Method and apparatus for wireless communication

Abstract

A method and apparatus for wireless communication are provided. The method comprises the following steps: the terminal equipment receives first configuration information, wherein the first configuration information is used for indicating an associated ID; the terminal equipment carries out model training and model reasoning on a first model according to the associated ID; wherein the first model corresponds to a first beam set and a second beam set, and the association ID is one or more of the following: a set ID used for selecting the first beam set and / or the second beam set; and the beam ID group comprises the ID of each beam in the first beam set and / or the second beam set.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and more specifically, to a method and device for wireless communication. Background Art

[0002] In some beam management scenarios, the terminal device can predict the downlink transmission beam through the model and report it to the network device. The model needs to be trained before it can be used for inference prediction. However, there is a certain time interval between the training phase and the inference phase. If the transmission conditions of the beam change, the model may be wrong during the inference phase. Therefore, how to ensure the consistency of the model in the training phase and the inference phase is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides a method and apparatus for wireless communication. The following introduces various aspects involved in the embodiments of the present application.

[0004] In a first aspect, a method for wireless communication is provided, comprising: a terminal device receives first configuration information, wherein the first configuration information is used to indicate an associated identification (identity, ID); the terminal device performs model training and model inference on a first model according to the associated ID; wherein the first model corresponds to a first beam set and a second beam set, and the associated ID is one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set.

[0005] According to a second aspect, a method for wireless communication is provided, comprising: a network device sends first configuration information, wherein the first configuration information is used to indicate an association ID; wherein the association ID is used by a terminal device to perform model training and model inference on a first model, the first model corresponds to a first beam set and a second beam set, and the association ID is one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set.

[0006] According to a third aspect, a device for wireless communication is provided, which is a terminal device and includes: a transceiver unit for receiving first configuration information, wherein the first configuration information is used to indicate an associated ID; a determination unit for performing model training and model inference on a first model according to the associated ID; wherein the first model corresponds to a first beam set and a second beam set, and the associated ID is one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set.

[0007] In a fourth aspect, a device for wireless communication is provided, which is a network device, comprising: a transceiver unit, used to send first configuration information, wherein the first configuration information is used to indicate an association ID; wherein the association ID is used by a terminal device to perform model training and model inference on a first model, and the first model corresponds to a first beam set and a second beam set, and the association ID is one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set.

[0008] In a fifth aspect, a communication device is provided, comprising a memory and a processor, wherein the memory is used to store a program, and the processor is used to call the program in the memory to execute the method described in the first aspect or the second aspect.

[0009] In a sixth aspect, a device is provided, comprising a processor, configured to call a program from a memory to execute the method described in the first aspect or the second aspect.

[0010] In a seventh aspect, a chip is provided, comprising a processor for calling a program from a memory so that a device equipped with the chip executes the method described in the first aspect or the second aspect.

[0011] According to an eighth aspect, a computer-readable storage medium is provided, on which a program is stored, wherein the program enables a computer to execute the method as described in the first aspect or the second aspect.

[0012] According to a ninth aspect, a computer program product is provided, comprising a program, wherein the program enables a computer to execute the method described in the first aspect or the second aspect.

[0013] In a tenth aspect, a computer program is provided, wherein the computer program enables a computer to execute the method as described in the first aspect or the second aspect.

[0014] The terminal device in the embodiment of the present application can determine the association ID according to the first configuration information to perform model training and model inference on the first model based on the association ID. In other words, the terminal device will perform beam measurement and beam prediction based on the same ID. When the beam transmission conditions change, the association ID may change, and the terminal device can update the first model in time to improve the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a wireless communication system applied in the embodiments of the present application.

[0016] Figure 2 It is a schematic diagram of the model processing process applied in the embodiment of the present application.

[0017] Figure 3 It is a flowchart of model reasoning on the terminal device side used in the embodiment of the present application.

[0018] Figure 4 It is a flowchart of a method for wireless communication provided in an embodiment of the present application.

[0019] Figure 5 yes Figure 4 A schematic diagram of a possible implementation of the method shown.

[0020] Figure 6 yes Figure 4 A schematic diagram of another possible implementation of the method shown.

[0021] Figure 7 yes Figure 4 A schematic diagram of yet another possible implementation of the method shown.

[0022] Figure 8 yes Figure 4 A schematic diagram of yet another possible implementation of the method shown.

[0023] Fig. 9 yes Figure 4 A schematic diagram of yet another possible implementation of the method shown.

[0024] Fig.10 yes Figure 4 A flowchart of a possible implementation of the method shown.

[0025] Fig.11 It is a structural diagram of a device for wireless communication provided in an embodiment of the present application.

[0026] Fig.12 It is a structural diagram of another device for wireless communication provided in an embodiment of the present application.

[0027] Fig.13 It is a structural diagram of a wireless communication device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application are described below in conjunction with the accompanying drawings.

[0029] Figure 1 FIG. 1 is a schematic diagram of the architecture of a wireless communication system 100 used in an embodiment of the present application. Figure 1 As shown, the wireless communication system 100 may include a network device 110 and a terminal device 120. The network device 110 may be a device that communicates with the terminal device 120. The network device 110 may provide communication coverage for a specific geographical area and may communicate with terminal devices located in the coverage area.

[0030] Figure 1 One network device and two terminal devices are shown exemplarily. Optionally, the wireless communication system 100 may include multiple network devices and each network device may include other number of terminal devices within its coverage area, which is not limited. That is, the wireless communication system may include one or more network devices, each of which may support wireless communication of one or more terminal devices.

[0031] In the embodiments of the present application, Figure 1 The communication system shown may also include other network entities such as a mobility management entity (MME), an access and mobility management function (AMF), and a network controller, but this is not limited in the embodiments of the present application.

[0032] It should be understood that the embodiments of the present application can be applied to various communication systems. For example, the embodiments of the present application can be applied to global system of mobile communication (GSM) system, code division multiple access (CDMA) system, wideband code division multiple access (WCDMA) system, general packet radio service (GPRS) system, long term evolution (LTE) system, advanced long term evolution (LTE-A) system, new radio (NR) system, NR system evolution system, LTE-based access to unlicensed spectrum (LTE-U) system, NR-based access to unlicensed spectrum (NR-U) system, universal mobile telecommunication system (UMTS), wireless local area network (WLAN) system, wireless fidelity (WiFi) system, fifth generation communication (5G) system. The embodiments of the present application can also be applied to other communication systems, such as the sixth-generation (6G) mobile communication system, or future communication systems such as satellite communication systems.

[0033] Traditional communication systems support a limited number of connections and are easy to implement. However, with the development of communication technology, communication systems can not only support traditional cellular communications, but also support one or more other types of communications. For example, a communication system can support one or more of the following communications: device to device (D2D) communication, machine to machine (M2M) communication, machine type communication (MTC), enhanced machine type communication (eMTC), vehicle to vehicle (V2V) communication, and vehicle to everything (V2X) communication, etc. The embodiments of the present application can also be applied to communication systems that support the above communication methods.

[0034] The communication system in the embodiment of the present application can be applied to a carrier aggregation (CA) scenario, a dual connectivity (DC) scenario, or a standalone (SA) networking scenario.

[0035] The communication system in the embodiment of the present application can be applied to an unlicensed spectrum. The unlicensed spectrum can also be considered as a shared spectrum. Alternatively, the communication system in the embodiment of the present application can also be applied to an authorized spectrum. The authorized spectrum can also be considered as a dedicated spectrum.

[0036] The embodiments of the present application can be applied to a non-terrestrial network (NTN) system. As an example, the NTN system can be a 4G-based NTN system, a NR-based NTN system, an Internet of Things (IoT)-based NTN system, or a narrowband Internet of Things (NB-IoT)-based NTN system.

[0037] The wireless communication system in the embodiment of the present application can use the following resources to support wireless communication with one or more communication devices: time resources (e.g., symbols, sub-time slots, time slots, subframes, frames, etc.) or frequency resources (e.g., subcarriers, carriers). Additionally, the wireless communication system can support wireless communication across various radio access technologies (radio access technology, RAT), including third generation (3G) radio access technology, fourth generation (4G) radio access technology, fifth generation (5G) radio access technology, and other suitable radio access technologies beyond 5G.

[0038] The terminal device in the embodiments of the present application may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station (MS), mobile terminal (MT), remote station, remote terminal, mobile device, user terminal, terminal, user communication equipment, wireless communication equipment, user agent or user device, etc.

[0039] In some embodiments, the terminal device in the embodiment of the present application may be a device that provides voice and / or data connectivity to a user, and may be used to connect people, objects, and machines, such as a handheld device with wireless connection function, a vehicle-mounted device, etc. The terminal device in the embodiment of the present application may be a mobile phone, a tablet computer (Pad), a laptop computer, a PDA, a mobile internet device (MID), a wearable device, a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical surgery, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, etc. Optionally, the UE may be used to act as a base station. For example, the UE may act as a scheduling entity that provides sidelink signals between UEs in V2X or D2D, etc. For example, a cellular phone and a car communicate with each other using sidelink signals. Cellular phones and smart home devices communicate with each other without relaying the communication signal through a base station.

[0040] In some embodiments, the terminal device may be a station (STATION, ST) in a WLAN. In some embodiments, the terminal device may be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA) device, a handheld device with wireless communication function, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a next-generation communication system (such as an NR system), or a terminal device in a future-evolved public land mobile network (PLMN) network, etc.

[0041] The network device in the embodiment of the present application may be a device for communicating with a terminal device, and the network device may also be referred to as an access network device or a radio access network device. The network device may be, for example, a base station. The network device in the embodiment of the present application may refer to a radio access network (RAN) node (or device) that connects a terminal device to a wireless network. The base station can broadly cover various names as follows, or be replaced with the following names, such as: NodeB, evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmission point (TRP), transmission point (TP), master station (MeNB), secondary station (SeNB), multi-standard radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, base band unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), positioning node, network communication equipment, etc. The base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof. The base station may also refer to a communication module, modem or chip used to be set in the aforementioned equipment or device. The base station may also be a mobile switching center and a device that performs the base station function in D2D, V2X, and M2M communications, a network-side device in a 6G network, and a device that performs the base station function in a future communication system. The base station may support networks with the same or different access technologies. The embodiments of the present application do not limit the specific technology and specific device form used by the network equipment.

[0042] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move based on the location of the mobile base station. In other examples, a helicopter or drone can be configured to act as a device that communicates with another base station.

[0043] In some deployments, the network device in the embodiments of the present application may refer to a CU or a DU, or the network device includes a CU and a DU. The gNB may also include an AAU.

[0044] The network equipment and terminal equipment can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on the water surface; they can also be deployed on aircraft, balloons and satellites in the air. The embodiments of the present application do not limit the scenarios in which the network equipment and terminal equipment are located.

[0045] It should be understood that all or part of the functions of the communication device in the present application may also be implemented by software functions running on hardware, or by virtualization functions instantiated on a platform (eg, a cloud platform).

[0046] In an embodiment of the present application, a network device may provide services for a cell, and a terminal device may communicate with the network device through transmission resources (e.g., frequency domain resources, or spectrum resources) used by the cell. The cell may be a cell corresponding to a network device (e.g., a base station). The cell may belong to a macro base station or a base station corresponding to a small cell. The small cells here may include: metro cells, micro cells, pico cells, femto cells, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.

[0047] It should be understood that the device with communication function in the network / system in the embodiment of the present application can be referred to as a communication device. Figure 1 Taking the wireless communication system 100 shown as an example, the communication equipment may include a network device 110 and a terminal device 120 with communication functions, and may also include other devices in the wireless communication system 100, such as a network controller, a mobile management entity and other network entities, which is not limited in the embodiments of the present application.

[0048] For ease of understanding, some relevant technical knowledge involved in the embodiments of the present application is first introduced. The following related technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all belong to the protection scope of the embodiments of the present application. The embodiments of the present application include at least part of the following contents.

[0049] In a wireless communication system, a terminal device may acquire a beam (also referred to as a spatial beam) to achieve a wireless connection between the terminal device and a wireless network. Exemplarily, the terminal device may perform beam scanning for an available beam sent by a wireless network and measure the properties of the beam. The properties of the beam include signal strength and signal quality. Exemplarily, after performing beam scanning, the terminal device may also perform beam refinement to achieve a set of potentially narrower beams for wireless connection to the wireless network. The beam may not only achieve a wireless connection between the terminal device and the wireless network, but also achieve high directional accuracy and high signal quality for wireless signal transmission between the terminal device and the wireless network.

[0050] With the development of communication technology, research on artificial intelligence (AI) / machine learning (ML) technology based on the air interface of communication systems (e.g., NR systems) has become one of the directions. The goals of this research include exploring how to enhance the advantages of the air interface. For example, the relevant performance of the air interface can be enhanced by enhancing support for AI / ML algorithms. For another example, the complexity and / or overhead of the air interface can be reduced by enhancing support for AI / ML algorithms.

[0051] Research on AI / ML technology can also enhance beam management (BM) related functions. As an example, AI / ML enhancements related to beam management can support reducing overhead and reducing beam measurement and reporting latency. As an example, applying AI / ML models can predict beams to improve the transmission efficiency of the air interface.

[0052] The entire process of applying AI / ML models for enhancement includes: model training, model reasoning, and model monitoring. In this process, after being trained, the AI / ML model can generate a set of outputs based on a set of inputs. The input can be a set of beam measurements, and the output can be a set of beam groups that are different from / larger than the input.

[0053] During model inference, the AI / ML model can predict the best beam from a set of different / larger beam groups based on a set of beam measurements.

[0054] In some embodiments, the AI / ML model may be located at the terminal device side or the terminal device may perform model training and / or model reasoning. The model may be referred to as a UE-side model. For example, the AI ​​model may be located at the terminal device or the terminal device may perform AI model training and / or generate the best beam using AI model reasoning.

[0055] As an example, the terminal device may use the beams in Set B (also called beam group B) as input to the ML model. The ML model may predict the best beam in Set A (also called beam group A), which is not fully measured by the terminal device.

[0056] In the above example, set B may be the beam group that the terminal device first measures. Set B may be multiple beams sent by a base station (e.g., a gNB). Each beam may correspond to a different direction or angle to cover multiple spatial directions. Each beam may also correspond to a measurement signal for obtaining a measurement value, such as reference signal received power (RSRP). The purpose of set B is to provide preliminary environmental information and channel conditions for the model.

[0057] In the above example, set A may be the group of beams that need to be predicted. The number of beams in set A is usually larger than that in set B, or the beam directions may be more concentrated. The AI / ML model can predict the best beam in set A through measurements of set B to improve the efficiency of data transmission.

[0058] Optionally, the beams of set A and set B can be in the same frequency range. The selection of set B can be given by the base station or determined by the terminal device itself. The relationship between set A and set B can be: set A and set B are different (set B is not a subset of set A), or set B is a subset of set A (set A and set B are different), or set A and set B are the same. For the first two, set B can be transmitted in both the measurement window and the prediction window, or it can be transmitted only in the measurement window. The last case can save the reference signal (RS) transmission overhead, and set B as a measurement resource can be transmitted only in the measurement window.

[0059] Optionally, 64 or even more beams can be used as the size of beam set A. For future-oriented networks, network devices will be able to transmit 64 more highly directional narrow beams. More narrow beams can also scan a larger set A, for example, the number of beams in set A may be as high as 256.

[0060] Optionally, the network device may send a channel state information reference signal (CSI-RS) or a synchronization signal block (SSB) as a reference signal. It should be understood that SSB may also represent a synchronization signal / physical broadcast channel block (SS / PBCH block). SSB includes a primary synchronization signal (PSS) and a secondary synchronization signal (SSS).

[0061] Optionally, the terminal device estimates the channel quality of each beam by measuring the RSRP of the received CSI-RS / SSS.

[0062] Optionally, during the model training process, the AI / ML model can adjust its weights by minimizing the loss function so that the model can accurately predict the best beam in set A from the RSRP measurement results of set B.

[0063] In some embodiments, AI / ML may be located at a network device (such as a base station) or the network device may perform model training and / or model reasoning. The model may be referred to as a network-side model. For example, the AI ​​model may be located at a base station or the base station may perform AI model training and / or generate the best beam using AI model reasoning.

[0064] In some embodiments, the network can fully control the data collection process for model training on the terminal device side, including the initiation, termination and management of data collection and data transmission.

[0065] In the model monitoring process, AI / ML model monitoring is used for at least the following purposes: model activation, deactivation, selection, switching, rollback, and update (including retraining). Model monitoring can also be referred to as the process of monitoring the inference performance of AI / ML models. There must be a time interval between the training process and the inference process of the model. When radio parameters / conditions change in the network, there is a high probability of errors in the inference phase of the model, so the model needs to be continuously corrected and trained based on the results of model monitoring.

[0066] Compared with other beam management technologies, through AI / ML-enabled beam management, terminal devices can experience reduced latency, reduced overhead, reduced power consumption, and improved signal quality based on beam prediction.

[0067] For ease of understanding, the following Figure 2 , the entire process of model processing on the terminal device side is introduced. Figure 2 This is explained from the perspective of the interaction between the terminal device (such as UE) and the network (NW). The network side uses 4 beams as an example. Figure 2 It can be seen that the whole process may include a model training process, a model inference process and a reporting process. The model training process includes steps S210 and S220, and the model inference process includes steps S230 and S240.

[0068] See also Figure 2 In step S210, the terminal device reports training-related information (UE report training-related information).

[0069] In step S220, the network side performs beam scanning based on four beams.

[0070] In step S230, the terminal device reports inference-related information (UEreport inference-related information).

[0071] In step S240, the network side selects two beams from four beams to perform beam scanning according to the report from the terminal device.

[0072] In step S250, the terminal device reports the optimal K beams (top-K beam report).

[0073] In step S260, the network side performs beam scanning according to the beam in the beam report. Figure 2 It can be seen that the two beams used for beam scanning at step S260 on the network side may be different from the two beams used for beam scanning at step S240 .

[0074] In step S270, the terminal device sends a beam report to facilitate the network side to determine the optimal beam.

[0075] In step S280, the network side sends a beam indication to the terminal device.

[0076] Combination of the above Figure 2This article introduces the process of processing the model when it is located on the terminal device side. In order to complete the entire process, the terminal device needs to collect and analyze data. As mentioned above, the data collection process can include data collection, initiation of data transmission, transmission, and management. The following introduces data collection and analysis using the terminal device side model as an example.

[0077] For the terminal device side model, the behavior of the terminal device or the content of the terminal device report may vary depending on whether it is a measurement resource for training, reasoning, or monitoring. In other words, the content may be different for different terminal device behaviors or terminal device reports. Therefore, when configuring AI / ML operation-related measurements for the terminal device side model, it is necessary to indicate to the terminal device the purpose of the measurement configuration or the implied terminal device behavior (such as training, reasoning, monitoring, or non-AI / ML operation).

[0078] As an example, for the training process, the terminal device may only need to measure the beams of the configured resources. For example, the model input and labels generated according to set A / set B can be used for internal training on the terminal device side, and there is no need to report to the network device.

[0079] As an example, for the inference process, the terminal device may need to measure the transmit (Tx) beams from set B and use them as model input, while the terminal device should also report the model output regarding the predicted beam / RSRP value.

[0080] For data collection, in the embodiment of the present application, data collection can be started / triggered by network configuration, and data collection executed by terminal device request can also be supported. Data collection for different purposes can correspond to different behaviors of terminal devices.

[0081] As an example, the collected data can be used to train AI models or give terminal devices a preliminary understanding of network performance. When performing data collection based on a training data set, the terminal device needs to be equipped with a mechanism to determine when and what data to collect for effective training. Since the network side may not know the specific data requirements of the terminal device or the best time for data collection, the terminal device can autonomously trigger the collection process and report the relevant configuration. For example, a terminal device-side AI / ML model, whose input may include layer 1 (L1) RSRP of 8 SSBs, and whose output may include predicted L1-RSRP of 32 CSI RSs. The 8 SSBs can correspond to set B for training purposes, and the 32 CSI RSs can correspond to set A for training purposes.

[0082] Taking the AI / ML model in the above example as an example, in order to collect data for training the AI / ML model, the network can configure the first SSB resource set and the first CSI-RS resource set to the terminal device. In the first SSB resource set, the SSB resource indicator (resource indicator, RI) or resource index (resource index, RI) can be from 1 to 8. SSBRI can be defined based on the ID associated with each SSB in the SSB resource set. In the first CSI-RS resource set, the CSI-RS resource indicator or index (CSI-RS RI, CRI) can be from 1 to 32. CRI can be defined based on the ID associated with each CSI-RS in the CSI-RS resource set.

[0083] Optionally, the CSI-RS in the first CSI-RS resource set will share the same periodicity with the SSB in the first SSB resource set.

[0084] Optionally, the terminal device can collect L1-RSRP about SSB and CSI-RS at various time occasions, and upload the data to the over the top (OTT) server for offline model training.

[0085] Optionally, during the model training process, the L1-RSRP corresponding to SSB RI=m (1≤m≤8) in the first SSB resource set can be used to determine the input value of the mth input feature of the AI / ML model, and the L1-RSRP corresponding to CRI=n (1≤n≤32) in the first CSI-RS resource set can be used to determine the value label associated with the nth output feature of the AI / ML model. This offline trained AI / ML model can then be downloaded back to the terminal device for future model reasoning.

[0086] As an example, for the terminal device side model, the training data can be generated by the terminal device. The termination point of the training data can include the terminal device or the terminal device side OTT server. Operation administration and maintenance (OAM) or the core network can be used to collect the situation of the terminal device side model training.

[0087] In the above example, the OTT server on the terminal device side collects data and trains the model. Therefore, the OTT server on the terminal device side knows what data it needs. Leaving the data collection on the terminal device side to the OTT server means that the OTT server can collect the data it needs more directly. The required data type does not need to be specified, and when / what to transmit is determined by the terminal device, which can provide full flexibility for terminal device / chipset vendors to train and implement their specific AI / ML models. The terminal device can use the same method to send training data to the OTT server on the terminal device side for model training. This process is transparent to the network side and does not require control / visibility from the network / mobile network operator (MNO).

[0088] In the above example, the terminal device itself should be responsible for protecting data privacy and obtaining user consent. More specifically, the training data is reported via the user plane (UP), that is, reported from the application-level data collection client of the terminal device to the application server (that is, the OTT server on the terminal device side). The terminal device vendor can install its own data collection client at the terminal device to collect lower-layer data and report it to its own application server for AI / ML model training. For the data collection application client at the terminal device, data transmission from the terminal device to the application server of the terminal device vendor can be supported in accordance with relevant regulations without the participation of the network side. The collected data type / format does not need to be specified, which provides full flexibility for the terminal device to train and implement its specific AI / ML model.

[0089] As an example, after the terminal device collects the training data, it can first transmit it to a data collection server (within the MNO) for terminal device side model training. Then, the training data can be transmitted from the data collection server to the OTT server (outside the MNO). That is, the terminal device can collect data and transmit it to the server for data collection for terminal device side model training. The data collection server for terminal device side model training can be marked as the first entity.

[0090] In the above example, the first entity can choose whether to send the collected data to the OTT server. The ownership of the OTT server is different from that of the OTT server in the previous example. The terminal device can transmit the collected data to the first entity within the MNO through the UP, and then the first entity can choose to forward the collected data to the OTT server, which depends on the implementation of the terminal device.

[0091] The following still takes the terminal device side model as an example to introduce the model reasoning and reporting process, which also includes the beam management reasoning process.

[0092] For the inference process of the model on the terminal device side, the terminal device measures the RS of the set B beams and predicts the Top-K beams of set A, and reports the prediction results to the network. In the traditional L1-RSRP report, the terminal device should report the L1-RSRP value of the channel measurement resource (CMR) associated with the CSI report. However, for artificial intelligence-based beam management, the beam group used for measurement (set B) and the beam group used for reporting (set A) may be different, so in order to notify the terminal device to report the prediction results, the association between set A and set B should be indicated to the terminal device.

[0093] It should be noted that the model training process also needs to confirm the association between set A and set B. For model training, the measurement results of set B will be used as model input, and the measurement results of set A will be used as true value labels. Therefore, the method of associating set B with set A can be to configure both the resources of set A and the resources of set B to the terminal device in advance. However, for model reasoning, the terminal device only needs to measure the beams in set B, and the resources or indexes of the beams in set A will only be used for reporting. Therefore, the key issue in the model reasoning process is how to represent the resources or indexes of set A.

[0094] During the inference process, the terminal device will not measure the beams in set A, so there is no need to configure the RS for the beams in set A. However, in some scenarios, the RS for the beams in set A can be configured to the terminal device for other purposes, such as performance monitoring. In this case, both the resource set of set B and the resource set of set A will be configured to the terminal device. It can be seen that the network can configure the resource set of set B and the resource set of set A based on different purposes. Taking CSI resources as an example, the terminal device needs to know the intention of the configured CSI resources (whether it is for training, inference, monitoring or non-AI / ML operations), because the corresponding CSI report types may be different.

[0095] As an example, for monitoring, depending on the discussion of monitoring types, the terminal device may need to measure the monitoring resources configured by the network equipment and report the model output / labels, or report the calculated metrics (e.g., beam prediction accuracy). For traditional non-AI / ML operations, the terminal device may need to measure Tx beams and report the measured beams / RSRP.

[0096] During the inference process of the AI / ML model, the measurements based on the Set B beams can be used as model input. In addition, the beam ID information can also be provided as input to the model. Based on the model output, the Top-1 / Top-K beams among Set A of beams can be obtained, or the Top-1 / Top-K beams among Set A of beams can be determined using the predicted L1-RSRP (depending on the label).

[0097] Optionally, the output of the model is, for example, the probability of each beam in Set A to be the Top-1 beam, or the predicted L1-RSRPs.

[0098] In some scenarios, the terminal device can support beam prediction in the spatial domain and / or time domain, namely BM-Case1 and / or BM-Case2. Based on AI / ML enhancements, beam prediction in the spatial domain (BM-Case1) and beam prediction in the time domain (BM-Case2) can reduce the overhead of the terminal device and reduce beam measurement and reporting delays.

[0099] BM-Case1 is to perform spatial domain downlink (DL) beam prediction for Set A based on the measurement results of Set B. For BM-case1, the measurements based on Set B of beams are used as model input to predict the Top-1 / Top-K beams in Set A.

[0100] BM-Case2 is a time-domain DL beam prediction for Set A based on historical measurements of Set B. For BM-Case2, measurements based on Set B of beams at historic time instance(s) can be used as model input to predict the time-domain DL beams for Set A. The prediction of DL Tx beams and the prediction of DL Tx / Rx beams can also be used to evaluate the performance of the prediction.

[0101] For BM-Case1 and BM-Case2, the terminal device can report the prediction results to the NW based on the output of the terminal device side model, or the NW can predict the Top-1 / Top-K beams based on the measurement report of the set B of the NW side model.

[0102] Still taking the terminal device side model as an example, for BM-Case1 with the terminal device side AI / ML model, the traditional transmission configuration indicator (TCI) state mechanism can be used to perform beam indication of the beam. Optionally, the terminal device can report the measurement results of more than 4 beams in one reporting instance. For BM-Case2 with the terminal device side AI / ML model, the terminal device can report the following information of AI / ML model reasoning to NW in one reporting example: the beams used for N future times based on the output of AI / ML model reasoning; timestamp information corresponding to the reported beam; information about the measurement values ​​of multiple past time instances.

[0103] In the reasoning process of beam management, beam indication is also an important information. After the terminal device reports the Top-K predicted beam, the NW will further indicate the beam used for the second step measurement or the terminal device can directly trigger the second measurement. Unlike BM-case1, BM-Case2 can obtain Top-K beams corresponding to multiple time instances. The beam indication on the network device side needs to consider the time information.

[0104] As an example, the network device may use multiple indications to indicate beams to the terminal device. The advantage of this is that the network device may select a more appropriate beam according to real-time channel changes.

[0105] Combine the following Figure 3 , taking the UE side model as an example, the model reasoning process based on beam prediction is introduced based on the interaction between UE and gNB.

[0106] See also Figure 3 In step S310, the gNB performs scanning based on the beams in the sparse set B (sparse Set Bsweeping).

[0107] In step S320, the model on the UE side performs Top-K beam prediction. The UE inputs the measurement value of the set B beam into the AI ​​model, and outputs the index and L1-RSRP of the top K best beams among all measured beams and the beam ID or beam index. The UE side model can also output the best beam LI-RSRP and the difference between the RSRP of the other K-1 beams and the best beam L1-RSRP to the base station. The K value can be configured by the base station, or the UE can determine the value of K based on the mobile speed, location information, and business mode.

[0108] In step S330, report CRI and / or predicted quality of Top K beam.

[0109] In step S340, the gNB may continue to initiate scanning with the best Top-K beams (Top-K beam sweeping). The Top-K beam sweeping procedure can be configured by the gNB. Step S340 is an optional step.

[0110] In step S350, the UE reports the beam quality (beam quality report). The UE may report the actual top 1 best beam, or may report the index and L1-RSRP of part or all of the top K best beams. The UE may reuse the traditional beam reporting mechanism.

[0111] In step S360, the gNB indicates the beam used for DL ​​data transmission. The gNB can reuse the traditional TCI beam indication mechanism.

[0112] Combination of the above Figure 2 and Figure 3 The process of beam management or beam prediction based on AI / ML models is introduced. Beam management enhancement based on AI / ML still faces some problems that need to be solved or studied.

[0113] As an example, when a terminal device uses an AI algorithm to process beams detected and measured from a wireless network to infer other beams that may have higher strength and / or higher quality, the terminal device needs to ensure the consistency of the model in the training phase and the inference phase. As mentioned above, there is a certain time interval between the training phase and the inference phase, and changes in radio parameters / conditions may cause errors in the inference phase.

[0114] As an example, as can be seen from the previous text, set A usually has more beams than set B. Among them, set B usually represents a measurement set or a partial set of downlink reference signals, which is used to assist the network side in beam prediction and selection. However, the terminal device may not need to know whether the beam scanning performed by the network device is for the beams of set A or for the beams of set B, or even for non-AI / ML-based BM. Therefore, how the terminal device determines the beams in set B among the beams sent by the network device is a question that needs to be considered.

[0115] Based on this, an embodiment of the present application proposes a method for wireless communication. In this method, the terminal device can determine the association ID corresponding to the first model based on the first configuration information, and then select the beam set for model training and model inference based on the association ID. In other words, the terminal device will perform beam measurement and beam prediction based on the same beam ID. When the beam transmission condition changes, the association ID of the beam set will be updated, and the terminal device can update the first model according to the new association ID to improve the prediction accuracy.

[0116] Combine the following Figure 4 The method for wireless communication proposed in the embodiment of the present application is described in detail. Figure 4 It is introduced from the perspective of the interaction between terminal devices and network devices.

[0117] The terminal device may be any of the terminal devices described above. For example, the terminal device is a UE.

[0118] In some embodiments, the terminal device supports function enhancement based on AI / ML operation. For example, the terminal device has the function of enhancing beam management based on AI / ML operation.

[0119] In some embodiments, a first model is deployed on the terminal device to predict the beam sent by the network device or other terminal devices. As an example, the first model is a model that supports AI algorithms or ML, that is, the first model is an AI / ML model.

[0120] As an example, the beam prediction implemented by the first model may be BM-Case1 described above, or BM-Case2, or other beam prediction types in the future, which are not limited here.

[0121] In some embodiments, the first model is deployed on the terminal device side. That is, the first model is not on the first device, but on a server that communicates with the first device. For example, the first model is on an OTT server that directly communicates with the terminal device.

[0122] In some embodiments, the first model may be one of multiple models deployed on the terminal device side. The first model may correspond to a network device that currently sends a beam or a beam set to the terminal device to predict the transmission beam of the network device. For example, the terminal device side may develop multiple models, such as models 1, 2, 3, and 4, and the first model may be any of the four models.

[0123] As an example, multiple models may be used to predict transmit beams for different scenarios or different beam transmitting devices.

[0124] The network device may be any network device that communicates with the terminal device, for example, a base station.

[0125] In some embodiments, the network device may monitor the process of the terminal device processing the first model. For example, the network device may determine whether the current stage is the training stage or the inference stage of the first model based on the report sent by the terminal device.

[0126] In some embodiments, the network device supports AI / ML operations. A first model may be deployed on the network device side.

[0127] In some embodiments, the network device may send multiple beams to the terminal device multiple times to facilitate the terminal device to perform measurements, and perform model training and model inference on the first model based on the measurement results.

[0128] See also Figure 4 In step S410, the terminal device receives first configuration information. As an example, the terminal device may receive first configuration information sent by the network device. As an example, the terminal device may receive first configuration information indicated by higher layer signaling.

[0129] The first configuration information is used to indicate the association ID, which can also be called the correlation ID. By introducing the association ID mechanism, the consistency of the model training and reasoning process can be determined. As an example, the association ID can be associated with the beam direction through a reference signal. The terminal device can estimate the channel state based on the CSI-RS signal in different directions, and the first model can obtain the measurement values ​​and association information of different beams through the signal index to ensure the consistency of the training and reasoning process.

[0130] As an example, during the training and inference process of the first model, the association ID is used to ensure the consistency between the first beam set and the second beam set. Different measurement scenarios, mobile environments, and time-varying wireless channel characteristics may cause changes in the measurement data, so the association ID can help ensure that the data remains consistent and relevant during the training and inference process.

[0131] As an example, the first configuration information may also indicate relevant configuration of data collection. The association ID is used for the data collection.

[0132] In some embodiments, the first configuration information may directly indicate the association ID. For example, the first configuration information includes the association ID. In some embodiments, the first configuration information may be used to determine the association ID. For example, the first configuration information may instruct the terminal device to determine the association ID based on the cell or the corresponding network device.

[0133] As an example, the association ID can be introduced under multiple frameworks. The multiple frameworks may include the CSI framework and other frameworks related to the beam reference signal. The multiple frameworks may also be included within the framework and outside the framework. Taking the CSI framework as an example, the multiple frameworks may be included within the CSI framework or outside the CSI framework. In the scenario where the association ID is introduced in the CSI framework, the association ID can directly use the resource index of the CSI, for example, the association ID is CSI-ResourceConfigID. When the association ID is introduced outside the CSI framework, since the association ID corresponds to the "beam-related attributes" of the network device, the indication mechanism of the association ID can be similar to the beam indication, where the TCI state is activated and indicated.

[0134] As an example, the association ID may be sent and marked together with the reference signal carried by the beam. For example, in the CSI framework, the association ID may be sent and marked together with the CSI reference signal (CSI-RS), so that the first model can use the same association ID to determine the selection of the beam in both the measurement phase (training phase) and the inference phase. For another example, the association ID may be sent and marked together with the SSB.

[0135] In the above example, the network device can send multiple sets of CSI-RS, and the terminal device receives these signals in different spatial directions. The terminal device can measure the channel state information (such as RSRP, SINR, etc.) of each beam and assign an associated ID (such as CSI-RS resource index, beam ID) to each beam.

[0136] In the above example, each CSI resource may correspond to a specific set of resource blocks, symbols, and antenna port combinations for channel measurement and beam selection. The network device may identify these resources by a CSI resource index (CSI-RS index), thereby indirectly identifying the beam. For example, a CSI resource may correspond to a specific set of resource blocks, such as H resource blocks, where H is a positive integer. The index of each resource block increases in order from low to high, i.e., index1, index2, ..., indexH. The terminal device may use each beam ID in the CSI resource beam group to associate with a CSI-RS resource index.

[0137] As an example, the association ID can be carried in the CSI resource configuration (CSI-ResourceConfig) related to beam transmission, which will be combined with Figures 5 to 7 The invention is exemplified by the following examples.

[0138] As an example, the association ID can be carried in the CSI periodic trigger state list (CSI-AperiodicTriggerStateList), which will be combined with Figures 8 to 10The invention is exemplified by the following examples.

[0139] As an example, the association ID may be carried in radio resource control (RRC) signaling. For example, when the association ID is an ID of a beam set, the association ID may be carried in the RRC signaling.

[0140] In the above example, the network device can use the RRC configuration association ID list at the beginning. When the network device selects a new association ID to use (different from the current association ID), the network device can use a different signal (for example, new downlink control information (DCI)) to indicate the index of the target association ID in the list to the terminal device. After receiving the indication message, the terminal device can assume that all RS resources are sent with the new NW configuration. The advantage of this mechanism is that the signaling indicated by the association ID is separate from the signaling triggered by the report. Therefore, by indicating the association ID in advance, the terminal device can have more time to switch to the corresponding model based on the association ID. In some scenarios, this mechanism can ensure that the terminal device only runs one model at a time.

[0141] The association ID is an ID used for the first model and / or the function in the first model mentioned above. During the training and reasoning process of the first model, the association ID is used to ensure the consistency between the measurement beam set and the prediction beam set corresponding to the first model. The measurement beam set may also be referred to as the first beam set, i.e., the set B mentioned above. The prediction beam set may also be referred to as the second beam set, i.e., the set A mentioned above.

[0142] The association ID may be one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set. In other words, the association ID may be divided into a set ID and an ID for each beam, so as to be applied to different scenarios or different requirements.

[0143] In some embodiments, the set ID may be an ID corresponding to both the first beam set and the second beam set. That is, the first beam set and the second beam set corresponding to the first model have the same associated ID to facilitate data collection, training and reasoning by the terminal device.

[0144] As an example, when the association ID is a set ID, the first configuration information may be sent via a resource index of a beam. For example, when the association ID is a set ID, the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

[0145] In the above example, the first configuration information can be carried in one of the following information: resource configuration information shared by the first beam set and the second beam set; a configuration report shared by the first beam set and the second beam set; first resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

[0146] Taking the resource configuration information of CSI-RS as an example, the resource configuration information can be indicated by CSI-ResourceConfig. Each CSI-RS resource has its unique ID (CSI-RS resource index). The ID information can correspond to the associated ID. The first model can identify the CSI-RS measurement values ​​associated with different beam directions through the index.

[0147] As an example, the first configuration information may be carried in a trigger list associated with a first report, and the first report is determined according to the inference result of the first model. Exemplarily, in order to save the use of CSI-ResourceConfig, the network device may send the first configuration information when triggering / activating the first report to allocate / indicate the association ID. Then, the association ID is shared in the triggered / activated first report.

[0148] In the above example, the first configuration information can be carried in one of the following trigger lists: a total list of related report configuration information; a first list of related report configuration information corresponding to the first beam set and a second list of related report configuration information corresponding to the second beam set.

[0149] In some embodiments, the beam ID group may include an ID of each beam in the first beam set and / or the second beam set. In some scenarios, the configuration of the beam ID group may facilitate the terminal device to select a suitable model for beam prediction.

[0150] As an example, assume that there are two different base station beam codebook (CB) configurations, CB1 and CB2, on the NW side. Without knowing the base station beam codebook configuration, the terminal device side may develop models 1, 2, 3, and 4. Each model is developed based on some assumed base station beam shape. Exemplarily, assume that model 1 happens to work for CB1 (perhaps because the base station beam shape assumed in model 1 training matches CB1), and model 2 happens to work for CB2. Assume that the base station uses CB1 during inference. Before inference, the terminal device needs to blindly try all 4 models and monitor their performance. When the terminal device finds that model 1 works well, it will continue to use model 1 during inference. It should be understood that there is no model identification in this scenario; the NW does not know that there are four models at the terminal device, and models 1-4 are just internal models on the terminal device side. Obviously, this monitoring-based trial-and-error approach is inefficient because it incurs overhead in terms of latency and power consumption of the terminal device. When training the model, the beam shape assumed by the terminal device may also not match the codebook of the actual base station. Therefore, the scheme of selecting the best model based on monitoring may have poor performance compared with other solutions based on offline / over-the-air alignment or model transmission. To address this problem, an ID selected to each beam is introduced as an identifier for model training and inference.

[0151] As an example, when the association ID is an ID beam group, the association ID is a beam ID group, the beam ID group is represented by a beam association ID matrix, and the beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

[0152] In the above example, the beam direction identifier (e.g., beam ID) can be used to identify and measure the ID of each beam in the first beam set. In order to better determine the relationship between the first beam set and the second beam set, the network device can construct a beam association ID matrix M to represent the radiation relationship between the first beam set and the second beam set. For example, each element M(i,j) in the M matrix can represent the correlation between the beam Bi in the first beam set (set B) and the beam Aj in the second beam set (set A). Through machine learning methods, the beam association ID matrix can be learned during the training process and used for beam prediction during inference.

[0153] It should be understood that in a wireless communication system, the channel state changes over time. Therefore, the association ID relationship between the first beam set and the second beam set may change over time. To address this issue, the terminal device can use a sliding window strategy. By using the sliding window strategy, the terminal device can update the beam association ID in real time based on the latest CSI-RS measurement data. After each update, the network device and the terminal device can communicate using the latest association ID.

[0154] In the above example, for the M matrix, the terminal device can use historical measurement data to calculate the spatial correlation between beams. If beams Bi and Aj exhibit similar channel states (such as RSRP values) in past measurements, M(i, j) has a larger correlation value.

[0155] As an example, the beam association ID matrix is ​​determined according to the resource index of the first beam set and the second beam set. That is, the beam association ID matrix may correspond to the resource index matrix of the first beam set and the second beam set. For example, the network device may construct a beam association ID matrix M1 based on the CSI beam resource index to represent the mapping relationship between the two beam sets. Each element M1(i,j) in the M1 matrix may represent the correlation between each beam resource index in the first beam set and the beam Aj in the second beam set.

[0156] In some embodiments, the association ID may include a set ID and a beam ID group. For example, the ID of the entire beam ID group is the set ID.

[0157] In some embodiments, the association ID remains unchanged during the model training and reasoning phases. As an example, when a terminal device measures the first beam set of beams at different times and performs AI / ML reasoning, it is generally believed that the same association ID as in the training phase can be maintained under a time-varying channel. In order to keep the association ID unchanged, it is necessary to ensure the spatiotemporal correlation and beam feature stability. Spatiotemporal correlation means that the time variability of the channel is relatively small within a shorter time scale. In this scenario, the measurement value (first beam set) of the terminal device has a certain correlation with the previous training data, allowing reasoning under the same association ID matrix for each beam. Beam feature stability means that certain beam IDs or beam directions may be relatively stable for a certain terminal device within a certain period of time. In this scenario, the terminal device can assume that its environment is similar to that during training during the reasoning phase, the correlation of the beams is still maintained, and the association ID matrix will not change significantly.

[0158] In some embodiments, the environment during the model training and reasoning phases may change, but the change is relatively small and the association ID does not need to be updated. Optionally, the association ID may determine whether to adjust based on one or more of the following information: the time interval between model training and model reasoning; the channel time variability corresponding to the model training and model reasoning; whether the beam characteristics of the first beam set and the second beam set are stable. When the time interval is small, or the channel time variability is small, or the beam characteristics are stable, the association ID is not adjusted even if the environment changes.

[0159] In some embodiments, the association ID is used for any of the multiple updated versions of the first model. As an example, after the first model is trained, the first model corresponds to the association ID in the subsequent reasoning process. However, once the first model is trained and updated based on the new association ID, the association ID corresponding to the first model is the new association ID, not the previous association ID.

[0160] In some embodiments, the association ID may be determined based on a large number of static IDs preconfigured by the network device. The first configuration information may be used by the terminal device to determine the association ID corresponding to the first model from a large number of static IDs.

[0161] In some embodiments, the association ID can be dynamically configured to avoid the unlimited growth of the number of association IDs. For example, within a NW, there may be multiple sets of NW-side related configurations, each for different cells on the NW side. In addition, over time, new NW-side related configurations or new environmental changes may appear, which makes the trained first model no longer feasible.

[0162] As an example, the network device can dynamically generate the association ID based on the network demand and network load of the cell. That is, the association ID is dynamically generated based on the current network demand and user connection status to avoid generating a large number of static IDs for each possible configuration or scenario in advance.

[0163] As an example, the network device pre-generates a certain number of association IDs and is configured to generate new association IDs in a first scenario. For example, the network device only generates new association IDs when a specific configuration or area changes, rather than pre-generating association IDs for every possible scenario.

[0164] As an example, to reduce the number of preconfigured association IDs, association IDs may be reused based on geographically or topologically adjacent areas. For example, between multiple tracking area update (TAU) areas, certain association IDs may be reused in the same or similar network configurations, thereby reducing the number of association IDs.

[0165] As an example, the network device can classify and hierarchically manage the association IDs according to different network configuration levels. The network configuration levels may include core network configuration, RAN configuration, user plane configuration, etc. In this way, different ranges of association IDs can be allocated at different levels, reducing the number of association IDs required at each layer.

[0166] As an example, in a scenario where the number of association IDs is fixed, the length of the association ID representation is reduced to reduce signaling overhead. For example, using a shorter identifier or using a hash function to map a wide range of association IDs to shorter bits for representation can significantly reduce the size of the association ID during signaling transmission, thereby reducing the signaling load.

[0167] As an example, the network device triggers the sending of the first configuration information based on the second type of trigger event. That is, the network device can update the association ID in an event-triggered manner. The second type of trigger event can be a specific event such as a user location change, a service demand change, etc.

[0168] In some embodiments, the association ID is related to the cell where the terminal device is located. That is, the association ID indicated by the first configuration information is the cell association ID. The ID of each cell (e.g., a physical cell identity (PCI)) may correspond to the association ID of each cell. For example, the association ID may be just a number between 0 and N. For another example, the association ID may be associated with the PCI.

[0169] As an example, the association ID may be maintained within the NW. The network device may initiate a data collection session and assign an identifier (such as the association ID) to the data collection session, and the terminal device may collect data attributed to the association ID. For example, in response to a request from the terminal device, the network device may initiate a data collection session. In the request, the terminal device may indicate its preference for set A and set B.

[0170] As an example, the association ID may be adjusted according to the network demand and network load of the cell. The association ID corresponding to the first model is not static, but may be updated due to changes in the network environment to avoid prediction errors. When the association ID is updated, the first configuration information may be used by the terminal device to determine the updated association ID and determine whether to update the first model.

[0171] In some embodiments, the association ID is related to the transmitting device of the first model predicted beam. Taking the first model predicting the transmitting beam of the network device as an example, the association ID is related to the additional conditions on the network (NW) side, which helps to identify the beam transmitting device according to the association ID.

[0172] As an example, over time, as long as the same NW-side additional conditions are met, the network device can send the corresponding association ID to the terminal device, enabling the terminal device to collect more data. The terminal device side can collect data over time using the given association ID to train the first model. Once the first model is fully trained, the terminal device can notify the network device that the first model is ready for inference. Then, whenever the NW-side additional conditions in the inference process match the additional conditions in the training data collection process, the corresponding association ID will be signaled from the network device to the terminal device, so that the terminal device can use the corresponding trained first model for that specific association ID.

[0173] As an example, the additional conditions associated with the association ID may include NW-side proprietary information. The NW-side proprietary information may be related to beam transmission parameters and / or beam transmission resources. Exemplarily, the NW-side proprietary information may include 3dB beam width, beam aiming direction, beam angle, transmission filter, corresponding resource order (corresponding to beam) of the first beam set and the second beam set, Tx power, antenna height, antenna downward tilt parameters, and deployment scenarios, etc.

[0174] As an example, when the terminal device accesses, it can report one or more association IDs corresponding to the first model. In the scenario where the network device is a base station, if the terminal device accesses base station B, and base station B has different NW-side additional conditions from the last accessed base station A, the terminal device needs to report the association ID. Base station B needs to identify the NW-side additional conditions corresponding to the association ID. In some scenarios, different base stations can identify the additional conditions behind different association IDs. However, this may be difficult between different base station vendors.

[0175] In some embodiments, the association ID configured on the network side may have multiple corresponding relationships with the network device to adapt to different application scenarios.

[0176] As an example, the association ID is one of multiple association IDs, and the multiple association IDs correspond to multiple network devices one by one. For example, each base station is configured with a unique association ID so that the base station can be identified according to the association ID. In this example, multiple network devices can belong to the same vendor or different vendors. Multiple association IDs corresponding to the same vendor belong to an ID group.

[0177] In the above example, different base station vendors should use different association ID sets. The base station vendor can assign a base station-specific ID / ID set to each base station. If a base station receives an association ID from a terminal device and the association ID does not belong to the base station vendor's association ID set, the base station will not send the inference configuration of the first model on the terminal device side.

[0178] As an example, the association ID is one of a plurality of association IDs, and the plurality of association IDs correspond to a network device. For example, a plurality of association IDs may be configured in a cell, and therefore, the network device of the cell corresponds to the plurality of association IDs.

[0179] As an example, when the association ID is a set ID, the association ID may be a global association ID. The global association ID may correspond to the ID of a TAU area. Optionally, the association ID may correspond to one or more network devices within the first TAU ​​area. That is, multiple network devices within the first TAU ​​area are configured with the same association ID. It should be noted that a TAU area usually has one area ID. When multiple network devices correspond to one association ID, different network devices can be distinguished by the area ID and PCI of the TAU area.

[0180] In step S420, the terminal device performs model training and model inference on the first model according to the association ID. That is, the terminal device can perform the training and inference process of the first model based on the same association ID.

[0181] In some embodiments, after the NW sends the data collection related configuration and its associated ID, the terminal device collects data corresponding to the associated ID, and trains the first model based on the collected data corresponding to the associated ID at the terminal device. For example, the first beam set includes all measured beam information, and each beam, together with its measurement value and associated ID, constitutes input data of the first model. Based on the historical measurement data of the first beam set, the first model is established and trained to predict the future optimal beam of the second beam set.

[0182] In some embodiments, during the inference process, it is necessary to clarify the association IDs of the beams in the first beam set and the beams in the second beam set so as to ensure the correctness of the first model input and prediction. The association ID can be regarded as a bridge between different measurement value sets (such as the first beam set) and beam prediction result sets (such as the second beam set).

[0183] In some embodiments, the association ID needs to be maintained. For example, if the terminal device uses an association ID to collect data sets from the cell and train its model. After a long time, when the terminal device re-enters the cell, the NW may have changed its NW-related configuration and the mapping of the terminal device ID of each cell. However, when the ID value is the same, the terminal device still believes that its model is feasible. In this case, since the network configuration has changed, if the terminal device believes that the first model is still valid, it is possible that the prediction is wrong. Therefore, it is actually necessary to retrain or update the first model at this time.

[0184] Optionally, the change in network configuration may include a change in the downtilt angle of the antenna, a change in the number of antennas, a change in the coverage shape of the beam, a change in the beam structure, or other changes in network conditions.

[0185] Optionally, the first model can determine whether to update based on a first type of trigger event. The first type of trigger event includes one or more of the following: a change in the hash value bound to the association ID; a change in the network configuration corresponding to the association ID indicated by RRC signaling or determined by the terminal device; a change in the network configuration exceeding a set threshold; a change in the location and / or association ID of the terminal device after re-accessing the cell.

[0186] As an example, the configuration of the current network device can generate a hash value and bind it to the associated ID. The terminal device can store the hash value during training and compare it with the current hash value during inference. If a hash value change is found, that is, the configuration has changed, triggering the first model update.

[0187] As an example, the network device can explicitly notify the terminal device of configuration changes through RRC signaling and recommend the first model to be updated. This method can ensure that the terminal device can update the first model in a timely manner without relying on passive observation of the terminal device.

[0188] As an example, the terminal device can determine whether the network configuration corresponding to the associated ID has changed based on self-model verification, comparison of environmental information, etc. For example, the terminal device can perform self-model verification regularly, that is, detect whether there are obvious differences through changes in network performance during the reasoning process (such as throughput, latency, channel quality, etc.). If the reasoning result does not match the expected model performance, it can trigger retraining and updating of the first model. For another example, the terminal device can record network-related environmental information (such as signal strength, channel status, etc.) during training, and compare this information during reasoning. If the difference is large, it can be inferred that the network configuration may have changed, thereby triggering retraining or model updating.

[0189] As an example, the terminal device can determine whether the model needs to be fine-tuned based on the magnitude of the change in the network configuration. When the magnitude of the change in the network configuration exceeds or equals the set threshold, the first model is retrained and updated; when the magnitude of the change in the network configuration is lower than the set threshold, the first model is not retrained and updated for the time being. For example, for small changes in network conditions, the terminal device can fine-tune the model through online learning (incremental learning). For another example, for the change thresholds of certain key parameters (such as antenna downtilt angle, beam strength, etc.), when the changes are within a certain range, the terminal device only needs to make fine adjustments instead of completely rebuilding the model.

[0190] As an example, the network device can classify the degree of change of the network configuration and notify the terminal device through RRC signaling. For example, when the network configuration changes slightly, the network device can mark "minor change" to indicate that the terminal device only needs to perform local model fine-tuning. When the network configuration changes significantly, the network device can mark "significant change" to instruct the terminal device to perform a complete model update or retraining.

[0191] As an example, the terminal device may determine whether to update the first model based on changes in its own position and / or changes in the associated ID. For example, the associated ID corresponds to the beam ID information. If the position of the terminal device has not changed over a period of time or is still near the original position after re-accessing the cell, and corresponds to the same beam ID information, the terminal device does not trigger the first model update. If the terminal device is still near the original position after re-accessing the cell, but has different beam ID information, the terminal device triggers the first model update. If the terminal device is not near the original position after re-accessing the cell and has different beam ID information, the terminal device triggers the first model update; if the terminal device is not near the original position when re-accessing the cell and has the same beam ID information, the terminal device does not trigger the first model update.

[0192] As an example, the terminal device can update the first model in real time or dynamically. The update of the first model may include a local parameter update of the first model. For example, when the beam shape changes, the terminal device can collect inference data in real time and update the model parameters locally without large-scale model retraining. For another example, the terminal device can dynamically adjust the model according to changes in physical conditions in the environment (such as changes in signal strength, interference levels, etc.). In this case, the terminal device does not rely entirely on the configuration notification of the network device, but automatically adjusts the model parameters to adapt to the new network conditions by sensing and measuring the local environment.

[0193] As an example, the terminal device may be one of multiple terminal devices participating in federated learning. When the first model is a local model in federated learning, the terminal device may send update information of the first model to other terminal devices in the multiple terminal devices. For example, the terminal device may share updates of its local model with the network end (or other terminal devices) through federated learning so as to perform collaborative learning when network conditions change. In this way, multiple terminal devices can collaboratively update the model under different network conditions, avoiding a single terminal device from taking on all model update work, while improving the efficiency of model fine-tuning.

[0194] Combination of the above Figure 4The present invention introduces an embodiment of a method for using an association ID to maintain consistency in the training and reasoning phases. As can be seen from the foregoing, the association ID can be indicated by resource configuration information or a trigger list associated with the first report. Figures 5 to 10 In the illustrated embodiment, the respective first configuration information may be sent in a variety of ways.

[0195] Figures 5 to 7 Indicates multiple possible ways to configure the first beam set (set B) and the first beam set (set A) association ID in CSI-ResourceConfig. This configuration method is configured based on the CSI-RS resource index for model reasoning. Figures 5 to 7 In the example, the number of reports is determined by the parameter "predicted_cri" or "predicted_cri_rsrp".

[0196] See also Figure 5 , the first beam set and the second beam set are configured in the same CSI resource configuration parameter (CSI-ResourceConfig) in CSI-ReportConfig, and the association ID is also in the resource configuration parameter. Figure 6 , the first beam set and the second beam set are configured as different CSI resource sets in CSI-ReportConfig, that is, different CSI-ResourceConfig; but the two resource sets share an association ID indicated in CSI-ReportConfig. Figure 6 compared to, Figure 7 The first beam set and the second beam set in the are configured with different CSI resource sets, and each resource set indicates an association ID.

[0197] Depend on Figures 5 to 7 It can be seen that Figure 5 and Figure 6 The association ID in is shared between the first beam set and the second beam set. Figure 6 and Figure 7 , the resource sets of the first beam set and the second beam set are configured separately, but Figure 7 The two association IDs in correspond to different CSI-ResourceConfigIDs (i.e., CSI-ResourceConfigA and CSI-ResourceConfigB). Figure 6 One of the associated IDs is configured in CSI-ReportConfig. Figure 6 and Figure 7 Different configuration methods, Figure 6Each CSI-ResourceConfigID in can construct a CSI-ResourceConfig, or can be configured in different CSI-ReportConfigs; Figure 7 The association ID in is configured in each CSI-ResourceConfigID, so different association IDs do not need to be additionally indicated in CSI-ReportConfig, which is more beneficial in terms of reporting configuration overhead.

[0198] Figures 8 to 9 Indicates multiple possible ways to configure the association ID when triggering aperiodic (AP) reporting. Figure 8 and Fig. 9 Two examples of how to indicate the correlation ID in multiple reports when a set of non-periodic reports is triggered.

[0199] Figure 8 It shows that in the CSI framework, after the CSI-ReportConfig is configured, it can be further grouped in the CSI-AperiodicTriggerStateList. CSI-AperiodicTriggerStateList is a list of CSI. In each CSI non-periodic trigger state, there is an associated ReportConfigInfoList (i.e. associatedReportConfigInfoList). This list contains an associated ReportConfigInfo list, and each associated ReportConfigInfo contains the CSI-ReportConfigID and the related resource set (ResourceSet) and quasi co-location (QCL) information used by this CSI-ReportConfig.

[0200] Optionally, the NW may use DCI to indicate the CSI-AperiodicTriggerState in the CSI-AperiodicTriggerStateList, and then all CSI-ReportConfig information contained in the CSI-AperiodisTriggerState will be triggered.

[0201] To indicate the association ID of the report, the association ID can be configured in CSI-AperiodicTriggerState. For example, the association ID can be configured directly under associatedReportConfigInfoList, and the same association ID can be shared between multiple CSI-ReportConfigs, such as Figure 8 Alternatively, the association ID can be configured in each associated ReportConfigInfo, so each CSI-ReportConfig has its own dedicated association ID, such as Fig. 9 shown.

[0202] Fig.10 The figure shows the entire interaction process between the NW and the UE in the current non-periodic (AP) report triggering method. In step S1010, the association ID is configured together with the RRC signal that configures the CSI-AperiodicTriggerStateList. The CSI-ReportConfig does not change between different association IDs. In step S1020, the DCI can trigger an AP report, such as CSI-AperiodicTriggerState. The NW can indicate the association ID when triggering a non-periodic report. For all alternatives for configuring the association ID within the CSI framework, the NW may configure reports that require different association IDs at the same time. In step S1030, if the UE is triggered by these reports, the UE performs measurements and reports that trigger the AP report. Fig.10 As shown, there is a model switching / backoff time between step S1020 and step S1030. The terminal device may need to run multiple models simultaneously or frequently switch between models to use the correct model to generate the corresponding beam report.

[0203] The above describes a method for indicating the association ID according to the resource configuration information or the report trigger list. The following describes a method for configuring the first beam set and the second beam set resources of the network device by taking the beam carrying CSI-RS as an example.

[0204] In some embodiments, for the terminal device side model, the network device needs to configure the first beam set for measurement and the second beam set for prediction and reporting. The two beam sets can have multiple association modes. Optionally, the network device can provide 4 options to configure the association between the beams in the first beam set and the beams in the second beam set for inference result reporting.

[0205] As an example, when the first beam set and the second beam set are used to send CSI-RS, the resources of the first beam set are indicated by the first parameter, and the resources of the second beam set are indicated by the second parameter. The first parameter and the second parameter are one of the following: the first parameter and the second parameter are different CSI-ResourceConfigIDs; the first parameter and the second parameter are the same CSI-ResourceConfigID; the first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; the first parameter is a dedicated information element (IE) of the first beam set, and the second parameter is CSI-RS-ResourceSetList.

[0206] As an embodiment, the network device configures a CSI-ResourceConfigID for the first beam set and the second beam set respectively (eg Figure 6 or Figure 7 ). The first beam set and the second beam set require separate resource sets to distinguish between resources used for measurement and resources used for prediction. The terminal device needs to report the measurement results of the second beam set as well as the first beam set, and the network device can evaluate whether the prediction results of the second beam set are accurate based on the measurement results of the first beam set. The network device can also decide whether to use the measurement results of the first beam set or the measurement results of the second beam set after machine learning. The terminal device needs to associate the two CSI-ReportConfigs together to understand which is the beam set used for reporting.

[0207] As an embodiment, the network device configures a CSI-ResourceConfigID (such as Figure 5 ), that is, the resource sets of the first beam set and the second beam set are configured in the same CSI-ResourceConfigId. Although it is configured in one CSI-ResourceConfigID, it can support the report configuration of the two beam sets. For example, the first beam set and the second beam set can use different CSI-ReportSets, respectively. In other words, separate resource sets are configured for the first beam set and the second beam set, and the terminal device can easily distinguish between the resources used for measurement and the resources used for prediction.

[0208] As an embodiment, the network device configures a CSI-ResourceConfigID for the first beam set; the second beam set is configured by multiple resource sets and can be a virtual set. This can be achieved by simply expanding the number of configurable resource sets in the resource setting. In this embodiment, the second beam set cannot be configured through CSI resources / ResourceSet / ResourceConfig because these IEs are defined based on CSI resources, in which the physical mapping of resources is specified. The second beam set is always configured with a corresponding resource ID / resource set ID / ResourceConfigID for training and performance evaluation. The second beam set can also be determined from the associated function / association ID.

[0209] As an embodiment, the second beam set may be configured through the existing CSI-RS-ResourceSetList. The first beam set may be configured through a new IE, which may be in the form of a bitmap, a bitmap ID, a group of RS IDs or a resource set ID.

[0210] Combination of the above Figures 1 to 10 , describes in detail the method embodiment of the present application. Figures 11 to 13 , describes the device embodiment of the present application in detail. It should be understood that the description of the device embodiment corresponds to the description of the method embodiment, so the parts not described in detail can refer to the previous method embodiment.

[0211] Fig.11 1100 is a schematic block diagram of a device for wireless communication according to an embodiment of the present application. The device 1100 may be any terminal device described above. Fig.11 The device 1100 shown includes a transceiver unit 1110 and a processing unit 1120 .

[0212] The transceiver unit 1110 can be used to receive first configuration information, where the first configuration information is used to indicate an associated ID; the determination unit 1120 can be used to perform model training and model inference on the first model according to the associated ID; wherein the first model corresponds to a first beam set and a second beam set, and the associated ID is one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set.

[0213] Optionally, the association ID is one of a plurality of association IDs, and the plurality of association IDs correspond one-to-one to a plurality of network devices.

[0214] Optionally, the association ID is a set ID, and the association ID corresponds to one or more network devices in the first TAU ​​area.

[0215] Optionally, the association ID is one of multiple association IDs, and the multiple association IDs correspond to one network device.

[0216] Optionally, the association ID is related to the cell where the terminal device is located, and the association ID is adjusted according to the network demand and network load of the cell.

[0217] Optionally, the association ID is a set ID, and the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

[0218] Optionally, the first configuration information is carried in one of the following information: resource configuration information shared by the first beam set and the second beam set; a configuration report shared by the first beam set and the second beam set; first resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

[0219] Optionally, the transceiver unit 1110 is further used to send a first report based on the inference result of the first model; wherein the association ID is a set ID, and the first configuration information is carried in a trigger list related to the first report.

[0220] Optionally, the first configuration information is carried in one of the following trigger lists: a total list of related report configuration information; a first list of related report configuration information corresponding to the first beam set and a second list of related report configuration information corresponding to the second beam set.

[0221] Optionally, the association ID is a set ID, and the association ID is carried in RRC signaling.

[0222] Optionally, the association ID is a beam ID group, the beam ID group is represented by a beam association ID matrix, and the beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

[0223] Optionally, the beam association ID matrix is ​​determined according to beam resource indexes of the first beam set and the second beam set.

[0224] Optionally, the association ID determines whether to adjust based on one or more of the following information: the time interval between model training and model reasoning; the channel time variability corresponding to the model training and model reasoning; whether the beam characteristics of the first beam set and the second beam set are stable.

[0225] Optionally, the first beam set and the second beam set are used to send CSI-RS, the resources of the first beam set are indicated by a first parameter, and the resources of the second beam set are indicated by a second parameter, and the first parameter and the second parameter are one of the following: the first parameter and the second parameter are different CSI-ResourceConfigIDs; the first parameter and the second parameter are the same CSI-ResourceConfigID; the first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; the first parameter is a dedicated information element for the first beam set, and the second parameter is CSI-RS-ResourceSetList.

[0226] Optionally, the first model determines whether to update based on a first type of trigger event, and the first type of trigger event includes one or more of the following: a change in the hash value bound to the associated ID; an RRC signaling indication or terminal device determination that the network configuration corresponding to the associated ID has changed; the change in the network configuration exceeds a set threshold; the location of the terminal device changes and / or the associated ID changes after re-accessing the cell.

[0227] Optionally, the first model is a local model in federated learning, the terminal device is one of multiple terminal devices participating in federated learning, and the transceiver unit is also used to send update information of the first model to other terminal devices among the multiple terminal devices.

[0228] Optionally, the first model is an artificial intelligence or machine learning model.

[0229] Fig.12 12 is a schematic block diagram of another apparatus for wireless communication according to an embodiment of the present application. The apparatus 1200 may be any network device described above. Fig.12 The illustrated device 1200 includes a transceiver unit 1210 .

[0230] The transceiver unit 1210 can be used to send first configuration information, where the first configuration information is used to indicate an association ID; wherein the association ID is used by the terminal device to perform model training and model inference on the first model, and the first model corresponds to the first beam set and the second beam set, and the association ID is one or more of the following: a set ID for selecting the first beam set and / or the second beam set; a beam ID group, including the ID of each beam in the first beam set and / or the second beam set.

[0231] Optionally, the association ID is one of a plurality of association IDs, and the plurality of association IDs correspond one-to-one to a plurality of network devices.

[0232] Optionally, the association ID is a set ID, and the association ID corresponds to one or more network devices in the first TAU ​​area.

[0233] Optionally, the association ID is one of multiple association IDs, and the multiple association IDs correspond to one network device.

[0234] Optionally, the association ID is related to the cell where the terminal device is located, and the association ID is adjusted according to the network demand and network load of the cell.

[0235] Optionally, the association ID is a set ID, and the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

[0236] Optionally, the first configuration information is carried in one of the following information: resource configuration information shared by the first beam set and the second beam set; a configuration report shared by the first beam set and the second beam set; first resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

[0237] Optionally, the transceiver unit 1210 is also used to receive a first report sent by the terminal device based on the inference result of the first model; wherein the association ID is a set ID, and the first configuration information is carried in a trigger list related to the first report.

[0238] Optionally, the first configuration information is carried in one of the following trigger lists: a total list of related report configuration information; a first list of related report configuration information corresponding to the first beam set and a second list of related report configuration information corresponding to the second beam set.

[0239] Optionally, the association ID is a set ID, and the association ID is carried in RRC signaling.

[0240] Optionally, the association ID is a beam ID group, the beam ID group is represented by a beam association ID matrix, and the beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

[0241] Optionally, the beam association ID matrix is ​​determined according to beam resource indexes of the first beam set and the second beam set.

[0242] Optionally, the association ID determines whether to adjust based on one or more of the following information: the time interval between model training and model reasoning; the channel time variability corresponding to the model training and model reasoning; whether the beam characteristics of the first beam set and the second beam set are stable.

[0243] Optionally, the first beam set and the second beam set are used to send CSI-RS, the resources of the first beam set are indicated by a first parameter, and the resources of the second beam set are indicated by a second parameter, and the first parameter and the second parameter are one of the following: the first parameter and the second parameter are different CSI-ResourceConfigIDs; the first parameter and the second parameter are the same CSI-ResourceConfigID; the first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; the first parameter is a dedicated information element for the first beam set, and the second parameter is CSI-RS-ResourceSetList.

[0244] Optionally, the first model determines whether to update based on a first type of trigger event, and the first type of trigger event includes one or more of the following: a change in the hash value bound to the associated ID; an RRC signaling indication or terminal device determination that the network configuration corresponding to the associated ID has changed; the change in the network configuration exceeds a set threshold; the location of the terminal device changes and / or the associated ID changes after re-accessing the cell.

[0245] Optionally, the first model is a local model in federated learning, the terminal device is one of multiple terminal devices participating in the federated learning, and update information of the first model is sent to other terminal devices among the multiple terminal devices except the terminal device.

[0246] Optionally, the first model is an artificial intelligence or machine learning model.

[0247] Fig.13 Shown is a schematic diagram of the structure of a communication device according to an embodiment of the present application. Fig.13 The dotted line in the figure indicates that the unit or module is optional. The device 1300 can be used to implement the method described in the above method embodiment. The device 1300 can be a chip, a terminal device or a network device.

[0248] The device 1300 may include one or more processors 1310. The processor 1310 may support the device 1300 to implement the method described in the method embodiment above. The processor 1310 may be a general-purpose processor or a special-purpose processor. For example, the processor may be a central processing unit (CPU). Alternatively, the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0249] The apparatus 1300 may further include one or more memories 1320. The memory 1320 stores a program, which can be executed by the processor 1310, so that the processor 1310 executes the method described in the above method embodiment. The memory 1320 may be independent of the processor 1310 or integrated in the processor 1310.

[0250] The apparatus 1300 may further include a transceiver 1330. The processor 1310 may communicate with other devices or chips through the transceiver 1330. For example, the processor 1310 may transmit and receive data with other devices or chips through the transceiver 1330.

[0251] The present application also provides a computer-readable storage medium for storing a program. The computer-readable storage medium can be applied to a terminal device or a network device provided in the present application, and the program enables a computer to execute the method performed by the terminal device or the network device in each embodiment of the present application.

[0252] The computer-readable storage medium may be any available medium that can be read by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid state disk (SSD)).

[0253] The embodiment of the present application also provides a computer program product. The computer program product includes a program. The computer program product can be applied to the terminal device or network device provided in the embodiment of the present application, and the program enables the computer to execute the method performed by the terminal device or network device in each embodiment of the present application.

[0254] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function 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 a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0255] The embodiment of the present application also provides a computer program. The computer program can be applied to the terminal device or network device provided in the embodiment of the present application, and the computer program enables a computer to execute the method executed by the terminal or network device in each embodiment of the present application.

[0256] The terms "system" and "network" in this application can be used interchangeably. In addition, the terms used in this application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application. The terms "first", "second", "third" and "fourth" in the specification and claims of this application and the accompanying drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions.

[0257] In the embodiments of the present application, the "indication" mentioned can be a direct indication, an indirect indication, or an indication of an association relationship. For example, A indicates B, which can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association relationship between A and B.

[0258] In the embodiments of the present application, the term "corresponding" may indicate that there is a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship of indication and being indicated, configuration and being configured, etc.

[0259] In the embodiments of the present application, "pre-definition" or "pre-configuration" can be implemented by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in a device (for example, including a terminal device and a network device), and the present application does not limit the specific implementation method. For example, pre-definition can refer to what is defined in the protocol.

[0260] In the embodiments of the present application, the “protocol” may refer to a standard protocol in the communication field, for example, it may include an LTE protocol, an NR protocol, and related protocols used in future communication systems, and the present application does not limit this.

[0261] In the embodiments of the present application, determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0262] In the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0263] In the embodiments of the present application, the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0264] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0265] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0266] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0267] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for wireless communication, characterized in that: include: The terminal device receives first configuration information, where the first configuration information is used to indicate an association identifier ID; The terminal device performs model training and model inference on the first model according to the association ID; The first model corresponds to the first beam set and the second beam set, and the association ID is one or more of the following: A set ID for selecting the first beam set and / or the second beam set; A beam ID group includes an ID of each beam in the first beam set and / or the second beam set.

2. The method according to claim 1, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond one-to-one to a plurality of network devices.

3. The method according to claim 1, characterized in that The association ID is a set ID, and the association ID corresponds to the first tracking area to update one or more network devices in the TAU area.

4. The method according to claim 1, characterized in that: The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond to a network device.

5. The method according to any one of claims 1 to 4, characterized in that The association ID is related to the cell where the terminal device is located, and the association ID is adjusted according to the network demand and network load of the cell.

6. The method according to any one of claims 1 to 5, characterized in that The association ID is the set ID, and the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

7. The method according to claim 6, characterized in that The first configuration information is carried in one of the following information: Resource configuration information shared by the first beam set and the second beam set; a configuration report common to the first beam set and the second beam set; First resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

8. The method according to any one of claims 1 to 5, characterized in that The method further comprises: The terminal device sends a first report according to the inference result of the first model; The association ID is the set ID, and the first configuration information is carried in a trigger list related to the first report.

9. The method according to claim 8, characterized in that The first configuration information is carried in one type of information of the following trigger list: A general list of relevant report configuration information; A first related report configuration information list corresponding to the first beam set and a second related report configuration information list corresponding to the second beam set.

10. The method according to any one of claims 1 to 5, characterized in that The association ID is the set ID, and the association ID is carried in radio resource control RRC signaling.

11. The method according to any one of claims 1 to 5, characterized in that The association ID is the beam ID group, and the beam ID group is represented by a beam association ID matrix. The beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

12. The method according to claim 11, characterized in that The beam association ID matrix is ​​determined according to beam resource indexes of the first beam set and the second beam set.

13. The method according to any one of claims 1 to 12, characterized in that The association ID determines whether to adjust based on one or more of the following information: The time interval between the model training and the model inference; Channel time variability corresponding to the model training and the model reasoning; Whether beam characteristics of the first beam set and the second beam set are stable.

14. The method according to any one of claims 1 to 13, characterized in that The first beam set and the second beam set are used to send a channel state information reference signal CSI-RS, resources of the first beam set are indicated by a first parameter, and resources of the second beam set are indicated by a second parameter, and the first parameter and the second parameter are one of the following: The first parameter and the second parameter are different CSI-ResourceConfigIDs; The first parameter and the second parameter are the same CSI-ResourceConfigID; The first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; The first parameter is a dedicated information element of the first beam set, and the second parameter is CSI-RS-ResourceSetList.

15. The method according to any one of claims 1 to 14, characterized in that The first model determines whether to update based on a first type of triggering event, where the first type of triggering event includes one or more of the following: The hash value bound to the association ID changes; RRC signaling indicates or the terminal device determines that the network configuration corresponding to the association ID has changed; The change in network configuration exceeds the set threshold; The location change and / or association ID change of the terminal device after re-accessing the cell.

16. The method according to claim 15, characterized in that The first model is a local model in federated learning, the terminal device is one of a plurality of terminal devices participating in the federated learning, and the method further includes: The terminal device sends update information of the first model to other terminal devices among the multiple terminal devices.

17. The method according to any one of claims 1 to 16, characterized in that The first model is an artificial intelligence or machine learning model.

18. A method for wireless communication, characterized in that: include: The network device sends first configuration information, where the first configuration information is used to indicate an association identifier ID; The association ID is used by the terminal device to perform model training and model inference on the first model, the first model corresponds to the first beam set and the second beam set, and the association ID is one or more of the following: A set ID for selecting the first beam set and / or the second beam set; A beam ID group includes an ID of each beam in the first beam set and / or the second beam set.

19. The method according to claim 18, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond one-to-one to a plurality of network devices.

20. The method according to claim 18, characterized in that The association ID is a set ID, and the association ID corresponds to the first tracking area to update one or more network devices in the TAU area.

21. The method according to claim 18, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond to a network device.

22. The method according to any one of claims 18 to 21, characterized in that The association ID is related to the cell where the terminal device is located, and the association ID is adjusted according to the network demand and network load of the cell.

23. The method according to any one of claims 18 to 22, characterized in that The association ID is the set ID, and the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

24. The method according to claim 23, characterized in that The first configuration information is carried in one of the following information: Resource configuration information shared by the first beam set and the second beam set; a configuration report common to the first beam set and the second beam set; First resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

25. The method according to any one of claims 18 to 22, characterized in that The method further comprises: The network device receives a first report sent by the terminal device according to the inference result of the first model; The association ID is the set ID, and the first configuration information is carried in a trigger list related to the first report.

26. The method according to claim 25, characterized in that The first configuration information is carried in one type of information of the following trigger list: A general list of relevant report configuration information; A first related report configuration information list corresponding to the first beam set and a second related report configuration information list corresponding to the second beam set.

27. The method according to any one of claims 18 to 22, characterized in that The association ID is the set ID, and the association ID is carried in radio resource control RRC signaling.

28. The method according to any one of claims 18 to 22, characterized in that The association ID is the beam ID group, and the beam ID group is represented by a beam association ID matrix. The beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

29. The method according to claim 28, characterized in that The beam association ID matrix is ​​determined according to beam resource indexes of the first beam set and the second beam set.

30. The method according to any one of claims 18 to 29, characterized in that in, The association ID determines whether to adjust based on one or more of the following information: The time interval between the model training and the model inference; Channel time variability corresponding to the model training and the model reasoning; Whether beam characteristics of the first beam set and the second beam set are stable.

31. The method according to any one of claims 18 to 30, characterized in that The first beam set and the second beam set are used to send a channel state information reference signal CSI-RS, resources of the first beam set are indicated by a first parameter, and resources of the second beam set are indicated by a second parameter, and the first parameter and the second parameter are one of the following: The first parameter and the second parameter are different CSI-ResourceConfigIDs; The first parameter and the second parameter are the same CSI-ResourceConfigID; The first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; The first parameter is a dedicated information element of the first beam set, and the second parameter is CSI-RS-ResourceSetList.

32. The method according to any one of claims 18 to 31, characterized in that The first model determines whether to update based on a first type of triggering event, where the first type of triggering event includes one or more of the following: The hash value bound to the association ID changes; RRC signaling indicates or the terminal device determines that the network configuration corresponding to the association ID has changed; The change in network configuration exceeds the set threshold; The location change and / or association ID change of the terminal device after re-accessing the cell.

33. The method according to claim 32, characterized in that The first model is a local model in federated learning, the terminal device is one of multiple terminal devices participating in the federated learning, and update information of the first model is sent to other terminal devices among the multiple terminal devices except the terminal device.

34. The method according to any one of claims 18 to 33, characterized in that The first model is an artificial intelligence or machine learning model.

35. A device for wireless communication, characterized in that: The device is a terminal device, and the device includes: A transceiver unit, configured to receive first configuration information, where the first configuration information is used to indicate an association identifier ID; A determination unit, configured to perform model training and model inference on the first model according to the association ID; The first model corresponds to the first beam set and the second beam set, and the association ID is one or more of the following: A set ID for selecting the first beam set and / or the second beam set; A beam ID group includes an ID of each beam in the first beam set and / or the second beam set.

36. The device according to claim 35, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond one-to-one to a plurality of network devices.

37. The device according to claim 35, characterized in that The association ID is a set ID, and the association ID corresponds to the first tracking area to update one or more network devices in the TAU area.

38. The device according to claim 35, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond to a network device.

39. The device according to any one of claims 35 to 38, characterized in that The association ID is related to the cell where the terminal device is located, and the association ID is adjusted according to the network demand and network load of the cell.

40. The device according to any one of claims 35 to 39, characterized in that The association ID is the set ID, and the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

41. The device according to claim 40, characterized in that The first configuration information is carried in one of the following information: Resource configuration information shared by the first beam set and the second beam set; a configuration report common to the first beam set and the second beam set; First resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

42. The device according to any one of claims 35 to 39, characterized in that The transceiver unit is also used to send a first report according to the inference result of the first model; wherein the association ID is the set ID, and the first configuration information is carried in a trigger list related to the first report.

43. The device according to claim 42, characterized in that The first configuration information is carried in one type of information of the following trigger list: A general list of relevant report configuration information; A first related report configuration information list corresponding to the first beam set and a second related report configuration information list corresponding to the second beam set.

44. The device according to any one of claims 35 to 39, characterized in that The association ID is the set ID, and the association ID is carried in radio resource control RRC signaling.

45. The device according to any one of claims 35 to 39, characterized in that The association ID is the beam ID group, and the beam ID group is represented by a beam association ID matrix. The beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

46. ​​The device according to claim 45, characterized in that The beam association ID matrix is ​​determined according to beam resource indexes of the first beam set and the second beam set.

47. The device according to any one of claims 35 to 46, characterized in that The association ID determines whether to adjust based on one or more of the following information: The time interval between the model training and the model inference; Channel time variability corresponding to the model training and the model reasoning; Whether beam characteristics of the first beam set and the second beam set are stable.

48. The device according to any one of claims 35 to 47, characterized in that The first beam set and the second beam set are used to send a channel state information reference signal CSI-RS, resources of the first beam set are indicated by a first parameter, and resources of the second beam set are indicated by a second parameter, and the first parameter and the second parameter are one of the following: The first parameter and the second parameter are different CSI-ResourceConfigIDs; The first parameter and the second parameter are the same CSI-ResourceConfigID; The first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; The first parameter is a dedicated information element of the first beam set, and the second parameter is CSI-RS-ResourceSetList.

49. The device according to any one of claims 35 to 48, characterized in that The first model determines whether to update based on a first type of triggering event, where the first type of triggering event includes one or more of the following: The hash value bound to the association ID changes; RRC signaling indicates or the terminal device determines that the network configuration corresponding to the association ID has changed; The change in network configuration exceeds the set threshold; The location change and / or association ID change of the terminal device after re-accessing the cell.

50. The device according to claim 49, characterized in that The first model is a local model in federated learning, the terminal device is one of multiple terminal devices participating in the federated learning, and the transceiver unit is also used to send update information of the first model to other terminal devices among the multiple terminal devices.

51. The device according to any one of claims 35 to 50, characterized in that The first model is an artificial intelligence or machine learning model.

52. A device for wireless communication, characterized in that: The device is a network device, and the device includes: A transceiver unit, configured to send first configuration information, where the first configuration information is used to indicate an association identifier ID; The association ID is used by the terminal device to perform model training and model inference on the first model, the first model corresponds to the first beam set and the second beam set, and the association ID is one or more of the following: A set ID for selecting the first beam set and / or the second beam set; A beam ID group includes an ID of each beam in the first beam set and / or the second beam set.

53. The device according to claim 52, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond one-to-one to a plurality of network devices.

54. The device according to claim 52, characterized in that The association ID is a set ID, and the association ID corresponds to the first tracking area to update one or more network devices in the TAU area.

55. The device according to claim 52, characterized in that The association ID is one of a plurality of association IDs, and the plurality of association IDs correspond to a network device.

56. The device according to any one of claims 52 to 55, characterized in that The association ID is related to the cell where the terminal device is located, and the association ID is adjusted according to the network demand and network load of the cell.

57. The device according to any one of claims 52 to 56, characterized in that The association ID is the set ID, and the first configuration information is carried in resource configuration information of the first beam set and / or the second beam set.

58. The device according to claim 57, characterized in that The first configuration information is carried in one of the following information: Resource configuration information shared by the first beam set and the second beam set; a configuration report common to the first beam set and the second beam set; First resource configuration information corresponding to the first beam set and second resource configuration information corresponding to the second beam set.

59. The device according to any one of claims 52 to 56, characterized in that The transceiver unit is further configured to receive a first report sent by the terminal device according to the inference result of the first model; The association ID is the set ID, and the first configuration information is carried in a trigger list related to the first report.

60. The device according to claim 59, characterized in that The first configuration information is carried in one type of information of the following trigger list: A general list of relevant report configuration information; A first related report configuration information list corresponding to the first beam set and a second related report configuration information list corresponding to the second beam set.

61. The device according to any one of claims 52 to 56, characterized in that The association ID is the set ID, and the association ID is carried in radio resource control RRC signaling.

62. The device according to any one of claims 52 to 56, characterized in that The association ID is the beam ID group, and the beam ID group is represented by a beam association ID matrix. The beam association ID matrix is ​​used to indicate a mapping relationship between the first beam set and the second beam set.

63. The device according to claim 62, characterized in that The beam association ID matrix is ​​determined according to beam resource indexes of the first beam set and the second beam set.

64. The device according to any one of claims 52 to 63, characterized in that The association ID determines whether to adjust based on one or more of the following information: The time interval between the model training and the model inference; Channel time variability corresponding to the model training and the model reasoning; Whether beam characteristics of the first beam set and the second beam set are stable.

65. The device according to any one of claims 52 to 64, characterized in that The first beam set and the second beam set are used to send a channel state information reference signal CSI-RS, resources of the first beam set are indicated by a first parameter, and resources of the second beam set are indicated by a second parameter, and the first parameter and the second parameter are one of the following: The first parameter and the second parameter are different CSI-ResourceConfigIDs; The first parameter and the second parameter are the same CSI-ResourceConfigID; The first parameter is CSI-ResourceConfigID, and the second parameter is a configuration parameter of multiple resource sets; The first parameter is a dedicated information element of the first beam set, and the second parameter is CSI-RS-ResourceSetList.

66. The device according to any one of claims 52 to 65, characterized in that The first model determines whether to update based on a first type of triggering event, where the first type of triggering event includes one or more of the following: The hash value bound to the association ID changes; RRC signaling indicates or the terminal device determines that the network configuration corresponding to the association ID has changed; The change in network configuration exceeds the set threshold; The location change and / or association ID change of the terminal device after re-accessing the cell.

67. The device according to claim 66, characterized in that The first model is a local model in federated learning, the terminal device is one of multiple terminal devices participating in the federated learning, and update information of the first model is sent to other terminal devices among the multiple terminal devices except the terminal device.

68. The device according to any one of claims 52 to 67, characterized in that The first model is an artificial intelligence or machine learning model.

69. A communication device, characterized in that: The invention comprises a memory and a processor, wherein the memory is used to store a program, and the processor is used to call the program in the memory to execute the method as claimed in any one of claims 1 to 34.

70. A device, characterized in that The invention comprises a processor, which is used to call a program from a memory to execute the method as claimed in any one of claims 1 to 34.

71. A chip, characterized in that: It comprises a processor, which is used to call a program from a memory, so that a device equipped with the chip executes a method as claimed in any one of claims 1 to 34.

72. A computer-readable storage medium, characterized in that A program is stored thereon, the program causing a computer to execute the method according to any one of claims 1 to 34.

73. A computer program product, characterized in that The method comprises a program which causes a computer to execute the method according to any one of claims 1 to 34.

74. A computer program, characterized in that The computer program causes a computer to execute the method according to any one of claims 1 to 34.

Images