basis selection information, and the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix.

In one possible implementation, the spatial domain basis and the frequency domain basis satisfy the third option; the channel feedback information also includes frequency domain basis selection information, and the frequency domain basis selection information indicates partial vectors in the frequency domain basis matrix, and the partial vectors are used to determine the coefficient matrix.

In a possible implementation manner, the channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis.

In one possible implementation, the spatial basis is determined based on spatial oversampling factor selection information, and the spatial oversampling factor selection information is preconfigured or predefined; or, the method further includes: sending the spatial oversampling factor selection information to the second device.

In one possible implementation, before receiving the channel feedback information from the second device, the method further includes: receiving first channel feedback information from the second device, where the first channel feedback information is obtained by a first AI model based on the first signal, and the first AI model is located on the second device; and receiving the channel feedback information from the second device includes: receiving second channel feedback information from the second device, where the second channel feedback information is used to monitor the performance of the first AI model and/or the second AI model, the second AI model is located on the first device, and the second AI model is used to recover the first channel feedback information.

In one possible implementation, it further includes: sending configuration information to the second device, the configuration information including first indication information, the first indication information being used to indicate the first option, the second option or the third option; after receiving channel feedback information from the second device, it further includes: determining a precoding matrix based on the channel feedback information and the first option, the second option or the third option indicated by the first indication information.

In a possible implementation, the configuration information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis.

In one possible implementation, the first indication information indicates the second option and the number of spatial bases, and the number of spatial bases is less than the number of all vectors in the spatial basis matrix; the channel feedback information also includes spatial basis selection information, and the spatial basis selection information indicates some vectors in the spatial basis matrix, and the some vectors are determined according to the number of spatial bases; the first option, the second option or the third option indicated by the channel feedback information and the first indication information includes: determining the precoding matrix according to the quantization information of the non-zero coefficients in the channel feedback information and the non-zero coefficient selection information, the spatial basis vectors indicated by the spatial basis selection information and all vectors in the frequency domain basis matrix.

In one possible implementation, the first indication information indicates the third option and the frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; the channel feedback information also includes frequency domain basis selection information, and the frequency domain basis selection information indicates part of the vectors in the frequency domain basis matrix, and the part of the vectors is determined according to the frequency domain basis selection ratio; the first option, the second option or the third option indicated by the channel feedback information and the first indication information includes: determining the precoding matrix according to the quantization information of the non-zero coefficients in the channel feedback information and the non-zero coefficient selection information, all vectors in the spatial domain basis matrix and the frequency domain basis vector indicated by the frequency domain basis selection information.

In one possible implementation, before sending the configuration information to the second device, it also includes: receiving terminal capability information from the second device, the terminal capability information indicating that the second device does not support the codebook characteristics, or the terminal capability information indicates that the second device has the ability to determine the coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix.

In a third aspect, a channel feedback information transmission method is provided, which can be applied to a second device, the method comprising: receiving a first signal from a first device; sending first channel feedback information to the first device, the first channel feedback information being obtained by a first AI model based on the first signal, the first AI model being located in the second device; receiving third channel feedback information from the first device, the third channel feedback information being recovered by a second AI model from the first channel feedback information, the second AI model being located in the first device, the third channel feedback information comprising quantization information of non-zero coefficients in a coefficient matrix and non-zero coefficient selection information, the non-zero coefficient selection information indicating the positions of the non-zero coefficients, the coefficient matrix being obtained based on a spatial basis and a frequency domain basis. The spatial basis and the frequency domain basis satisfy the first option, the second option, or the third option; the first option being: the spatial basis is all vectors in a spatial basis matrix, and the frequency domain basis is all vectors in a frequency domain basis matrix; the second option being: the spatial basis is a portion of vectors in a spatial basis matrix, and the frequency domain basis is all vectors in a frequency domain basis matrix; the third option being: the spatial basis is all vectors in a spatial basis matrix, and the frequency domain basis is a portion of vectors in a frequency domain basis matrix.

In a possible implementation, the method further includes: receiving configuration information from the first device, where the configuration information includes first indication information, and the first indication information is used to determine the first option, the second option, or the third option.

In one possible implementation, before receiving the configuration information from the first device, it also includes: sending terminal capability information to the first device, the terminal capability information indicating that the second device does not support the codebook characteristics, or the terminal capability information indicates that the second device has the ability to determine the coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix.

In a fourth aspect, a channel feedback information transmission method is provided, which can be applied to a second device, the method comprising: sending a first signal to a first device; receiving first channel feedback information from the first device, the first channel feedback information being obtained by a first AI model based on the first signal, the first AI model being located in the second device; sending third channel feedback information to the first device, the third channel feedback information being recovered by a second AI model from the first channel feedback information, the second AI model being located in the first device, the third channel feedback information comprising quantization information of non-zero coefficients in a coefficient matrix and non-zero coefficient selection information, the non-zero coefficient selection information indicating the position of the non-zero coefficients, the coefficient matrix being obtained based on a spatial basis and a frequency domain basis. The spatial basis and the frequency domain basis satisfy the first option, the second option, or the third option; the first option being: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option being: the spatial basis is a portion of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option being: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is a portion of the vectors in the frequency domain basis matrix.

In a possible implementation, the method further includes: sending configuration information to the second device, where the configuration information includes first indication information, and the first indication information is used to determine the first option, the second option, or the third option.

In one possible implementation, before sending the configuration information to the second device, it also includes: receiving terminal capability information from the first device, the terminal capability information indicating that the second device does not support the codebook characteristics, or the terminal capability information indicates that the second device has the ability to determine the coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix.

Based on the third aspect or the fourth aspect above, in a possible implementation manner, the configuration information further includes spatial oversampling factor selection information.

Based on the third aspect or the fourth aspect above, in a possible implementation method, the first indication information is used to determine the second option, and the first indication information indicates the number of spatial bases, and the number of spatial bases is less than the number of all vectors in the spatial basis matrix; the channel feedback information also includes spatial basis selection information, and the spatial basis selection information indicates some vectors in the spatial basis matrix, and the some vectors are determined based on the number of spatial bases.

Based on the above-mentioned third aspect or fourth aspect, in a possible implementation method, the first indication information is used to determine the third option, and the first indication information indicates the frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; the channel feedback information also includes frequency domain basis selection information, and the frequency domain basis selection information indicates a partial vector in the frequency domain basis matrix, and the partial vector is determined according to the frequency domain basis selection ratio.

Based on the above-mentioned third aspect or fourth aspect, in a possible implementation method, the spatial basis and the frequency domain basis satisfy the second option; the third channel feedback information also includes spatial basis selection information, and the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix.

Based on the above-mentioned third aspect or fourth aspect, in a possible implementation method, the spatial domain basis and the frequency domain basis satisfy the third option; the third channel feedback information also includes frequency domain basis selection information, and the frequency domain basis selection information indicates a partial vector in the frequency domain basis matrix, and the partial vector is used to determine the coefficient matrix.

Based on the third aspect or the fourth aspect above, in a possible implementation manner, the third channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis.

Based on the third aspect or the fourth aspect above, in a possible implementation method, the spatial basis is determined according to spatial oversampling factor selection information, the spatial oversampling factor selection information is based on the configuration from the first device, or the spatial oversampling factor selection information is preconfigured, or the spatial oversampling factor selection information is predefined.

In the fifth aspect, a communication method is provided, which can be applied to a second device, the method comprising: sending terminal capability information to a first device, the terminal capability information being used to determine a channel feedback information transmission method; and receiving configuration information from the first device, the configuration information being used to indicate a channel feedback information transmission method that matches the terminal capability information.

In the sixth aspect, a communication method is provided, which can be applied to a first device, the method including: receiving terminal capability information from the first device, the terminal capability information being used to determine a channel feedback information transmission method; and sending configuration information to the first device, the configuration information being used to indicate a channel feedback information transmission method that matches the terminal capability information.

Based on the above-mentioned fifth aspect or sixth aspect, in a possible implementation method, the terminal capability information indicates that the second device supports the first channel feedback information transmission method or supports the codebook characteristics; the configuration information includes first indication information, and the first indication information indicates a first option, and the first option is: determining the spatial basis for non-zero coefficients as all vectors in the spatial basis matrix, and the frequency domain basis for determining the non-zero coefficients as all vectors in the frequency domain basis matrix.

Based on the above-mentioned fifth aspect or sixth aspect, in a possible implementation method, the terminal capability information indicates that the second device supports a second channel feedback information transmission method or supports codebook characteristics; the configuration information includes first indication information, and the first indication information indicates a second option, and the second option is: the spatial basis used to determine the non-zero coefficient is a part of the vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficient is all vectors in the frequency domain basis matrix.

Based on the fifth aspect or the sixth aspect above, in a possible implementation, the first indication information further indicates the number of spatial bases, and the number of spatial bases is smaller than the number of all vectors in the spatial basis matrix.

Based on the above-mentioned fifth aspect or sixth aspect, in a possible implementation method, the terminal capability information indicates that the second device supports a third channel feedback information transmission method or supports codebook characteristics; the configuration information includes first indication information, and the first indication information indicates a third option, and the third option is: the spatial basis used to determine the non-zero coefficient is all vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficient is part of the vectors in the frequency domain basis matrix.

Based on the fifth or sixth aspect above, in a possible implementation manner, the first indication information further indicates a frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1.

Based on the fifth or sixth aspect above, in a possible implementation manner, the configuration information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis.

In the seventh aspect, a communication device is provided, comprising a unit or module for executing the method as described in any one of the first aspect, or a unit or module for executing the method as described in any one of the second aspect, or a unit or module for executing the method as described in any one of the third aspect, or a unit or module for executing the method as described in any one of the fourth aspect, or a unit or module for executing the method as described in any one of the fifth aspect, or a unit or module for executing the method as described in any one of the sixth aspect.

In an eighth aspect, a communication device is provided, comprising: one or more processors configured to execute the method as described in any one of the first aspect, or to execute the method as described in any one of the second aspect, or to execute the method as described in any one of the third aspect, or to execute the method as described in any one of the fourth aspect, or to execute the method as described in any one of the fifth aspect, or to execute the method as described in any one of the sixth aspect.

In the ninth aspect, a readable storage medium is provided, wherein the readable storage medium stores a program or instruction. When the program or instruction is run on a device, the device executes the method as described in any one of the first aspect, or executes the method as described in any one of the second aspect, or executes the method as described in any one of the third aspect, or executes the method as described in any one of the fourth aspect, or executes the method as described in any one of the fifth aspect, or executes the method as described in any one of the sixth aspect.

In the tenth aspect, a chip system is provided, comprising a processor for supporting a device to implement the method as described in any one of the first aspect, or to implement the method as described in any one of the second aspect, or to implement the method as described in any one of the third aspect, or to implement the method as described in any one of the fourth aspect, or to implement the method as described in any one of the fifth aspect, or to implement the method as described in any one of the sixth aspect.

In the eleventh aspect, a program product is provided, comprising a program; when the program is executed by a processor, the method described in any one of the first aspect, or the method described in any one of the second aspect, or the method described in any one of the third aspect, or the method described in any one of the fourth aspect, or the method described in any one of the fifth aspect, or the method described in any one of the sixth aspect is implemented.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a neuron structure in an embodiment of the present application;

FIG2 is a schematic diagram of the layer relationship of the neural network in an embodiment of the present application;

FIG3 is a schematic diagram of a CSI feedback framework based on AE in an embodiment of the present application;

FIG4 is a schematic diagram of a flow chart of a base station performing model monitoring according to a related art;

FIG5 is a schematic diagram of a process of performing model monitoring by a terminal provided by the related art;

FIG6 is a schematic diagram of a communication system architecture applicable to an embodiment of the present application;

FIG7 is a schematic diagram of a simplified communication system architecture applicable to an embodiment of the present application;

FIG8 is a schematic diagram of another communication system architecture applicable to an embodiment of the present application;

FIG9 is a schematic diagram of a possible application framework in a communication system according to an embodiment of the present application;

FIG10 is a schematic diagram of a possible application framework in a communication system according to an embodiment of the present application;

FIG11 is a schematic diagram of a flow chart of a CSI transmission method provided in an embodiment of the present application;

FIG12a is a flow chart of a CSI transmission and a model monitoring method performed by a network side according to an embodiment of the present application;

FIG12 b is a flow chart of another CSI transmission and model monitoring method performed by the network side provided in an embodiment of the present application;

FIG13a is a flow chart of a CSI transmission method and a model monitoring method performed by a terminal side according to an embodiment of the present application;

FIG13 b is a flow chart of another CSI transmission method and a model monitoring method performed by a terminal side according to an embodiment of the present application;

FIG14 is a flow chart of a communication method provided in an embodiment of the present application;

FIG15 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

FIG16 is a schematic structural diagram of another communication device provided in an embodiment of the present application;

FIG17 is a schematic structural diagram of another communication device provided in an embodiment of the present application.

DETAILED DESCRIPTION

In order to understand the present application more clearly, some technologies and technical terms involved in the present application are first explained.

(1) AI, machine learning, AI models, and neural networks

AI refers to the intelligence exhibited by machines created by humans. Generally, AI refers to the technology that replicates human intelligence through ordinary computer programs. AI can be defined as machines or computers that mimic humans and possess cognitive functions associated with human thinking, such as learning and problem-solving. AI is able to learn from past experiences, make rational decisions, and respond quickly. The goal of AI is to understand intelligence by building computer programs capable of symbolic reasoning or deduction.

Machine learning is a path to artificial intelligence (AI), specifically using machine learning to solve AI problems. Machine learning theory primarily involves the design and analysis of algorithms that enable computers to automatically "learn." Machine learning algorithms automatically analyze data to identify patterns and use these patterns to make predictions about unknown data. Because learning algorithms involve extensive statistical theory, machine learning is particularly closely linked to inferential statistics, also known as statistical learning theory.

An AI model is an algorithm or computer program that implements AI functions. It represents the mapping relationship between the model's input and output. An AI model can be a neural network or other machine learning model.

Neural networks (NNs) are a specific implementation of machine learning. According to the universal approximation theorem, NNs can theoretically approximate any continuous function, enabling them to learn arbitrary mappings. Therefore, NNs can accurately abstractly model complex, high-dimensional problems.

The idea of neural networks comes from the neuron structure of brain tissue. Each neuron performs a weighted summation operation on the input value, and generates an output through an activation function. Figure 1 shows an example of the structure of a neuron. As shown in Figure 1, for example, the input of a neuron is x = [x ₀ , x ₁ ,…, x _n ], and the weights corresponding to each input are wt = [wt ₀ , wt ₁ ,…, wt _n ], where wt _i is the weight of x _i , used to weight x _i , and the bias of the weighted summation is b. The form of the activation function can be diverse. For example, the activation function of a neuron is:
y＝f(z)＝max(0,z)…………………………(1)

The output of this neuron is:

For another example, the activation function of a neuron is:
y=f(z)=z……………………(3)

The output of this neuron is:

b, wt _i , and _xi can be decimals, integers (eg, 0, positive integers, or negative integers), or complex numbers. The activation functions of different neurons in a neural network can be the same or different.

A neural network generally includes a multi-layer structure, and each layer may include one or more neurons. Increasing the depth and/or width of a neural network can improve the expressive power of the neural network and provide more powerful information extraction and abstract modeling capabilities for complex systems. The depth of a neural network can refer to the number of layers the neural network includes, and the number of neurons included in each layer can be referred to as the width of the layer. In one implementation, the neural network includes an input layer and an output layer. The input layer of the neural network processes the input received by the neurons and passes the result to the output layer, which then obtains the output result of the neural network. In another implementation, the neural network includes an input layer, a hidden layer, and an output layer, as shown in Figure 2. The input layer of the neural network processes the input received by the neurons and passes the result to the middle hidden layer. The hidden layer then passes the calculation result to the output layer or an adjacent hidden layer, and finally the output layer obtains the output result of the neural network. A neural network can include one or more hidden layers connected in sequence, without limitation.

During the training process of a neural network, a loss function can be defined. The loss function describes the gap or difference between the output value of the neural network and the ideal target value. This application does not limit the specific form of the loss function. The training process of a neural network is the process of adjusting the neural network parameters so that the value of the loss function is less than the threshold value or meets the target requirements. Among them, the neural network parameters include, for example, one or more of the number of layers of the neural network, the width (i.e., the number of neurons in the layer), the weights of the neurons, and the activation functions of the neurons.

(2) CSI and CSI Transmission Method

Currently, in long-term evolution (LTE) and new radio (NR) communication systems, base stations need to obtain downlink CSI to determine the resource, MCS, precoding, and other configurations for the downlink data channels of scheduled terminals. In time-division duplex (TDD) systems, due to the reciprocity of uplink and downlink channels, base stations can obtain uplink CSI by measuring uplink reference signals, thereby inferring more accurate downlink CSI, for example, using uplink CSI as downlink CSI. In frequency-division duplex (FDD) systems, uplink and downlink reciprocity cannot be guaranteed. Downlink CSI is obtained by terminals measuring downlink reference signals, such as CSI-RS or SSB. Therefore, terminals need to generate CSI reports according to protocol predefined methods or base station configurations and send these reports to the base station to obtain downlink CSI.

In this application, the meaning of CSI is broader than that of CSI in traditional solutions and is not limited to channel quality indication (CQI), precoding matrix indicator (PMI), rank indicator (RI), or CSI-RS resource indicator (CRI). It can also be one or more of channel response information (such as channel response matrix, frequency domain channel response information, time domain channel response information), weight information corresponding to the channel response, reference signal receiving power (RSRP), or signal to interference plus noise ratio (SINR). RI indicates the number of downlink transmission layers recommended by the terminal device, CQI indicates the modulation and coding scheme supported by the current channel conditions determined by the terminal device, and PMI indicates the precoding recommended by the terminal device. The number of precoding layers indicated by PMI may correspond to RI. Among them, CSI can be replaced by channel information, and CSI report can be replaced by channel feedback information or channel information report or channel information feedback.

The representation of CSI can also be called a transmission mode, transmission format, data format, or quantization mode. This application describes the CSI transmission mode as an example. Currently, CSI transmission modes include scalar quantization, codebook-based, and auto-encoder (AE)-based modes.

(1) Scalar quantization method

The scalar quantization method refers to quantizing each element (or value) in the precoding matrix using a quantization method such as Float32, Float16, or Nbit (N bits), and sending the quantized information when feeding back CSI.

The scalar quantization method has a large transmission overhead. For example, for a precoding matrix with 13 subbands, 32 ports, and 1 layer, Float32 quantization requires 26,624 bits, Float16 requires 13,312 bits, and 8-bit quantization requires 6,656 bits.

(2) Codebook-based approach

The codebook can be understood as a representation of a precoding matrix. A codebook-based approach refers to using a codebook to represent the CSI. For example, the CSI can be represented using a codebook based on codebook parameters of a Rel-16 Enhanced Type II codebook (hereinafter referred to as the R16 codebook).

Taking the R16 codebook as an example, the precoding matrix (or eigenmatrix or eigenvector matrix) of each layer can be regarded as the multiplication of three matrices: the spatial basis matrix, the coefficient matrix, and the frequency basis matrix. For example, the codebook (or precoding matrix) of layer 1 can be expressed as:

In formula (5), W ^l is the l-th layer precoding matrix, whose dimension is N _p ×N ₃ , which can be expressed as That is, a complex matrix of dimension _Np × _N3 , where _{Np = 2N1N2} , is the number of ports, _N1 and _N2 are configured by the base station, representing the number of ports in the horizontal and vertical directions respectively, and _N3 is the number of subbands, which is determined by the bandwidth and subband size configured by the system.

In formula (5), is the spatial basis matrix (composed of spatial basis vectors), usually a discrete Fourier transform (DFT) matrix or generated by a DFT matrix, with a dimension of N _p ×N _p , which can be expressed as The specific form is related to the antenna port (such as CSI-RS port) form (or arrangement), for example It is related to the number of antenna ports in the horizontal and vertical directions. For example, It can be expressed as:

Where kron() represents the Kronecker product of matrices, i.e., the tensor product. DFT() represents the DFT operation.

In formula (5), is the frequency domain basis matrix (composed of frequency domain basis vectors), usually a DFT matrix or generated by a DFT matrix, for example The dimension is N ₃ ×N ₃ and can be expressed as

In formula (5), is the coefficient matrix (or projection matrix, or weight matrix), whose dimension is N _p ×N ₃ , which can be expressed as

Based on the above formula (5), the l-th layer codebook It can also be expressed as the following formula (7):

Where l represents the number of layers, l = 1, 2, 3 _{, 4. i 2, 5, l} is the phase coefficient indication information, used to indicate the phase coefficient. The parameter t ranges from 0 to N ₃ - 1. γ _{t, l} is the power normalization factor.

L is related to the number of spatial basis vectors. A spatial basis matrix consists of 2L vectors, where L is an integer greater than or equal to 1. The first L vectors correspond to one polarization direction, and the last L vectors correspond to the other polarization direction. L is configured by the base station.

is the spatial basis, which is a vector in the orthogonal basis, such as the vector in the DFT matrix, which can be expressed as Among them, _N1 and _N2 are antenna port forms, _O1 and _O2 are spatial oversampling granularity, and _N1 and _N2 , _O1 and _O2 are all configured by the base station. and The spatial basis selection information is expressed, and _q1 and _q2 are expressed by the spatial oversampling factor selection information.

is the frequency domain basis, is a vector in the orthogonal basis, such as the vector in the DFT matrix, which can be expressed as The information is represented by the frequency domain basis selection. _{M υ} is the number of frequency domain basis vectors, _pυ is the selection ratio of the frequency domain basis, which is configured by the base station; R is the number of precoding matrix indicator (PMI) subbands (numberOfPMI-SubbandsPerCQI-Subband) for each channel quality indicator (CQI) subband, which is configured by the base station.

is the broadband amplitude coefficient, which is 4-bit quantized information; is the sub-band amplitude coefficient, which is 3-bit quantization information; is the phase coefficient, which is 4-bit quantized information.

Since the coefficient matrix contains many coefficients with smaller values, the spatial basis matrix, frequency domain basis matrix, and coefficient matrix can be reduced in dimensionality, retaining only the spatial and frequency domain basis corresponding to the coefficients with larger values in the coefficient matrix. For example, for each codebook layer, 2L vectors can be selected from the _Np vectors of the spatial basis matrix, M vectors (M is an integer greater than or equal to 1) can be selected from the _N3 vectors of the frequency domain basis matrix, and K coefficients (K is an integer greater than or equal to 1) can be selected from the _Np * _N3 coefficients in the coefficient matrix. Each coefficient can be quantized using Q bits (Q is an integer greater than 0), where the selected coefficients are called non-zero coefficients.

The base station may configure codebook parameters for the terminal, and the terminal may select a spatial basis, a frequency basis, or coefficients based on the codebook parameters configured by the base station. For example, the codebook parameters configured by the base station for the terminal may include L, a selection ratio _pυ for the frequency basis, and a selection ratio β for non-zero coefficients.

For example, Table 1 shows the R16 codebook parameters. Wherein, p _υ (0<p _υ <1) represents the selection ratio of the frequency domain basis, R is a parameter configured by the base station; β (0<β<1) represents the selection ratio of non-zero coefficients. The total number of non-zero coefficients in each layer does not exceed K ₀ , and the total number of non-zero coefficients in all layers cannot exceed 2K ₀ . in, is the ceiling operator, Represents the rounding of a real number x to the nearest integer not less than x.

Table 1

For example, Table 2 shows an enhanced codebook parameter. Compared with the R16 codebook parameter, the enhanced codebook parameter may have more non-zero coefficients, so the CSI accuracy is higher, but the corresponding CSI transmission overhead is higher.

Table 2

In Tables 1 and 2 above, υ represents the number of layers (or streams), υ = 1, 2, 3, or 4. For example, according to the parameter combination index with a value of 6 in Table 1, when υ = 1 or 2, the selection ratio of the frequency domain basis p _υ = 0.5, and when υ = 3 or 4, the selection ratio of the frequency domain basis p _υ = 0.25.

To reduce signaling overhead, the base station may send an index of a codebook parameter combination (eg, parameter combination index 6 in Table 1, or parameter combination index 9 in Table 2) to the terminal. The index of the codebook parameter combination may indicate a set of codebook parameters.

The terminal can select the spatial basis, frequency basis, and non-zero coefficients based on the codebook parameters configured by the base station. The terminal can also select the spatial oversampling factor, quantize the non-zero coefficients, and perform index reordering. The index reordering operation includes placing the frequency basis where the maximum coefficient is located in the first column and performing cyclic shifts on the other columns accordingly. Accordingly, when providing CSI feedback, the terminal reports information such as the spatial basis selection information, the spatial oversampling factor selection information, the frequency basis selection information, the non-zero coefficient selection information, and the quantization information of the non-zero coefficients.

Exemplarily, the non-zero coefficient selection information is in the form of a bitmap, the number of bits in the bitmap is equal to the number of coefficients in the coefficient matrix, and the bit with a value of 1 in the bitmap indicates that the coefficient at the corresponding position in the coefficient matrix is the non-zero coefficient of the feedback, that is, the non-zero coefficient selection information can indicate the position of the feedback non-zero coefficient in the coefficient matrix.

Exemplarily, the non-zero coefficients may include quantized amplitude coefficients and/or phase coefficients, and the amplitude coefficients include wideband amplitude coefficients and/or sub-band amplitude coefficients.

(3) AE-based CSI feedback

The AE model consists of two sub-models: an encoder and a decoder. AE generally refers to a network structure composed of these two sub-models. The encoder and decoder of an AE are typically trained together and can be used in conjunction with each other. The AE model can also be called a bilateral model, a dual-end model, or a collaborative model.

At present, AI technology can be introduced into wireless communication networks to realize network intelligence. For example, a specific application of introducing AI technology into wireless communication networks is CSI transmission based on AI models (or CSI feedback based on AE). For example, the terminal side compresses and quantizes CSI through an encoder (CSI generator), and the base station side recovers CSI through a decoder (CSI reconstructor). Figure 3 shows a schematic diagram of a CSI feedback framework based on AE. Among them, the input of the CSI generator can be called input CSI or true CSI or original CSI, the feedback from the terminal to the base station can be called CSI feedback or CSI report or compressed CSI, and the output of the CSI reconstructor can be called output CSI or recovered CSI or reconstructed CSI.

(3) Model monitoring (or model monitoring)

Model monitoring refers to detecting the performance of the AI model and determining whether the AI model is working properly. If the AI model performance is poor, you can switch to non-AI mode, replace the AI model, or update the AI model.

One model monitoring method, called intermediate key performance indicator (KPI) monitoring, can be used to test the accuracy of AI model outputs. This monitoring method involves obtaining the AI model output and determining whether the AI model's performance meets requirements by comparing the output with the corresponding label or ground-truth. Intermediate KPI monitoring can determine whether the AI model meets requirements based on one or more of the following KPI metrics: generalized cosine similarity (GCS), square generalized cosine similarity (SGCS), and normalized mean square error (NMSE).

Model monitoring can be performed by a network device (such as a base station) or by a terminal.

For example, Figure 4 shows a schematic diagram of a process for performing model monitoring by a base station. As shown in Figure 4, the terminal obtains the true CSI (see step 402) based on the reference signal sent by the base station (see step 401). The terminal compresses and quantizes the true CSI based on the CSI generator to obtain CSI feedback (that is, the CSI feedback is obtained based on AE), and sends the CSI feedback to the base station (see step 403). The terminal also sends the true CSI to the base station (see step 404). The base station uses a CSI reconstructor to recover the CSI feedback from the terminal to obtain recovered CSI, and determines whether the performance of the CSI generator and/or CSI reconstructor meets the requirements based on the recovered CSI and the true CSI from the terminal (see step 405).

For example, Figure 5 shows a schematic diagram of a process for performing model monitoring by a terminal. As shown in Figure 5, the terminal obtains true CSI (see step 502) based on the reference signal sent by the base station (see step 501). The terminal compresses and quantizes the true CSI based on the CSI generator to obtain CSI feedback (that is, the CSI feedback is obtained based on AE), and sends the CSI feedback to the base station (see step 503). The base station uses a CSI reconstructor to recover the CSI feedback from the terminal to obtain recovered CSI (see step 504), and sends the recovered CSI to the terminal (see step 505). The terminal determines whether the performance of the CSI generator and/or CSI reconstructor meets the requirements based on the recovered CSI and the true CSI (see step 506).

In step 404 of the process shown in FIG4 , to reduce transmission overhead, a terminal supporting the codebook feature can compress and quantize the true CSI using a codebook-based approach and send the compressed and quantized CSI to the base station. Using a codebook-based approach to quantize the true CSI has low overhead. For example, using the codebook parameters corresponding to a paramCombination-r16 value of 8 for the R16 codebook (see the R16 codebook parameters corresponding to the parameter combination index of 8 in Table 1), the quantized true CSI is less than 500 bits. To improve the accuracy of the true CSI, the values of one or more of the parameters, such as parameter L, frequency domain floor selection ratio p _υ , and non-zero coefficient selection ratio β, can be appropriately increased, thereby improving the accuracy of the true CSI. However, the codebook is an independent and optional feature. Some terminals may support AI-CSI features (for example, support for compression and quantization of true CSI based on a CSI generator) but not the codebook feature. For such terminals, true CSI cannot be transmitted using a codebook-based method. Instead, scalar quantization is used to transmit true CSI, resulting in high transmission overhead.

In step 505 of the process shown in Figure 5 , to reduce transmission overhead, the base station can also compress and quantize the recovered CSI using a codebook-based approach and send the compressed and quantized CSI to the terminal. Similarly, for terminals that do not support codebook features, the base station can only transmit the recovered CSI using scalar quantization, resulting in higher transmission overhead.

Similar problems also exist in the CSI feedback scenario in wireless communication systems that do not adopt AI technology. For example, for terminals that do not support codebook features, such as terminals that do not support the R16 codebook feature, CSI feedback cannot be performed based on a codebook, such as the R16 codebook, resulting in high CSI transmission overhead.

To this end, an embodiment of the present application provides a channel feedback information transmission method and a related device that can implement the method, which can be used for terminals that do not support codebook characteristics, so that they can take into account both accuracy and transmission overhead when transmitting channel feedback information.

In the embodiment of the present application, the channel feedback information may indicate channel quality, and be used for resource scheduling, etc. An example of the channel feedback information is CSI, which is not limited in the present application.

In the embodiment of the present application, "not supporting codebook characteristics" can be understood as not supporting the transmission of channel feedback information in a codebook-based manner. For example, taking CSI as an example, for a terminal to send CSI to a base station, a terminal that does not support codebook characteristics does not have the ability to perform one or more of the following operations when quantizing the true value CSI: selecting a spatial basis, selecting a spatial oversampling factor, selecting a frequency domain basis, index rearrangement, etc. For another example, still taking CSI as an example, for a terminal to receive CSI sent by a base station, a terminal that does not support codebook characteristics, after receiving the recovered CSI sent by the base station in a codebook-based manner, does not have one or more of the following capabilities: determining the spatial basis matrix according to the spatial basis selection information, determining the frequency domain basis matrix according to the frequency domain basis selection information, determining the spatial basis matrix according to the spatial oversampling factor, index rearrangement, recovering the precoding matrix according to the spatial basis matrix, the frequency domain basis matrix and the coefficient matrix, and therefore cannot obtain the corresponding coefficient matrix according to the recovered CSI.

In the channel feedback information transmission method provided in the embodiment of the present application, when reducing the dimension of the measured coefficient matrix, all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix can be used, thereby eliminating the need to perform spatial basis and/or frequency domain basis selection operations. In this way, even for terminals that do not support codebook characteristics, channel feedback information can be transmitted, and both accuracy and transmission overhead can be taken into account when transmitting channel feedback information.

The embodiments of the present application are described in detail below with reference to the accompanying drawings.

The technical solutions provided in this application can be applied to various communication systems, such as fifth-generation (5G) or NR systems, LTE systems, LTE FDD systems, LTE TDD systems, wireless local area networks (WLAN) systems, satellite communication systems, future communication systems, or integrated systems of multiple systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems.

A device in a communication system can send signals to or receive signals from another device. These signals may include information, signaling, or data. The term "device" can also be replaced by an entity, network entity, network element, communication device, communication module, node, or communication node. This application uses devices as an example for description. For example, a communication system may include at least one terminal and at least one network device. A network device can send downlink signals to a terminal, and/or a terminal can send uplink signals to a network device.

Figure 6 is a schematic diagram of the architecture of a communication system 1000 used in an embodiment of the present application. As shown in Figure 6, the communication system includes an access network 100 and a core network 200. Optionally, the communication system 1000 may also include the Internet 300. The access network 100 may include at least one access network device (such as 110a and 110b in Figure 6) and at least one terminal (such as 120a-120j in Figure 6). The terminal is wirelessly connected to the access network device, and the access network device is wirelessly or wiredly connected to the core network. The core network device and the access network device may be independent, distinct physical devices, or the functions of the core network device and the logical functions of the access network device may be integrated into the same physical device, or a single physical device may integrate some of the functions of the core network device and some of the functions of the access network device. Terminals and access network devices may be interconnected via wired or wireless connections. Figure 6 is merely a schematic diagram. The communication system may also include other network devices, such as wireless relay devices and wireless backhaul devices, which are not shown in Figure 6.

The network device in the embodiment of the present application may include a device for communicating with the terminal device, and the network device may include an access network device or a wireless access network device, such as the network device may be a base station. The access network device in the embodiment of the present application may refer to a wireless access network (radio access network, RAN) node (or device) that connects the terminal device to the wireless network. The base station can broadly cover the following various names, or be replaced with the following names, such as: base station (BS), node B (NodeB), evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmission point (TRP), transmitting point (TP), main station, auxiliary station, multi-standard wireless (motor slide retainer, MSR) node, home station, etc. A RAN node may include a base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, etc. A base station may be a macro base station, micro base station, relay node, donor node, or similar, or a combination thereof. A base station may also refer to a communication module, modem, or chip used within the aforementioned devices or apparatuses. A base station may also be a mobile switching center, a device that performs base station functions in D2D, V2X, and M2M communications, or a device that performs base station functions in future communication systems. A base station may support networks with the same or different access technologies. Optionally, a RAN node may also be a server, wearable device, vehicle, or vehicle-mounted device. For example, the access network device in vehicle-to-everything (V2X) technology may be a roadside unit (RSU). The embodiments of this application do not limit the specific technology and device form used by the network device.

In some deployments, the network devices mentioned in the embodiments of the present application may include a CU, a DU, or both a CU and a DU, or a device including a control plane CU node (central unit-control plane (CU-CP)), a user plane CU node (central unit-user plane (CU-UP)), and a DU node. For example, the network devices may include a gNB-CU-CP, a gNB-CU-UP, and a gNB-DU.

In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes implementing portions of the base station's functionality. For example, a RAN node can be a CU, DU, CU-CP, CU-UP, or RU. The CU and DU can be separate or included in the same network element, such as the BBU. The RU can be included in a radio frequency device or radio unit, such as an RRU, AAU, or RRH.

A RAN node may support one or more types of fronthaul interfaces, with different fronthaul interfaces corresponding to DUs and RUs with different functionalities. If the fronthaul interface between the DU and RU is a common public radio interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and RU is another type of interface, compared to CPRI, some downlink and/or uplink baseband functions, such as precoding, digital beamforming (BF), or inverse fast Fourier transform (IFFT)/cyclic prefix (CP) for downlink, are moved from the DU to the RU. For uplink, digital beamforming (BF), or one or more fast Fourier transform (FFT)/cyclic prefix (CP) removal are moved from the DU to the RU. In one possible implementation, the interface can be an enhanced common public radio interface (eCPRI). In the eCPRI architecture, the division between the DU and RU is different, corresponding to different types (Categories) of eCPRI, such as eCPRI Category A, B, C, D, E, and F.

Taking eCPRI Cat A as an example, for downlink transmission, based on layer mapping, the DU is configured to implement layer mapping and one or more functions before it (i.e., one or more of coding, rate matching, scrambling, modulation, and layer mapping), while other functions after layer mapping (for example, one or more of resource element (RE) mapping, digital beamforming (BF), or inverse fast Fourier transform (IFFT)/adding a cyclic prefix (CP)) are moved to the RU for implementation. For uplink transmission, the DU is configured to perform demapping and one or more of the preceding functions (i.e., decoding, rate matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and demapping), with demapping being the key division. Other functions after demapping (e.g., one or more of digital BF or fast Fourier transform (FFT)/CP removal) are implemented in the RU. For a functional description of the DU and RU corresponding to various types of eCPRI, please refer to the eCPRI protocol and will not be elaborated on here.

In one possible design, the processing unit used to implement baseband functions in the BBU is called a baseband high (BBH) unit, and the processing unit used to implement baseband functions in the RRU/AAU/RRH is called a baseband low (BBL) unit.

In different systems, CU (or CU-CP and CU-UP), DU or RU may have different names, but those skilled in the art will understand their meanings. For example, in an open radio access network (open RAN, ORAN/O-RAN) system, CU may also be called O-CU (open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, CU-UP may also be called O-CU-UP, and RU may also be called O-RU. Any of the CU (or CU-CP, CU-UP), DU and RU in this application may be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

In the embodiments of the present application, the device for implementing the functions of the network device can be a network device; it can also be a device that can support the network device to implement the functions, such as a chip system, a hardware circuit, a software module, or a hardware circuit and a software module. The device can be installed in the network device or used in conjunction with the network device. In the embodiments of the present application, only the device for implementing the functions of the network device is used as an example to illustrate, and does not constitute a limitation on the solutions of the embodiments of the present application.

A terminal may also be called terminal equipment, user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication equipment, user equipment, or user device. It can be a device with wireless transceiver capabilities.

A terminal can be a device that provides voice/data, for example, a handheld device with wireless connection function, a vehicle-mounted device, etc. At present, some examples of terminals are: mobile phones, tablet computers, laptops, PDAs, mobile internet devices (MID), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart Wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, wearable devices, terminal devices in 5G networks or terminal devices in future evolved public land mobile networks (PLMNs), etc., are not limited to these in the embodiments of the present application.

As an example and not a limitation, in the embodiments of the present application, the terminal may also be a wearable device. Wearable devices may also be called wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include full-featured, large-sized, and independent of smartphones to achieve complete or partial functions, such as smart watches or smart glasses, as well as devices that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

In the embodiments of the present application, the apparatus for realizing the functions of the terminal may be a terminal device, or may be an apparatus capable of supporting the terminal in realizing the functions, such as a chip system, which may be installed in the terminal device or used in conjunction with the terminal device. In the embodiments of the present application, the chip system may be composed of a chip, or may include a chip and other discrete devices. In the embodiments of the present application, only the terminal device is used as an example for explanation, and the embodiments of the present application are not limited to the solutions of the embodiments of the present application.

The terminal can be fixed or movable. For example, the terminal can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; it can also be deployed on the water surface (such as ships, etc.); it can also be deployed in the air (such as airplanes, balloons and artificial satellites). The base station can be fixed or mobile. For example, a helicopter or a drone can be configured to act as a mobile base station, and one or more cells can move according to the location of the mobile base station. In other examples, a helicopter or a drone can be configured to be used as a device for communicating with another base station. The embodiments of the present application do not limit the application scenarios of network devices and terminals.

The roles of network devices and terminals can be relative. For example, the helicopter or drone 120i in Figure 6 can be configured as a mobile network device. With respect to the terminal 120j accessing the wireless access network 100 via 120i, drone 120i is a network device. However, with respect to network device 110a, 120i is a terminal, meaning that communication between 110a and 120i occurs via a wireless air interface protocol. Of course, communication between 110a and 120i can also occur via an interface protocol between network devices. In this case, 120i is also a network device relative to 110a. Therefore, both network devices and terminals can be collectively referred to as communication devices. 110a and 110b in Figure 6 can be referred to as communication devices with network device functionality, while 120a-120j in Figure 6 can be referred to as communication devices with terminal functionality.

Terminals and network devices can be hardware devices, or software functions running on dedicated hardware, software functions running on general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities including dedicated or general-purpose hardware devices and software functions. This application does not limit the specific form of terminals and network devices.

It should be understood that the number and type of each device in the communication system shown in Figure 6 are for illustration only, and the present application is not limited to this. In actual applications, the communication system may also include more terminals, more access network devices, and other network elements, such as core network equipment, network management and/or network elements for implementing artificial intelligence functions.

Based on the system architecture shown in Figure 6, Figure 7 shows a simplified communication system architecture applicable to an embodiment of the present application. As shown in Figure 7, the communication system includes a network device 110 and at least one terminal, such as the terminal 120 and the terminal 130 shown in Figure 7. The network device 110 and the terminal (such as the terminal 120 and the terminal 130) can communicate via a wireless link. The communication devices in the communication system, for example, the network device 110 and the terminal 120, can communicate via multi-antenna technology.

Optionally, an AI module may be configured in the network device 110. Optionally, an AI module may also be configured in the terminal 120 and/or the terminal 130. The AI module is used to perform AI-related operations, such as constructing a training data set, training an AI model, or monitoring an AI model.

Based on the system architecture shown in Figure 6, Figure 8 shows another simplified communication system architecture diagram applicable to the embodiment of the present application. Compared with the communication system shown in Figure 7, the communication system shown in Figure 8 also includes an AI network element 140. The AI network element 140 is used to perform AI-related operations, such as building a training data set or training an AI model or AI model monitoring. In the embodiment of the present application, the AI network element can also be referred to as an AI node, or an AI entity, or an AI module, which is not limited in this application.

In one possible implementation, the network device 110 may send data related to the training of the AI model to the AI network element 140, which constructs a training data set and trains the AI model. For example, the data related to the training of the AI model may include data reported by the terminal. The AI network element 140 may send the results of operations related to the AI model to the network device 110, and forward them to the terminal through the network device 110. For example, the results of operations related to the AI model may include at least one of the following: an AI model that has completed training, an evaluation result or a test result of the model, etc. Exemplarily, a portion of the trained AI model may be deployed on the network device 110, and another portion may be deployed on the terminal. Alternatively, the trained AI model may be deployed on the network device 110. Alternatively, the trained AI model may be deployed on the terminal.

It should be understood that FIG8 illustrates only the example of a direct connection between AI network element 140 and network device 110. In other scenarios, AI network element 140 may also be connected to a terminal. Alternatively, AI network element 140 may be connected to both network device 110 and a terminal simultaneously. Alternatively, AI network element 140 may be connected to network device 110 through a third-party network element. This embodiment of the present application does not limit the connection relationship between the AI network element and other network elements.

It should be noted that Figures 7 and 8 are simplified schematic diagrams for ease of understanding. For example, the communication system may also include other devices, such as wireless relay devices and/or wireless backhaul devices, which are not shown in Figures 7 and 8. In actual applications, the communication system may include multiple network devices and multiple terminals. The embodiments of the present application do not limit the number of network devices and terminals included in the communication system.

In some other communication systems provided in embodiments of the present application, an AI network element can be deployed in one or more of the following locations in the communication system: access network equipment, terminals, or core network equipment. The AI network element can also be deployed independently, for example, in a location other than any of the aforementioned devices, such as a host or cloud server in an over-the-top (OTT) system. The AI network element can communicate with other devices in the communication system, such as one or more of the following: network equipment, terminals, or core network elements.

It can be understood that AI network elements can be independent devices, or they can be integrated into the same device to implement different functions, or they can be network elements in hardware devices, or they can be software functions running on dedicated hardware, or they can be virtualized functions instantiated on a platform (for example, a cloud platform). The embodiments of this application do not limit the specific form of the above-mentioned AI network elements.

It is understood that the embodiments of the present application do not limit the number of AI network elements. For example, when there are multiple AI network elements, the multiple AI network elements can be divided based on function, for example, different AI network elements are responsible for different functions.

Figure 9 is a schematic diagram of a possible application framework in a communication system provided by an embodiment of the present application. As shown in Figure 9, network elements in the communication system are connected through interfaces (e.g., NG, Xn) or air interfaces. These network elements, such as core network equipment, access network equipment (e.g., RAN nodes), terminals, or one or more devices in operations, administration, and management (OAM) are provided with one or more AI modules (for clarity, only one is shown in Figure 9). The access network equipment can serve as a separate RAN node or include multiple RAN nodes, for example, including CU and DU. The CU and/or DU can also be provided with one or more AI modules. Optionally, the CU can also be split into CU-CP and CU-UP. One or more AI modules are provided in the CU-CP and/or CU-UP.

The AI module is used to implement the corresponding AI function. The AI modules deployed in different network elements may be the same or different. The model of the AI module can implement different functions according to different parameter configurations. The model of the AI module can be configured based on one or more of the following parameters: structural parameters (such as the number of neural network layers, the width of the neural network, the connection relationship between layers, the weight of the neuron, the activation function of the neuron, or at least one of the bias in the activation function), input parameters (such as the type of input parameters and/or the dimension of the input parameters), or output parameters (such as the type of output parameters and/or the dimension of the output parameters). Among them, the bias in the activation function can also be called the bias of the neural network.

An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or on the same node or device.

The network device may be a network device equipped with one or more AI modules. The network device may be a core network device, an access network device (RAN node), or one or more OAM devices as shown in FIG9 . For example, the AI module may be a RAN intelligent controller (RIC). The AI module may obtain subsets from multiple terminals from a RAN node (e.g., a CU, CU-CP, CU-UP, DU, and/or RU), reorganize them into training dataset #2, and perform training based on training dataset #2.

FIG10 is a schematic diagram of a possible application framework in a communication system. As shown in FIG10 , the communication system includes an RIC. For example, the RIC may be the AI module shown in FIG9 , which is used to implement AI-related functions. The RIC includes a near-real-time RIC (near-real time RIC, near-RT RIC) and a non-real-time RIC (non-real time RIC, Non-RT RIC). Among them, the non-real-time RIC mainly processes non-real-time information, such as data that is not sensitive to time delay, and the delay of the data can be in the order of seconds. The real-time RIC mainly processes near-real-time information, such as data that is relatively sensitive to time delay, and the delay of the data is in the order of tens of milliseconds.

The near real-time RIC is used for model training and reasoning. For example, it is used to train an AI model and use the AI model for reasoning. The near real-time RIC can obtain network-side and/or terminal-side information from a RAN node (e.g., CU, CU-CP, CU-UP, DU, and/or RU) and/or a terminal. This information can be used as training data or reasoning data. Optionally, the near real-time RIC can deliver the reasoning result to the RAN node and/or the terminal. Optionally, the reasoning result can be exchanged between the CU and the DU, and/or between the DU and the RU. For example, the near real-time RIC delivers the reasoning result to the DU, and the DU sends it to the RU.

The non-real-time RIC is also used for model training and reasoning. For example, it is used to train an AI model and use the model for reasoning. The non-real-time RIC can obtain network-side and/or terminal-side information from RAN nodes (such as CU, CU-CP, CU-UP, DU and/or RU) and/or terminals. This information can be used as training data or reasoning data, and the reasoning results can be submitted to the RAN node and/or terminal. Optionally, the reasoning results can be exchanged between the CU and the DU, and/or between the DU and the RU. For example, the non-real-time RIC submits the reasoning results to the DU, and the DU sends it to the RU.

The near real-time RIC and non-real-time RIC may also be separately configured as a network element. Optionally, the near real-time RIC and non-real-time RIC may also be part of other devices. For example, the near real-time RIC is configured in a RAN node (e.g., a CU and/or DU), while the non-real-time RIC is configured in an OAM, a cloud server, a core network device, or other network device.

Using the channel feedback information transmission method provided in the embodiments of the present application, terminals that support matrix multiplication and scalar quantization capabilities can transmit channel feedback information. Compared to the current codebook-based transmission method, the channel feedback information transmission method provided in the embodiments of the present application does not require one or more of the following operations: basis selection (including spatial basis selection and/or frequency basis selection), oversampling factor selection, and index permutation, thereby reducing the terminal capability requirements for channel feedback information transmission. In addition, compared to the scalar quantization method, the channel feedback information transmission method provided in the embodiments of the present application can save the transmission overhead of channel feedback information.

The following describes the channel feedback information transmission method provided in the embodiments of the present application. For details, please refer to the following channel feedback information transmission method 1 to channel feedback information transmission method 7.

(1) Channel feedback information transmission method 1

When channel feedback information transmission mode 1 is adopted, the spatial basis used to determine the coefficient matrix is all vectors in the spatial basis matrix, or the full set of the spatial basis matrix. Taking the R16 codebook as an example, in the embodiment of the present application, a spatial basis matrix of dimension _Np × _Np is used to determine the coefficient matrix. Accordingly, _the relevant parameters for the spatial basis selection operation are L = _21Np = _N1N2 . Among them, _N1 and _N2 are configured by the network device, and _the spatial basis matrix is configured, preconfigured, or predefined by the network device. Therefore, there is no need to perform the spatial basis selection operation, and accordingly, there is no need to transmit the spatial basis selection information.

When channel feedback information transmission mode 1 is used, the frequency domain basis used to determine the coefficient matrix is all vectors in the frequency domain basis matrix, or in other words, the complete set of the frequency domain basis matrix. Taking the R16 codebook as an example, in the embodiment of the present application, a frequency domain basis matrix of dimension _N3 × _N3 is used to determine the coefficient matrix. Accordingly, the relevant parameter _Mυ = _N3 for the frequency domain basis selection operation, _N3 is configured by the network device, and the frequency domain basis matrix is configured, preconfigured, or predefined by the network device. Therefore, there is no need to perform the frequency domain basis selection operation, and accordingly, there is no need to transmit the frequency domain basis selection information.

When channel feedback information transmission mode 1 is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is configured, preconfigured, or predefined by the network device. Therefore, there is no need to perform a spatial oversampling factor selection operation, and accordingly, there is no need to transmit the spatial oversampling factor selection information. Taking the R16 codebook as an example, _q1 and _q2 used to determine the spatial basis are determined based on the spatial oversampling factor selection information. In the embodiment of the present application, _q1 and _q2 can use default values or values configured by the network device, such as _q1 = 0 and _q2 = 0. In other words, _q1 and _q2 do not need to be represented by the spatial oversampling factor selection information, and therefore, there is no need to perform a spatial oversampling factor selection operation.

That is, when channel feedback information transmission mode 1 is adopted, the channel feedback information transmitter selects non-zero coefficients and performs quantization, but does not select a spatial basis, select a frequency basis, or determine spatial oversampling factor selection information. Accordingly, the channel feedback information includes non-zero coefficient selection information and non-zero coefficient quantization information, but does not include spatial basis selection information, spatial oversampling factor selection information, or frequency basis selection information.

The coefficient matrix is usually relatively sparse, with some elements having relatively small values or being 0. In order to reduce feedback, the terminal may quantize the elements with larger values (called non-zero coefficients) and send them to the network device. That is, the non-zero coefficients in the channel feedback information may be all or part of the coefficients in the coefficient matrix that are not 0 (or have large values). Optionally, the non-zero coefficients may include amplitude coefficients and/or phase coefficients, and the amplitude coefficients may include broadband amplitude coefficients and/or sub-band amplitude coefficients. Optionally, the quantization method used for the non-zero coefficients may include float32, float16, Nbit, etc., which is not limited in this application.

The non-zero coefficient selection information is used to indicate the position of the non-zero coefficient in the coefficient matrix. Optionally, the representation method of the non-zero coefficient selection information can be the same as the representation method of the non-zero coefficient selection information in the CSI feedback method based on the R16 codebook. For example, the non-zero coefficient selection information can be a bitmap whose length (i.e., the number of bits) is the same as the number of elements in the coefficient matrix. Each bit indicates a coefficient in the coefficient matrix. Different bit values can indicate whether the coefficient at the corresponding position is fed back as a non-zero coefficient, or in other words, the position of each non-zero coefficient in the channel feedback information in the coefficient matrix can be determined based on the bitmap. Taking the R16 codebook as an example, when channel feedback information transmission method 1 is adopted, the dimension of the coefficient matrix is _Np × _N3 . Therefore, the length of the above bitmap is _Np × _N3 = _{2N1N2N3 .} The coefficients corresponding to the bits with a value of 1 in the bitmap are non- _zero coefficients in the channel feedback information, and the coefficients corresponding to the bits with a value of 0 in the bitmap are non-zero coefficients in the channel feedback information.

It should be understood that the non-zero coefficient selection information may also be represented in other ways, which is not limited in this application.

In one possible implementation, taking the R16 codebook as an example, when channel feedback information transmission mode 1 is adopted, the spatial basis used to determine the coefficient matrix is the full set of spatial basis matrices, that is, L = N ₁ N ₂ , and the frequency domain basis used to determine the coefficient matrix is the full set of frequency domain basis matrices, that is, M _υ = N ₃ . Based on the above formula (7), the l-th layer codebook can be the following formula (8):

Where l represents the number of layers, l = 1, 2, 3, 4. _{γ t,l} is the power normalization factor,

for When i ranges from 0 to N ₁ N ₂ -1, and Get all the combinations of {0,1,…N ₁ -1} and {0,1,…N ₂ -1} respectively, for example in, is the floor operator, It means adjusting the real number x to the nearest integer not greater than x; mod() is the remainder operation, and mod(x,y) represents the remainder of x divided by y.

When f ranges from 0 to N ₃ -1, Get all the values of {0,1,…N ₃ -1}, for example

In another possible implementation, still taking the R16 codebook as an example, the amplitude coefficient does not distinguish between the full band and the subband. That is, the channel feedback information includes the subband amplitude coefficient but does not include the width amplitude coefficient. Then the l-th layer codebook can be simplified to the following formula (9) based on formula (8):

In another possible implementation, still taking the R16 codebook as an example, the coefficients do not distinguish between amplitude and phase. That is, the amplitude and phase of the coefficients can be uniformly quantized, for example, by float32 or float16 or N-bit quantization. Then, the l-th layer codebook can be simplified to the following formula (10) based on formula (9):

in, Information is represented by quantized coefficients.

Based on the implementation principle of the above-mentioned channel feedback information transmission method 1, the terminal can implement the channel feedback information transmission method 1 through coefficient projection and coefficient quantization. That is to say, a terminal that supports matrix multiplication and scalar quantization can implement the above-mentioned channel feedback information transmission method 1. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to perform operations such as basis selection (including spatial basis and frequency domain basis), spatial oversampling factor selection and index rearrangement, thereby reducing the channel feedback information transmission requirements for terminal capabilities. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

(2) Channel Feedback Information Transmission Method 2

When the second channel feedback information transmission method is adopted, the spatial basis used to determine the coefficient matrix is all vectors in the spatial basis matrix, or the full set of the spatial basis matrix. Therefore, there is no need to perform the spatial basis selection operation, and accordingly there is no need to transmit the spatial basis selection information.

When the second channel feedback information transmission method is adopted, the frequency domain basis used to determine the coefficient matrix is all vectors in the frequency domain basis matrix, or the full set of the frequency domain basis matrix. Therefore, there is no need to perform the frequency domain basis selection operation, and accordingly, there is no need to transmit the frequency domain basis selection information.

When the second channel feedback information transmission mode is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is determined by the transmitter of the channel feedback information. Therefore, the channel feedback information includes the spatial oversampling factor selection information.

That is, when channel feedback information transmission mode 2 is adopted, the channel feedback information transmitter selects non-zero coefficients and performs quantization, and determines spatial oversampling factor selection information, but does not select a spatial basis or a frequency domain basis. Accordingly, the channel feedback information includes non-zero coefficient selection information, non-zero coefficient quantization information, and spatial oversampling factor selection information, but does not include spatial basis selection information or frequency domain basis selection information.

The spatial domain oversampling factor selection information is used to determine the spatial domain basis. Taking the first layer codebook of the R16 codebook shown in formula (7) as an example, the spatial domain oversampling factor selection information can indicate q ₁ and q _2. According to q ₁ and q ₂ , That is to determine the airspace basis.

In one possible implementation, the spatial oversampling factor selection information may include one or more parameters, which may indicate or determine q ₁ and q ₂ , thereby determining the spatial basis. Taking the R16 codebook as an example, the spatial oversampling factor selection information in the current protocol includes the following parameter i _1,1 , i.e., q ₁ and q ₂ may be jointly indicated by the parameter i _1,1 . For example, if the value ranges of q ₁ and q ₂ are both 0, 1, 2, and 3, q ₁ and q ₂ may each be represented by 2 bits, i _1,1 being 4 bits, with the first two bits representing q ₁ and the last two bits representing q ₂ .

The method for expressing the non-zero coefficient selection information and the quantization method of the non-zero coefficients can refer to the relevant content in the first channel feedback information transmission method, and will not be repeated here.

Based on the implementation principle of the above-mentioned channel feedback information transmission method 2, the terminal can implement the channel feedback information transmission method 2 through coefficient projection and coefficient quantization. That is to say, the terminal that supports matrix multiplication and scalar quantization can implement the above-mentioned channel feedback information transmission method 2. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to perform operations such as basis selection (including spatial basis and frequency domain basis) and index rearrangement, thereby reducing the requirements of the terminal capability for channel feedback information transmission. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

(3) Channel feedback information transmission method 3

When channel feedback information transmission mode 3 is used, the spatial basis used to determine the coefficient matrix is a portion of the vectors in the spatial basis matrix, or a subset of the spatial basis matrix. Taking the R16 codebook as an example, in the embodiment of the present application, 2L vectors can be selected from the spatial basis matrix of dimension _Np × _Np to determine the coefficient matrix. That is, L < N ₁ N ₂ . N ₁ and N ₂ are configured by the network device, and the spatial basis matrix is configured, preconfigured, or predefined by the network device. Accordingly, the channel feedback information includes spatial basis selection information, which indicates 2L spatial basis selection information selected from the spatial basis matrix.

When the third channel feedback information transmission method is adopted, the frequency domain basis used to determine the coefficient matrix is all vectors in the frequency domain basis matrix, or the full set of the frequency domain basis matrix. Therefore, there is no need to perform the frequency domain basis selection operation, and accordingly, there is no need to transmit the frequency domain basis selection information.

When channel feedback information transmission method three is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is configured, preconfigured or predefined by the network device, so there is no need to perform the spatial oversampling factor selection operation, and accordingly there is no need to transmit the spatial oversampling factor selection information.

That is, when channel feedback information transmission mode 3 is adopted, the channel feedback information transmitter selects non-zero coefficients and performs quantization, selects a spatial basis, and does not select a frequency basis or determine spatial oversampling factor selection information. Accordingly, the channel feedback information includes non-zero coefficient selection information, non-zero coefficient quantization information, and spatial basis selection information, but does not include frequency basis selection information or spatial oversampling factor selection information.

The spatial basis selection information indicates the spatial basis selected from the spatial basis matrix. Taking the first layer codebook of the R16 codebook shown in formula (7) as an example, the spatial basis selection information may indicate and pass and You can determine the corresponding That is, determining or indicating the airspace basis.

In a possible implementation, the spatial basis selection information may include one or more parameters, which may indicate or determine and Then, the spatial basis can be indicated or determined. Taking the R16 codebook as an example, the spatial basis selection information in the current protocol includes the following parameters: i _1,2 , that is, the parameters i _1,2 can be used to jointly indicate and For example, Where C(dx,dy) is predefined and can be obtained based on the values of dx and dy. Where dx = N ₁ N ₂ -1-n ⁽ⁱ⁾ and dy = Li.

The method for expressing the non-zero coefficient selection information and the quantization method of the non-zero coefficients can refer to the relevant content in the first channel feedback information transmission method, and will not be repeated here.

In one possible implementation, taking the R16 codebook as an example, when channel feedback information transmission mode 3 is adopted, the spatial basis used to determine the coefficient matrix is a subset of the spatial basis matrix, that is, L<N ₁ N ₂ , and the frequency domain basis used to determine the coefficient matrix is the full set of the frequency domain basis, that is, M _υ =N ₃ . Based on the above formula (7), the l-th layer codebook can be the following formula (11):

In another possible implementation, the amplitude coefficient may not distinguish between the full band and the sub-band. That is, the channel feedback information includes the sub-band amplitude coefficient but does not include the width amplitude coefficient. This can further simplify the above formula (11). For the specific implementation, please refer to the relevant content in the first channel feedback information transmission method.

In another possible implementation, the coefficients do not distinguish between amplitude and phase, that is, the amplitude and phase of the coefficients can be uniformly quantized, which can further simplify the above formula (11). For specific implementation, please refer to the relevant content in the first channel feedback information transmission method.

Based on the implementation principle of the above-mentioned channel feedback information transmission method three, the terminal can implement the channel feedback information transmission method three through coefficient projection, coefficient quantization, and spatial basis selection. In other words, a terminal that supports matrix multiplication, scalar quantization, and spatial basis selection can implement the above-mentioned channel feedback information transmission method three. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to perform operations such as frequency domain basis selection and index permutation, thereby reducing the requirements for terminal capabilities for channel feedback information transmission. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

(IV) Channel Feedback Information Transmission Method 4

When channel feedback information transmission method 4 is adopted, the spatial basis used to determine the coefficient matrix is all vectors in the spatial basis matrix, or the full set of the spatial basis matrix. Therefore, there is no need to perform the spatial basis selection operation, and accordingly there is no need to transmit the spatial basis selection information.

When channel feedback information transmission mode 4 is used, the frequency domain basis used to determine the coefficient matrix is a portion of the vectors in the frequency domain basis matrix, or in other words, a subset of the frequency domain basis matrix. Taking the R16 codebook as an example, in the embodiment of the present application, M _v vectors can be selected from the frequency domain basis matrix of dimension N ₃ ×N ₃ to determine the coefficient matrix, where M _v <N _3. Accordingly, the channel feedback information includes frequency domain basis selection information, which indicates the M _v frequency domain basis selected from the frequency domain basis matrix.

When channel feedback information transmission mode 4 is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is configured, preconfigured or predefined by the network device, so there is no need to perform the spatial oversampling factor selection operation, and accordingly there is no need to transmit the spatial oversampling factor selection information.

That is, when channel feedback information transmission mode 4 is adopted, the channel feedback information transmitter selects non-zero coefficients and performs quantization, selects a frequency domain basis, and does not select a spatial domain basis or determine spatial oversampling factor selection information. Accordingly, the channel feedback information includes non-zero coefficient selection information, non-zero coefficient quantization information, and frequency domain basis selection information, but does not include spatial domain basis selection information or spatial oversampling factor selection information.

The frequency domain basis selection information indicates the frequency domain basis selected from the frequency domain basis matrix. Taking the first layer codebook of the R16 codebook shown in formula (7) as an example, the frequency domain basis selection information can indicate pass The corresponding frequency domain basis vector can be determined

The method for expressing the non-zero coefficient selection information and the quantization method of the non-zero coefficients can refer to the relevant content in the first channel feedback information transmission method, and will not be repeated here.

In one possible implementation, taking the R16 codebook as an example, when channel feedback information transmission mode 4 is adopted, the spatial basis used to determine the coefficient matrix is the full set of spatial basis matrices, that is, L = N ₁ N ₂ , and the frequency domain basis used to determine the coefficient matrix is a subset of the frequency domain basis, that is, M _υ <N ₃ . Based on the above formula (7), the l-th layer codebook can be the following formula (12):

In another possible implementation, the amplitude coefficient may not distinguish between the full band and the sub-band. That is, the channel feedback information includes the sub-band amplitude coefficient but does not include the width amplitude coefficient. This can further simplify the above formula (11). For the specific implementation, please refer to the relevant content in the first channel feedback information transmission method.

In another possible implementation, the coefficients do not distinguish between amplitude and phase, that is, the amplitude and phase of the coefficients can be uniformly quantized, which can further simplify the above formula (11). For specific implementation, please refer to the relevant content in the first channel feedback information transmission method.

Based on the implementation principle of the above-mentioned channel feedback information transmission mode 4, the terminal can implement the channel feedback information transmission mode 4 through coefficient projection, coefficient quantization and frequency domain basis selection. That is to say, a terminal that supports matrix multiplication, scalar quantization and frequency domain basis matrix selection can implement the above-mentioned channel feedback information transmission mode 4. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to perform operations such as selecting a spatial basis and determining spatial oversampling factor selection information, thereby reducing the requirements of the terminal capability for channel feedback information transmission. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

(V) Channel Feedback Information Transmission Mode 5

When channel feedback information transmission mode 5 is used, the spatial basis used to determine the coefficient matrix is a subset of vectors in the spatial basis matrix, or a subset of the spatial basis matrix. Accordingly, the channel feedback information includes spatial basis selection information indicating 2L spatial basis selections selected from the spatial basis matrix.

When channel feedback information transmission method five is adopted, the frequency domain basis used to determine the coefficient matrix is all vectors in the frequency domain basis matrix, or the full set of the frequency domain basis matrix. Therefore, there is no need to perform the frequency domain basis selection operation, and accordingly there is no need to transmit the frequency domain basis selection information.

When channel feedback information transmission mode five is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is determined by the transmitter of the channel feedback information, so the channel feedback information includes the spatial oversampling factor selection information.

That is, when channel feedback information transmission mode 5 is adopted, the channel feedback information transmitter selects and quantizes non-zero coefficients, selects a spatial basis, and determines spatial oversampling factor selection information, but does not select a frequency domain basis. Accordingly, the channel feedback information includes non-zero coefficient selection information, non-zero coefficient quantization information, spatial basis selection information, and spatial oversampling factor selection information, but does not include frequency domain basis selection information.

The method for expressing the non-zero coefficient selection information and the quantization method of the non-zero coefficients can refer to the relevant content in the first channel feedback information transmission method, and will not be repeated here.

The representation method of the spatial basis selection information can refer to the relevant content in the third channel feedback information transmission method, which will not be repeated here.

For the representation of the spatial oversampling factor selection information, reference may be made to the relevant content in the second channel feedback information transmission method, which will not be described in detail.

Based on the implementation principle of the above-mentioned channel feedback information transmission mode 5, the terminal can implement the channel feedback information transmission mode 5 through coefficient projection, coefficient quantization, spatial basis selection, and spatial oversampling factor selection information determination. In other words, a terminal that supports matrix multiplication, scalar quantization, spatial basis matrix selection, and spatial oversampling factor selection information determination can implement the above-mentioned channel feedback information transmission mode 5. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to perform operations such as frequency domain basis selection, thereby reducing the requirements of the terminal capability for channel feedback information transmission. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

(VI) Channel Feedback Information Transmission Method 6

When channel feedback information transmission method six is adopted, the spatial basis used to determine the coefficient matrix is all vectors in the spatial basis matrix, or the full set of the spatial basis matrix. Therefore, there is no need to perform the spatial basis selection operation, and accordingly there is no need to transmit the spatial basis selection information.

When channel feedback information transmission mode 6 is used, the frequency domain basis used to determine the coefficient matrix is a partial vector in the frequency domain basis matrix, or in other words, a subset of the frequency domain basis matrix. Accordingly, the channel feedback information includes frequency domain basis selection information indicating _M frequency domain basis selected from the spatial domain basis matrix.

When channel feedback information transmission mode 6 is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is determined by the transmitter of the channel feedback information, so the channel feedback information includes the spatial oversampling factor selection information.

That is, when channel feedback information transmission mode 6 is adopted, the channel feedback information transmitter selects non-zero coefficients and performs quantization, selects a frequency domain basis, and does not select a spatial domain basis or determine spatial domain oversampling factor selection information. Accordingly, the channel feedback information includes non-zero coefficient selection information, non-zero coefficient quantization information, frequency domain basis selection information, and spatial domain oversampling factor selection information, but does not include spatial domain basis selection information.

The method for expressing the non-zero coefficient selection information and the quantization method of the non-zero coefficients can refer to the relevant content in the first channel feedback information transmission method, and will not be repeated here.

For the representation of the frequency domain basis selection information, reference may be made to the relevant content in the fourth channel feedback information transmission method, which will not be described in detail.

For the representation of the spatial oversampling factor selection information, reference may be made to the relevant content in the second channel feedback information transmission method, which will not be described in detail.

Based on the implementation principle of the above-mentioned channel feedback information transmission mode 6, the terminal can implement the channel feedback information transmission mode 6 through coefficient projection, coefficient quantization, frequency domain basis selection, and spatial domain oversampling factor selection information determination. In other words, a terminal that supports matrix multiplication, scalar quantization, frequency domain basis matrix selection, and spatial domain oversampling factor selection information determination can implement the above-mentioned channel feedback information transmission mode 6. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to perform operations such as spatial domain basis selection, thereby reducing the requirements of the terminal capability for channel feedback information transmission. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

(VII) Channel Feedback Information Transmission Method 7

When channel feedback information transmission mode 7 is used, the spatial basis used to determine the coefficient matrix is a subset of vectors in the spatial basis matrix, or a subset of the spatial basis matrix. Accordingly, the channel feedback information includes spatial basis selection information indicating 2L spatial basis selection information selected from the spatial basis matrix.

When channel feedback information transmission mode 7 is used, the frequency domain basis used to determine the coefficient matrix is a partial vector in the frequency domain basis matrix, or in other words, a subset of the frequency domain basis matrix. Accordingly, the channel feedback information includes frequency domain basis selection information indicating _M frequency domain basis selection information selected from the spatial domain basis matrix.

When channel feedback information transmission method seven is adopted, the spatial basis oversampling factor selection information used to determine the spatial basis is configured, preconfigured or predefined by the network device, so there is no need to perform the spatial oversampling factor selection operation, and accordingly there is no need to transmit the spatial oversampling factor selection information.

That is, when channel feedback information transmission mode 7 is adopted, the channel feedback information transmitter selects and quantizes non-zero coefficients, selects spatial basis vectors, and selects frequency basis vectors, but does not determine spatial oversampling factor selection information. Accordingly, the channel feedback information includes non-zero coefficient selection information, non-zero coefficient quantization information, spatial basis selection information, and frequency basis selection information, but does not include spatial oversampling factor selection information.

The method for expressing the non-zero coefficient selection information and the quantization method of the non-zero coefficients can refer to the relevant content in the first channel feedback information transmission method, and will not be repeated here.

The representation method of the spatial basis selection information can refer to the relevant content in the third channel feedback information transmission method, which will not be repeated here.

For the representation of the frequency domain basis selection information, reference may be made to the relevant content in the fourth channel feedback information transmission method, which will not be described in detail.

For the representation of the spatial oversampling factor selection information, reference may be made to the relevant content in the second channel feedback information transmission method, which will not be described in detail.

Based on the implementation principle of the above-mentioned channel feedback information transmission mode seven, the terminal can implement channel feedback information transmission mode seven through coefficient projection, coefficient quantization, spatial basis selection, and frequency domain basis selection. In other words, a terminal that supports matrix multiplication, scalar quantization, spatial basis selection, and frequency domain basis selection can implement the above-mentioned channel feedback information transmission mode seven. Compared with the current codebook-based CSI transmission method, the terminal in the embodiment of the present application does not need to determine operations such as spatial oversampling factor selection information, thereby reducing the requirements for terminal capabilities for channel feedback information transmission. In addition, compared with scalar quantization of all coefficients in the coefficient matrix and transmission of quantization information, the embodiment of the present application can save the transmission overhead of channel feedback information.

It should be understood that although the channel feedback information transmission method provided in some embodiments of the present application is described using the R16 codebook as an example, the R16 codebook can also be replaced with other codebooks, such as the R17 codebook or the R18 codebook, etc., and its implementation principle is the same as that of the R16 codebook, and the requirements for terminal capabilities can also be reduced.

The channel feedback information transmission method provided in the embodiments of the present application is described below with reference to the accompanying drawings.

Based on the system architecture shown in any of Figures 6 to 10, Figure 11 shows a flow chart of a channel feedback information transmission method provided in an embodiment of the present application. The flow is described by taking the second device sending channel feedback information to the first device as an example.

The first device may be a network-side device, for example, the first device may be a base station, and the second device may be a terminal-side device. The network-side device may be a network device, or a module (such as a chip) in a network device, or software containing network device functions (such as a control subsystem), or other devices that communicate with the network device, such as an AI network element, which is a server, such as an OTT device or a cloud server, etc., or a combination of one or more of the above. The terminal-side device may be a terminal device, or a module (such as a chip) in a terminal device, or software containing terminal device functions (such as a control subsystem), or other devices that communicate with the terminal device, such as an AI network element, which is a server, such as an OTT device or a cloud server, etc., or a combination of one or more of the above.

As shown in Figure 11, the process may include the following steps:

Step 1101: A first device sends a first signal to a second device.

The first signal is used for channel measurement, and the channel measurement result can be used for resource scheduling, etc. In one possible implementation, the first signal can be a reference signal, such as a CSI-RS or SSB, which is not limited in this application.

Step 1102: The second device obtains channel feedback information according to the first signal.

In one possible implementation, the second device measures the first signal to obtain signal measurement information, determines a coefficient matrix based on the spatial domain basis and the frequency domain basis and the signal measurement information, then selects non-zero coefficients from the coefficient matrix, quantizes the non-zero coefficients, and obtains channel feedback information.

The channel feedback information is used to indicate the quantization information and non-zero coefficient selection information of the non-zero coefficients in the coefficient matrix. The non-zero coefficient selection information is used to indicate the position of the non-zero coefficients fed back via the channel feedback information in the coefficient matrix. The representation of the non-zero coefficient selection information can refer to the relevant content in the aforementioned embodiments. The non-zero coefficients can include quantized amplitude coefficients and/or phase coefficients, and the amplitude coefficients can include wideband amplitude coefficients and/or subband amplitude coefficients. For the relevant description of the non-zero coefficients and the quantization method, refer to the relevant content in the aforementioned embodiments.

It should be understood that "channel feedback information is used to indicate quantization information and non-zero coefficient selection information of non-zero coefficients in the coefficient matrix" can also be understood as: the channel feedback information includes quantization information and non-zero coefficient selection information of the non-zero coefficients in the coefficient matrix. This application does not limit the representation method of the channel feedback information.

In one possible implementation, the number of non-zero coefficients fed back via the channel feedback information is determined by the second device. Optionally, the channel feedback information further indicates the number of non-zero coefficients fed back via the channel feedback information, or the channel feedback information further includes information indicating the number of non-zero coefficients.

In another possible implementation, the number of non-zero coefficients to be fed back via the channel feedback information is configured by the first device, or is preconfigured or predefined. In this case, the channel feedback information sent by the second device may not need to indicate the number of non-zero coefficients, or the channel feedback information may not include information indicating the number of non-zero coefficients.

In another possible implementation, the number of non-zero coefficients to be fed back via the channel feedback information is determined by the second device based on an accuracy requirement of the channel feedback information configured by the first device. A higher accuracy requirement indicates a greater number of non-zero coefficients. Optionally, the accuracy requirement of the channel feedback information may be represented by an SGCS threshold.

In one possible implementation, the quantization mode of the non-zero coefficients is determined by the second device. Optionally, the channel feedback information sent by the second device further indicates the quantization mode of the non-zero coefficients, or the channel feedback information further includes information indicating the quantization mode of the non-zero coefficients.

In another possible implementation, the quantization method of the non-zero coefficients is configured by the first device, or is preconfigured or predefined. In this case, the channel feedback information sent by the second device may not need to indicate the quantization method of the non-zero coefficients, or the channel feedback information may not include information indicating the quantization method of the non-zero coefficients.

In another possible implementation, the quantization method of the non-zero coefficients is determined by the second device according to the accuracy requirement of the channel feedback information configured by the first device.

In an embodiment of the present application, the coefficient matrix can be obtained based on the spatial basis and the frequency domain basis, wherein at least one of the spatial basis and the frequency domain basis is all the vectors in the basis matrix. Specifically, the spatial basis and the frequency domain basis used to determine the coefficient matrix satisfy the first option, the second option or the third option. The first option is: the spatial basis used to determine the coefficient matrix is all the vectors in the spatial basis matrix, and the frequency domain basis used to determine the coefficient matrix is all the vectors in the frequency domain basis matrix; the second option is: the spatial basis used to determine the coefficient matrix is part of the vectors in the spatial basis matrix, and the frequency domain basis used to determine the coefficient matrix is all the vectors in the frequency domain basis matrix; the third option is: the spatial basis used to determine the coefficient matrix is all the vectors in the spatial basis matrix, and the frequency domain basis used to determine the coefficient matrix is part of the vectors in the frequency domain basis matrix.

For example, if the second device determines the coefficient matrix using all vectors in the spatial basis matrix and all vectors in the frequency domain basis matrix, that is, the spatial basis and frequency domain basis used to determine the coefficient matrix satisfy the first option, then the second device does not need to perform spatial basis selection and frequency domain basis selection operations. Accordingly, the channel feedback information does not include spatial basis selection information and frequency domain basis selection information. For specific implementation methods, please refer to the relevant content of Channel Feedback Information Transmission Mode 1 or Channel Feedback Information Transmission Mode 2.

For another example, if the second device uses some vectors in the spatial basis matrix and all vectors in the frequency domain basis matrix to determine the coefficient matrix, that is, the spatial and frequency domain basis used to determine the coefficient matrix meet the second option, then the second device performs the spatial basis selection operation without performing the frequency domain basis selection operation. Accordingly, the channel feedback information includes spatial basis selection information but does not include frequency domain basis selection information. For specific implementation methods, please refer to the relevant content of Channel Feedback Information Transmission Mode 3 or Channel Feedback Information Transmission Mode 5.

For another example, if the second device uses all vectors in the spatial basis matrix and some vectors in the frequency domain basis matrix to determine the coefficient matrix, that is, the spatial and frequency domain basis used to determine the coefficient matrix meet the third option, then the second device performs the frequency domain basis selection operation without performing the spatial domain basis selection operation. Accordingly, the channel feedback information includes frequency domain basis selection information but does not include spatial domain basis selection information. For specific implementation methods, please refer to the relevant content of Channel Feedback Information Transmission Mode 4 or Channel Feedback Information Transmission Mode 6.

In one possible implementation, the spatial basis and frequency domain basis used to determine the coefficient matrix satisfy the fourth option, and the fourth option is: the spatial basis used to determine the coefficient matrix is a partial vector in the spatial basis matrix, and the frequency domain basis used to determine the coefficient matrix is a partial vector in the frequency domain basis matrix. If the second device uses partial vectors in the spatial basis matrix and partial vectors in the frequency domain basis matrix to determine the coefficient matrix, that is, the spatial basis and frequency domain basis used to determine the coefficient matrix satisfy the fourth option, the second device performs a spatial basis selection operation and a frequency domain basis selection operation, and accordingly, the channel feedback information includes spatial basis selection information and frequency domain basis selection information. For specific implementation methods, please refer to the relevant content of channel feedback information transmission method seven.

In one possible implementation, the second device may select a spatial basis from the spatial basis matrix based on the number of spatial basis configured by the first device. For example, taking the R16 codebook as an example, if the number of spatial basis configured by the first device is L, the second device may select 2L spatial basis from the spatial basis matrix. In another possible implementation, the second device may select a spatial basis from the spatial basis matrix based on the accuracy requirement of the channel feedback information configured by the first device. For example, the higher the accuracy requirement, the greater the number of spatial basis selected.

In a possible implementation, the second device may select a frequency domain basis from the frequency domain basis matrix according to the number of frequency domain basis configured by the first device. For example, taking the R16 codebook as an example, if the frequency domain basis selection ratio configured by the first device is p _υ , then the second device may select from the frequency domain basis matrix In another possible implementation, the second device may select a frequency domain basis from the frequency domain basis matrix according to the accuracy requirement of the channel feedback information configured by the first device. For example, the higher the accuracy requirement, the more frequency domain basis is selected.

In one possible implementation, the channel feedback information sent by the second device also includes spatial oversampling factor selection information. For example, if the spatial oversampling factor selection information is determined by the second device, the second device may send the determined spatial oversampling factor selection information to the first device, so that the first device determines the spatial basis based on the spatial oversampling factor selection information. For the representation of the spatial oversampling factor selection information, refer to the relevant content in Channel Feedback Information Transmission Mode 2, Channel Feedback Information Transmission Mode 5, or Channel Feedback Information Transmission Mode 6.

In another possible implementation, the spatial oversampling factor selection information is configured, preconfigured, or predefined by the first device. Accordingly, the second device may determine the spatial basis based on the spatial oversampling factor selection information, and may not need to send the spatial oversampling factor selection information to the first device.

Step 1103: The second device sends channel feedback information to the first device.

After receiving the channel feedback information, the second device can obtain the precoding matrix based on the channel feedback information, the spatial basis and the frequency basis. Taking the first layer of the R16 codebook as an example, the second device can obtain the precoding matrix based on the channel feedback information and the spatial basis matrix. The conjugate transposed matrix of and frequency domain basis matrix The conjugate transposed matrix of Restore the precoding matrix W ^l of the lth layer.

Based on the method shown in Figure 11 above, for the terminal to send channel feedback information to the network device, the terminal may not need to perform one or more of the following operations: spatial basis selection operation, spatial oversampling factor selection operation, frequency domain basis selection operation, and accordingly, the terminal does not need to send the corresponding selection information to the network device. Similarly, for the terminal to receive channel feedback information sent by the network device, the network device does not need to send one or more of the following information to the terminal: spatial basis selection information, spatial oversampling factor selection information, frequency domain basis selection information, and accordingly, the terminal does not need to perform the corresponding selection operation. For terminals that do not support the codebook feature, the channel feedback information transmission method provided in the embodiment of the present application can be used to take into account both accuracy and transmission overhead when transmitting channel feedback information.

Based on the process shown in Figure 11 above, in one possible implementation, the first device may also send configuration information to the second device to configure the second device to use the above-mentioned method to feedback channel feedback information. In this implementation, before the first device sends the first signal to the second device, the second device sends configuration information to the first device, and the configuration information includes first indication information, and the first indication information is used to determine the above-mentioned first option, second option, third option, or fourth option. Accordingly, in step 1102, the second device can determine the first option, second option, third option, or fourth option based on the first indication information, and determine the channel feedback information based on the first option, second option, third option, or fourth option. In step 1103, the first device can determine the precoding matrix based on the channel feedback information and the first option, second option, third option, or fourth option indicated by the first indication information. For example, if the first indication information indicates the second option, the first device can determine the precoding matrix based on the quantization information of the non-zero coefficients and the non-zero coefficient selection information in the channel feedback information, the spatial basis vector indicated by the spatial basis selection information, and all vectors in the frequency domain basis matrix; for another example, if the first indication information indicates the second option, the first device can determine the precoding matrix based on the quantization information of the non-zero coefficients and the non-zero coefficient selection information in the channel feedback information, all vectors in the spatial basis matrix, and the frequency domain basis vector indicated by the frequency domain basis selection information.

In one possible implementation, the first indication information may be an index or number of the first option, the second option, the third option, or the fourth option. For example, the first indication information is 2-bit information, and when the value of the 2-bit information is 0, it indicates the first option; when the value of the 2-bit information is 1, it indicates the second option; and when the value of the 2-bit information is 2, it indicates the third option. For another example, the first indication information is 3-bit information, and when the value of the 3-bit information is 0, it indicates the first option; when the value of the 3-bit information is 1, it indicates the second option; when the value of the 3-bit information is 2, it indicates the third option; and when the value of the 3-bit information is 3, it indicates the fourth option. This approach can reduce signaling overhead.

Exemplarily, if the first indication information indicates the first option, the second device may adopt the above-mentioned channel feedback information transmission mode one or the above-mentioned channel feedback information transmission mode two; if the first indication information indicates the second option, the second device may adopt the above-mentioned channel feedback information transmission mode three or the above-mentioned channel feedback information transmission mode five; if the first information indicates the third option, the second device may adopt the above-mentioned channel feedback information transmission mode four or the above-mentioned channel feedback information transmission mode six; if the first indication information indicates the fourth option, the second device may adopt the above-mentioned channel feedback information transmission mode seven.

Optionally, when the first information is used to indicate the index or number of the first option, the second option, the third option, or the fourth option, the configuration information may further include information for determining the number of spatial domain bases and/or the number of frequency domain bases. For example, when the first indication information indicates the second option, the configuration information also includes the number of spatial domain bases configured by the first device; for another example, when the first indication information indicates the third option, the configuration information also includes the frequency domain base selection ratio configured by the first device.

In another possible implementation, the first indication information may be a first parameter combination, a second parameter combination, a third parameter combination, or a fourth parameter combination, each parameter combination including the number of spatial bases and the frequency domain base selection ratio configured by the first device. For example, taking the R16 codebook as an example:

The first parameter combination includes: the number of spatial basis L=1/2N _p (i.e., indicating that all vectors in the spatial basis matrix are used to determine the coefficient matrix), and the frequency domain basis selection ratio p _υ =1 (i.e., indicating that all vectors in the frequency domain basis matrix are used to determine the coefficient matrix);

The second parameter combination includes: the number of spatial basis L<1/2N _p (i.e., indicating that some vectors in the spatial basis matrix are used to determine the coefficient matrix), and the frequency domain basis selection ratio p _υ =1 (i.e., indicating that all vectors in the frequency domain basis matrix are used to determine the coefficient matrix);

The third parameter combination includes: the number of spatial basis L=1/2N _p (i.e., indicating that all vectors in the spatial basis matrix are used to determine the coefficient matrix), and the frequency domain basis selection ratio p _υ <1 (i.e., indicating that some vectors in the frequency domain basis matrix are used to determine the coefficient matrix);

The fourth parameter combination includes: the number of spatial basis L<1/2N _p (i.e., indicating that the coefficient matrix is determined by using some vectors in the spatial basis matrix), and the frequency domain basis selection ratio p _υ <1 (i.e., indicating that the coefficient matrix is determined by using some vectors in the frequency domain basis matrix).

Exemplarily, if the first indication information is a first parameter combination, the second device may adopt the above-mentioned channel feedback information transmission mode one or the above-mentioned channel feedback information transmission mode two, and may determine to use all vectors in the spatial basis matrix according to the number of spatial basis L in the first parameter combination, and determine to use all vectors in the frequency domain basis matrix according to the frequency domain basis selection ratio p _υ ; if the first indication information is a second parameter combination, the above-mentioned channel feedback information transmission mode three or the above-mentioned channel feedback information transmission mode five may be adopted, and may select 2L vectors from the spatial basis matrix according to the number of spatial basis L in the first parameter combination, and determine to use all vectors in the frequency domain basis matrix according to the frequency domain basis selection ratio p _υ ; if the first indication information is a third parameter combination, the second device may adopt the above-mentioned channel feedback information transmission mode four or the above-mentioned channel feedback information transmission mode six, and may determine to use all vectors in the spatial basis matrix according to the number of spatial basis L in the first parameter combination, and select 2L vectors from the frequency domain basis matrix according to the frequency domain basis selection ratio p _υ. vectors; if the first indication information is the fourth parameter combination, the second device may adopt the above-mentioned channel feedback information transmission mode seven, and may select 2L vectors from the spatial basis matrix according to the number L of spatial basis in the first parameter combination, and select 2L vectors from the frequency domain basis matrix according to the frequency domain basis selection ratio p _υ. vectors.

Optionally, the above parameter combination may further include a non-zero coefficient selection ratio β configured by the first device, and the non-zero coefficient selection ratio β is used to determine a maximum number K ₀ of non-zero coefficients fed back for each layer. Accordingly, for each layer of the codebook, the number of non-zero coefficients selected by the second device for feedback does not exceed the maximum number K ₀ .

In another possible implementation, the first indication information may be an index of the first parameter combination, the second parameter combination, the third parameter combination, or the fourth parameter combination. The first device and the second device may preconfigure or predefine a correspondence between the parameter combinations and the indexes, and the second device may obtain the corresponding parameter combination by querying the correspondence based on the parameter combination index configured on the first device.

It should be understood that the number of spatial bases L in the second parameter combination may have multiple values, and accordingly, the number of second parameter combinations may also be multiple. Similarly, the number of third parameter combinations may also be multiple, and the number of fourth parameter combinations may also be multiple.

In a possible implementation, the configuration information sent by the first device may further include spatial oversampling factor selection information. A method for expressing the spatial oversampling factor selection information may refer to the relevant content in the first channel feedback information transmission method described above.

In one possible implementation, the second device may also report terminal capability information to the first device. Specifically, before the first device sends the configuration information to the second device, the second device sends the terminal capability information to the first device. Accordingly, the first device sends the configuration information to the second device based on the terminal capability information sent by the second device, causing the second device to send channel feedback information using the channel feedback information transmission method described above.

For example, in the embodiment of the present application, the terminal capability information may indicate one or more of the following:

Option 1: The terminal capability information indicates that the second device does not support the codebook feature, or does not support the codebook feature defined by the current protocol, such as not supporting the R16 codebook feature;

Option 2: The terminal capability information indicates that the second device does not support determining a coefficient matrix based on some vectors in the spatial basis matrix and some vectors in the frequency domain basis matrix;

Option 3: The terminal capability information indicates that the second device supports determining the coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix. In other words, the terminal capability information indicates that the second device supports one of the above-mentioned channel feedback information transmission modes one to seven. Optionally, the terminal capability information may include second indication information, and the second indication information indicates one of the above-mentioned channel feedback information transmission modes one to seven. For example, the second indication information is 3-bit information. When the 3-bit information is 0, it indicates that the second device supports the above-mentioned channel feedback information transmission mode one. When the 3-bit information is 1, it indicates that the second device supports the above-mentioned channel feedback information transmission mode two, and so on.

Optionally, the terminal capability information may also indicate that the second device supports codebook characteristics. In this case, the first device may also configure the second device to use one of the channel feedback information transmission modes 1 to 7 to feed back channel feedback information through configuration information.

A possible implementation manner in which the second device sends the terminal capability information to the first device and the first device sends the above configuration information to the second device according to the terminal capability information sent by the second device can refer to the process shown in FIG14 .

The channel feedback information transmission method provided in the embodiment of the present application can also be applied to a model monitoring scenario implemented on the network side. The following describes the process in the model monitoring scenario executed by the network side in conjunction with Figures 12a and 12b.

Based on the system architecture shown in Figure 7, Figure 9 or Figure 10, see Figure 12a, a flow chart of a model monitoring scenario performed by the network side provided in an embodiment of the present application. The process is described using a network device and a terminal as an example. The terminal includes a first AI model, and the first AI model is used to compress the channel feedback information, or to compress and quantize the channel feedback information. The network device includes a second AI model, and the second AI model can restore the received channel feedback information or the dequantized channel feedback information. The first AI model and the second AI model can refer to the architecture shown in Figure 3. The description of the network device and the terminal can refer to the aforementioned embodiment and will not be repeated here.

As shown in Figure 12a, the process includes the following steps:

Step 1201: The network device sends a first signal to the terminal.

For the specific implementation of this step, please refer to step 1101 in Figure 11.

Step 1202: The terminal sends first channel feedback information to the network device, where the first channel feedback information is obtained by the first AI model according to the first signal.

In one possible implementation, the terminal measures the first signal to obtain signal measurement information, which is then input into a first AI model to obtain first channel feedback information output by the first AI model. Alternatively, the first channel feedback information is obtained by the first AI model based on true channel feedback information. The true channel feedback information may be signal measurement information obtained directly by measurement without quantization or compression.

Step 1203: The terminal sends second channel feedback information to the network device.

In this step, the terminal determines second channel feedback information based on signal measurement information obtained by measuring the first signal, and sends the second channel feedback information to the network device. For a specific implementation of this step, see steps 1102 and 1103 in Figure 11. That is, the terminal can use one of the above-mentioned channel feedback information transmission modes 1 to 7 to send the second channel feedback information to the network device.

Step 1204: The network device recovers the first channel feedback information to obtain third channel feedback information, and performs model monitoring based on the third channel feedback information and the second channel feedback information, that is, monitors the performance of the first AI model and/or the second AI model.

In this step, on the network device side, the first channel feedback information is input into the second AI model to obtain the third channel feedback information output by the second AI model. The network device determines whether the performance of the first AI model and/or the second AI model meets the requirements by comparing the difference between the KPI indicator of the third channel feedback information and the performance KPI indicator of the second channel feedback information. Optionally, the KPI indicator may include one or more of the following: GCS, SGCS, mean square error (MSE) or NMSE, etc., which are not limited in this application.

For example, when the KPI indicator is GCS, the KPI indicator satisfies the following formula (13):

in, is the predicted channel feedback information of the i-th resource unit, which is obtained based on the third channel feedback information; w _i is the true channel feedback information of the i-th resource unit, which is obtained based on the second channel feedback information. _N is the number of resource units. _‖wi‖ represents the modulus or norm of matrix w _i .

For example, when the KPI indicator is SGCS, the KPI indicator satisfies the following formula (14):

in, is the predicted channel feedback information of the i-th resource unit, which is obtained based on the third channel feedback information; w _i is the true channel feedback information of the i-th resource unit, which is obtained based on the second channel feedback information.

For example, when the KPI indicator is MSE, the KPI indicator satisfies the following formula (15):

in, is the predicted channel feedback information of the i-th resource unit, which is obtained based on the third channel feedback information; w _i is the true channel feedback information of the i-th resource unit, which is obtained based on the second channel feedback information.

For example, when the KPI indicator is NMSE, the KPI indicator satisfies the following formula (16):

in, is the predicted channel feedback information of the i-th resource unit, which is obtained based on the third channel feedback information; w _i is the true channel feedback information of the i-th resource unit, which is obtained based on the second channel feedback information.

Optionally, the process shown in Figure 12a may also include a step in which the network device sends configuration information to the terminal. Further, it may also include a step in which the terminal reports terminal capability information to the network device. For specific implementation methods, please refer to the relevant content in the aforementioned embodiments.

The process shown in FIG12a can also be applied to a channel state prediction scenario. In this scenario, the terminal can use the channel response obtained based on historical signal measurements to obtain channel feedback information and send it to the network device to predict the channel state.

Exemplarily, in a possible channel state prediction and model monitoring scenario, a first AI model is provided on the terminal side, which is used to predict channel feedback information based on one or more second signals (the one or more second signals are no later than the first signal, or the second signal is a historical signal). The terminal sends the predicted channel feedback information (here referred to as the first channel feedback information) to the network device. The terminal device also uses the method provided in the embodiment of the present application to send second channel feedback information to the network device. The second channel feedback information is obtained based on the first signal. The second channel feedback information is the true value channel feedback information corresponding to the first channel feedback information. In other words, the second channel feedback information can be used to monitor the performance of the first AI model. On the network device side, the channel state prediction performance of the first AI can be evaluated based on the first channel feedback information and the second channel feedback information.

Exemplarily, in another possible scenario of channel state prediction and model monitoring, a first AI model is provided on the terminal side, and the terminal obtains predicted channel feedback information based on one or more second signals (the one or more second signals are no later than the first signal), and then uses the first AI model to compress and quantize the first channel feedback information. The terminal sends the compressed and quantized channel feedback information (here referred to as the first channel feedback information) to the network device. The terminal device also uses the method provided in the embodiment of the present application to send second channel feedback information to the network device. The second channel feedback information is obtained based on the first signal, and the second channel feedback information is the true value channel feedback information corresponding to the first channel feedback information. On the network device side, the first channel state information is restored by the second AI model, and the performance of the first AI is evaluated based on the restored channel state information and the second channel feedback information, or the performance of the second AI model is evaluated, or the performance of the first AI model and the second AI model is evaluated.

Exemplarily, in another possible scenario of channel state prediction and model monitoring, a first AI model is provided on the terminal side, and the terminal uses the first AI model and obtains predicted channel feedback information based on one or more second signals (the one or more second signals are no later than the first signal), and the predicted channel feedback information is compressed and quantized by the first AI model (referred to as the first channel state information here). The terminal sends the first channel feedback information to the network device. The terminal device also uses the method provided in the embodiment of the present application to send second channel feedback information to the network device. The second channel feedback information is obtained based on the first signal, and the second channel feedback information is the true value channel feedback information corresponding to the first channel feedback information. On the network device side, the first channel state information is restored by the second AI model, and the performance of the first AI is evaluated based on the restored channel state information and the second channel feedback information, or the performance of the second AI model is evaluated, or the performance of the first AI model and the second AI model is evaluated.

Optionally, an association relationship may be established between the predicted channel feedback information and the true channel feedback information obtained based on the first signal. For example, for the same channel, an association relationship may be established between the predicted channel feedback information and the true channel feedback information. For another example, an association relationship may be established between the channel feedback information predicted based on the first signal at the first time (or first time period) in history and the true channel feedback information obtained based on the current first signal.

Based on the system architecture shown in any of Figures 6 to 10, refer to Figure 12b for another flow chart of a model monitoring scenario performed by the network side provided in an embodiment of the present application. This process is described using a network device and a terminal as an example. The terminal includes a first AI model, which is used to compress the channel feedback information, or to compress and quantize the channel feedback information. The AI network element includes a second AI model, which can restore the received channel feedback information or the dequantized channel feedback information. The first AI model and the second AI model can refer to the architecture shown in Figure 3. The AI network element has a model monitoring function. The description of the network device and the terminal can refer to the aforementioned embodiment and will not be repeated here.

As shown in Figure 12b, the process includes the following steps:

Step 1211: The network device sends a first signal to the terminal.

For the specific implementation of this step, please refer to step 1101 in Figure 11.

Step 1212a: The terminal sends first channel feedback information to the network device, where the first channel feedback information is obtained by the first AI model based on the first signal.

For the specific implementation of this step, please refer to step 1202 in Figure 12a.

Step 1212b: The network device sends the received first channel feedback information to the AI network element.

Step 1213a: The terminal sends second channel feedback information to the network device.

For the specific implementation of this step, please refer to step 1203 in Figure 12a.

Step 1213b: The network device sends the second channel feedback information to the AI network element.

Step 1214: The AI network element recovers the first channel feedback information based on the second AI model to obtain third channel feedback information, and performs model monitoring based on the third channel feedback information and the second channel feedback information, that is, monitors the performance of the first AI model and/or the second AI model.

For the specific implementation of this step, please refer to step 1204 in Figure 12a.

Optionally, the process shown in Figure 12b may also include a step in which the network device sends configuration information to the terminal. Further, it may also include a step in which the terminal reports terminal capability information to the network device. For specific implementation methods, please refer to the relevant content in the aforementioned embodiments.

The process shown in Figure 12b above can also be applied to the channel state prediction scenario. In the channel state prediction scenario, the terminal can use the channel response obtained based on historical signal measurements to obtain channel feedback information and send it to the network device to achieve channel state prediction. Specifically, the implementation method in the channel prediction and model monitoring scenarios can refer to the aforementioned embodiment. The difference is that the operation of performing channel feedback information recovery on the network side can be implemented by the AI network element set on the network side.

The above-mentioned channel feedback information transmission method provided in the embodiment of the present application can also be applied to the model monitoring scenario implemented on the terminal side. The process in the model monitoring scenario executed by the terminal side is explained below in combination with Figures 13a and 13b.

Based on the system architecture shown in any of the figures in Figures 6 to 10, see Figure 13a, a flow chart of a model monitoring scenario performed by the terminal side provided in an embodiment of the present application. The process is described using a network device and a terminal as an example. The terminal includes a first AI model, which is used to compress and quantize channel feedback information. The network device includes a second AI model, which can recover the received channel feedback information. The first AI model and the second AI model can refer to the architecture shown in Figure 3. The description of the network device and the terminal can refer to the aforementioned embodiment and will not be repeated here.

As shown in Figure 13a, the process includes the following steps:

Step 1301: The network device sends a first signal to the terminal.

For the specific implementation of this step, please refer to step 1101 in Figure 11.

Step 1302: The terminal sends first channel feedback information to the network device. The first channel feedback information is obtained by the first AI model based on the first signal, or in other words, the first channel feedback information is obtained by the first AI model based on true channel feedback information. The true channel feedback information here can be signal measurement information obtained directly without quantization or compression.

For the specific implementation of this step, please refer to step 1202 in Figure 12a.

Step 1303: The network device recovers the first channel feedback information to obtain third channel feedback information.

In this step, on the network device side, the first channel feedback information is input into the second AI model to obtain the third channel feedback information output by the second AI model.

Step 1304: The network device sends the third channel feedback information to the terminal.

In this step, the network device may use one of the channel feedback information transmission modes 1 to 7 to send the third channel feedback information to the terminal.

Step 1305: The terminal performs model monitoring based on the third channel feedback information and the true channel feedback information, that is, monitors the performance of the first AI model and/or the second AI model.

For the specific implementation of model monitoring, please refer to the relevant content in step 1204 in Figure 12a.

Optionally, the process shown in Figure 13a may also include a step in which the network device sends configuration information to the terminal. Further, it may also include a step in which the terminal reports terminal capability information to the network device. For specific implementation methods, please refer to the relevant content in the aforementioned embodiments.

The process shown in FIG. 13a can also be applied to a channel state prediction scenario. In this scenario, the terminal sends first channel feedback information to the network device based on the first signal received historically. Accordingly, the network device can determine a precoding matrix based on the first channel feedback information, thereby using the first signal received historically to predict the precoding matrix.

Based on the system architecture shown in any of Figures 6 to 10, refer to Figure 13b for another flow chart of a model monitoring scenario performed by the terminal side provided in an embodiment of the present application. The process is described using a network device and a terminal as an example. The terminal includes a first AI model, which is used to compress and quantize channel feedback information. The network device includes a second AI model, which can restore the received channel feedback information. The first AI model and the second AI model can refer to the architecture shown in Figure 3. The architecture also includes an AI network element, which has a model monitoring function. The description of the network device and the terminal can refer to the aforementioned embodiment and will not be repeated here.

As shown in Figure 13b, the process includes the following steps:

Step 1311: The network device sends a first signal to the terminal.

For the specific implementation of this step, please refer to step 1101 in Figure 11.

Step 1312a: The terminal sends first channel feedback information to the network device. The first channel feedback information is obtained by the first AI model based on the first signal, or in other words, the first channel feedback information is obtained by the first AI model based on true channel feedback information. The true channel feedback information here can be signal measurement information obtained directly without quantization or compression.

For the specific implementation of this step, please refer to step 1202 in Figure 12a.

Step 1312b: The terminal sends the true value channel feedback information to the AI network element.

Optionally, the terminal may quantize the true channel feedback information before sending it to the AI network element to reduce signaling overhead.

Step 1313: The network device recovers the first channel feedback information to obtain third channel feedback information.

In this step, on the network device side, the first channel feedback information is input into the second AI model to obtain the third channel feedback information output by the second AI model.

Step 1314a: The network device sends the third channel feedback information to the terminal.

In this step, the network device may use one of the channel feedback information transmission modes 1 to 7 to send the third channel feedback information to the terminal.

Step 1314b: The terminal sends the third channel feedback information to the AI network element.

Step 1315: The AI network element performs model monitoring based on the third channel feedback information and the true channel feedback information, that is, monitors the performance of the first AI model and/or the second AI model.

For the specific implementation of model monitoring, please refer to the relevant content in step 1204 in Figure 12a.

Optionally, the process shown in Figure 13b may also include a step in which the network device sends configuration information to the terminal. Further, it may also include a step in which the terminal reports terminal capability information to the network device. For specific implementation methods, please refer to the relevant content in the aforementioned embodiments.

The process shown in FIG13b above can also be applied to a channel state prediction scenario. In this scenario, the terminal sends the first channel feedback information to the network device, which is determined by the terminal based on a first signal received historically. Accordingly, the network device can determine the precoding matrix based on the first channel feedback information, thereby using the historical first signal to predict the precoding matrix.

Based on the system architecture shown in any of Figures 6 to 10, an embodiment of the present application also provides a communication method, in which the second device can report terminal capability information to the first device, and the first device can send configuration information to the second device based on the terminal capability information to configure the transmission of channel feedback information of the second device.

Refer to Figure 14, which is a flow chart of a communication method provided in an embodiment of the present application. The process is described by taking the second device sending channel feedback information to the first device as an example. The first device may be a network side device, for example, the first device may be a base station, and the second device may be a terminal side device. The network side device may be a network device, or a module (such as a chip) in a network device, or software (such as a control subsystem) containing the functions of a network device, etc. The terminal side device may be a terminal device, or a module (such as a chip) in a terminal device, or software (such as a control subsystem) containing the functions of a terminal device, etc.

As shown in Figure 14, the process may include the following steps:

Step 1401: The second device sends terminal capability information to the first device, where the terminal capability information is used to determine a channel feedback information transmission method.

The terminal capability information may also be understood as being used to indicate a channel feedback information transmission mode supported or not supported by the second device, or being used to indicate the channel feedback information transmission capability of the second device.

In one possible implementation, the channel feedback information transmission capability may include one or more of the following:

Capability 1: Supporting codebook characteristics, or supporting codebooks and codebook parameters. The second device can transmit channel feedback information based on the codebook.

Taking support for the R16 codebook and R16 codebook parameters as an example, support for the R16 codebook means that the second device supports the transmission of channel feedback information in the format (or method) of the R16 codebook, that is, the second device supports processing the channel feedback information into the channel feedback information format corresponding to the R16 codebook. Support for the R16 codebook parameters means that the second device supports the configuration parameters of the R16 codebook, that is, the terminal supports processing the channel feedback information into the channel feedback information format corresponding to the parameter, including the number of spatial basis, the proportion of the number of frequency domain basis, the proportion of non-zero coefficients, etc. The R16 codebook parameters can be shown in Table 1.

Taking support for the R16 codebook and enhanced codebook parameters as an example, support for the R16 codebook means that the second device supports the transmission of channel feedback information in the format (or manner) of the R16 codebook, that is, the second device supports processing the channel feedback information into a channel feedback information format corresponding to the R16 codebook. Support for enhanced codebook parameters means that the second device supports processing the channel feedback information into a channel feedback information format corresponding to the enhanced codebook parameters. The enhanced codebook parameters can be shown in Table 2.

Capability 2: Supporting the channel feedback information transmission mode provided in the embodiment of the present application. For example, supporting one of the channel feedback information transmission modes 1 to 7 provided in the embodiment of the present application.

Exemplarily, the terminal capability information indicates that the second device supports determining a coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix. In other words, the terminal capability information indicates that the second device supports one of the above-mentioned channel feedback information transmission modes one to seven. Optionally, the terminal capability information may include second indication information, and the second indication information indicates one of the above-mentioned channel feedback information transmission modes one to seven. For example, the second indication information is 3-bit information. When the 3-bit information is 0, it indicates that the second device supports the above-mentioned channel feedback information transmission mode one. When the 3-bit information is 1, it indicates that the second device supports the above-mentioned channel feedback information transmission mode two, and so on.

It can be understood that the terminal capability information indicates capability 2, and it can also be understood that the terminal capability information indicates that the second device does not support the codebook characteristics, or does not support the codebook characteristics defined by the current protocol, such as not supporting the R16 codebook characteristics, or the terminal capability information indicates that the second device does not support determining the coefficient matrix based on partial vectors in the spatial basis matrix and partial vectors in the frequency domain basis matrix.

Capability 3: Support codebook reception.

Taking the R16 codebook as an example, supporting codebook reception means that the second device supports receiving channel feedback information in the R16 codebook format, that is, the second device can process the channel feedback information in the R16 codebook format into original channel feedback information (the original channel feedback information here refers to the channel feedback information in the form of a precoding matrix, not the unprocessed channel feedback information). The second device receiving the channel feedback information can be used for the second device side model monitoring scenario. For example, referring to Figure 13a, the first device sends the recovered channel feedback information to the second device, and the channel feedback information can be sent in the R16 codebook format. The second device receiving the channel feedback information can also be used in the data set transmission scenario. For example, the first device sends the data set to the second device, and the channel feedback information in the data set can be represented in the R16 codebook format. The reception of the R16 codebook has lower requirements on the terminal capability than the transmission of the R16 codebook. For example, the second device assigns the non-zero coefficients to the coefficient matrix according to their positions, and then restores the coefficient matrix to the precoding matrix according to the basis selection information. There is no need to perform operations such as basis selection. Therefore, the second device that does not support R16 codebook transmission may support R16 codebook reception.

Capability 4: Supporting CSI scalar quantization. That is, the second device supports directly performing scalar quantization on elements in the channel feedback information.

Step 1402: The first device sends configuration information to the second device, where the configuration information indicates a channel feedback information transmission mode that matches the terminal capability information.

In this step, the first device can determine the channel feedback information transmission mode corresponding to the terminal capability information sent by the second device, and configure the transmission of the channel feedback information for the second device according to the channel feedback information transmission mode, for example, configuring an appropriate channel feedback information transmission mode and transmission parameters for the second device.

Optionally, the first device may further configure the second device to transmit the channel feedback information according to at least one of an accuracy requirement and a transmission overhead requirement of the channel feedback information.

Exemplarily, if the terminal capability information indicates the above-mentioned capability 1 and capability 4, the first device can configure the second device to send channel feedback information in accordance with the R16 codebook format, and can configure the codebook parameters for the second device according to Table 1, such as paramCombination6 in Table 1 (see the codebook parameters corresponding to the parameter combination with a value of 6 in Table 1).

For example, if the terminal capability information indicates the above-mentioned capability 1 and capability 4, the first device can configure the second device to send channel feedback information in accordance with the R16 codebook format, and can configure the codebook parameters for the second device according to Table 2, such as paramCombination9 in Table 2 (see the enhanced codebook parameters corresponding to the parameter combination with a value of 9 in Table 2).

Exemplarily, if the terminal capability information indicates the above-mentioned capability 1 and capability 4, the first device may also configure the second device to send channel feedback information according to the channel feedback information transmission method provided in the embodiment of the present application.

Exemplarily, if the terminal capability information indicates the above-mentioned capability 2 and capability 4, the first device may configure the second device to send channel feedback information according to the channel feedback information transmission method provided in the embodiment of the present application.

For example, if the terminal capability information indicates that the second device supports a first channel feedback information transmission mode (for example, the above-mentioned channel feedback information transmission mode one or the above-mentioned channel feedback information transmission mode two), the configuration information includes first indication information, and the first indication information indicates a first option, which is: determining the spatial basis for non-zero coefficients as all vectors in the spatial basis matrix, and determining the frequency domain basis for non-zero coefficients as all vectors in the frequency domain basis matrix.

For another example, if the terminal capability information indicates that the second device supports a second channel feedback information transmission mode (e.g., the above-mentioned channel feedback information transmission mode 3 or the above-mentioned channel feedback information transmission mode 5), the configuration information includes first indication information, where the first indication information indicates a second option, where the second option is: a spatial basis for determining non-zero coefficients is a portion of vectors in a spatial basis matrix, and a frequency domain basis for determining non-zero coefficients is all vectors in a frequency domain basis matrix. Optionally, the first indication information further indicates the number of spatial basis, where the number of spatial basis is less than the number of all vectors in the spatial basis matrix.

For another example, if the terminal capability information indicates that the second device supports a third channel feedback information transmission mode (e.g., the channel feedback information transmission mode 4 or the channel feedback information transmission mode 6 described above), the configuration information includes first indication information, where the first indication information indicates a third option, where the third option is: the spatial basis used to determine the non-zero coefficients is all vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficients is a portion of the vectors in the frequency domain basis matrix. Optionally, the first indication information further indicates a frequency domain basis selection ratio, where the frequency domain basis selection ratio is greater than 0 and less than 1.

Optionally, the configuration information further includes spatial oversampling factor selection information, where the spatial oversampling factor selection information is used to determine the spatial basis.

Exemplarily, if the terminal capability information indicates the above-mentioned capability 3 and capability 4, the first device may configure the second device to receive channel feedback information according to the channel feedback information transmission method provided in the embodiment of the present application.

Exemplarily, if the terminal capability information indicates the above-mentioned capability 1 and capability 4, the first device may also configure the second device to receive channel feedback information according to the channel feedback information transmission method provided in the embodiment of the present application.

The contents of the configuration information can be referred to in the above embodiments and will not be described in detail.

It is understood that in order to implement the functions in the above embodiments, the network devices and terminal devices include hardware structures and/or software modules corresponding to the execution of each function. Those skilled in the art should readily appreciate that, in combination with the units and method steps of each example described in the embodiments disclosed in this application, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in hardware or in a manner driven by computer software depends on the specific application scenario and design constraints of the technical solution.

Figures 15 and 16 are schematic diagrams of the structures of possible communication devices provided in embodiments of the present application. These communication devices can be used to implement the functions of the network-side device (first device) or the terminal-side device (second device) in the above-mentioned method embodiments, and thus can also achieve the beneficial effects of the above-mentioned method embodiments. In embodiments of the present application, the communication device can be the above-mentioned device or a module (such as a chip) in the above-mentioned device.

As shown in Figure 15, the communication device 1500 includes a processing unit 1510 and a transceiver unit 1520. The communication device 1500 is used to implement the functions of the terminal side device or the network side device in the method embodiment shown in any of Figures 11, 12a, 12b, 13a or 13b.

When the communication device 1500 is used to implement the functions of the terminal-side device (second device) in the method embodiment shown in Figure 11, Figure 12a, or Figure 12b: the transceiver unit 1520 is used to receive a first signal from the first device; the processing unit 1510 is used to send channel feedback information to the first device through the transceiver unit 1520, the channel feedback information being obtained based on the first signal, the channel feedback information indicating quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, the non-zero coefficient selection information indicating the position of the non-zero coefficient, and the coefficient matrix being obtained based on a spatial basis and a frequency domain basis. The spatial basis and the frequency domain basis satisfy the first option, the second option, or the third option; the first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option is: the spatial basis is a portion of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is a portion of the vectors in the frequency domain basis matrix.

When the communication device 1500 is used to implement the functions of the network-side device (first device) in the method embodiment shown in Figure 11, Figure 12a, or Figure 12b: the transceiver unit 1520 is used to send a first signal to the second device; the transceiver unit 1520 is also used to receive channel feedback information from the second device, the channel feedback information being obtained based on the first signal, the channel feedback information indicating quantization information of non-zero coefficients in a coefficient matrix and non-zero coefficient selection information, the non-zero coefficient selection information indicating the position of the non-zero coefficient, and the channel feedback information, spatial basis, and frequency domain basis being used to determine the precoding matrix. The spatial basis and the frequency domain basis satisfy the first option, the second option, or the third option; the first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option is: the spatial basis is a portion of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is a portion of the vectors in the frequency domain basis matrix.

When the communication device 1500 is used to implement the function of the terminal side device (second device) in the method embodiment shown in Figure 13a or Figure 13b: the transceiver unit 1520 is used to receive a first signal from the first device; the processing unit 1510 is used to send first channel feedback information to the first device through the transceiver unit 1520, the first channel feedback information is obtained by a first artificial intelligence AI model according to the first signal, and the first AI model is located in the second device; the transceiver unit 1520 is used to receive third channel feedback information from the first device, the third channel feedback information is obtained by recovering the first channel feedback information by a second AI model, the second AI model is located in the first device, the third channel feedback information includes quantization information of non-zero coefficients in a coefficient matrix and non-zero coefficient selection information, the non-zero coefficient selection information indicates the position of the non-zero coefficient, and the coefficient matrix is obtained according to a spatial domain basis and a frequency domain basis. Among them, the spatial basis and the frequency domain basis satisfy the first option, the second option or the third option; the first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix.

When the communication device 1500 is used to implement the functions of the network-side device (first device) in the method embodiment shown in Figure 13a or Figure 13b: the processing unit 1510 is used to send a first signal to the first device through the transceiver unit 1520; the transceiver unit 1520 is used to receive first channel feedback information from the first device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model based on the first signal, and the first AI model is located in the second device; the processing unit 1510 is used to send third channel feedback information to the first device through the transceiver unit 1520, where the third channel feedback information is obtained by recovering the first channel feedback information by a second AI model, and the second AI model is located in the first device, and the third channel feedback information includes quantization information of non-zero coefficients in a coefficient matrix and non-zero coefficient selection information, where the non-zero coefficient selection information indicates the positions of the non-zero coefficients, and the coefficient matrix is obtained based on a spatial domain basis and a frequency domain basis. Among them, the spatial basis and the frequency domain basis satisfy the first option, the second option or the third option; the first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; the third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix.

When the communication device 1500 is used to implement the function of the terminal side device (second device) in the method embodiment shown in Figure 14: the processing unit 1510 is used to send terminal capability information to the first device through the transceiver unit 1520, and the terminal capability information is used to determine the channel feedback information transmission method; the transceiver unit 1520 is used to receive configuration information from the first device, and the configuration information is used to indicate the channel feedback information transmission method that matches the terminal capability information.

When the communication device 1500 is used to implement the function of the network side device (first device) in the method embodiment shown in Figure 14: the transceiver unit 1520 is used to receive terminal capability information from the first device, and the terminal capability information is used to determine the channel feedback information transmission method; the processing unit 1510 is used to send configuration information to the first device through the transceiver unit, and the configuration information is used to indicate the channel feedback information transmission method that matches the terminal capability information.

A more detailed description of the processing unit 1510 and the transceiver unit 1520 can be directly obtained by referring to the relevant description in the method embodiment shown in the above drawings, and will not be repeated here.

Some embodiments of the present application provide a communication device 1600 , which includes a processing circuit 1610 .

The processing circuit 1610 may include one or more processors, or all or part of the circuits in one or more processors for controlling or processing functions.

The communication device 1600 may further include a communication circuit 1620 .

When communication device 1600 is a network device or a terminal device, communication circuit 1620 may be a transceiver, a transceiver circuit, or an interface circuit. When communication device 1600 is a chip for a network device or a terminal device, communication circuit 1620 may be a transceiver circuit or an interface circuit. When communication device 1600 is a server, communication circuit 1620 may be a transceiver circuit or an interface circuit.

Optionally, the communication device 1600 may further include a memory 1630. The memory 1630 is used to store instructions executed by the processor, or to store input data required by the processor to execute instructions, or to store data generated after the processor executes instructions.

When the communication device 1600 is used to implement the method shown in the above figures, the processor is used to implement the functions of the above processing unit, and the communication circuit is used to implement the functions of the above transceiver unit.

When the communication device 1600 is a chip implemented in the aforementioned device, the chip implements the functions of the corresponding device in the aforementioned method embodiment. The chip receives information from other modules in the device (such as a radio frequency module or antenna), where the information is sent to the device by other modules; or the chip sends information to other modules in the device (such as a radio frequency module or antenna).

When the communication device 1600 is a module applied to a mobile node, the module implements the functions of the mobile node in the above method embodiments. The module receives information from other modules (such as a radio frequency module or antenna), which is information sent by the terminal to the device; or the module sends information to other modules in the device (such as a radio frequency module or antenna), which is information sent by the device to the terminal. The module here can be the baseband chip of the device, or it can be a DU or other module. The DU here can be a DU under the open radio access network (O-RAN) architecture.

It is understood that the processor in the embodiments of the present application may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

In this application, another example of a communication device is provided, which includes at least one processor and at least one memory, the at least one processor and the at least one memory being coupled, the at least one memory being used to store instructions, and when the instructions are executed by the at least one processor, the communication device performs the method in the above-mentioned embodiment. Taking the communication device including a processor and a memory as an example, as shown in Figure 17, the communication device 1700 includes a processor 1710 and a memory 1730. The processor 1710 and the memory 1730 are coupled, and the memory 1730 stores instructions. When the instructions stored in the memory 1730 are executed by the processor 1710, the communication device 1700 performs the method performed by the terminal-side device or the network-side device in the above-mentioned embodiment.

It should be understood that the processor 1710 and the memory 1730 may also be integrated together, such as in one chip.

The method steps in the embodiments of the present application can be implemented in hardware or in software instructions that can be executed by a processor. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disk, mobile hard disk, CD-ROM or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. The storage medium can also be an integral part of the processor. The processor and storage medium can be located in an ASIC. In addition, the ASIC can be located in a network side device or a terminal side device. The processor and storage medium can also exist in a network side device or a terminal side device as discrete components.

In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are performed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a server, a network device, a user device, or other programmable device. The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video disk; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or nonvolatile storage medium, or may include both volatile and nonvolatile types of storage media.

In the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and/or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

In this application, "at least one" means one or more, and "more" means two or more. "And/or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and/or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. In the text description of this application, the character "/" generally indicates that the previous and next associated objects are in an "or" relationship; in the formula of this application, the character "/" indicates that the previous and next associated objects are in a "division" relationship. "Including at least one of A, B and C" can mean: including A; including B; including C; including A and B; including A and C; including B and C; including A, B and C.

It is understood that the various numbers used in the embodiments of this application are merely for ease of description and are not intended to limit the scope of the embodiments of this application. The order of the sequence numbers of the above-mentioned processes does not necessarily imply a specific order of execution; the order of execution of the processes should be determined by their functions and inherent logic.

Claims (63)

A channel feedback information transmission method, characterized by being applied to a second device, comprising: receiving a first signal from a first device; Sending channel feedback information to the first device, where the channel feedback information is obtained based on the first signal, the channel feedback information indicating quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, the non-zero coefficient selection information indicating positions of the non-zero coefficients, and the coefficient matrix is obtained based on a spatial domain basis and a frequency domain basis; The spatial domain basis and the frequency domain basis satisfy the first option, the second option or the third option; The first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; The second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all the vectors in the frequency domain basis matrix; The third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix. The method according to claim 1, wherein the spatial domain basis and the frequency domain basis satisfy the second option; The channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix. The method of claim 1, wherein the spatial domain basis and the frequency domain basis satisfy the third option; The channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates partial vectors in the frequency domain basis matrix, and the partial vectors are used to determine the coefficient matrix. The method according to any one of claims 1 to 3, characterized in that the channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. The method according to any one of claims 1 to 3 is characterized in that the spatial basis is determined according to spatial oversampling factor selection information, the spatial oversampling factor selection information is based on a configuration from the first device, or the spatial oversampling factor selection information is preconfigured, or the spatial oversampling factor selection information is predefined. The method according to any one of claims 1 to 5, further comprising: Sending first channel feedback information to the first device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model based on the first signal, and the first AI model is located on the second device; The sending channel feedback information to the first device includes: Sending second channel feedback information to the first device, where the second channel feedback information is used to monitor performance of the first AI model and/or the second AI model, where the second AI model is located on the first device, and the second AI model is used to recover the first channel feedback information. The method according to any one of claims 1 to 6, further comprising: Sending first channel feedback information to the first device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model based on one or more second signals, the first AI model is located on the second device, the one or more second signals are no later than the first signal, and the first channel feedback information is predicted channel feedback information; The sending channel feedback information to the first device includes: Sending second channel feedback information to the first device, where the second channel feedback information is true channel feedback information corresponding to the first channel feedback information, and the second channel feedback information is used to monitor the performance of the first AI model and/or the second AI model. The second AI model is located on the first device, and the second AI model is used to recover the first channel feedback information. The method according to any one of claims 1 to 7, further comprising: receiving configuration information from the first device, where the configuration information includes first indication information, where the first indication information is used to determine the first option, the second option, or the third option; The sending channel feedback information to the first device includes: Determine the first option, the second option, or the third option according to the first indication information; Determine the channel feedback information according to the first option, the second option, or the third option; Sending the channel feedback information to the first device. The method according to claim 8 is characterized in that the configuration information also includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. The method according to claim 8 or 9, wherein the first indication information is used to determine the second option, the first indication information indicates the number of spatial basis, and the number of spatial basis is less than the number of all vectors in the spatial basis matrix; The channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates a portion of vectors in the spatial basis matrix, where the portion of vectors is determined according to the number of spatial basis; The coefficient matrix is determined according to the spatial basis vector indicated by the spatial basis selection information and all vectors in the frequency domain basis matrix. The method according to claim 8 or 9, wherein the first indication information is used to determine the third option, the first indication information indicates a frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; The channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates a portion of vectors in the frequency domain basis matrix, where the portion of vectors is determined according to the frequency domain basis selection ratio; The coefficient matrix is determined according to the frequency domain basis vector indicated by the frequency domain basis selection information and all vectors in the spatial domain basis matrix. The method according to any one of claims 8 to 11, characterized in that, before receiving the configuration information from the first device, the method further comprises: Terminal capability information is sent to the first device, where the terminal capability information indicates that the second device does not support the codebook feature, or the terminal capability information indicates that the second device has the ability to determine a coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix. A channel feedback information transmission method, characterized by being applied to a first device, comprising: sending a first signal to a second device; receiving channel feedback information from the second device, where the channel feedback information is obtained based on the first signal, the channel feedback information indicating quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, the non-zero coefficient selection information indicating positions of the non-zero coefficients, and the channel feedback information, a spatial domain basis, and a frequency domain basis are used to determine a precoding matrix; The spatial domain basis and the frequency domain basis satisfy the first option, the second option or the third option; The first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; The second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all the vectors in the frequency domain basis matrix; The third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix. The method of claim 13, wherein the spatial domain basis and the frequency domain basis satisfy the second option; The channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix. The method of claim 13, wherein the spatial domain basis and the frequency domain basis satisfy the third option; The channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates a portion of vectors in the frequency domain basis matrix, and the portion of vectors is used to determine the coefficient matrix. The method according to any one of claims 13 to 15, wherein the channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. The method according to any one of claims 13-14, characterized in that the spatial basis is determined according to spatial oversampling factor selection information, and the spatial oversampling factor selection information is preconfigured or predefined; or The method further comprises: Send spatial oversampling factor selection information to the second device. The method according to any one of claims 13 to 17, further comprising: receiving first channel feedback information from the second device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model according to the first signal, and the first AI model is located in the second device; The receiving channel feedback information from the second device includes: Second channel feedback information is received from the second device, where the second channel feedback information is used to monitor performance of the first AI model and/or the second AI model, the second AI model is located in the first device, and the second AI model is used to recover the first channel feedback information. The method according to any one of claims 13 to 18, further comprising: receiving first channel feedback information from the second device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model based on one or more second signals, the first AI model is located on the second device, the one or more second signals are no later than the first signal, and the first channel feedback information is predicted channel feedback information; The receiving channel feedback information from the second device includes: Receive second channel feedback information from the second device, where the second channel feedback information is true channel feedback information corresponding to the first channel feedback information, and the second channel feedback information is used to monitor performance of the first AI model and/or the second AI model. The second AI model is located on the first device, and the second AI model is used to recover the first channel feedback information. The method according to any one of claims 13 to 19, further comprising: Sending configuration information to the second device, where the configuration information includes first indication information, where the first indication information is used to indicate the first option, the second option, or the third option; After receiving the channel feedback information from the second device, the method further includes: A precoding matrix is determined according to the channel feedback information and the first option, the second option, or the third option indicated by the first indication information. The method as claimed in claim 20 is characterized in that the configuration information also includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. The method according to claim 20 or 21, wherein the first indication information indicates the second option and a number of spatial basis, and the number of spatial basis is less than the number of all vectors in the spatial basis matrix; The channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates a portion of vectors in the spatial basis matrix, where the portion of vectors is determined according to the number of spatial basis; The determining a precoding matrix according to the channel feedback information and the first option, the second option, or the third option indicated by the first indication information includes: The precoding matrix is determined according to the quantization information of the non-zero coefficients in the channel feedback information and the non-zero coefficient selection information, the spatial basis vector indicated by the spatial basis selection information and all vectors in the frequency domain basis matrix. The method according to claim 20 or 21, wherein the first indication information indicates the third option and a frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; The channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates a portion of vectors in the frequency domain basis matrix, where the portion of vectors is determined according to the frequency domain basis selection ratio; The determining a precoding matrix according to the channel feedback information and the first option, the second option, or the third option indicated by the first indication information includes: The precoding matrix is determined according to the quantization information of the non-zero coefficients in the channel feedback information and the non-zero coefficient selection information, all vectors in the spatial basis matrix and the frequency domain basis vector indicated by the frequency domain basis selection information. The method according to any one of claims 20 to 23, wherein before sending the configuration information to the second device, the method further comprises: Receive terminal capability information from the second device, where the terminal capability information indicates that the second device does not support codebook characteristics, or the terminal capability information indicates that the second device has the ability to determine a coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix. A method for transmitting channel feedback information, characterized in that the method comprises: receiving a first signal from a first device; Sending first channel feedback information to the first device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model based on the first signal, and the first AI model is located on the second device; receiving third channel feedback information from the first device, where the third channel feedback information is recovered by a second AI model from the first channel feedback information, where the second AI model is located in the first device, the third channel feedback information including quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, where the non-zero coefficient selection information indicates positions of the non-zero coefficients, where the coefficient matrix is obtained based on a spatial domain basis and a frequency domain basis; The spatial domain basis and the frequency domain basis satisfy the first option, the second option or the third option; The first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; The second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all the vectors in the frequency domain basis matrix; The third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix. The method according to claim 25, further comprising: Configuration information is received from the first device, where the configuration information includes first indication information, and the first indication information is used to determine the first option, the second option, or the third option. The method of claim 26, wherein before receiving the configuration information from the first device, the method further comprises: Terminal capability information is sent to the first device, where the terminal capability information indicates that the second device does not support the codebook feature, or the terminal capability information indicates that the second device has the ability to determine a coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix. The method according to any one of claims 26-27 is characterized in that the configuration information also includes spatial oversampling factor selection information. The method according to any one of claims 26 to 28, wherein the first indication information is used to determine the second option, the first indication information indicates a number of spatial basis, and the number of spatial basis is less than the number of all vectors in the spatial basis matrix; The channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates a portion of vectors in the spatial basis matrix, and the portion of vectors is determined according to the number of spatial basis. The method according to any one of claims 26 to 28, wherein the first indication information is used to determine the third option, the first indication information indicates a frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; The channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates a portion of vectors in the frequency domain basis matrix, and the portion of vectors is determined according to the frequency domain basis selection ratio. The method according to any one of claims 25 to 27, wherein the spatial domain basis and the frequency domain basis satisfy the second option; The third channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix. The method according to any one of claims 25 to 27, wherein the spatial domain basis and the frequency domain basis satisfy the third option; The third channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates partial vectors in the frequency domain basis matrix, and the partial vectors are used to determine the coefficient matrix. The method according to any one of claims 25 to 32, wherein the third channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. The method according to any one of claims 25 to 32, characterized in that the spatial basis is determined according to spatial oversampling factor selection information, the spatial oversampling factor selection information is based on a configuration from the first device, or the spatial oversampling factor selection information is preconfigured, or the spatial oversampling factor selection information is predefined. A method for transmitting channel feedback information, characterized in that the method comprises: sending a first signal to a first device; receiving first channel feedback information from the first device, where the first channel feedback information is obtained by a first artificial intelligence (AI) model according to the first signal, and the first AI model is located on the second device; sending third channel feedback information to the first device, where the third channel feedback information is recovered by a second AI model from the first channel feedback information, where the second AI model is located in the first device, the third channel feedback information including quantization information of non-zero coefficients and non-zero coefficient selection information in a coefficient matrix, where the non-zero coefficient selection information indicates positions of the non-zero coefficients, where the coefficient matrix is obtained based on a spatial domain basis and a frequency domain basis; The spatial domain basis and the frequency domain basis satisfy the first option, the second option or the third option; The first option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is all vectors in the frequency domain basis matrix; The second option is: the spatial basis is part of the vectors in the spatial basis matrix, and the frequency domain basis is all the vectors in the frequency domain basis matrix; The third option is: the spatial basis is all vectors in the spatial basis matrix, and the frequency domain basis is part of the vectors in the frequency domain basis matrix. The method of claim 35, further comprising: Configuration information is sent to the second device, where the configuration information includes first indication information, and the first indication information is used to determine the first option, the second option, or the third option. The method of claim 36, wherein before sending the configuration information to the second device, the method further comprises: Receive terminal capability information from the first device, where the terminal capability information indicates that the second device does not support codebook characteristics, or the terminal capability information indicates that the second device has the ability to determine a coefficient matrix based on all vectors in the spatial basis matrix and/or all vectors in the frequency domain basis matrix. The method according to any one of claims 36-37 is characterized in that the configuration information also includes spatial oversampling factor selection information. The method according to any one of claims 36 to 38, wherein the first indication information is used to determine the second option, the first indication information indicates a number of spatial basis, and the number of spatial basis is less than the number of all vectors in the spatial basis matrix; The channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates a portion of vectors in the spatial basis matrix, and the portion of vectors is determined according to the number of spatial basis. The method according to any one of claims 36 to 38, wherein the first indication information is used to determine the third option, the first indication information indicates a frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1; The channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates a portion of vectors in the frequency domain basis matrix, and the portion of vectors is determined according to the frequency domain basis selection ratio. The method according to any one of claims 35 to 37, wherein the spatial domain basis and the frequency domain basis satisfy the second option; The third channel feedback information further includes spatial basis selection information, where the spatial basis selection information indicates partial vectors in the spatial basis matrix, and the partial vectors are used to determine the coefficient matrix. The method according to any one of claims 35 to 37, wherein the spatial domain basis and the frequency domain basis satisfy the third option; The third channel feedback information further includes frequency domain basis selection information, where the frequency domain basis selection information indicates partial vectors in the frequency domain basis matrix, and the partial vectors are used to determine the coefficient matrix. The method according to any one of claims 35 to 42, wherein the third channel feedback information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. The method according to any one of claims 35 to 42, characterized in that the spatial basis is determined according to spatial oversampling factor selection information, the spatial oversampling factor selection information is based on a configuration from the first device, or the spatial oversampling factor selection information is preconfigured, or the spatial oversampling factor selection information is predefined. A communication method, characterized in that the method comprises: Sending terminal capability information to the first device, where the terminal capability information is used to determine a channel feedback information transmission mode; Receive configuration information from the first device, where the configuration information is used to indicate a channel feedback information transmission mode that matches the terminal capability information. The method of claim 45, wherein the terminal capability information indicates that the second device supports a first channel feedback information transmission mode or supports a codebook characteristic; The configuration information includes first indication information, which indicates a first option, which is: the spatial basis for determining non-zero coefficients is all vectors in the spatial basis matrix, and the frequency domain basis for determining non-zero coefficients is all vectors in the frequency domain basis matrix. The method of claim 45, wherein the terminal capability information indicates that the second device supports a second channel feedback information transmission mode or supports a codebook characteristic; The configuration information includes first indication information, which indicates a second option, which is: the spatial basis used to determine the non-zero coefficients is a part of the vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficients is all vectors in the frequency domain basis matrix. The method as claimed in claim 47 is characterized in that the first indication information also indicates the number of spatial bases, and the number of spatial bases is less than the number of all vectors in the spatial basis matrix. The method of claim 45, wherein the terminal capability information indicates that the second device supports a third channel feedback information transmission mode or supports a codebook characteristic; The configuration information includes first indication information, which indicates a third option, and the third option is: the spatial basis used to determine the non-zero coefficients is all vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficients is part of the vectors in the frequency domain basis matrix. The method as claimed in claim 49 is characterized in that the first indication information also indicates the frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1. The method according to any one of claims 45 to 50 is characterized in that the configuration information also includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. A communication method, characterized in that the method comprises: receiving terminal capability information from the first device, where the terminal capability information is used to determine a channel feedback information transmission mode; Configuration information is sent to the first device, where the configuration information is used to indicate a channel feedback information transmission mode that matches the terminal capability information. The method of claim 52, wherein the terminal capability information indicates that the second device supports the first channel feedback information transmission mode or supports a codebook characteristic; The configuration information includes first indication information, which indicates a first option, which is: the spatial basis for determining non-zero coefficients is all vectors in the spatial basis matrix, and the frequency domain basis for determining non-zero coefficients is all vectors in the frequency domain basis matrix. The method of claim 52, wherein the terminal capability information indicates that the second device supports a second channel feedback information transmission mode or supports a codebook characteristic; The configuration information includes first indication information, which indicates a second option, which is: the spatial basis used to determine the non-zero coefficients is a part of the vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficients is all vectors in the frequency domain basis matrix. The method as claimed in claim 54 is characterized in that the first indication information also indicates the number of spatial bases, and the number of spatial bases is less than the number of all vectors in the spatial basis matrix. The method of claim 52, wherein the terminal capability information indicates that the second device supports a third channel feedback information transmission mode or supports a codebook characteristic; The configuration information includes first indication information, which indicates a third option, and the third option is: the spatial basis used to determine the non-zero coefficients is all vectors in the spatial basis matrix, and the frequency domain basis used to determine the non-zero coefficients is part of the vectors in the frequency domain basis matrix. The method as claimed in claim 56 is characterized in that the first indication information also indicates the frequency domain basis selection ratio, and the frequency domain basis selection ratio is greater than 0 and less than 1. The method according to any one of claims 52 to 57 is characterized in that the configuration information further includes spatial oversampling factor selection information, and the spatial oversampling factor selection information is used to determine the spatial basis. A communication device, characterized in that it includes a unit or module for executing the method as described in any one of claims 1 to 12, or includes a unit or module for executing the method as described in any one of claims 13 to 24, or includes a unit or module for executing the method as described in any one of claims 25 to 34, or includes a unit or module for executing the method as described in any one of claims 35 to 44, or includes a unit or module for executing the method as described in any one of claims 45 to 51, or includes a unit or module for executing the method as described in any one of claims 52 to 58. A communication device, characterized in that it includes one or more processors for executing programs or instructions in at least one memory so that the device performs the method described in any one of claims 1 to 12, the method described in any one of claims 13 to 24, the method described in any one of claims 25 to 34, the method described in any one of claims 35 to 44, the method described in any one of claims 45 to 51, or the method described in any one of claims 52 to 58. A readable storage medium, characterized in that it is used to store a program or instruction, when the program or instruction is executed, the method according to any one of claims 1 to 12, the method according to any one of claims 13 to 24, the method according to any one of claims 25 to 34, the method according to any one of claims 35 to 44, the method according to any one of claims 45 to 51, or the method according to any one of claims 52 to 58 is implemented. A computer program, characterized in that it includes a program or instruction, and when the program or instruction is executed, the method according to any one of claims 1 to 12, the method according to any one of claims 13 to 24, the method according to any one of claims 25 to 34, the method according to any one of claims 35 to 44, the method according to any one of claims 45 to 51, or the method according to any one of claims 52 to 58 is implemented. A communication system, characterized in that it includes one or more of an apparatus for executing the method as described in any one of claims 1 to 12, an apparatus for executing the method as described in any one of claims 13 to 24, an apparatus for executing the method as described in any one of claims 25 to 34, an apparatus for executing the method as described in any one of claims 35 to 44, an apparatus for executing the method as described in any one of claims 45 to 51, or an apparatus for executing the method as described in any one of claims 52 to 58.