A CSI transmitting and receiving method and apparatus

Abstract

A CSI sending and receiving method and device, in which a terminal sends CSI of a target layer to an access network device according to an indication of the access network device, the target layer being any layer. In contrast to a conventional CSI feedback in which a terminal sends a PMI of n layers with the strongest quality to an access network device, the CSI fed back by the terminal can meet the needs of the access network device, reducing unnecessary air interface overhead of the terminal sending CSI.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a CSI transmission and reception method and apparatus. Background Technology

[0002] In wireless communication networks, such as mobile communication networks, the services supported by the networks are becoming increasingly diverse, thus requiring increasingly diverse demands. For example, networks need to support ultra-high speeds, ultra-low latency, and / or massive connectivity. This characteristic makes network planning, network configuration, and / or resource scheduling increasingly complex. Furthermore, as network functions become more powerful, such as supporting higher spectrum levels, supporting higher-order multiple-input multiple-output (MIMO) technologies, supporting beamforming, and / or supporting beam management and other new technologies, these new demands, scenarios, and characteristics bring unprecedented challenges to network planning, operation, and efficient operation. To meet these challenges, artificial intelligence (AI) technology can be introduced into wireless communication networks to achieve network intelligence. Based on this, how to send and receive channel state information (CSI) in AI-powered wireless communication networks is a problem worthy of research. Summary of the Invention

[0003] This application provides a CSI transmission and reception method and apparatus to meet the CSI requirements of artificial intelligence wireless communication networks.

[0004] In a first aspect, a CSI transmission method is provided, applied to a first communication device, the first communication device being a terminal, or a chip or circuit applied in the terminal, comprising: receiving first information from a second communication device, the first information including indication information for instructing the first communication device to transmit channel state information (CSI) of a target layer; and sending a first CSI report to the second communication device, the first CSI report including the CSI of the target layer.

[0005] Through the above implementation, the terminal sends a target-layer CSI to the access network device according to the device's instructions. This target layer can be any layer. Compared to the traditional CSI feedback where the terminal sends the highest quality n-layer PMI to the access network device, the terminal's feedback CSI can meet the access network device's needs, reducing the air interface overhead of sending unnecessary CSI.

[0006] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix, or the compression and / or quantization information of the channel response of the target layer.

[0007] Through the above implementation, the terminal sends the target layer feature vector matrix or channel response compression and / or quantization information to the access network device. Compared with the terminal directly sending the target layer feature vector matrix or channel response to the access network device, the air interface overhead of the terminal sending CSI can be reduced.

[0008] In one implementation, the CSI of the target layer further includes one or more of the following indication information: eigenvalues ​​of the feature vector matrix of the target layer, singular values, or the target layer itself.

[0009] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0010] In the above implementation, the CSI fed back by the terminal to the access network equipment can be called true CSI. True CSI and traditional CSI can be fed back together. For example, a new type of CSI report can be designed that can simultaneously feed back true CSI and traditional CSI. The joint feedback of true CSI and traditional CSI can improve the efficiency of CSI feedback.

[0011] In one implementation, the method further includes: sending a second CSI report to the second communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0012] In the above implementation, the CSI fed back by the terminal to the access network device can be called the truth CSI. The truth CSI and the traditional CSI can be fed back independently. For example, a new CSI type can be designed for feeding back truth CSI, etc. The feedback of truth CSI and traditional CSI does not affect each other.

[0013] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, threshold values ​​of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, transmission period of the CSI of the target layer, or transmission pattern of the CSI of the target layer.

[0014] In one implementation, the CSI of the target layer sent by the first communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0015] In the above implementation, the first CSI report reported by the terminal to the access network device can be used for AI-related data collection, such as model training, model inference, model testing, model verification, or model monitoring. The terminal sends the corresponding target layer CSI to the access network device according to the target layer indicated by the access network device, which can meet the AI ​​data needs of the access network device.

[0016] In one implementation, the first information is configuration information, which is used to configure the first communication device to send the CSI of the target layer.

[0017] In a second aspect, a CSI receiving method is provided, which is a method corresponding to the first aspect. The beneficial effects can be found in the first aspect. The method is applied to a second communication device, which is an access network device, or a chip or circuit in the access network device, etc. The method includes: sending first information to a first communication device, the first information including indication information for instructing the first communication device to send CSI of the target layer channel state information; and receiving a first CSI report from the first communication device, the first CSI report including the CSI of the target layer.

[0018] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix, or the compression and / or quantization information of the channel response of the target layer.

[0019] In one implementation, the CSI of the target layer further includes one or more of the following indication information: eigenvalues, singular values, or the target layer itself.

[0020] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0021] In one implementation, the method further includes: receiving a second CSI report from the first communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0022] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, threshold values ​​of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, transmission period of the CSI of the target layer, or transmission pattern of the CSI of the target layer.

[0023] In one implementation, the CSI of the target layer received by the second communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0024] In one implementation, the first information is configuration information, which is used to configure the first communication device to send the CSI of the target layer.

[0025] Thirdly, a CSI transmission method is provided. In one implementation, it is applied to a first communication device, which is a terminal, or a chip or circuit applied in the terminal. Alternatively, the first communication device is an access network device, or a chip or circuit applied in the access network device, comprising: sending first information to a second communication device, the first information including indication information for instructing the first communication device to transmit channel state information (CSI) of the target layer; and sending a first CSI report to the second communication device, the first CSI report including the CSI of the target layer.

[0026] In the above implementation, the terminal can proactively send target layer CSI to the access network device. For example, the terminal sends target layer indication information to the access network device; the terminal sends target layer CSI to the access network device. Alternatively, the access network device can proactively send target layer CSI to the terminal. For example, the access network device sends target layer indication information to the terminal; the access network device sends target layer CSI to the terminal, etc. The terminal or access network device can use the collected target layer CSI for AI processing to meet the AI's requirements for CSI data.

[0027] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix of the target layer, or the compression and / or quantization information of the channel response of the target layer.

[0028] In one implementation, the CSI of the target layer may further include one or more of the following indication information: eigenvalues, singular values, or the target layer itself.

[0029] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0030] In one implementation, the method further includes: sending a second CSI report to the second communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0031] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following of the target layer: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, a threshold value of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, the transmission period of the CSI of the target layer, or the transmission pattern of the CSI of the target layer.

[0032] In one implementation, the CSI of the target layer sent by the first communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0033] In one implementation, the first information is indication information, which is used to instruct the first communication device to send the CSI of the target layer.

[0034] Fourthly, a CSI receiving method is provided, which is the counterpart to the aforementioned third aspect. The beneficial effects of this fourth aspect are described in the third aspect. It is applied to a second communication device, which is an access network device, or a chip or circuit applied within an access network device, or a terminal, or a chip or circuit applied within a terminal, etc. The method includes: receiving first information from a first communication device, the first information including indication information for instructing the first communication device to transmit channel state information (CSI) of the target layer; and receiving a first CSI report from the first communication device, the first CSI report including the CSI of the target layer.

[0035] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix of the target layer, or the compression and / or quantization information of the channel response of the target layer.

[0036] In one implementation, the CSI of the target layer may further include one or more of the following indication information: features of the feature vector matrix of the target layer, singular values, or the target layer itself.

[0037] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0038] In one implementation, the method further includes: receiving a second CSI report from the first communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0039] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following of the target layer: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, a threshold value of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, the transmission period of the CSI of the target layer, or the transmission pattern of the CSI of the target layer.

[0040] In one implementation, the CSI of the target layer received by the second communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0041] In one implementation, the first information is indication information, which is used to instruct the first communication device to send the CSI of the target layer.

[0042] Fifthly, an apparatus is provided, comprising a unit or module corresponding to performing the method described in any one of the first to fourth aspects, wherein the unit or module may be implemented by hardware circuitry, or by software, or by a combination of hardware circuitry and software.

[0043] A sixth aspect provides an apparatus comprising a processor and an interface circuit, the processor being configured to communicate with other devices via the interface circuit and to perform the methods described in any one of the first to fourth aspects. The processor may include one or more processors.

[0044] A seventh aspect provides an apparatus comprising a processor coupled to a memory for executing a program stored in the memory to perform the methods described in any one of the first to fourth aspects. The memory may be located within or outside the apparatus. Furthermore, there may be one or more processors.

[0045] Eighthly, an apparatus is provided, including a processor and a memory; the memory is used to store computer instructions, and when the apparatus is in operation, the processor executes the computer instructions stored in the memory to cause the apparatus to perform the methods described in any one of the first to fourth aspects above.

[0046] Ninth aspect, a chip system is provided, comprising: a processor or circuitry for performing the methods described in any one of the first to fourth aspects.

[0047] In a tenth aspect, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a communication device, cause the methods described in any one of the first to fourth aspects to be performed.

[0048] Eleventhly, a computer program product is provided, comprising a computer program or instructions that, when executed by a device, cause the methods described in any one of the first to fourth aspects to be performed.

[0049] In a twelfth aspect, a system is provided, comprising a first communication device that performs the method of the first aspect and a second communication device that performs the method of the second aspect, or comprising a first communication device that performs the method of the third aspect and a second communication device that performs the method of the fourth aspect. Attached Figure Description

[0050] Figure 1 A schematic diagram of the communication system provided in the embodiments of this application;

[0051] Figure 2 A schematic diagram illustrating the deployment of an AI model provided in an embodiment of this application;

[0052] Figure 3 Another schematic diagram illustrating the deployment of an AI model provided in an embodiment of this application;

[0053] Figure 4 A schematic diagram of the access network device architecture provided in an embodiment of this application;

[0054] Figure 5 This is another schematic diagram of the access network device architecture provided in the embodiments of this application;

[0055] Figure 6 A schematic diagram illustrating the application architecture of the AI ​​model provided in the embodiments of this application;

[0056] Figure 7 A schematic diagram of a neuron provided in an embodiment of this application;

[0057] Figure 8 A schematic diagram of the layer relationships of a neural network provided in an embodiment of this application;

[0058] Figure 9 A flowchart illustrating CSI transmission and reception provided in this application embodiment;

[0059] Figure 10 A schematic diagram illustrating the use of a model to send and receive CSI, provided for embodiments of this application;

[0060] Figure 11 Another flowchart for CSI transmission and reception provided in the embodiments of this application;

[0061] Figure 12 Another flowchart for CSI transmission and reception provided in the embodiments of this application;

[0062] Figure 13 A schematic diagram of the device provided in the embodiments of this application;

[0063] Figure 14 Another schematic diagram of the device provided in the embodiments of this application. Detailed Implementation

[0064] Figure 1 This is a schematic diagram of the architecture of the communication system 1000 to which this application can be applied. For example... Figure 1 As shown, the communication system 1000 includes a wireless access network 100 and a core network 200. Optionally, the communication system 1000 may also include an Internet 300.

[0065] The wireless access network 100 may include at least one access network device (such as...) Figure 1 100a and 110b in the above), may also include at least one terminal (such as Figure 1 (Referring to 120a-120j in the original text). Terminals connect wirelessly to access network equipment, which in turn connects to the core network wirelessly or via a wired connection. Core network equipment and access network equipment can be independent physical devices, or the functions of the core network equipment and the logical functions of the access network equipment can be integrated onto the same physical device, or a single physical device can integrate some of the functions of both core network and access network equipment. Terminals can connect to each other, and access network equipment can connect to each other, via wired or wireless connections. Figure 1 This is just a schematic diagram. The communication system 1000 may also include other network devices, such as wireless repeaters and wireless backhaul devices. Figure 1 It is not shown in the middle.

[0066] Access network equipment can be a base station, an evolved NodeB (eNodeB), a transmission reception point (TRP), a next-generation NodeB (gNB) in a 5G mobile communication system, access network equipment in an open radio access network (O-RAN), a next-generation base station in a 6G mobile communication system, a base station in a future mobile communication system, or an access node in a wireless fidelity (WiFi) system; or it can be a module or unit that performs some of the functions of a base station, such as a central unit (CU), a distributed unit (DU), a central unit control plane (CU-CP) module, or a central unit user plane (CU-UP) module. Access network equipment can also be a macro base station (such as...). Figure 1 110a in the text), can also be a micro base station or an indoor station (such as... Figure 1 In section 110b), it can also be a relay node or a donor node, etc. This application does not limit the specific technology or equipment form used in the access network equipment.

[0067] In this application, the apparatus for implementing the functions of an access network device can be the access network device itself; it can also be an apparatus capable of supporting the access network device in implementing the functions, such as a chip system, hardware circuit, software module, or hardware circuit plus software module. This apparatus can be installed in the access network device or can be used in conjunction with the access network device. In this application, the chip system can be composed of chips or can include chips and other discrete components. For ease of description, the following description uses the example of an access network device as the apparatus for implementing the functions of the access network device to describe the technical solution provided in this application.

[0068] 1) Protocol layer structure.

[0069] Communication between access network devices and terminals follows a specific protocol layer structure. This protocol layer structure can include a control plane protocol layer structure and a user plane protocol layer structure. For example, the control plane protocol layer structure can include the functions of protocol layers such as the radio resource control (RRC) layer, the packet data convergence protocol (PDCP) layer, the radio link control (RLC) layer, the media access control (MAC) layer, and the physical layer. Similarly, the user plane protocol layer structure can include the functions of protocol layers such as the PDCP layer, the RLC layer, the MAC layer, and the physical layer. In one possible implementation, a service data adaptation protocol (SDAP) layer can be included above the PDCP layer.

[0070] Optionally, the protocol layer structure between the access network device and the terminal may also include an artificial intelligence (AI) layer for transmitting AI-related data.

[0071] 2) Central unit (CU) and distributed unit (DU).

[0072] Access network equipment may include CUs and DUs. Multiple DUs can be centrally controlled by a single CU. As an example, the interface between the CU and DU can be called an F1 interface. The control plane (CP) interface can be F1-C, and the user plane (UP) interface can be F1-U. This application does not limit the specific names of the interfaces. CUs and DUs can be distinguished according to the protocol layer of the wireless network: for example, the functions of the PDCP layer and above are located in the CU, and the functions of the protocol layers below the PDCP layer (such as RLC and MAC layers) are located in the DU; or, for example, the functions of the PDCP layer and above are located in the CU, and the functions of the PDCP layer and below are located in the DU, without restriction.

[0073] The above division of CU and DU processing functions according to protocol layers is merely an example; other methods can also be used. For instance, CU or DU can be divided into functions with more protocol layers, or they can be divided into partial processing functions with protocol layers. In one implementation, some functions of the RLC layer and the protocol layer functions above the RLC layer are located in the CU, while the remaining functions of the RLC layer and the protocol layer functions below the RLC layer are located in the DU. In another implementation, the functions of CU or DU can be divided according to service type or other system requirements, such as by latency. Functions that need to meet latency requirements are located in the DU, while functions that do not need to meet this latency requirement are located in the CU. In another design, the CU can also have one or more core network functions. For example, the CU can be located on the network side for convenient centralized management. In another design, the radio unit (RU) of the DU is remotely located. Optionally, the RU can have radio frequency functions.

[0074] Optionally, DU and RU can be partitioned at the physical layer (PHY). For example, DU can implement higher-level functions in the PHY layer, and RU can implement lower-level functions. When used for transmission, the PHY layer functions may include one or more of the following: adding cyclic redundancy check (CRC) codes, channel coding, rate matching, scrambling, modulation, layer mapping, precoding, resource mapping, physical antenna mapping, or radio frequency (RF) transmission functions. When used for reception, the PHY layer functions may include one or more of the following: CRC check, channel decoding, rate matching descrambling, descrambling, demodulation, layer mapping demapping, channel detection, resource demapping, physical antenna demapping, or RF reception functions. Higher-level functions in the PHY layer may include a subset of the PHY layer's functions, for example, functions closer to the MAC layer; lower-level functions in the PHY layer may include another subset of the PHY layer's functions, for example, functions closer to the RF functions. For example, higher-level functions in the PHY layer may include adding CRC codes, channel coding, rate matching, scrambling, modulation, and layer mapping, while lower-level functions may include precoding, resource mapping, physical antenna mapping, and radio frequency transmission. Alternatively, higher-level functions in the PHY layer may include adding CRC codes, channel coding, rate matching, scrambling, modulation, layer mapping, and precoding, while lower-level functions may include resource mapping, physical antenna mapping, and radio frequency transmission. For example, higher-level functions in the PHY layer may include CRC checksum, channel decoding, rate matching de-matching, decoding, demodulation, and layer mapping de-matching, while lower-level functions may include channel detection, resource de-mapping, physical antenna de-mapping, and radio frequency reception. Alternatively, higher-level functions in the PHY layer may include CRC checksum, channel decoding, rate matching de-matching, decoding, demodulation, layer mapping de-matching, and channel detection, while lower-level functions may include resource de-mapping, physical antenna de-mapping, and radio frequency reception.

[0075] For example, the functionality of a CU can be implemented by a single entity or by different entities. For instance, the functionality of the CU can be further divided, separating the control plane and user plane and implementing them through different entities: a control plane CU entity (i.e., the CU-CP entity) and a user plane CU entity (i.e., the CU-UP entity). The CU-CP and CU-UP entities can be coupled with a DU to jointly complete the functions of the access network device.

[0076] Optionally, any one of DU, CU, CU-CP, CU-UP, and RU can be a software module, a hardware structure, or a combination of software and hardware structures, without limitation. The different entities can exist in different forms, without limitation. For example, DU, CU, CU-CP, and CU-UP are software modules, and RU is a hardware structure. These modules and the methods they execute are also within the protection scope of the embodiments of this application.

[0077] In one possible implementation, the access network device includes CU-CP, CU-UP, DU, and RU. For example, the execution entity in the embodiments of this application may include DU, or DU and RU, or CU-CP, DU and RU, or CU-UP, DU and RU, without limitation. The methods executed by each module are also within the protection scope of the embodiments of this application.

[0078] A terminal can also be referred to as a terminal device, user equipment (UE), mobile station, mobile terminal, etc. Terminals can be widely used in communication scenarios across various environments, including but not limited to one or more of the following: device-to-device (D2D), vehicle-to-everything (V2X), machine-type communication (MTC), Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, or smart cities. Terminals can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, or smart home devices, etc. This application does not limit the specific technologies or device forms used in the terminal.

[0079] In this application, the device for implementing the terminal's functions can be a terminal itself; it can also be a device capable of supporting the terminal in implementing those functions, such as a chip system, hardware circuit, software module, or a hardware circuit plus a software module. This device can be installed in the terminal or used in conjunction with the terminal. For ease of description, the following description uses the example of a terminal as the device for implementing the terminal's functions to illustrate the technical solution provided in this application.

[0080] Access network equipment and terminals can be fixed-location or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; on water; or in the air on aircraft, balloons, and satellites. This application does not limit the application scenarios for access network equipment and terminals. Access network equipment and terminals can be deployed in the same or different scenarios; for example, both can be deployed on land; or the access network equipment can be deployed on land and the terminal on water, etc., and so on.

[0081] The roles of access network devices and terminals can be relative, for example, Figure 1 The helicopter or drone 120i can be configured as a mobile access network device. For terminals 120j accessing the wireless access network 100 via 120i, terminal 120i is an access network device; however, for access network device 110a, 120i is a terminal, meaning that 110a and 120i communicate via a wireless air interface protocol. Alternatively, 110a and 120i can communicate via an interface protocol between access network devices; in this case, 120i is also an access network device relative to 110a. Therefore, both access network devices and terminals can be collectively referred to as communication devices. Figure 1 110a and 110b in this context can be referred to as communication devices with access network equipment functions. Figure 1 The 120a-120j in the text can be referred to as communication devices with terminal functions.

[0082] Communication between access network devices and terminals, between access network devices, and between terminals can be conducted using licensed spectrum, unlicensed spectrum, or both simultaneously. Communication can be conducted using spectrum below 6 GHz, spectrum above 6 GHz, or both simultaneously. This application does not limit the spectrum resources used for wireless communication.

[0083] In this application, the foregoing Figure 1In the communication system shown, introducing independent network elements, such as AI network elements or AI nodes, can enable AI-related operations. AI network elements can be directly connected to access network equipment in the communication system, or indirectly connected through third-party network elements. These third-party network elements can be core network elements such as authentication management functions (AMF) or user plane functions (UPF). Alternatively, AI functions, AI modules, or AI entities can be built into other network elements in the communication system to implement AI-related operations. These other network elements can be access network equipment, core network equipment, or network management systems, etc. In this case, the network element performing AI-related operations can be a network element with built-in AI functions. Operation administration and maintenance (OAM) is used for the operation, management, and maintenance of access network equipment and / or core network equipment.

[0084] like Figure 2 or Figure 3 As shown, at least one device among core network equipment, access network equipment, terminals, or OAM (Operational Access Management) can be deployed with an AI model to implement corresponding functions. In this application, the AI ​​models deployed in different nodes can be the same or different. Differences in models include at least one of the following: different structural parameters of the model, for example, different number of layers and / or different weights; different input parameters of the model; or different output parameters of the model. Among these, differences in the model's input parameters and / or output parameters can be described as differences in the model's functionality. Figure 2 The difference is that, in Figure 3 In this architecture, the functions of access network devices are divided into CUs and DUs. Optionally, CUs and DUs can be CUs and DUs under an O-RAN architecture. One or more AI models can be deployed in a CU. And / or, one or more AI models can be deployed in a DU. Optionally, further, [the following can be added] Figure 3 The CU is split into CU-CP and CU-UP. Optionally, one or more AI models can be deployed in CU-CP. And / or, one or more AI models can be deployed in CU-UP. Optionally, Figure 2 or Figure 3 In this context, the OAM of access network equipment and the OAM of core network equipment can be deployed independently.

[0085] In this application, the access network equipment may adopt an O-RAN architecture. The following illustrative description of the O-RAN architecture is not intended to limit this application.

[0086] In the first design, refer to Figure 4The access network equipment includes near real-time access network intelligent controllers (RICs), CUs, DUs, and RUs. The near real-time RIC is used for model training and inference. For example, the near real-time RIC can train an AI model and use the AI ​​model for inference. For instance, the near real-time RIC can obtain network-side or terminal-side information from one or more of the CUs, DUs, RUs, or terminals, which can be used as training data or inference data. For example, the aforementioned information can be used as training data, and the near real-time RIC can use the collected training data to train an AI model. Alternatively, the aforementioned information can be used as inference data, and the near real-time RIC can perform model inference based on the collected inference data and the AI ​​model to determine the inference result. Optionally, the near real-time RIC can send the inference result to one or more of the CUs, DUs, RUs, or terminals. Optionally, the CU and DU can exchange inference results. For example, the near real-time RIC sends the inference result to the CU, and the CU forwards the inference result to the DU. Optionally, the DU and RU can exchange inference results. For example, the near real-time RIC sends the inference result to the DU, or the near real-time RIC sends the inference result to the CU, the CU forwards the inference result to the DU, and the DU forwards the inference result to the RU.

[0087] In the first design, the access network equipment includes a near real-time RIC. There is no restriction on whether non-real-time RICs are included outside the access network equipment. For example, non-real-time RICs may or may not be included outside the access network equipment.

[0088] In the second design, refer to Figure 4 In addition to access network equipment, non-real-time RICs are included. For example, a non-real-time RIC can be located in the OAM (Operational Access Management) or core network equipment, without limitation. A non-real-time RIC can train an AI model and use the AI ​​model for inference. Optionally, a non-real-time RIC can collect network-side or terminal-side information from one or more of the CU (Cybernetic Unit), DU (Distributed Utility Unit), RU (Remote Root Unit), or terminals, which can be used as training data or inference data. For example, using this information as training data, the non-real-time RIC can train an AI model. Alternatively, using this information as inference data, the non-real-time RIC can use the inference data and the AI ​​model to determine the inference result. Optionally, a non-real-time RIC can send the inference result to one or more of the CU, DU, RU, or terminals. Optionally, CUs and DUs can exchange inference results. DUs and RUs can exchange inference results.

[0089] In the second design, non-real-time RICs are included in addition to the access network equipment. There is no restriction on whether near real-time RICs are included in the access network equipment. For example, the access network equipment may or may not include near real-time RICs.

[0090] In the third design, refer to Figure 4 The access network equipment includes near real-time RICs, while non-real-time RICs are included outside the access network equipment. Similar to the first design described above, the near real-time RIC can perform model training and inference. And / or, similar to the second design described above, the non-real-time RIC can perform model training and inference. And / or, the non-real-time RIC can perform model training, and the near real-time RIC can perform model inference. For example, the non-real-time RIC can send the trained AI model to the near real-time RIC, which then uses the AI ​​model for model inference. Optionally, the non-real-time RIC and / or the near real-time RIC can collect network-side or terminal-side information from one or more of the CU, DU, RU, or terminal, which can be used as training data or inference data. For example, using this information as training data, the non-real-time RIC uses the training data to train the AI ​​model. Using this information as inference data, the near real-time RIC uses the AI ​​model and inference data to determine the inference result. Optionally, the near real-time RIC can send the inference result to one or more of the CU, DU, RU, or terminal. Optionally, the CU and DU can exchange inference results. DU and RU can interactively infer results.

[0091] Figure 5 This application provides an alternative O-RAN architecture. (Relative) Figure 4 , Figure 5 The CU was separated into CU-CP and CU-UP, etc.

[0092] In the embodiments of this application, the second communication device can perform AI-related operations on the received channel state information (CSI). The following description of AI technology is not intended to limit the scope of this application.

[0093] An AI model is a concrete implementation of AI functions, representing the mapping relationship between the model's input and output. AI models can be neural networks, linear regression models, decision tree models, support vector machines (SVMs), Bayesian networks, Q-learning models, or other machine learning models. In this application, AI functions may include one or more of the following: data collection (collecting training data and / or inference data), data preprocessing, model training (model learning), model information dissemination (configuring model information), model validation, model inference, or inference result dissemination, etc. In this application, the AI ​​model may be simply referred to as a model.

[0094] like Figure 6The diagram illustrates an application architecture for an AI model. The data source stores training and inference data. The model training host analyzes or trains the AI ​​model using the training data provided by the data source and deploys the AI ​​model on the model inference host. Optionally, the model training host can also update the AI ​​model already deployed on the model inference host. The model inference host can also feed back relevant information about the deployed model to the model training host, enabling the model training host to optimize or update the deployed AI model.

[0095] In this process, the AI ​​model is learned through model training nodes, which essentially learns the mapping relationship between the model's input and output using training data. The model inference nodes use the AI ​​model to perform inference based on the inference data provided by the data source, obtaining the inference result. This method can also be described as follows: the model inference nodes input inference data into the AI ​​model, and the AI ​​model outputs the inference result. This inference result can indicate the configuration parameters used (executed) by the execution object, and / or the operations performed by the execution object. The inference result can be uniformly planned by the actor entity and sent to one or more execution objects (e.g., network entities) for execution. Optionally, the execution entity or execution object can feed back the measurement results of its collected parameters or measurements to the data source; this process can be called performance feedback, and the fed-back parameters can serve as training data or inference data. Optionally, feedback information related to model performance can be determined based on the inference results output by the model inference node, and this feedback information can be fed back to the model inference node. The model inference node can then feed back the model's performance information to the model training node based on this feedback information, so that the model training node can optimize or update the deployed AI model. This process can be called model feedback.

[0096] AI models can be neural networks or other machine learning models. Taking neural networks as an example, they are a specific implementation of machine learning techniques. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings. Therefore, neural networks can accurately model complex, high-dimensional problems abstractly.

[0097] The idea behind neural networks originates from the neuronal structure of the brain. Each neuron performs a weighted summation of its input values, and then passes the weighted sum through an activation function to produce the output. For example... Figure 7 The diagram shown is a schematic of a neuron structure. Assume the neuron's input is x = [x0, x1, ..., x...]. nThe weights corresponding to each input are w = [w, w1, ..., w]. n The bias of the weighted summation is b. The activation function can take many forms. Suppose the activation function of a neuron is: y = f(z) = max(0,z), then the output of this neuron is: For example, if the activation function of a neuron is y = f(z) = z, the output of that neuron is: x i w i , and b can take various possible values, such as decimals, integers (including 0, positive integers, or negative integers), or complex numbers. The activation functions of different neurons in a neural network can be the same or different.

[0098] Neural networks typically consist of multiple layers, each containing one or more neurons. Increasing the depth and / or width of a neural network enhances its expressive power, providing more robust information extraction and abstract modeling capabilities for complex systems. The depth of a neural network refers to the number of layers it comprises, while the number of neurons in each layer is referred to as the width of that layer. Figure 8 The diagram illustrates the layer relationships of a neural network. In one implementation, the neural network includes an input layer and an output layer. The input layer processes the received input through neurons and then passes the result to the output layer, which obtains the output of the neural network. In another implementation, the neural network includes an input layer, hidden layers, and an output layer. The input layer processes the received input through neurons and then passes the result to the hidden layer. The hidden layer then passes the result to the output layer or an adjacent hidden layer, and finally, the output layer obtains the output of the neural network. A neural network can include one or more layers of sequentially connected hidden layers, without limitation. During the training process of the neural network, a loss function can be defined. The loss function describes the difference between the output value of the neural network and the ideal target value. This application does not limit the specific form of the loss function. The training process of a neural network involves adjusting the neural network parameters, such as the number of layers, width, weights of neurons, and / or parameters in the activation functions of neurons, to make the value of the loss function less than a threshold value or meet the target requirements.

[0099] In communication systems such as Long Term Evolution (LTE) and New Radio (NR), access network equipment needs to acquire downlink channel state information (CSI) to determine the downlink data channel resources, modulation and coding scheme (MCS), and precoding configurations for scheduling terminals. In Time Division Duplex (TDD) systems, due to the reciprocity of uplink and downlink channels, access network equipment can measure the uplink reference signal to acquire the uplink CSI and then infer the downlink CSI. For example, the uplink CSI can be used as the downlink CSI. In Frequency Division Duplex (FDD) systems, uplink and downlink reciprocity cannot be guaranteed, and terminals measure downlink reference signals to acquire the downlink CSI. For example, terminals measure downlink reference signals such as the channel state information reference signal (CSI-RS) or the synchronizing signal / physical broadcast channel block (SSB) to acquire the downlink CSI. In one implementation, the terminal generates a CSI report according to a predefined protocol or access network device configuration. This CSI report carries the downlink CSI measured by the terminal and is reported to the access network device so that the access network device can obtain the downlink CSI.

[0100] In the NR protocol, the downlink CSI feedback process includes: the access network device sending a CSI-RS to the terminal. The terminal performs channel and interference measurements based on the CSI-RS to obtain measurement results; based on the measurement results, it determines feedback quantities such as the rank indicator (RI), channel quality indicator (CQI), and precoding matrix indicator (PMI). The terminal reports the downlink CSI to the access network device, which includes information such as the RI, CQI, and PMI measured by the terminal. Specifically, the RI indicates the number of downlink transmission layers suggested by the terminal, the CQI indicates the MCS supported by the current downlink channel conditions suggested by the terminal, and the PMI indicates the precoding scheme suggested by the terminal.

[0101] This application involves precoding, and the precoding technology is described below.

[0102] Precoding is an important way to improve communication speed. Precoding involves the transmitter using multiple antennas (or antenna ports) to assign weights to the transmitted data from each antenna, achieving beamforming and making the transmitted signal more closely match the channel conditions. For example, in downlink precoding, if the downlink signal transmitted by the access network device is X, the downlink channel is H, the downlink precoding matrix is ​​W, and the noise is N0, then the downlink signal transmitted by the access network device is: S = WX; the downlink signal received by the terminal is: Y = HWX + N0.

[0103] The layer number fed back by the PMI corresponds to the RI. For example, if the RI indicates the number of downlink transmission layers n, then the PMI indicates the PMI for layer n. For each layer PMI, the access network device can use the corresponding PMI to precode the downlink signal, achieving beamforming to transmit the downlink signal to the terminal. In other words, by using each layer PMI for precoding, the access network device can form a corresponding beam, and by using the corresponding beam to transmit the downlink signal to the terminal, beamforming can be achieved, improving the transmission quality of the downlink signal.

[0104] The aforementioned CSI feedback mechanism is designed for downlink precoding, and the layer number of the PMI fed back by the terminal is consistent with the layer number indicated by the RI. For example, if the RI indicates n layers for downlink transmission, the terminal needs to feed back the PMI corresponding to each of the n layers to the access network device. In other words, the terminal needs to feed back the PMI of layer n to the access network device. After the introduction of AI, the PMI may be used not only for downlink precoding but also for data collection in AI. In one implementation, the access network device may not need the PMI of layer n. Therefore, in the aforementioned CSI feedback mechanism, where the terminal decides to feed back the PMI of the strongest layer n in the feedback channel, it may not be flexible enough to meet the data collection needs related to AI on the access network device side.

[0105] This application provides a CSI (Content Sense Interface) sending and receiving method, including: a first communication device receiving first information from a second communication device, the first information including indication information for instructing the first communication device to send CSI at a target layer; the first communication device sending a first CSI report to the second communication device, the first CSI report including the CSI at the target layer; for example, the first communication device is a terminal, or a chip or circuit applied in the terminal; the second communication device is an access network device, or a chip or circuit applied in the access network device. In this method, the access network device sends indication information on which layers of CSI to send to the terminal, and the terminal sends the corresponding layer of CSI to the access network device according to the indication. The terminal sends the corresponding layer of CSI to the access network device according to the indication, thereby meeting the data collection needs related to AI on the access network device side.

[0106] like Figure 9As shown, taking the first communication device as the terminal and the second communication device as the access network equipment as an example, a process is provided, including:

[0107] Step 900: The terminal receives first information from the access network device, the first information including indication information for instructing the terminal to send CSI of the target layer.

[0108] For example, the first information can be configuration information used to configure the terminal to send CSI reports at the target layer. This first information can be carried in an RRC message, MAC CE, or downlink control information (DCI), among others, without limitation. In one implementation, the access network device can send an RRC message to the terminal, which is used to configure the CSI report; configuring the CSI report can be referred to as CSI report configuration. The terminal sends a CSI report to the access network device according to the CSI report configuration. In this embodiment, the first information can be carried in the RRC message configuring the CSI report.

[0109] In another implementation, the access network device configures multiple CSI reports for the terminal, each configured CSI report being referred to as a CSI report configuration. Then, the access network device sends a MAC CE or DCI to the terminal to activate one of the CSI report configurations. The terminal then sends a CSI report to the access network device according to the activated CSI report configuration. The first information can be carried in the MAC CE or DCI activating the CSI report. In yet another implementation, the access network device can send a DCI scheduling the CSI report to the terminal. Upon receiving the scheduling of this DCI, the terminal sends a CSI report to the access network device. Optionally, the CSI report can be transmitted over a data channel, such as over a physical uplink shared channel (PUSCH), without limitation. The first information can be carried in the DCI scheduling the CSI report, etc. Or,

[0110] The first information can be an independent piece of information. It can indicate the index of its associated CSI report configuration, or the index of the first information can be indicated in the CSI report configuration, thus establishing a relationship between the first information and the CSI report. In other words, it indicates the specific CSI report to which the first information applies. For example, the terminal sends a CSI report to the access network device according to the CSI report configuration. This CSI report configuration can configure the time-frequency resources and methods for the terminal to send the CSI report, without limitation. The access network device can instruct the terminal to send the target layer CSI using the first information in step 900. In this embodiment, by establishing the correspondence between the CSI configuration report and the first information, the terminal can be instructed to send the target layer CSI in which one or more CSI reports. Alternatively, the correspondence between the first information and the CSI report can be predefined, protocol-specified, or defaulted. For example, there are multiple types of CSI reports; the target layer CSI can be sent in one or more types of CSI reports predefined. Or, when the first information takes effect, it can apply to predefined, protocol-specified, or default CSI reports.

[0111] As previously explained, the first information includes indication information for instructing the terminal to send CSIs of the target layers. The target layer can be any layer, such as one or more layers, which can be consecutive or discontinuous, without limitation. The first information can instruct the terminal to send CSIs of the target layers using a bitmap. For example, the access network device can use N bits to instruct the terminal which target layers' CSIs to send. Each bit represents a layer; 1 indicates the terminal needs to send the corresponding layer's CSI, 0 indicates the terminal does not need to send the corresponding layer's CSI, and vice versa. For example, if the first information sent by the access network device to the terminal includes the indication information 1100, it indicates that the terminal needs to send CSIs of the first and second layers. The first and second layers can be considered the target layers. For the third and fourth layers, the terminal does not need to send the corresponding CSIs. Alternatively, the first information can instruct the terminal to send CSIs of the target layers using a specific indication method. For example, using N bits to instruct the terminal to send 2... NThe first information indicates which layers' CSIs are being sent. For example, 00 indicates sending the first-layer CSI, 01 indicates sending the second-layer CSI, 10 indicates sending the third-layer CSI, and 11 indicates sending the fourth-layer CSI, etc. Alternatively, the first information can specify which layers' CSIs the terminal will not send. The remaining layers without a specific indication are considered the layers that require CSIs, i.e., the unindicated remaining layers are considered the target layers. For example, a 2-bit instruction can be used to indicate which layers' CSIs the terminal will not send. For instance, 00 indicates not sending the first-layer CSI, 01 indicates not sending the second-layer CSI, 10 indicates not sending the third-layer CSI, and 11 indicates not sending the fourth-layer CSI. If the first information sent by the access network device to the terminal includes 00, the terminal determines not to send the first-layer CSI and sends the second to fourth-layer CSIs; the second to fourth layers are considered the target layers. Alternatively, multiple layer combinations can be predefined, and the first information can specify which layer combination's CSI the terminal will send. For example, four layer combinations are predefined: Layer 1, Layer 2 and Layer 3, Layer 1 and Layer 3, Layer 3 and Layer 4, etc. Two bits can be used to indicate which layer combination's CSI the terminal should send. The layer combination indicated by the access network device that requires CSI transmission can be considered the target layer. Alternatively, the access network device can indicate layer combinations that do not require CSI transmission; for layer combinations not indicated by the access network device, the terminal considers them to be the target layers requiring CSI transmission.

[0112] Step 901: The terminal sends a first CSI report to the access network device. The first CSI report includes the CSI of the target layer.

[0113] As explained above, a correlation can be established between the first information and the CSI report. For a CSI report with an established correlation, there are no restrictions on whether the target layer's CSI is sent when the CSI report is submitted. In this embodiment, it can be considered as establishing a correlation between the first information and the first CSI report.

[0114] In this embodiment, the CSI sent by the terminal to the access network device can be a true CSI, which can be a feature vector matrix, a channel response, or compressed and / or quantized information of the channel response, or compressed and / or quantized information of the feature vector matrix, etc., without limitation. That is, in this embodiment, the target layer CSI sent by the terminal to the access network device includes one or more of the following indication information: the target layer channel response, the target layer feature vector matrix, compressed and / or quantized information of the target layer feature vector matrix, or compressed and / or quantized information of the target layer channel response, etc.

[0115] In one implementation, the access network device sends a downlink reference signal to the terminal. The terminal measures the downlink reference signal to determine the downlink channel response H. Eigenvalue decomposition or singular value decomposition is performed on the downlink channel response H or its covariance matrix to obtain the eigenvector matrix V of the downlink channel. Each row or column of this eigenvector matrix V can be called an eigenvector. For example, the eigenvectors can be sorted according to the magnitude of the eigenvalues ​​or singular values, such as from largest to smallest or smallest to largest, to obtain the eigenvectors of the first layer, the second layer, the third layer, etc. Optionally, considering the frequency domain dimension, adding the frequency domain dimension to each eigenvector yields the eigenvector matrix of the first layer, the eigenvector matrix of the second layer, the eigenvector matrix of the third layer, etc. Alternatively, without considering the frequency domain dimension, the eigenvectors of the first layer, the eigenvectors of the second layer, and the eigenvectors of the third layer can also be called the eigenvector matrix of the first layer, the eigenvector matrix of the second layer, and the eigenvector matrix of the third layer, respectively. Optionally, the correspondence between the access network device and the terminal regarding the feature vector matrix and the layer can be agreed upon in advance in some way, so that their understanding of the correspondence between the feature vector matrix and the layer is consistent. For example, they can agree to sort the feature vector matrix from largest to smallest according to the size of the eigenvalues ​​or singular values, so as to obtain the feature vector matrix of the first layer, the feature vector matrix of the second layer, and the feature vector matrix of the third layer, etc.

[0116] In another implementation, such as Figure 10As shown, after obtaining the target layer's feature vector matrix V, the terminal can input V into a CSI generator, whose output is a compressed target layer feature vector matrix Z. The terminal then sends the compressed target layer feature vector matrix Z to the access network device. The access network device inputs the received compressed target layer feature vector matrix Z into a CSI reconstructor, whose output is the recovered target layer feature vector matrix V'. Through this implementation, the terminal does not directly send the target layer feature vector matrix to the access network device, but instead sends a compressed target layer feature vector matrix, reducing CSI transmission overhead. It is understood that both the CSI generator and the CSI reconstructor are pre-trained models. The CSI generator's function includes compressing the target layer feature vector matrix, and the CSI reconstructor's function includes recovering the compressed target layer feature vector matrix. Alternatively, the terminal quantizes the target layer feature vector matrix and sends the quantized target layer feature vector matrix to the access network device. The access network device then recovers the target layer feature vector matrix from the quantized target layer feature vector matrix. Quantization is the process of transforming the target layer's feature vector matrix from continuous values ​​to discrete values. Alternatively, the terminal can compress and quantize the target layer's feature vector matrix; for example, it can compress first and then quantize, or quantize first and then compress, or compress and quantize simultaneously, without restriction. The terminal sends the compressed and quantized target layer feature vector matrix to the access network device; the access network device then restores the original target layer feature vector matrix.

[0117] In another implementation, the access network device measures the downlink reference signal to determine the downlink channel response H; it then performs eigenvalue decomposition or singular value decomposition on the downlink channel response H or its covariance matrix to obtain the channel response for each layer. Taking singular value decomposition (SVD) as an example, SVD(H) = U, S, V, i.e., H = U * S * V; where V is the eigenvector matrix, and the channel response H of the i-th layer is: U(:,i) * S(i) * V(i,:); where U(:,i) represents the i-th column of U, S(i) represents the i-th eigenvalue, and V(i,:) represents the i-th row of V. The terminal can send the target layer's channel response to the access network device. Alternatively, the terminal can compress and / or quantize the target layer's channel response and send the compressed and / or quantized target layer's channel response to the access network device. The access network device then recovers the compressed and / or quantized target layer's channel response.

[0118] Optionally, the CSI of the target layer sent by the terminal to the access network device may also include one or more of the following indication information: eigenvalues, singular values, or the target layer itself, of the characteristic vector matrix of the channel response of the target layer.

[0119] In one implementation, when the first information includes indication information indicating the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer's channel response, the terminal, in addition to sending the indicated target layer's feature vector matrix to the access network device, also sends the corresponding eigenvalues ​​or singular values ​​of the target layer's feature vector matrix to the access network device. Taking eigenvalues ​​as an example, the eigenvalues ​​corresponding to each layer's feature vectors in the target layer can include the eigenvalues ​​corresponding to each feature vector in that layer's feature vector matrix. For example, if a layer's feature vector matrix in the target layer includes three feature vectors, the terminal can send the eigenvalues ​​corresponding to each of those three feature vectors to the access network device. Alternatively, one eigenvalue can be sent for each layer's feature vectors in the target layer. For example, the terminal can calculate the average or weighted average of the eigenvalues ​​of all feature vectors included in the layer's feature vector matrix, and send the average calculation result of the eigenvalues ​​to the access network device. In another implementation, the layer of eigenvalues ​​or singular values ​​sent by the terminal can be different from the layer of CSI transmission mentioned above. That is, the eigenvalues ​​or singular values ​​sent by the terminal may not be the eigenvalues ​​or singular values ​​of the target layer's feature vector matrix. For example, the eigenvalues ​​or singular values ​​of the eigenvector matrices of which layers the terminal sends to the access network device can be predefined, protocol-specified, or default. For instance, it could be predefined that the terminal sends the eigenvalues ​​or singular values ​​of the eigenvector matrices of all layers to the access network device.

[0120] For example, an access network device can send indication information to a terminal device to instruct the terminal to send characteristic values ​​or singular values, etc. This indication information can be configuration information, described as the access network device configuring the terminal to send characteristic values ​​or singular values, etc. In this embodiment, the indication information for instructing the terminal to send the target layer's CSI and the indication information for instructing the terminal to send characteristic values ​​or singular values ​​can be carried in the same information. For example, the first information in step 900 may include, in addition to the indication information for instructing the terminal to send the target layer's CSI, also the indication information for instructing the terminal to send the characteristic values ​​or singular values ​​of the target layer's feature vector matrix. Alternatively, the indication information for instructing the terminal to send the target layer's CSI and the indication information for instructing the terminal to send characteristic values ​​or singular values ​​can be independent. Further, the indication information for characteristic values ​​or singular values ​​can carry the target layer's indication information, etc., to instruct the terminal to send the target layer's characteristic values ​​or singular values ​​to the access network device. Alternatively, the layer to which the terminal sends the CSI and the layer to which the terminal sends the characteristic values ​​or singular values ​​can be different. For example, the terminal sends the characteristic values ​​or singular values ​​of the first layer to the access network device. The first layer is not exactly the same as the target layer. This first layer can be indicated by the access device, or it can be preset, specified by the protocol, or default, etc., without restriction. For example, the terminal sends characteristic values ​​or singular values ​​of all layers to the access network device according to a preset.

[0121] In one implementation, the access network device can also indicate a threshold to the terminal, which instructs the terminal to send a feature vector matrix whose eigenvalues ​​or singular values ​​are greater than or equal to the threshold. This threshold can be included in the first information in step 900 without limitation. Taking eigenvalues ​​as an example, "eigenvalue greater than or equal to the threshold" could mean that the eigenvalues ​​corresponding to all eigenvectors in a certain layer's feature vector matrix are greater than or equal to the threshold; or, that N eigenvectors in a certain layer's feature vector matrix have eigenvalues ​​greater than or equal to the threshold; or, that a eigenvalue greater than or equal to the threshold can be calculated from the eigenvalues ​​corresponding to all eigenvector matrices in a certain layer's feature vector matrix. As previously explained, the access network device can instruct the terminal device to send the CSI of the target layer, which includes the feature vector matrix of the target layer. In one implementation, the target layer indicated by the access network device may differ from the layer calculated based on the aforementioned threshold. In this case, the two can be further intersected, and the terminal can send the feature vector matrix of the intersecting layer to the access network device. For example, the access network device instructs the terminal to send feature vector matrices for layers 1, 3, and 5, which can be considered target layers. Based on the threshold values ​​indicated by the access network device, layers 2, 3, and 6 are determined to meet the conditions. The terminal can take the intersection of these two sets and send the feature vector matrix for layer 3 to the access network device. In another implementation, the terminal can send the feature vector matrices for the target layers and the feature vector matrices that meet the threshold conditions to the access network device, without restriction. Continuing with the above example, the terminal can send the feature vector matrices for the indicated target layers (1, 3, 5) and the feature vector matrices for layers that meet the threshold conditions (2, 3, 6) to the access network device.

[0122] In one implementation, the access network device can also instruct the terminal on the transmission period or transmission pattern of the target layer CSI. This instruction information can be carried in the first information in step 900, without limitation. The terminal transmits the target layer CSI to the access network device according to the transmission period indicated by the access network device. Further, the target layer includes at least one layer, and the transmission period for each layer's CSI can be the same or different. If the transmission period for each layer's CSI is the same, a single transmission period can be configured for all layers of the target layer; alternatively, the transmission period for each layer's CSI in the target layer can be configured independently, and the transmission periods for each layer's CSI in the independently configured target layer can be the same or different, without limitation. The transmission pattern is used to indicate which layers' CSI the terminal needs to transmit each time. For example, the transmission pattern can be {layer 1, layer 2 and layer 3, layer 1 and layer 3, layer 3 and layer 4}. Then, in the first transmission period, the terminal can transmit the layer 1 CSI to the access network device. In the second transmission period, the terminal transmits the layer 2 and layer 3 CSIs to the access network device. Similarly, in the third and fourth transmission cycles, the terminal sends the CSIs of Layer 1 and Layer 3, and the CSIs of Layer 3 and Layer 4, respectively, to the access network device. For example, the transmission pattern can be determined based on the target layer, and in each transmission cycle, the terminal sends the CSIs of a portion of the target layer to the access network device according to the transmission pattern. For example, the target layer includes Layers 1 to 4. Then, as described above, in each transmission cycle, the terminal sends the CSIs of a portion of the target layer. For example, in the third transmission cycle, the terminal sends the CSIs of Layer 1 and Layer 3, etc., to the access network device.

[0123] In one implementation, in addition to the indication information for instructing the terminal to send the CSI of the target layer, the first information in step 900 above also includes indication information for instructing the terminal to send one or more of the following: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, the threshold of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, the transmission period of the CSI of the target layer, or the transmission pattern of the CSI of the target layer, etc.

[0124] In this embodiment, the CSI sent by the terminal to the access network device is a channel response, compressed information of the channel response, an eigenvector matrix, or compressed information corresponding to the eigenvector matrix. This differs from the traditional CSI sent by the terminal to the access network device, which includes RI, CQI, and PMI. For ease of distinction, the CSI sent by the terminal in this embodiment can be referred to as ground truth CSI, while the traditional CSI sent by the terminal can be referred to as traditional CSI.

[0125] In one implementation, truth-based CSI and conventional CSI can be sent independently. As previously explained, the terminal can send a first CSI report to the access network device, which includes the target layer CSI, and the target layer CSI can be truth-based CSI. Furthermore, the terminal can also send a second CSI report to the access network device, which includes conventional CSI, i.e., the second CSI report includes one or more of the following indication information: RI, CQI, or PMI, where the number of layers in the PMI is determined based on the RI. Optionally, a new CSI report type (report quantity) can be defined for sending truth-based CSI, such as cir-tCSI. In this new type of CSI report, which layers' CSI are sent is indicated by the access network device and is not affected by the RI determined by the terminal.

[0126] In another implementation, truth-based CSI and traditional CSI can be sent together, meaning they can be sent in a single CSI report. The first CSI report mentioned above, in addition to including the target layer's CSI (truth-based CSI), may also include indication information of at least one of the following: RI, CQI, or PMI, etc., where the number of PMI layers is determined based on the RI. Optionally, a new CSI report type can be defined for sending truth-based CSI and traditional CSI, such as cri-RI-PMI-CQI-tCSI, or cri-RI-CQI-tCSI, or cri-RI-LI-PMI-CQI-tCSI, or cri-RSRP-tCSI, or cri-SINR-tCSI, etc.

[0127] exist Figure 9 In this process, the terminal sends a target layer CSI to the access network device. This target layer can be any layer, not limited to the highest quality layer (n) in traditional CSI transmission. Upon receiving the target layer CSI, the access network device can use it as AI-related data. For example, the target layer CSI can be used for one or more of the following: model training, model inference, model testing, model validation, or model monitoring.

[0128] In one implementation, the access network device can use the target layer's CSI as training data for model training. Model training is a crucial part of machine learning; the essence of machine learning is learning certain features from a training set (composed of training data) to minimize the difference between the model's output and the ideal target value under training on that set. Typically, even with the same network structure, models trained on different training sets may have different weights and / or outputs. Therefore, the composition and selection of the training set, to some extent, determine the model's performance. And / or, the access network device can use the target layer's CSI as inference data for model inference. For example, the target layer's CSI is input into the model as inference data, and the model outputs the inference result. And / or, the access network device can use the target layer's CSI as test data for model testing. For example, after model training, a test set (composed of test data) can be used to test the trained model. For example, to evaluate the model's generalization ability, determine if the model's performance meets requirements, or decide whether the model is usable. Alternatively, the access network device can use the target layer's CSI as validation data to validate the model. Model validation typically occurs during model training. For example, after each one or more iterations (EPOCH) of training, a validation set (composed of validation data) can be used to validate the current model, monitoring its training status, such as whether it is underfitting, overfitting, or has converged, and determining whether to terminate training. Optionally, during model validation, the model's hyperparameters can be adjusted. These hyperparameters can refer to at least one of the following parameters: the number of layers in the neural network, the number of neurons, the activation function, or the loss function, etc. And / or, the access network device can use the target layer's CSI as monitoring data to monitor the model. For example, during model inference, the model's performance can be monitored. If the accuracy of the inference result is less than a threshold, information can be fed back to the model training node, allowing the training node to optimize or update the deployed model; this process can also be called model feedback.

[0129] exist Figure 9In this process, the access network device may only need to perform AI-related operations on the target layer. For example, if a model performs poorly at the target layer, the access network device can collect the target layer's CSI as training data for model training. Alternatively, if the access network device only wants to monitor the model's performance at the target layer, it can collect the target layer's CSI as monitoring data for model monitoring. The access network device can send first information to the terminal, instructing the terminal to send the target layer's CSI. The terminal then sends the target layer's CSI to the access network device according to the instruction. This allows the access network device to flexibly collect CSI from different layers based on different needs. Furthermore, the terminal sends the corresponding layer's CSI according to the access network device's instruction, ensuring that the CSI sent by the terminal meets the access network device's requirements and reducing the air interface overhead of sending unnecessary layer CSI.

[0130] This application provides another CSI sending and receiving method, which differs from the previous method, as can be seen below. Figure 11 As explained below, the terminal can proactively send the target layer CSI to the access network device. For example, the terminal can send target layer indication information to the access network device; the terminal can send the target layer CSI to the access network device. Alternatively, see below. Figure 12 The description states that the access network device can proactively send target layer CSI to the terminal. For example, the access network device sends target layer indication information to the terminal; the access network device sends target layer CSI to the terminal. The method includes: a first communication device sending first information to a second communication device, the first information including indication information of the first communication device sending target layer CSI; the first communication device sending a first CSI report to the second communication device, the first CSI report including the target layer CSI.

[0131] The first communication device is a terminal, or a chip or circuit used in the terminal. The second communication device is an access network device, or a chip or circuit used in the access network device. Taking the first communication device as the terminal and the second communication device as the access network device as an example, such as... Figure 11 As shown, a process is provided, including:

[0132] Step 1100: The terminal sends first information to the access network device, the first information including indication information for instructing the terminal to send the target layer CSI.

[0133] For example, the first information may be indication information, which is used to instruct the terminal to send CSI to the target layer.

[0134] Step 1101: The terminal sends a first CSI report to the access network device. The first CSI report includes the CSI of the target layer.

[0135] In one implementation, the first information and the first CSI report can be in the same message. For example, the first CSI report carries the first information (or target layer indication information), such as in the first part of the first CSI report. In another implementation, the first information and the first CSI report can be in different messages. The association between the first information and the first CSI report can be established by carrying an index of the first CSI report in the first information, or by carrying an index of the first information in the first CSI report. Alternatively, the association between the first information and the first CSI report can be predefined, protocol-defined, or default, without restriction. For example, there are multiple types of CSI reports, which can be predefined, and the first information indicates the layer sending the CSI in one or more types of CSI reports. Alternatively, during the effective period of the first information, the first information can indicate the layer of all CSI reports, or it can indicate the layer of predefined CSI reports, without restriction.

[0136] The way the first information indicates the target layer is similar to the previous text. Figure 9 The indication method for the target layer in the process is similar. For example, the target layer can be indicated by a bitmap. Alternatively, the target layer can be indicated by specific indication information, such as using 2 bits to indicate 4 layers. If the specific indication information is 00, it indicates that the indicated target layer is layer 1. Alternatively, multiple layer combinations can be pre-defined, and different layer combinations can be indicated by indication information. For example, if 4 layer combinations are pre-defined as layer 1, layer 2 and layer 3, layer 1 and layer 3, layer 3 and layer 4, etc., then 2 bits can be used to indicate each of these 4 layer combinations. For example, if the specific indication information is 10, it indicates that the indicated layer combination is layer 1 and layer 3, that is, the target layers are layer 1 and layer 3.

[0137] In this embodiment, the target layer CSI sent by the terminal to the access network device can be a true CSI. The target layer CSI includes one or more of the following indication information: the target layer's channel response, compression and / or quantization information of the target layer's channel response, the target layer's feature vector matrix, or compression and / or quantization information of the target layer's feature vector matrix. Further, the target layer CSI may also include one or more of the following indication information: eigenvalues ​​of the target layer's feature vector matrix, singular values, or the target layer itself. For example, there may be multiple target layers. When the terminal reports the target layer CSI to the access network device, it can indicate the layer corresponding to each CSI information. Taking the CSI information as a feature vector matrix as an example, the CSI information for each layer includes at least one feature vector matrix. When the terminal sends the target layer CSI to the access network device, it can indicate the layer corresponding to each feature vector matrix. If all feature vector matrices in the CSI information correspond to the same layer, then the CSI information can indicate a corresponding layer, or no indication may be given. Alternatively, if multiple feature vector matrices in the CSI information belong to the same layer, then these multiple feature vector matrices can indicate a corresponding layer.

[0138] and Figure 9 The process is similar. The first CSI report includes the target layer CSI, which is a truth CSI. The truth CSI and traditional CSI can be sent together. Furthermore, the first CSI report can also include traditional CSI. For example, the first CSI report can also include one or more of the following indication information: RI, CQI, or PMI, etc. The number of layers for PMI is determined based on RI. Alternatively, the target layer CSI and traditional CSI can be sent independently. For example, the terminal can also send a second CSI report to the access network equipment. The second CSI report includes one or more of the following indication information: RI, CQI, or PMI. Similarly, in traditional CSI, the number of layers for PMI is determined based on RI.

[0139] In one implementation: the terminal measures the downlink reference signal and obtains the measurement result; based on the measurement result, the terminal determines the CSI for different layers. The terminal uses the above... Figure 11The method in the process involves sending the target layer's CSI to the access network device. It's understood that the target layer can be any layer, not limited to the highest quality layer (n) in traditional CSI transmission. Since the access network device doesn't instruct the terminal to specifically send which layers' CSI, the terminal proactively sends the target layer's CSI information to the access network device. The terminal needs to also inform the access network device of the target layer of the sent CSI; that is, the terminal needs to send the first information from step 1100 above to the access network device, thereby ensuring that the access network device correctly understands the received target layer's CSI. Optionally, when receiving the target layer's CSI, the access network device can use it for one or more of the following: model training, model inference, model testing, model verification, or model monitoring, etc.

[0140] The first communication device is an access network device, or a chip or circuit used in an access network device. The second communication device is a terminal, or a chip or circuit used in a terminal. Taking the first communication device as an access network device and the second communication device as a terminal as an example, such as... Figure 12 As shown, a process is provided, including:

[0141] Step 1200: The access network device sends first information to the terminal, the first information including indication information for instructing the access network device to send the target layer CSI.

[0142] For example, the first information can be indication information, which is used to instruct the access network device to send the target layer's CSI. For the method of indicating the target layer with the first information, see [link to relevant documentation]. Figure 11 The explanation in the document.

[0143] Step 1201: The access network device sends the target layer CSI to the terminal.

[0144] In one implementation, the first information and the target layer's CSI can be in the same message. For example, the first information may be located in the target layer's CSI. Alternatively, the first information and the target layer's CSI can be in different messages. For example, the first information may carry indication information of the associated target layer's CSI, or the target layer's CSI may carry indication information of the associated first information. Alternatively, the association between the first information and the target layer's CSI can be predefined, protocol-defined, or default, etc., without restriction. Alternatively, during the effective period of the first information, the first information may indicate the layer of all CSI information sent by the access network device, or indicate the layer of predefined CSI information, etc.

[0145] For example, step 1202 above can also be described as: the access network device sends a first CSI report to the terminal, the first CSI report carrying the CSI of the target layer.

[0146] For details regarding the CSI of the target layer and the specific content included in the first information, please refer to the preceding text. Figure 11 The explanation in the document. Figure 12 The embodiments are the same as those described above. Figure 11 The main difference in the embodiments is that: Figure 11 In this process, the terminal actively sends the target layer's CSI to the access network device. To ensure the access network device correctly understands the received target layer's CSI, the terminal also needs to send the corresponding target layer's indication information back to the access network device. Figure 12 In the process, the access network device actively sends CSI to the terminal. In order for the terminal to correctly understand the CSI of the target layer it receives, the access network device also needs to send the target layer indication information corresponding to the CSI to the terminal.

[0147] Similar to the above, the CSI for the target layer is the truth CSI. Truth CSI and traditional CSI can be sent together. The first CSI report may also include one or more of the following indication information: RI, CQI, or PMI, etc., where the number of layers for the PMI is determined based on the RI. Alternatively, truth CSI and traditional CSI can be sent separately, and the access network device may also send one or more of the following indication information to the terminal: RI, CQI, or PMI, where the number of layers for the PMI is determined based on the RI.

[0148] Optionally, the terminal may use the received target layer CSI for one or more of the following: model training, model inference, model testing, model validation, or model monitoring. For example, in one implementation, models are deployed on both the access network device side and the terminal side, and these models need to be used in conjunction with each other. For example, as... Figure 10 As shown, a CSI generator is deployed on the terminal side, and a CSI reconstructor is deployed on the access network device side. The access network device can use all or some of the collected CSIs from all layers as training data to train the CSI reconstructor model. After the access network device completes training, it can select the target layer's CSIs from all layers and send them to the terminal. The terminal then uses the target layer's CSIs to train the terminal-side CSI generator model. In another implementation, the following approach is used... Figure 10 For example, if the performance of the model is monitored by the terminal, the access network device can restore the CSI of the target layer (i.e., Figure 10 The recovered V' is sent to the terminal, which compares the recovered V' with the original V to determine the model's performance, etc.

[0149] It is understood that, in the embodiments of this application:

[0150] 1. Within each process, focus on describing the differences between the various processes. The descriptions of different processes can be cross-referenced.

[0151] 2. There are no restrictions on the order of steps in each process. For example, in Figure 9 In the process, the first information and the first CSI report can be sent together, or the first CSI report can be sent first, followed by the first information, etc.

[0152] 3. Regarding "instruction," the following explanation is provided: Instruction may display or implicitly indicate the corresponding information, without limitation. For example, in the description of this application: the first CSI report includes instruction information for the target layer's CSI. Therefore, the instruction for the target layer's CSI can be displayed; for example, the first CSI report includes the target layer's CSI. Alternatively, the instruction for the target layer's CSI can be implicitly indicated. For example, if there is a correspondence between the target layer's CSI and other information, the first CSI report may carry other information that corresponds to the target layer's CSI to indicate the target layer's CSI, etc.

[0153] 4. In the description of this application, unless otherwise specified, CSI refers to downlink CSI. The above description mainly uses the transmission of downlink CSI as an example. It should be understood that this is not intended to limit this application. The solutions in the embodiments of this application can be used not only for the transmission of downlink CSI, but also for other scenarios, such as the transmission of uplink CSI.

[0154] It is understood that, in order to achieve the functions in the above embodiments, terminals and access network devices include hardware structures and / or software modules corresponding to perform each function. Those skilled in the art should readily recognize that, based on the units and method steps described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0155] Figure 13 and Figure 14 The diagram illustrates the possible structures of communication devices provided in the embodiments of this application. These communication devices can be used to implement one or more corresponding functions in the above method embodiments, such as functions implemented by one or more of the terminal or access network equipment, and thus may achieve the beneficial effects of the above method embodiments.

[0156] like Figure 13 As shown, the communication device 1300 includes a processing unit 1310 and a transceiver unit 1320. The communication device 1300 is used to implement one or more functions in the above method embodiments, such as functions implemented by one or more of a terminal or access network equipment.

[0157] In one design, the communication device 1300 is a terminal, or a chip or circuit applied in a terminal, etc., when the communication device 1300 is used to implement... Figure 9 The terminal functions as follows: a transceiver unit 1320 receives first information from a second communication device, the first information including indication information for instructing the first communication device to send Channel State Information (CSI) of the target layer; optionally, a processing unit 1310 generates a first CSI report; the transceiver unit 1320 is also used to send the first CSI report to the second communication device, the first CSI report including the CSI of the target layer. Optionally, the second communication device can be an access network device, or a chip or circuit applied in the access network device, etc.

[0158] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix, or the compression and / or quantization information of the channel response of the target layer.

[0159] In one implementation, the CSI of the target layer further includes one or more of the following indication information: eigenvalues ​​of the feature vector matrix of the target layer, singular values, or the target layer itself.

[0160] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0161] In one implementation, the transceiver unit 1320 is further configured to: send a second CSI report to the second communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0162] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, threshold values ​​of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, transmission period of the CSI of the target layer, or transmission pattern of the CSI of the target layer.

[0163] In one implementation, the CSI of the target layer sent by the first communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0164] In one implementation, the first information is configuration information, which is used to configure the first communication device to send the CSI of the target layer.

[0165] In another design, the communication device 1300 is an access network device, or a chip or circuit used in an access network device. When the communication device is used to implement... Figure 9 The access network device may optionally include: a processing unit 1310 for generating first information; a transceiver unit 1320 for sending the first information to a first communication device, the first information including indication information for instructing the first communication device to send Channel State Information (CSI) of the target layer; and receiving a first CSI report from the first communication device, the first CSI report including the CSI of the target layer. Optionally, the first communication device may be a terminal, or a chip or circuit applied in the terminal, etc.

[0166] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix, or the compression and / or quantization information of the channel response of the target layer.

[0167] In one implementation, the CSI of the target layer further includes one or more of the following indication information: eigenvalues, singular values, or the target layer itself.

[0168] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0169] In one implementation, the transceiver unit 1320 is further configured to receive a second CSI report from the first communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0170] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, threshold values ​​of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, transmission period of the CSI of the target layer, or transmission pattern of the CSI of the target layer.

[0171] In one implementation, the CSI of the target layer received by the second communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0172] In one implementation, the first information is configuration information, which is used to configure the first communication device to send the CSI of the target layer.

[0173] In one design, the communication device 1300 is a terminal, or a chip or circuit applied in a terminal. When the communication device is used to implement... Figure 11 When the communication device is used to implement the functions of a terminal, it is either an access network device or a chip or circuit applied in an access network device. Figure 12 The access network device may optionally include: a processing unit 1310 for generating first information; and a transceiver unit 1320 for sending the first information to a second communication device, the first information including indication information instructing the first communication device to send channel state information (CSI) of the target layer; and sending a first CSI report to the second communication device, the first CSI report including the CSI of the target layer. Optionally, the second communication device may be an access network device, or a chip or circuit used in an access network device. Alternatively, the second communication device may be a terminal, or a chip or circuit used in a terminal, etc.

[0174] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix of the target layer, or the compression and / or quantization information of the channel response of the target layer.

[0175] In one implementation, the CSI of the target layer may further include one or more of the following indication information: eigenvalues, singular values, or the target layer itself.

[0176] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0177] In one implementation, the transceiver unit 1320 is further configured to: send a second CSI report to the second communication device, the second CSI report including one or more of the following indication information: RI, CQI or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0178] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following of the target layer: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, a threshold value of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, the transmission period of the CSI of the target layer, or the transmission pattern of the CSI of the target layer.

[0179] In one implementation, the CSI of the target layer sent by the first communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0180] In one implementation, the first information is indication information, which is used to instruct the first communication device to send the CSI of the target layer.

[0181] In one design, the communication device 1300 is an access network device, or a chip or circuit used in an terminating access network device. When the communication device is used to implement... Figure 11 When the communication device is used to access the network, it may be a terminal, or a chip or circuit applied in the terminal. When the communication device is used to implement... Figure 12 The terminal functions as follows: a transceiver unit 1320 is configured to receive first information from a first communication device, the first information including indication information for instructing the first communication device to transmit channel state information (CSI) of the target layer; and to receive a first CSI report from the first communication device, the first CSI report including the CSI of the target layer. Optionally, a processing unit 1310 is configured to process the first information and the first CSI report, etc.

[0182] In one implementation, the CSI of the target layer includes one or more of the following indication information: the channel response of the target layer, the feature vector matrix of the target layer, the compression and / or quantization information of the feature vector matrix of the target layer, or the compression and / or quantization information of the channel response of the target layer.

[0183] In one implementation, the CSI of the target layer may further include one or more of the following indication information: features of the feature vector matrix of the target layer, singular values, or the target layer itself.

[0184] In one implementation, the first CSI report may also include one or more of the following indication information: rank indicator (RI), channel quality indicator (CQI), or precoding matrix index (PMI), wherein the number of layers of the PMI is determined based on the RI.

[0185] In one implementation, the transceiver unit 1320 is further configured to receive a second CSI report from the first communication device, the second CSI report including one or more of the following indication information: RI, CQI, or PMI, wherein the number of layers of the PMI is determined according to the RI.

[0186] In one implementation, the first information further includes indication information for instructing the first communication device to send one or more of the following of the target layer: eigenvalues ​​and singular values ​​of the feature vector matrix of the target layer, a threshold value of the eigenvalues ​​or singular values ​​of the feature vector matrix of the target layer, the transmission period of the CSI of the target layer, or the transmission pattern of the CSI of the target layer.

[0187] In one implementation, the CSI of the target layer received by the second communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

[0188] In one implementation, the first information is indication information, which is used to instruct the first communication device to send the CSI of the target layer.

[0189] A more detailed description of the processing unit 1310 and the transceiver unit 1320 can be obtained directly from the relevant descriptions in the above method embodiments, and will not be repeated here.

[0190] like Figure 14 As shown, the communication device 1400 includes a processor 1410 and an interface circuit 1420. The processor 1410 and the interface circuit 1420 are coupled to each other. It is understood that the interface circuit 1420 can be a transceiver or an input / output interface. Optionally, the communication device 1400 may further include a memory 1430 for storing instructions executed by the processor 1410, or storing input data required by the processor 1410 to execute instructions, or storing data generated after the processor 1410 executes instructions. Optionally, the processor 1410 can be used to implement one or more of the functions described in the above method embodiments.

[0191] Specifically, the processor 1410 can execute instructions in the memory 1430 to enable the communication device 1400 to perform one or more functions as described in the above method embodiments, such as functions implemented by one or more of a terminal or access network device.

[0192] When the communication device 1400 is used to implement Figure 9 , Figure 11 or Figure 12 In the method shown, the processor 1410 is used to implement the functions of the processing unit 1310, and the interface circuit 1420 is used to implement the functions of the transceiver unit 1320.

[0193] When the aforementioned communication device is a terminal or a chip applied to a terminal, the terminal or terminal chip can implement the functions of the terminal in the above method embodiments. The terminal chip receives information from other modules (such as radio frequency modules or antennas) in the terminal, which is sent to the terminal by the access network device; or, the terminal chip sends information to other modules (such as radio frequency modules or antennas) in the terminal, which is sent to the access network device by the terminal.

[0194] When the aforementioned communication device is an access network device or a module applied to an access network device, the access network device or access network device module can implement the functions of the access network device in the above method embodiments. The access network device module receives information from other modules (such as radio frequency modules or antennas) in the access network device, the information being sent by the terminal to the access network device; or, the access network device module sends information to other modules (such as radio frequency modules or antennas) in the access network device, the information being sent by the access network device to the terminal. Here, the access network device module can be the baseband chip of the access network device, or it can be a DU or other modules. The DU here can be a DU under an open radio access network (O-RAN) architecture.

[0195] It is understood that the aforementioned communication device can be a device in wireless communication or a chip applied to such a device, such as the aforementioned access network equipment or terminal, or it can also be a device in other communication systems such as a WIFI communication system that performs one or more functions of an access network equipment or terminal, etc., and is not limited in this application. For example, in a WIFI communication system, the device performing the function of an access network equipment can be an access node, or a chip or circuit applied to the access node, etc. The device performing the function of a terminal can be a terminal, or a chip or circuit applied to the terminal, etc.

[0196] It is understood that the processor in the embodiments of this application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0197] The method steps in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a base station or terminal. Of course, the processor and storage medium can also exist as discrete components in a data quality measurement device.

[0198] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video optical disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both types of storage media.

[0199] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0200] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates an "or" relationship between the preceding and following related objects; in the formulas of this application, the character " / " indicates a "division" relationship between the preceding and following related objects. "Including at least one of A, B, and C" can mean: including A; including B; including C; including A and B; including A and C; including B and C; including A, B, and C.

[0201] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers described above does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

Claims

1. A CSI sending method, characterized in that, The method is applied to a first communication device, and comprises: receiving first information from a second communication device, wherein the first information comprises indication information indicating that the first communication device transmits channel state information (CSI) of a target layer; transmitting a first CSI report to the second communication device, wherein the first CSI report comprises the CSI of the target layer; the CSI of the target layer comprises indication information of one or more of the following: a channel response of the target layer, an eigenvector matrix of the target layer, compression and / or quantization information of the eigenvector matrix, or compression and / or quantization information of the channel response of the target layer.

2. The method of claim 1, wherein, The CSI of the target layer further comprises indication information of one or more of the following: eigenvalues of the eigenvector matrix of the target layer, singular values, or the target layer.

3. The method of claim 1 or 2, wherein, The first CSI report further comprises indication information of one or more of the following: a rank indication (RI), a channel quality indication (CQI), or a precoding matrix index (PMI), wherein a number of layers of the PMI is determined according to the RI.

4. The method of claim 1 or 2, wherein, The method further comprises: transmitting a second CSI report to the second communication device, wherein the second CSI report comprises indication information of one or more of the following: the RI, the CQI, or the PMI, wherein a number of layers of the PMI is determined according to the RI.

5. The method of claim 1 or 2, wherein, The first information further comprises indication information indicating that the first communication device transmits one or more of the following: eigenvalues of the eigenvector matrix of the target layer, singular values, a threshold value of the eigenvalues or the singular values of the eigenvector matrix of the target layer, a transmission period of the CSI of the target layer, or a transmission pattern of the CSI of the target layer.

6. The method of claim 1 or 2, wherein, The CSI of the target layer transmitted by the first communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

7. The method of claim 1 or 2, wherein, The first information is configuration information, and the configuration information is used to configure the first communication device to transmit the CSI of the target layer.

8. A CSI receiving method, comprising: The method is applied to a second communication device, and comprises: transmitting first information to a first communication device, wherein the first information comprises indication information indicating that the first communication device transmits channel state information (CSI) of a target layer; receiving a first CSI report from the first communication device, wherein the first CSI report comprises the CSI of the target layer; the CSI of the target layer comprises indication information of one or more of the following: a channel response of the target layer, an eigenvector matrix of the target layer, compression and / or quantization information of the eigenvector matrix, or compression and / or quantization information of the channel response of the target layer.

9. The method of claim 8, wherein, The CSI of the target layer further comprises indication information of one or more of the following: eigenvalues of the eigenvector matrix of the target layer, singular values, or the target layer.

10. The method of claim 8 or 9, wherein, The first CSI report further comprises indication information of one or more of the following: a rank indication (RI), a channel quality indication (CQI), or a precoding matrix index (PMI), wherein a number of layers of the PMI is determined according to the RI.

11. The method of claim 8 or 9, wherein, The method further comprises: receiving a second CSI report from the first communication device, the second CSI report including indication information of one or more of the following: RI, CQI, or PMI, the number of layers of the PMI being determined according to the RI.

12. The method of claim 8 or 9, wherein, the first information further including indication information for indicating that the first communication device transmits one or more of the following: an eigenvalue of a feature vector matrix of the target layer, a singular value of a feature vector matrix of the target layer, a threshold value of an eigenvalue or a singular value of a feature vector matrix of the target layer, a transmission period of the CSI of the target layer, or a transmission pattern of the CSI of the target layer.

13. The method of claim 8 or 9, wherein, the CSI of the target layer received by the second communication device is used for one or more of the following: model training, model inference, model testing, model verification, or model monitoring.

14. The method of claim 8 or 9, wherein, the first information is configuration information for configuring the first communication device to transmit the CSI of the target layer.

15. An apparatus, comprising: comprising means for performing the method of any one of claims 1 to 7.

16. A communications device, characterized by comprising a processor and an interface circuit for receiving signals from other communication devices outside the communication device and transmitting to the processor or sending signals from the processor to other communication devices outside the communication device, the processor being configured to implement the method of any one of claims 1 to 7 by logic circuit or executable code instructions.

17. An apparatus, comprising: comprising means for performing the method of any one of claims 8 to 14.

18. A communications device, characterized by comprising a processor and an interface circuit for receiving signals from other communication devices outside the communication device and transmitting to the processor or sending signals from the processor to other communication devices outside the communication device, the processor being configured to implement the method of any one of claims 8 to 14 by logic circuit or executable code instructions.

19. A computer-readable storage medium, characterized in that, the storage medium stores a computer program or instructions, when the computer program or instructions are executed by a communication device, the method of any one of claims 1 to 7 is implemented, or the method of any one of claims 8 to 14 is implemented.

20. A system, comprising: comprising the device of claim 15 or 16, and the device of claim 17 or 18.

Images