WO2025166574A1 - Information processing method and device - Google Patents

Abstract

The present application relates to an information processing method and a device. The information processing method comprises: a first communication device receiving a plurality of pieces of first output information from a plurality of second communication devices, wherein each piece of first output information is obtained by means of a second communication device using a first information processing module to process first input information; on the basis of the plurality of pieces of first output information, the first communication device obtaining second input information; and the first communication device using a second information processing module to process the second input information to obtain second output information. In the embodiments of the present application, output information can be obtained on the basis of a plurality of pieces of first input information from a plurality of communication devices, thereby supporting joint feedback, and improving the information processing efficiency of a communication system.

Description

Information processing method and device Technical Field

The present application relates to the field of communications, and more specifically, to an information processing method and device.

Background Art

In an artificial intelligence (AI)-based channel state information (CSI) autoencoder solution, a CSI feedback system can include an encoder and a decoder, deployed on the user side and base station side, respectively. The user-side encoder's neural network compresses and encodes the CSI and then feeds it back to the base station via an air interface feedback link. The base station decoder recovers the compressed CSI and outputs complete feedback channel information.

Summary of the Invention

The embodiments of the present application provide an information processing method and device, which can improve information processing efficiency.

The present invention provides an information processing method, including:

The first communication device receives a plurality of first output information from a plurality of second communication devices; wherein one first output information is obtained by a second communication device processing the first input information using the first information processing module;

The first communication device obtains second input information according to the plurality of first output information;

The first communication device processes the second input information using a second information processing module to obtain second output information.

The present invention provides an information processing module training method, including:

The first communication device receives multiple CSIs from multiple second communication devices; wherein one CSI is obtained by one second communication device through CSI-RS measurement;

The first communication device uses the multiple CSIs to train multiple first information processing modules and second information processing modules that need to be trained, to obtain multiple trained first information processing modules and second information processing modules.

The present invention provides an information processing method, including:

The first communication device sends first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

The present invention provides an information processing method, including:

The second communication device receives first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

An embodiment of the present application provides a first communication device, including:

A first receiving unit is configured to receive a plurality of first output information from a plurality of second communication devices; wherein one first output information is obtained by a second communication device processing the first input information using the first information processing module;

The first processing unit is configured to obtain second input information according to the plurality of first output information; and process the second input information using a second information processing module to obtain second output information.

An embodiment of the present application provides a first communication device, including:

A second receiving unit is configured to receive a plurality of CSIs from a plurality of second communication devices; wherein one CSI is obtained by one second communication device through CSI-RS measurement;

The second processing unit is configured to train the plurality of first information processing modules and the second information processing modules that need to be trained using the plurality of CSIs to obtain the plurality of trained first information processing modules and the second information processing modules.

An embodiment of the present application provides a first communication device, including: a second sending unit, configured to send first indication information, where the first indication information is used to indicate that a second communication device is scheduled for multi-user transmission.

An embodiment of the present application provides a first communication device, including:

A second receiving unit is configured to receive a plurality of CSIs from a plurality of second communication devices; wherein one CSI is obtained by one second communication device through CSI-RS measurement;

The second processing unit is configured to train the plurality of first information processing modules and the second information processing modules that need to be trained using the plurality of CSIs to obtain the plurality of trained first information processing modules and the second information processing modules.

An embodiment of the present application provides a first communication device, including: a second sending unit, configured to send first indication information, where the first indication information is used to indicate that a second communication device is scheduled for multi-user transmission.

An embodiment of the present application provides a second communication device, including: a fourth receiving unit, configured to receive first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

An embodiment of the present application provides a communication device, comprising: a transceiver, a processor, and a memory. The memory is used to store a computer program, the transceiver is used to communicate with other devices, and the processor is used to call and execute the computer program stored in the memory, so that the communication device performs the above-mentioned information processing method.

An embodiment of the present application provides a chip for implementing the above-mentioned information processing method.

Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that the device equipped with the chip Execute the above information processing method.

An embodiment of the present application provides a computer-readable storage medium for storing a computer program. When the computer program is executed by a device, the device executes the above-mentioned information processing method.

An embodiment of the present application provides a computer program product, including computer program instructions, which enable a computer to execute the above-mentioned information processing method.

An embodiment of the present application provides a computer program, which, when executed on a computer, enables the computer to execute the above-mentioned information processing method.

In the embodiment of the present application, output information can be obtained based on multiple first input information from multiple communication devices, supporting joint feedback and improving the information processing efficiency of the communication system.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of an application scenario according to an embodiment of the present application.

Figure 2 is a schematic diagram of the neuron structure.

Figure 3 is a schematic diagram of a fully connected neural network.

Figure 4 is a schematic diagram of a convolutional neural network.

Figure 5 is a schematic diagram of LSTM.

Figure 6 is a schematic diagram of the AI-based CSI autoencoder framework.

FIG7 is a schematic flowchart of an information processing method according to an embodiment of the present application.

FIG8 is a schematic flowchart of an information processing module training method according to an embodiment of the present application.

FIG9 is a schematic flowchart of an information processing module training method according to another embodiment of the present application.

FIG10 is a schematic flowchart of an information processing method according to an embodiment of the present application.

FIG11 is a schematic flowchart of an information processing method according to another embodiment of the present application.

FIG12 is a schematic flowchart of an information processing method according to an embodiment of the present application.

FIG13 is a schematic flowchart of an information processing method according to another embodiment of the present application.

FIG14 is a schematic diagram of a CSI feedback framework based on AI multi-users.

FIG15 is a schematic diagram of a UE-side training method.

FIG16 is a schematic diagram of a network-side training method.

FIG17 is a schematic diagram of a two-stage multi-user CSI feedback signaling process.

FIG18 is a schematic diagram of a one-stage multi-user CSI feedback signaling process.

FIG19 is a schematic block diagram of a first communication device according to an embodiment of the present application.

FIG20 is a schematic block diagram of a first communication device according to an embodiment of the present application.

FIG21 is a schematic block diagram of a first communication device according to an embodiment of the present application.

FIG22 is a schematic block diagram of a second communication device according to an embodiment of the present application.

Figure 23 is a schematic block diagram of a communication device according to an embodiment of the present application.

Figure 24 is a schematic block diagram of a chip according to an embodiment of the present application.

Figure 25 is a schematic block diagram of a communication system according to an embodiment of the present application.

DETAILED DESCRIPTION

The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: Long Term Evolution (LTE) system, Advanced Long Term Evolution (LTE-A) system, New Radio (NR) system, NR system evolution system, LTE on unlicensed spectrum (LTE-U) system, NR on unlicensed spectrum (NR-U) system, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Wireless Fidelity (WiFi), 5th-Generation (5G) system or other communication systems.

Generally speaking, traditional communication systems support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communications, but will also support, for example, device-to-device (D2D) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), vehicle-to-vehicle (V2V) communication, or vehicle-to-everything (V2X) communication, etc. The embodiments of the present application can also be applied to these communication systems.

In one embodiment, the communication system in the embodiment of the present application can be applied to a carrier aggregation (CA) scenario, a dual connectivity (DC) scenario, and a standalone (SA) networking scenario.

In one embodiment, the communication system in the embodiment of the present application can be applied to an unlicensed spectrum, wherein the unlicensed spectrum can also be considered as a shared spectrum; or, the communication system in the embodiment of the present application can also be applied to an authorized spectrum, wherein the authorized spectrum can also be considered as an unshared spectrum.

The embodiments of the present application describe various embodiments in conjunction with network devices and terminal devices, wherein the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication equipment, user agent or user device, etc.

The terminal device can be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a wearable device, a terminal device in a next-generation communication system such as a NR network, or a terminal device in a future evolved Public Land Mobile Network (PLMN) network, etc.

In an embodiment of the present application, the terminal device can be deployed on land, including indoors or outdoors, handheld, wearable 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 satellites, etc.).

In the embodiments of the present application, the terminal device may be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city, or a wireless terminal device in a smart home, etc.

As an example and not a limitation, in the embodiment of the present application, the terminal device 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 those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those 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 an embodiment of the present application, the network device may be a device for communicating with a mobile device, and the network device may be an access point (AP) in a WLAN, an evolved base station (eNB or eNodeB) in an LTE, or a relay station or access point, or an in-vehicle device, a wearable device, a network device (gNB) in an NR network, or a network device in a future evolved PLMN network or a network device in an NTN network, etc.

As an example and not a limitation, in an embodiment of the present application, the network device may have a mobile feature, for example, the network device may be a mobile device. Optionally, the network device may be a satellite or a balloon station. For example, the satellite may be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, etc. Optionally, the network device may also be a base station set up in a location such as land or water.

In an embodiment of the present application, a network device can provide services for a cell, and a terminal device communicates with the network device through the transmission resources used by the cell (for example, frequency domain resources, or spectrum resources). The cell can be a cell corresponding to a network device (for example, a base station). The cell can belong to a macro base station or a base station corresponding to a small cell. The small cells here may include: metro cells, micro cells, pico cells, femto cells, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.

FIG1 exemplarily illustrates a communication system 100. The communication system includes a network device 110 and two terminal devices 120. In one embodiment, the communication system 100 may include multiple network devices 110, and each network device 110 may include a different number of terminal devices 120 within its coverage area, which is not limited in this embodiment of the present application.

In one embodiment, the communication system 100 may also include other network entities such as a Mobility Management Entity (MME) and an Access and Mobility Management Function (AMF), but this embodiment of the present application does not limit this.

Among them, the network equipment may include access network equipment and core network equipment. That is, the wireless communication system also includes multiple core networks for communicating with the access network equipment. The access network equipment can be an evolutionary base station (evolutional node B, abbreviated as eNB or e-NodeB) macro base station, micro base station (also called "small base station"), pico base station, access point (AP), transmission point (TP) or new generation base station (new generation Node B, gNodeB), etc. in a long-term evolution (LTE) system, a next-generation (mobile communication system) (next radio, NR) system or an authorized auxiliary access long-term evolution (LAA-LTE) system.

It should be understood that in the embodiments of the present application, a device having a communication function in a network/system may be referred to as a communication device. Taking the communication system shown in Figure 1 as an example, the communication device may include a network device and a terminal device having a communication function. The network device and the terminal device may be specific devices in the embodiments of the present application and will not be described in detail here. The communication device may also include other devices in the communication system, such as a network controller, a mobility management entity, and other network entities, which are not limited in the embodiments of the present application.

It should be understood that the terms "system" and "network" are often used interchangeably herein. The term "and/or" is simply a description of an association between related objects, indicating that three possible relationships exist. For example, "A and/or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character "/" generally indicates that the related objects are in an "or" relationship.

It should be understood that the "indication" mentioned in the embodiments of this application can be a direct indication, an indirect indication, or an indication of an association. For example, "A indicates B" can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association between A and B.

In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship between indication and being indicated, configuration and being configured, etc.

To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following relevant technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

1. Neural Networks and Machine Learning

A neural network is a computational model consisting of multiple interconnected neuron nodes. The connections between nodes represent weighted values, called weights, from input signals to output signals. Each node performs a weighted summation of different input signals and outputs the result through a specific activation function. Figure 2 shows the neuron structure.

A simple neural network is shown in Figure 3, which includes an input layer, a hidden layer, and an output layer. Different outputs can be generated by different connection methods, weights, and activation functions of multiple neurons, thereby fitting the mapping relationship from input to output. Each upper-level node is connected to all of its lower-level nodes. This fully connected model can also be called a deep neural network (DNN) in the embodiments of this application.

The basic structure of a convolutional neural network (CNN) consists of an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer, as shown in Figure 4. Each neuron in the convolutional kernel of a convolutional layer is locally connected to its input. The introduction of a pooling layer extracts the local maximum or average features of a layer, effectively reducing the number of network parameters and exploiting local features, enabling the CNN to converge quickly and achieve excellent performance.

A recurrent neural network (RNN) is a neural network that models sequential data and has achieved remarkable results in natural language processing applications such as machine translation and speech recognition. Specifically, the network memorizes information from past moments and uses it in the calculation of the current output. That is, the nodes between hidden layers are no longer disconnected but connected, and the input of the hidden layer includes not only the input layer but also the output of the hidden layer at the previous moment. Commonly used RNNs include structures such as the Long-Short Term Memory (LSTM) artificial neural network kernel and the Gate Recurrent Unit (GRU). Figure 5 shows a basic LSTM unit structure. Unlike RNNs that only consider the most recent state, the cell state of LSTM determines which states should be retained and which states should be forgotten, solving the defects of traditional RNNs in long-term memory.

2. Codebook-based Channel State Information (CSI) Feedback Scheme in NR

In NR systems, the CSI feedback scheme typically uses codebook-based eigenvector feedback to enable the base station to obtain downlink CSI. Specifically, the base station sends a downlink Channel State Information-Reference Signal (CSI-RS) to the user. The user uses the CSI-RS to estimate the downlink channel CSI and performs eigenvalue decomposition on the estimated downlink channel to obtain the eigenvector corresponding to the downlink channel.

Specifically, NR provides two codebook designs: Type 1 and Type 2. The Type 1 codebook is used for CSI feedback with conventional accuracy, primarily for transmission in single-user (SU)-multiple input multiple output (MIMO) scenarios, while the Type 2 codebook is primarily used to improve the transmission performance of multi-user (MU)-MIMO. Both Type 1 and Type 2 codebooks employ a two-stage codebook feedback scheme where W = W ₁ W ₂ , where W ₁ describes the wideband, long-term characteristics of the channel and determines a set of L discrete Fourier transform (DFT) beams; and W ₂ describes the subband, short-term characteristics of the channel. Specifically, for the Type 1 codebook, W ₂ selects a beam from the L DFT beams; for the Type 2 codebook, W ₂ linearly combines the L DFT beams in W ₁ and provides feedback in the form of amplitude and phase. Generally, the Type 2 codebook utilizes a higher number of feedback bits to obtain higher-precision CSI feedback performance.

3. AI-based CSI feedback method in the 18th generation version (Release 18, R18)

Given the tremendous success of AI technology, especially deep learning in computer vision and natural language processing, the communications field has begun to try to use deep learning to solve technical problems that are difficult to solve with traditional communication methods. For example, the neural network architecture commonly used in deep learning is nonlinear and data-driven. It can extract features from the actual channel matrix data and restore the channel matrix information compressed and fed back by the UE as much as possible on the base station side. While ensuring the restoration of channel information, it also provides the possibility of reducing the CSI feedback overhead on the UE side. The CSI feedback based on deep learning regards the channel information as an image to be compressed, and uses a deep learning autoencoder to compress and feedback the input channel information. The compressed channel image is fed back and reconstructed at the sending end, which can preserve the channel information to a greater extent.

AI-based CSI feedback is one of the main use cases of AI projects. The basic implementation framework of CSI feedback is as follows:

Using an AI-based CSI autoencoder approach, the entire feedback system is divided into an encoder and a decoder, deployed at the user's transmitting end and the base station's receiving end, respectively. After the user obtains channel information through channel estimation, it serves as the encoder's input. The encoder's neural network compresses and encodes the channel information matrix, and the compressed bit stream is fed back to the base station via the air interface feedback link. The base station recovers the channel information based on the feedback bit stream through the decoder and outputs the complete feedback channel information. The neural networks of the encoder and decoder shown in Figure 6 can adopt a DNN composed of multiple layers of fully connected layers, a CNN composed of multiple layers of convolutional layers, or an RNN with structures such as LSTM and GRU. Various neural network architectures such as residual and self-attention mechanisms can also be used to improve the performance of the encoder and decoder.

The above-mentioned CSI input and/or CSI output can be full channel information, or eigenvector information obtained based on full channel information. Therefore, the current channel information feedback methods based on deep learning are mainly divided into full channel information feedback and eigenvector feedback. Although the former can realize the compression and feedback of full channel information, the feedback bit stream overhead is high, and this feedback method is not supported in the existing NR system. As for the eigenvector-based feedback method, it is the feedback architecture supported by the current NR system, and the AI-based eigenvector feedback method can achieve higher CSI feedback accuracy with the same feedback bit overhead, or significantly reduce the feedback overhead while achieving the same CSI feedback accuracy.

4. Multi-user transmission in NR

MU-MIMO technology is a key NR technology. By simultaneously serving multiple users within the same time-frequency resources, it can significantly improve the system's spectral efficiency. Compared to SU-MIMO, the ratio of the number of antennas on the user side to the number of concurrent data streams (including the data streams that the user needs to receive and the data streams of co-scheduled users) is lower, and the channel matrix of the interference signal is generally difficult to estimate. Therefore, the performance of the MU-MIMO system is more dependent on the accuracy of the CSI acquisition and the degree of optimization of the precoding and scheduling algorithms. In current NR systems, the design of the Type 2 codebook is mainly aimed at enhancing MU-MIMO transmission, which can significantly improve CSI accuracy and thus greatly improve the performance of MU-MIMO transmission.

Taking two users as an example, the base station side usually schedules two users with relatively low channel correlation to perform multi-user transmission, so that the precoding matrix can better eliminate the interference between the two users and improve the gain of MU-MIMO.

The characteristics of codebook-based CSI feedback in NR include: CSI feedback in the 5G NR standard uses Type 1 and Type 2 codebook-based feedback. This codebook-based CSI feedback method has good generalization capabilities for different users and various channel scenarios. However, because the codebook is pre-set, it does not effectively utilize the correlation between different antenna ports and subbands. Therefore, under the same feedback overhead, the feedback performance is poor; or when the feedback performance is achieved, the feedback overhead is high.

The characteristics of AI-based CSI feedback include: The AI-based CSI feedback method can extract the correlation of feature vectors in the time domain and frequency domain, so it can achieve better feedback performance with lower feedback overhead. However, the encoder and decoder used in this solution both use neural network models. When the training set environment is consistent with the actual deployment environment of the model (i.e., the test set environment), the AI-based CSI feedback method can achieve significant gains in both feedback overhead and feedback performance. However, AI-based CSI feedback is also targeted at single-user, point-to-point CSI feedback. That is, the encoder deployed on the user side can only compress and feedback the CSI of the user, and the decoder on the base station side can only decode and recover the CSI of a single user.

The characteristics of multi-user transmission in NR include: Currently, multi-user transmission in NR requires the base station to first obtain the CSI reports of each user in the cell and then perform user scheduling based on the CSI information of multiple users. In this case, the performance gain of MU-MIMO is heavily dependent on the CSI feedback accuracy of individual users. Whether it is the codebook-based CSI feedback in NR or the AI-based CSI feedback in Release 18/R19, CSI feedback accuracy may be low in certain wireless channel environments, affecting the base station's multi-user scheduling results.

Furthermore, because the primary challenge in MU-MIMO transmission is interference between multiple users, the CSI feedback schemes in related technologies are point-to-point, with no joint feedback between multiple users or estimation of inter-user interference. In other words, CSI feedback and downlink precoding calculations are performed independently. Consequently, with limited uplink feedback overhead, the design is not optimized for multi-user transmission scenarios.

In summary, the use of AI technology to design CSI feedback and precoding methods for multiple users may bring potential performance gains to the performance improvement of MU-MIMO systems.

FIG7 is a schematic flow chart of an information processing method 700 according to an embodiment of the present application. The method can optionally be applied to the system shown in FIG1 , but is not limited thereto. The method includes at least part of the following contents.

S710. A first communication device receives a plurality of first output information from a plurality of second communication devices; wherein one piece of first output information is obtained by a second communication device processing first input information using a first information processing module;

S720: The first communication device obtains second input information according to the plurality of first output information;

S730: The first communication device uses a second information processing module to process the second input information to obtain second output information.

In some examples, the first communication device may be a network device such as a base station, and the second communication device may be a terminal device such as a UE. The first input information may be the CSI of the UE before being processed by the first information processing module, and the first output information may be the CSI of the first information processing module. Output information. The base station can receive first output information from multiple UEs, and then combine the first output information of the multiple UEs as second input information and input it into the second information processing module for processing. The second information processing module can output second output information.

In one embodiment, the first information processing module includes at least one of a first model, a first function, and a first characteristic.

In one embodiment, the second information processing module includes at least one of a second model, a second function, and a second characteristic.

For example, at least one of the first model, the first function, and the first characteristic may be related to a CSI encoder. The first model may be a first AI/machine learning (ML) model, and the first AI/ML model may be used for CSI encoding. The first function may be a first AI/ML function, and the first AI/ML function may include a CSI encoding function. The first characteristic may be a first AI/ML characteristic, and the first AI/ML characteristic may include a CSI encoding characteristic.

For another example, at least one of the second model, the second function, and the second characteristic may be related to a CSI decoder. For another example, the second model may be a second AI/ML model, which may be used for CSI decoding. The second function may be a second AI/ML function, which may include a CSI decoding function. The second characteristic may be a second AI/ML characteristic, which may include a CSI decoding characteristic.

In one embodiment, the first model includes a channel state information (CSI) encoder model, the first function includes a CSI encoder function, or the first characteristic includes a CSI encoder characteristic.

In one embodiment, the second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic.

For example, multiple UEs use a CSI encoder model to process measured first input information to obtain first output information, which is then sent to the base station. After receiving the first output information from the encoders of multiple users, the base station combines the multiple first output information to obtain combined second input information, and then processes it using the CSI decoder model to obtain second output information.

In one embodiment, the first input information includes CSI measured by the second communication device, and the first output information includes a bit sequence obtained by the second communication device processing the CSI using a first information processing module.

In one embodiment, the second input information includes information obtained by combining the multiple first input information, and the second output information includes a precoding vector obtained by the first communications device processing the second input information using the second information processing module. For example, after the UE measures and obtains CSI, it processes the CSI using a CSI encoder model to obtain an encoded bit sequence. The UE can then send the bit sequence to the base station. The base station can combine the bit sequences from multiple UEs and input them into a CSI decoder model corresponding to the encoder model on the UE to obtain a precoding vector for the CSI.

The embodiments of the present application can obtain output information based on multiple first input information from multiple communication devices, support joint feedback, and improve the information processing efficiency of the communication system.

FIG8 is a schematic flow chart of an information processing module training method 800 according to an embodiment of the present application. The method can optionally be applied to the system shown in FIG1 , but is not limited thereto. The method includes at least part of the following contents.

S810. A first communication device receives multiple CSIs from multiple second communication devices, wherein one CSI is obtained by one second communication device through CSI-RS measurement.

S820: The first communication device uses the multiple CSIs to train multiple first information processing modules and second information processing modules that need to be trained, to obtain multiple trained first information processing modules and second information processing modules.

In an embodiment of the present application, after the second communication device, such as a UE, performs CSI-RS measurement locally, it can send the CSI to the first communication device. After the first communication device receives multiple CSIs from multiple second communication devices, it can use the multiple CSIs to train the CSI model that needs to be trained. The CSI model may include multiple first information processing modules and second information processing modules. The multiple first information processing modules and second information processing modules can be jointly trained, and can be trained on the terminal side, such as the UE, the UE's computing unit, computing node or computing entity, or on the network side, such as the network device, the network device's computing unit, computing node or computing entity. The first information processing modules applicable to different terminal devices can be the same, not completely the same, or completely different. For example, UE1 corresponds to the CSI encoder model M1, UE2 and UE3 correspond to the CSI encoder model M2, and UE4 and UE5 correspond to the CSI encoder model M3. Among them, M1 and M2 are completely different, and M2 and M3 have the same structure but different parameters (an example of not completely the same). M1, M2 and M3 can be jointly trained with the same CSI decoder model.

In some examples, the first communication device may include a user-side computing unit, computing entity, or computing node, such as a server with strong computing capabilities. The second communication device may include a terminal device, such as a UE. During the data collection phase, the UE-side computing unit, computing entity, or computing node may receive a training dataset from multiple UEs. The dataset may include CSI measured by the multiple UEs using downlink CSI-RS.

In some examples, the first communication device may include a network device, a network-side computing unit, a computing entity, or a computing node. The second communication device may include a terminal device such as a UE, a user-side computing unit, a computing entity, or a computing node. For example, during the data collection phase, the base station may receive data sets for training from multiple UEs.

In one embodiment, the first information processing module includes at least one of a first model, a first function, and a first characteristic.

In one embodiment, the second information processing module includes at least one of a second model, a second function, and a second characteristic.

In one embodiment, the first model includes a channel state information (CSI) encoder model, the first function includes a CSI encoder function, or the first characteristic includes a CSI encoder.

In one embodiment, the second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic.

For the above modules, functions, characteristics, etc., please refer to the relevant description of the above method embodiment.

Figure 9 is a schematic flow chart of an information processing module training method 900 according to another embodiment of the present application. The method may include one or more features of the above-mentioned method 800. In one embodiment, in step S820, the first communications device uses the multiple CSIs to train multiple first information processing modules and second information processing modules that need to be trained, thereby obtaining multiple trained first information processing modules and second information processing modules, and further includes:

S910: The first communication device groups the multiple CSIs to obtain multiple first input information groups;

S920: Use the one or more first input information groups as inputs to multiple first information processing modules that need to be trained, and obtain second output information output by the second information processing module;

S930. Optimize the loss function according to each first input information group and its corresponding second output information to obtain a plurality of trained first information processing modules and second information processing modules.

In an embodiment of the present application, multiple methods can be used to group multiple first input information in a training set to obtain one or more first input information groups. For example, the first input information is randomly grouped. The first input information of UE1 and UE2 is divided into the first group, the first input information of UE2 and UE4 is divided into the second group, and the first input information of UE3 and UE5 is divided into the third group. For another example, multiple first input information are grouped based on a user correlation threshold. For example, by comparing the similarity of the CSI-RS measurement results in multiple first input information, several first input information with similar Reference Signal Received Power (RSRP) values are not grouped together, which can reduce interference between multiple users. For another example, by comparing the similarity of the CSI-RS measurement results in multiple first input information, several first input information with overlapping channel main directions are not grouped together.

In an embodiment of the present application, if the CSI model to be trained includes multiple first information processing modules, such as multiple CSI encoder models and second information processing modules, such as a CSI decoder model, each first input information group can be used as the input of the multiple first information processing modules to obtain the second output information output by the second information processing module. Each first input information group and its corresponding second output information are then substituted into the loss function formula to obtain the loss function calculation result. Determine whether the loss function calculation result converges. If it converges, training can be stopped. If it does not converge, the first information processing module and the second information processing module can be adjusted, such as adjusting the parameters of the CSI encoder module and/or the CSI decoder model. Then, continue training using the new or original first input information group.

In one embodiment, the method further comprises:

S940. The first communication device sends the multiple first information processing modules to multiple second communication devices and/or sends the second information processing module to a third communication device.

For example, after the training of the computing unit, computing entity or computing node on the UE side is completed, the trained CSI encoder model is distributed to one or more UEs, and the trained CSI decoder model is distributed to the base station.

For another example, after the base station completes training, it distributes the trained CSI encoder model to one or more UEs.

For another example, after the computing unit, computing entity or computing node on the base station side completes training, the trained CSI encoder model is distributed to one or more UEs, and the trained CSI decoder model is distributed to the base station.

In an embodiment of the present application, if the UE has strong computing performance, multiple first information processing modules and second information processing modules can also be trained on the UE. After training is completed, the UE distributes the trained CSI decoder model to the base station. In this case, the first communication device and the second communication device can be understood as different components of the UE.

FIG10 is a schematic flow chart of an information processing method 1000 according to an embodiment of the present application. The method can optionally be applied to the system shown in FIG1 , but is not limited thereto. The method includes at least part of the following contents.

S1010. A first communication device sends first indication information, where the first indication information is used to indicate that a second communication device is scheduled for multi-user transmission.

In an embodiment of the present application, a first communication device may send first indication information to a second communication device. If the first communication device is a network device, the second communication device may be a terminal device. For example, a base station sends the first indication information to one or more UEs, indicating that the one or more UEs are scheduled for multi-user transmission.

In one embodiment, the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback. For example, if the first indication information instructs the UE to activate a CSI encoder model, the UE uses the CSI encoder model to process the CSI measured by the UE to obtain a bit sequence, and feeds back the bit sequence to the base station.

In one embodiment, the first communication device may activate its own second information processing module. For example, the first communication device may activate the second information processing module upon sending first indication information to the first communication device. For another example, the first communication device may activate the second information processing module upon sending first indication information to the first communication device and receiving feedback from one or more first communication devices indicating that the first information processing module has been activated.

In one embodiment, the first information processing module includes at least one of a first model, a first function, and a first characteristic.

In one embodiment, the second information processing module includes at least one of a second model, a second function, and a second characteristic.

In one embodiment, the first model includes a channel state information (CSI) encoder model, the first function includes a CSI encoder function, or the first characteristic includes a CSI encoder.

In one embodiment, the second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic.

In one embodiment, the first indication information is further used to indicate at least one of the following: the number of multi-users scheduled simultaneously; and the time for the scheduled users to perform multi-user transmission.

In an embodiment of the present application, if the number of simultaneously scheduled multiple users indicated in the first indication information is greater than 1, the second communication device may select a corresponding model, function, or feature from the simultaneously scheduled multiple users for activation. For example, if the first indication information indicates that UE1 and UE2 are scheduled simultaneously, UE1 may activate its own model M1 after receiving the first indication information; and UE2 may activate its own model M2 after receiving the first indication information. For another example, if the first indication information indicates that the UEs where M1 and M2 are located are scheduled simultaneously, UE1 may activate its own model M1 after receiving the first indication information; and UE2 may activate its own model M2 after receiving the first indication information.

In this embodiment of the present application, if the time for the scheduled user to perform multi-user transmission indicated in the first indication information includes T time units, the second communications device may use the first information processing module to provide multi-user CSI feedback within the subsequent T time units. After T time units, the second communications device may disable the first information processing module and return to the initial state.

FIG11 is a schematic flow chart of an information processing method 1100 according to another embodiment of the present application. The method may include one or more features of the above method 1000. In one embodiment, the method further includes:

S1110: A first communication device receives first reporting information, where the first reporting information is used to report the CSI of the second communication device. This step may be performed before S1010 to trigger the first communication device to execute S1010. For example, if the network device receives first reporting information from one or more second communication devices, it may send the first indication information described above to the second communication devices.

In one embodiment, the method further comprises:

S1120: The first communication device receives second reporting information, which is used to report whether the first information processing module of the second communication device is in an activated state. After S1010, if the second communication device activates the first information processing module, it may send second reporting information to the first communication device to inform the first communication device that the second communication device has activated the first information processing module. In this case, the first communication device may activate its own second information processing module.

In one embodiment, the method further comprises:

S1130, the first communication device sends a second indication message, and the second indication message is used to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module. For example, the first communication device can send a second indication message to one or more second communication devices. If the second indication message indicates to shut down or deactivate the first information processing module, the second communication device that receives the second indication message can shut down or deactivate the first information processing module. If the second indication message indicates to activate a new first information processing module, the second communication device that receives the second indication message can activate or switch to the new first information processing module. For example, two CSI encoder models M1 and M2 are deployed in UE1. After receiving the first indication message, the CSI encoder model M1 is activated, and after receiving the second indication message, it switches to activating the CSI encoder model M2.

In one embodiment, the method further includes: the first communication device receiving capability information, where the capability information is used to indicate the multi-user CSI feedback-related capability of the second communication device. This step can be combined with one or more steps in the above-mentioned information processing method and training method embodiments.

In one embodiment, the multi-user CSI feedback-related capabilities include at least one of the following:

Whether it supports AI/ML-based multi-user CSI feedback capabilities;

Whether the capability of deploying the first information processing module is supported;

Whether the capability of receiving the first indication information and/or the second indication information is supported.

In this embodiment of the present application, the first communication device may receive capability information from one or more second communication devices.

The capability information of different second communication devices may be the same or different. For example, the network device receives capability information of UE1, UE2, and UE3. The capability information of UE1 includes the capability to support AI/ML-based multi-user CSI feedback. The capability information of UE2 includes the capability to support the deployment of a CSI encoder model. The capability information of UE3 includes the capability to support the reception of the first indication information and the second indication information.

FIG12 is a schematic flow chart of an information processing method 1200 according to an embodiment of the present application. The method can optionally be applied to the system shown in FIG1 , but is not limited thereto. The method includes at least part of the following contents.

S1210. A second communication device receives first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

In one implementation, the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback.

In one embodiment, the first indication information is further used to indicate at least one of the following: the number of multi-users scheduled simultaneously; and the time for the scheduled users to perform multi-user transmission.

FIG13 is a schematic flow chart of an information processing method 1300 according to another embodiment of the present application. The method may include one or more features of the above method. In one embodiment, the method further includes:

S1310: The second communication device sends first reporting information, where the first reporting information is used to report the CSI of the second communication device.

In one embodiment, the method further comprises:

S1320: The second communication device sends second reporting information, where the second reporting information is used to report whether the first information processing module of the second communication device is in an activated state.

In one embodiment, the method further comprises:

S1330: The second communication device receives second indication information, where the second indication information is used to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module.

In one embodiment, the method further includes: the second communication device sending capability information, where the capability information is used to indicate the multi-user CSI feedback-related capability of the second communication device. This step can be combined with one or more steps in the above-mentioned information processing method and training method embodiments.

In one embodiment, the multi-user CSI feedback-related capabilities include at least one of the following:

Whether it supports AI/ML-based multi-user CSI feedback capabilities;

Whether the capability of deploying the first information processing module is supported;

Whether the capability of receiving the first indication information and/or the second indication information is supported.

For specific examples of the second communication device executing methods 1200 and 1300 in this embodiment, reference may be made to the relevant descriptions about the second communication device in the above methods 1000 and 1100 , which will not be repeated here for the sake of brevity.

The communication method of the embodiment of the present application may include an AI-based multi-user CSI feedback and precoding method. For example, a first artificial intelligence/machine learning (AI/ML) model/function/feature is deployed on multiple user sides, and matched with a second AI/ML model/function/feature on the network side to achieve multi-user joint CSI feedback. For another example, the downlink precoding vector of each user is directly obtained on the network side. Unlike the point-to-point single-user CSI feedback method adopted in the LTE/NR standard, the solution of the embodiment of the present application can adopt joint feedback, taking into account the interference characteristics between multiple users, and directly implement the output of the optimal precoding vector on the network side through the AI/ML model/function/feature, thereby maximizing the spectrum efficiency of downlink MU-MIMO. The embodiment of the present application also provides a model training method and signaling process for multi-user CSI feedback, as well as a corresponding terminal capability reporting method, to support a multi-user CSI feedback method based on AI/ML.

Example 1: AI-based multi-user CSI feedback deployment framework

This embodiment provides an implementation framework for an AI-based multi-user CSI feedback method. Figure 14 illustrates the AI-based multi-user CSI feedback framework, using a two-user deployment scenario as an example. The user side can deploy a first AI/ML model/function/feature, whose inputs are the first inputs of user 1 and user 2, respectively, and whose outputs are the first outputs of user 1 and user 2, respectively. The network side can deploy a second AI/ML model/function/feature, whose inputs are the second inputs and second outputs, respectively.

Specifically, the first AI/ML model/function/feature can be a CSI encoder model obtained through training, or other implementation methods for realizing the CSI compression coding function, such as a filter, an implementation algorithm, etc. The first input of user 1/user 2 is the CSI obtained by local measurement of user 1/user 2, which can be a feature vector (such as the input structure discussed in the R18 AI/ML CSI project, and the embodiment of the present application can use this input information as an example), can be original channel information, or can be other input forms that characterize CSI, which are not limited here. The first output of user 1/user 2 can be a bit sequence after the first input information passes through the first AI/ML model/function/feature, which can be reported through an uplink feedback channel, such as a physical uplink control channel (PUCCH) or a physical uplink shared channel (PUSCH).

The second AI/ML model/function/feature can be a CSI decoder model obtained through training, or it can be other implementation methods for realizing CSI recovery decoding function and precoding, such as a filter matching the first AI/ML model/function/feature, an implementation algorithm, etc. (no more examples are given here). The second input combines the first outputs from user 1 and user 2. For example, the first output of user 1 is a bit stream of length _m1 , and the first output of user 2 is a bit stream of length _m2 , then the second input is a bit stream of length _m1 + _m2 . The second output is the downlink precoding vector for user 1 and user 2 respectively. The precoding vector can have different granularities in the frequency domain, for example, it can be the entire bandwidth, the subband granularity, the resource block (RB) granularity, or the subcarrier granularity.

This implementation method can be extended to scenarios with multiple users, that is, the first AI/ML model/function/feature can be deployed on more than 2 users, and the maximum number of users that can be supported depends on the maximum number of multi-user transmissions configured in the system.

Unlike point-to-point CSI feedback, the solution of this embodiment matches the first AI/ML model/function/feature deployed on multiple user sides with a second AI/ML model/function/feature on the network side to perform joint CSI feedback. In addition, the second output of the second AI/ML model/function/feature on the network side may not be the CSI corresponding to each user, but directly output the downlink precoding vector corresponding to each user. This integrated design of CSI feedback and downlink precoding first considers the interference between multiple users and can maximize Spectral efficiency of downlink MU-MIMO. In addition, this embodiment implements precoding design with different frequency domain granularities by deploying a second AI/ML model/function/feature on the network side, which also reduces the complexity of system scheduling and precoding vector calculation after obtaining CSI.

Example 2: AI-based multi-user CSI feedback training method

This embodiment provides an AI-based multi-user CSI feedback training method for obtaining a first AI/ML model/function/feature and a second AI/ML model/function/feature that can be deployed in Example 1. Specifically, this embodiment includes training methods for both the UE and network sides, as detailed in the following sub-embodiments.

Sub-embodiment 1: UE-side training method

This sub-embodiment provides a UE-side training method, i.e., both the first AI/ML model/function/feature and the second AI/ML model/function/feature are performed on the UE side. Considering the UE's training capabilities and the fact that the first AI/ML model/function/feature needs to be deployed on multiple UEs, the training process needs to be performed on the computing entity/unit/node on the UE side, as shown in Figure 15.

The above training method may include the following steps:

S1, Data Collection: The UE sends a data set for training to the UE-side computing entity/unit/node. The data set may include CSIs measured by multiple UEs using downlink CSI-RSs.

S2, UE Grouping: Specifically, the UE-side computing entity/unit/node groups the data reported by multiple UEs to form a first input information group for model training, for example, {UE-1's first input information, UE-2's first input information, ..., UE-K's first input information}. Each group constitutes a training sample. This grouping process can be implemented in various ways, as shown in Examples S2a and S2b below.

S2a, one implementation method is random grouping, that is, the K UEs in each group come from a random combination of all UEs in all data collection. The corresponding assumption of this grouping method is that the network side does not schedule the pairing of multiple users but performs random pairing. However, the channel correlation between multiple users is high. If the interference is large, they may not be suitable for scheduling multi-user transmission. However, the advantages of this grouping method include: the training set samples of this situation are included in the training process, which to a certain extent ensures that the first AI/ML model/function/feature and the second AI/ML model/function/feature obtained by training have good generalization performance for different user scheduling situations, and random grouping is relatively simple to process the data set.

S2b: Grouping based on user correlation thresholds. This means that the user correlation between the K UEs in each group is below a certain threshold. This grouping approach takes into account the realistic assumptions of user scheduling on the network side. Its advantage is that the training set samples corresponding to this grouping are more likely to occur during multi-user transmission, with minimal interference between multiple users, making them suitable for scheduling multi-user transmission. However, for different multi-user scheduling assumptions, the generalization of the first and second trained AI/ML models/functions/features may be poor.

S3, model training. Based on the framework of Figure 16, the first input information grouped in S2 is used as the input of the first AI/ML model/function/feature, and the loss function uses the negative of the spectrum efficiency calculated based on the second output and the first input information group. By optimizing the loss function, the first and second AI/ML models/functions/features can obtain the optimal second output (e.g., precoding vectors for different users on the network side) based on the first input information group.

S4, model distribution. The UE-side computing entity/unit/node distributes the first AI/ML model/function/feature to different UEs and distributes the second AI/ML model/function/feature to the network for deployment.

Sub-embodiment 2: Network-side training method

This sub-embodiment provides a joint training method on the network side, i.e., both the first AI/ML model/function/feature and the second AI/ML model/function/feature are performed on the network side. Given the network's strong training capabilities, the training process can be implemented directly on the network-side base station or on the network-side computing entity/unit/node, as shown in Figure 16.

The above training methods include:

S1, Data Reporting and Data Collection. Multiple UEs measure the downlink CSI-RS to obtain CSI and feed the collected data back to the network via an uplink channel. This uplink channel can be the PUSCH, PUCCH, or other channel resources dedicated to data reporting. The network transmits this data set to the network-side computing entity/unit/node.

The UE grouping in step S2 and the model training in step S3 are the same as those in sub-embodiment 1 and will not be described in detail here.

S4, model distribution. The network-side computing entity/unit/node distributes the first AI/ML model/function/feature to different UEs and distributes the second AI/ML model/function/feature to the network side for deployment.

As described above, the first and second AI/ML model/function/feature training methods on the UE side or the network side generally adopt offline training and online deployment methods in consideration of complexity. However, this embodiment can also support online training and AI/ML model/function/feature updates, which will not be described in detail here. Through this training method, the user and the network side can obtain the first AI/ML model/function/feature and the second AI/ML model/function/feature for multi-user CSI feedback based on AI/ML. This training method is different from the point-to-point user CSI feedback method in that it is necessary to obtain joint pairing data of multiple users for training during the data collection phase, so that the first AI/ML model/function/feature can obtain the interference information characteristics between users, and thus effectively suppress multi-user interference in the multi-user precoding vector calculation.

Example 3: Signaling process supporting multi-user CSI feedback

The point-to-point CSI feedback design does not support this multi-user CSI joint feedback method. Therefore, this embodiment provides a signaling process that supports multi-user CSI feedback.

Sub-embodiment 1: Two-stage multi-user CSI feedback signaling process

First, this embodiment supports a two-stage multi-user CSI feedback signaling process, as shown in Figure 17. The process may include:

S1701: A user sends a first report to the network. At this point, the user, such as a UE, is in an initial state, i.e., not scheduled for multi-user transmission. The first report should include at least the CSI of the user. This first report is point-to-point and can employ either a traditional codebook approach or an AI/ML-based CSI feedback approach.

S1702: The network side performs multi-user scheduling based on first reporting information of multiple users.

S1703, the network side sends a first indication message to each scheduled user, and the first indication message indicates that the user is scheduled for multi-user transmission, which is used to activate the first AI/ML function/model/feature on the user side and perform multi-user CSI feedback. The first indication message can also indicate the number K of multi-users scheduled at the same time, so that the user can select the corresponding first AI/ML model/function/feature for activation. The first indication message can also additionally indicate the time T when the user is scheduled for multi-user transmission. Within the subsequent T time units after receiving the first information indication, the user can use the first AI/ML model/function/feature for multi-user CSI feedback. After T time units, the user can turn off the first AI/ML model/function/feature and return to the initial state. If the first indication message does not additionally indicate the time T, the default information defaults to continuous activation, that is, the first AI/ML model/function/feature continues to work until the second indication message is received (see step S7);

S1704: The user activates the first AI/ML model/function/feature according to the received first instruction information;

S1705: The user sends second reporting information to the network, where the second reporting information includes at least the first output information of the plurality of users and feedback information confirming activation of the first AI/ML model/function/feature;

S1706: The network side activates the second AI/ML function/model/feature and obtains the second output information ultimately required by the network side based on the second reported information.

S1707: The network side sends second indication information to the user, used to instruct the user side to shut down the first AI/ML model/function/feature, or to instruct the user side to switch to a new first AI/ML model/function/feature.

The first reporting information and the second reporting information may be transmitted via UCI signaling or other dedicated uplink signaling for carrying AI-related reporting information, and reported via PUCCH or PUSCH. The first indication information and the second indication information may be transmitted via DCI, MAC CE, or RRC signaling, or via other dedicated downlink signaling for carrying AI-related indication information.

This two-stage multi-user CSI feedback process requires the network to schedule multiple users based on point-to-point user CSI feedback information, and then determine user pairing and the selection of the first AI/ML model/function/feature. The advantage of this solution is that the network can provide relatively reasonable multi-user scheduling based on the CSI feedback of a single user and provide reasonable guidance on the selection of the first AI/ML model/function/feature for multiple users, resulting in relatively good multi-user transmission performance.

Sub-embodiment 2: One-stage multi-user CSI feedback signaling process

This embodiment also supports a one-stage multi-user CSI feedback signaling process, as shown in FIG18 :

The main difference between this process and the two-stage multi-user CSI feedback signaling process is that the first stage does not need to wait for the user's first reported information to be reported (i.e., step S1701 can be omitted), and the network side can directly perform blind scheduling for multiple users. The subsequent process is basically the same as the two-stage process (i.e., steps S1801 to S1806 can be referred to the relevant descriptions of steps S1702 to S1707, respectively), and will not be repeated here.

The multi-user CSI feedback signaling process in this phase is simpler. The network side does not need to wait for each user's CSI report and can directly trigger the multi-user CSI feedback signaling process. However, this process may cause unreasonable multi-user scheduling results on the network side, resulting in significant interference between users. This places higher requirements on the first and second AI/ML functions/models/features, namely, that these first and second AI/ML functions/models/features can better achieve interference suppression between multiple users and output more matching precoding vectors.

Example 4: Terminal Capability Reporting

This embodiment provides a terminal capability reporting method to support the terminal in implementing an AI/ML-based multi-user CSI feedback method.

The first capability of the terminal includes the ability to support AI/ML-based multi-user CSI feedback, the ability to support the deployment of the first AI/ML model/function/feature, and the ability to support the reception of the first indication information and the second indication information. The specific reporting methods are as follows:

Method 1: The terminal reports the capability of supporting AI/ML-based multi-user CSI feedback (first capability). A terminal with the first capability can also support the first AI/ML model/function/feature and can also support receiving the first indication information and the second indication information.

Method 2: The terminal reports the capability of supporting the first AI/ML model/function/feature (first capability). The terminal with the first capability can also support receiving the first indication information and the second indication information, and can also support AI/ML-based multi-user CSI feedback;

Method 3: The terminal reports the capability (first capability) of supporting the reception of the first indication information and the second indication information. The terminal with the first capability can also support the deployment of the first AI/ML model/function/feature, and can also support multi-user CSI feedback based on AI/ML.

FIG19 is a schematic block diagram of a first communication device 1900 according to an embodiment of the present application. The first communication device 1900 may include:

The first receiving unit 1901 is configured to receive a plurality of first output information from a plurality of second communication devices; wherein one first output information is obtained by a second communication device processing the first input information using the first information processing module;

The first processing unit 1902 is configured to obtain second input information according to the plurality of first output information; and process the second input information using a second information processing module to obtain second output information.

In one embodiment, the first information processing module includes at least one of a first model, a first function, and a first characteristic.

In one embodiment, the second information processing module includes at least one of a second model, a second function, and a second characteristic.

In one embodiment, the first model comprises a channel state information (CSI) encoder model, the first function comprises a CSI encoder function, or the first characteristic comprises a CSI encoder characteristic; or

The second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic.

In one embodiment, the first input information includes CSI measured by the second communication device, and the first output information includes a bit sequence obtained by the second communication device processing the CSI using a first information processing module.

In one embodiment, the second input information includes information obtained by combining the multiple first input information, and the second output information includes a precoding vector obtained by the first communication device processing the second input information using the second information processing module.

FIG20 is a schematic block diagram of a first communication device 2000 according to an embodiment of the present application. The first communication device 2000 may include:

The second receiving unit 2001 is configured to receive multiple CSIs from multiple second communication devices; wherein one CSI is obtained by one second communication device through CSI-RS measurement;

The second processing unit 2002 is configured to train a plurality of first information processing modules and a second information processing module that need to be trained using the plurality of CSIs to obtain a plurality of trained first information processing modules and a second information processing module.

In one embodiment, the second processing unit 2002 is further configured to:

Grouping the multiple CSIs to obtain multiple first input information groups;

Using the one or more first input information groups as inputs to a plurality of first information processing modules that need to be trained, and obtaining second output information output by a second information processing module;

The loss function is optimized according to each first input information group and its corresponding second output information to obtain a plurality of trained first information processing modules and second information processing modules.

In one embodiment, the first communication device further includes:

The first sending unit 2003 is configured to send the multiple first information processing modules to multiple second communication devices and/or send the second information processing module to a third communication device.

In one embodiment, the first information processing module includes at least one of a first model, a first function, and a first characteristic.

In one embodiment, the second information processing module includes at least one of a second model, a second function, and a second characteristic.

In one embodiment, the first model comprises a channel state information (CSI) encoder model, the first function comprises a CSI encoder function, or the first characteristic comprises a CSI encoder characteristic; or

The second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic.

FIG21 is a schematic block diagram of a first communication device 2100 according to an embodiment of the present application. The first communication device 2100 may include:

The second sending unit 2101 is configured to send first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

In one implementation, the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback.

In one embodiment, the first indication information is further used to indicate at least one of the following: the number of multi-users scheduled simultaneously; and the time for the scheduled users to perform multi-user transmission.

In one embodiment, the first communication device further includes:

The third receiving unit 2102 is configured to receive first reporting information, where the first reporting information is used to report the CSI of the second communication device.

In one implementation, the third receiving unit 2102 is further configured to receive second reporting information, where the second reporting information is used to report whether the first information processing module of the second communication device is in an activated state.

In one embodiment, the second sending unit 2101 is further configured to send second indication information, where the second indication information is configured to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module.

In one implementation, the third receiving unit 2102 is further configured to receive capability information, where the capability information is used to indicate the multi-user CSI feedback-related capability of the second communication device.

In one embodiment, the multi-user CSI feedback-related capabilities include at least one of the following:

Whether it supports AI/ML-based multi-user CSI feedback capabilities;

Whether the capability of deploying the first information processing module is supported;

Whether the capability of receiving the first indication information and/or the second indication information is supported.

The first communication devices 1900, 2000, and 2100 of the embodiments of the present application can implement the corresponding functions of the first communication devices in the aforementioned method embodiments. The processes, functions, implementation methods, and beneficial effects corresponding to the various modules (submodules, units, or components, etc.) in the first communication devices 1900, 2000, and 2100 can be found in the corresponding descriptions in the aforementioned method embodiments, and will not be repeated here. It should be noted that the functions described in the various modules (submodules, units, or components, etc.) in the first communication devices 1900, 2000, and 2100 of the embodiments of the application are as follows: It can be implemented by different modules (sub-modules, units or components, etc.) or by the same module (sub-module, unit or component, etc.).

FIG22 is a schematic block diagram of a second communication device 2200 according to an embodiment of the present application. The second communication device 2200 may include:

The fourth receiving unit 2201 is configured to receive first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

In one implementation, the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback.

In one embodiment, the first indication information is further used to indicate at least one of the following: the number of multi-users scheduled simultaneously; and the time for the scheduled users to perform multi-user transmission.

In one embodiment, the second communication device further includes:

The third sending unit 2202 is configured to send first reporting information, where the first reporting information is used to report the CSI of the second communication device.

In one implementation, the third sending unit 2202 is further configured to send second reporting information, where the second reporting information is used to report whether the first information processing module of the second communication device is in an activated state.

In one embodiment, the fourth receiving unit 2201 is further configured to receive second indication information, where the second indication information is configured to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module.

In one implementation, the third sending unit 2202 is further configured to send capability information, where the capability information is used to indicate the multi-user CSI feedback-related capability of the second communication device.

In one embodiment, the multi-user CSI feedback-related capabilities include at least one of the following:

Whether it supports AI/ML-based multi-user CSI feedback capabilities;

Whether the capability of deploying the first information processing module is supported;

Whether the capability of receiving the first indication information and/or the second indication information is supported.

The second communication device 2200 of the embodiment of the present application can implement the corresponding functions of the second communication device in the aforementioned method embodiment. The processes, functions, implementation methods and beneficial effects corresponding to the various modules (sub-modules, units or components, etc.) in the second communication device 2200 can be found in the corresponding descriptions in the above-mentioned method embodiments, and will not be repeated here. It should be noted that the functions described in the various modules (sub-modules, units or components, etc.) in the second communication device 2200 of the embodiment of the application can be implemented by different modules (sub-modules, units or components, etc.) or by the same module (sub-module, unit or component, etc.).

Figure 23 is a schematic structural diagram of a communication device 2300 according to an embodiment of the present application. The communication device 2300 includes a processor 2310, which can call and run a computer program from a memory to enable the communication device 2300 to implement the method in the embodiment of the present application.

In one embodiment, the communication device 2300 may further include a memory 2320. The processor 2310 may call and execute a computer program from the memory 2320 to enable the communication device 2300 to implement the method in the embodiment of the present application.

The memory 2320 may be a separate device independent of the processor 2310 or may be integrated into the processor 2310 .

In one embodiment, the communication device 2300 may further include a transceiver 2330 , and the processor 2310 may control the transceiver 2330 to communicate with other devices. Specifically, the transceiver 2330 may send information or data to other devices, or receive information or data sent by other devices.

The transceiver 2330 may include a transmitter and a receiver. The transceiver 2330 may further include an antenna, and the number of antennas may be one or more.

In one embodiment, the communication device 2300 may be the first communication device of the embodiment of the present application, and the communication device 2300 may implement the corresponding processes implemented by the first communication device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

In one embodiment, the communication device 2300 may be the second communication device of the embodiment of the present application, and the communication device 2300 may implement the corresponding processes implemented by the second communication device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

24 is a schematic structural diagram of a chip 2400 according to an embodiment of the present application. The chip 2400 includes a processor 2410, which can call and execute a computer program from a memory to implement the method according to the embodiment of the present application.

In one embodiment, the chip 2400 may further include a memory 2420. The processor 2410 may call and execute a computer program from the memory 2420 to implement the method executed by the terminal device or the network device in the embodiment of the present application.

The memory 2420 may be a separate device independent of the processor 2410 , or may be integrated into the processor 2410 .

In one embodiment, the chip 2400 may further include an input interface 2430. The processor 2410 may control the input interface 2430 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

In one embodiment, the chip 2400 may further include an output interface 2440. The processor 2410 may control the output interface 2440 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

In one embodiment, the chip can be applied to the first communication device in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the first communication device in each method of the embodiment of the present application. For the sake of brevity, it will not be repeated here.

In one embodiment, the chip can be applied to the second communication device in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the second communication device in each method of the embodiment of the present application. For the sake of brevity, it will not be repeated here.

The chips used in the first communication device and the second communication device may be the same chip or different chips.

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

The processor mentioned above may be a general-purpose processor, a digital signal processor (DSP), a field programmable gate array (FPGA), an application-specific integrated circuit (ASIC), or other programmable logic devices, transistor logic devices, discrete hardware components, etc. The general-purpose processor mentioned above may be a microprocessor or any conventional processor, etc.

The memory mentioned above may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM).

It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

FIG25 is a schematic block diagram of a communication system 2500 according to an embodiment of the present application. The communication system 2500 includes a first communication device 2510 and a second network device 2520 .

In one embodiment, a first communication device 2510 is configured to receive multiple first output information from multiple second communication devices 2520. Each first output information is obtained by a second communication device processing first input information using a first information processing module. The first communication device obtains second input information based on the multiple first output information. The first communication device processes the second input information using a second information processing module to obtain second output information. The second communication device 2520 is configured to transmit the first output information.

In one embodiment, a first communication device 2510 is configured to receive multiple CSIs from multiple second communication devices, wherein each CSI is obtained by a second communication device through CSI-RS measurement. The first communication device uses the multiple CSIs to train multiple first information processing modules and second information processing modules that require training, thereby obtaining multiple trained first information processing modules and second information processing modules. The second communication device 2520 is configured to send the CSIs.

In one implementation, the first communication device 2510 is configured to send first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission.

In one implementation, the second communication device 2520 is configured to receive the first indication information.

The first communication device 2510 can be used to implement the corresponding functions implemented by the first communication device in the above method, and the second communication device 2520 can be used to implement the corresponding functions implemented by the second communication device in the above method. For the sake of brevity, they are not described here in detail.

In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function in accordance with the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims (57)

An information processing method, comprising: The first communication device receives a plurality of first output information from a plurality of second communication devices; wherein one first output information is obtained by a second communication device processing the first input information using the first information processing module; The first communication device obtains second input information according to the plurality of first output information; The first communication device uses a second information processing module to process the second input information to obtain second output information. The method according to claim 1, wherein the first information processing module includes at least one of a first model, a first function and a first characteristic, and the second information processing module includes at least one of a second model, a second function and a second characteristic. The method of claim 2, wherein the first model comprises a channel state information (CSI) encoder model, the first function comprises a CSI encoder function, or the first characteristic comprises a CSI encoder characteristic; or The second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic. The method according to any one of claims 1 to 3, wherein the first input information includes CSI measured by the second communication device, and the first output information includes a bit sequence obtained by the second communication device processing the CSI using a first information processing module. The method according to any one of claims 1 to 4, wherein the second input information includes information obtained by merging the multiple first input information, and the second output information includes a precoding vector obtained by the first communication device processing the second input information using the second information processing module. A method for training an information processing module, comprising: The first communication device receives multiple CSIs from multiple second communication devices; wherein one CSI is obtained by one second communication device through CSI-RS measurement; The first communication device uses the multiple CSIs to train multiple first information processing modules and second information processing modules that need to be trained to obtain multiple trained first information processing modules and second information processing modules. The method according to claim 6, wherein the first communications device uses the multiple CSIs to train multiple first information processing modules and second information processing modules that need to be trained to obtain the trained multiple first information processing modules and second information processing modules, further comprising: The first communication device groups the multiple CSIs to obtain multiple first input information groups; Using the one or more first input information groups as inputs to a plurality of first information processing modules that need to be trained, to obtain second output information output by a second information processing module; The loss function is optimized according to each first input information group and its corresponding second output information to obtain a plurality of trained first information processing modules and second information processing modules. The method according to claim 6 or 7, wherein the method further comprises: The first communication device sends the multiple first information processing modules to multiple second communication devices and/or sends the second information processing module to a third communication device. The method according to any one of claims 6 to 8, wherein the first information processing module includes at least one of a first model, a first function and a first characteristic, and the second information processing module includes at least one of a second model, a second function and a second characteristic. The method of claim 9, wherein the first model comprises a channel state information (CSI) encoder model, the first function comprises a CSI encoder function, or the first characteristic comprises a CSI encoder characteristic; or The second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic. An information processing method, comprising: The first communication device sends first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission. The method according to claim 11, wherein the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback. The method according to claim 11 or 12, wherein the first indication information is further used to indicate at least one of the following: The number of users scheduled simultaneously; The time during which the scheduled users perform multi-user transmission. The method according to any one of claims 11 to 13, wherein the method further comprises: The first communication device receives first reporting information, where the first reporting information is used to report the CSI of the second communication device. The method according to any one of claims 11 to 14, wherein the method further comprises: The first communication device receives second reporting information, and the second reporting information is used to report the first information processing of the second communication device. Check whether the management module is activated. The method according to any one of claims 11 to 15, wherein the method further comprises: The first communication device sends second indication information, where the second indication information is used to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module. The method according to any one of claims 11 to 16, wherein the method further comprises: The first communication device receives capability information, where the capability information is used to indicate a multi-user CSI feedback-related capability of the second communication device. The method according to any one of claims 11 to 17, wherein the multi-user CSI feedback-related capabilities include at least one of the following: Whether it supports AI/ML-based multi-user CSI feedback capabilities; Whether the capability of deploying the first information processing module is supported; Whether the capability of receiving the first indication information and/or the second indication information is supported. An information processing method, comprising: The second communication device receives first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission. The method according to claim 19, wherein the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback. The method according to claim 19 or 20, wherein the first indication information is further used to indicate at least one of the following: The number of users scheduled simultaneously; The time during which the scheduled users perform multi-user transmission. The method according to any one of claims 19 to 21, further comprising: The second communication device sends first reporting information, where the first reporting information is used to report the CSI of the second communication device. The method according to any one of claims 19 to 22, wherein the method further comprises: The second communication device sends second reporting information, where the second reporting information is used to report whether the first information processing module of the second communication device is in an activated state. The method according to any one of claims 19 to 23, wherein the method further comprises: The second communication device receives second indication information, where the second indication information is used to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module. The method according to any one of claims 19 to 24, further comprising: The second communication device sends capability information, where the capability information is used to indicate a multi-user CSI feedback-related capability of the second communication device. The method according to any one of claims 19 to 25, wherein the multi-user CSI feedback-related capabilities include at least one of the following: Whether it supports AI/ML-based multi-user CSI feedback capabilities; Whether the capability of deploying the first information processing module is supported; Whether the capability of receiving the first indication information and/or the second indication information is supported. A first communication device, comprising: A first receiving unit is configured to receive a plurality of first output information from a plurality of second communication devices; wherein one first output information is obtained by a second communication device processing the first input information using the first information processing module; The first processing unit is configured to obtain second input information according to the plurality of first output information; and process the second input information using a second information processing module to obtain second output information. The first communication device according to claim 27, wherein the first information processing module includes at least one of a first model, a first function, and a first characteristic, and the second information processing module includes at least one of a second model, a second function, and a second characteristic. The first communications device of claim 28, wherein the first model comprises a channel state information (CSI) encoder model, the first function comprises a CSI encoder function, or the first characteristic comprises a CSI encoder characteristic; or The second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic. The first communication device according to any one of claims 27 to 29, wherein the first input information includes CSI measured by the second communication device, and the first output information includes a bit sequence obtained by the second communication device processing the CSI using a first information processing module. The first communication device according to any one of claims 27 to 30, wherein the second input information includes information obtained by merging the multiple first input information, and the second output information includes a precoding vector obtained by the first communication device processing the second input information using the second information processing module. A first communication device, comprising: A second receiving unit is configured to receive a plurality of CSIs from a plurality of second communication devices; wherein one CSI is obtained by one second communication device through CSI-RS measurement; The second processing unit is configured to train the plurality of first information processing modules and second information processing modules that need to be trained using the plurality of CSIs to obtain the plurality of trained first information processing modules and second information processing modules. The first communication device according to claim 32, wherein the second processing unit is further configured to: Grouping the multiple CSIs to obtain multiple first input information groups; Using the one or more first input information groups as inputs to a plurality of first information processing modules that need to be trained, to obtain second output information output by a second information processing module; The loss function is optimized according to each first input information group and its corresponding second output information to obtain a plurality of trained first information processing modules and second information processing modules. The first communication device according to claim 32 or 33, wherein the first communication device further comprises: The first sending unit is configured to send the multiple first information processing modules to multiple second communication devices and/or send the second information processing module to a third communication device. The first communication device according to any one of claims 32 to 34, wherein the first information processing module includes at least one of a first model, a first function and a first characteristic, and the second information processing module includes at least one of a second model, a second function and a second characteristic. The first communications device of claim 35, wherein the first model comprises a channel state information (CSI) encoder model, the first function comprises a CSI encoder function, or the first characteristic comprises a CSI encoder characteristic; or The second model includes a CSI decoder model, the second function includes a CSI decoder function, or the second characteristic includes a CSI decoder characteristic. A first communication device, comprising: The second sending unit is configured to send first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission. The first communication device according to claim 37, wherein the first indication information is used to activate the first information processing module of the second communication device to perform multi-user CSI feedback. The first communications device according to claim 37 or 38, wherein the first indication information is further used to indicate at least one of the following: The number of users scheduled simultaneously; The time during which the scheduled users perform multi-user transmission. The first communication device according to any one of claims 37 to 39, wherein the first communication device further comprises: The third receiving unit is configured to receive first reporting information, where the first reporting information is used to report the CSI of the second communication device. According to the first communication device according to any one of claims 37 to 40, the third receiving unit is further used to receive second reporting information, and the second reporting information is used to report whether the first information processing module of the second communication device is in an activated state. According to the first communication device according to any one of claims 37 to 41, the second sending unit is further used to send second indication information, and the second indication information is used to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module. The first communication device according to any one of claims 37 to 42, wherein the third receiving unit is further used to receive capability information, wherein the capability information is used to indicate the multi-user CSI feedback-related capability of the second communication device. The first communications device according to any one of claims 37 to 43, wherein the multi-user CSI feedback-related capabilities include at least one of the following: Whether it supports AI/ML-based multi-user CSI feedback capabilities; Whether the capability of deploying the first information processing module is supported; Whether the capability of receiving the first indication information and/or the second indication information is supported. A second communication device, comprising: The fourth receiving unit is used to receive first indication information, where the first indication information is used to indicate that the second communication device is scheduled for multi-user transmission. The second communication device according to claim 45, wherein the first indication information is used to activate a first information processing module of the second communication device to perform multi-user CSI feedback. The second communication device according to claim 45 or 46, wherein the first indication information is further used to indicate at least one of the following: The number of users scheduled simultaneously; The time during which the scheduled users perform multi-user transmission. The second communication device according to any one of claims 45 to 47, wherein the second communication device further comprises: The third sending unit is configured to send first reporting information, where the first reporting information is used to report the CSI of the second communication device. The second communication device according to any one of claims 45 to 48, wherein the third sending unit is further used to send second reporting information, and the second reporting information is used to report whether the first information processing module of the second communication device is in an activated state. The second communication device according to any one of claims 45 to 49, wherein the fourth receiving unit is further used to receive second indication information, and the second indication information is used to instruct the second communication device to shut down the first information processing module and/or switch to a new first information processing module. The second communication device according to any one of claims 45 to 50, wherein the third sending unit is further used to send capability information, wherein the capability information is used to indicate the multi-user CSI feedback-related capability of the second communication device. The second communications device according to any one of claims 45 to 51, wherein the multi-user CSI feedback-related capabilities include at least one of the following: Whether it supports AI/ML-based multi-user CSI feedback capabilities; Whether the capability of deploying the first information processing module is supported; Whether the capability of receiving the first indication information and/or the second indication information is supported. A communication device comprises: a transceiver, a processor and a memory, wherein the memory is used to store a computer program, the transceiver is used to communicate with other devices, and the processor is used to call and run the computer program stored in the memory so that the terminal communication device executes the method as described in any one of claims 1 to 26. A chip comprises: a processor configured to call and run a computer program from a memory, so that a device equipped with the chip executes the method according to any one of claims 1 to 26. A computer-readable storage medium for storing a computer program, wherein when the computer program is executed by a device, the device is caused to perform the method according to any one of claims 1 to 26. A computer program product comprising computer program instructions, wherein the computer program instructions enable a computer to execute the method according to any one of claims 1 to 26. A computer program causing a computer to execute the method according to any one of claims 1 to 26.