Communication method, apparatus and system

Abstract

The present application relates to a communication method, apparatus and system. A terminal device receives first information from a network device. The terminal device uses an AI model or an AI function to process the first information, the AI model or the AI function being used for implementing some or all of the functions of a first receiver of the terminal device. The terminal device determines the performance of the AI model or the AI function on the basis of the processing result of the first information. For an AI model or an AI function for implementing a receiving function, embodiments of the present application provide a mode for determining the performance of the AI model or the AI function.

Description

Communication method, device and system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on September 6, 2023, with application number 202311149742.6 and application name “A Communication Method, Device and System”, the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present application relates to the field of communication technology, and in particular to a communication method, device, and system. Background Art

[0004] Currently, the application of artificial intelligence (AI) in communications, particularly air interfaces, has been extensively researched. For example, research projects have been initiated on air interface AI, focusing on the basic framework for its use and several typical AI air interface use cases (AI channel state information (CSI) feedback, AI beam management, and AI positioning). Considering that future communication systems will be further integrated with AI technology, more traditional communication modules can be replaced by AI, with AI reception algorithms being one possible approach.

[0005] An AI receiving algorithm utilizes an AI model to replace some or all functional modules in a traditional receiver. Compared to traditional receiving algorithms, AI models can be closely integrated with specific application scenarios and conditions. For example, AI models can be trained specifically for these scenarios or conditions, resulting in superior performance. Alternatively, while achieving performance similar to that of traditional receiving algorithms, the use of AI models can reduce computational complexity and further reduce the power consumption of the receiving device, thereby achieving energy savings.

[0006] When using AI models to implement reception functions, the performance of the AI ​​models is closely related to the reception performance. However, there is currently no consensus on how to determine the performance of the AI ​​models used to implement reception functions.

[0007] Summary of the Invention

[0008] Embodiments of the present application provide a communication method, device, and system, which provide a method for determining the performance of an AI model or AI function.

[0009] In a first aspect, a first communication method is provided, which can be executed by a terminal device, or by other devices including terminal device functions, or by a chip system (or, chip) or other functional module, which can realize the functions of the terminal device, and the chip system or functional module is, for example, set in the terminal device. The terminal device may include a first receiver. The method includes: receiving first information from a network device; processing the first information using an AI model or an AI function, wherein the AI ​​model or the AI ​​function is used to realize part or all of the functions of the first receiver; and determining the performance of the AI ​​model or the AI ​​function based on the processing result of the first information.

[0010] In an embodiment of the present application, after receiving the first information, the terminal device uses an AI model or AI function to process the first information, and thus the performance of the AI ​​model or AI function can be determined based on the processing result. That is, for the AI ​​model or AI function used to implement the receiving function, the embodiment of the present application provides a method for determining its performance. For example, if it is determined that the performance of the AI ​​model or AI function is poor, corresponding measures can be taken, such as adjusting the performance of the AI ​​model or AI function, or reselecting or retraining the AI ​​model or AI function, thereby improving the receiving performance.

[0011] In an optional embodiment, the amount of information in the first information is less than or equal to a first threshold. Reducing the amount of information in the first information is beneficial to reducing the complexity of processing the first information by the terminal device.

[0012] In an optional embodiment, the method further includes: receiving downlink control information from the network device, the downlink control information being used to schedule the first information, wherein the downlink control information includes first indication information, the first indication information being used to indicate that the first information is used to determine the performance of the AI ​​model or the AI ​​function; or, the downlink control information is scrambled using a first RNTI, the first RNTI being used to indicate that the first information is used to determine the performance of the AI ​​model or the AI ​​function. The downlink control information may indicate the purpose of the first information by explicitly including the first indication information, making the indication clearer; or, the downlink control information may indicate the purpose of the first information by implicitly indicating the purpose of the first information by scrambling, thereby achieving the purpose of the indication without adding additional indication overhead.

[0013] In an optional embodiment, the downlink control information is further used to indicate a transmission resource for the first information. The transmission resource for the first information can be predefined by a protocol, or agreed upon in advance by the network device and the terminal device, or can also be indicated by the downlink control information, which is a more flexible approach.

[0014] In an optional embodiment, the downlink control information indicates the transmission resource of the first information in the following manner: the downlink control information is used to indicate the location of the transmission resource of the first information; or, the downlink control information includes the index of the transmission resource of the first information; or, the downlink control information is also used to schedule second information, the second information is service data or control information, and the downlink control information is also used to indicate the offset of the transmission resource of the first information relative to the transmission resource of the second information. The downlink control information can directly indicate the location of the transmission resource of the first information, which is a clear indication; or the downlink control information can include the index of the transmission resource of the first information. For example, if the network device and the terminal device have pre-agreed on the resources for transmitting test information, or the protocol has pre-defined the resources for transmitting test information, then the downlink control information only needs to include the index of the corresponding resource, which can save the overhead of the downlink control information; or, if the downlink control information also schedules the second information, then the downlink control information can also indicate the offset of the transmission resource of the first information relative to the transmission resource of the second information, thereby also realizing the indication of the transmission resource of the first information; or, if the downlink control information also schedules the second information, and the offset of the transmission resource of the first information relative to the transmission resource of the second information can also be pre-agreed on by the network device and the terminal device, or pre-defined by the protocol, then the downlink control information does not need to indicate the offset, thereby saving the overhead of the downlink control information. It can be seen that the method of indicating the transmission resource of the first information is more flexible.

[0015] In an optional embodiment, the first information and the second information are independently encoded. Independently encoding the first information and the second information can reduce the impact of information used to determine the performance of the AI ​​model or AI function on ordinary business.

[0016] In an optional embodiment, the transmission resource of the first information is preconfigured or predefined by the network device. If the network device preconfigured or the protocol predefined, it is not necessary to indicate the transmission resource of the first information through downlink control information, which can save downlink control information overhead; or, even if the network device preconfigured or the protocol predefined, the transmission resource of the first information can still be indicated through downlink control information. For example, if multiple transmission resources are preconfigured or predefined, the downlink control information can indicate some or all of the transmission resources, which makes the selection of the transmission resource of the first information more flexible.

[0017] In an optional embodiment, the method further includes: sending a first request message to the network device, the first request message being used to request determination of the performance of the AI ​​model or the AI ​​function; or receiving a second indication message from the network device, the second indication message being used to instruct determination of the performance of the AI ​​model or the AI ​​function. The detection (or monitoring) process of the AI ​​model or AI function can be initiated by the terminal device, or by the network device, which is more flexible.

[0018] In an optional implementation, the first request information is included in a MAC CE, or in a UCI, or in an RRC message.

[0019] In an optional embodiment, the first request information also includes first auxiliary information, which is auxiliary information used to determine the performance of the AI ​​model or the AI ​​function. When the terminal device requests to start the detection process of the AI ​​model or AI function, it can report the first auxiliary information to the network device to help the network device determine the test information accordingly, so that the test information determined by the network device is more in line with the real-time needs of the terminal device. Among them, the embodiment of the present application refers to the information used to determine the AI ​​model or AI function as test information or monitoring information, etc., and this article mainly takes test information as an example.

[0020] In an optional embodiment, the method further includes: receiving first configuration information from the network device, the first configuration information configuring parameters for information used to determine the performance of the AI ​​model or the AI ​​function. The network device may configure the parameters of the test information to the terminal device, so that the terminal device can determine the test information accordingly and receive and process the test information.

[0021] In an optional embodiment, the parameters of the information used to determine the performance of the AI ​​model or the AI ​​function include one or more of the following: type information of the information used to determine the performance of the AI ​​model or the AI ​​function; information about how the information used to determine the performance of the AI ​​model or the AI ​​function is generated; or information about how the information used to determine the performance of the AI ​​model or the AI ​​function is sent. The parameters of the test information may also include other parameters, such as the size of the test information, which are not limited thereto.

[0022] In an optional embodiment, the first configuration information is also used to configure a true value; or, the first configuration information is also used to configure a method for determining a true value; wherein the true value includes label information corresponding to the first information, and the label information is used by the terminal device to determine the performance of the AI ​​model or the AI ​​function based on the processing result of the first information and the label information. If the terminal device uses a direct detection method to detect the performance of the AI ​​model or AI function, the terminal device can obtain a true value. For example, the true value can be understood as the original information. For example, if the terminal device receives and processes the first information completely accurately, the information obtained should be exactly the same as the true value. Thus, the terminal device can determine the performance of the AI ​​model or AI function based on the processing result of the first information and the true value.

[0023] In an optional implementation, the true value and the first information are both data, or both are sequences, or both are reference signals.

[0024] In an optional embodiment, the method further includes: receiving first configuration information from the network device, the first configuration information including third indication information, the third indication information being used to indicate that the first information is only used to determine the performance of the AI ​​model or the AI ​​function. The network device may indicate to the terminal device that the first information is not used for other purposes, and the terminal device may use the first information only to determine the performance of the AI ​​model or AI function and not for other purposes.

[0025] In an optional embodiment, the first information is only used to determine the performance of the AI ​​model or the AI ​​function. For example, if the first information is not used for other purposes, or if it is pre-agreed upon between the network device and the terminal device, or pre-defined by the protocol, the network device does not need to indicate the terminal device, thereby saving indication overhead.

[0026] In an optional embodiment, the first configuration information is included in a unicast message; or, the first configuration information is included in a multicast message; or, the first configuration information is included in a broadcast message. The first configuration information can be sent to the terminal device via unicast, multicast, or broadcast, thereby reducing transmission overhead.

[0027] In an optional implementation, the broadcast message is a SIB or an RRC message.

[0028] In an optional embodiment, the method further includes: sending second auxiliary information to the network device, where the second auxiliary information is used to indicate one or more of the following: whether the terminal device supports the AI ​​model or the AI ​​function, the manner in which the terminal device performs AI model or AI function detection, or auxiliary information used to determine the performance of the AI ​​model or the AI ​​function. The terminal device can send the second auxiliary information to the network device to assist the network device in determining test information accordingly, so that the test information determined by the network device better meets the needs of the terminal device.

[0029] In an optional embodiment, the auxiliary information of the information used to determine the performance of the AI ​​model or the AI ​​function includes one or more of the following: type information of the AI ​​model or the AI ​​function; size of the information used to determine the performance of the AI ​​model or the AI ​​function; type information of the information used to determine the performance of the AI ​​model or the AI ​​function; generation method information of the information used to determine the performance of the AI ​​model or the AI ​​function; or sending method information of the information used to determine the performance of the AI ​​model or the AI ​​function.

[0030] In an optional embodiment, the terminal device further includes a second receiver, and determining the performance of the AI ​​model or the AI ​​function based on the processing result of the first information includes: using the AI ​​model or the AI ​​function and the second receiver to process the first information, respectively, and determining the performance of the AI ​​model or the AI ​​function based on the processing result of the AI ​​model or the AI ​​function and the processing result of the second receiver; or using the AI ​​model or the AI ​​function to process the first information, and determining the performance of the AI ​​model or the AI ​​function based on tag information corresponding to the first information and the processing result of the AI ​​model or the AI ​​function. The terminal device can use different methods to determine the performance of the AI ​​model or AI function, which is more flexible.

[0031] In an optional embodiment, the first information is included in a PDSCH; or the first information is included in a MAC PDU, and the MAC header of the MAC PDU is used to indicate that the first information is information used to determine the performance of the AI ​​model or the AI ​​function. The first information can be physical layer information, MAC layer information, or information of a higher protocol layer, without limitation.

[0032] In an optional embodiment, the first information is included in the PDSCH, and the method further includes: not sending HARQ feedback information corresponding to the first information. Because the first information is not used for other purposes, the terminal device does not need to send HARQ feedback information for the first information, thereby saving transmission overhead.

[0033] In an optional embodiment, the first information is included in the MAC PDU, and the method further includes: the MAC layer of the terminal device not delivering the first information to an upper layer of the MAC layer; or, the MAC layer of the terminal device discarding the MAC PDU. Because the first information is not used for other purposes, the terminal device does not need to process the first information as ordinary information, thereby reducing the impact of the first information on ordinary services.

[0034] In an optional implementation, the first information is included in the PDSCH, and the first information is broadcast information.

[0035] In an optional embodiment, the first information is information specifically used to determine the performance of the AI ​​model or the AI ​​function, or is business data. The first information may not be used for other purposes, but only for determining the performance of the AI ​​model or AI function; alternatively, the first information may be business data, which is equivalent to using business data to test the AI ​​model or AI function, eliminating the need for additional information specifically used to test the AI ​​model or AI function. This can simplify the model testing process and expand the scope of business data utilization.

[0036] In a second aspect, a second communication method is provided, which can be performed by a network device, or by other devices including network device functions, or by a chip system (or, chip) or other functional module, which can implement the functions of the network device, and the chip system or functional module is, for example, provided in the network device. The network device is, for example, an access network device, for example, one implementation of the access network device is a base station. The method includes: sending first information, wherein the first information is used to determine the performance of an AI model or an AI function, and the AI ​​model or the AI ​​function is used to implement part or all of the functions of a first receiver of a terminal device.

[0037] In an optional implementation, the information volume of the first information is less than or equal to a first threshold.

[0038] In an optional embodiment, the method further includes: sending downlink control information to the terminal device, the downlink control information being used to schedule the first information, wherein the downlink control information includes first indication information, and the first indication information is used to indicate that the first information is used to determine the AI ​​model or the AI ​​function performance; or, the downlink control information is encrypted by a first RNTI, and the first RNTI is used to indicate that the first information is used to determine the AI ​​model or the AI ​​function performance.

[0039] In an optional implementation, the downlink control information is further used to indicate a transmission resource of the first information.

[0040] In an optional embodiment, the downlink control information indicates the transmission resource of the first information in the following manner: the downlink control information is used to indicate the location of the transmission resource of the first information; or, the downlink control information includes the index of the transmission resource of the first information; or, the downlink control information is also used to schedule second information, the second information is service data or control information, and the downlink control information is also used to indicate the offset of the transmission resource of the first information relative to the transmission resource of the second information.

[0041] In an optional embodiment, the first information and the second information are independently encoded.

[0042] In an optional implementation manner, the transmission resource of the first information is preconfigured or predefined by the network device.

[0043] In an optional embodiment, the method further includes: receiving first request information from the terminal device, the first request information being used to request determination of the performance of the AI ​​model or the AI ​​function; or, sending second indication information to the terminal device, the second indication information being used to indicate determination of the performance of the AI ​​model or the AI ​​function.

[0044] In an optional implementation, the first request information is included in a MAC CE, or in a UCI, or in an RRC message.

[0045] In an optional embodiment, the first request information also includes first auxiliary information, where the first auxiliary information is auxiliary information used to determine the AI ​​model or the AI ​​function performance.

[0046] In an optional embodiment, the method further includes: sending first configuration information to the terminal device, wherein the first configuration information configures parameters of information used to determine the performance of the AI ​​model or the AI ​​function.

[0047] In an optional embodiment, the parameters of the information used to determine the performance of the AI ​​model or the AI ​​function include one or more of the following: type information of the information used to determine the performance of the AI ​​model or the AI ​​function; generation method information of the information used to determine the performance of the AI ​​model or the AI ​​function; or transmission resource information of the information used to determine the performance of the AI ​​model or the AI ​​function.

[0048] In an optional embodiment, the first configuration information is also used to configure the true value; or, the first configuration information is also used to configure the method for determining the true value; wherein, the true value includes label information corresponding to the first information, and the label information is used by the terminal device to determine the performance of the AI ​​model or the AI ​​function based on the processing result of the first information and the label information.

[0049] In an optional implementation, the true value and the first information are both data, or both are sequences, or both are reference signals.

[0050] In an optional embodiment, the method further includes: receiving first configuration information from the network device, the first configuration information including third indication information, and the third indication information is used to indicate that the first information is only used to determine the performance of the AI ​​model or the AI ​​function.

[0051] In an optional embodiment, the first information is only used to determine the performance of the AI ​​model or the AI ​​function.

[0052] In an optional implementation, the first configuration information is included in a unicast message; or, the first configuration information is included in a multicast message; or, the first configuration information is included in a broadcast message.

[0053] In an optional implementation, the broadcast message is a SIB or an RRC message.

[0054] In an optional embodiment, the method further includes: receiving second auxiliary information from the terminal device, the second auxiliary information being used to indicate one or more of the following: whether the terminal device supports the AI ​​model or the AI ​​function, the manner in which the terminal device performs AI model or AI function detection, and auxiliary information of information used to determine the performance of the AI ​​model or the AI ​​function.

[0055] In an optional embodiment, the auxiliary information of the information used to determine the performance of the AI ​​model or the AI ​​function includes one or more of the following: type information of the AI ​​model or the AI ​​function; size of the information used to determine the performance of the AI ​​model or the AI ​​function; type information of the information used to determine the performance of the AI ​​model or the AI ​​function; generation method information of the information used to determine the performance of the AI ​​model or the AI ​​function; or sending method information of the information used to determine the performance of the AI ​​model or the AI ​​function.

[0056] In an optional embodiment, the first information is included in the PDSCH; or, the first information is included in the MAC PDU, and the MAC header of the MAC PDU is used to indicate that the first information is information used to determine the performance of the AI ​​model or the AI ​​function.

[0057] In an optional implementation, the first information is included in the PDSCH, and the first information is broadcast information.

[0058] In an optional embodiment, the first information is information specifically used to determine the performance of the AI ​​model or the AI ​​function, or is business data.

[0059] Regarding the technical effects brought about by the second aspect or various optional implementations, reference may be made to the introduction to the technical effects of the first aspect or corresponding implementations.

[0060] In a third aspect, a communication device is provided. The communication device may be the terminal device described in any one of the first to second aspects. The communication device has the functions of the terminal device described above. The communication device may be, for example, a terminal device, or a larger device including a terminal device, or a functional module within the terminal device, such as a baseband device or a chip system. In one optional implementation, the communication device includes a baseband device and a radio frequency device. In another optional implementation, the communication device includes a processing unit (sometimes also referred to as a processing module) and a transceiver unit (sometimes also referred to as a transceiver module). The transceiver unit is capable of performing both transmitting and receiving functions. When the transceiver unit performs the transmitting function, it may be referred to as a transmitting unit (sometimes also referred to as a transmitting module); when the transceiver unit performs the receiving function, it may be referred to as a receiving unit (sometimes also referred to as a receiving module). The transmitting unit and the receiving unit may be the same functional module, referred to as a transceiver unit, which is capable of both transmitting and receiving functions; alternatively, the transmitting unit and the receiving unit may be different functional modules, with the transceiver unit being a collective term for these functional modules.

[0061] In an optional embodiment, the transceiver unit (or, the receiving unit) is used to receive first information from a network device; the processing unit is used to process the first information using an AI model or an AI function, and the AI ​​model or the AI ​​function is used to implement part or all of the functions of the first receiver; the processing unit is also used to determine the performance of the AI ​​model or the AI ​​function based on the processing result of the first information.

[0062] In an optional embodiment, the communication device also includes a storage unit (sometimes also referred to as a storage module), and the processing unit is used to couple with the storage unit and execute the program or instructions in the storage unit, so as to enable the communication device to perform the functions of the terminal device described in any one of the first to second aspects above.

[0063] In a fourth aspect, a communication device is provided. The communication device may be the network device described in any one of the first to second aspects above. The communication device has the functions of the above-mentioned network device. The communication device is, for example, a network device, or a larger device including a network device, or a functional module in a network device, such as a baseband device or a chip system. In an optional implementation, the communication device includes a baseband device and a radio frequency device. In another optional implementation, the communication device includes a processing unit (sometimes also referred to as a processing module) and a transceiver unit (sometimes also referred to as a transceiver module). For the implementation of the transceiver unit, please refer to the introduction of the fourth aspect.

[0064] In an optional embodiment, the transceiver unit (or, the sending unit) is used to send first information, where the first information is used to determine the performance of an AI model or an AI function, and the AI ​​model or the AI ​​function is used to implement part or all of the functions of the first receiver of the terminal device.

[0065] In an optional embodiment, the communication device also includes a storage unit (sometimes also referred to as a storage module), and the processing unit is used to couple with the storage unit and execute the program or instructions in the storage unit, enabling the communication device to perform the functions of the network device described in any one of the first to second aspects above.

[0066] In a fifth aspect, a communication device is provided. The communication device may be a first terminal device, or a chip or chip system used in the first terminal device. The communication device includes a communication interface and a processor, and optionally, a memory. The memory is used to store a computer program. The processor is coupled to the memory and the communication interface. When the processor reads the computer program or instructions, the communication device executes the method performed by the terminal device in the above aspects.

[0067] In a sixth aspect, a communication device is provided. The communication device may be a network device, or a chip or chip system used in a network device. The communication device includes a communication interface and a processor, and optionally, a memory. The memory is used to store a computer program. The processor is coupled to the memory and the communication interface. When the processor reads the computer program or instructions, the communication device executes the method performed by the network device in the above aspects.

[0068] In a seventh aspect, a communication system is provided, comprising a terminal device and a network device, wherein the terminal device is configured to execute the methods described in the aforementioned aspects, and the network device is configured to execute the methods described in the aforementioned aspects. For example, the terminal device may be implemented using the communication apparatus described in the third or fifth aspect, and the network device may be implemented using the communication apparatus described in the fourth or sixth aspect.

[0069] In an eighth aspect, a computer-readable storage medium is provided, which is used to store computer programs or instructions. When the computer-readable storage medium is executed, the methods executed by the terminal device or network device in the above aspects are implemented.

[0070] In a ninth aspect, a computer program product comprising instructions is provided, which, when executed on a computer, enables the methods described in the above aspects to be implemented.

[0071] In a tenth aspect, a chip system is provided, comprising a processor and an interface, wherein the processor is configured to call and execute instructions from the interface so that the chip system implements the methods in the above aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] FIG1 is a schematic diagram of a conventional receiving and processing algorithm;

[0073] FIG2 is a schematic diagram of using an AI model to replace functional modules in a traditional receiver;

[0074] FIG3 is a schematic diagram of the structure of a communication network used in an embodiment of the present application;

[0075] FIG4 is a flow chart of a communication method provided in an embodiment of the present application;

[0076] FIG5 is a schematic diagram of transmission resources of first information and second information in an embodiment of the present application;

[0077] 6A and 6B are schematic diagrams of two implementations of the first field and the second field in an embodiment of the present application;

[0078] FIG7 is a schematic diagram of a device provided in an embodiment of the present application;

[0079] FIG8 is a schematic diagram of another device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0081] In the embodiments of the present application, the number of nouns, unless otherwise specified, means "singular noun or plural noun", that is, "one or more". "At least one" means one or more, and "plural" means two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. For example, A / B means: A or B. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c means: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0082] The ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish between multiple objects, and are not used to limit the size, content, sequence, timing, priority or importance of multiple objects. For example, the first information and the second information can be the same information or different information, and this name does not indicate the difference in resources occupied by the two information, sending order, sending end / receiving end, content, size, application scenario, priority or importance, etc. In addition, the numbering of the steps in the various embodiments introduced in this application is only for distinguishing different steps, and is not used to limit the order between the steps. For example, S401 may occur before S402, or may occur after S402, or may occur at the same time as S402.

[0083] Below, some terms or concepts in the embodiments of the present application are explained to facilitate understanding by those skilled in the art.

[0084] In the embodiment of the present application, the terminal device is a device with wireless transceiver function, which can be a fixed device, a mobile device, a handheld device (such as a mobile phone), a wearable device, an in-vehicle device, or a wireless device built into the above device (such as a communication module, a modem, or a chip system, etc.). The terminal device is used to connect people, objects, machines, etc., and can be widely used in various scenarios, such as but not limited to the following scenarios: sensing scenarios, cellular communications, device-to-device communication (D2D), vehicle to everything (V2X), machine-to-machine / machine-type communications (M2M / MTC), Internet of Things (IoT), virtual reality (VR), augmented reality (AR), industrial control, self-driving, remote medical, smart grid, smart furniture, smart office, smart wearable, smart transportation, smart city, drones, robots and other scenarios. The terminal device may sometimes be referred to as user equipment (UE), terminal, access station, UE station, remote station, wireless communication device, or user equipment, etc.

[0085] In the embodiments of the present application, the communication device for implementing the terminal device function may be a terminal device, or may be a device capable of supporting the terminal device to implement the function, such as a chip system, which may be installed in the terminal device. In the technical solutions provided in the embodiments of the present application, the technical solutions provided in the embodiments of the present application are described by taking the terminal device as an example in which the device for implementing the terminal device function is a terminal device. In addition, for ease of description, the terminal device in the embodiments of the present application is described by taking a UE as an example.

[0086] The network devices in the embodiments of the present application include, for example, access network devices, and / or core network devices. The access network device is a device with wireless transceiver functions, which is used to communicate with the terminal device. The access network devices include but are not limited to base stations (base transceiver station (BTS), node B (Node B), evolved node B (eNodeB) / eNB, or the next generation node B (gNodeB) / gNB), transmission reception points (TRP), base stations subsequently evolved by the third generation partnership project (3GPP), access nodes in wireless fidelity (Wi-Fi) systems, wireless relay nodes, wireless backhaul nodes, etc. The base station can be: a macro base station, a micro base station, a pico base station, a small station, a relay station, etc. Multiple base stations can support networks with the same access technology, or they can support networks with different access technologies. The base station can include one or more co-station or non-co-station transmission and receiving points. The access network device may also be a wireless controller, a centralized unit (CU), and / or a distributed unit (DU) in a cloud radio access network (CRAN) scenario. The access network device may also be a server, etc. For example, the network device in the V2X technology may be a road side unit (RSU). The following describes the access network device using a base station as an example. The base station can communicate with the terminal device, or it can communicate with the terminal device through a relay station. The terminal device can communicate with multiple base stations in different access technologies. The core network device is used to implement functions such as mobility management, data processing, session management, policy and billing. The names of the devices that implement core network functions in systems with different access technologies may be different, and the embodiments of the present application are not limited to this. Taking the fifth generation mobile communication technology (5G) system as an example, the core network equipment includes: access and mobility management function (AMF), session management function (SMF), policy control function (PCF) or user plane function (UPF), etc.

[0087] In the embodiments of the present application, the communication device for implementing the network device function may be a network device, or may be a device capable of supporting the network device to implement the function, such as a chip system, which may be installed in the network device. In the technical solutions provided in the embodiments of the present application, the technical solutions provided in the embodiments of the present application are described by taking the network device as an example of the device for implementing the network device function being a network device.

[0088] The application of AI in communications, especially in air interfaces, has been widely studied. For example, a research project has been launched on air interface AI, and the basic framework for the use of air interface AI and several typical AI air interface use cases / features (AI CSI feedback, AI beam management, AI positioning) have been studied. When the network side uses AI on the UE side, it can be based on the granularity of the AI ​​function or the granularity of the AI ​​model. In the embodiment of the present application, the AI ​​function refers to an AI feature / feature group enabled by one or more configurations (such as radio resource control (RRC) configuration), which can be determined by the applicable conditions supported by the UE. For example, CSI measurement resource configuration 1 and CSI reporting configuration 1 meet the applicable conditions for the UE to use AI for AI CSI feedback, then the AI ​​CSI reporting function based on CSI measurement resource configuration 1 and CSI reporting configuration 1 can be called an AI function. The AI ​​model is the specific implementation of the AI ​​function, and the AI ​​model characterizes the mapping relationship between the input and output of the model. The AI ​​model can be a neural network, a linear regression model, a decision tree model, a support vector machine (SVM), a Bayesian network, a Q learning model or other machine learning models. In an embodiment of the present application, the use process involved in the AI ​​function or AI model may include at least one of the following: data collection (collecting training data and / or reasoning data), data preprocessing, model training (or model learning), model information release (configuration model information), model delivery / transmission, model management (activation, deactivation, switching, fallback), model verification (verification of the trained model), model reasoning (reasoning using the trained model), or reasoning result release. Among them, reasoning can also be called prediction. Optionally, the AI ​​model may include a machine learning (ML) model.

[0089] Considering that future communication systems will be further closely integrated with AI technology, more traditional communication modules can also be replaced by AI, among which AI receiving algorithm is a possible direction.

[0090] Please refer to Figure 1, which shows a schematic diagram of a traditional receiver processing algorithm. Each small box in Figure 1 can be understood as a functional module or one or more processing steps. According to Figure 1, after receiving the downlink signal, the typical processing algorithm includes RF processing and baseband processing. RF processing includes multiple functional modules such as low-noise amplifiers, downconversion, amplifiers, and analog-to-digital conversion. Baseband processing also includes multiple functional modules such as fast Fourier transform (FFT) / cyclic prefix (CP) removal, synchronization, channel estimation, equalization / demodulation, and channel decoding. Traditional receiver algorithms (such as those shown in Figure 1) are modular in design. Specifically, they target a single functional module by establishing a mathematical model and then finding the optimal or suboptimal solution for that module. This essentially results in a local optimal solution for the entire receiver, not necessarily a global optimal solution. However, if an AI model replaces some or all of the functional modules in a traditional receiver, such as one or more of the small boxes in Figure 1, a global optimal solution can be achieved for the entire receiver.

[0091] AI models can replace any one or more functional modules in a traditional receiver. For example, among all the functional modules included in the RF processing and baseband processing in Figure 1, any one or more functional modules can be replaced by AI models. One AI model can replace one functional module; alternatively, one AI model can replace multiple functional modules, for example, using a fusion approach so that one AI model performs the functions of multiple functional modules. For example, referring to Figure 2, the three functional modules of a traditional receiver—channel estimation, equalization / demodulation, and channel decoding—can be replaced by AI models. For example, three AI models can be used to replace each of these three functional modules, or a single AI model can be used to replace all three. The functional modules of channel estimation, equalization / demodulation, and channel decoding in Figure 2 can also be referred to as different features. When AI is enabled for these features under specific configurations, they can be referred to as corresponding AI functions. For example, if AI can be used for channel estimation under bandwidth configuration 1, then channel estimation under bandwidth configuration 1 can be referred to as an AI function. For another example, if AI can be used for equalization / demodulation under modulation configuration 1, then equalization / demodulation under modulation configuration 1 can be referred to as an AI function. An AI function may include one or more AI models, all of which can implement the AI ​​function. The specific AI model to be used can be determined by the UE itself, that is, the AI ​​model is transparent to the network device, but the AI ​​function is not transparent to the network device. In the embodiments of the present application, the management granularity of the network device when the UE uses AI reception is not limited. From the network side, the UE can implement the functional modules in the receiver at the granularity of the AI ​​model, or at the granularity of the AI ​​function.

[0092] When using AI models to implement reception functions, the performance of the AI ​​models is closely related to the reception performance. However, there is currently no consensus on how to determine the performance of the AI ​​models used to implement reception functions.

[0093] In view of this, after receiving the first information, the UE in the embodiment of the present application can use the AI ​​model or AI function to process the first information, so that the performance of the AI ​​model or AI function can be determined based on the processing result. That is, for the AI ​​model or AI function used to implement the receiving function, the embodiment of the present application provides a method for determining its performance. Furthermore, if it is determined that the performance of the AI ​​model or AI function is poor, corresponding measures can be taken, such as adjusting the performance of the AI ​​model or AI function, turning off the AI ​​model or AI function, reselecting or retraining the AI ​​model or AI function, or falling back to a non-AI receiving function, etc., thereby improving the receiving performance.

[0094] The technical solutions provided in the embodiments of the present application can be applied to the fourth generation mobile communication technology (the 4th generation, 4G) system, such as the long term evolution (long term evolution, LTE) system, or can be applied to the 5G system, such as the new radio (new radio, NR) system, or can also be applied to the next generation mobile communication system or other similar communication systems, such as the sixth generation mobile communication technology (the 6th generation, 6G) system, etc., without specific limitation. In addition, the technical solutions provided in the embodiments of the present application can be applied to D2D scenarios, such as NR-D2D scenarios, etc., or can be applied to V2X scenarios, such as NR-V2X scenarios, etc. For example, it can be used in the fields of factory manufacturing, whole house intelligence, intelligent driving, assisted driving, or intelligent networked vehicles.

[0095] In the current discussion, the AI ​​model of the air interface can be divided into two categories: unilateral model and bilateral model. The bilateral model means that the network side (such as the base station) and the UE side each have a model, and the models on both ends work together to complete a function. Therefore, the models on both ends can also be regarded as two parts of an overall model. The unilateral model means that the AI ​​model is only deployed on the network side or the UE side. Compared with the bilateral model, the unilateral model is simpler in terms of model management. The AI ​​model or AI function for implementing the receiving function involved in the various embodiments of the present application can be a unilateral model deployed on the UE side. Of course, there is no limitation to this. For example, the AI ​​model or AI function for implementing the receiving function involved in the various embodiments of the present application can also be a unilateral model deployed on other devices (such as the network side), or the AI ​​model or AI function for implementing the receiving function involved in the various embodiments of the present application can also be a bilateral model deployed on both sides (such as the network side and the UE side) when considering that the transmitting end also participates in the joint design.

[0096] Figure 3 is a communication network architecture applicable to an embodiment of the present application. Figure 3 includes a network device and a UE, and the network device and the UE are capable of communicating. For example, the UE deploys an AI model or an AI function, and the AI ​​model or the AI ​​function is capable of implementing a receiving algorithm. For example, the UE may include a first receiver, wherein some or all functional modules within the first receiver may be replaced by the AI ​​model or the AI ​​function, or the AI ​​model or the AI ​​function may implement the functions of some or all functional modules within the first receiver, or the AI ​​model or the AI ​​function may implement some or all functions of the first receiver.

[0097] In order to better introduce the embodiments of the present application, the methods provided by the embodiments of the present application are described below in conjunction with the accompanying drawings. In the various embodiments of the present application, the two features "AI model" and "AI function" can be replaced with each other. In the various embodiments of the present application, the device that "deploys" the AI ​​model may refer to the device that runs the AI ​​model. For example, "deployment" can also be understood as "running" or "using" or "storing". For example, in Figure 3, the UE is the device that deploys the AI ​​model. The UE can run the AI ​​model, perform reasoning and other processes through the AI ​​model, and obtain the reasoning results output by the AI ​​model. In various embodiments of the present application, the AI ​​model or AI function can be obtained by training the UE; or, the AI ​​model or AI function can be obtained by training the network device, and after the training is completed, the network device can send the parameters of the AI ​​model or AI function to the UE, so that the UE can deploy the AI ​​model or AI function; or, the AI ​​model or AI function can be obtained by training the third-party device, and after the training is completed, the third-party device can send the parameters of the AI ​​model or AI function to the UE, so that the UE can deploy the AI ​​model or AI function, or after the training is completed, the third-party device can send the parameters of the AI ​​model or AI function to the network device, and the network device then sends the parameters of the AI ​​model or AI function to the UE, so that the UE can deploy the AI ​​model or AI function. The third-party device is, for example, an operation, administration and maintenance (OAM), an over the top server (OTT server), or other devices.

[0098] Unless otherwise specified herein, all steps indicated by dashed lines in the accompanying drawings corresponding to the various embodiments of the present application are optional steps. The methods provided in the various embodiments of the present application can be applied to the network architecture shown in FIG3 . For example, the UE involved in the various embodiments of the present application can be the UE in FIG3 ; the network device described in the various embodiments of the present application can be the network device in FIG3 .

[0099] An embodiment of the present application provides a communication method. Please refer to Figure 4, which is a flowchart of the method.

[0100] S401: The UE sends second auxiliary information to the network device. Correspondingly, the network device receives the second auxiliary information from the UE. The second auxiliary information may also have other names, such as second requirement information or second expectation information, which are not limited thereto.

[0101] The UE may include a first receiver, and some or all functions within the first receiver may be implemented by an AI model or AI function. For example, the AI ​​model or AI function may implement the functions of some or all functional modules within the first receiver. Therefore, the first receiver may also be referred to as an AI receiver, or may have other names. In the embodiment of the present application, the UE may determine the performance of the AI ​​model or AI function.

[0102] In an embodiment of the present application, the information used to determine the performance of the AI ​​model or AI function may be referred to as test information, or as monitoring information, or may have other names, and will be introduced below using "test information" as an example. The second auxiliary information may include auxiliary information of the test information, or the network device may determine the test information based on the second auxiliary information and / or other factors, or the second auxiliary information may indicate the UE's expectations for the test information. The test information may include the first information to be introduced later, for example, the network device may determine the first information based on the second auxiliary information and / or other information, or the second auxiliary information may indicate the UE's expectations for the first information.

[0103] Optionally, the second auxiliary information may indicate (or include) one or more of the following: whether the UE supports the AI ​​model or AI function, the way in which the UE detects (monitors) the AI ​​model or AI function, or auxiliary information of the test information. Among them, detecting the AI ​​model or AI function can also be understood as monitoring the AI ​​model or AI function. Optionally, whether the UE supports the AI ​​model or AI function may also be related to the type of UE, or the network device may determine it through the registration information of the UE. In this case, whether the UE supports the AI ​​model or AI function may not be indicated by the second auxiliary information. For example, the second auxiliary information may not indicate (or include) the information that "UE supports the AI ​​model or AI function".

[0104] The way in which the UE performs AI model or AI function detection, for example, includes a direct detection method (or a direct monitoring method) or an indirect detection method (or an indirect monitoring method). Among them, the direct detection method can be a method of detection with the help of a true value (or a label), wherein the true value and the label can be the same feature, that is, the "true value" is the "label"; or, the true value can include the label. For example, the label can be the original information. In the direct detection method, the network device performs a series of processing processes such as modulation and mapping on the original information, and then sends the processed original information to the UE; the UE receives the processed original information, and processes the processed original information using the AI ​​model or AI function (for example, using the first receiver to process the processed original information), such as performing radio frequency processing and baseband processing, etc., to obtain information A. The UE can determine the performance of the AI ​​model or AI function based on the information A and the original information.

[0105] The indirect detection method may not rely on the true value (or, label). In the indirect detection method, the network device performs a series of processing processes such as modulation and mapping on the original information, and then sends the processed original information to the UE; the UE receives the processed original information, and processes the processed original information using the AI ​​model or AI function to obtain information A, and the UE may also include a second receiver, and the UE also uses the second receiver to process the processed original information to obtain information B. The UE can determine the performance of the AI ​​model or AI function based on the information A and information B. Among them, all functions of the second receiver are not implemented by any AI model or AI function. It can be understood that the second receiver is a traditional receiver. For example, an implementation method of the second receiver can refer to Figure 1. Optionally, the first receiver and the second receiver can be two different hardware circuits; or the two receivers can also be partially the same hardware circuit, but some of the functional modules (software or hardware) therein are different.

[0106] For example, if the UE includes a first receiver and a second receiver, the UE may support a direct detection method or an indirect detection method, or the UE may support both detection methods; or, if the UE includes a first receiver but does not include a second receiver, the UE may support a direct detection method but may not support an indirect detection method.

[0107] Optionally, the way in which the UE performs AI model or AI function detection may also be related to the type of UE, or the network device may determine it through the registration information of the UE, and the second auxiliary information may not indicate the way in which the UE performs AI model or AI function detection.

[0108] Auxiliary information of the test information can be used to determine the test information. The auxiliary information, for example, includes (or indicates) one or more of the following: type information of the AI ​​model or AI function, size of the test information, type information of the test information, information on how the test information is generated, or information on how the test information is sent.

[0109] The type information of the AI ​​model or AI function may indicate the functions supported by the AI ​​model or AI function. For example, refer to Table 1, which is an example of a receiver type. A row in Table 1 represents a type, and Table 1 includes the type of functional modules in a traditional receiver and the type of the AI ​​model or AI function. For example, the type of the AI ​​model or AI function in the embodiment of the present application may be type 1 or type 2 in Table 1, or may be other types in addition to Table 1. In addition, type 1, type 2, etc. in Table 1 are just examples of type names, and the corresponding types may also have other names, for example, they may also be called type A, type B, etc. In addition, for module functions using AI, it may further include conditions supported under the use of AI, thereby enabling the configuration of the corresponding AI function.

[0110] Table 1

[0111] In Table 1, Type 1 is a type of AI model or AI function, corresponding to the channel estimation function in baseband processing. Types 2, 7, or 8 are other types of AI models or AI functions, with Type 2 or Type 7 corresponding to the equalization / demodulation function in baseband processing, and Type 8 corresponding to the channel decoding function in baseband processing. Types 3 through 6 are types of functional modules within a conventional receiver, for example, Type 3 corresponds to the channel decoding function in baseband processing, and so on. "Fused" in Table 1 means that this type of AI model or AI function can implement the function of a functional module within the first receiver, or can also implement the functions of multiple functional modules within the first receiver, that is, the functions of multiple functional modules within the first receiver are considered to be integrated into a single AI model or AI function. "Independent" in Table 1 means that this type of AI model or AI function can implement the corresponding baseband or RF function within the first receiver. For example, the equalization / demodulation function in Type 2 can be integrated with the channel estimation function and implemented through a single AI model or AI function. The input of the AI ​​model or AI function can be the received signal, and the output can be the equalized / demodulated signal. For example, the equalization / demodulation function in type 7 only implements the equalization / demodulation function of AI and still needs to rely on the output result of the previous channel estimation module. The input of the AI ​​model or AI function is the received signal and the estimated channel, and the output is the equalized / demodulated signal.

[0112] For example, the network device can determine the parameters of the test information based on the type information of the AI ​​model or AI function, and the test information can be determined based on the parameters. For example, if the type information of the AI ​​model or AI function is type 2 in Table 1, the network device can determine that the type of the test information is a predefined bit stream (or a sequence), and the type information of the test information can be a parameter of the test information. For another example, if the type information of the AI ​​model or AI function is type 8 in Table 1, the network device can determine that the type of the test information is a transport block (TB), and the bits in the TB can be predefined or non-predefined (such as randomly generated).

[0113] Alternatively, the type information of the AI ​​model or AI function used by the UE may also be predefined by the protocol, or the UE has been reported to the network device in advance (for example, the UE reports during the registration process), or is preconfigured in the network device and the UE, then the auxiliary information of the test information may not indicate the type information of the AI ​​model or AI function.

[0114] The size of the test information can be measured by the data size, for example, the size can be the size of the transmission block TB occupied by the test information or the number of bytes or bits of the TB occupied. Alternatively, the size of the test information can also be the length of the sequence or the number of bits of the sequence. Alternatively, the size of the test information can also be measured by the transmission resources occupied by the test information, for example, the size can be the number of resource elements (RE) occupied by the test information, or the number of resource blocks (RB), or the number of orthogonal frequency division multiplexing (OFDM) symbols, etc. The size of the test information can be a specific value, that is, the size of the test information can be equal to the value; or the size of the test information can also be a value range, that is, the size of the test information can be within the value range. In the process of detecting the AI ​​model or AI function, the network device may send the test information once, or may send the test information multiple times, then the size of the test information can indicate the size of the test information once, or may indicate the total size or average size of multiple test information.

[0115] Alternatively, the size of the test information may be predefined by a protocol, or determined by the network device itself, or preconfigured in the network device and the UE, or determined by negotiation between the UE and the network device, etc., and the auxiliary information of the test information may not indicate the size of the test information.

[0116] Optionally, the amount of test information (i.e., the size of the test information) may be less than or equal to a first threshold. For example, the amount of test information sent by the network device at one time (e.g., the first test information to be introduced later) may be less than or equal to the first threshold; or, the sum of the amount of test information sent by the network device multiple times may be less than or equal to the first threshold. That is, the amount of test information may be small, thereby reducing the complexity of UE's reception and processing. The first threshold is, for example, configured by the network device, or pre-configured in the network device, or predefined by a protocol. For example, when the network device determines the first threshold, it may be determined based on the auxiliary information of the test information in the second auxiliary information, for example, the network device sets the first threshold to be less than or equal to the size of the test information in the auxiliary information of the test information.

[0117] The type information of the test information may indicate the type of the test information, for example, the type of the test information may be a reference signal, a sequence (or a bit stream), data (or a physical downlink shared channel (PDSCH)), a TB, or control information, etc. Alternatively, the type information of the test information may also be predefined by a protocol, or determined by the network device itself, or preconfigured in the network device and the UE, or determined by negotiation between the UE and the network device, etc., then the auxiliary information of the test information may also not indicate the type information of the test information. The reference signal may, for example, include one or more of a channel state information (CSI)-reference signal (RS), a demodulation reference signal (DMRS), a positioning reference signal (PRS), a phase tracking reference signal (PTRS), or a reference signal such as a physical broadcast channel (PBCH) or a synchronization signal block (SSB).

[0118] The test information generation method information may indicate the test information generation method, for example, indicating the protocol layer used to generate the test information, and / or indicating whether the test data is randomly generated or generated according to a first rule. The protocol layer used to generate the test information is, for example, a physical (PHY) layer, a media access control (MAC) layer, or any other protocol layer above the MAC layer, such as a radio link control (RLC) layer. Optionally, if the generation method information indicates that the test data is generated according to a first rule, the second auxiliary information may further include information about the first rule. For example, if the first rule includes one or more formulas, the second auxiliary information may include the one or more formulas, so that the network device can generate the test data according to the first rule. Alternatively, the first rule may also be predefined by a protocol, or preconfigured in the network device, or determined by negotiation between the UE and the network device, etc., in which case the second auxiliary information may not necessarily include information about the first rule.

[0119] Alternatively, the generation method information of the test information may be predefined by a protocol, or determined by the network device itself, or preconfigured in the network device and the UE, or determined by negotiation between the UE and the network device, etc., and the auxiliary information of the test information may not indicate the generation method information of the test information.

[0120] The test information transmission mode information may indicate the transmission mode of the test information, for example, indicating one or more of the following: the transmission resource of the test information, the transmission period or transmission time interval of the test information, the modulation mode of the test information, the coding mode of the test information, the precoding mode of the test information, or the multi-user (MU) pairing mode. The transmission resource of the test information is the resource used to send the test information, which may include time domain resources and / or frequency domain resources. For example, indicating the transmission resource of the test information by the second auxiliary information is equivalent to agreeing on the transmission resource of the test information between the UE and the network device, so that the network device can send the test information through the agreed resource, and the UE can also receive the test information through the agreed resource.

[0121] Alternatively, the information on the sending method of the test information may be predefined by a protocol, or determined by the network device itself, or preconfigured in the network device and the UE, or determined by negotiation between the UE and the network device, etc., and the auxiliary information of the test information may not indicate the information on the sending method of the test information.

[0122] Optionally, the first receiver of the UE may use one AI model or AI function, or may use multiple AI models or AI functions. For example, different functional modules in the first receiver may be replaced with different AI models or AI functions. Alternatively, optionally, the UE may further include a third receiver. In addition to some or all functional modules in the first receiver being replaced by AI models or AI functions, some or all functional modules in the third receiver may also be replaced by AI models or AI functions. The AI ​​model or AI function used to replace the functional module in the first receiver and the AI ​​model or AI function used to replace the functional module in the third receiver may be different AI models or AI functions. Then, the UE can report a set of auxiliary information (such as second auxiliary information) for all AI models or AI functions used by the UE, and the test information determined by the network device based on the second auxiliary information can be used for any AI model or AI function of the UE; or, the UE only reports auxiliary information for one of the AI ​​models or AI functions used by the UE, for example, the second auxiliary information is reported for an AI model or AI function in the first receiver, and the test information determined by the network device based on the second auxiliary information can be used for the AI ​​model or AI function in the first receiver of the UE, but cannot be used for other AI models or AI functions used by the UE; or, the UE can report auxiliary information separately for all AI models or AI functions used by the UE, for example, the second auxiliary information is reported for an AI model or AI function in the first receiver, and other auxiliary information is reported for other AI models or AI functions. The network device can determine the test information separately according to different auxiliary information, and these test information are also applied to the corresponding AI models or AI functions respectively.

[0123] Optionally, whether the UE detects the AI ​​model or AI function, or whether the UE uses test data, may be related to factors such as the UE capability and / or the AI ​​function or the capability of the AI ​​function. For example, for an AI model or AI function with strong generalization, it may not be necessary to perform model detection frequently. In this case, the UE can use ordinary business data to perform model detection, such as using an indirect detection method. Since it is ordinary business data, the network device does not need to generate test information, so the UE does not need to send the second auxiliary information; or, for an AI model or AI function with weaker generalization capability, it may be necessary to perform model detection more frequently. In this case, the UE can send the second auxiliary information to the network device.

[0124] As can be seen from the above introduction, the network device does not need to obtain the second auxiliary information, or the network device can obtain the second auxiliary information through other methods except S401, so S401 is an optional step.

[0125] S402: The network device sends first configuration information to the UE. Correspondingly, the UE receives the first configuration information from the network device.

[0126] The first configuration information may configure parameters of the test information. For example, the test information parameters configured by the first configuration information may include (or indicate) one or more of the following: test information type information, test information size, test information generation method information, test information transmission method information, third indication information, a true value, or true value determination method information.

[0127] The type of test information configured by the first configuration information and the type of test information indicated by the second auxiliary information may be the same or different. For example, if the network device receives the second auxiliary information and configures the parameters of the test information only based on the second auxiliary information, the type of test information configured by the first configuration information and the type of test information indicated by the second auxiliary information may be the same; for another example, if the network device does not receive the second auxiliary information, the type of test information configured by the first configuration information and the type of test information indicated by the second auxiliary information may be the same or different; or, if the network device receives the second auxiliary information but configures the parameters of the test information based on the second auxiliary information and other information (for example, if the network device receives auxiliary information from multiple UEs, and the auxiliary information of each UE indicates the parameters of the test information desired by each UE, the network device may configure the parameters of the test information based on the auxiliary information of the multiple UEs), the type of test information configured by the first configuration information and the type of test information indicated by the second auxiliary information may be the same or different.

[0128] Similarly, the size of the test information configured by the first configuration information and the size of the test information indicated by the second auxiliary information may be the same or different. Please refer to the introduction of the type information of the test information in the previous paragraph.

[0129] Similarly, the test information generation method information configured by the first configuration information and the test information generation method information indicated by the second auxiliary information may be the same or different. Please refer to the introduction of the test information type information in the previous paragraph.

[0130] Similarly, the information on the sending mode of the test information configured by the first configuration information and the information on the sending mode of the test information indicated by the second auxiliary information may be the same or different. Please refer to the introduction of the type information of the test information in the previous text. Optionally, for example, the information on the sending mode of the test information configured by the first configuration information may indicate the transmission resources of the test information. For example, if the transmission resources of the test information are agreed upon between the UE and the network device, the network device may select part or all of the resources from the agreed resources and indicate them to the UE through the first configuration information, so that the network device can send the test information through the resources indicated by the first configuration information, and the UE can also receive the test information through the resources indicated by the first configuration information. The first configuration information may directly indicate the location of the transmission resource, for example, including information on time domain resources and / or frequency domain resources; or, when agreeing on the transmission resource, the UE and the network device may set an index for each transmission resource. The network device may indicate one or more indexes through the first configuration information, thereby indicating one or more transmission resources. This indication method can save the overhead of the first configuration information.

[0131] The third indication information may indicate that the test information is only used to determine the performance of the AI ​​model or AI function. It can also be understood that, through the third indication information, the UE can determine that the test information is only used to determine the information of the AI ​​model or AI function, and not for other purposes. For example, the test information is a reference signal. Reference signals are generally used for channel estimation and other purposes. If a reference signal is test information, the UE can determine that the reference signal is only used to determine the performance of the AI ​​model or AI function, and is not used for channel estimation and other purposes (or the UE is not required or expected to perform channel estimation based on the reference signal). Through the third indication information, the test information can be distinguished from ordinary information (such as service data or reference signals used for channel estimation and other performance), so that the test information does not affect normal services. Alternatively, the test information is only used to determine the performance of the AI ​​model or AI function and is not used for other purposes. This feature can also be predefined by the protocol, or preconfigured in the UE, or determined by negotiation between the UE and the network device. As long as the UE determines that a piece of information is test information, it can be determined that the information is only used to determine the performance of the AI ​​model or AI function. Then the first configuration information may not need to include the third indication information.

[0132] The method for determining the true value can be used to determine the true value, that is, even if the UE does not directly obtain the true value, the UE knows the method for determining the true value and can determine the true value accordingly. The previous article introduced two model detection methods, namely the direct detection method and the indirect detection method. If the UE uses the indirect detection method, the first configuration information does not need to configure the true value, nor does it need to configure the method for determining the true value; or, if the UE uses the direct detection method, the first configuration information can configure the true value, or configure the method for determining the true value, so that the UE can obtain the true value, so that the performance of the AI ​​model or AI function can be determined based on the true value and the processing result of the AI ​​model or AI function on the test information. Among them, the true value and the test information are the same type of information, for example, the true value and the test information can both be data, or both are control information, or both are sequences, or both are reference signals.

[0133] Optionally, if the UE uses a direct detection method, the true value can be determined by the second receiver. Alternatively, it can be understood that the method for determining the true value can be a method for determining it by the second receiver. For example, the UE uses the second receiver to process the first information and can obtain the processing result of the second receiver. If the second receiver processes the first information correctly, the UE can assume that the second receiver has determined the true value. For example, the true value is the processing result of the second receiver, or the true value is determined based on the processing result of the second receiver.

[0134] The first configuration information can be sent via unicast, for example, included in a unicast message; or, the first configuration information can be sent via multicast, for example, included in a multicast message; or, the first configuration information can be sent via broadcast, for example, included in a broadcast message. For example, if the network device determines the first configuration information for a certain UE, and the test information configured in the first configuration information is used to determine the performance of the AI ​​model or AI function of the UE, then the first configuration information can be sent to the UE via unicast. Alternatively, if the network device determines the first configuration information for a group of UEs, and the test information configured in the first configuration information can be used to determine the performance of the AI ​​model or AI function of each UE in the group, then the first configuration information can be sent to the group of UEs via multicast. Alternatively, if the network device determines the first configuration information for multiple UEs (there may not be a group relationship between the multiple UEs, for example, the multiple UEs are part or all of the UEs covered by the network device), and the test information configured in the first configuration information can be used to determine the performance of the AI ​​model or AI function of the multiple UEs, then the first configuration information can be sent via broadcast.

[0135] If the first configuration information is included in a broadcast message, the first configuration information is optionally included in a message of an RRC layer, for example. For example, the first configuration information is included in a system information block (SIB), such as an SIB message such as SIB1 or SIB2; for another example, the broadcast message is included in a common RRC configuration such as PDSCH-common-configuration in an RRC message.

[0136] Alternatively, the parameters of the test information may be predefined by a protocol, or preconfigured in the network device and the UE, or determined by negotiation between the UE and the network device, etc. Then S402 may not be necessary, and is therefore an optional step.

[0137] S403: The network device sends first information to the UE. Correspondingly, the UE receives the first information from the network device.

[0138] The first information may be test information, for example, the first information is test information sent once by the network device. After entering the model detection phase, the network device may send test information, and the UE may receive test information from the network device. Specifically, for the UE, after sending the second auxiliary information in S401 to the network device, it may be considered that the model detection phase has been entered, or the test information has been started to be received; for the network device, after receiving the second auxiliary information in S401, it may be considered that the model detection phase has been entered, or the test information has been started to be sent. Alternatively, for the UE, after receiving the first configuration information in S402, it may be considered that the model detection phase has been entered, or the test information has been started to be received; for the network device, after sending the first configuration information in S402 to the UE, it may be considered that the model detection phase has been entered, or the test information has been started to be sent. Alternatively, the UE and the network device may also determine when to enter the model detection phase, or when to start sending and receiving test information, by other means.

[0139] For example, the UE may send a first request message to the network device, and the first request message may request to determine the performance of the AI ​​model or AI function. If the UE uses only one AI model or AI function, the first request message may request to determine the performance of the AI ​​model or AI function; or if the UE uses multiple AI models or AI functions, the first request message may request to determine the performance of some or all of the AI ​​models or AI functions. For the UE, sending the first request message can be considered to have entered the model detection phase, or can start receiving test information. For the network device, receiving the first request message can be considered to have entered the model detection phase, or can start sending test information. Optionally, the first request message may be included in a MAC control element (CE), or included in uplink control information (UCI), or included in a radio resource control (RRC) message, or can also be included in other types of messages. The RRC message is, for example, a UE assistance information message, or can also be a message of other RRC layers.

[0140] Optionally, the UE may send a first request message to the network device when a first condition is met, and the first condition may be a model detection condition. The first condition may be determined by the UE itself, or sent by the network device to the first UE, or predefined through a protocol, or preconfigured in the UE, or determined by negotiation between the UE and the network device, etc. For example, the first condition includes one or more of the following: a reception error rate using the first receiver within a first time period is greater than or equal to a second threshold, a reception error rate using the first receiver within the first time period is greater than or equal to the second threshold and the signal quality is greater than a third threshold, the UE switches scenarios, or a detection period is reached.

[0141] The reception error rate of the first receiver is, for example, an error rate of a final decoding result obtained using the first receiver, or a demodulation error rate obtained using the first receiver. The second threshold is, for example, configured by the network device, or preconfigured in the UE, or determined by negotiation between the UE and the network device, or predefined by a protocol.

[0142] For example, if the reception error rate of the first receiver used in the first time period is greater than or equal to the second threshold, the UE can test the performance of the AI ​​model or AI function. If the performance of the AI ​​model or AI function is poor, the UE can switch to using other AI models or AI functions, or use a second receiver for reception to improve the reception accuracy. Alternatively, if the reception error rate of the first receiver used in the first time period is greater than or equal to the second threshold, and the signal quality is greater than a third threshold, it is likely that there is a problem with the first receiver. Therefore, the UE can test the performance of the AI ​​model or AI function. If the performance of the AI ​​model or AI function is poor, the UE can switch to using other AI models or AI functions, or use a second receiver for reception to improve the reception accuracy. The third threshold is, for example, configured by the network device, or pre-configured in the UE, or determined by negotiation between the UE and the network device, or predefined by a protocol, etc.

[0143] Alternatively, the UE switches the scenario, for example, the UE moves from indoors to outdoors. At this time, the performance of the AI ​​receiver (such as the first receiver) may not be guaranteed. The UE can test the performance of the AI ​​model or AI function. If the performance of the AI ​​model or AI function is poor, the UE can switch to using other AI models or AI functions, or use a second receiver for reception to improve the reception accuracy in the new scenario.

[0144] Alternatively, the UE is configured with periodic detection, requiring the performance of the AI ​​model or AI function to be detected periodically. When the detection period arrives, the UE can trigger the model detection process. This detection period can be configured by the network device, pre-configured in the UE and network device, determined by negotiation between the UE and network device, or predefined by a protocol.

[0145] Optionally, the first request information may also include first auxiliary information, and the first auxiliary information may include auxiliary information of the test information. It can also be understood that the network device can determine the test information based on the first auxiliary information. The content included in the first auxiliary information may be similar to the content included in the second auxiliary information described above. For example, the first auxiliary information may indicate (or include) one or more of the following: whether the UE supports the AI ​​model or AI function, how the UE performs AI model or AI function detection, or auxiliary information of the test information. The values ​​of the specific parameters included in the first auxiliary information may be the same as or different from the values ​​of the specific parameters included in the second auxiliary information. For example, both the first auxiliary information and the second auxiliary information include auxiliary information of the test information, and the auxiliary information of the test information included in the first auxiliary information and the second auxiliary information both include information on how the test information is generated. The information on how the test information is generated included in the first auxiliary information may indicate that the test information is generated through the physical layer, and the information on how the test information is generated included in the second auxiliary information may also indicate that the test information is generated through the physical layer; or, the information on how the test information is generated included in the first auxiliary information may indicate that the test information is generated through the MAC layer, and the information on how the test information is generated included in the second auxiliary information may indicate that the test information is generated through the physical layer.

[0146] The first auxiliary information and the second auxiliary information may be independent of each other. For example, the UE may report the first auxiliary information instead of the second auxiliary information; or the UE may report the second auxiliary information instead of the first auxiliary information; or the UE may report the second auxiliary information but a period of time has passed between the time the UE reports the second auxiliary information and the time it expects to perform model checking. During this period, some of the UE's expectations may have changed. Therefore, the UE may indicate its latest expectations for test information through the first auxiliary information.

[0147] Alternatively, the first assistance information may be used to modify or supplement the second assistance information. For example, although the UE reports the second assistance information, when it expects to perform model checking, there is a period of time between the reporting of the second assistance information and the time when some expectations of the UE may change. Therefore, the UE may modify the values ​​of some or all parameters in the second assistance information through the first assistance information, or supplement the content not included in the second assistance information through the first assistance information.

[0148] If the network device receives first auxiliary information from the UE after having already sent the first configuration information, and the first auxiliary information is different from the second auxiliary information, the network device may optionally reconfigure the parameters of the test information based on the first auxiliary information, obtain the second configuration information, and send the second configuration information. The UE may then use the second configuration information to receive the test information. The parameters included in the second configuration information are similar to those included in the first configuration information, except that the values ​​may be different. The second configuration information is sent in a similar manner to the first configuration information.

[0149] Optionally, if the first request information is included in a MAC CE or RRC message, the first request information may include the first auxiliary information; alternatively, if the first request information is included in UCI, the first request information may not include the first auxiliary information. This is because the capacity of a MAC CE or RRC message is relatively large and can carry more information, whereas the capacity of UCI is relatively small and may not be able to carry much content. Alternatively, whether the first request information includes the first auxiliary information may not be related to the signaling type but may be related to the UE's expectations.

[0150] Alternatively, the UE may not need to send the first request information to the network device, but the network device may send the second indication information to the UE, and the second indication information may indicate the determination of the performance of the AI ​​model or AI function. If the UE uses only one AI model or AI function, the second indication information may indicate the determination of the performance of the AI ​​model or AI function; or if the UE uses multiple AI models or AI functions, the second indication information may indicate the determination of the performance of some or all of the AI ​​models or AI functions. For the UE, upon receiving the second indication information, it can be considered that the model detection phase has been entered, or the test information can be started to be received. For the network device, upon sending the second indication information, it can be considered that the model detection phase has been entered, or the test information can be started to be sent. Optionally, the second indication information may be included in the MAC CE, or included in the downlink control information (DCI), or included in the RRC message, or may also be included in other types of messages.

[0151] Alternatively, the UE and the network device do not need to send information for indicating the start of determining the performance of the AI ​​model or AI function, that is, the UE and the network device do not need to transmit information for starting the model detection phase, but enter the model detection phase through other means. For example, when the second condition is met, both the UE and the network device can determine to enter the model detection phase, or determine to start sending and receiving test information. The second condition can be pre-configured in the UE and the network device, or pre-configured by the network device, or determined by negotiation between the UE and the network device, or pre-defined by a protocol, etc. The second condition may be the same as or different from the aforementioned first condition.

[0152] The first information may be generated by different protocol layers. For example, if the first information is generated by the physical layer, the first information may be, for example, a PDSCH, or the first information may be carried in the PDSCH. Alternatively, if the first information is generated by the MAC layer, the first information may be, for example, a MAC protocol data unit (PDU), or the first information may be included in a MAC PDU. Alternatively, the first information may be generated by a higher protocol layer, in which case the implementation of the first information may be different accordingly.

[0153] In addition, the first information can be sent in a unicast manner, for example, included in a unicast message; or, the first information can be sent in a multicast manner, for example, included in a multicast message; or, the first information can be sent in a broadcast manner, for example, included in a broadcast message. For example, if the first information is used by a certain UE to determine the performance of the AI ​​model or AI function used by the UE, the first information can be sent to the UE in a unicast manner. For another example, the first information can be used by a group of UEs to determine the performance of the AI ​​model or AI function used by each UE in the group of UEs, and the first information can be sent to the group of UEs in a multicast manner. For another example, the first information can be used for the performance of the AI ​​model or AI function used by each UE in multiple UEs (there does not necessarily exist a group relationship between these multiple UEs, for example, these multiple UEs are all or part of the UEs covered by the network device), and the first information can be sent in a broadcast manner.

[0154] Optionally, before S403, the method may further include S404, where the network device may send DCI to the UE, and correspondingly, the UE receives the DCI from the network device. The DCI may schedule the first information, and further, the DCI may indicate that the first information is test information.

[0155] As an example of the DCI indicating that the first information is test information, the DCI includes first indication information, which may indicate that the first information is test information, or indicate that the first information is used to determine the performance of an AI model or AI function. This method can be considered an explicit indication method. Optionally, if the UE uses multiple AI models or AI functions, the DCI may indicate which AI model or AI functions the first information is specifically used for. The first indication information is, for example, carried in a first field, which may be a newly added field in the DCI or a reserved field in the DCI. The first field may occupy one or more bits. Taking the first field occupying 1 bit as an example, if the value of the first field is "0", it indicates that the first information is not test information, or indicates that the first information is not used to determine the performance of the AI ​​model or AI function; alternatively, if the value of the first field is "1", it indicates that the first information is test information, or indicates that the first information is used to determine the performance of the AI ​​model or AI function.

[0156] As another example of the DCI indicating that the first information is test information, the DCI is scrambled by a first radio network temporary indentifier (RNTI), and the first RNTI may indicate that the first information is test information, or indicate that the first information is used to determine the performance of the AI ​​model or AI function. This method can be considered as an implicit indication method. Optionally, if the UE uses multiple AI models or AI functions, different AI models or AI functions may be indicated by different RNTIs, or some of the AI ​​models or AI functions may also be indicated by one RNTI. In an embodiment of the present application, the correspondence between the AI ​​model or AI function and the RNTI may be configured by the network device (for example, the correspondence is included in the first configuration information), or pre-configured in the network device and the UE, or determined by negotiation between the UE and the network device, or predefined by a protocol, etc.

[0157] As another example of the DCI indicating that the first information is test information, the DCI indicates whether the first information is test information by the amount of information scheduled. For example, if the amount of information scheduled by the DCI is less than a fourth threshold, it indicates that the DCI is scheduling test information; alternatively, if the amount of information scheduled by the DCI is greater than or equal to the fourth threshold, it indicates that the DCI is not scheduling test information. This method can be considered an implicit indication method. If the UE sends second auxiliary information to the network device, and the second auxiliary information indicates the size of the test information, then the fourth threshold is, for example, less than or equal to the size of the test information indicated by the second auxiliary information; alternatively, if the UE sends first auxiliary information to the network device, and the first auxiliary information indicates the size of the test information, then the fourth threshold is, for example, less than or equal to the size of the test information indicated by the first auxiliary information. For example, if the test information is generated by the MAC layer or by another protocol layer above the MAC layer, the DCI may use this indication method, or may use either of the two aforementioned indication methods; alternatively, if the test information is generated by the physical layer, then either of the two aforementioned indication methods may be used.

[0158] Optionally, the DCI may also indicate the transmission resources of the first information. For example, the DCI may indicate the location of the transmission resources of the first information, such as indicating the time domain resources and / or frequency domain resources of the first information. Alternatively, the DCI may include an index of the transmission resources of the first information. For example, if the UE and the network device have agreed on the transmission resources of the test information and set indexes for the agreed resources respectively, the DCI may include the corresponding index, so that the UE can determine which resources the transmission resources of the first information actually include. Alternatively, the DCI may indicate the offset of the transmission resources of the first information relative to the transmission resources of the second information, where the second information is the information scheduled by the DCI. That is, the DCI may schedule the first information and the second information, and the second information is ordinary business data or control information, etc., which is equivalent to the DCI scheduling ordinary business data or control information, and scheduling test information. Then the DCI may indicate the transmission resources of the second information. Then the DCI may indicate the offset of the transmission resources of the first information relative to the transmission resources of the second information, thereby indicating the transmission resources of the first information. The offset of the transmission resources of the first information relative to the transmission resources of the second information, for example, includes the offset of the time domain resources of the first information relative to the time domain resources of the second information, and / or includes the offset of the frequency domain resources of the first information relative to the frequency domain resources of the second information. The offset of the time domain resources of the first information relative to the time domain resources of the second information, for example, is the offset of the starting time domain resources of the first information relative to the starting time domain resources of the second information, or is the offset of the ending time domain resources of the first information relative to the ending time domain resources of the second information; the offset of the frequency domain resources of the first information relative to the frequency domain resources of the second information, for example, is the offset of the starting frequency domain resources of the first information relative to the starting frequency domain resources of the second information, or is the offset of the ending frequency domain resources of the first information relative to the ending frequency domain resources of the second information. If the DCI schedules the first information and the second information, then when sending the second information, the network device may perform rate matching on the transmission resources of the first information, bypassing the transmission resources of the first information to map the second information. Optionally, the network device may independently encode the first information and the second information, for example, independently perform channel coding, to reduce the impact of the test information on the service.

[0159] Please refer to Figure 5, which is an example of the transmission resources of the first information and the transmission resources of the second information. The horizontal axis of Figure 5 represents time and the vertical axis represents frequency. For example, the " / " part in Figure 5 represents PDSCH 1, and PDSCH1 represents the first information; the horizontal line part in Figure 5 represents PDSCH2, and PDSCH 2 represents the second information; in addition, the physical downlink control channel (PDCCH) shown in Figure 5 may include DCI. The DCI may indicate the offset between the starting time domain position of PDSCH 1 and the starting time domain position of PDSCH 2 (such as offset A shown in Figure 5), for example, the offset A is one or more OFDM symbols, or one or more time slots; in addition, the DCI may also indicate the offset between the starting frequency domain position of PDSCH 1 and the starting frequency domain position of PDSCH 2 (such as offset B shown in Figure 5), for example, the offset B is one or more subcarriers.

[0160] Alternatively, the DCI may not necessarily indicate the transmission resources of the first information. For example, the DCI also schedules the second information, and the DCI indicates the transmission resources of the second information, and the offset of the transmission resources of the first information relative to the transmission resources of the second information may be pre-configured by the network device, or pre-configured in the UE and the network device, or pre-defined by the protocol, etc., then the DCI may not necessarily indicate the transmission resources of the first information, and the UE can determine the transmission resources of the first information based on the transmission resources of the second information and the predicted offset. For another example, the transmission resources of the first information are pre-configured by the network device or pre-defined by the protocol, for example, the network device pre-configures or the protocol pre-defines a transmission resource, or pre-configures or pre-defines a transmission resource and a pre-configured or pre-defined resource period, and this transmission resource is, for example, a transmission resource within the first resource period, then the UE can determine the transmission resources of the first information based on the information pre-configured or pre-defined by the network device, without the need for DCI indication.

[0161] Taking the example of a DCI including first indication information, optionally, if the first indication information indicates that the first information is test information, the DCI may also indicate the transmission resources for the first information; alternatively, if the first information indicates that the first information is not test information, the DCI may not necessarily indicate the transmission resources for the first information. For example, the DCI may carry the first indication information via the first field and indicate the transmission resources for the first information via the second field. The second field may be a newly added field in the DCI or a reserved field in the DCI. Refer to Figures 6A and 6B for two examples of such DCI. In Figure 6A, the value of the first indication information is "1," indicating that the first information is test information. In this case, the second field is not empty, indicating the transmission resources for the first information. The second field may occupy one or more bits. Figure 6A uses an example where the second field occupies three bits and the value of the second field is "001," but this is not limiting. In Figure 6B, the value of the first indication information is "0," indicating that the first information is not test information. In this case, the second field may be empty (or the DCI may not include the second field). In this case, the second field does not indicate the transmission resources for the first information.

[0162] For example, if the first information is data, a reference signal or a sequence, the network device may send the DCI to the UE; or, if the first information is control information, the control information may be the DCI, that is, the first information itself may indicate that the first information is test information.

[0163] For another example, if the first information is physical layer information, the network device may send the DCI to the UE, and the DCI may indicate whether the first information is test information. Alternatively, if the first information is MAC layer information, the network device may send the DCI to the UE, and the DCI may indicate whether the first information is test information, or the DCI may not indicate whether the first information is test information, or the network device may not send the DCI to the UE. If the first information is MAC layer information, the first information is, for example, included in a MAC PDU, or the first information is a MAC PDU. If the first information is MAC layer information, the network device processes the first information at the physical layer in the same manner as it processes ordinary information. However, if the network device wants to send the first information and ordinary information at the same time (for example, sending the first information and the second information at the same time, the second information being ordinary data or control information, etc.), the first information and the second information may be included in different MAC PDUs, respectively. Optionally, the MAC PDU including the first information may indicate whether the first information is test information. For example, the MAC header of the MAC PDU may indicate that the first information is test information, or that the first information is information used to determine the performance of an AI model or AI function. For example, the network device may add fourth indication information to the MAC header to indicate that the first information is test information, or that the first information is information used to determine the performance of an AI model or AI function. The fourth indication information, for example, occupies one or more bits, and the fourth indication information may be indicated in a manner similar to the first indication information, and will not be further described.

[0164] S405: The UE uses an AI model or an AI function (or uses a first receiver) to process the first information.

[0165] S406. The UE determines the performance of the AI ​​model or AI function based on the processing result of the first information.

[0166] S405 and S406 are introduced together as follows.

[0167] As previously mentioned, test information may be generated by different protocol layers, and the UE may process test information generated by different protocol layers differently, as illustrated below. If the test information is data or control information, the network device processes it similarly to ordinary data or control information. For example, the network device may also perform channel coding, modulation, and mapping on the test information. Similarly, if the test information is a reference signal or sequence, the network device processes it similarly to ordinary reference signals or sequences.

[0168] 1. The test information is physical layer information, or the test information is information generated by the physical layer of the network device. For example, the test information is included in the PDSCH, or the first information is the PDSCH.

[0169] This description will be made using the first information in the test information as an example. If the UE determines that the first information is test information, the UE may process the first information using the AI ​​model or AI function to be tested to obtain a processing result. The AI ​​model or AI function may replace one or more functional modules in the first receiver. Depending on the functional modules replaced, the "processing" referred to herein may also vary. Taking the structure shown in Figure 1 as an example when the functional modules in the first receiver are not replaced by the AI ​​model or AI function, for example, the functional modules replaced by the AI ​​model or AI function include the analog-to-digital conversion module in the RF processing in Figure 1, but do not include the modules after the analog-to-digital conversion module. In this case, the processing here refers to analog-to-digital conversion, and the processing result obtained refers to the analog-to-digital conversion result. For another example, the functional modules replaced by the AI ​​model or AI function include the channel estimation module in the baseband processing in Figure 1, but do not include the modules after the channel estimation module (such as the equalization / demodulation module, etc.). In this case, the processing here refers to channel estimation, and the processing result obtained refers to the channel estimation result. For another example, the functional modules replaced by the AI ​​model or AI function include the equalization / demodulation module in the baseband processing in Figure 1, but do not include the modules after the equalization / demodulation module (such as the channel decoding module, etc.). In this case, the processing here refers to demodulation, and the processing result obtained refers to the demodulation result. For another example, the functional modules replaced by the AI ​​model or AI function include the channel decoding module in the baseband processing in Figure 1. In this case, the processing here refers to channel decoding, and the processing result obtained refers to the channel decoding result.

[0170] If the UE uses a direct detection method, after obtaining the processing result, the UE can determine the performance of the AI ​​model or AI function based on the processing result and the true value (or label). For example, the functional module replaced by the AI ​​model or AI function is the equalization / demodulation module in the baseband processing of the first receiver, and the module after the equalization / demodulation module (such as the channel decoding module, etc.) is not replaced. Then the true value can be a data block (or bit stream) that has been channel-coded but not modulated, and the processing result can be a data block (or bit stream) that has been equalized / demodulated by the AI ​​model or AI function but not channel-decoded. When the direct detection method is adopted, the UE can compare whether each bit in the processed data block is the same as the corresponding bit in the true value, so as to count the number of erroneous bits. Through one or more direct detections, the bit error rate of direct detection can be further counted. If the bit error rate is higher than the threshold A (the threshold A can be determined by the UE itself, or configured by the network device, or predefined by the protocol, or determined by negotiation between the UE and the network device, etc.), it is considered that the performance of the AI ​​model or AI function does not meet the requirements.

[0171] Alternatively, if the UE uses an indirect detection method, in addition to using the AI ​​model or AI function (or the first receiver) to process the first information, the UE also uses the second receiver to process the first information, then the UE can obtain the processing result of the first receiver (or the AI ​​model or AI function), and obtain the processing result of the second receiver. The UE can determine the performance of the AI ​​model or AI function based on these two processing results. For example, in each indirect detection, the UE counts whether the first information processed by the first receiver and the first information processed by the second receiver can pass the cyclic redundancy check (CRC), for example, the number of CRC check errors, that is, the number of block errors. After one or more indirect detections, the block error rates after processing by the two receivers can be further counted separately. If the block error rate after processing by the first receiver is greater than the fifth threshold, and the block error rate after processing by the second receiver is less than the sixth threshold, it is considered that the performance of the AI ​​model or AI function does not meet the requirements, where the fifth threshold and the sixth threshold may be the same or different.

[0172] If the test information is physical layer information, and the test information is no longer used for other purposes except for determining the performance of the AI ​​model or AI function, then optionally, the UE may not need to send hybrid automatic repeat request (HARQ) feedback information corresponding to the test information, and the HARQ feedback information may include, for example, a positive acknowledgement (ACK) or a negative acknowledgement (NACK). For example, if the first information is test information, the UE may not need to send the HARQ feedback information corresponding to the first information. Because the test information is not used for other purposes, it only has the function of determining the performance of the AI ​​model or AI function for the UE, and may not have more functions for the network device. Therefore, the network device may not pay attention to whether the UE successfully receives the test information. Therefore, the UE may not need to send the HARQ feedback information corresponding to the test information, thereby reducing signaling overhead. Optionally, the physical layer may not submit the processing result of the test information to the MAC layer of the UE, and / or the physical layer may discard the processing result of the test information.

[0173] 2. The test information is MAC layer information, or the test information is information generated by the MAC layer of the network device. For example, the test information is included in a MAC PDU, or the test information is a MAC PDU.

[0174] Let's take the first information in the test information as an example. As mentioned above, for the test information of the MAC layer, the network device can indicate whether the first information is test information through DCI and through the MAC header; or the network device can indicate whether the first information is test information through DCI without indicating whether the first information is test information through the MAC header; or the network device can indicate whether the first information is test information through the MAC header without indicating whether the first information is test information through DCI. Due to different indication methods, the UE's processing process may also be different, which are introduced below.

[0175] (1) The network device indicates whether the first information is test information through the DCI and indicates whether the first information is test information through the MAC header.

[0176] The physical layer of the UE first receives the first information, and according to the indication of the DCI, the physical layer can determine whether the first information is test information. If the UE uses a direct detection method and the DCI indicates that the first information is test information, the physical layer can use the AI ​​model or AI function (or, use the first receiver) to process the first information to obtain a processing result, and the physical layer can submit the processing result to the MAC layer of the UE, or the physical layer of the UE can submit the processing result and the true value (or, label) to the MAC layer. The MAC layer can determine whether the first information is test information based on the MAC header of the processing result. If it is determined that the first information is test information, the MAC layer can determine the performance of the AI ​​model or AI function based on the processing result and the true value (or, label). Optionally, since the first information is test information, the MAC layer may no longer submit the processing result of the first information to the upper layer of the MAC layer, and / or the MAC layer may discard the processing result of the first information.

[0177] Alternatively, if the UE uses an indirect detection method and the DCI indicates that the first information is test information, the physical layer may use an AI model or AI function (or, use a first receiver) to process the first information to obtain a processing result A, and the physical layer may use a second receiver to process the first information to obtain a processing result B. The physical layer may submit the processing result A and the processing result B to the MAC layer of the UE. The MAC layer may determine whether the first information is test information based on the MAC header of the processing result A. For example, if it is determined that the first information is test information, the MAC layer may determine the performance of the AI ​​model or AI function based on the processing result A and the processing result B. Optionally, since the first information is test information, the MAC layer may no longer submit the two processing results of the first information to the upper layer of the MAC layer, and / or the MAC layer may discard the processing result A and the processing result B of the first information.

[0178] Alternatively, if the DCI indicates that the first information is not test information, the physical layer may use an AI model or AI function (or, use a first receiver) to process the first information to obtain a processing result A, or use a second receiver to process the first information to obtain a processing result B. The physical layer may deliver the processing result A or the processing result B to the MAC layer of the UE. The MAC layer may determine whether the first information is test information based on the MAC header of the processing result A. For example, if it is determined that the first information is not test information, the MAC layer may continue to process the first information according to the processing result A or the processing result B, and deliver the processing result obtained by the MAC layer to the upper layer of the MAC layer, such as the RLC layer.

[0179] (2) The network device indicates whether the first information is test information through the DCI, and does not indicate whether the first information is test information through the MAC header.

[0180] The physical layer of the UE first receives the first information, and according to the indication of the DCI, the physical layer can determine whether the first information is test information. If the UE uses a direct detection method, and the DCI indicates that the first information is test information, the physical layer can use an AI model or an AI function (or, use a first receiver) to process the first information to obtain a processing result, and the physical layer can determine the performance of the AI ​​model or AI function based on the processing result and the true value (or, label). Since the MAC header does not indicate whether the first information is test information, that is, the MAC layer is not required to determine whether the first information is test information, the physical layer may optionally not submit the processing result to the MAC layer, and / or the physical layer may discard the processing result.

[0181] Alternatively, if the UE uses an indirect detection method and the DCI indicates that the first information is test information, the physical layer may use an AI model or AI function (or use a first receiver) to process the first information to obtain a processing result A, and the physical layer may use a second receiver to process the first information to obtain a processing result B. The physical layer may determine the performance of the AI ​​model or AI function based on the processing result A and the processing result B. Since the MAC header does not indicate whether the first information is test information, that is, the MAC layer is not required to determine whether the first information is test information, the physical layer may optionally not submit the processing result A and the processing result B to the MAC layer, and / or the physical layer may discard the processing result A and the processing result B.

[0182] Alternatively, if the DCI indicates that the first information is not test information, the physical layer may use the AI ​​model or AI function (or, use the first receiver) to process the first information to obtain a processing result A, or use the second receiver to process the first information to obtain a processing result B. The physical layer may deliver the processing result A or the processing result B to the MAC layer of the UE. The MAC layer may continue to process the first information according to the processing result A or the processing result B, and deliver the processing result obtained by the MAC layer to the upper layer of the MAC layer, such as the RLC layer.

[0183] (3) The network device indicates whether the first information is test information through the MAC header, and does not indicate whether the first information is test information through the DCI.

[0184] The UE's physical layer first receives the first information. The physical layer cannot determine whether the first information is test information, so the physical layer processes the first information in the same manner as it processes ordinary information. For example, the physical layer first processes the first information using an AI model or AI function (or, using the first receiver). If the processing is successful, the physical layer may submit the processing result to the UE's MAC layer. The MAC layer may determine whether the first information is test information based on the MAC header of the first information. If the first information is test information, the MAC layer may use the processing result to determine the performance of the AI ​​model or AI function. For example, if the first information is test information and the UE uses direct detection, the MAC layer may determine the performance of the AI ​​model or AI function based on the processing result and the true value (or, label). Alternatively, if the first information is test information and the UE uses indirect detection, then although the MAC layer has not received the processing result of the first information by the second receiver, it does have the processing result of the AI ​​model or AI function on the first information, and the MAC layer can also determine the performance of the AI ​​model or AI function based on this processing result. Alternatively, if the MAC layer determines that the first information is not test information, the MAC layer may continue to process the first information based on the processing result, and submit the processing result obtained by the MAC layer to an upper layer of the MAC layer, such as the RLC layer. If the first information is test information, the MAC layer may optionally no longer submit the processing result of the first information to an upper layer of the MAC layer, and / or the MAC layer may discard the processing result of the first information.

[0185] Alternatively, the physical layer first processes the first information using an AI model or AI function (or using a first receiver). If the processing fails, the physical layer may then use a second receiver to process the first information. The physical layer may deliver the processing result of the second receiver (referred to as processing result B) to the MAC layer of the UE. Optionally, the processing result of the AI ​​model or AI function (for example, referred to as processing result A, which is a failure or an error) may also be delivered to the MAC layer of the UE, and the MAC layer may receive two processing results. The MAC layer may determine whether the first information is test information based on the MAC header of the first information. If the first information is test information, the MAC layer may use the processing result to determine the performance of the AI ​​model or AI function. For example, if the first information is test information and the UE uses a direct detection method, the MAC layer may determine the performance of the AI ​​model or AI function based on the processing result A and the true value (or, label); or, if the first information is test information and the UE uses an indirect detection method, the MAC layer may determine the performance of the AI ​​model or AI function based on the processing result A and the processing result B. Alternatively, if the MAC layer determines that the first information is not test information, the MAC layer may continue to process the first information based on the processing result A or the processing result B, and submit the processing result obtained by the MAC layer to the upper layer of the MAC layer, such as the RLC layer. Wherein, if the first information is test information, optionally, the MAC layer may no longer submit the processing result A and the processing result B of the first information to the upper layer of the MAC layer, and / or, the MAC layer may discard the processing result A and the processing result B of the first information.

[0186] Alternatively, after receiving the first information, the physical layer may process the first information using an AI model or an AI function (or using a first receiver) to obtain a processing result A, and process the first information using a second receiver to obtain a processing result B. The physical layer may deliver the processing result A and the processing result B to the MAC layer of the UE, and the MAC layer may receive two processing results. The MAC layer may determine whether the first information is test information based on the MAC header of the first information. If the first information is test information, the MAC layer may use the processing result to determine the performance of the AI ​​model or AI function. For example, if the first information is test information and the UE uses a direct detection method, the MAC layer may determine the performance of the AI ​​model or AI function based on the processing result A and the true value (or, label); or, if the first information is test information and the UE uses an indirect detection method, the MAC layer may determine the performance of the AI ​​model or AI function based on the processing result A and the processing result B. Alternatively, if the MAC layer determines that the first information is not test information, the MAC layer may continue to process the first information according to the processing result A or the processing result B, and submit the processing result obtained by the MAC layer to an upper layer of the MAC layer, such as the RLC layer. If the first information is test information, the MAC layer may optionally no longer submit the processing result of the first information (e.g., processing result A or processing result B) to an upper layer of the MAC layer, and / or the MAC layer may discard the processing result of the first information.

[0187] The above describes some processing methods of the UE for the first information, which are not limited to the above processing methods in actual applications. For example, the test information may also be generated by the RLC layer or a higher protocol layer, and the UE processing method may also be different, which is not specifically limited.

[0188] For example, for a pre-monitoring (or, pre-detection) scenario, the test information can be MAC layer information or higher layer information. The pre-monitoring scenario means that the UE is currently using the second receiver, and the UE may be about to switch to using the first receiver, or the UE expects to determine the performance of the first receiver. In the process of using the second receiver, the UE can use the test information to detect the performance of the first receiver (or, detect the AI ​​model or AI function).

[0189] For another example, for a non-pre-monitoring (or, non-pre-detection) scenario, the test information can be physical layer information. For example, in a non-pre-monitoring scenario, the UE is currently using a first receiver (or, using an AI model or AI function). During the use of the first receiver, the UE can use the test information to detect the performance of the first receiver (or, detect the AI ​​model or AI function). For example, if it is determined that the performance of the first receiver is poor, the UE may consider adjusting the performance of the first receiver or switching the receiver to improve the reception performance.

[0190] Optionally, the network device may send test information within a first duration, and at the end of the first duration, the network device may stop sending test information. Alternatively, if the UE determines the performance of the AI ​​model or AI function, or the UE believes that it is no longer necessary to determine the performance of the AI ​​model or AI function, the UE may also send a stop indication to the network device; upon receiving the stop indication, the network device may stop sending test information. For example, although the first duration has not ended, if the network device receives the stop indication, the network device may stop sending test information, thereby saving the transmission overhead of the test information.

[0191] In the foregoing introduction, the embodiment of the present application sets test information specifically for determining the performance of the AI ​​model or AI function, which can not only determine the performance of the AI ​​model or AI function, but also reduce the impact on other services. Alternatively, the embodiment of the present application may not set test information specifically for determining the performance of the AI ​​model or AI function, for example, using ordinary information (such as business data, control information, or reference signals or sequences of other functions, etc.) to determine the performance of the AI ​​model or AI function, for example, the aforementioned first information can be ordinary information. If this method is adopted, it is not necessary for the UE and the network device to perform steps such as determining the test information (or determining the configuration of the test information), for example, it is not necessary to perform steps such as S401, S402 and S404, which can reduce the interaction process between the UE and the network device, save signaling overhead, and improve the efficiency of model detection.

[0192] After receiving the first information, the UE in the embodiment of the present application can use the AI ​​model or AI function to process the first information, so that the performance of the AI ​​model or AI function can be determined based on the processing result. That is, for the AI ​​model or AI function used to implement the receiving function, the embodiment of the present application provides a method for determining its performance. Optionally, if it is determined that the performance of the AI ​​model or AI function is poor, corresponding measures can be taken, such as adjusting the performance of the AI ​​model or AI function, shutting down the AI ​​model or AI function, reselecting or retraining the AI ​​model or AI function, or falling back to a non-AI receiving function, thereby improving the receiving performance. Optionally, if it is determined that the performance of the AI ​​model or AI function is poor, the measures taken can be determined by the UE itself; or, after determining the performance of the AI ​​model or AI function, the UE can also report the performance of the AI ​​model or AI function to the network device, and the network device can then instruct the UE which measures to perform. The test information in the embodiment of the present application can be generated by different protocol layers, which is more flexible. In addition, the test information in the embodiments of the present application can be sent in a unicast manner, or in a multicast or broadcast manner, so that the model detection process can be performed in a single UE or in parallel in multiple UEs, which is more flexible. Moreover, sending the test information in a multicast or broadcast manner is beneficial to saving signaling overhead compared to a unicast manner.

[0193] FIG7 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application. The communication device 700 may be the UE or the circuit system of the UE described in the embodiment shown in FIG4 , and is used to implement the method corresponding to the UE in the above method embodiment. Alternatively, the communication device 700 may be the network device or the circuit system of the network device described in the embodiment shown in FIG4 , and is used to implement the method corresponding to the network device in the above method embodiment. For example, one circuit system is a chip system.

[0194] The communication device 700 includes at least one processor 701. Processor 701 can be used for internal processing of the device to implement certain control processing functions. Optionally, processor 701 includes instructions. Optionally, processor 701 can store data. Optionally, different processors can be independent devices, located in different physical locations, or on different integrated circuits. Optionally, different processors can be integrated into one or more processors, for example, on one or more integrated circuits.

[0195] Optionally, the communication device 700 includes one or more memories 703 for storing instructions. Optionally, data may also be stored in the memories 703. The processor and memory may be provided separately or integrated together.

[0196] Optionally, the communication device 700 includes a communication line 702 and at least one communication interface 704. Since the memory 703, the communication line 702 and the communication interface 704 are all optional, they are indicated by dotted lines in FIG7 .

[0197] Optionally, the communication device 700 may further include a transceiver and / or an antenna. The transceiver may be used to send information to or receive information from other devices. The transceiver may be referred to as a transceiver, a transceiver circuit, an input / output interface, etc., and is used to implement the transceiver function of the communication device 700 via an antenna. Optionally, the transceiver includes a transmitter and a receiver. For example, the transmitter may be used to generate a radio frequency signal from a baseband signal, and the receiver may be used to convert the radio frequency signal into a baseband signal.

[0198] The processor 701 may include a general-purpose central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present application.

[0199] Communication link 702 may include a pathway for transmitting information between the aforementioned components.

[0200] The communication interface 704 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), wired access network, etc.

[0201] The memory 703 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 703 may exist independently and be connected to the processor 701 via the communication line 702. Alternatively, the memory 703 may be integrated with the processor 701.

[0202] The memory 703 is used to store computer-executable instructions for executing the solution of the present application, and the execution is controlled by the processor 701. The processor 701 is used to execute the computer-executable instructions stored in the memory 703, thereby implementing the steps performed by the UE or network device in the embodiment shown in FIG.

[0203] Optionally, the computer-executable instructions in the embodiments of the present application may also be referred to as application code, which is not specifically limited in the embodiments of the present application.

[0204] In a specific implementation, as an embodiment, the processor 701 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 7 .

[0205] In a specific implementation, as an embodiment, the communication device 700 may include multiple processors, such as the processor 701 and the processor 705 in FIG7 . Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. The processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0206] When the device shown in FIG7 is a chip, such as a chip of a UE or a chip of a network device, the chip includes a processor 701 (and may also include a processor 705), a communication circuit 702, and a communication interface 704. Optionally, the chip may include a memory 703. Specifically, the communication interface 704 may be an input interface, a pin, or a circuit. The memory 703 may be a register, a cache, or the like. The processor 701 and the processor 705 may be a general-purpose CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the communication method of any of the above embodiments.

[0207] The embodiment of the present application can divide the functional modules of the device according to the above-mentioned method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. For example, in the case of dividing each functional module according to each function, Figure 8 shows a schematic diagram of a device, and the device 800 can be the UE or network device involved in the above-mentioned various method embodiments, or a chip in the UE or a chip in the network device. The device 800 includes a sending unit 801, a processing unit 802 and a receiving unit 803.

[0208] It should be understood that the device 800 can be used to implement the steps performed by the UE or network device in the communication method of the embodiment of the present application. The relevant features can refer to the embodiment shown in Figure 4 above and will not be repeated here.

[0209] Optionally, the functions / implementation processes of the sending unit 801, the receiving unit 803, and the processing unit 802 in FIG8 may be implemented by the processor 701 in FIG7 calling computer-executable instructions stored in the memory 703. Alternatively, the functions / implementation processes of the processing unit 802 in FIG8 may be implemented by the processor 701 in FIG7 calling computer-executable instructions stored in the memory 703, and the functions / implementation processes of the sending unit 801 and the receiving unit 803 in FIG8 may be implemented by the communication interface 704 in FIG7.

[0210] Optionally, when the device 800 is a chip or a circuit, the functions / implementation processes of the sending unit 801 and the receiving unit 803 can also be implemented through pins or circuits.

[0211] The present application also provides a computer-readable storage medium, which stores a computer program or instruction. When the computer program or instruction is executed, the method performed by the UE or network device in the aforementioned method embodiment is implemented. In this way, the functions described in the above embodiments can be implemented in the form of software functional units and sold or used as independent products. Based on this understanding, the technical solution of the present application can be essentially or in other words, the part that contributes or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. Storage media include various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0212] The present application also provides a computer program product, which includes: computer program code, which, when running on a computer, enables the computer to execute the method executed by the UE or the network device in any of the aforementioned method embodiments.

[0213] An embodiment of the present application further provides a processing device, including a processor and an interface; the processor is used to execute the method executed by the UE or network device involved in any of the above method embodiments.

[0214] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. 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 includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0215] The various illustrative logic units and circuits described in the embodiments of the present application can be implemented or operated by a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any conventional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.

[0216] The steps of the methods or algorithms described in the embodiments of the present application can be directly embedded in hardware, software units executed by a processor, or a combination of the two. The software unit can be stored in RAM, flash memory, ROM, erasable programmable read-only memory (EPROM), EEPROM, registers, hard disks, removable disks, CD-ROMs, or other storage media in any form known in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Alternatively, the storage medium can also be integrated into the processor. The processor and storage medium can be provided in an ASIC, which can be provided in a terminal device. Alternatively, the processor and storage medium can also be provided in different components in the terminal device.

[0217] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0218] The contents of the various embodiments of this application can refer to each other. If there is no special explanation and logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other. The technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0219] It is understood that in the embodiments of the present application, the UE and / or the network device may perform some or all of the steps in the embodiments of the present application. These steps or operations are merely examples. In the embodiments of the present application, other operations or variations of various operations may also be performed. In addition, the various steps may be performed in a different order than those presented in the embodiments of the present application, and it is possible that not all operations in the embodiments of the present application need to be performed.

Claims

1. A communication method, characterized in that: Applied to a terminal device, the terminal device includes a first receiver, and the method includes: receiving first information from a network device; Processing the first information using an artificial intelligence (AI) model or an AI function, wherein the AI ​​model or the AI ​​function is used to implement part or all of the functions of the first receiver; The performance of the AI ​​model or the AI ​​function is determined according to the processing result of the first information.

2. The method according to claim 1, characterized in that The information volume of the first information is less than or equal to a first threshold.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: receiving downlink control information from the network device, wherein the downlink control information is used to schedule the first information, wherein: The downlink control information includes first indication information, where the first indication information is used to indicate that the first information is used to determine the performance of the AI ​​model or the AI ​​function; or, The downlink control information is scrambled by a first wireless network temporary identifier RNTI, where the first RNTI is used to indicate that the first information is used to determine the performance of the AI ​​model or the AI ​​function.

4. The method according to claim 3, characterized in that The downlink control information is also used to indicate a transmission resource of the first information.

5. The method according to claim 4, characterized in that The downlink control information indicates the transmission resource of the first information in the following manner: The downlink control information is used to indicate the location of the transmission resource of the first information; or, The downlink control information includes an index of a transmission resource of the first information; or, The downlink control information is further used to schedule second information, where the second information is service data or control information. The downlink control information is further used to indicate an offset of a transmission resource of the first information relative to a transmission resource of the second information.

6. The method according to claim 5, characterized in that The first information and the second information are independently encoded respectively.

7. The method according to any one of claims 1 to 3, characterized in that: The transmission resource of the first information is preconfigured or predefined by the network device.

8. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: Sending first request information to the network device, where the first request information is used to request to determine the performance of the AI ​​model or the AI ​​function; or, Receive second indication information from the network device, where the second indication information is used to indicate the performance of determining the AI ​​model or the AI ​​function.

9. The method according to claim 8, characterized in that The first request information is included in a media access control MAC control element CE, or in uplink control information UCI, or in a radio resource control RRC message.

10. The method according to claim 8 or 9, characterized in that: The first request information also includes first auxiliary information, where the first auxiliary information is auxiliary information used to determine the performance of the AI ​​model or the AI ​​function.

11. The method according to any one of claims 1 to 10, characterized in that: The method further comprises: Receive first configuration information from the network device, wherein the first configuration information configures parameters of information used to determine performance of the AI ​​model or the AI ​​function.

12. The method according to claim 11, characterized in that The parameters used to determine the information of the AI ​​model or the AI ​​function performance include one or more of the following: type information of the information used to determine the performance of the AI ​​model or the AI ​​function; information about how the information used to determine the performance of the AI ​​model or the AI ​​function is generated; or, Information about the sending method of the information used to determine the performance of the AI ​​model or the AI ​​function.

13. The method according to claim 11 or 12, characterized in that: The first configuration information is also used to configure a true value; or, The first configuration information is also used to configure a method for determining a true value; Among them, the true value includes label information corresponding to the first information, and the label information is used by the terminal device to determine the performance of the AI ​​model or the AI ​​function based on the processing result of the first information and the label information.

14. The method according to claim 13, characterized in that The true value and the first information are both data, or both are sequences, or both are reference signals.

15. The method according to any one of claims 1 to 14, characterized in that: The method further comprises: Receive first configuration information from the network device, where the first configuration information includes third indication information, and the third indication information is used to indicate that the first information is only used to determine the performance of the AI ​​model or the AI ​​function.

16. The method according to any one of claims 1 to 14, characterized in that: The first information is only used to determine the performance of the AI ​​model or the AI ​​function.

17. The method according to any one of claims 11 to 16, characterized in that: The first configuration information is included in a unicast message; or, The first configuration information is included in a multicast message; or, The first configuration information is included in a broadcast message.

18. The method according to claim 17, characterized in that The broadcast message is a system information block SIB, or an RRC message.

19. The method according to any one of claims 1 to 18, characterized in that: The method further comprises: Sending second auxiliary information to the network device, wherein the second auxiliary information is used to indicate one or more of the following: whether the terminal device supports the AI ​​model or the AI ​​function, the way in which the terminal device performs AI model or AI function detection, or auxiliary information of information used to determine the performance of the AI ​​model or the AI ​​function.

20. The method according to claim 19, characterized in that The auxiliary information used to determine the performance of the AI ​​model or the AI ​​function includes one or more of the following: Type information of the AI ​​model or the AI ​​function; the size of the information used to determine the performance of the AI ​​model or the AI ​​function; type information of the information used to determine the performance of the AI ​​model or the AI ​​function; information about how the information used to determine the performance of the AI ​​model or the AI ​​function is generated; or, Information about the sending method of the information used to determine the performance of the AI ​​model or the AI ​​function.

21. The method according to any one of claims 1 to 20, characterized in that: The terminal device further includes a second receiver, which determines the performance of the AI ​​model or the AI ​​function according to the processing result of the first information, including: using the AI ​​model or the AI ​​function and the second receiver to process the first information respectively, and determining the performance of the AI ​​model or the AI ​​function according to the processing result of the AI ​​model or the AI ​​function and the processing result of the second receiver; or, The first information is processed using the AI ​​model or the AI ​​function, and the performance of the AI ​​model or the AI ​​function is determined according to the label information corresponding to the first information and the processing result of the AI ​​model or the AI ​​function.

22. The method according to any one of claims 1 to 21, characterized in that: The first information is included in a physical downlink shared channel PDSCH; or, The first information is included in a MAC protocol data unit PDU, and a MAC header of the MAC PDU is used to indicate that the first information is information used to determine the performance of the AI ​​model or the AI ​​function.

23. The method according to claim 22, characterized in that The first information is included in the PDSCH, and the method further includes: No hybrid automatic repeat request HARQ feedback information corresponding to the first information is sent.

24. The method according to claim 22, characterized in that The first information is included in the MAC PDU, and the method further includes: The MAC layer of the terminal device does not submit the first information to an upper layer of the MAC layer; or, The MAC layer of the terminal device discards the MAC PDU.

25. The method according to claim 22 or 23, characterized in that The first information is included in the PDSCH, and the first information is broadcast information.

26. The method according to any one of claims 1 to 25, characterized in that: The first information is information specifically used to determine the performance of the AI ​​model or the AI ​​function, or is business data.

27. A communication method, characterized in that: Applied to a network device, the method comprises: Sending first information, wherein the first information is used to determine the performance of an AI model or an AI function, and the AI ​​model or the AI ​​function is used to implement part or all of the functions of a first receiver of a terminal device.

28. The method according to claim 27, characterized in that The information volume of the first information is less than or equal to a first threshold.

29. The method according to claim 27 or 28, characterized in that The method further comprises: Sending downlink control information to the terminal device, where the downlink control information is used to schedule the first information, wherein: The downlink control information includes first indication information, where the first indication information is used to indicate that the first information is used to determine the AI ​​model or the AI ​​function performance; or, The downlink control information is encrypted by a first RNTI, and the first RNTI is used to indicate that the first information is used to determine the AI ​​model or the AI ​​function performance.

30. A communication device, characterized in that: The communication device comprises a processing unit and a transceiver unit, wherein the processing unit is coupled to the transceiver unit to execute the method according to any one of claims 1 to 26, or execute the method according to any one of claims 27 to 29.

31. A communication device, characterized in that: The communication device includes a processor and a memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the communication device performs the method as described in any one of claims 1 to 26, or the communication device performs the method as described in any one of claims 27 to 29.

32. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and when the computer program is run on a computer, the computer is caused to execute the method according to any one of claims 1 to 26, or the computer is caused to execute the method according to any one of claims 27 to 29.

33. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is run on a computer, the computer is caused to execute the method according to any one of claims 1 to 26, or the computer is caused to execute the method according to any one of claims 27 to 29.

34. A chip system, characterized in that: The chip system comprises: A processor and an interface, wherein the processor is used to call and run instructions from the interface, and when the processor executes the instructions, the method according to any one of claims 1 to 26 is implemented, or the method according to any one of claims 27 to 29 is implemented.

35. A communication system, characterized in that: The communication system includes a terminal device and a network device, wherein: The terminal device is used to execute the method according to any one of claims 1 to 26; The network device is used to execute the method according to any one of claims 27 to 29.

Images