Wireless communication method, terminal device and network device

Abstract

The embodiment of the invention provides a wireless communication method, terminal equipment and network equipment, which can realize AI-based wave beam time domain prediction. The wireless communication method comprises: a network device obtaining a first data set, the first data set comprising identification information of spatial filters of K1 time units and link quality information of the K1 time units, and K1 being a positive integer; the network equipment inputs the first data set into a first network model and outputs first prediction information, the first prediction information comprises identification information of spatial filters of F1 time units and link quality information of the F1 time units, at least one check point is set in the F1 time units, and F1 is a positive integer; the network equipment determines whether the spatial filter output by the first network model is used in n time units after the first check point according to the link quality check result of the first check point, the first check point belongs to at least one check point, n is a positive integer, and n is smaller than F1.

Description

Wireless communication method, terminal device and network device Technical Field

[0001] The embodiments of the present application relate to the field of communications, and more specifically, to a wireless communication method, terminal device, and network device. Background Art

[0002] New Radio (NR) systems introduce a beam management mechanism. Downlink beam management includes downlink beam scanning, optimal beam reporting by the terminal, and downlink beam indication by the network. However, when a terminal device is in rapid motion, frequent beam switching is required, frequently triggering the beam management process and incurring significant overhead and latency.

[0003] Summary of the Invention

[0004] The embodiments of the present application provide a wireless communication method, terminal device, and network device, which can realize AI-based beam time domain prediction, reduce beam management overhead, and ensure the stability of link quality.

[0005] In a first aspect, a wireless communication method is provided, the method comprising:

[0006] The network device obtains a first data set, wherein the first data set includes identification information of a spatial filter of K1 time units and link quality information of K1 time units, where K1 is a positive integer;

[0007] The network device inputs the first data set into a first network model and outputs first prediction information, wherein the first prediction information includes identification information of a spatial filter of F1 time units and link quality information of F1 time units, wherein at least one check point is set in the F1 time unit, and F1 is a positive integer;

[0008] The network device determines whether to use the spatial filter output by the first network model within n time units after the first check point based on the link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F1.

[0009] In a second aspect, a wireless communication method is provided, the method comprising:

[0010] The terminal device obtains a third data set, where the third data set includes identification information of the spatial filter of K3 time units and link quality information of K3 time units, where K3 is a positive integer;

[0011] The terminal device inputs the third data set into the second network model and outputs second prediction information, wherein the second prediction information includes identification information of the spatial filter and link quality information of F3 time units, wherein at least one check point is set in the F3 time units, and F3 is a positive integer;

[0012] The terminal device determines whether to use the spatial filter output by the second network model within n time units after the first check point based on the link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F3.

[0013] In a third aspect, a network device is provided for executing the method in the first aspect.

[0014] Specifically, the network device includes a functional module for executing the method in the above-mentioned first aspect.

[0015] In a fourth aspect, a terminal device is provided for executing the method in the second aspect.

[0016] Specifically, the terminal device includes a functional module for executing the method in the above-mentioned second aspect.

[0017] In a fifth aspect, a network device is provided, comprising a processor and a memory; the memory is used to store computer programs, and the processor is used to call and run the computer programs stored in the memory, so that the network device executes the method in the above-mentioned first aspect.

[0018] In a sixth aspect, a terminal device is provided, comprising a processor and a memory; the memory is used to store computer programs, and the processor is used to call and run the computer programs stored in the memory, so that the terminal device executes the method in the above-mentioned second aspect.

[0019] In a seventh aspect, a device is provided for implementing the method in any one of the first to second aspects above.

[0020] Specifically, the apparatus includes: a processor, configured to call and run a computer program from a memory, so that a device equipped with the apparatus executes the method in any one of the first to second aspects described above.

[0021] In an eighth aspect, a computer-readable storage medium is provided for storing a computer program, wherein the computer program enables a computer to execute the method in any one of the first to second aspects above.

[0022] In a ninth aspect, a computer program product is provided, comprising computer program instructions, wherein the computer program instructions enable a computer to execute the method in any one of the first to second aspects above.

[0023] In a tenth aspect, a computer program is provided, which, when executed on a computer, enables the computer to execute the method in any one of the first to second aspects above.

[0024] Through the technical solution of the first aspect above, the network device implements time-domain prediction of the spatial filter based on the first network model, as well as the spatial filter based on the link quality verification prediction, and determines the spatial filter to be used through a spatial filter scanning method within a period of time after the verification fails. It is possible to monitor the link quality predicted by the AI ​​model in real time, ensuring the performance of the matching spatial filter between the network device and the terminal device, reducing the spatial filter measurement and feedback overhead required for spatial filter tracking during the movement of the terminal device, reducing the spatial filter switching delay, while ensuring the stability of the link quality, avoiding the degradation of the spatial filter quality caused by the prediction errors of the AI ​​model, and achieving better spatial filter time-domain prediction performance.

[0025] Through the technical solution of the second aspect above, the terminal device implements time-domain prediction of the spatial filter based on the second network model, as well as the spatial filter based on the link quality verification prediction, and determines the spatial filter to be used through a spatial filter scanning method within a period of time after the verification fails. It is possible to monitor the link quality predicted by the AI ​​model in real time, ensuring the performance of the matching spatial filter between the network device and the terminal device, reducing the spatial filter measurement and feedback overhead required for spatial filter tracking during the movement of the terminal device, reducing the spatial filter switching delay, while ensuring the stability of the link quality, avoiding the degradation of the spatial filter quality caused by the prediction errors of the AI ​​model, and achieving better spatial filter time-domain prediction performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] FIG1 is a schematic diagram of a communication system architecture applied in an embodiment of the present application.

[0027] FIG2 is a schematic diagram showing the connections of neurons in a neural network.

[0028] FIG3 is a schematic diagram of a fully connected neural network.

[0029] FIG4 is a schematic structural diagram of a convolutional neural network.

[0030] FIG5 is a schematic structural diagram of an LSTM unit.

[0031] FIG6 is a schematic diagram of a downlink beam scanning process.

[0032] FIG7 is a schematic diagram of another downlink beam scanning process.

[0033] FIG8 is a schematic flowchart of a wireless communication method provided according to an embodiment of the present application.

[0034] FIG9 is a schematic diagram of a link quality check provided in an embodiment of the present application.

[0035] FIG10 is a schematic diagram of another link quality check provided in an embodiment of the present application.

[0036] Figure 11 is a schematic diagram of an LSTM network model provided in an embodiment of the present application.

[0037] 12 to 15 are schematic flow charts of prediction and verification provided in embodiments of the present application.

[0038] FIG16 is a schematic flowchart of another wireless communication method provided according to an embodiment of the present application.

[0039] FIG17 is a schematic diagram of another link quality verification provided in an embodiment of the present application.

[0040] FIG18 is a schematic diagram of another link quality verification provided in an embodiment of the present application.

[0041] 19 to 21 are schematic flow charts of prediction and verification provided in embodiments of the present application.

[0042] Figure 22 is a schematic block diagram of a network device provided according to an embodiment of the present application.

[0043] Figure 23 is a schematic block diagram of a terminal device provided according to an embodiment of the present application.

[0044] Figure 24 is a schematic block diagram of a communication device provided according to an embodiment of the present application.

[0045] Figure 25 is a schematic block diagram of a device provided according to an embodiment of the present application.

[0046] Figure 26 is a schematic block diagram of a communication system provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0047] The following will describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. With respect to the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0048] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: 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), 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 on unlicensed spectrum, NR-based access to unlicensed spectrum (NR-U) system on unlicensed spectrum, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Internet of Things (IoT), Wireless Fidelity (WFI) system. Fidelity, WiFi), fifth-generation communication (5th-Generation, 5G) system, sixth-generation communication (6G) system or other communication systems.

[0049] Generally speaking, traditional communication systems support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communications, but will also support, for example, device-to-device (D2D) communication, machine-to-machine (M2M) communication, machine type communication (MTC), vehicle-to-vehicle (V2V) communication, sidelink (SL) communication, vehicle-to-everything (V2X) communication, etc. The embodiments of the present application can also be applied to these communication systems.

[0050] In some embodiments, the communication system in the embodiments of the present application can be applied to a carrier aggregation (CA) scenario, a dual connectivity (DC) scenario, an independent (SA) networking scenario, or a non-standalone (NSA) networking scenario.

[0051] In some embodiments, the communication system in the embodiments of the present application can be applied to an unlicensed spectrum, where the unlicensed spectrum can also be considered as a shared spectrum; or, the communication system in the embodiments of the present application can also be applied to an authorized spectrum, where the authorized spectrum can also be considered as an unshared spectrum.

[0052] In some embodiments, the communication system in the embodiments of the present application can be applied to the FR1 frequency band (corresponding to the frequency band range of 410MHz to 7.125GHz), can also be applied to the FR2 frequency band (corresponding to the frequency band range of 24.25GHz to 52.6GHz), and can also be applied to new frequency bands such as high-frequency bands corresponding to the frequency band range of 52.6GHz to 71GHz or the frequency band range of 71GHz to 114.25GHz.

[0053] The embodiments of the present application describe various embodiments in conjunction with network devices and terminal devices, wherein the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.

[0054] The terminal device can be a station (ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a next-generation communication system such as an NR network, or a terminal device in a future evolved Public Land Mobile Network (PLMN) network, etc.

[0055] In an embodiment of the present application, the terminal device can be deployed on land, including indoors or outdoors, handheld, wearable or vehicle-mounted; it can also be deployed on the water surface (such as ships, etc.); it can also be deployed in the air (such as airplanes, balloons and satellites, etc.).

[0056] In an embodiment of the present application, the terminal device may be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city or a wireless terminal device in a smart home, an in-vehicle communication device, a wireless communication chip / application specific integrated circuit (ASIC) / system on chip (SoC), etc.

[0057] As an example and not a limitation, in the embodiment of the present application, the terminal device may also be a wearable device. Wearable devices may also be called wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0058] In an embodiment of the present application, the network device may be a device for communicating with a mobile device. The network device may be an access point (AP) in WLAN, a base station (BTS) in GSM or CDMA, a base station (NodeB, NB) in WCDMA, an evolved base station (eNB or eNodeB) in LTE, or a relay station or access point, or a network device or base station (gNB) or a transmission reception point (TRP) in a vehicle-mounted device, a wearable device, and an NR network, or a network device in a future evolved PLMN network or a network device in an NTN network, etc.

[0059] As an example and not a limitation, in an embodiment of the present application, the network device may have a mobile feature, for example, the network device may be a mobile device. In some embodiments, the network device may be a satellite or a balloon station. For example, the satellite may be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, etc. In some embodiments, the network device may also be a base station set up in a location such as land or water.

[0060] In an embodiment of the present application, the network device can provide services for a cell, and the terminal device communicates with the network device through the transmission resources used by the cell (for example, frequency domain resources, or spectrum resources). The cell can be a cell corresponding to the network device (for example, a base station). The cell can belong to a macro base station or a base station corresponding to a small cell. The small cells here may include: metro cells, micro cells, pico cells, femto cells, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.

[0061] For example, a communication system 100 used in an embodiment of the present application is shown in FIG1 . The communication system 100 may include a network device 110, which may be a device that communicates with a terminal device 120 (or a communication terminal or terminal). The network device 110 may provide communication coverage for a specific geographic area and may communicate with terminal devices within the coverage area.

[0062] FIG1 exemplarily shows a network device and two terminal devices. In some embodiments, the communication system 100 may include multiple network devices and each network device may include other numbers of terminal devices within its coverage area, which is not limited in the embodiments of the present application.

[0063] In some embodiments, the communication system 100 may further include other network entities such as a network controller and a mobility management entity, which is not limited in the embodiments of the present application.

[0064] It should be understood that in the embodiments of the present application, a device having a communication function in a network / system may be referred to as a communication device. Taking the communication system 100 shown in FIG1 as an example, the communication device may include a network device 110 and a terminal device 120 having a communication function. The network device 110 and the terminal device 120 may be the specific devices described above and will not be described in detail here. The communication device may also include other devices in the communication system 100, such as a network controller, a mobility management entity, and other network entities, which are not limited in the embodiments of the present application.

[0065] It should be understood that the terms "system" and "network" are often used interchangeably herein. The term "and / or" is simply a description of an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " generally indicates that the related objects are in an "or" relationship.

[0066] It should be understood that this document relates to a first communication device and a second communication device. The first communication device can be a terminal device, such as a mobile phone, machine facilities, customer premises equipment (CPE), industrial equipment, vehicles, etc.; the second communication device can be a peer communication device of the first communication device, such as a network device, mobile phone, industrial equipment, vehicles, etc. In the embodiments of the present application, the first communication device can be a terminal device, and the second communication device can be a network device (i.e., uplink communication or downlink communication); alternatively, the first communication device can be a first terminal, and the second communication device can be a second terminal (i.e., sideline communication).

[0067] The terms used in the embodiments of this application are intended only to explain the specific embodiments of this application and are not intended to limit this application. The terms "first," "second," "third," and "fourth," etc. in the specification and claims of this application and the accompanying drawings are used to distinguish different objects, not to describe a specific order. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions.

[0068] It should be understood that the "indication" mentioned in the embodiments of this application can be a direct indication, an indirect indication, or an indication of an association. For example, "A indicates B" can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association between A and B.

[0069] In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship between indication and being indicated, configuration and being configured, etc.

[0070] In the embodiments of the present application, "pre-definition" or "pre-configuration" may be implemented by pre-storing corresponding codes, tables, or other methods that can be used to indicate relevant information in a device (e.g., a terminal device and a network device). The present application does not limit the specific implementation method. For example, pre-definition may refer to information defined in a protocol.

[0071] In the embodiments of the present application, the “protocol” may refer to a standard protocol in the communications field, for example, it may be an evolution of an existing LTE protocol, NR protocol, Wi-Fi protocol, or a protocol related to other communication systems. The present application does not limit the protocol type.

[0072] To facilitate understanding of the technical solutions of the embodiments of the present application, the technical solutions of the present application are described in detail below through specific embodiments. The following related technologies can be combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the scope of protection of the embodiments of the present application. The embodiments of the present application include at least part of the following contents.

[0073] To facilitate a better understanding of the embodiments of the present application, the neural network and machine learning related to the present application are explained.

[0074] A neural network (NN) is a computational model consisting of multiple interconnected neuron nodes, as shown in Figure 2. The connections between nodes represent the weighted values ​​from input signals to output signals, called weights. Each node performs a weighted summation on different input signals and outputs them through a specific activation function.

[0075] A simple neural network, as shown in Figure 3, consists of an input layer, a hidden layer, and an output layer. By using different connections, weights, and activation functions across multiple neurons, different outputs can be generated, thereby fitting the mapping from input to output. Each node in the previous level is connected to all of its next-level nodes. This fully connected model is also referred to as a deep neural network (DNN) in this application.

[0076] The basic structure of a convolutional neural network (CNN) consists of an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer, as shown in Figure 4. Each neuron in the convolutional kernel of a convolutional layer is locally connected to its input. The introduction of a pooling layer extracts the local maximum or average features of a layer, effectively reducing the number of network parameters and exploiting local features, enabling the convolutional neural network to converge quickly and achieve excellent performance.

[0077] A recurrent neural network (RNN) is a neural network that models sequential data and has achieved remarkable results in natural language processing applications such as machine translation and speech recognition. Specifically, the network device memorizes information from past moments and uses it in the calculation of the current output. This means that the nodes between hidden layers are no longer disconnected but connected, and the input to the hidden layer includes not only the input layer but also the output of the hidden layer at the previous moment. Common RNNs include structures such as long short-term memory (LSTM) and gated recurrent unit (GRU). Figure 5 shows a basic LSTM cell structure, which can include a tanh activation function. Unlike RNNs, which only consider the most recent state, the cell state of an LSTM determines which states should be retained and which should be forgotten, addressing the shortcomings of traditional RNNs in long-term memory. This application only uses RNNs as an example and does not exclude the possibility of replacing RNNs with other network structures such as DNNs and CNNs, and expanding the application using the same working method.

[0078] To facilitate a better understanding of the embodiments of the present application, the beam management related to the present application is explained.

[0079] In the NR system, millimeter wave frequency band communications are introduced, and corresponding beam management mechanisms are also introduced, including beam management that can be divided into uplink and downlink. This application mainly focuses on reducing the overhead and latency related to downlink beam management. Therefore, the main introduction is to downlink beam management. Specifically, downlink beam management includes downlink beam scanning, optimal beam reporting on the UE side, and downlink beam indication on the network side.

[0080] The downlink beam scanning process refers to the network device scanning different transmit beam directions using downlink reference signals. The UE can use different receive beams for measurement, thereby traversing all beam pairs. The UE calculates the Layer 1 Reference Signal Received Power (L1-RSRP) value corresponding to each beam pair.

[0081] The downlink reference signal includes a synchronization signal block (Synchronization Signal Block, SSB) and / or a channel state information reference signal (Channel State Information Reference Signal, CSI-RS).

[0082] The downlink beam scanning process may be the P1 process (or downlink full scanning process) shown in FIG. 6 and the P3 process shown in FIG. 7 .

[0083] As shown in FIG6 , the network device may traverse all transmit beams to send downlink reference signals, and the UE side traverses all receive beams to perform measurements and determine corresponding measurement results.

[0084] As shown in FIG7 , the network device may send a downlink reference signal using a specific transmit beam, and the UE side traverses all receive beams to perform measurements and determine corresponding measurement results.

[0085] Beam reporting means that the UE compares the L1-RSRP values ​​of all measured beam pairs, selects the K transmit beams with the highest L1-RSRP, and then reports them to the network as uplink control information. Here, L1-RSRP can also be replaced by other beam link indicators, such as Layer 1 Signal to Interference plus Noise Ratio (L1-SINR) / Layer 1 Reference Signal Receiving Quality (L1-RSRQ). After decoding the UE's beam reporting, the network device can use Media Access Control (MAC) or Downlink Control Information (DCI) signaling to carry the Transmission Configuration Indicator (TCI) status (including the transmit beam with SSB or CSI-RS as the reference) to complete the beam indication to the UE. The UE uses the receive beam corresponding to the transmit beam for reception.

[0086] The above is the basic process of beam management used in the NR system. When the UE moves, the NR system does not support beam tracking and beam prediction for UE mobile scenarios. Therefore, the network and UE need to periodically or aperiodically trigger the beam management process to readjust to match the transmit beam and receive beam.

[0087] To facilitate a better understanding of the embodiments of the present application, the AI-based beam time domain prediction related to the present application is explained.

[0088] In artificial intelligence (AI)-based beam time-domain prediction, the measurement results of the past K>=1 moments are used as the input of the AI ​​model, and the prediction results of the next F moments are used as the output of the AI ​​model. The AI ​​model can be deployed on the network side or the UE side. The measurement results of the AI ​​model can be L1-RSRP, and the corresponding transmit beam index, receive beam index or beam pair index, and can also include other auxiliary information. By designing a matching signaling process and supporting AI functions, AI-based beam time-domain prediction can be achieved, so that in the next F moments, the beam management process in the NR system is no longer required, and beam tracking in UE mobile scenarios can be directly achieved.

[0089] In order to facilitate a better understanding of the embodiments of the present application, the problems solved by the present application are explained.

[0090] When the UE moves, beam switching is required. However, the NR system currently does not support beam tracking and beam prediction for UE mobility scenarios. Therefore, the network and UE need to periodically or aperiodically trigger the beam management process to readjust the transmit and receive beams.

[0091] For the downlink full scan process, also known as the P1 process, the UE needs to traverse all combinations of transmit and receive beams, which incurs significant overhead and latency. For example, if the network equipment deploys 64 different downlink transmit beams in the FR2 frequency band (carried by up to 64 SSBs), the UE uses multiple antenna panels (including a single receive beam panel) to simultaneously scan the receive beams, and each antenna panel has four receive beams. Therefore, the UE needs to measure at least 256 beam pairs, resulting in 256 resources of downlink resource overhead.

[0092] From a time perspective, each SSB cycle is approximately 20ms, so four SSB cycles are required to complete the measurement of four receiving beams (assuming that multiple receiving antenna panels can be used for beam scanning), which means at least 80ms is required.

[0093] Therefore, when the UE is in a fast-moving scenario, frequent beam switching is required. The existing NR system beam management technology solution will frequently trigger the beam management process, resulting in huge overhead and latency. Therefore, it is necessary to design a beam switching and beam tracking solution for UE mobile scenarios.

[0094] In addition, the AI-based beam time-domain prediction framework can effectively reduce the huge overhead and latency caused by frequent beam management, enabling low-cost beam tracking. However, the limited generalization of the AI ​​model itself, as well as the complex changes in the channel environment during UE mobility, can lead to a mismatch between the training set and the actual deployment environment, which can lead to AI-based beam time-domain prediction errors. That is, within F moments in the future, the optimal transmit beam, optimal receive beam, or optimal beam pair predicted by the AI ​​model may not be the optimal choice. The corresponding link quality may even be so poor that normal communication between the network and the UE cannot be achieved.

[0095] Therefore, how to improve the robustness of AI-based beam time-domain prediction and ensure the stability of link quality is also a problem that needs to be solved.

[0096] In order to solve the above technical problems, this application proposes a beam time domain prediction solution based on an AI model. While deploying the AI ​​model on the network or UE side to realize beam time domain prediction, a link quality verification and beam management fallback mechanism based on link quality threshold is designed, so that the network and UE can monitor the link quality of the AI-based beam time domain prediction in real time, thereby ensuring the performance of the matching beam pairs between the network and the UE. This application deploys the AI ​​model online on the network or UE side and designs a matching link quality verification and beam management fallback mechanism. While reducing the beam measurement and feedback overhead required for beam tracking during UE mobility and reducing the beam switching delay, it ensures the stability of the link quality, avoids the degradation of beam quality caused by prediction errors of the AI ​​model, and achieves better beam time domain prediction performance.

[0097] The technical solution of this application is described in detail below through specific embodiments.

[0098] FIG8 is a schematic flowchart of a wireless communication method 200 according to an embodiment of the present application. The wireless communication method 200 may be performed by a network device in the communication system shown in FIG1 . As shown in FIG8 , the wireless communication method 200 includes at least part of the following:

[0099] S210: The network device obtains a first data set, where the first data set includes identification information of a spatial filter of K1 time units and link quality information of K1 time units, where K1 is a positive integer.

[0100] S220: The network device inputs the first data set into a first network model and outputs first prediction information, wherein the first prediction information includes identification information of a spatial filter and link quality information of F1 time units, wherein at least one check point is set in the F1 time unit, and F1 is a positive integer;

[0101] S230, the network device determines whether to use the spatial filter output by the first network model within n time units after the first check point based on the link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F1.

[0102] In the application embodiments, a spatial filter may also be referred to as a beam, or a spatial relation, or a spatial setting, or a spatial domain filter, or a reference signal.

[0103] In the embodiment of the present application, the first check point may be any one of the at least one check point, and the network device or the terminal device performs link quality verification at the at least one check point in turn.

[0104] In this embodiment of the present application, one time unit corresponds to one spatial filter identification information and one link quality information, where the link quality information may be the link quality information measured under the spatial filter. That is, in the first data set, K1 time units correspond to K1 spatial filter identification information and K1 link quality information, respectively; in the first prediction information, F1 time units correspond to F1 spatial filter identification information and F1 link quality information, respectively.

[0105] In the application embodiment, when predicting the beam time domain based on the AI ​​model (i.e., the first network model), a link quality threshold is set on the network side or the terminal side to determine whether the link quality of the optimal transmit beam or the optimal beam pair predicted by the AI ​​model within F1 time units meets the standard; when the AI ​​model prediction result does not meet the standard, a reasonable beam management fallback mechanism is designed so that the network device and the terminal device can match the correct beam pair. This application can reduce the beam management overhead based on the beam time domain prediction of the AI ​​model while ensuring the stability of the link quality.

[0106] In some embodiments, the time unit may include, but is not limited to, one of the following: time slot, frame, subframe, symbol, millisecond, microsecond.

[0107] In some embodiments, the spatial filter includes an optimal transmit spatial filter (such as an optimal transmit beam), or the spatial filter includes an optimal transmit spatial filter and an optimal receive spatial filter (such as a beam pair consisting of an optimal transmit beam and an optimal receive beam).

[0108] In some embodiments, the transmit spatial filter may also be referred to as a transmit beam (Tx beam) or a transmitting-end spatial domain filter, and the above terms may be interchangeable.

[0109] In some embodiments, the receive spatial filter may also be referred to as a receive beam (Rx beam) or a receive-end spatial domain filter, and the above terms may be interchangeable.

[0110] In some embodiments, the combination of the transmit spatial filter and the receive spatial filter may also be referred to as a beam pair, a spatial filter pair, or a spatial filter bank, and the above terms may be used interchangeably.

[0111] In some embodiments, the identification information of the spatial filter may be an index or an identification of the spatial filter.

[0112] For example, the identification information of the transmit spatial filter may be an index or an identification of the transmit spatial filter.

[0113] For another example, the identification information of the receiving spatial filter may be an index or an identification of the receiving spatial filter.

[0114] For another example, the identification information of the combination of the transmit spatial filter and the receive spatial filter may be a combination index.

[0115] In some embodiments, the link quality information may include but is not limited to at least one of the following:

[0116] Layer 1 Reference Signal Receiving Power (L1-RSRP), Layer 1 Reference Signal Receiving Quality (L1-RSRQ), Layer 1 Signal to Interference plus Noise Ratio (L1-SINR).

[0117] In some embodiments, when the spatial filter includes an optimal transmit spatial filter, the first data set is obtained by the network device through spatial filter scanning. For example, the network device obtains identification information of the optimal transmit spatial filter for K1 time units through spatial filter scanning, and uses the link measurement information and identification information of the optimal transmit spatial filter for the past K1 time units as the first data set.

[0118] In some embodiments, when the spatial filter includes an optimal transmit spatial filter and an optimal receive spatial filter, the first data set is obtained by the network device using identification information of the optimal receive spatial filter reported by the terminal device and the optimal transmit spatial filter determined by spatial filter scanning. For example, the terminal device reports identification information of the optimal receive spatial filter for the past K1 time units, and the network device obtains identification information of the optimal transmit spatial filter for K1 time units through spatial filter scanning, and uses the link measurement information, identification information of the optimal transmit spatial filter, and identification information of the optimal receive spatial filter for the past K1 time units as the first data set.

[0119] In some embodiments, when the number of the at least one check point is greater than or equal to 2, different check points are separated by n time units. Of course, the intervals between different check points may also be different. For example, check point 1 and check point 2 are separated by n1 time units, and check point 2 and check point 3 are separated by n2 time units. Whether to use the spatial filter output by the first network model within the next n1 time units may be determined based on the check result of check point 1, and whether to use the spatial filter output by the first network model within the next n2 time units may be determined based on the check result of check point 2, and so on.

[0120] In some embodiments, the above S230 may specifically include:

[0121] If the link quality check at the first check point succeeds, the network device uses the spatial filter output by the first network model within n time units after the first check point; and / or,

[0122] In the case that the link quality check at the first check point fails, the network device determines the spatial filter to be used by means of spatial filter scanning within n time units after the first check point.

[0123] In some embodiments, the K1 time units and the F1 time units may be consecutive time units. Specifically, for example, the F1 time units are the F1 time units that are consecutive after the K1 time units.

[0124] In some embodiments, the K1 time units may be measurement times, and the F1 time units may be prediction times.

[0125] In some embodiments, K1 ≥ F1. Optionally, the specific values ​​of K1 and F1 may be configured by the network device, or the specific values ​​of K1 and F1 may be agreed upon by a protocol. Alternatively, the specific values ​​of K1 and F1 may be determined by model parameters of the first network model.

[0126] In some embodiments, the link quality verification result of the first verification point is determined based on the link quality information of the first verification point and a link quality threshold.

[0127] Among them, the link quality information of the first verification point is: when the network device sends a reference signal using the optimal transmit spatial filter corresponding to the first verification point, the link quality information measured by the terminal device using the optimal receive spatial filter corresponding to the first verification point.

[0128] In some embodiments, the reference signal may be CSI-RS or SSB.

[0129] In some embodiments, the optimal transmit spatial filter corresponding to the first verification point may be the optimal transmit spatial filter of the time unit to which the first verification point belongs predicted by the first network model.

[0130] Specifically, when the link quality information of the first verification point is greater than or equal to the link quality threshold, the link quality verification of the first verification point is successful; and / or, when the link quality information of the first verification point is less than the link quality threshold, the link quality verification of the first verification point fails.

[0131] Specifically, for example, at verification point q0, the network side or the terminal side compares the link quality r0 of the terminal device after accessing the spatial filter (such as a transmit beam, a receive beam, or a beam pair) of the time unit to which verification point q0 belongs based on the link quality threshold (r_th). When r0 ≥ r_th, the verification is successful; when r0 < r_th, the verification fails, indicating that there may be an error in the spatial filter of the time unit to which verification point q0 belongs predicted by the first network model.

[0132] Verification successful: When verification point q0 is successful, the spatial filter prediction results of the first network model can be continued to be used for the next n time units (including the time unit where verification point q0 is located) to achieve the fast access of the terminal device and the network device until the next verification point is encountered.

[0133] Verification failure: When the verification point q0 fails, the spatial filter prediction results of the subsequent n time units (including the time unit where the verification point q0 is located) are considered unreliable, and the network device and the terminal device cannot directly access the corresponding spatial filter (beam or beam pair) through the prediction results of the first network model. Therefore, the n time units from the verification point q0 to the next verification point, that is, after the verification point q0, will be set as a fallback period. During this fallback period, the traditional beam management method is adopted to achieve beam pair matching and terminal device access through beam measurement and reporting. Until the next verification point of the verification point q0, if the verification is successful, the beam prediction results of the first network model are restored; if the verification continues to fail, the above process is repeated until the next verification point or the end of F1 time units (that is, the prediction time).

[0134] In the beam management fallback mechanism based on link quality verification proposed in this application, the verification configuration group {q0,n} can significantly impact the performance of the solution. When n is configured too large, the verification density is too low, which may cause errors in the spatial filter predicted by the first network model to be missed. However, when n is configured too small, the verification density is too high, which may result in additional verification signaling overhead. Therefore, the verification configuration group can be dynamically configured by the network device or terminal device based on actual conditions.

[0135] In some embodiments, the optimal receiving spatial filter corresponding to the first check point is determined by the terminal device based on the optimal transmitting spatial filter corresponding to the first check point. Specifically, for example, the network device can trigger a periodic terminal device receiving spatial filter scanning process (i.e., the P3 process as shown in Figure 7): a CSI-RS resource set is transmitted, and its repetition (Repetition) is set to be turned on or activated (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the optimal transmitting spatial filter corresponding to the first check point predicted by the first network model, and the terminal device measures the CSI-RS through different receiving spatial filters and selects the optimal receiving spatial filter.

[0136] In some embodiments, the optimal receive spatial filter corresponding to the first check point is indicated by the network device to the terminal device. Specifically, for example, when the spatial filter includes an optimal transmit spatial filter and an optimal receive spatial filter, the network device may indicate to the terminal device the optimal receive spatial filter for the time unit to which the first check point belongs, as predicted by the first network model.

[0137] In some embodiments, the link quality verification result of the first verification point is determined by the network device based on the link quality information of the first verification point reported by the terminal device and the link quality threshold. That is, the verification can be performed by the network device.

[0138] In some embodiments, the link quality check result of the first check point is determined by the terminal device based on the link quality information of the first check point and the link quality threshold and reported to the network device. That is, the check can be performed by the terminal device.

[0139] In some embodiments, the link quality threshold is determined based on an average value or a weighted average value of the link quality information of the K1 time units.

[0140] In some embodiments, the link quality threshold is determined based on the following formula 1:

[0141] r_th=(Q1+Q2+…+Q K1 ) / K1–X Formula 1

[0142] Among them, r_th represents the link quality threshold, Q1 to Q K1 They represent the link quality information of the K1 time units, and X represents the link quality threshold offset value. That is, Q1 represents the link quality information of the first time unit in the K1 time units, Q2 represents the link quality information of the second time unit in the K1 time units, ..., Q K1 Indicates the link quality information of the K1th time unit among K1 time units.

[0143] In some embodiments, the link quality threshold is determined based on the following formula 2:

[0144] r_th=(a1*Q1+a2*Q2+…+a K1 *Q K1 ) / K1–X Formula 2

[0145] Among them, r_th represents the link quality threshold, Q1 to Q K1 Respectively represent the link quality information of the K1 time units, a1 to a K1 They represent the weighted weights of the K1 time units, and a1+a2+…+a K1 =1, X represents the link quality threshold offset value. That is, Q1 represents the link quality information of the first time unit in K1 time units, Q2 represents the link quality information of the second time unit in K1 time units, ..., Q K1 represents the link quality information of the K1th time unit in K1 time units. a1 represents the weighted weight of the first time unit in K1 time units, a2 represents the weighted weight of the second time unit in K1 time units, ..., a K1 represents the weighted weight of the K1th time unit among K1 time units.

[0146] In some embodiments, in the above formula 2, a1<=a2<=…<=a K1 This is because the closer the moment is to the check point, the closer the channel environment and link quality are, so the weighted weight should be larger.

[0147] In some embodiments, in the above formulas 1 and 2, the value of X is determined based on the channel quality. Specifically, X>0. The setting of X is because in an actual system, the change in the link quality of the beam may fade as the terminal device moves. Therefore, a deviation in the beam quality within a certain range is allowed. If X is not set or X is set too low, it may cause each check to fail. Even when the beam or beam pair predicted by the first network model is correct, the link quality threshold requirement cannot be met. Therefore, the value of X can be configured according to different channel qualities. For example, the channel quality can be evaluated based on CSI-RS or SSB.

[0148] In some embodiments, in the above formula 1 and formula 2, the value of X is configured by the network device, or the value of X is configured by the terminal device.

[0149] In some embodiments, the link quality threshold is agreed upon by a protocol, or the link quality threshold is configured by the network device.

[0150] In some embodiments, the unit of the link quality threshold (r_th) is decibel (dB) or decibel milliwatt (dBm), which is not limited in this application.

[0151] In some embodiments, when the link quality check at the first check point fails, the network device sends a first indication message, wherein the first indication message is used to instruct the terminal device to determine the spatial filter to be used by spatial filter scanning within n time units after the first check point.

[0152] In some embodiments, the K1 time units may be measurement time, the F1 time units may be prediction time, and the K1 time units and the F1 time units may constitute a complete prediction cycle.

[0153] Specifically, as shown in FIG9 , the network device determines whether to use the spatial filter output by the first network model within n time units after check point 0 based on the link quality check result of check point 0, where the starting position of check point 0 is located at time unit q0 among F1 time units. For example, when the check point 0 check succeeds, the network device determines whether to use the spatial filter output by the first network model within n time units after check point 0; when the check point 0 check fails, the network device determines the spatial filter to use by scanning the spatial filter within n time units after check point 0. That is, when the check point 0 check fails, the traditional beam management method is used within n time units after check point 0 to achieve beam pair matching and terminal device access through beam measurement and reporting. The n time units after check point 0 can also be referred to as the backoff time or backoff period.

[0154] In some embodiments, as shown in FIG9 , one or more check points are set within F1 time units. The check points are configured by setting the starting position of the 0th check point to time unit q0, and setting the check interval between adjacent check points to n>=1. Therefore, a check configuration group {q0,n} can determine the check start time and check density within F1 time units. This check configuration group {q0,n} can be configured by a network device or a terminal device.

[0155] In some embodiments, the starting position of the at least one check point and the value of n are configured by the network device, or the starting position of the at least one check point and the value of n are agreed upon by a protocol. For example, the network device may indicate the check point configuration (including the starting position of the check point and the value of n) to the terminal device via a Media Access Control Control Element (MAC CE) or DCI.

[0156] In some embodiments, the starting position of the at least one check point and the value of n are determined by the network device or terminal device based on the check configuration group; wherein the check configuration group includes multiple check configuration types, and each check configuration type in the multiple check configuration types corresponds to a starting position of the check point and a value of n.

[0157] In some embodiments, the verification configuration group is agreed upon by a protocol, or the verification configuration group is pre-configured by a network device.

[0158] For example, the check point configuration group can be as shown in Table 1. The check configuration group {q0,n} can have N pre-configured configurations to form a check configuration group, and is indicated by adding a log2N bit indication field in the MAC CE or DCI, for example, N=4 in Table 1.

[0159] Table 1

[0160] Configuration type checkpoint q0n0 time unit 111 time unit 132 time unit F1 / 213 time unit F1 / 23

[0161] For another specific example, the checkpoint configuration group may be as shown in Table 2.

[0162] Table 2

[0163] Configuration type checkpoint q0 checkpoint q1n0 time unit 1 time unit 311 time unit 1 time unit 532 time unit 2 time unit 413 time unit 2 time unit 63

[0164] In an embodiment of the present application, the measurement time of the first K1 time units and the prediction time of the last F1 time units are statically and periodically spaced, as shown in Figure 9. When the moving speed of the terminal device is uneven, or the moving direction of the terminal device changes, it may result in that within the prediction time of the last F1 time units, the first network model prediction accuracy of the relatively early time units is higher, which enables the network device and the terminal device to match the correct spatial filter with better link quality; while the relatively late time units may have poor prediction performance, because the spatial filters that the network devices and terminal devices should adopt in these time units and the spatial and temporal correlation between the selected spatial filters and the corresponding link quality measurement results within the measurement time of the first K1 time units are relatively weak. The first network model may be more inaccurate in the process of extracting the correlation to generate predictions due to changes in the movement pattern of the terminal device. Specifically, it may cause verification failures to always occur at the relatively late verification points within the measurement time of the F1 time units, and it is necessary to fall back for beam scanning, and it is impossible to effectively use the first network model to reduce the beam scanning overhead and access delay. In this case, the embodiment of the present application further proposes a dynamic beam management fallback mechanism based on link quality verification, that is, the measurement time of K1 time units and the prediction time of F1 time units are dynamically shifted and updated.

[0165] In some embodiments, after the first check point verification is completed, the network device updates the first data set based on the link quality verification result of the first check point; the network device inputs the updated first data set into the first network model, outputs the updated first prediction information, and determines the updated first check point in the updated F1 time units included in the updated first prediction information; the network device determines whether to use the spatial filter output by the first network model within n time units after the updated first check point in the updated F1 time units based on the link quality verification result of the updated first check point.

[0166] In some embodiments, when the link quality check of the first check point is successful, the updated K1 time units included in the updated first data set include m time units in the F1 time units, and the updated K1 time units do not include the first m time units in the K1 time units, and the updated first check point is located in the mth time unit of the updated F1 time unit; wherein, the m time units in the F1 time unit are the time unit in the F1 time unit to which the first check point belongs and the time unit before, and m is a positive integer.

[0167] For example, assuming the first checkpoint is checkpoint 0, when the check is successful at checkpoint 0 (located at the q0th time unit among F1 time units), it is considered that all spatial filter prediction results before checkpoint 0 are correct and can enable the network device and the terminal device to be paired normally. At this time, the measurement time and the prediction time are shifted and updated. As shown in Figure 10, the link quality information and predicted spatial filter of the first q0 time units within the prediction time are also used as inputs to the first network model, and the link quality information and identification information of the earliest q0 time units in the past K1 time units included in the first data set are discarded. The identification information and link quality information of the updated spatial filter of the K1 time units are re-input into the first network model to obtain the updated prediction results of the first network model for the future F1 time units. At the same time, checkpoint 0 and the corresponding q0th time unit are updated.

[0168] In some embodiments, when the link quality check of the first check point fails, the updated K1 time units included in the updated first data set include m+n time units in the F1 time unit, and the updated K1 time units do not include the first m+n time units in the K1 time unit, and the updated first check point is located at the m+nth time unit of the updated F1 time unit; wherein, the m time units in the F1 time unit are the time unit in the F1 time unit to which the first check point belongs and the time unit before, and m is a positive integer.

[0169] For example, assuming the first checkpoint is checkpoint 0, when the check fails at checkpoint 0 (located at the q0th time unit among F1 time units): the n time units after checkpoint 0 (including the q0th time unit) enter the backoff time, and the traditional beam management method is used to match the beam pair between the network device and the terminal device. At this time, the measurement time and the prediction time are also shifted and updated. As shown in Figure 10, the link quality information and predicted spatial filter of the first q0+n time units within the prediction time are used as the input of the first network model, and the link quality information and identification information of the earliest q0+n time units in the past K1 time units included in the first data set are discarded. The identification information and link quality information of the updated spatial filter of the K1 time units are re-input into the first network model to obtain the updated prediction results of the first network model for the next F1 time units. At the same time, checkpoint 0 and the corresponding q0th time unit are updated.

[0170] In this embodiment, the dynamic beam management fallback mechanism based on link quality verification can use the correct and available prediction results of the first network model as the input of the first network model in real time, and update the output results of the first network model in real time, so that the first network model can more accurately predict the spatial filter in the short period of time in the future.

[0171] In some embodiments, the starting position of the first check point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first check point and the value of n are dynamically configured by the network device based on the check results in multiple time units in the past.

[0172] For example, in a dynamic beam management fallback mechanism based on link quality verification, the starting position of the first verification point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first verification point and the value of n are dynamically configured by the network device based on the verification results in multiple time units in the past.

[0173] In some embodiments, assuming the starting position of the first checkpoint is q0, the checkpoint configuration group {q0, n} also significantly affects the performance of this embodiment, especially the configuration of q0. When q0 is configured to be small (i.e., the first checkpoint within F1 time units is too early, for example, the checkpoint is performed in the first time unit), optimal first network model prediction performance can be guaranteed, but frequent first network model online inference is required to update the prediction results. When q0 is configured to be large, the number of first network model online inferences is small, and a single prediction result can be used for a long time, but the accuracy of the spatial filter predicted by the first network model cannot be guaranteed to be stable. Considering that the checkpoint configuration group {q0, n} is dynamically configured by MAC CE or DCI, the network device can dynamically adjust it based on the specific movement of the terminal device and the checkpoint results of multiple time units in the past. When the checkpoint results of multiple time units in the past are all successful, a larger q0 can be configured to reduce the number of first network model inferences. When a checkpoint fails, a smaller q0 can be configured to increase the number of first network model inferences to ensure the prediction performance of the first network model.

[0174] In some embodiments, the first network model may be an RNN, for example, an RNN-like structure such as LSTM, GRU, etc. Of course, the first network model may also be other model structures such as DNN, CNN, etc., which is not limited in this application.

[0175] In some embodiments, the first network model may be obtained through offline training, or through online training, or through a combination of offline and online training. For example, the network device first obtains a static training result through offline training, and further uses the offline trained model to predict the optimal beam or beam pair. In subsequent measurements and / or reporting by the terminal device, the network device may continue to collect more measurement data, and then use the measurement results to continue training the first network model to optimize model parameters to achieve better prediction results.

[0176] In some embodiments, the first network model is trained by the network device.

[0177] In some embodiments, the network device obtains a second data set; wherein the second data set includes at least one of the following: identification information of a spatial filter of K2 time units, link quality information of K2 time units, identification information of a spatial filter of F2 time units, and link quality information of F2 time units; wherein K2 and F2 are both positive integers;

[0178] The network device trains the first network model according to the second data set to obtain model parameters of the first network model.

[0179] In some embodiments, K2 ≥ K1, and / or, F2 ≥ F1.

[0180] In some embodiments, during the model training phase, the input of the first network model may be the index of the beam or beam pair of K2 time units and the corresponding link quality information, the label may be the index of the beam or beam pair with the best link quality (or the best measurement result) of F2 time units, and the output may be the index of the beam or beam pair with the best link quality of F2 time units.

[0181] In some embodiments, the first network model is trained by another device. Specifically, for example, the network device receives model parameter information of the trained first network model. The other device may be a terminal device, a third-party device, or a core network device, which is not limited in this application.

[0182] In some embodiments, the network side or the terminal side deploys an offline training method for the first network model, and the constructed data set (i.e., the second data set) may include the following parts:

[0183] Input: optimal transmit beam index or beam pair index of consecutive K2 time units, and corresponding link quality information L1-RSRP;

[0184] Label: The optimal transmit beam index or beam pair index for the consecutive F2 time units after the input consecutive K2 time units, and the corresponding link quality information L1-RSRP.

[0185] During the offline training process, the network device and the terminal device perform periodic beam selection for multiple consecutive K2+F2 time units according to the beam management protocol while the terminal device is moving, and the data set is reported by the terminal device to the network device.

[0186] For the first network model, this application extracts the correlation between the beam time domain sequence of the measurement time of the past K time units and the prediction time of the next F time units, so LSTM is used for prediction. The specific structure is shown in Figure 11.

[0187] During offline training, the first network model can be implemented by directly selecting the optimal transmit beam or beam pair index from multiple candidate transmit beams or beam pairs, or by predicting the L1-RSRP of all candidate transmit beams or beam pairs and selecting the best one as the output result. The embodiments of the present application support both implementation methods. Since the implementation logic of the second method is very straightforward, the network complexity is relatively high, and the predicted L1-RSRP of other transmit beams or beam pairs cannot be used, the first implementation method is used as an example below.

[0188] For the training process of the optimal transmit beam index or beam pair index, a cross entropy loss function can be used:

[0189] Transmit beam index as label: Taking 64 candidate transmit beams as an example, the label of the optimal transmit beam is one-hot encoded. For example, if the optimal transmit beam is mt, the label can be encoded as [0,..,0,1,0,..0]. This vector is a 64-bit vector, where only the mt position is 1 and all other positions are 0. The model can be trained using the multi-class cross entropy loss function.

[0190] Beam pair index as label: Taking 64 candidate transmit beams and 4 candidate receive beams, a total of 256 candidate beam pairs, as an example, the optimal beam pair is one-hot encoded. For example, for the optimal beam pair {mt,nt}, the label can be encoded as [0,..,0,1,0,..0]. This vector is a 256-length vector, where only the mt*64+nt position is 1, and all other positions are 0. The model can be trained using the multi-class cross entropy loss function.

[0191] Considering that network devices typically have significantly greater computing power than terminal devices, the offline training phase is preferably implemented on the network side. For example, assuming the offline training phase is implemented on the network side, after the offline training is completed, the network device can transfer the offline trained network model from the network device to the terminal side through model download to implement online inference and prediction.

[0192] The technical solution of this application is described in detail below through specific embodiments.

[0193] In Example 1, the network side performs inference prediction based on the first network model and performs verification. As shown in Figure 12, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and optimal transmit beam of the past K1 time units as input, and outputs the optimal transmit beam of the next F1 time units. Verification of the checkpoints is implemented on the network side. Specifically, verification and prediction can be implemented through the following Sa-0 to Sa-9.

[0194] Sa-0, the network device indicates the checkpoint configuration to the terminal device via MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-set configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field in the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0195] Sa-1 and Sa-2 are implemented within the measurement time of K1 time units. The network device can directly obtain the transmission beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the first network model.

[0196] Sa-3, the network device directly obtains the transmission beam that should be adopted by the network side in the next F1 time units through the output of the first network model, but the terminal device cannot obtain the receiving beam information that should be adopted within the next F1 time units.

[0197] Sa-4, therefore, Sa-4 is implemented within the next F1 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): transmit a CSI-RS resource set with its repetition set to ON, that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the first network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0198] Sa-5: Based on the Sa-0 configuration, the terminal device reports L1-RSRP at the check point.

[0199] Sa-6: The network performs a link quality check. The link quality threshold calculation method and the value of X during this check are determined by the network. If the check fails, Sa-7 through Sa-9 are triggered.

[0200] Sa-7: The network device indicates through DCI that the link quality information verification has failed and needs to fall back to beam scanning.

[0201] Sa-8, during the backoff time, performs beam scanning and reselects the optimal transmit beam and the optimal receive beam.

[0202] Sa-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0203] Specifically, the Sa-8 and Sa-9 processes are repeated until the next checkpoint.

[0204] In Example 2, the network side performs inference prediction based on the first network model and performs verification. As shown in Figure 13, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and optimal beam pair of the past K1 time units as input, and uses the optimal beam pair of the next F1 time units as output. Verification of the checkpoints is implemented on the network side. Specifically, verification and prediction can be implemented through the following steps Sb-0 to Sb-9.

[0205] Sb-0: The network device indicates the checkpoint configuration to the terminal device via MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-set configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field in the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0206] Sb-1 and Sb-2 are implemented within the measurement time of K1 time units, and the network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process. Since the beam management does not support the terminal device to report the optimal receiving beam, that is, the optimal beam selection of the terminal device is transparent to the network device. Therefore, in order to support the network device to obtain the optimal beam pair index, it is necessary to report the optimal receiving beam index during the Sb-2 process. Since the network side only needs to know the receiving beam index number reported by the terminal device, and does not need to know the corresponding binding relationship between the optimal receiving beam index and the optimal receiving beam actually used, a field for indicating the optimal receiving beam index can be added to the uplink control information (UCI). For example, when there are 4 candidate receiving beams, 2 bits are used to indicate the optimal receiving beam index number.

[0207] Sb-3, the network device directly obtains the transmission beam that should be adopted by the network side and the receiving beam index that should be adopted by the terminal side for the next F1 time units through the output of the first network model.

[0208] Sb-4, the network device indicates to the terminal device the optimal receiving beam to be used. The network device can indicate to the terminal device by adding an information field for indicating the optimal receiving beam index in the DCI, and the terminal device receives according to the receiving beam corresponding to the optimal receiving beam index.

[0209] Sb-5: Based on the Sb-0 configuration, the terminal device reports L1-RSRP at the check point.

[0210] Sb-6: The network performs a link quality check. The link quality threshold calculation method and the value of X during this check are determined by the network. If the check fails, Sb-7 through Sb-9 are triggered.

[0211] Sb-7: The network device indicates through DCI that the link quality information verification has failed and needs to fall back to perform beam scanning.

[0212] Sb-8, within the backoff time, performs beam scanning and reselects the optimal transmit beam and the optimal receive beam.

[0213] Sb-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0214] Specifically, the Sb-8 and Sb-9 processes are repeated until the next checkpoint.

[0215] It should be noted that the embodiment using the optimal transmit beam as the input and output of the first network model can be more compatible with the beam management protocol, and only needs to periodically trigger the P3 process based on the optimal transmit beam output on the network side of the first network model; the embodiment using the optimal beam pair as the input and output of the first network model needs to add the reporting and indication of the optimal receive beam index to the beam management protocol, but does not require an additional P3 process to achieve the optimal receive beam pair matching. However, under different input configurations, the process based on link quality verification is compatible with the network-side AI model reasoning and verification method, and can share a set of signaling processes.

[0216] In Example 3, the network side performs inference prediction based on the first network model, and the terminal side performs verification. As shown in Figure 14, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and optimal transmit beam of the past K1 time units as input, and outputs the optimal transmit beam of the next F1 time units. Verification of the checkpoint is then implemented on the terminal side. Specifically, verification and prediction can be implemented through the following Sc-0 to Sc-8.

[0217] In Sc-0, the network device indicates the checkpoint configuration to the terminal device via a MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-defined configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field in the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0218] Sc-1 and Sc-2 are implemented within the measurement time of K1 time units. The network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the first network model.

[0219] Sc-3, the network device directly obtains the transmission beam that should be adopted by the network side in the next F1 time units through the output of the first network model, but the terminal device cannot obtain the receiving beam information that should be adopted within the next F1 time units.

[0220] Sc-4, therefore, is implemented within the next F1 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): transmit a CSI-RS resource set with its repetition set to on (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the first network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0221] Sc-5: Based on the configuration in Sc-0, the terminal device locally verifies the link quality. The link quality threshold calculation method and the value of X during this verification process can be configured by the network. Optionally, since the terminal device can also obtain link quality information for the past K1 time units, the link quality threshold and the value of X can also be determined by the terminal device itself.

[0222] Sc-6: The terminal device reports a link quality information verification failure indication through UCI and needs to fall back to perform beam scanning.

[0223] Sc-7: During the backoff time, beam scanning is performed to reselect the optimal transmit beam and the optimal receive beam.

[0224] Sc-8: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0225] Specifically, the Sc-7 and Sc-8 processes are repeated until the next checkpoint.

[0226] In Example 4, the network side performs inference prediction based on the first network model, and the terminal side performs verification. The network side deploys the first network model, takes the link measurement information (L1-RSRP) and optimal beam pair of the past K1 time units as input, and uses the optimal beam pair of the next F1 time units as output. The terminal side implements verification of the checkpoint. Specifically, verification and prediction can be implemented through the following Sd-0 to Sd-8.

[0227] In step Sd-0, the network device indicates the checkpoint configuration to the terminal device via a MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-defined configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field to the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0228] Sd-1 and Sd-2 are implemented within the measurement time of K1 time units, and the network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process. Since the beam management does not support the terminal device to report the optimal receiving beam, that is, the optimal beam selection of the terminal device is transparent to the network device. Therefore, in order to support the network device to obtain the optimal beam pair index, it is necessary to report the optimal receiving beam index during the Sd-2 process. Since the network side only needs to know the receiving beam index number reported by the terminal device, and does not need to know the corresponding binding relationship between the optimal receiving beam index and the optimal receiving beam actually used, a field for indicating the optimal receiving beam index can be added to the UCI. For example, when there are 4 candidate receiving beams, 2 bits are used to indicate the optimal receiving beam index number.

[0229] Sd-3, the network device directly obtains the transmission beam that should be adopted by the network side and the receiving beam index that should be adopted by the terminal side for the next F1 time units through the output of the first network model.

[0230] Sd-4, the network device indicates to the terminal device the optimal receiving beam to be used. The network device can indicate to the terminal device by adding an information field for indicating the optimal receiving beam index in the DCI, and the terminal device receives according to the receiving beam corresponding to the optimal receiving beam index.

[0231] In Sd-5, based on the configuration in Sd-0, the terminal device locally verifies the link quality. The link quality threshold calculation method and the value of X during this verification process can be configured by the network. Optionally, since the terminal device can also obtain link quality information for the past K1 time units, the link quality threshold and the value of X can also be determined by the terminal device itself.

[0232] Sb-6: The terminal device reports a link quality information verification failure indication through UCI and needs to fall back to perform beam scanning.

[0233] Sd-7, within the backoff time, performs beam scanning and reselects the optimal transmit beam and the optimal receive beam.

[0234] Sd-8: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0235] Specifically, the Sd-7 and Sd-8 processes are repeated until the next checkpoint.

[0236] In Example 5, the network side performs inference prediction based on the first network model, and performs verification. In the dynamic beam management fallback mechanism based on link quality verification, as shown in Figure 15, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and the optimal transmit beam of the past K1 time units as input, and uses the optimal transmit beam of the next F1 time units as output, and implements verification of the verification point on the network side. Specifically, verification and prediction can be achieved through the following Se-0 to Se-10.

[0237] Se-0, the network device indicates the checkpoint configuration to the terminal device through MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can be dynamically configured based on the mobility of the terminal device, or based on the checkpoint results in multiple past time units.

[0238] Se-1 and Se-2 are implemented within the measurement time of K1 time units. The network device can directly obtain the transmission beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the first network model.

[0239] Se-3, the network device directly obtains the transmission beam that should be adopted by the network side in the next F1 time units through the output of the first network model, but the terminal device cannot obtain the receiving beam information that should be adopted within the next F1 time units.

[0240] Se-4, therefore, Se-4 is implemented within the next F1 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): transmit a CSI-RS resource set with its repetition (Repetition) set to on (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the first network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0241] Se-5, the terminal device reports the link quality information of all predicted time units before the cutoff check point q0 to the network device, which is used to dynamically update the measurement time and the prediction time, thereby determining the input of the updated first network model.

[0242] Se-6, the network side performs link quality verification; the link quality threshold calculation method and the value of X in this verification process are determined by the network side, and the network equipment dynamically updates the measurement time and prediction time based on the verification results, thereby determining the input of the updated first network model.

[0243] Se-7: The network device indicates verification success or failure to the terminal device. If verification fails, the device enters the fallback time and updates the verification configuration group {q0,n} after the fallback beam scan is complete. If verification succeeds, the device directly updates the verification configuration group {q0,n}. Finally, after the dynamic update of the measurement time and prediction time, the network can re-predict the optimal transmit beam using the first network model.

[0244] Se-8, when the verification fails, beam scanning is performed within the backoff time to reselect the optimal transmit beam and the optimal receive beam; when the verification succeeds, the verification configuration group {q0,n} indication is directly updated.

[0245] Se-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0246] Se-10, after the fallback beam scan is completed, the network device updates the verification configuration group {q0,n} indication.

[0247] In Example 6, the network side performs inference prediction based on the first network model and performs verification. In a dynamic beam management fallback mechanism based on link quality verification, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and the optimal beam pair for the past K1 time units as input, and uses the optimal beam pair for the next F1 time units as output. Verification of the checkpoints is implemented on the network side. Specifically, verification and prediction can be implemented through the following Sf-0 to Sf-10.

[0248] Sf-0, the network device indicates the checkpoint configuration to the terminal device via MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can be dynamically configured based on the mobility of the terminal device, or based on the checkpoint results in multiple past time units.

[0249] Sf-1 and Sf-2 are implemented within the measurement time of K1 time units. The network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the first network model. Since the beam management does not support the terminal device to report the optimal receiving beam, that is, the optimal beam selection of the terminal device is transparent to the network device. Therefore, in order to support the network device to obtain the optimal beam pair index, it is necessary to report the optimal receiving beam index during the Sf-2 process. Since the network side only needs to know the receiving beam index number reported by the terminal device, and does not need to know the corresponding binding relationship between the optimal receiving beam index and the optimal receiving beam actually used, a field for indicating the optimal receiving beam index can be added to the UCI. For example, when there are 4 candidate receiving beams, 2 bits are used to indicate the optimal receiving beam index number.

[0250] Sf-3, the network device directly obtains the transmit beam that should be adopted by the network side and the receive beam index that should be adopted by the terminal side for the next F1 time units through the output of the first network model.

[0251] Sf-4, the network device indicates to the terminal device the optimal receiving beam to be used. The network device can indicate to the terminal device by adding an information field for indicating the optimal receiving beam index in the DCI, and the terminal device receives according to the receiving beam corresponding to the optimal receiving beam index.

[0252] Sf-5, the terminal device reports the link quality information of all predicted time units before the cutoff check point q0 to the network device, which is used to dynamically update the measurement time and the prediction time, thereby determining the input of the updated first network model.

[0253] Sf-6, the network side performs link quality verification; the link quality threshold calculation method and the value of X during this verification process are determined by the network side, and the network device dynamically updates the measurement time and prediction time based on the verification results, thereby determining the input of the updated first network model.

[0254] In step 7, the network device indicates verification success or failure to the terminal device. If verification fails, the device enters the backoff time and updates the verification configuration group {q0,n} after the backoff beam scan is complete. If verification succeeds, the device directly updates the verification configuration group {q0,n}. Finally, after the dynamic update of the measurement time and prediction time, the network can re-predict the optimal transmit beam using the first network model.

[0255] Sf-8, when the verification fails, beam scanning is performed within the backoff time to reselect the optimal transmit beam and the optimal receive beam; when the verification succeeds, the verification configuration group {q0,n} indication is directly updated.

[0256] Sf-9: Within the backoff time, the terminal device reports the optimal beam pair and the corresponding L1-RSRP.

[0257] Sf-10, after the fallback beam scan is completed, the network device updates the verification configuration group {q0,n} indication.

[0258] In Example 7, the network side performs inference prediction based on the first network model, and the terminal side performs verification. In the dynamic beam management fallback mechanism based on link quality verification, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and the optimal transmit beam of the past K1 time units as input, and uses the optimal transmit beam of the next F1 time units as output. Verification of the checkpoint is implemented on the terminal side. Specifically, verification and prediction can be implemented through the following Sg-0 to Sg-10.

[0259] Sg-0, the network device indicates the checkpoint configuration to the terminal device via MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can be dynamically configured based on the mobility of the terminal device, or based on the checkpoint results in multiple past time units.

[0260] Sg-1 and Sg-2 are implemented within the measurement time of K1 time units. The network device can directly obtain the transmission beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the first network model.

[0261] Sg-3, the network device directly obtains the transmission beam that should be adopted by the network side in the next F1 time units through the output of the first network model, but the terminal device cannot obtain the receiving beam information that should be adopted within the next F1 time units.

[0262] Sg-4, therefore, Se-4 is implemented within the next F1 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): transmit a CSI-RS resource set with its repetition set to ON, that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the first network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0263] At step Sg-5, the terminal device reports link quality information for all predicted time units up to the checkpoint q0 to the network device for dynamically updating the measurement and prediction times, thereby determining the input to the updated first network model. Based on the check results, the network device dynamically updates the measurement and prediction times, thereby determining the input to the updated first network model.

[0264] Sg-6: The terminal device locally verifies the link quality. The link quality threshold calculation method and the value of X during this verification process can be configured by the network. Optionally, since the terminal device can also obtain link quality information for the past K1 time units, the link quality threshold and the value of X can also be determined by the terminal device itself.

[0265] In step Sg-7, the network device indicates verification success or failure to the terminal device. If verification fails, the device enters the fallback time and updates the verification configuration group {q0,n} after the fallback beam scan is complete. If verification succeeds, the device directly updates the verification configuration group {q0,n}. Finally, after the dynamic update of the measurement time and prediction time, the network can re-predict the optimal transmit beam using the first network model.

[0266] Sg-8, when the verification fails, beam scanning is performed within the backoff time to reselect the optimal transmit beam and the optimal receive beam; when the verification succeeds, the verification configuration group {q0,n} indication is directly updated.

[0267] Sg-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0268] Sg-10: After the fallback beam scan is completed, the network device updates the verification configuration group {q0,n} indication.

[0269] In Example 8, the network side performs inference prediction based on the first network model, and the terminal side performs verification. In the dynamic beam management fallback mechanism based on link quality verification, the network side deploys the first network model, takes the link measurement information (L1-RSRP) and the optimal beam pair of the past K1 time units as input, and uses the optimal beam pair of the next F1 time units as output. Verification of the checkpoint is implemented on the terminal side. Specifically, verification and prediction can be implemented through the following Sh-0 to Sh-10.

[0270] Sh-0, the network device indicates the check point configuration to the terminal device through MAC CE or DCI. Specifically, the check configuration group {q0,n} can be dynamically configured based on the mobility of the terminal device, or based on the check results in multiple time units in the past.

[0271] Sh-1 and Sh-2 are implemented within the measurement time of K1 time units, and the network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process. Since the beam management does not support the terminal device to report the optimal receiving beam, that is, the optimal beam selection of the terminal device is transparent to the network device. Therefore, in order to support the network device to obtain the optimal beam pair index, it is necessary to report the optimal receiving beam index in the Sh-2 process. Since the network side only needs to know the receiving beam index number reported by the terminal device, and does not need to know the corresponding binding relationship between the optimal receiving beam index and the optimal receiving beam actually used, a field for indicating the optimal receiving beam index can be added to the UCI. For example, when there are 4 candidate receiving beams, 2 bits are used to indicate the optimal receiving beam index number.

[0272] Sh-3, the network device directly obtains the transmission beam that should be adopted by the network side and the receiving beam index that should be adopted by the terminal side for the next F1 time units through the output of the first network model.

[0273] Sh-4, the network device indicates to the terminal device the optimal receiving beam to be used. The network device can indicate to the terminal device by adding an information field for indicating the optimal receiving beam index in the DCI, and the terminal device receives according to the receiving beam corresponding to the optimal receiving beam index.

[0274] Sh-5: The terminal device reports link quality information for all predicted time units up to the checkpoint q0 to the network device for dynamically updating the measurement time and prediction time, thereby determining the input to the updated first network model. Based on the check result, the network device dynamically updates the measurement time and prediction time, thereby determining the input to the updated first network model.

[0275] Sh-6: The terminal device locally verifies the link quality. The link quality threshold calculation method and the value of X during this verification process can be configured by the network. Optionally, since the terminal device can also obtain link quality information for the past K1 time units, the link quality threshold and the value of X can also be determined by the terminal device itself.

[0276] In Sh-7, the network device indicates verification success or failure to the terminal device. If verification fails, the device enters the fallback time and updates the verification configuration group {q0,n} after the fallback beam scan is complete. If verification succeeds, the device directly updates the verification configuration group {q0,n}. Finally, after dynamically updating the measurement time and prediction time, the network can re-predict the optimal transmit beam using the first network model.

[0277] Sh-8, when the verification fails, beam scanning is performed within the backoff time to reselect the optimal transmit beam and the optimal receive beam; when the verification succeeds, the verification configuration group {q0,n} indication is directly updated.

[0278] Sh-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0279] Sh-10, after the fallback beam scan is completed, the network device updates the verification configuration group {q0,n} indication.

[0280] Therefore, in an embodiment of the present application, the network device implements a time-domain prediction of the spatial filter based on the first network model, and a spatial filter based on the link quality verification prediction, and determines the spatial filter to be used by a spatial filter scanning method within a period of time after the verification fails. It is possible to monitor the link quality predicted by the AI ​​model in real time, ensuring the performance of the spatial filter matched between the network device and the terminal device, reducing the spatial filter measurement and feedback overhead required for spatial filter tracking during the movement of the terminal device, reducing the spatial filter switching delay, and ensuring the stability of the link quality. It avoids the degradation of the spatial filter quality caused by the prediction error of the AI ​​model, and achieves better spatial filter time-domain prediction performance.

[0281] FIG16 is a schematic flowchart of a wireless communication method 300 according to an embodiment of the present application. The wireless communication method 300 may be performed by a terminal device in the communication system shown in FIG1 . As shown in FIG16 , the wireless communication method 300 includes at least part of the following:

[0282] S310: The terminal device obtains a third data set, where the third data set includes identification information of the spatial filter of K3 time units and link quality information of K3 time units, where K3 is a positive integer;

[0283] S320: The terminal device inputs the third data set into the second network model and outputs second prediction information, wherein the second prediction information includes identification information of the spatial filter and link quality information of F3 time units, wherein at least one check point is set in the F3 time units, and F3 is a positive integer;

[0284] S330, the terminal device determines whether to use the spatial filter output by the second network model within n time units after the first check point based on the link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F3.

[0285] In the application embodiments, a spatial filter may also be referred to as a beam, or a spatial relation, or a spatial setting, or a spatial domain filter, or a reference signal.

[0286] In an embodiment of the present application, the first check point may be any one of the at least one check point, and the terminal device or the network device performs link quality verification at the at least one check point in turn.

[0287] In this embodiment of the present application, one time unit corresponds to one spatial filter identification information and one link quality information, where the link quality information may be the link quality information measured under the spatial filter. That is, in the third data set, K3 time units correspond to K3 spatial filter identification information and K3 link quality information, respectively; in the second prediction information, F3 time units correspond to F3 spatial filter identification information and F3 link quality information, respectively.

[0288] In the embodiment of the application, when predicting the beam time domain based on the AI ​​model (i.e., the second network model), a link quality threshold is set on the network side or the terminal side to determine whether the link quality of the optimal transmit beam or the optimal beam pair predicted by the AI ​​model within F3 time units meets the standard; when the AI ​​model prediction result does not meet the standard, a reasonable beam management fallback mechanism is designed so that the network device and the terminal device can match the correct beam pair. This application can reduce the beam management overhead based on the beam time domain prediction of the AI ​​model while ensuring the stability of the link quality.

[0289] In some embodiments, the time unit may include, but is not limited to, one of the following: time slot, frame, subframe, symbol, millisecond, microsecond.

[0290] In some embodiments, the spatial filter includes an optimal receive spatial filter (such as an optimal receive beam), or the spatial filter includes an optimal transmit spatial filter (such as an optimal transmit beam), or the spatial filter includes an optimal transmit spatial filter and an optimal receive spatial filter (such as a beam pair consisting of an optimal transmit beam and an optimal receive beam).

[0291] In some embodiments, the transmit spatial filter may also be referred to as a transmit beam (Tx beam) or a transmitting-end spatial domain filter, and the above terms may be interchangeable.

[0292] In some embodiments, the receive spatial filter may also be referred to as a receive beam (Rx beam) or a receive-end spatial domain filter, and the above terms may be interchangeable.

[0293] In some embodiments, the combination of the transmit spatial filter and the receive spatial filter may also be referred to as a beam pair, a spatial filter pair, or a spatial filter bank, and the above terms may be used interchangeably.

[0294] In some embodiments, the identification information of the spatial filter may be an index or an identification of the spatial filter.

[0295] For example, the identification information of the transmit spatial filter may be an index or an identification of the transmit spatial filter.

[0296] For another example, the identification information of the receiving spatial filter may be an index or an identification of the receiving spatial filter.

[0297] For another example, the identification information of the combination of the transmit spatial filter and the receive spatial filter may be a combination index.

[0298] In some embodiments, the link quality information may include but is not limited to at least one of the following:

[0299] Layer 1 Reference Signal Receiving Power (L1-RSRP), Layer 1 Reference Signal Receiving Quality (L1-RSRQ), Layer 1 Signal to Interference plus Noise Ratio (L1-SINR).

[0300] In some embodiments, when the spatial filter includes an optimal receiving spatial filter, the third data set is obtained by the terminal device through spatial filter scanning. For example, the terminal device obtains identification information of the optimal receiving spatial filter for K3 time units through spatial filter scanning, and uses the link measurement information and identification information of the optimal receiving spatial filter for the past K3 time units as the third data set.

[0301] In some embodiments, when the spatial filter includes an optimal transmit spatial filter, the third data set is obtained by the network device through spatial filter scanning and indicated to the terminal device. That is, the third data set can be obtained from the network device. For example, the network device obtains identification information of the optimal transmit spatial filter for K3 time units through spatial filter scanning, and uses the link measurement information and identification information of the optimal transmit spatial filter for the past K3 time units as the third data set.

[0302] In some embodiments, when the spatial filter includes an optimal transmit spatial filter and an optimal receive spatial filter, the third data set is obtained by the terminal device using identification information of the optimal transmit spatial filter indicated by the network device and the optimal receive spatial filter determined by spatial filter scanning. Specifically, for example, the network device indicates identification information of the optimal transmit spatial filter for the past K3 time units, and the terminal device obtains identification information of the optimal receive spatial filter for K3 time units through spatial filter scanning. The link measurement information, identification information of the optimal transmit spatial filter, and identification information of the optimal receive spatial filter for the past K3 time units are used as the third data set.

[0303] In some embodiments, when the number of the at least one check point is greater than or equal to 2, different check points are separated by n time units. Of course, the intervals between different check points may also be different. For example, check point 1 and check point 2 are separated by n1 time units, and check point 2 and check point 3 are separated by n2 time units. Whether to use the spatial filter output by the second network model within the next n1 time units may be determined based on the check result of check point 1, and whether to use the spatial filter output by the second network model within the next n2 time units may be determined based on the check result of check point 2, and so on.

[0304] In some embodiments, the above S330 may specifically include:

[0305] If the link quality check at the first check point succeeds, the terminal device uses the spatial filter output by the second network model within n time units after the first check point; and / or,

[0306] In the case where the link quality check at the first check point fails, the terminal device determines the spatial filter to be used by means of spatial filter scanning within n time units after the first check point.

[0307] In some embodiments, the K3 time units and the F3 time units may be consecutive time units. Specifically, for example, the F3 time units are the F1 time units that are consecutive after the K3 time units.

[0308] In some embodiments, the K3 time units may be measurement times, and the F3 time units may be prediction times.

[0309] In some embodiments, K3 ≥ F3. Optionally, the specific values ​​of K3 and F3 may be configured by the network device, or the specific values ​​of K3 and F3 may be agreed upon by a protocol. Alternatively, the specific values ​​of K3 and F3 may be determined by model parameters of the second network model.

[0310] In some embodiments, the link quality check result of the first check point is determined based on the link quality information of the first check point and a link quality threshold;

[0311] The link quality information of the first check point is: when the network device uses the optimal transmit spatial filter corresponding to the first check point to send a reference signal, the terminal device uses the optimal receive spatial filter corresponding to the first check point to measure the link quality information.

[0312] In some embodiments, the reference signal may be a CSI-RS or an SSB.

[0313] In some embodiments, the optimal receiving spatial filter corresponding to the first check point may be the optimal receiving spatial filter of the time unit to which the first check point belongs predicted by the second network model. The terminal device may report the optimal receiving spatial filter corresponding to the first check point to the network device.

[0314] In some embodiments, the optimal transmit spatial filter corresponding to the first check point may be the optimal transmit spatial filter for the time unit to which the first check point belongs, as predicted by the second network model. The terminal device may report the optimal transmit spatial filter corresponding to the first check point to the network device.

[0315] In some embodiments, the optimal transmit spatial filter corresponding to the first check point may be the optimal transmit spatial filter of the time unit to which the first check point belongs, as indicated by the network device.

[0316] Specifically, when the link quality information of the first verification point is greater than or equal to the link quality threshold, the link quality verification of the first verification point succeeds; and / or, when the link quality information of the first verification point is less than the link quality threshold, the link quality verification of the first verification point fails.

[0317] Specifically, for example, at the verification point q0, the network side or the terminal side compares the link quality r0 after the terminal device applies a spatial filter (such as a transmit beam, a receive beam, or a beam pair) in the time unit to which the verification point q0 belongs, based on the link quality threshold (r_th). When r0 ≥ r_th, the verification is successful; when r0 < r_th, the verification fails, indicating that there may be an error in the spatial filter predicted by the second network model for the time unit to which the verification point q0 belongs.

[0318] Verification successful: When the verification point q0 is successful, the spatial filter prediction results of the second network model can be continued to be used for the subsequent n time units (including the time unit where the verification point q0 is located), enabling the rapid access of the terminal device and the network device until the next verification point is encountered.

[0319] Verification failed: When the verification point q0 fails, the spatial filter prediction results for the subsequent n time units (including the time unit where the verification point q0 is located) are considered unreliable. The network device and the terminal device cannot directly access the corresponding spatial filter (beam or beam pair) through the prediction results of the second network model. Therefore, for the period from the verification point q0 to the next verification point, that is, the n time units after the verification point q0, it is set as a fallback period. During this fallback period, traditional beam management methods are used to achieve the matching of beam pairs and the access of the terminal device through beam measurement and reporting. Until the next verification point of the verification point q0, if the verification is successful, the beam prediction results of the second network model are restored; if the verification continues to fail, the above process is repeated until the next-next verification point or the end of F3 time units (i.e., the prediction time).

[0320] Under the beam management fallback mechanism based on link quality verification proposed in this application, the verification configuration group {q0, n} can significantly affect the performance of the solution. When n is configured too large, the verification density is too low, which may cause undetected errors in the spatial filter predicted by the second network model; but when n is configured too small, the verification density is too high, which may result in additional verification signaling overhead. Therefore, the verification configuration group can be dynamically configured by the network device or the terminal device according to the actual situation.

[0321] In some embodiments, the optimal receiving spatial filter corresponding to the first check point is determined by the terminal device based on the optimal transmitting spatial filter corresponding to the first check point. Specifically, for example, the network device can trigger a periodic terminal device receiving spatial filter scanning process (i.e., the P3 process as shown in Figure 7): a CSI-RS resource set is transmitted, and its repetition (Repetition) is set to be turned on or activated (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the optimal transmitting spatial filter corresponding to the first check point predicted by the second network model, and the terminal device measures the CSI-RS through different receiving spatial filters and selects the optimal receiving spatial filter.

[0322] In some embodiments, the optimal transmit spatial filter corresponding to the first check point is indicated to the network device by the terminal device. Specifically, for example, when the spatial filter includes an optimal transmit spatial filter and an optimal receive spatial filter, the terminal device may report the optimal receive spatial filter for the time unit to which the first check point belongs, as predicted by the second network model, to the network device.

[0323] In some embodiments, the optimal transmit spatial filter corresponding to the first check point is indicated by the network device to the terminal device.

[0324] In some embodiments, the link quality verification result of the first verification point is determined by the network device based on the link quality information of the first verification point reported by the terminal device and the link quality threshold. That is, the verification can be performed by the network device.

[0325] In some embodiments, the link quality check result of the first check point is determined by the terminal device based on the link quality information of the first check point and the link quality threshold and reported to the network device. That is, the check can be performed by the terminal device.

[0326] In some embodiments, the link quality threshold is determined based on an average value or a weighted average value of the link quality information of the K3 time units.

[0327] In some embodiments, the link quality threshold is determined based on the following formula 3:

[0328] r_th=(Q1+Q2+…+Q K3 ) / K3–X Formula 3

[0329] Among them, r_th represents the link quality threshold, Q1 to Q K3They represent the link quality information of the K3 time units respectively, and X represents the link quality threshold offset value. That is, Q1 represents the link quality information of the first time unit in the K3 time units, Q2 represents the link quality information of the second time unit in the K3 time units, ..., Q K3 Indicates the link quality information of the K3th time unit among K3 time units.

[0330] In some embodiments, the link quality threshold is determined based on the following formula 4:

[0331] r_th=(a1*Q1+a2*Q2+…+a K3 *Q K3 ) / K3–X Formula 4

[0332] Among them, r_th represents the link quality threshold, Q1 to Q K3 Respectively represent the link quality information of the K3 time units, a1 to a K3 They represent the weighted weights of the K3 time units, and a1+a2+…+a K3 =1, X represents the link quality threshold offset value. That is, Q1 represents the link quality information of the first time unit in K3 time units, Q2 represents the link quality information of the second time unit in K3 time units, ..., Q K3 represents the link quality information of the K3th time unit in the K3 time units. a1 represents the weighted weight of the first time unit in the K3 time units, a2 represents the weighted weight of the second time unit in the K3 time units, ..., a K3 represents the weighted weight of the K3th time unit among the K3 time units.

[0333] In some embodiments, in the above formula 4, a1<=a2<=…<=a K3 This is because the closer the moment is to the check point, the closer the channel environment and link quality are, so the weighted weight should be larger.

[0334] In some embodiments, in the above formulas 3 and 4, the value of X is determined based on the channel quality. Specifically, X>0. The setting of X is because in an actual system, the change in the link quality of the beam may fade as the terminal device moves. Therefore, a deviation in the beam quality within a certain range is allowed. If X is not set or X is set too low, it may cause each check to fail. Even when the beam or beam pair predicted by the second network model is correct, the link quality threshold requirement cannot be met. Therefore, the value of X can be configured according to different channel qualities. For example, the channel quality can be evaluated based on CSI-RS or SSB.

[0335] In some embodiments, in the above formula 3 and formula 4, the value of X is configured by the network device, or the value of X is configured by the terminal device.

[0336] In some embodiments, the link quality threshold is agreed upon by a protocol, or the link quality threshold is configured by the network device.

[0337] In some embodiments, the unit of the link quality threshold (r_th) is decibel (dB) or decibel milliwatt (dBm), which is not limited in this application.

[0338] In some embodiments, when the link quality check at the first check point fails, the terminal device sends a second indication message, wherein the second indication message is used to instruct the network device to determine the spatial filter to be used by spatial filter scanning within n time units after the first check point.

[0339] In some embodiments, the K3 time units may be measurement time, the F3 time units may be prediction time, and the K3 time units and the F3 time units may constitute a complete prediction cycle.

[0340] Specifically, as shown in FIG17 , the terminal device determines whether to use the spatial filter output by the second network model within n time units after check point 0 based on the link quality check result of check point 0, where the starting position of check point 0 is located at time unit q0 among F3 time units. For example, when the check point 0 check succeeds, the terminal device determines whether to use the spatial filter output by the second network model within n time units after check point 0; when the check point 0 check fails, the terminal device determines the spatial filter to use by scanning the spatial filter within n time units after check point 0. That is, when the check point 0 check fails, the traditional beam management method is used within n time units after check point 0 to achieve beam pair matching and terminal device access through beam measurement and reporting. The n time units after check point 0 can also be referred to as the backoff time or backoff period.

[0341] In some embodiments, as shown in FIG17 , one or more checkpoints are set within F3 time units. The checkpoints are configured by setting the starting position of the 0th checkpoint to time unit q0, and setting the check interval between adjacent checkpoints to n>=1. Therefore, a checkpoint configuration group {q0,n} can determine the checkpoint start time and checkpoint density within F3 time units. This checkpoint configuration group {q0,n} can be configured by a network device or a terminal device.

[0342] In some embodiments, the starting position of the at least one checkpoint and the value of n are configured by the network device, or the starting position of the at least one checkpoint and the value of n are agreed upon by a protocol. For example, the network device may indicate the checkpoint configuration (including the starting position of the checkpoint and the value of n) to the terminal device via a MAC CE or a DCI.

[0343] In some embodiments, the starting position of the at least one check point and the value of n are determined by the network device or terminal device based on the check configuration group; wherein the check configuration group includes multiple check configuration types, and each check configuration type in the multiple check configuration types corresponds to a starting position of the check point and a value of n.

[0344] In some embodiments, the verification configuration group is agreed upon by a protocol, or the verification configuration group is pre-configured by a network device.

[0345] For example, the check point configuration group may be as shown in Table 1 or Table 2 above.

[0346] In an embodiment of the present application, the measurement time of the first K3 time units and the prediction time of the last F3 time units are statically and periodically spaced, as shown in Figure 17. When the moving speed of the terminal device is uneven, or the moving direction of the terminal device changes, it may result in that within the prediction time of the last F3 time units, the second network model prediction accuracy of the relatively early time units is higher, which enables the network device and the terminal device to match the correct spatial filter with better link quality; while the relatively late time units may have poor prediction performance, because the spatial filters that the network devices and terminal devices should adopt in these time units and the spatial and temporal correlation between the selected spatial filters and the corresponding link quality measurement results within the measurement time of the first K3 time units are relatively weak. The second network model may be more inaccurate in the process of extracting the correlation to generate predictions due to changes in the movement pattern of the terminal device. Specifically, it may cause verification failures to always occur at the relatively late verification points within the measurement time of the F3 time units, and it is necessary to fall back for beam scanning, and it is impossible to effectively use the second network model to reduce the beam scanning overhead and access delay. In this case, the embodiment of the present application further proposes a dynamic beam management fallback mechanism based on link quality verification, that is, the measurement time of K3 time units and the prediction time of F3 time units are dynamically shifted and updated.

[0347] In some embodiments, after the first check point verification is completed, the terminal device updates the third data set based on the link quality verification result of the first check point; the terminal device inputs the updated third data set into the second network model, outputs the updated second prediction information, and determines the updated first check point in the updated F3 time units included in the updated second prediction information; the terminal device determines whether to use the spatial filter output by the second network model within n time units after the updated first check point in the updated F3 time units based on the link quality verification result of the updated first check point.

[0348] In some embodiments, when the link quality check of the first check point is successful, the updated K3 time units included in the updated third data set include m time units of the F3 time units, and the updated K3 time units do not include the first m time units of the K3 time units, and the updated first check point is located at the mth time unit of the updated F3 time units; wherein, the m time units in the F3 time units are the time units in the F3 time units to which the first check point belongs and the previous time units, and m is a positive integer.

[0349] For example, assuming the first checkpoint is checkpoint 0, when the check succeeds at checkpoint 0 (located at the q0th time unit among F3 time units), it is considered that all spatial filter prediction results before checkpoint 0 are correct and can enable the network device and terminal device to pair normally. At this time, the measurement time and prediction time are shifted and updated. As shown in Figure 18, the link quality information and predicted spatial filter of the first q0 time units within the prediction time are also used as inputs to the second network model, and the link quality information and identification information of the earliest q0 time units among the past K3 time units included in the third data set are discarded. The identification information and link quality information of the updated spatial filter of the K3 time units are re-input into the second network model to obtain the updated prediction results of the second network model for the next F3 time units. At the same time, checkpoint 0 and the corresponding q0th time unit are updated.

[0350] In some embodiments, when the link quality check of the first check point fails, the updated K3 time units included in the updated third data set include m+n time units of the F3 time units, and the updated K3 time units do not include the first m+n time units of the K3 time units, and the updated first check point is located at the m+nth time unit of the updated F3 time units; wherein, the m time units in the F3 time units are the time units in the F3 time units to which the first check point belongs and the previous time units, and m is a positive integer.

[0351] For example, assuming the first checkpoint is checkpoint 0, when the check fails at checkpoint 0 (located at the q0th time unit among F3 time units): the n time units after checkpoint 0 (including the q0th time unit) enter the backoff time, and the traditional beam management method is used to match the beam pair between the network device and the terminal device. At this time, the measurement time and the prediction time are also shifted and updated. As shown in Figure 18, the link quality information and predicted spatial filter of the first q0+n time units within the prediction time are used as the input of the second network model, and the link quality information and identification information of the earliest q0+n time units among the past K3 time units included in the third data set are discarded. The identification information and link quality information of the updated spatial filter of the K3 time units are re-input into the second network model to obtain the updated prediction results of the second network model for the next F3 time units. At the same time, checkpoint 0 and the corresponding q0th time unit are updated.

[0352] In this embodiment, the dynamic beam management fallback mechanism based on link quality verification can use the correct and available prediction results of the second network model as the input of the second network model in real time, and update the output results of the second network model in real time, so that the second network model can more accurately predict the spatial filter in the short period of time in the future.

[0353] In some embodiments, the starting position of the first check point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first check point and the value of n are dynamically configured by the terminal device based on the verification results in multiple past time units.

[0354] For example, in a dynamic beam management fallback mechanism based on link quality verification, the starting position of the first verification point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first verification point and the value of n are dynamically configured by the network device based on the verification results in multiple time units in the past.

[0355] In some embodiments, assuming the starting position of the first checkpoint is q0, the checkpoint configuration group {q0, n} also significantly affects the performance of this embodiment, especially the configuration of q0. When q0 is configured to be small (i.e., the first checkpoint within F3 time units is too early, for example, the checkpoint is performed in the first time unit), optimal second network model prediction performance can be guaranteed, but frequent second network model online inference is required to update the prediction results. When q0 is configured to be large, the second network model online inference frequency is small, and a single prediction result can be used for a long time, but the accuracy of the spatial filter predicted by the second network model cannot be guaranteed to be stable. Considering that the checkpoint configuration group {q0, n} is dynamically configured by MAC CE or DCI, the network device can dynamically adjust it based on the specific movement of the terminal device and the checkpoint results of multiple time units in the past. When the checkpoint results of multiple time units in the past are all successful, a larger q0 can be configured to reduce the number of second network model inferences. When a checkpoint fails, a smaller q0 can be configured to increase the number of second network model inferences to ensure the prediction performance of the second network model.

[0356] In some embodiments, the second network model can be an RNN, for example, an RNN-like structure such as LSTM, GRU, etc. Of course, the second network model can also be other model structures such as DNN, CNN, etc., which is not limited in this application.

[0357] In some embodiments, the second network model may be obtained through offline training, or through online training, or through a combination of offline and online training. For example, the network device first obtains a static training result through offline training, and further uses the offline trained model to predict the optimal beam or beam pair. In subsequent measurements and / or reporting by the terminal device, the network device may continue to collect more measurement data, and then use the measurement results to continue training the second network model to optimize model parameters to achieve better prediction results.

[0358] In some embodiments, the second network model is trained by the terminal device.

[0359] In some embodiments, the terminal device obtains a fourth data set; wherein the fourth data set includes at least one of the following: identification information of the spatial filter of K4 time units, link quality information of K4 time units, identification information of the spatial filter of F4 time units, and link quality information of F4 time units; wherein K4 and F4 are both positive integers; the terminal device trains the second network model according to the fourth data set to obtain model parameters of the second network model.

[0360] In some embodiments, K4 ≥ K3, and / or, F4 ≥ F3.

[0361] In some embodiments, during the model training phase, the input of the second network model may be the index of the beam or beam pair of K4 time units and the corresponding link quality information, the label may be the index of the beam or beam pair with the best link quality (or the best measurement result) of F4 time units, and the output may be the index of the beam or beam pair with the best link quality of F4 time units.

[0362] In some embodiments, the second network model is trained by another device. Specifically, for example, the terminal device receives the trained model parameter information of the second network model. The other device may be a network device, a third-party device, or a core network device, which is not limited in this application.

[0363] In some embodiments, the network side or the terminal side deploys an offline training method for the second network model, and the constructed data set (i.e., the fourth data set) may include the following parts:

[0364] Input: optimal transmit beam index or beam pair index of consecutive K4 time units, and corresponding link quality information L1-RSRP;

[0365] Label: The optimal transmit beam index or beam pair index for the consecutive F4 time units after the input consecutive K4 time units, and the corresponding link quality information L1-RSRP.

[0366] During the offline training process, the network device and the terminal device perform periodic beam selection for multiple consecutive K4+F4 time units according to the beam management protocol while the terminal device is moving, and the data set is reported by the terminal device to the network device.

[0367] For the second network model, this application extracts the correlation between the beam time domain sequence of the measurement time of the past K time units and the prediction time of the next F time units, so LSTM is used for prediction. The specific structure is shown in Figure 11.

[0368] During offline training, the second network model can be implemented by directly selecting the optimal transmit beam or beam pair index from multiple candidate transmit beams or beam pairs, or by predicting the L1-RSRP of all candidate transmit beams or beam pairs and selecting the best one as the output result. The embodiments of the present application support both implementation methods. Since the implementation logic of the second method is very straightforward, the network complexity is relatively high, and the predicted L1-RSRP of other transmit beams or beam pairs cannot be used, the first implementation method is used as an example below.

[0369] For the training process of the optimal transmit beam index or beam pair index, a cross entropy loss function can be used:

[0370] Transmit beam index as label: Taking 64 candidate transmit beams as an example, the label of the optimal transmit beam is one-hot encoded. For example, if the optimal transmit beam is mt, the label can be encoded as [0,..,0,1,0,..0]. This vector is a 64-bit vector, where only the mt position is 1 and all other positions are 0. The model can be trained using the multi-class cross entropy loss function.

[0371] Beam pair index as label: Taking 64 candidate transmit beams and 4 candidate receive beams, a total of 256 candidate beam pairs, as an example, the optimal beam pair is one-hot encoded. For example, for the optimal beam pair {mt,nt}, the label can be encoded as [0,..,0,1,0,..0]. This vector is a 256-length vector, where only the mt*64+nt position is 1, and all other positions are 0. The model can be trained using the multi-class cross entropy loss function.

[0372] Considering that the computing power of network devices is usually much better than that of terminal devices, the offline training phase is preferably implemented on the network side.

[0373] For example, assuming that the offline training phase is implemented on the network side, after the offline training of the network device is completed, it is necessary to transfer the offline trained second network model from the network device to the terminal side through model download and realize online reasoning and prediction.

[0374] The technical solution of this application is described in detail below through specific embodiments.

[0375] In Example 9, the terminal side performs inference prediction based on the first network model, and the network side performs verification. As shown in Figure 19, the terminal side deploys the first network model, takes the link measurement information (L1-RSRP) and optimal transmit beam of the past K3 time units as input, and uses the optimal transmit beam of the next F3 time units as output. Verification of the checkpoint is implemented on the network side. Specifically, verification and prediction can be implemented through the following Si-0 to Si-10.

[0376] At Si-0, the network device indicates the checkpoint configuration to the terminal device via a MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-defined configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field to the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0377] Si-1 and Si-2 are implemented within the measurement time of K3 time units. The network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the second network model.

[0378] Si-3, the terminal device predicts the optimal transmission beam and corresponding L1-RSRP for the next F3 time units through the second network model.

[0379] Si-4: The terminal device reports the prediction results to the network device, namely the optimal transmit beam and corresponding L1-RSRP for F3 time units.

[0380] Si-5, therefore, is implemented within the next F3 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): a CSI-RS resource set is transmitted, and its repetition (Repetition) is set to on (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the second network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0381] Si-6, based on the Si-0 configuration, the terminal device reports L1-RSRP at the check point.

[0382] In step Si-7, the network performs a link quality check. The calculation method for the link quality threshold and the value of X during this check are determined by the network. If the check fails, steps Si-8 to Si-9 are triggered.

[0383] Si-8, the network device indicates through DCI that the link quality information verification has failed and needs to fall back to perform beam scanning.

[0384] Si-9, during the backoff time, performs beam scanning and reselects the optimal transmit beam and the optimal receive beam.

[0385] Si-10: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0386] Specifically, the Si-9 and Si-10 processes are repeated until the next checkpoint.

[0387] In Example 10, the terminal side performs inference prediction based on the second network model, and the network side performs verification. As shown in Figure 20, the terminal side deploys the second network model, takes the link measurement information (L1-RSRP) and optimal beam pair of the past K3 time units as input, and uses the optimal beam pair of the next F3 time units as output. The network side verifies the checkpoint. Specifically, verification and prediction can be achieved through the following Sj-0 to Sj-9.

[0388] At step Sj-0, the network device indicates the checkpoint configuration to the terminal device via a MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-defined configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field to the MAC CE or DCI, where N = 4, as shown in Table 1 above.

[0389] Sj-1 and Sj-2 are implemented within the measurement time of K3 time units. The network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process. Since the beam management does not support the terminal device to report the optimal receiving beam, that is, the optimal beam selection of the terminal device is transparent to the network device. Therefore, in order to support the network device to obtain the optimal beam pair index, it is necessary to report the optimal receiving beam index during the Sj-2 process. Since the network side only needs to know the receiving beam index number reported by the terminal device, and does not need to know the corresponding binding relationship between the optimal receiving beam index and the optimal receiving beam actually used, a field for indicating the optimal receiving beam index can be added to the UCI. For example, when there are 4 candidate receiving beams, 2 bits are used to indicate the optimal receiving beam index number.

[0390] Sj-3, the terminal device directly obtains the transmission beam that the network side should adopt in the next F3 time units and the receiving beam index that the terminal side should adopt through the output of the second network model.

[0391] Sj-4, the terminal device reports the optimal transmission beam to be used to the network device. The terminal device can indicate the network device by adding an information field for indicating the optimal transmission beam index in the UCI. The network device transmits according to the transmission beam corresponding to the optimal transmission beam index.

[0392] Sj-5: Based on the Sj-0 configuration, the terminal device reports L1-RSRP at the check point.

[0393] Sj-6: The network performs a link quality check. The link quality threshold calculation method and the value of X during this check are determined by the network. If the check fails, Sj-7 through Sj-9 are triggered.

[0394] Sj-7: The network device indicates through DCI that the link quality information verification has failed and needs to fall back to perform beam scanning.

[0395] Sj-8, during the backoff time, performs beam scanning and reselects the optimal transmit beam and the optimal receive beam.

[0396] Sj-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0397] Specifically, the Sj-8 and Sj-9 processes are repeated until the next checkpoint.

[0398] When the AI ​​model is deployed on the terminal side, the link quality measurement information and the optimal receiving beam index of the past K3 time units can also be used as the input of the second network model, and the optimal receiving beam index to be used in the next F3 time units can be output. The link quality check and fallback beam scanning can be realized by reporting L1-RSRP at the check point.

[0399] It should be noted that using the optimal transmit beam and the optimal beam pair as inputs of the second network model can achieve good compatibility with the beam management protocol.

[0400] In Example 11, the terminal side performs inference prediction based on the second network model and performs verification. As shown in Figure 21, the terminal side deploys the second network model, takes the link measurement information (L1-RSRP) and optimal transmit beam of the past K3 time units as input, and uses the optimal transmit beam of the next F3 time units as output. The terminal side verifies the checkpoint. Specifically, verification and prediction can be achieved through the following Sk-0 to Sk-9.

[0401] In Sk-0, the network device indicates the checkpoint configuration to the terminal device via MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-set configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field in the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0402] Sk-1 and Sk-2 are implemented within the measurement time of K3 time units. The network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the second network model.

[0403] Sk-3: The terminal device predicts the optimal transmit beam and corresponding L1-RSRP for the next F3 time units through the second network model.

[0404] Sk-4: The terminal device reports the prediction results to the network device, namely the optimal transmit beam and corresponding L1-RSRP of F3 time units.

[0405] Sk-5, therefore, Sk-5 is implemented in the next F3 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): transmit a CSI-RS resource set with its repetition set to on (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the second network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0406] In Sk-6, the terminal device locally verifies the link quality. The link quality threshold calculation method and the value of X during this verification process can be configured by the network. Optionally, since the terminal device can also obtain link quality information for the past K3 time units, the link quality threshold and the value of X can also be determined by the terminal device itself.

[0407] Sk-7: The terminal device reports a link quality information verification failure indication through UCI and needs to fall back to perform beam scanning.

[0408] Sk-8, during the backoff time, performs beam scanning and reselects the optimal transmit beam and the optimal receive beam.

[0409] Sk-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0410] Specifically, the Sk-8 and Sk-9 processes are repeated until the next checkpoint.

[0411] In Example 12, the terminal side performs inference prediction based on the second network model and performs verification. The terminal side deploys the second network model, takes the link measurement information (L1-RSRP) and optimal beam pair of the past K3 time units as input, and uses the optimal beam pair of the next F3 time units as output. The terminal side verifies the checkpoint. Specifically, verification and prediction can be implemented through the following steps S1-0 to S1-8.

[0412] S1-0: The network device indicates the checkpoint configuration to the terminal device via a MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can have N pre-defined configurations, forming a checkpoint configuration group table. This is indicated by adding a log2N bit indication field to the MAC CE or DCI. For example, in Table 1 above, N = 4.

[0413] Sl-1 and Sl-2 are implemented within the measurement time of K3 time units, and the network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process. Since the beam management does not support the terminal device to report the optimal receiving beam, that is, the optimal beam selection of the terminal device is transparent to the network device. Therefore, in order to support the network device to obtain the optimal beam pair index, it is necessary to report the optimal receiving beam index during the Sl-2 process. Since the network side only needs to know the receiving beam index number reported by the terminal device, and does not need to know the corresponding binding relationship between the optimal receiving beam index and the optimal receiving beam actually used, a field for indicating the optimal receiving beam index can be added to the UCI. For example, when there are 4 candidate receiving beams, 2 bits are used to indicate the optimal receiving beam index number.

[0414] Sl-3, the terminal device directly obtains the transmission beam that should be adopted by the network side and the receiving beam index that should be adopted by the terminal side for the next F3 time units through the output of the second network model.

[0415] Sl-4, the terminal device indicates to the network device the optimal transmission beam to be used. The terminal device can indicate to the network device by adding an information field for indicating the optimal transmission beam index in the UCI, and the network device transmits according to the transmission beam corresponding to the optimal transmission beam index.

[0416] In S1-5, based on the configuration in S1-0, the terminal device locally verifies the link quality. The link quality threshold calculation method and the value of X used in this verification process can be configured by instructions from the network. Optionally, since the terminal device can also obtain link quality information for the past K3 time units, the link quality threshold and the value of X can also be determined by the terminal device itself.

[0417] Sl-6, the terminal device reports a link quality information verification failure indication through UCI and needs to fall back to perform beam scanning.

[0418] Sl-7, within the backoff time, beam scanning is performed to reselect the optimal transmit beam and the optimal receive beam.

[0419] Sl-8: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0420] Specifically, the S1-7 and S1-8 processes are repeated until the next checkpoint.

[0421] In Example 12, the terminal side performs inference prediction based on the second network model, and the network side performs verification. In the dynamic beam management fallback mechanism based on link quality verification, the link measurement information (L1-RSRP) and the optimal transmit beam for the past K3 time units are used as input, and the optimal transmit beam for the next F3 time units is used as output. The network side verifies the verification point. Specifically, verification and prediction can be implemented through the following Sm-0 to Sm-10.

[0422] In Sm-0, the network device indicates the checkpoint configuration to the terminal device via MAC CE or DCI. Specifically, the checkpoint configuration group {q0,n} can be dynamically configured based on the mobility of the terminal device, or based on the checkpoint results in multiple past time units.

[0423] Sm-1 and Sm-2 are implemented within the measurement time of K3 time units. The network device can directly obtain the transmit beam and the corresponding L1-RSRP in each time unit through the beam scanning process, and use them as the input of the second network model.

[0424] Sm-3: The terminal device predicts the optimal transmit beam and corresponding L1-RSRP for the next F3 time units through the second network model.

[0425] In Sm-4, the terminal device reports the prediction results to the network device, namely the optimal transmit beam and the corresponding L1-RSRP for F3 time units.

[0426] Sm-5, therefore, is implemented within the next F3 time units to determine the optimal receiving beam on the terminal side. Specifically, the network device can trigger a periodic terminal device receiving beam scanning process (P3 process): transmit a CSI-RS resource set with its repetition set to on (ON), that is, all CSI-RS resources in the CSI-RS resource set are sent using the transmit beam at the fth moment predicted by the second network model. The terminal device measures the CSI-RS through different receiving beams and selects the optimal receiving beam.

[0427] Sm-6, the network side performs link quality verification; the link quality threshold calculation method and the value of X during this verification process are determined by the network side, and the network device dynamically updates the measurement time and prediction time based on the verification results, thereby determining the input of the updated first network model.

[0428] In step Sm-7, the network device indicates verification success or failure to the terminal device. If verification fails, the device enters the backoff time and updates the verification configuration group {q0,n} after the backoff beam scan is complete. If verification succeeds, the device directly updates the verification configuration group {q0,n}. Finally, after dynamically updating the measurement and prediction times, the terminal can re-predict the optimal transmit beam using the second network model.

[0429] Sm-8, when the verification fails, beam scanning is performed within the backoff time to reselect the optimal transmit beam and the optimal receive beam; when the verification succeeds, the verification configuration group {q0,n} indication is directly updated.

[0430] Sm-9: Within the backoff time, the terminal device reports the optimal transmit beam and the corresponding L1-RSRP.

[0431] Sm-10: After the fallback beam scan is completed, the network device updates the verification configuration group {q0,n} indication.

[0432] It should be noted that the terminal side performs inference prediction based on the second network model and performs verification on the terminal side. In the dynamic beam management fallback mechanism based on link quality verification, the link measurement information (L1-RSRP) and the optimal beam pair of the past K3 time units are used as input, and the optimal beam pair of the next F3 time units is used as output. The verification of the checkpoint is implemented on the terminal side. The specific process can be referred to the above embodiment and will not be repeated here.

[0433] Therefore, in an embodiment of the present application, the terminal device implements a time-domain prediction of the spatial filter based on the second network model, and a spatial filter based on the link quality verification prediction, and determines the spatial filter to be used by a spatial filter scanning method within a period of time after the verification fails. It is possible to monitor the link quality predicted by the AI ​​model in real time, ensuring the performance of the spatial filter matched between the network device and the terminal device, reducing the spatial filter measurement and feedback overhead required for spatial filter tracking during the movement of the terminal device, reducing the spatial filter switching delay, and ensuring the stability of the link quality. It avoids the degradation of the spatial filter quality caused by the prediction error of the AI ​​model, and achieves better spatial filter time-domain prediction performance.

[0434] The above text, in combination with Figures 8 to 21, describes in detail the method embodiment of the present application. The following text, in combination with Figures 22 to 26, describes in detail the device embodiment of the present application. It should be understood that the device embodiment and the method embodiment correspond to each other, and similar descriptions can refer to the method embodiment.

[0435] Figure 22 shows a schematic block diagram of a network device 400 according to an embodiment of the present application. As shown in Figure 22, the network device 400 includes:

[0436] An acquiring unit 410 is configured to acquire a first data set, wherein the first data set includes identification information of spatial filters of K1 time units and link quality information of K1 time units, where K1 is a positive integer;

[0437] a processing unit 420 configured to input the first data set into a first network model and output first prediction information, wherein the first prediction information includes identification information of a spatial filter and link quality information for F1 time units, wherein at least one check point is set in the F1 time unit, and F1 is a positive integer;

[0438] The processing unit 420 is further configured to determine, based on a link quality check result of a first check point, whether to use a spatial filter output by the first network model within n time units after the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F1.

[0439] In some embodiments, the spatial filter comprises an optimal transmit spatial filter; or

[0440] The spatial filter includes an optimal transmission spatial filter and an optimal reception spatial filter.

[0441] In some embodiments, when the number of the at least one check point is greater than or equal to 2, different check points are separated by n time units.

[0442] In some embodiments, the processing unit 420 is specifically configured to:

[0443] If the link quality check at the first check point succeeds, using the spatial filter output by the first network model within n time units after the first check point; and / or,

[0444] In the case that the link quality check at the first check point fails, the spatial filter to be used is determined by spatial filter scanning within n time units after the first check point.

[0445] In some embodiments, after the first check point verification is completed, the processing unit 420 is further configured to update the first data set according to the link quality verification result of the first check point;

[0446] The processing unit 420 is further configured to input the updated first data set into the first network model, output updated first prediction information, and determine an updated first check point in the updated F1 time units included in the updated first prediction information;

[0447] The processing unit 420 is further configured to determine whether to use the spatial filter output by the first network model within n time units after the updated first check point in the updated F1 time units according to the link quality check result of the updated first check point.

[0448] In some embodiments, when the link quality check of the first check point succeeds, the updated K1 time units included in the updated first data set include m time units among the F1 time units, and the updated K1 time units do not include the first m time units among the K1 time units, and the updated first check point is located at the mth time unit of the updated F1 time units; and / or,

[0449] In the case where the link quality check of the first check point fails, the updated K1 time units included in the updated first data set include the m+n time units of the F1 time units, and the updated K1 time units do not include the first m+n time units of the K1 time units, and the updated first check point is located at the m+nth time unit of the updated F1 time units;

[0450] The m time units in the F1 time units are the time unit to which the first check point belongs and the time unit before it in the F1 time units, and m is a positive integer.

[0451] In some embodiments, the starting position of the first check point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first check point and the value of n are dynamically configured by the network device based on the check results in multiple time units in the past.

[0452] In some embodiments, the starting position of the at least one checkpoint and the value of n are configured by the network device, or the starting position of the at least one checkpoint and the value of n are agreed upon by a protocol.

[0453] In some embodiments, the starting position of the at least one check point and the value of n are determined by the network device or the terminal device based on the check configuration group;

[0454] The verification configuration group includes a plurality of verification configuration types, and each of the plurality of verification configuration types corresponds to a starting position of a verification point and a value of n.

[0455] In some embodiments, the verification configuration group is agreed upon by a protocol, or the verification configuration group is pre-configured by a network device.

[0456] In some embodiments, the link quality check result of the first check point is determined based on the link quality information of the first check point and a link quality threshold;

[0457] The link quality information of the first check point is:

[0458] When the network device uses the optimal transmit spatial filter corresponding to the first check point to send a reference signal, the terminal device uses the optimal receive spatial filter corresponding to the first check point to measure link quality information.

[0459] In some embodiments, if the link quality information of the first verification point is greater than or equal to the link quality threshold, the link quality verification of the first verification point succeeds; and / or,

[0460] In a case where the link quality information of the first verification point is less than the link quality threshold, the link quality verification of the first verification point fails.

[0461] In some embodiments, the link quality check result of the first check point is determined by the network device based on the link quality information of the first check point reported by the terminal device and the link quality threshold; or,

[0462] The link quality verification result of the first verification point is determined by the terminal device based on the link quality information of the first verification point and the link quality threshold and reported to the network device.

[0463] In some embodiments, the link quality threshold is determined based on an average value or a weighted average value of the link quality information of the K1 time units.

[0464] In some embodiments, the link quality threshold is determined based on the following formula:

[0465] r_th=(Q1+Q2+…+Q K1 ) / K1–X;

[0466] Among them, r_th represents the link quality threshold, Q1 to Q K1 They respectively represent the link quality information of the K1 time units, and X represents the link quality threshold offset value.

[0467] In some embodiments, the link quality threshold is determined based on the following formula:

[0468] r_th=(a1*Q1+a2*Q2+…+a K1 *Q K1 ) / K1–X;

[0469] Among them, r_th represents the link quality threshold, Q1 to Q K1Respectively represent the link quality information of the K1 time units, a1 to a K1 They represent the weighted weights of the K1 time units, and a1+a2+…+a K1 =1, X represents the link quality threshold offset value.

[0470] In some embodiments, a1<=a2<=…<=a K1 .

[0471] In some embodiments, the value of X is determined based on channel quality.

[0472] In some embodiments, the value of X is configured by the network device, or the value of X is configured by the terminal device.

[0473] In some embodiments, the link quality threshold is agreed upon by a protocol, or the link quality threshold is configured by the network device.

[0474] In some embodiments, the optimal receiving spatial filter corresponding to the first check point is determined by the terminal device based on the optimal transmitting spatial filter corresponding to the first check point, or the optimal receiving spatial filter corresponding to the first check point is indicated to the terminal device by the network device.

[0475] In some embodiments, when the link quality check at the first check point fails, the network device 400 further includes:

[0476] The communication unit 430 is configured to send first indication information, wherein the first indication information is used to instruct the terminal device to determine the spatial filter to be used by means of spatial filter scanning within n time units after the first check point.

[0477] In some embodiments, when the spatial filter includes an optimal transmission spatial filter, the first data set is acquired by the network device through spatial filter scanning; or,

[0478] In the case where the spatial filter includes an optimal transmitting spatial filter and an optimal receiving spatial filter, the first data set is obtained by the network device through the identification information of the optimal receiving spatial filter reported by the terminal device and the optimal transmitting spatial filter determined by spatial filter scanning.

[0479] In some embodiments, the acquisition unit 410 is further configured to acquire a second data set; wherein the second data set includes at least one of the following: identification information of the spatial filter of K2 time units, link quality information of K2 time units, identification information of the spatial filter of F2 time units, and link quality information of F2 time units; wherein K2 and F2 are both positive integers;

[0480] The processing unit 420 is further configured to train the first network model according to the second data set to obtain model parameters of the first network model.

[0481] In some embodiments, the network device 400 further includes:

[0482] The communication unit 430 is configured to receive the trained model parameter information of the first network model.

[0483] In some embodiments, the communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system on chip. The processing unit may be one or more processors.

[0484] It should be understood that the network device 400 according to the embodiment of the present application may correspond to the network device in the embodiment of the method of the present application, and the above-mentioned and other operations and / or functions of each unit in the network device 400 are respectively for implementing the corresponding processes of the network device in the method 200 shown in Figure 8. For the sake of brevity, they will not be repeated here.

[0485] FIG23 shows a schematic block diagram of a terminal device 500 according to an embodiment of the present application. As shown in FIG23 , the terminal device 500 includes:

[0486] An acquiring unit 510 is configured to acquire a third data set, wherein the third data set includes identification information of the spatial filter of K3 time units and link quality information of K3 time units, where K3 is a positive integer;

[0487] a processing unit 520 configured to input the third data set into the second network model and output second prediction information, wherein the second prediction information includes identification information of the spatial filter and link quality information of F3 time units, wherein at least one check point is set in the F3 time units, and F3 is a positive integer;

[0488] The processing unit 520 is further configured to determine, based on a link quality check result of a first check point, whether to use a spatial filter output by the second network model within n time units after the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F3.

[0489] In some embodiments, the spatial filter comprises an optimal receive spatial filter; or,

[0490] The spatial filter comprises an optimal transmit spatial filter; or,

[0491] The spatial filter includes an optimal transmission spatial filter and an optimal reception spatial filter.

[0492] In some embodiments, when the number of the at least one check point is greater than or equal to 2, different check points are separated by n time units.

[0493] In some embodiments, the processing unit 520 is specifically configured to:

[0494] If the link quality check at the first check point succeeds, using the spatial filter output by the second network model within n time units after the first check point; and / or,

[0495] In the case that the link quality check at the first check point fails, the spatial filter to be used is determined by spatial filter scanning within n time units after the first check point.

[0496] In some embodiments, after the first check point verification is completed, the processing unit 520 is further configured to update the third data set according to the link quality verification result of the first check point;

[0497] The processing unit 520 is further configured to input the updated third data set into the second network model, output updated second prediction information, and determine an updated first check point in the updated F3 time units included in the updated second prediction information;

[0498] The processing unit 520 is further configured to determine whether to use the spatial filter output by the second network model within n time units after the updated first check point in the updated F3 time units according to the link quality check result of the updated first check point.

[0499] In some embodiments, when the link quality check of the first check point succeeds, the updated K3 time units included in the updated third data set include m time units among the F3 time units, and the updated K3 time units do not include the first m time units among the K3 time units, and the updated first check point is located at the mth time unit of the updated F3 time units; and / or,

[0500] In a case where the link quality check of the first check point fails, the updated K3 time units included in the updated third data set include the m+n time units of the F3 time units, and the updated K3 time units do not include the first m+n time units of the K3 time units, and the updated first check point is located at the m+nth time unit of the updated F3 time units;

[0501] Among them, m time units in the F3 time units are the time unit to which the first check point belongs and the time unit before it in the F3 time units, and m is a positive integer.

[0502] In some embodiments, the starting position of the first check point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first check point and the value of n are dynamically configured by the terminal device based on the verification results in multiple past time units.

[0503] In some embodiments, the starting position of the at least one checkpoint and the value of n are configured by the network device, or the starting position of the at least one checkpoint and the value of n are agreed upon by a protocol.

[0504] In some embodiments, the starting position of the at least one check point and the value of n are determined by the network device or the terminal device based on a check configuration group;

[0505] The verification configuration group includes a plurality of verification configuration types, and each of the plurality of verification configuration types corresponds to a starting position of a verification point and a value of n.

[0506] In some embodiments, the verification configuration group is agreed upon by a protocol, or the verification configuration group is pre-configured by a network device.

[0507] In some embodiments, the link quality check result of the first check point is determined based on the link quality information of the first check point and a link quality threshold;

[0508] The link quality information of the first check point is:

[0509] When the network device uses the optimal transmit spatial filter corresponding to the first check point to send a reference signal, the terminal device uses the link quality information measured by the optimal receive spatial filter corresponding to the first check point.

[0510] In some embodiments, if the link quality information of the first verification point is greater than or equal to the link quality threshold, the link quality verification of the first verification point succeeds; and / or,

[0511] In a case where the link quality information of the first verification point is less than the link quality threshold, the link quality verification of the first verification point fails.

[0512] In some embodiments, the link quality check result of the first check point is determined by the network device based on the link quality information of the first check point and the link quality threshold reported by the terminal device and indicated to the terminal device; or,

[0513] The link quality check result of the first check point is determined by the terminal device based on the link quality information of the first check point and the link quality threshold.

[0514] In some embodiments, the link quality threshold is determined based on an average value or a weighted average value of the link quality information of the K3 time units.

[0515] In some embodiments, the link quality threshold is determined based on the following formula:

[0516] r_th=(Q1+Q2+…+Q K3 ) / K3–X;

[0517] Among them, r_th represents the link quality threshold, Q1 to Q K3 They respectively represent the link quality information of the K3 time units, and X represents the link quality threshold offset value.

[0518] In some embodiments, the link quality threshold is determined based on the following formula:

[0519] r_th=(a1*Q1+a2*Q2+…+a K3 *Q K3 ) / K3–X;

[0520] Among them, r_th represents the link quality threshold, Q1 to Q K3 Respectively represent the link quality information of the K3 time units, a1 to a K3 They represent the weighted weights of the K3 time units, and a1+a2+…+a K3 =1, X represents the link quality threshold offset value.

[0521] In some embodiments, a1<=a2<=…<=a K3 .

[0522] In some embodiments, the value of X is determined based on channel quality.

[0523] In some embodiments, the value of X is configured by the network device, or the value of X is configured by the terminal device.

[0524] In some embodiments, the link quality threshold is agreed upon by a protocol, or the link quality threshold is configured by a network device.

[0525] In some embodiments, the optimal transmit spatial filter corresponding to the first check point is indicated by the network device to the terminal device.

[0526] In some embodiments, when the link quality check at the first check point fails, the terminal device 500 further includes:

[0527] The communication unit 530 is configured to send second indication information, wherein the second indication information is used to instruct the network device to determine a spatial filter to be used by means of spatial filter scanning within n time units after the first check point.

[0528] In some embodiments, when the spatial filter includes an optimal receiving spatial filter, the third data set is acquired by the terminal device through spatial filter scanning; or

[0529] In the case where the spatial filter includes an optimal transmit spatial filter, the third data set is obtained by the terminal device from the network device; or,

[0530] In the case where the spatial filter includes an optimal transmitting spatial filter and an optimal receiving spatial filter, the third data set is obtained by the terminal device through the identification information of the optimal transmitting spatial filter indicated by the network device, and the optimal receiving spatial filter determined by spatial filter scanning.

[0531] In some embodiments, the acquisition unit 510 is further configured to acquire a fourth data set; wherein the fourth data set includes at least one of the following: identification information of the spatial filter of K4 time units, link quality information of K4 time units, identification information of the spatial filter of F4 time units, and link quality information of F4 time units; wherein K4 and F4 are both positive integers;

[0532] The processing unit 520 is further configured to train the second network model according to the fourth data set to obtain model parameters of the second network model.

[0533] In some embodiments, the terminal device 500 further includes:

[0534] The communication unit 530 is used to receive the trained model parameter information of the second network model.

[0535] In some embodiments, the communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system on chip. The processing unit may be one or more processors.

[0536] It should be understood that the terminal device 500 according to the embodiment of the present application may correspond to the terminal device in the embodiment of the method of the present application, and the above-mentioned and other operations and / or functions of each unit in the terminal device 500 are respectively for realizing the corresponding processes of the terminal device in the method 300 shown in Figure 16. For the sake of brevity, they will not be repeated here.

[0537] Figure 24 is a schematic structural diagram of a communication device 600 provided in an embodiment of the present application. The communication device 600 shown in Figure 24 includes a processor 610, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0538] In some embodiments, as shown in FIG24 , the communication device 600 may further include a memory 620. The processor 610 may call and execute a computer program from the memory 620 to implement the method in the embodiment of the present application.

[0539] The memory 620 may be a separate device independent of the processor 610 , or may be integrated into the processor 610 .

[0540] In some embodiments, as shown in FIG. 24 , the communication device 600 may further include a transceiver 630 , and the processor 610 may control the transceiver 630 to communicate with other devices. Specifically, the transceiver 630 may send information or data to other devices, or receive information or data sent by other devices.

[0541] The transceiver 630 may include a transmitter and a receiver. The transceiver 630 may further include an antenna, and the number of antennas may be one or more.

[0542] In some embodiments, the processor 610 may implement the functions of a processing unit in a terminal device, or the processor 610 may implement the functions of a processing unit in a network device, which will not be described in detail here for the sake of brevity.

[0543] In some embodiments, the transceiver 630 may implement the functions of a communication unit in a terminal device, which will not be described in detail here for the sake of brevity.

[0544] In some embodiments, the transceiver 630 may implement the function of a communication unit in a network device, which will not be described in detail here for the sake of brevity.

[0545] In some embodiments, the communication device 600 may specifically be a network device of an embodiment of the present application, and the communication device 600 may implement the corresponding processes implemented by the network device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0546] In some embodiments, the communication device 600 may specifically be a terminal device of an embodiment of the present application, and the communication device 600 may implement the corresponding processes implemented by the terminal device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0547] Figure 25 is a schematic structural diagram of an apparatus according to an embodiment of the present application. The apparatus 700 shown in Figure 25 includes a processor 710, which can call and execute a computer program from a memory to implement the method according to the embodiment of the present application.

[0548] In some embodiments, as shown in FIG25 , the apparatus 700 may further include a memory 720. The processor 710 may call and execute a computer program from the memory 720 to implement the method in the embodiment of the present application.

[0549] The memory 720 may be a separate device independent of the processor 710 , or may be integrated into the processor 710 .

[0550] In some embodiments, the apparatus 700 may further include an input interface 730. The processor 710 may control the input interface 730 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips. Optionally, the processor 710 may be located inside or outside the chip.

[0551] In some embodiments, the processor 710 may implement the functions of a processing unit in a terminal device, or the processor 710 may implement the functions of a processing unit in a network device, which will not be described in detail here for the sake of brevity.

[0552] In some embodiments, the input interface 730 may implement the function of a communication unit in a terminal device, or the input interface 730 may implement the function of a communication unit in a network device.

[0553] In some embodiments, the apparatus 700 may further include an output interface 740. The processor 710 may control the output interface 740 to communicate with other devices or chips, specifically, to output information or data to other devices or chips. Optionally, the processor 710 may be located inside or outside the chip.

[0554] In some embodiments, the output interface 740 may implement the function of a communication unit in a terminal device, or the output interface 740 may implement the function of a communication unit in a network device.

[0555] In some embodiments, the device can be applied to the network equipment in the embodiments of the present application, and the device can implement the corresponding processes implemented by the network equipment in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0556] In some embodiments, the apparatus can be applied to the terminal device in the embodiments of the present application, and the apparatus can implement the corresponding processes implemented by the terminal device in the various methods in the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0557] In some embodiments, the device mentioned in the embodiments of the present application may also be a chip, such as a system-on-chip, a system-on-chip, a chip system, or a system-on-chip chip.

[0558] FIG26 is a schematic block diagram of a communication system 800 provided in an embodiment of the present application. As shown in FIG26 , the communication system 800 includes a terminal device 810 and a network device 820 .

[0559] Among them, the terminal device 810 can be used to implement the corresponding functions implemented by the terminal device in the above method, and the network device 820 can be used to implement the corresponding functions implemented by the network device in the above method. For the sake of brevity, they will not be repeated here.

[0560] It should be understood that the processor of the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiment can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0561] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0562] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0563] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0564] In some embodiments, the computer-readable storage medium can be applied to the network device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0565] In some embodiments, the computer-readable storage medium can be applied to the terminal device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0566] An embodiment of the present application also provides a computer program product, including computer program instructions.

[0567] In some embodiments, the computer program product can be applied to the network device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0568] In some embodiments, the computer program product can be applied to the terminal device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0569] The embodiment of the present application also provides a computer program.

[0570] In some embodiments, the computer program can be applied to the network device in the embodiments of the present application. When the computer program runs on a computer, the computer executes the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0571] In some embodiments, the computer program can be applied to the terminal device in the embodiments of the present application. When the computer program runs on the computer, the computer executes the corresponding processes implemented by the terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0572] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0573] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0574] In the several embodiments provided in this 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 merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, 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.

[0575] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0576] 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.

[0577] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. In view of this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0578] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A wireless communication method, characterized in that: include: The network device obtains a first data set, wherein the first data set includes identification information of a spatial filter of K1 time units and link quality information of K1 time units, where K1 is a positive integer; The network device inputs the first data set into a first network model and outputs first prediction information, wherein the first prediction information includes identification information of a spatial filter of F1 time units and link quality information of F1 time units, wherein at least one check point is set in the F1 time unit, and F1 is a positive integer; The network device determines, based on a link quality check result of a first check point, whether to use a spatial filter output by the first network model within n time units after the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F1.

2. The method according to claim 1, wherein The spatial filter comprises an optimal transmit spatial filter; or The spatial filter includes an optimal transmission spatial filter and an optimal reception spatial filter.

3. The method according to claim 1 or 2, wherein: When the number of the at least one check point is greater than or equal to 2, different check points are spaced apart by n time units.

4. The method according to any one of claims 1 to 3, characterized in that The network device determines, based on a link quality check result of a first check point, whether to use a spatial filter output by the first network model within n time units after the first check point, including: If the link quality check at the first check point succeeds, the network device uses the spatial filter output by the first network model within n time units after the first check point; and / or, In a case where the link quality check at the first check point fails, the network device determines a spatial filter to be used by spatial filter scanning within n time units after the first check point.

5. The method according to any one of claims 1 to 4, characterized in that After the first checkpoint check is completed, the method further includes: The network device updates the first data set according to the link quality verification result of the first verification point; The network device inputs the updated first data set into the first network model, outputs updated first prediction information, and determines an updated first check point in the updated F1 time units included in the updated first prediction information; The network device determines whether to use the spatial filter output by the first network model within n time units after the updated first check point in the updated F1 time units according to the link quality check result of the updated first check point.

6. The method according to claim 5, wherein If the link quality check of the first check point succeeds, the updated K1 time units included in the updated first data set include m time units among the F1 time units, and the updated K1 time units do not include the first m time units among the K1 time units, and the updated first check point is located in the mth time unit of the updated F1 time units; and / or In a case where the link quality check of the first check point fails, the updated K1 time units included in the updated first data set include m+n time units of the F1 time units, and the updated K1 time units do not include the first m+n time units of the K1 time units, and the updated first check point is located at the m+nth time unit of the updated F1 time unit; The m time units in the F1 time units are the time unit to which the first check point belongs and the time unit before the F1 time units, and m is a positive integer.

7. The method according to claim 5 or 6, wherein: The starting position of the first check point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first check point and the value of n are dynamically configured by the network device based on the check results in multiple past time units.

8. The method according to any one of claims 1 to 4, characterized in that The starting position of the at least one check point and the value of n are configured by the network device, or the starting position of the at least one check point and the value of n are agreed upon by a protocol.

9. The method according to any one of claims 1 to 4, characterized in that The starting position of the at least one check point and the value of n are determined by the network device or the terminal device based on the check configuration group; The verification configuration group includes a plurality of verification configuration types, and each of the plurality of verification configuration types corresponds to a starting position of a verification point and a value of n.

10. The method according to claim 9, wherein The verification configuration group is agreed upon by a protocol, or the verification configuration group is pre-configured by a network device.

11. The method according to any one of claims 1 to 10, characterized in that The link quality verification result of the first verification point is determined based on the link quality information of the first verification point and a link quality threshold; The link quality information of the first check point is: When the network device uses the optimal transmit spatial filter corresponding to the first check point to send a reference signal, the terminal device uses the link quality information measured by the optimal receive spatial filter corresponding to the first check point.

12. The method according to claim 11, wherein If the link quality information of the first verification point is greater than or equal to the link quality threshold, the link quality verification of the first verification point succeeds; and / or, When the link quality information of the first verification point is less than the link quality threshold, the link quality verification of the first verification point fails.

13. The method according to claim 11 or 12, The link quality verification result of the first verification point is determined by the network device based on the link quality information of the first verification point reported by the terminal device and the link quality threshold; or, The link quality verification result of the first verification point is determined by the terminal device based on the link quality information of the first verification point and the link quality threshold and reported to the network device.

14. The method according to any one of claims 11 to 13, characterized in that The link quality threshold is determined based on an average value or a weighted average value of the link quality information of the K1 time units.

15. The method according to claim 14, wherein The link quality threshold is determined based on the following formula: r_th=(Q1+Q2+…+Q K1 ) / K1–X; Among them, r_th represents the link quality threshold, Q1 to Q K1 They respectively represent the link quality information of the K1 time units, and X represents the link quality threshold offset value.

16. The method according to claim 14, wherein The link quality threshold is determined based on the following formula: r_th=(a1*Q1+a2*Q2+…+a K1 *Q K1 ) / K1–X; Among them, r_th represents the link quality threshold, Q1 to Q K1 Respectively represent the link quality information of the K1 time units, a1 to a K1 They represent the weighted weights of the K1 time units, and a1+a2+…+a K1 =1, X represents the link quality threshold offset value.

17. The method according to claim 16, wherein <h2 style=";text-align:left;direction:ltr">a1<=a2<=…<=a<h2 style=";text-align:left;direction:ltr"> K1 <h2 style=";text-align:left;direction:ltr"> 。 18. The method according to any one of claims 15 to 17, characterized in that The value of X is determined based on the channel quality.

19. The method according to any one of claims 15 to 18, characterized in that The value of X is configured by the network device, or the value of X is configured by the terminal device.

20. The method according to any one of claims 11 to 13, characterized in that The link quality threshold is agreed upon by a protocol, or the link quality threshold is configured by the network device.

21. The method according to any one of claims 11 to 20, characterized in that The optimal receiving spatial filter corresponding to the first check point is determined by the terminal device based on the optimal transmitting spatial filter corresponding to the first check point, or the optimal receiving spatial filter corresponding to the first check point is indicated to the terminal device by the network device.

22. The method according to any one of claims 1 to 21, characterized in that In the case where the link quality check of the first check point fails, the method further includes: The network device sends first indication information, wherein the first indication information is used to instruct the terminal device to determine the spatial filter to be used by spatial filter scanning within n time units after the first check point.

23. The method according to any one of claims 1 to 22, characterized in that In the case where the spatial filter includes an optimal transmission spatial filter, the first data set is acquired by the network device through spatial filter scanning; or In the case where the spatial filter includes an optimal transmitting spatial filter and an optimal receiving spatial filter, the first data set is obtained by the network device through the identification information of the optimal receiving spatial filter reported by the terminal device, and the optimal transmitting spatial filter determined by spatial filter scanning.

24. The method according to any one of claims 1 to 23, characterized in that The method further comprises: The network device obtains a second data set; wherein the second data set includes at least one of the following: identification information of the spatial filter of K2 time units, link quality information of K2 time units, identification information of the spatial filter of F2 time units, and link quality information of F2 time units; wherein K2 and F2 are both positive integers; The network device trains the first network model according to the second data set to obtain model parameters of the first network model.

25. The method according to any one of claims 1 to 23, characterized in that The method further comprises: The network device receives the trained model parameter information of the first network model.

26. A wireless communication method, characterized in that: include: The terminal device obtains a third data set, wherein the third data set includes identification information of the spatial filter of K3 time units and link quality information of K3 time units, where K3 is a positive integer; The terminal device inputs the third data set into a second network model and outputs second prediction information, wherein the second prediction information includes identification information of the spatial filter and link quality information of F3 time units, wherein at least one check point is set in the F3 time units, and F3 is a positive integer; The terminal device determines whether to use the spatial filter output by the second network model within n time units after the first check point based on the link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F3.

27. The method according to claim 26, wherein The spatial filter comprises an optimal reception spatial filter; or, The spatial filter comprises an optimal transmit spatial filter; or, The spatial filter includes an optimal transmission spatial filter and an optimal reception spatial filter.

28. The method according to claim 26 or 27, wherein When the number of the at least one check point is greater than or equal to 2, different check points are spaced apart by n time units.

29. The method according to any one of claims 26 to 28, characterized in that The terminal device determines, based on a link quality check result of a first check point, whether to use a spatial filter output by the second network model within n time units after the first check point, including: If the link quality check at the first check point succeeds, the terminal device uses the spatial filter output by the second network model within n time units after the first check point; and / or, In a case where the link quality check at the first check point fails, the terminal device determines a spatial filter to be used by spatial filter scanning within n time units after the first check point.

30. The method according to any one of claims 26 to 29, wherein After the first checkpoint check is completed, the method further includes: The terminal device updates the third data set according to the link quality verification result of the first verification point; The terminal device inputs the updated third data set into the second network model, outputs updated second prediction information, and determines an updated first check point in the updated F3 time units included in the updated second prediction information; The terminal device determines whether to use the spatial filter output by the second network model within n time units after the updated first check point in the updated F3 time units based on the link quality check result of the updated first check point.

31. The method of claim 30, wherein: If the link quality check of the first check point succeeds, the updated K3 time units included in the updated third data set include m time units of the F3 time units, and the updated K3 time units do not include the first m time units of the K3 time units, and the updated first check point is located in the mth time unit of the updated F3 time units; and / or, In a case where the link quality check of the first check point fails, the updated K3 time units included in the updated third data set include the m+n time units of the F3 time units, and the updated K3 time units do not include the first m+n time units of the K3 time units, and the updated first check point is located at the m+nth time unit of the updated F3 time units; Among them, m time units among the F3 time units are the time unit to which the first check point belongs and the time unit before it among the F3 time units, and m is a positive integer.

32. The method according to claim 30 or 31, wherein The starting position of the first check point and the value of n are dynamically configured by the network device based on the mobility of the terminal device, or the starting position of the first check point and the value of n are dynamically configured by the terminal device based on the check results in multiple past time units.

33. The method according to any one of claims 26 to 29, wherein The starting position of the at least one check point and the value of n are configured by the network device, or the starting position of the at least one check point and the value of n are agreed upon by a protocol.

34. The method according to any one of claims 26 to 29, wherein The starting position of the at least one check point and the value of n are determined by the network device or the terminal device based on the check configuration group; The verification configuration group includes a plurality of verification configuration types, and each of the plurality of verification configuration types corresponds to a starting position of a verification point and a value of n.

35. The method of claim 34, wherein: The verification configuration group is agreed upon by a protocol, or the verification configuration group is pre-configured by a network device.

36. The method according to any one of claims 26 to 35, wherein The link quality verification result of the first verification point is determined based on the link quality information of the first verification point and a link quality threshold; The link quality information of the first check point is: When the network device uses the optimal transmit spatial filter corresponding to the first check point to send a reference signal, the terminal device uses the link quality information measured by the optimal receive spatial filter corresponding to the first check point.

37. The method of claim 36, wherein: If the link quality information of the first verification point is greater than or equal to the link quality threshold, the link quality verification of the first verification point succeeds; and / or, When the link quality information of the first verification point is less than the link quality threshold, the link quality verification of the first verification point fails.

38. The method of claim 36 or 37, The link quality verification result of the first verification point is determined by the network device based on the link quality information of the first verification point reported by the terminal device and the link quality threshold, and is indicated to the terminal device; or, The link quality check result of the first check point is determined by the terminal device based on the link quality information of the first check point and the link quality threshold.

39. The method according to any one of claims 36 to 38, wherein The link quality threshold is determined based on an average value or a weighted average value of the link quality information of the K3 time units.

40. The method of claim 39, wherein The link quality threshold is determined based on the following formula: r_th=(Q1+Q2+…+Q K3 ) / K3–X; Among them, r_th represents the link quality threshold, Q1 to Q K3 They respectively represent the link quality information of the K3 time units, and X represents the link quality threshold offset value.

41. The method of claim 39, wherein: The link quality threshold is determined based on the following formula: r_th=(a1*Q1+a2*Q2+…+a K3 *Q K3 ) / K3–X; Among them, r_th represents the link quality threshold, Q1 to Q K3 Respectively represent the link quality information of the K3 time units, a1 to a K3 They represent the weighted weights of the K3 time units, and a1+a2+…+a K3 =1, X represents the link quality threshold offset value.

42. The method of claim 41, wherein <h2 style=";text-align:left;direction:ltr">a1<=a2<=…<=a<h2 style=";text-align:left;direction:ltr"> K3 <h2 style=";text-align:left;direction:ltr"> 。 43. The method according to any one of claims 40 to 42, wherein The value of X is determined based on the channel quality.

44. The method according to any one of claims 40 to 43, wherein The value of X is configured by the network device, or the value of X is configured by the terminal device.

45. The method according to any one of claims 36 to 38, wherein The link quality threshold is agreed upon by a protocol, or the link quality threshold is configured by a network device.

46. ​​The method according to any one of claims 36 to 45, wherein The optimal transmit spatial filter corresponding to the first check point is indicated by the network device to the terminal device.

47. The method according to any one of claims 26 to 46, wherein In the case where the link quality check of the first check point fails, the method further includes: The terminal device sends second indication information, wherein the second indication information is used to instruct the network device to determine the spatial filter to be used by spatial filter scanning within n time units after the first check point.

48. The method according to any one of claims 26 to 47, wherein In the case where the spatial filter includes an optimal receiving spatial filter, the third data set is acquired by the terminal device through spatial filter scanning; or In the case where the spatial filter includes an optimal transmit spatial filter, the third data set is obtained by the terminal device from a network device; or In the case where the spatial filter includes an optimal transmitting spatial filter and an optimal receiving spatial filter, the third data set is obtained by the identification information of the optimal transmitting spatial filter indicated by the terminal device through the network device, and the optimal receiving spatial filter determined by spatial filter scanning.

49. The method according to any one of claims 26 to 48, wherein The method further comprises: The terminal device obtains a fourth data set; wherein the fourth data set includes at least one of the following: identification information of the spatial filter of K4 time units, link quality information of K4 time units, identification information of the spatial filter of F4 time units, and link quality information of F4 time units; wherein K4 and F4 are both positive integers; The terminal device trains the second network model according to the fourth data set to obtain model parameters of the second network model.

50. The method according to any one of claims 26 to 48, wherein The method further comprises: The terminal device receives the trained model parameter information of the second network model.

51. A network device, characterized in that: include: An acquiring unit, configured to acquire a first data set, wherein the first data set includes identification information of a spatial filter of K1 time units and link quality information of K1 time units, where K1 is a positive integer; a processing unit, configured to input the first data set into a first network model and output first prediction information, wherein the first prediction information includes identification information of a spatial filter of F1 time units and link quality information of F1 time units, wherein at least one check point is set in the F1 time unit, and F1 is a positive integer; The processing unit is further configured to determine whether to use the spatial filter output by the first network model within n time units after the first check point based on a link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F1.

52. A terminal device, characterized in that: include: an acquiring unit, configured to acquire a third data set, wherein the third data set includes identification information of the spatial filter of K3 time units and link quality information of K3 time units, where K3 is a positive integer; a processing unit, configured to input the third data set into a second network model, and output second prediction information, wherein the second prediction information includes identification information of the spatial filter and link quality information of F3 time units, wherein at least one check point is set in the F3 time units, and F3 is a positive integer; The processing unit is further configured to determine whether to use the spatial filter output by the second network model within n time units after the first check point based on a link quality check result of the first check point, wherein the first check point belongs to the at least one check point, n is a positive integer, and n<F3.

53. A terminal device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, so that the terminal device executes the method according to any one of claims 1 to 25.

54. A network device, characterized in that include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, so that the network device executes the method according to any one of claims 26 to 50.

55. A chip, characterized in that: include: A processor, configured to call and execute a computer program from a memory, so that a device equipped with the chip executes the method according to any one of claims 1 to 25.

56. A chip, characterized in that The device comprises: a processor configured to call and run a computer program from a memory, so that a device equipped with the chip executes the method according to any one of claims 26 to 50.

57. A computer-readable storage medium, characterized in that For storing a computer program, when the computer program is executed, the method according to any one of claims 1 to 25 is implemented.

58. A computer-readable storage medium, characterized in that For storing a computer program, when said computer program is executed, the method according to any one of claims 26 to 50 is implemented.

59. A computer program product, characterized in that The method comprises computer program instructions, which, when executed, implement the method according to any one of claims 1 to 25.

60. A computer program product, characterized in that The method comprises computer program instructions which, when executed, implement the method according to any one of claims 26 to 50.

61. A computer program, characterized in that When the computer program is executed, the method according to any one of claims 1 to 25 is implemented.

62. A computer program, characterized in that When the computer program is executed, the method according to any one of claims 26 to 50 is implemented.

Images