Method and apparatus for wireless communication

Abstract

A method and apparatus for wireless communication are provided. The method includes: a terminal device receiving first configuration information, the first configuration information being used for indicating an inference configuration; the terminal device performing measurement prediction and / or measurement event prediction according to the inference configuration; based on a result of the measurement prediction and / or the measurement event prediction, the terminal device determining whether to send a predicted measurement report; wherein the inference configuration comprises a triggering manner of a prediction event, the triggering manner of the prediction event being determined according to an entering condition and / or a confidence threshold of the prediction event.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a method and apparatus for wireless communication. Background Technology

[0002] Based on the measurement configurations issued by network devices, terminal devices perform radio resource management (RRM) measurements on the serving cell and neighboring cells, and trigger measurement reports based on measurement events to support functions such as mobility control, cell reselection, and handover. With the significant increase in the number of cells and frequencies requiring monitoring, overhead and latency present considerable challenges. Therefore, artificial intelligence (AI) / machine learning (ML) technologies have been introduced. However, while maintaining the traditional RRM measurement and measurement event framework, how to utilize AI / ML to predict RRM measurements and measurement events is a technical issue that needs to be considered. Summary of the Invention

[0003] This application provides a method and apparatus for wireless communication. The various aspects related to the embodiments of this application are described below.

[0004] In a first aspect, a method for wireless communication is provided, comprising: a terminal device receiving first configuration information, the first configuration information being used to indicate inference configuration; the terminal device performing measurement prediction and / or measurement event prediction according to the inference configuration; and based on the results of the measurement prediction and / or the measurement event prediction, the terminal device determining whether to send a predicted measurement report; wherein the inference configuration includes a triggering mode for the predicted event, the triggering mode for the predicted event being determined according to an entry condition and / or a confidence threshold for the predicted event.

[0005] In a second aspect, a method for wireless communication is provided, comprising: a network device sending first configuration information to a terminal device; and the network device making a mobility decision based on a predicted measurement report sent by the terminal device; wherein the first configuration information is used to indicate an inference configuration, the inference configuration is used by the terminal device to perform measurement prediction and / or measurement event prediction, the result of the measurement prediction and / or the measurement event prediction is used by the terminal device to determine whether to send a predicted measurement report, and the inference configuration includes a triggering method for the predicted event, the triggering method for the predicted event being determined based on an entry condition and / or a confidence threshold for the predicted event.

[0006] Thirdly, an apparatus for wireless communication is provided, the apparatus being a terminal device, the apparatus comprising: a receiving unit for receiving first configuration information, the first configuration information being used to indicate inference configuration; a first processing unit for performing measurement prediction and / or measurement event prediction according to the inference configuration; and a second processing unit for determining whether to send a predicted measurement report based on the results of the measurement prediction and / or the measurement event prediction; wherein the inference configuration includes a triggering method for the predicted event, the triggering method for the predicted event being determined according to the entry condition and / or confidence threshold of the predicted event.

[0007] Fourthly, an apparatus for wireless communication is provided, the apparatus being a network device, the apparatus comprising: a transmitting unit for transmitting first configuration information to a terminal device; and a processing unit for performing mobility decisions based on a predicted measurement report transmitted by the terminal device; wherein the first configuration information is used to instruct inference configuration, the inference configuration being used by the terminal device to perform measurement prediction and / or measurement event prediction, the result of the measurement prediction and / or the measurement event prediction being used by the terminal device to determine whether to send a predicted measurement report, and the inference configuration including a triggering method for the predicted event, the triggering method for the predicted event being determined based on the entry conditions and / or confidence threshold of the predicted event.

[0008] Fifthly, a communication device is provided, including a memory and a processor, the memory for storing a program, and the processor for calling the program in the memory to perform the method as described in the first or second aspect.

[0009] A sixth aspect provides an apparatus including a processor for calling a program from memory to perform the method as described in the first or second aspect.

[0010] A seventh aspect provides a chip including a processor for calling a program from memory, causing a device on which the chip is mounted to perform the method as described in the first or second aspect.

[0011] Eighthly, a computer-readable storage medium is provided having a program stored thereon that causes a computer to perform the method as described in the first or second aspect.

[0012] Ninth aspect, a computer program product is provided, including a program that causes a computer to perform the method as described in the first or second aspect.

[0013] In a tenth aspect, a computer program is provided that causes a computer to perform the method as described in the first or second aspect.

[0014] In this embodiment, the terminal device performs measurement prediction and / or measurement event prediction according to the inference configuration indicated by the first configuration information, and determines whether to send a predicted measurement report based on the prediction results. The triggering method of the predicted event, included in the inference configuration, is determined based on the entry conditions and / or confidence thresholds of the predicted event. Under the traditional RRM measurement and measurement event framework, network devices can configure the triggering conditions of predicted events and the sending of measurement reports based on entry conditions and / or confidence thresholds, thereby improving the efficiency and performance of mobility control and RRM. Attached Figure Description

[0015] Figure 1 This is a system architecture example diagram of a wireless communication system to which the embodiments of this application are applicable.

[0016] Figure 2 This is a schematic diagram of the network architecture applicable to the embodiments of this application.

[0017] Figure 3A and Figure 3B This is a schematic diagram of the structure of the wireless protocol stack applicable to the embodiments of this application.

[0018] Figure 4 This is a schematic diagram of a neuron in a neural network to which the embodiments of this application apply.

[0019] Figure 5 This is a schematic diagram of the neural network to which the embodiments of this application apply.

[0020] Figure 6 This is a schematic diagram of a convolutional neural network applicable to the embodiments of this application.

[0021] Figure 7 This is a flowchart illustrating the lifecycle management process of AI / ML.

[0022] Figure 8 This is a flowchart illustrating a method for wireless communication proposed in an embodiment of this application.

[0023] Figure 9 This is a schematic diagram illustrating the prediction of a measurement event in advance based on a duration T.

[0024] Figure 10 This is a schematic diagram illustrating the prediction of target measurement events within the first window.

[0025] Figure 11 This is a flowchart illustrating the process of predicting measured events.

[0026] Figure 12 This is a schematic diagram of one type of state machine transition.

[0027] Figure 13 yes Figure 8The flowchart illustrates one implementation of the method shown.

[0028] Figure 14 yes Figure 8 A flowchart illustrating another implementation of the method shown.

[0029] Figure 15 This is a schematic diagram of a device for wireless communication provided in an embodiment of this application.

[0030] Figure 16 This is a schematic diagram of another device for wireless communication provided in an embodiment of this application.

[0031] Figure 17 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0033] Communication system architecture

[0034] Figure 1 This is a system architecture example diagram of a wireless communication system 100 applicable to embodiments of this application. The wireless communication system 100 may include a network device 110 and a terminal device 120. The network device 110 may be a device that communicates with the terminal device 120. The network device 110 may provide communication coverage for a specific geographical area and may communicate with the terminal device 120 located within that coverage area.

[0035] Figure 1 An example is shown of a network device and multiple terminal devices, such as... Figure 1 Terminal devices 120a to 120j are included. Optionally, the wireless communication system 100 may include multiple network devices, and each network device may include other numbers of terminal devices within its coverage area; this embodiment does not limit this.

[0036] Optionally, the wireless communication system 100 may also include other network entities such as a network controller and a mobility management entity, which is not limited in this embodiment.

[0037] It should be understood that the technical solutions of the embodiments of this application can be applied to various communication systems, such as: 5th-generation (5G) systems or new radio (NR) systems, long-term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, advanced long-term evolution (LTE-A) systems, enhanced 5G (5G advanced) systems, etc. The technical solutions provided in this application can also be applied to future communication systems, such as 6th-generation (6G) mobile communication systems, satellite communication systems, etc.

[0038] The communication system in this application embodiment can be applied to carrier aggregation (CA) scenarios, dual connectivity (DC) scenarios, and standalone (SA) network deployment scenarios.

[0039] The terminal device in this application embodiment can also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station (MS), mobile terminal (MT), remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user device. The terminal device in this application embodiment can be a device that provides voice and / or data connectivity to a user, and can be used to connect people, objects, and machines, such as a handheld device with wireless connectivity, vehicle-mounted device, etc. The terminal device in the embodiments of this application may be a mobile phone, tablet computer, laptop computer, handheld computer, camera equipment, mobile internet device (MID), wearable device, virtual reality (VR) device, augmented reality (AR) device, wireless terminal in industrial control, wireless terminal in self-driving, wireless terminal in remote medical surgery, wireless terminal in smart grid, wireless terminal in transportation safety, wireless terminal in smart city, wireless terminal in smart home, etc. Optionally, the terminal device may be used to act as a base station. For example, the terminal device may act as a scheduling entity, providing sidelink signals between UEs in vehicle-to-everything (V2X) or device-to-device (D2D) connections. For example, cellular phones and cars communicate with each other using sidelink signals. Cellular phones and smart home devices can communicate without relaying communication signals through base stations.

[0040] The network device in this application embodiment can be a device for communicating with terminal devices. This network device can also be called an access network device or a radio access network device, such as a base station (BS). In this application embodiment, the network device can refer to a radio access network (RAN) node or a next-generation RAN (NG-RAN) node (or device) that connects user equipment to a wireless network. A base station can broadly encompass, or be replaced by, various names including: NodeB, evolved NodeB (eNB), next-generation NodeB (gNB), relay station, transmitting and receiving point (TRP), transmitting point (TP), master station (MeNB), secondary station (SeNB), multi-mode radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), positioning node, etc. A base station can be a macro base station, micro base station, relay node, donor node, or similar, or a combination thereof. A base station can also refer to a communication module, modem, or chip installed within the aforementioned equipment or apparatus. Base stations can also be mobile switching centers, devices that perform base station functions in D2D, V2X, and machine-to-machine (M2M) communications, network-side devices in 6G networks, and devices that perform base station functions in future communication systems. Base stations can support networks using the same or different access technologies. The embodiments of this application do not limit the specific technologies or device forms used in the network equipment.

[0041] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move depending on the location of the mobile base station. In other examples, a helicopter or drone can be configured as a device to communicate with another base station.

[0042] In some deployments, the network device in this application embodiment may refer to a CU or a DU, or the network device may include both a CU and a DU. The gNB may also include an AAU.

[0043] Network devices and terminal devices can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on airplanes, balloons, and satellites. This application does not limit the scenario in which the network devices and terminal devices are located.

[0044] It should be understood that all or part of the functions of the communication device in this application can also be implemented by software functions running on hardware, or by virtualization functions instantiated on a platform (e.g., a cloud platform).

[0045] Figure 2 A schematic diagram of a network architecture 200 according to an embodiment of this application is illustrated. This network architecture 200 describes the network architecture of a 5G NR / LTE / LTE-A system, which can also be referred to as a 5G system (5GS) / evolved packet system (EPS) network architecture. The network architecture 200 includes at least one of the following: network device 110, terminal device 120, 5G core network (5GC) / evolved packet core (EPC) 210, home subscriber server (HSS) / unified data management (UDM) 220, and Internet service 230. Figure 2 The network devices and terminal devices in the diagram are illustrated using RAN and UE as examples, respectively.

[0046] like Figure 2As shown, network device 110 provides user plane and control plane protocol termination to terminal device 120. Network device 110 is connected to 5GC / EPC 210 via an S1 / NG interface. 5GC / EPC 210 includes a mobility management entity (MME) / authentication management field (AMF) / session management function (SMF) 211, other MMEs / AMFs / SMFs 214, a service gateway (S-GW) / user plane function (UPF) 212, and a packet data network gateway (P-GW) / UPF 213. MME / AMF / SMF 211 is the control node that handles signaling between terminal device 120 and 5GC / EPC 210. Generally, MME / AMF / SMF 211 provides bearer and connection management. All user Internet Protocol (IP) packets are transmitted through the S-GW / UPF212, which is itself connected to the P-GW / UPF213. The P-GW provides UE IP address allocation and other functions. The P-GW / UPF213 is connected to Internet service 230. Internet service 230 includes operator-compliant Internet Protocol services, specifically including the Internet, intranet, IP multimedia subsystem (IMS), and packet-switched streaming services. It is evident that network architecture 200 provides packet-switched services; however, those skilled in the art will readily understand that the various concepts presented herein can be extended to networks providing circuit-switched services or other cellular networks.

[0047] Figure 3A and Figure 3B The following are schematic diagrams of the wireless protocol stack structure of one embodiment of this application. Figure 3A and Figure 3B This introduction uses the 5G wireless protocol stack as an example. The 5G wireless protocol stack is divided into two planes: the user plane (UP) protocol stack and the control plane (CP) protocol stack. The user plane protocol stack contains the protocol suite used for user data transmission, while the control plane protocol stack contains the protocol suite used for control signaling transmission in the 5G system. The specific names of each protocol stack layer are as follows:

[0048] like Figure 3AAs shown, the user plane protocol stack, from top to bottom, includes: the Service Data Adaptation Protocol (SDAP) layer, the Packet Data Convergence Protocol (PDCP) layer, the Radio Link Control (RLC) layer, the Medium Access Control (MAC) layer, and the Physical (PHY) layer.

[0049] like Figure 3B As shown, the control plane protocol stack, from top to bottom, includes: non-access stratum (NAS); radio resource control (RRC) layer, PDCP layer, RLC layer, MAC layer, and PHY layer.

[0050] It should be understood that the different layers in the above protocol stack have different functions, and they work together through inter-layer interaction to achieve communication between terminal devices and network devices. With the development of artificial intelligence technology, AI-assisted computing has permeated the processing implementation methods of the above protocol stack. For example, the scheduling algorithm of the MAC layer and the encoding / decoding algorithm of the PHY layer can apply artificial intelligence algorithms to improve the performance of communication algorithms.

[0051] As an example, Figure 3A and Figure 3B The wireless protocol architecture described herein is applicable to the terminal devices used in this application, such as UEs.

[0052] As an example, Figure 3A and Figure 3B The wireless protocol architecture described herein is applicable to network devices used in this application, such as gNBs.

[0053] It should be understood that the interpretation of the terminology in the embodiments of this application may refer to the TS36, TS37 and TS38 series of specifications of the 3rd generation partnership project (3GPP), but may also refer to the specifications of the Institute of Electrical and Electronics Engineers (IEEE).

[0054] To facilitate understanding, some related technical knowledge involved in the embodiments of this application is first introduced. The following related technologies are optional solutions and can be arbitrarily combined with the technical solutions of the embodiments of this application, all of which fall within the protection scope of the embodiments of this application. The embodiments of this application include at least some of the following contents.

[0055] Neural Networks

[0056] AI research, exemplified by neural networks, has achieved significant results in many fields and will continue to impact people's lives and work for a long time to come. A neural network can be understood as a computational model composed of multiple interconnected neurons. In a neural network, the connection strength between nodes can be represented as the weighted values ​​corresponding to the input signals, also known as parameters. Each neuron performs a weighted summation of different input signals and outputs the result through a specific activation function. Neurons can achieve nonlinear mappings depending on the activation function.

[0057] by Figure 4 Taking the neuron shown as an example, the input of the neuron can be denoted as A, and each dimension of the input can be denoted as a. j The corresponding weighted value is denoted as w. j Where j takes values ​​of 1, 2, ..., n. The neuron's input can also be configured with a bias term to adjust the output, such as... Figure 4 The constant 1 in the input (corresponding to the weighting value denoted as b) is used. The weighting value, along with the summation units (SUs), enhances or weakens the input. The output of the SU can be input into the activation function f to obtain the output t.

[0058] Common neural networks include convolutional neural networks (CNN), recurrent neural networks (RNN), and deep neural networks (DNN).

[0059] The following text combines Figure 5 This application describes the neural network to which the embodiments are applicable. Figure 5 The neural network shown can be divided into three categories according to the position of different layers: input layer 510, hidden layer 520, and output layer 530. Generally speaking, the first layer is the input layer 510, the last layer is the output layer 530, and the intermediate layers between the first and last layers are hidden layers 520.

[0060] The input layer 510 is used to input data, which may be, for example, a received signal received by a receiver. The hidden layer 520 is used to process the input data, for example, to decompress the received signal. The hidden layer may also be called an intermediate layer. The output layer 530 is used to output the processed output data, for example, to output the decompressed signal.

[0061] See Figure 5 A neural network consists of multiple layers, each containing multiple neurons. Neurons between layers can be fully connected or partially connected. For connected neurons, the output of a neuron in one layer can serve as the input to a neuron in the next layer.

[0062] To facilitate understanding, we will use CNN as an example below, combined with... Figure 6 Examples of multiple layers in a neural network are provided. A CNN is a deep neural network with convolutional structures. For example... Figure 6 As shown, the structure of a CNN may include an input layer 610, a convolutional layer 620, a pooling layer 630, a fully connected layer 640, and an output layer 650. The convolutional layer 620, pooling layer 630, and fully connected layer 640 are the intermediate layers of this CNN.

[0063] It should be noted that, as Figure 6 The CNN shown is only an example of a convolutional neural network. In specific applications, convolutional neural networks can also exist in the form of other network models, and this application does not limit them.

[0064] RRM measurement

[0065] In communication systems (e.g., NR and subsequent 6G wireless communication systems), network devices (e.g., gNB) typically issue measurement configurations to enable terminal devices (e.g., UE) to perform RRM measurements on the serving cell and neighboring cells (also known as neighboring cells), and trigger measurement reports based on measurement events to support functions such as mobility control, cell reselection, and handover.

[0066] As an example, RRM measurement includes the measurement of the received signal to determine the received signal power, received signal quality, etc.

[0067] As an example, the measurement events include events A1-A6, as detailed below:

[0068] A1: The quality of the service area is higher than the threshold (quality is better than the threshold);

[0069] A2: The quality of the service area is below the threshold (quality is worse than the threshold);

[0070] A3: The quality of a neighboring cell is more than one offset higher than the currently serving cell;

[0071] A4: The quality of a neighboring cell is higher than an absolute threshold;

[0072] A5: The quality of the serving cell is below threshold 1 and the quality of a neighboring cell is above threshold 2 (double threshold event);

[0073] A6: The quality of another neighboring cell is one offset higher than the quality of the current (or selected) neighboring cell (commonly used in CA / multi-cell scenarios).

[0074] With increasingly dense network deployments, continuous expansion of carriers and frequency bands, and widespread adoption of multi-antenna and beamforming technologies, the number of cells and frequency points that terminal devices need to monitor has increased significantly. Traditional "measurement + reporting" mechanisms based solely on periodicity and events face significant challenges in terms of measurement overhead, feedback overhead, and latency, which can easily limit further improvements in mobility performance and user experience.

[0075] To address the aforementioned technical challenges, standardization organizations such as 3GPP have begun introducing air interface enhancement mechanisms based on artificial intelligence / machine learning (AI / ML). This mechanism aims to utilize AI / ML models to predict channel quality, RRM measurements, and mobility events, thereby achieving advance and refined control of mobility decisions without requiring a significant increase in actual measurements and reporting. For example, by deploying trained AI / ML models on the terminal or network equipment side, it is possible to predict RRM measurement trends over a future period based on historical measurement results and environmental information, or predict the future triggering of a specific measurement event, enabling the network to anticipate potential handover needs and prepare resources accordingly.

[0076] As an example, the AI / ML model on the terminal device side can support RRM measurement prediction, that is, implement the RRM measurement prediction function. The RRM measurement prediction function, for example, takes the received traditional cell measurement signal as input and uses the AI / ML model to predict future signal conditions as output. Use cases for RRM measurement prediction include time-domain prediction case A, time-domain prediction case B, and frequency-domain prediction. The predictions in time-domain prediction case A (also called case A) and time-domain prediction case B (also called case B) depend on time-domain correlation, while frequency-domain prediction depends on frequency-domain correlation.

[0077] In the above embodiments, the output of time-domain prediction case A is typically a measurement value for a future time period. The output of time-domain prediction case B is typically a measurement value for the current time instance. For skip mode example 1 of time-domain prediction cases A and B, the input / output is the same, the difference being that time-domain prediction case B can skip multiple measurement instances to derive the next prediction result. For skip mode example 2 of time-domain prediction case B, the input within the observation window (OW) is discontinuous, which differs from skip mode example 1 of time-domain prediction cases A and B.

[0078] In the above embodiments, frequency domain prediction refers to calculating the predicted frequency based on the measured frequency. The measured frequency and the predicted frequency cannot differ too much; otherwise, the terminal device will be unable to make the prediction. Therefore, the terminal device can support inter-frequency prediction, but the frequency difference should be less than a threshold. This threshold is a maximum frequency difference. For example, a common frequency domain prediction scenario could be a co-location scenario.

[0079] As an example, the AI / ML model on the terminal device side can support measurement event prediction. The measurement events A1-A6 mentioned above are included in the scope of measurement event prediction. In the scenario of measurement event prediction, the prediction window (PW) is usually a sliding window, and the terminal device needs to perform prediction multiple times.

[0080] In the above embodiments, the prediction of measurement events can be achieved through two prediction methods: direct measurement event prediction and indirect measurement event prediction. Direct measurement events can be predicted directly by an AI / ML model, meaning the model's output is the probability of the event occurring. In RRM measurement event detection, indirect measurement event prediction refers to using the prediction output of the RRM measurement prediction function as input to the relevant measurement event detection framework, thereby enabling the detection of future handover events. Relevant measurement events include, for example, existing 5G NR events, 6G events, etc.

[0081] In the above embodiments, the prediction of measurement events can be determined based on the prediction results of RRM measurement prediction. RRM measurement prediction allows for better prediction of RRM measurement events. This means that these two functions are closely related in AI / ML-based switching operations. In some cases, RRM measurement prediction and measurement event prediction can use the same AI model. In other cases, because the terminal device needs to post-process the prediction results of RRM measurement prediction in measurement event prediction, different AI models can be used for RRM measurement prediction and measurement event prediction.

[0082] As an example, AI / ML models can perform operations such as activation, deactivation, inference, and monitoring through lifecycle management (LCM) and performance monitoring mechanisms. The following section will combine... Figure 7 This paper provides an exemplary illustration of the overall LCM process for the air interface AI / ML. Figure 7 This is presented from the perspective of the interaction between the terminal device (e.g., UE) and the network (NW). The dashed line indicates that this process is optional.

[0083] See Figure 7 In step S710, the network side sends a capability query (UECapabilityEnqiry) to the terminal device.

[0084] In step S720, the terminal device sends capability information (UECapabilityInformation) to the network side.

[0085] In step S730, the network side sends an RRC Reconfiguration message to the terminal device.

[0086] In step S740, the terminal device sends an applicable functionality reporting report to the network side.

[0087] In step S750, the network side sends an RRC reconfiguration message to the terminal device. This step is optional.

[0088] In step S760, the terminal device and the network side jointly perform operations such as AI / ML activation / deactivation / inference / monitoring.

[0089] like Figure 7 As shown, the terminal device and the network side will interact by querying capabilities and reporting capability information. Through these interactions, the network device can configure resources and functions based on the capabilities of the terminal device.

[0090] The above text combined Figure 7 This paper introduces the predictive capabilities and related monitoring required for introducing AI / ML models into RRM measurements and measurement events. However, the predictions and LCM of AI / ML models may impact the RRM measurement and measurement event framework. Maintaining the existing RRM measurement and measurement event framework (e.g., A1-A6) presents several challenges for AI / ML model-based predictions and LCM.

[0091] For example, how to organically combine the predicted measurement results or predicted events output by AI / ML with the decision logic of existing events.

[0092] For example, how can the protocol layer support unified configuration and capability indicators for various prediction modes on the terminal device side, such as time-domain prediction case A, time-domain prediction case B, and frequency-domain prediction, including prediction window length, prediction time resolution, number of supported cells, and frequency range?

[0093] For example, how can we use model LCM and performance monitoring mechanisms to adjust or roll back the prediction function in a timely manner when the wireless environment or model performance changes, so as to avoid incorrect predictions leading to incorrect switching or service interruption?

[0094] For example, in the scenario of predicting measurement events, the terminal device needs to perform predictions multiple times. How can we define the triggering conditions for the terminal device to report inference results to avoid excessive signaling overhead due to frequent changes in prediction results?

[0095] In summary, while maintaining the traditional RRM measurement and measurement event framework, how to use AI / ML to predict RRM measurements and measurement events is a technical issue that needs to be considered.

[0096] To address the aforementioned issues, this application proposes a method for wireless communication. In this method, a terminal device performs measurement prediction and / or measurement event prediction based on an inference configuration indicated by first configuration information, and determines whether to send a predicted measurement report based on the prediction results. The inference configuration includes a triggering method for the predicted event determined based on the entry conditions and / or confidence thresholds of the predicted event. Within the traditional RRM measurement and measurement event framework, network devices can configure the triggering conditions for predicted events and the transmission of measurement reports based on entry conditions and / or confidence thresholds, thereby improving the efficiency and performance of mobility control and RRM.

[0097] To facilitate understanding, the following will be combined with... Figure 8 The methods proposed in the embodiments of this application will be described in detail. Figure 8 It is presented from the perspective of the interaction between terminal devices and network devices.

[0098] The terminal device can be any of the terminals or terminal-side devices mentioned above, such as a UE.

[0099] In some embodiments, the terminal device can communicate with the network device. As one embodiment, the terminal device can receive configuration information sent by the network device to perform RRM measurements and / or determine measurement events.

[0100] In some embodiments, the terminal device may be a communication device deployed with an AI / ML model. The AI / ML model deployed on the terminal device can be used for inference or prediction. When the AI / ML model is deployed on a terminal-side device corresponding to the terminal device, the terminal-side device may be an auxiliary device communicating with the terminal device, or it may be another communication device relative to the network device side. For example, the terminal-side device may be a network relay.

[0101] As an example, the AI / ML model can be a model related to measurement prediction and / or measurement event prediction. Measurement prediction includes RRM measurement prediction. The AI / ML model can also combine non-AI algorithms for inference or prediction.

[0102] As one example, the terminal device can deploy an AI / ML model for measurement prediction and measurement event prediction. In other words, the terminal device can perform measurement prediction and measurement event prediction using an AI / ML model.

[0103] As one example, the terminal device can deploy a first AI / ML model for measurement prediction and a second AI / ML model for measurement event prediction. That is, the terminal device performs measurement prediction and measurement event prediction using two different AI / ML models respectively.

[0104] As an example, the AI / ML model deployed on the terminal device can support time-domain prediction and frequency-domain prediction. For instance, the AI / ML model can support time-domain prediction case A and time-domain prediction case B, as well as frequency-domain prediction for co-located cases.

[0105] As an example, the AI / ML model deployed on the terminal device side can support layer 3 (L3) cell-level prediction.

[0106] In some embodiments, the terminal device may be equipped with at least one transmitting antenna and at least one receiving antenna for wireless transmission.

[0107] In some embodiments, the cell where the terminal device is located is the first cell. The terminal device communicates with the network device corresponding to the first cell.

[0108] The network device can be any of the network devices or network (NW) side devices described above, such as a base station. The network device can be the network device corresponding to the first cell. The network device can provide services to all terminal devices in the first cell.

[0109] In some embodiments, the network device may deploy an AI / ML model and train the AI / ML model. As one embodiment, the network device may collect data to construct a dataset related to the AI / ML model. As another embodiment, the network device may train the AI / ML model based on the collected data and send the trained model to the terminal device.

[0110] In some embodiments, for RRM measurement prediction and measurement event prediction, the network device can specify signaling and protocol aspects to enable LCM function management of the AI / ML model on the terminal device side. Through the interaction between the terminal device and the network device, it is also possible to implement the terminal device capability request and response process, and the applicability reporting of complete and partial configuration methods.

[0111] As an example, the applicability report may include Figure 7 The applicable functions are shown in the report.

[0112] As one example, for applicability reports, the terminal device can receive the complete inference configuration and / or partial inference configuration in an RRC configuration message or an RRC reconfiguration message. For example, the network device can configure the complete inference configuration, and the terminal device reports the applicability status to the network via an RRC Reconfiguration Complete message, an RRC Resume Complete message, and UE assistance information (UAI). Alternatively, the network device can configure the partial inference configuration via an OtherConfig message, and the terminal device can report the applicability status to the network via RRC Reconfiguration Complete and UAI or update the applicability via UAI. Then, the network device provides the complete inference configuration based on the received applicability status.

[0113] As an example, after receiving the inference configuration (in full or in part) via an RRC reconfiguration message, the terminal device can report its applicability in the initial suitability report via an RRC configuration or reconfiguration completion message. If the inference configuration is not applicable, the terminal device's report may include a flag indicating that it prefers to send it to the NW. When the terminal device indicates that the inference configuration is not applicable, the network device is expected to publish it, i.e., it does not support the terminal device's autonomous publication. Alternatively, the terminal device can update the suitability via UAI. For these configurations, FFS on RRC reconfiguration has been completed.

[0114] As an example, when transitioning to the RRC_INACTIVE state, the end device retains the inference configuration (full and / or partial inference configuration). The end device does not automatically release the inference configuration when it becomes unsuitable. However, the end device can automatically fall back to a non-AI operation. Alternatively, the end device can add an applicability report for the inference configuration in RRCResumeComplete.

[0115] As an example, an RRC reconfiguration complete message can include the applicability / inapplicability status of all predicted configurations included in the previous RRC reconfiguration message. Network devices can also introduce a flag in the OtherConfig carried in the RRC configuration to indicate whether applicability reporting via UAI is enabled. When an inferred configuration becomes inapplicable, the terminal device should report its inapplicability via UAI. During mobility, the target network device can check its applicability—i.e., RRC reconfiguration—by including the inferred configuration of the target cell in the handover command message, and the terminal device can respond with an RRC reconfiguration complete message.

[0116] It should be noted that in the embodiments of this application, the model can also be replaced by a function, such as an AI / ML model being an AI / ML function.

[0117] See Figure 8 , Figure 8 The process shown includes steps S810 to S830, which are described below. It should be noted that... Figure 8 The method shown may also include other steps, which are not limited in this application embodiment.

[0118] In step S810, the terminal device receives first configuration information. The first configuration information comes from the network device.

[0119] The first configuration information is used to indicate the inference configuration. This first configuration information may include relevant parameters for the inference configuration, enabling the terminal device to perform related inference or prediction based on the AI / ML model. For example, measurement prediction and measurement event prediction based on the AI / ML model.

[0120] In some embodiments, the first configuration information can be determined based on first capability information sent by the terminal device. The first capability information can indicate the terminal device's capabilities for measurement prediction and measurement event prediction. The terminal device's capabilities for RRM measurement prediction and measurement event prediction can be understood as its capabilities regarding AI mobility. The system can define the terminal device's capabilities separately for RRM measurement prediction and measurement event prediction. The terminal device's capabilities can also be shared during the RRM measurement prediction and measurement event prediction processes.

[0121] As one embodiment, measurement prediction may include reference signal received power (RSRP) measurement prediction, reference signal received quality (RSRQ) measurement prediction, signal to interference plus noise ratio (SINR) measurement prediction, etc., and this application embodiment does not limit this.

[0122] As an example, measurement event prediction can be used to predict the likelihood of triggering a predicted event. That is, by predicting a predicted event, the probability or likelihood of the predicted event being triggered can be output. The predicted event can be a definite or specified event, or it can be an indefinite event. When the predicted event is a specified or configured event, it is also called the target measurement event.

[0123] As an example, the predicted event may include at least one of events pA1-pA6. pA1-pA6 are labels for predicted versions A1-A6. In the AI ​​scenario, A1-A6 correspond to pA1-pA6 to represent predicted events A1…A6. The decision logic is still based on RRM measurements such as RSRP / RSRQ / SINR. RRM measurements may include predicted RRM measurements and / or measured RRM measurements.

[0124] As one embodiment, the predicted event may include at least one of events A1-A6. The cell quality in events A1-A6 can be determined based on measured results, predicted results, or both.

[0125] In some embodiments, the first capability information includes at least one of the following: capability information related to time-domain prediction; capability information related to frequency-domain prediction; and capability information related to measurement event prediction. That is, the capabilities of the terminal device can be defined separately for time-domain prediction, frequency-domain prediction, and measurement event prediction. Among these, the capability information related to time-domain prediction and the capability information related to frequency-domain prediction belong to the capability information related to measurement prediction, i.e., frequency-domain prediction capability.

[0126] As an example, time-domain prediction may include time-domain prediction within the same frequency range. The minimum set of capabilities for a terminal device to perform time-domain prediction within the same frequency range includes at least one of the following: the maximum PW length that the terminal device can support for RSRP measurement prediction; the frequency list (or frequency range) that the terminal device can support for RSRP measurement prediction, which may be provided in FR; and the maximum total number of cells (including serving cells and neighboring cells) that the terminal device can support for RSRP measurement prediction.

[0127] Network devices can configure the PW length based on the capabilities of the terminal devices. PW length configuration in the time domain can support duration configuration; for example, the NW can configure the PW length as a single duration value (e.g., xx milliseconds). PW length configuration can also support configuring the number of instances and time intervals; for example, the NW can configure the number of time instances to be predicted within the PW and the time interval between consecutive time instances.

[0128] As an example, frequency domain prediction may include inter-frequency prediction. The basic capabilities of a terminal device for inter-frequency prediction include at least one of the following: for each target (predicted) frequency, a range or list of input frequencies that the terminal device needs to measure; for each frequency, the terminal device can support a maximum number of neighboring cells for RSRP measurement prediction (neighboring cells are all juxtaposed with the serving cell); the terminal device can support a maximum total number of cells (including serving cells and neighboring cells) for RSRP measurement prediction.

[0129] As an example, the basic capability of the terminal device to perform measurement event prediction includes at least one of the following: indicating whether it supports measurement event prediction for each predicted event type (pA1, ..., pA6) (pA1 represents predicted event A1); and for each event type, indicating the maximum prediction time length (PW) supported by the terminal device for measurement event prediction. The maximum PW time for each event type can be the same, or each event type can have its own prediction time length for the maximum PW.

[0130] As an example, time-domain prediction includes multiple use case types, and the capability information related to time-domain prediction includes multiple sub-information corresponding to the multiple use case types. For example, time-domain prediction includes Case A and Case B. As mentioned above, the inputs and / or outputs of Case A and Case B may be different; therefore, time-domain prediction Case A and Time-domain prediction Case B can each have separate capability information, i.e., sub-information. That is, the capability information related to time-domain prediction includes first sub-information and second sub-information. The first sub-information is used to indicate the capability supporting frequency-domain prediction Case A, and the second sub-information is used to indicate the capability supporting frequency-domain prediction Case B.

[0131] In some embodiments, predicted events can be determined based on predictions made using RRM measurements. That is, different prediction results correspond to different predicted events. In this case, it is not necessary to introduce different capability information to indicate support for different event types separately. Therefore, support for different event types is indicated by the same capability information.

[0132] In other embodiments, the terminal device needs to post-process the measurement prediction results in the measurement event prediction process. Therefore, different capability information (sub-information) can be introduced to indicate support for different event types respectively. Alternatively, the network can be configured with one or more combinations of capabilities for different predicted events, and the terminal device can report which events the prediction capability only supports.

[0133] As one example, the capability information related to the prediction of measurement events includes multiple sub-information items. These sub-information items correspond one-to-one with multiple predicted events, or multiple sub-information items correspond one-to-one with combinations of multiple predicted events.

[0134] As one embodiment, the first capability information may include a combination of capability information related to measurement event prediction and capability information related to measurement prediction. The network device can determine whether the terminal device supports measurement event prediction for a specific use case (e.g., time domain, frequency domain) based on the relevant combination reported by the terminal device. If the terminal device supports measurement event prediction, then the terminal device should be able to perform at least one of the prediction scenarios, such as time domain prediction case A, time domain prediction case B, and frequency domain prediction.

[0135] For example, network devices or terminal devices can indicate support for measurement event prediction in time-domain prediction case A through the terminal device's capabilities for time-domain prediction case A and its capabilities for measurement event prediction.

[0136] For example, network devices or terminal devices can indicate support for measurement event prediction in time-domain prediction case B through the terminal device's capabilities for time-domain prediction case B and its capabilities for measurement event prediction.

[0137] For example, network devices or terminal devices can indicate support for measurement event prediction and frequency domain prediction through the time-frequency domain prediction capability and the measurement event prediction capability of the terminal device.

[0138] Inference configuration can also be called measurement configuration or inference parameter configuration. The inference configuration in the embodiments of this application can be applied to all the following use cases: intra-frequency prediction, time-domain prediction case A, time-domain prediction case B, and inter-frequency prediction.

[0139] As one embodiment, the inference configuration can be used to instruct the terminal device to predict the measurement objects, and / or, the terminal device to predict the combination of measurement objects / cell combinations.

[0140] In some embodiments, inference configuration can be used to configure information related to cell measurement prediction. For each serving cell and neighboring cell configured for prediction, the network device can configure the required measurement objects for the terminal device based on the terminal device's prediction capabilities.

[0141] In some embodiments, the RRM prediction report can be configured based on existing events (A1-A6) and predicted events (e.g., pA1-pA6). A1-A6 are six standard RRM measurement events in 5G NR. pA1-pA6 are simply labels for the predicted versions of A1-A6. Thus, one or more predicted events in the inference configuration are determined based on RRM measurement events, maintaining the framework of existing RRM events.

[0142] The inference configuration includes the triggering method for predicted events to indicate triggering conditions. The triggering method for predicted events can be determined based on the entry conditions and / or confidence thresholds of the predicted events. As one embodiment, the triggering method for predicted events can be determined solely based on the entry conditions and / or confidence thresholds of the predicted events. As another embodiment, in addition to entry conditions and / or confidence thresholds, the triggering method for predicted events is also determined based on the prediction time. That is, the triggering method for predicted events is determined based on the prediction time and the entry conditions and / or confidence thresholds.

[0143] As an example, when the triggering method of the predicted event is determined solely based on the entry conditions and / or confidence threshold of the predicted event, the first configuration information includes one or more of the following: type information of the predicted event or the measurement event corresponding to the predicted event; entry condition information; parameter information of measurement prediction and / or measurement event prediction; confidence threshold information; and flag information of the triggering method of the predicted event.

[0144] In another embodiment, when the triggering method of the predicted event is determined based on the prediction time and the entry conditions and / or confidence threshold, the first configuration information also includes timer information for the predicted event. That is, the first configuration information includes one or more of the following: type information of the predicted event or the corresponding measurement event; information on the entry conditions; parameter information for measurement prediction and / or measurement event prediction; information on the confidence threshold; timer information for the predicted event; and flag information for the triggering method of the predicted event.

[0145] As an example, the type information of the predicted event or the corresponding measured event is used to indicate that the type of the predicted event is one of A1 to A6, or one of its predicted versions (e.g., pA1 to pA6).

[0146] As an example, entry condition information is used to indicate the entry condition parameters for an event, such as threshold values, biases, hysteresis values, etc. The system can define an entry condition for each predicted event (e.g., an RRM measurement event or a predicted version).

[0147] As an example, parameter information for measuring forecasts and / or measuring event forecasts is used to indicate the forecast lead time or forecast time window (e.g., T). pred This includes prediction time windows, prediction periods, etc. The prediction time window can also be called the first window. The duration of the prediction lead time can be called the first duration.

[0148] As an example, timer information for a predicted event indicates the duration of timer T for that event. This timer information can also be called T-information or T-condition. The timer can be timed based on a first duration or a first window. The duration of the timer can also be represented by T. The timer information can indicate the prediction interval corresponding to the predicted event, such as interval T.

[0149] As an example, the confidence threshold information is used to indicate the confidence threshold Thr.

[0150] As an example, the flag information is used to indicate the triggering method of the predicted event, such as an entry condition method.

[0151] For example, the first configuration information may include one or more of the following: event type information, event condition parameter information, T information, prediction-related parameter information, confidence threshold information, and triggering method flag information.

[0152] Inference configuration is used by the terminal device to perform measurement prediction and / or measurement event prediction. The terminal device makes relevant predictions based on the inference configuration.

[0153] In some embodiments, the measurement prediction is an RRM measurement prediction, and the inference configuration of the RRM measurement prediction may be carried in the RRM measurement configuration. The RRM measurement report indicated by the RRM measurement configuration includes predicted measurement results and / or measured measurement results. Predicted measurement results may also be referred to as predicted values ​​or predicted measurement values, and measured measurement results may also be referred to as measured values, measured values, or actual measurement results.

[0154] As an example, it is confirmed that the inference parameter configuration for RRM measurement prediction can be included in the existing RRM measurement configuration. In this case, the RRM measurement report needs to be configured with both AI and non-AI results. For example, the terminal device may only obtain predictions for a subset of neighboring cells, while for other cells, it may need to perform and provide the actual measurement results. Alternatively, the terminal device may estimate insufficient prediction accuracy based on performance monitoring and may provide measured values ​​instead of predicted values. Furthermore, the terminal device may need to provide both measured and predicted measurement results simultaneously for network-side performance monitoring.

[0155] As an example, the predicted measurement results and the measured measurement results are associated based on at least one of the following identifiers (IDs): measurement ID, measurement object ID, cell ID, and physical cell identifier (PCI).

[0156] For example, the same measurement ID can be associated with both actual and predicted measurement configurations. The associated ID should be configurable for both training and inference. End devices may need to report both actual and predicted measurement results associated with the same measurement configuration. Network devices should be able to distinguish between predicted and actual measurements. When a measurement ID is associated with both actual and predicted measurement configurations, the end device needs to indicate whether the corresponding measurement result is actual or predicted.

[0157] For example, the ID used to associate predicted measurement results with measured measurement results is called the association ID. The association ID is unique within a public land mobile network (PLMN). The association ID can be 12 bits, 24 bits, 36 bits, 48 ​​bits, etc. An association ID can be associated with one or more cells. Network devices can use the parameter allowedCellsToAddModList in existing RRC messages or the newly configured parameter predictedCellsToAddModList to indicate the list of selectable cells for time-domain or frequency-domain prediction. The newly configured parameter can be found in the RRC-related information.

[0158] For example, the actual measurement configuration and the predicted measurement configuration may use the same measurement configuration. This means that the terminal device needs to report both the actual and predicted measurement results associated with the same measurement ID. This includes cases where the predicted and actual measurements are included in a single report message, as well as cases where different report messages include either the predicted or actual measurements. However, the network device needs to know whether the reported result is a predicted or actual measurement; therefore, an indication is required in the measurement report to indicate whether a particular result is a predicted or actual measurement.

[0159] In some embodiments, the inference configuration for measurement event prediction includes the following elements: a predefined length of the processing window (e.g., PW) associated with the AI / ML model used for prediction, and event-related information, which may include the event type and associated event-specific parameters. When measurement event prediction is for RRM measurement event prediction, the event prediction report may include RRM measurement values, time-related information about the predicted event (e.g., the time / window of occurrence), RRM measurement data, and time information related to the predicted event, such as the expected time of occurrence or the occurrence window. Regarding the reporting process for measurement events, it should be noted that when a measurement event is determined and the terminal device is configured with time-domain prediction case B and frequency-domain prediction, the terminal device can report the measurement event in a manner similar to conventional processes. For indirect measurement event prediction, the measurement-related configuration will primarily follow the configuration for RRM measurement prediction. However, the key difference lies in when and how the measurement report is sent. Regarding time-domain prediction, if future measurement results can be predicted, then the predicted event at a future time can be predicted.

[0160] As an example, the terminal device needs to have the basic capability to perform event prediction. The terminal device can choose the type of event prediction to use, either indirectly measuring event prediction or directly measuring event prediction.

[0161] In some embodiments, the first configuration information may be carried in RRC signaling. For example, the first configuration information may be an RRC reconfiguration message or an RRC configuration message. The RRC reconfiguration message or RRC configuration message may include inference configuration.

[0162] In step S820, the terminal device performs measurement prediction and / or measurement event prediction based on the inference configuration. Measurement prediction and measurement event prediction are implemented using the same AI / ML model, or they are implemented using different AI / ML models.

[0163] In some embodiments, measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result. The result of RRM measurement prediction is the predicted RRM measurement value. That is, the result of measurement prediction includes the predicted measurement result and / or the corresponding confidence level. The result of measurement prediction and / or the measured measurement result is used to predict measurement events. For example, the predicted measurement result and / or the measured measurement result are used to predict measurement events. Or, the predicted measurement result and its confidence level and / or the measured measurement result are used to predict measurement events.

[0164] As an example, the measurement prediction results may include: predicted RRM measurements m at multiple future times. pred (cell,t+kΔ); the confidence value corresponding to the predicted RRM measurement or the probability information related to the triggering of the predicted event. mpred (cell, t+kΔ) represents the RRM measurement value of the terminal device for cell at the k-th prediction time t+kΔ in the future, for example, predicting RSRP / RSRQ / SINR. Here, t is the current time; Δ is the prediction time step; and k is a non-negative integer index (k = 1, 2, ...), representing the k-th future sampling point.

[0165] Optionally, Δ is also called the sampling interval, such as 10ms or 20ms, which is determined by the network device configuration or the terminal device itself.

[0166] As one example, the measurement prediction results are used to determine the valid measurement value for making a decision on the predicted event. This valid measurement value can be solely the predicted measurement result, or it can be a combination of the predicted and measured measurement results. In other words, the valid measurement value can be a predicted value, or a fusion of predicted and measured values.

[0167] For example, the effective measurement value can be m(cell,t+kΔ)=α×m1(cell,t+kΔ)+(1-α)×m2(cell,t0), where α is the fusion weight between 0 and 1, which can be determined by network configuration or local policy; m(cell,t+kΔ) represents the effective RRM measurement value for cell at the k-th prediction time t+kΔ in the future; m1(cell,t+kΔ) is the predicted measurement result obtained based on the AI / ML model; m2(cell,t0) is the measured measurement result obtained at time t0; cell is the cell identifier, which can be the serving cell or a neighboring cell, corresponding to the PCI, cell ID or measurement object ID of the serving cell or a neighboring cell; t represents the current reference time / baseline time when making the prediction.

[0168] As an example, the measurement prediction results are used to determine the confidence index for interval T. The confidence index can be the minimum confidence value of each prediction point within interval T, or it can be the average or weighted average of the confidence values ​​of each prediction point within interval T.

[0169] For example, the confidence index could be the minimum confidence value of each prediction point within the T interval. T =min∈[t index ,t index +T]conf(k).

[0170] In some embodiments, the terminal device performs measurement prediction for a second time at a first time point to obtain a predicted measurement result for the second time point. The predicted measurement result for the second time point is used to evaluate the measurement event at the second time point. The second time point is later than the first time point, and the duration between the first time point and the second time point is a first duration. The first duration can be determined based on predefined information or network device configuration information.

[0171] As an example, the first duration can be the timer duration T mentioned above. The first duration can represent the duration of the interval T.

[0172] As one example, the configuration of the first duration can instruct the terminal device to make an advance prediction of whether a target measurement event will be met at a single future time (e.g., t+T). The first duration can represent a "lead time".

[0173] As an example, the terminal device can make an advance prediction before the first duration (duration T), such as Figure 9 As shown. See also Figure 9 The terminal device predicts the RRM measurement result at time t+T and evaluates event Ax at time t+T based on the predicted RRM test result. Relative to the current time, the terminal device performs an event prediction for event Ax a first time interval in advance.

[0174] exist Figure 9 In the method shown, the terminal device at the current time t only cares about a fixed future time: t+T. When the network device needs to know the triggering status of the Ax event at least a first duration in advance, after receiving the advance prediction measurement report, the network device can perform proactive handover preparation to improve mobility performance.

[0175] In other embodiments, the terminal device performs measurement predictions on multiple times within a first window at a first moment to obtain multiple predicted measurement results. These multiple predicted measurement results are used to determine one or more predicted times that trigger a target measurement event within the first window, or the distribution information of the target measurement event triggering within the first window. This distribution information represents the probability and / or temporal distribution of the target measurement event triggering.

[0176] As one example, the length of the first window is a second duration, which is less than or equal to the first duration. For instance, the length of the first window is less than or equal to the duration of interval T. Interval T may include multiple sampling times or multiple sampling points.

[0177] As one embodiment, the terminal device can predict the time when a target measurement event will be fulfilled, that is, predict the time when the target measurement event will be triggered. The terminal device can predict at time t when the event is expected to be fulfilled within a certain period, for example, (t, t+T). Figure 10 As shown.

[0178] See Figure 10At the current time t, the terminal device is concerned with a time interval: (t, t+T'), where T' is the length of the first window (the second duration), which can be less than or equal to T. Within this time interval, the device predicts when event Ax will be satisfied. The prediction of event Ax is performed within the time interval from t to t+T'; that is, the terminal device repeatedly examines the prediction results at many points along the time axis (t, t+T') to find t' (the time of the first satisfaction) or to obtain the entire time distribution. Figure 10 In this case, t' = t + T'.

[0179] Figure 10 The second duration T' in the formula is used to indicate the time when the terminal device will predict the satisfaction time of the target measurement event within the time period (t, t+T'). T' represents the length of the prediction window triggered by the event, and T is the length of the entire prediction window.

[0180] Figure 10 The method shown can be viewed as multiple operations of a pre-prediction method. The terminal device is not limited to a single moment t+T, but rather predicts the fulfillment of a target measurement event within a pre-configured future time period (e.g., from t to t+T'). The terminal device needs to determine the predicted time (e.g., t') when the target event is first fulfilled within this time period, or determine temporal information representing the probability distribution of the target test event triggering within this time period. The terminal device can report the predicted time and / or temporal distribution, along with the corresponding predicted measurement value, confidence level, and other information, to the network device. This allows the network device to flexibly select the actual time point to trigger mobility operations throughout the entire time period, or to make more refined scheduling and switching decisions based on the prediction reliability of different candidate time points.

[0181] In some embodiments, measurement event prediction is performed on one or more prediction events based on the triggering method of the prediction events. The one or more prediction events include at least one target measurement event. The at least one target measurement event corresponds to one of the following: a measurement object, a cell, a combination of measurement objects, or a combination of cells.

[0182] In step S830, based on the results of measurement prediction and / or measurement event prediction, the terminal device determines whether to send a prediction measurement report. The prediction measurement report may include the results of measurement prediction or the results of measurement event prediction. For RRM, the prediction measurement report is also the RRM prediction report. When the prediction measurement report includes the results of measurement event prediction, the report can also be called an event prediction report. For measurement event prediction, the terminal device explicitly indicates in the inference report when the predicted event will occur. For measurement event prediction, since the prediction window is sliding, the terminal device needs to perform predictions multiple times. If the terminal device reports that an event is predicted every time, the signaling overhead will increase. To avoid sending reports frequently, the terminal device can send the prediction measurement report only when certain conditions are met.

[0183] As one example, the predictive measurement report sent by the terminal device is used by the network device to make mobility decisions, such as handover preparation.

[0184] As an example, if the network device configures the maximum PW length and the predicted occurrence time of the event is within the configured window, the terminal device sends an event prediction report. Alternatively, if the network device configures an event prediction report for the terminal device and sets a series of values ​​for the predicted occurrence time of the event, the terminal device sends an event prediction report if the predicted occurrence time of the event is within this range. For example, if the terminal device's AI / ML model determines that the predicted event will occur at a future time within the network device's configured window based on actual measurements of the serving cell and / or neighboring cells, as well as other information available at the terminal device, the terminal device sends an event prediction report to the network device.

[0185] In some embodiments, the prediction measurement report is also reported when at least one of the following conditions is met: the new trigger time of the predicted event changes by more than a first threshold relative to the most recently reported trigger time; the target cell or the type of the predicted event changes; the prediction result of the predicted event within the prediction window changes from triggered to non-triggered, or from non-triggered to triggered; or the confidence level of the prediction measurement result does not fall within a pre-configured confidence threshold range.

[0186] As an example, the first threshold can be determined based on predefined information or the configuration information of the network device.

[0187] As an example, the type of event to be predicted may change, for example, from predicting event A3 to predicting event A5.

[0188] As an example, a change in the prediction result can refer to a difference between the current prediction and the previous (or a specified) prediction. For example, the predicted event in the previous report was predicted to trigger within PW, but this prediction finds that it will not trigger within PW. This prediction finding that the event will not trigger within PW includes: this prediction finding that there are no longer any moments within PW that meet the entry conditions, or that all confidence levels are below the threshold.

[0189] As an example, the confidence level of the predicted measurement result not falling within the confidence threshold interval can be understood as the confidence level of the predicted measurement result crossing the confidence threshold interval. For example, two confidence thresholds can be introduced: a high threshold and a low threshold. H Low threshold Conf L (Conf H >Conf L ).

[0190] In the above embodiments, the two confidence thresholds can form a confidence hysteresis. The terminal device can then determine the confidence level based on the previous confidence level (Conf). prev ) and the confidence level (Conf) new The relationship between the magnitude of the two confidence thresholds determines whether to send a predictive measurement report and the content of the report.

[0191] For example, from Conf prev <Conf L Become Conf new >Conf H At that time, the terminal device sends an update stating that "the prediction has become reliable".

[0192] For example, from Conf prev >Conf H Become Conf new <Conf L At that time, the terminal device sends an update stating that "the prediction is no longer reliable".

[0193] For example, the small fluctuations in the middle (in Conf) L ~Conf H (between) no report is triggered.

[0194] As one example, the terminal device can update the results by sending multiple prediction measurement reports. In this case, the prediction measurement report can also be called an updated inference report. The update cycle can be determined based on predefined or configuration information. For example, the network device can further control the update frequency of the inference report by configuring parameters such as the minimum update interval, the maximum silence interval, and the time quantization granularity.

[0195] As an example, the terminal device may send a prediction measurement report only when any of the following occurs: the change in the new prediction event trigger time relative to the previously reported prediction event trigger time exceeds a first threshold; the target cell or the type of the prediction measurement event changes; the prediction result changes from triggering the prediction event within the prediction window to no longer triggering the prediction event within the prediction window or vice versa; the prediction confidence level crosses a pre-configured confidence level threshold range.

[0196] As an example, the prediction measurement report may include at least one of the following: a measurement identifier and event type indicating the corresponding prediction event; indication information for indicating that this report is based on the prediction results, such as prediction indication bits, prediction event type identifiers, etc.; prediction measurement results and / or valid measurement values ​​of one or more cells; and confidence index and / or prediction trigger time information carrying the T interval.

[0197] As an example, for time-domain prediction case A, one or more instances of PW prediction measurement results for each cell can be reported in a single measurement report message. For time-domain prediction case B, the measurement report may include the latest measurement results (whether actual or predicted). For frequency-domain prediction, the measurement report may include the latest measurement results for the predicted cell (existing measurement reports remain unchanged).

[0198] To facilitate understanding, the following will be combined with... Figure 11 The prediction process for measuring events (predicting events) is illustrated by example. Figure 11 It is also introduced from the perspective of the interaction between terminal devices and network devices.

[0199] See Figure 11 In step S1110, the terminal device receives an RRC reconfiguration message from the network device, which includes inference configuration. The inference configuration includes event-related information. After receiving one or more complete inference configurations via the RRC reconfiguration message, the terminal device should retain all complete inference configurations, regardless of whether the complete inference structure is applicable, until the network explicitly releases it.

[0200] In step S1120, the terminal device sends an RRC reconfiguration complete message to the network device.

[0201] In step S1130, the terminal device predicts a measurement event.

[0202] In step S1140, the terminal device sends an event prediction report, also known as a prediction measurement report, to the network device. The event prediction report may include time-related information about the predicted event.

[0203] In step S1150, the network device performs handover preparation. Due to the event prediction report in step S1140, the network device can perform handover preparation in advance.

[0204] In step S1160, the measurement event is satisfied based on the actual measurement.

[0205] In step S1170, the terminal device sends a measurement report to the network device.

[0206] In step S1180, the terminal device receives an RRC reconfiguration message containing a handover command from the network device. After receiving the measurement report from the terminal device, the network device sends a handover command to the terminal device. This step can be triggered by traditional measurement events based on actual measurement results, or by publishing corresponding handover preparation-related configurations and resources based on time-related information of predicted events in the event prediction report.

[0207] In step S1190, the terminal device sends an RRC reconfiguration complete message to the network device.

[0208] like Figure 11 As shown, before reporting an event prediction, the terminal device needs to determine whether the predicted event satisfies the predicted RRM measurement results (and optional actual measurement results), i.e., whether the predicted event is predictable. For indirect measurement event prediction, there are two possible triggering conditions for the event prediction report. One possible triggering condition is that the predicted RRM measurement results (and optional actual measurement results) satisfy the input conditions corresponding to the predicted event. The other possible triggering condition is that the predicted RRM measurement results (and optional actual measurement results) satisfy the entry conditions of the predicted event, i.e., all measurements (including the predicted measurement and the actual measurement over a certain period T) are satisfied.

[0209] It should be noted that if the network device sends an RRC configuration (handover command) upon receiving an event prediction report, handover latency can be further reduced. If the handover is triggered based on actual measurement results, and the measurement event may not actually be completed in the future, the terminal device can release the stored RRC configuration.

[0210] The above text combined Figures 8 to 11 This paper introduces various implementation methods for terminal devices to perform measurement prediction and / or measurement event prediction based on initial configuration information. These methods are based on AI / ML for prediction, enabling terminal devices to predict RRM measurement values ​​and events within the existing RRM measurement and event framework. Furthermore, through standardized capability indications, configuration and reporting mechanisms, and model monitoring and rollback mechanisms, these methods address issues such as high measurement and reporting overhead, difficulty in controlling prediction event signaling overhead, and lack of unified management during model training and inference phases. This improves the efficiency and performance of mobility control and RRM while ensuring protocol controllability and backward compatibility.

[0211] In some embodiments, measurement event prediction can be controlled based on transitions between multiple states. These multiple states can be states for one or more predicted events. The multiple states can include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state. It should be understood that these five states are merely examples, and the multiple states may also include other states.

[0212] The definitions of idle state, candidate state, predicted activation state, reported state, and cleared state are described below with examples.

[0213] In the idle state, the terminal device has not yet detected any trend that meets the triggering conditions of the predicted event. For example, there are no valid prediction candidates, and no corresponding predicted event has been reported to the network side. The idle state can be the initial state by default, or it can be the return state after the predicted event has been cleared.

[0214] In a candidate state, the terminal device has detected that the entry condition for a predicted event (e.g., a target measurement event) on the prediction timeline has been met, but the combination of the entry condition and condition T has not yet been met, or the confidence level of the T interval has not yet been evaluated. Candidate states are typically used to represent potential predicted events that are about to be triggered.

[0215] In the prediction activation state, the terminal device has determined the following results on the prediction timeline: the entry condition and condition T will be met within a certain future time period, and the corresponding confidence index is not lower than the threshold; or, the entry condition will be met at a certain future moment, and the confidence index at that moment is not lower than the threshold. In this case, the prediction event is considered prediction activation, but the corresponding measurement report has not yet been generated or sent; or the terminal device can maintain a prediction validity timer (e.g., T) in this state. valid ), used to limit the effective duration of prediction results.

[0216] In the reported state, the terminal device has sent at least one predictive measurement report to the network device based on events in the predictive active state. The network side can perform or prepare to perform mobility decisions based on this report. In this state, the terminal device can continuously monitor whether the predictive conditions still hold, and can trigger report updates or revocations if necessary.

[0217] The clear state characterizes the release process of predicted or reported events. The clear state can be entered by the terminal device based on local determination or by explicit instructions from the network device. After performing the corresponding clearing operation (e.g., resetting internal variables, stopping related timers), it can return to the idle state.

[0218] As an example, the prediction of measurement events can be controlled based on the transitions between the five states defined above. When conditions are met between the five states, a state transition occurs, changing from one state to another.

[0219] As one example, the measurement event prediction can determine multiple states to be transitioned based on the first capability information reported by the terminal device. The first capability information can indicate whether all five states are supported, or only one or a few of them. That is, the terminal device does not need to fully support all five states. For example, the terminal device may only support the idle state, predict the active state, and the reported state. Alternatively, the terminal device may only support the idle state, predict the active state, and the cleared state.

[0220] In some embodiments, multiple states correspond to a single state machine, meaning that the transitions between multiple states are managed by a single state machine. A state machine can control the transitions between multiple states based on transition conditions and / or instructions from network devices. Based on the state machine, the control of predictive measurement events can be achieved. Terminal devices can utilize finite state machines to uniformly manage the prediction, decision-making, and reporting of measurement events, thereby ensuring prediction reliability while reducing the probability of false triggers.

[0221] As one example, a state machine is used to maintain at least one target measurement event. For instance, for each measurement event (e.g., A3 / pA3) and each measurement object / cell combination, the terminal device may maintain a separate state machine. Alternatively, depending on the implementation, a state may manage multiple measurement events or multiple measurement objects / cell combinations in a combined manner.

[0222] As one example, the terminal device may maintain one or more state machines for the target measurement event. The overall structure of the state machine may be determined based on the states supported by the terminal device. For example, the state machine may include at least the following five states: S0, idle state; S1, candidate state; S2, predicted activation state; S3, reported state; S4, cleared state.

[0223] To facilitate understanding, the following will be combined with... Figure 12 The transitions and triggering conditions between multiple states are illustrated by example. Figure 12 The structure and transition method of the state machine are shown. Figure 12 The state machine in the example includes five states, namely S0 to S4 mentioned above.

[0224] See Figure 12 The transition from an idle state to other states involves two transition paths: S0→S1 (idle → candidate) and S0→S2 (idle → predicted activation).

[0225] The condition for a state machine to transition from an idle state to a candidate state is the condition for the prediction to be met. For example, based on the prediction results output by the AI / ML model, the terminal device determines that the entry condition for the target event is met for the first time at a certain future moment on the prediction timeline (e.g., the condition that A3's neighboring cells are better than the serving cell). Another example is that, for a certain future moment, the predicted confidence value is not lower than a second threshold value. The second threshold value can be the same as or slightly lower than the final threshold, for example, Conf... thr1 .

[0226] Terminal devices can directly transition from an idle state to a prediction active state when certain transition conditions are met, omitting candidate states. These transition conditions may include entry conditions and / or confidence levels. For example, for a given future time, the prediction result meets the entry conditions for the predicted event. Another example is that the corresponding confidence value is greater than or equal to a confidence threshold. Thr For example, network configuration instructions can be set to a timer with no set duration.

[0227] Furthermore, if no prediction result satisfying the entry condition and with a confidence level not lower than the minimum threshold is detected on the prediction timeline, the state machine remains in the idle state. In other words, the result of determining whether S0 should transition also includes S0→S0 (remaining idle).

[0228] like Figure 12 As shown, the transition from candidate state to other states includes two transition paths: S1→S2 (candidate → predicted activation) and S1→S4 (candidate → clearing).

[0229] When using the entry condition + T + confidence threshold scheme, the event transitions from the candidate state to the prediction activation state if all of the following conditions are met: On the prediction timeline, the continuous time length from the future moment when the entry condition is first met is not shorter than the prediction interval corresponding to T; or, within the prediction interval T, the entry condition of the event is continuously met at all or most of the sampling points; or, the confidence index (e.g., minimum or average value) calculated based on the corresponding confidence value within the prediction interval T is not lower than the confidence threshold.

[0230] Once the state machine transitions to the prediction activation state, the terminal device can record information such as the target cell of the predicted event, the expected trigger time, and the corresponding confidence index.

[0231] If any of the following situations occur while in the candidate state, the state machine can roll back to the idle state: Subsequent prediction results indicate that the entry condition for the predicted event no longer holds on the prediction timeline, meaning there is no longer a future moment that satisfies the entry condition; or, although a future moment that satisfies the entry condition still exists, the confidence value is lower than a preset threshold when recalculating the confidence index for interval T; or, the candidate state duration exceeds a preset upper limit, or the prediction validity timer expires. Upon rolling back to the idle state, the terminal device can clear temporary variables associated with the candidate event.

[0232] like Figure 12 As shown, the transition from the predicted active state to other states includes two transition paths: S2→S3 (predicted active → reported) and S2→S4 (predicted active → cleared).

[0233] The state machine can transition from the prediction activation state to the reported state when one of the following transition conditions is met: The terminal device successfully generates a measurement report and sends at least one measurement report containing prediction event information, provided that the constraints such as the reporting cycle and the number of reporting times are met; the measurement report carries information such as the prediction event identifier, the prediction measurement result, and optional confidence index and prediction trigger time.

[0234] If any of the following occurs before or during the preparation for reporting, the state machine can transition from the prediction activation state to the clear state: subsequent prediction results indicate that the target event will no longer trigger within the originally planned prediction period, or the predicted trigger time has substantially shifted; or the confidence value recalculated for the updated prediction result is below the threshold; or the prediction validity timer expires, indicating that the previous prediction conclusion is invalid. In this case, the terminal device can choose not to send the prediction measurement report again, or send a revocation message as needed.

[0235] like Figure 12 As shown, the transition from a reported state to other states includes a transition path, namely S3→S4 (reported → cleared).

[0236] The state machine can transition from the reported state to the cleared state when any of the following transition conditions are met: the terminal device determines, based on new prediction results and / or measured results, that the reported predicted event is no longer valid (e.g., the prediction trigger time has passed but the event has not actually triggered, or the network has completed or abandoned the corresponding mobility operation); or, the terminal device receives an explicit indication from the network device (e.g., releasing or updating the measurement configuration) indicating that the predicted event no longer needs to be maintained; or, the reporting validity timer within the terminal device expires. Upon transitioning to the cleared state, the terminal device may optionally send update or cancellation information to the network device to reflect the change in the prediction conclusion.

[0237] like Figure 12 As shown, the transition from the clear state to other states includes a transition path, namely S4→S0 (clear → idle).

[0238] In the clear state, the state machine returns to the idle state after the terminal device completes the following operations: clearing cached information such as prediction results, confidence indices, and trigger times related to the target measurement event; or stopping or resetting the timer associated with the prediction event; or restoring to the default measurement and prediction monitoring logic.

[0239] After returning to the idle state, the terminal device can restart a new round of state machine processes based on the new prediction results.

[0240] As an example, if the triggering method of the predicted event is determined solely by the entry condition and confidence threshold, without considering prediction time-to-trigger (TTT) or other prediction time-related information, the state machine can be configured to allow S0 to directly jump to S2, thus bypassing the candidate state S1. In this case, the "confidence level at the entry condition time" can be directly used as the basis for determining whether to transition to the prediction activation state.

[0241] By setting up a state machine, the terminal device can manage predicted measurement events using the state machine structure. The terminal device can be compatible with multiple triggering strategies for predicted events (such as schemes with or without TTT) within the same framework, and uses confidence thresholds and timers for filtering and clearing during state transitions, thereby improving the stability and controllability of predicted events.

[0242] As discussed above, predicted events can be triggered in various ways, corresponding to multiple combinations of triggering conditions. For ease of understanding, the following two examples illustrate different triggering methods for predicted events.

[0243] Example 1

[0244] In Embodiment 1, the triggering method for the predicted event is determined based on the entry conditions and / or confidence threshold. That is, Embodiment 1 is a predicted event triggering method based on entry conditions and / or confidence thresholds. When the triggering method is determined based on the entry conditions and confidence thresholds, the terminal device triggers the predicted event only based on the prediction result meeting the event's entry conditions and confidence thresholds, without displaying the duration of the prediction check on the prediction timeline, which helps to obtain a greater lead time.

[0245] Figure 13 The diagram illustrates the process by which the triggering method for the predicted event is determined based on the entry conditions and confidence threshold. Figure 13 The steps shown are performed by the terminal device; the steps performed by the network device are not shown, but will be explained below. The terminal device supports AI / ML-based RRM measurement prediction and triggers prediction event reporting when the prediction result meets the entry conditions for a measurement event and the confidence level is not lower than a threshold.

[0246] See Figure 13 In step S1310, the terminal device receives the predicted event configuration. Correspondingly, the network device sends the predicted event configuration.

[0247] As an example, the terminal device receives first configuration information from the network device. This first configuration information is as described above. The triggering method flag indicates that the predicted event triggering method used is the entry-only condition + confidence threshold method. The first configuration information may omit parameters related to the timing duration condition for checking the prediction timeline, or the duration T may be configured to zero.

[0248] In step S1320, the terminal device collects the measured RRM and performs AI / ML prediction. Through this prediction, valid measurement values ​​can be generated.

[0249] As an example, the terminal device can collect measured RRM values ​​and update historical sequences.

[0250] As an example, the terminal device can predict RRM measurements over a future period based on an AI / ML model, generating predicted RRM measurements and corresponding confidence levels. Depending on the network configuration, the terminal device can use only the predicted RRM measurements or fuse the predicted and measured values ​​to obtain effective measurements for event decision-making.

[0251] In step S1330, the terminal device determines the entry conditions and selects the earliest time when they are met. This determination can be performed on the prediction timeline.

[0252] As an example, the terminal device checks the valid measurements at each sampling time within a future time period on the prediction timeline. Specifically, for each future sampling time k, the terminal device can determine whether the entry condition is met based on the event's entry condition. If at least one sampling time meets the entry condition, the earliest prediction time index t that meets the entry condition is determined. index The corresponding confidence value is obtained; otherwise, the terminal device does not trigger the prediction event.

[0253] In step S1340, the terminal device makes a prediction event decision based on a confidence threshold. This confidence threshold corresponds to the entry condition time.

[0254] As an example, the terminal device compares the confidence value Target1 at the time of entry with the confidence threshold Target configured in the network. If Target1 is greater than or equal to Target, the terminal device determines that the predicted event will occur in the future and switches the predicted event state from idle state to predicted event active state; if Target1 is less than Target, the terminal device considers the reliability of the prediction result insufficient and does not trigger the predicted event or initiate a predicted measurement report.

[0255] In step S1350, the terminal device triggers a predictive measurement report for early decision-making on the network side.

[0256] As an example, when the terminal device determines that a predicted event is activated, it triggers the sending of a measurement report message, provided that the reporting cycle limit is met. This measurement report may include: indication information indicating that this report is a predicted event report; and / or, the corresponding event type and measurement identifier; and / or, the predicted RRM measurement result or valid measurement value of at least one target cell; and / or, information carrying the predicted trigger time, such as the expected occurrence of a predicted event at a certain time.

[0257] From the network side's perspective, it can make advance decisions based on early prediction information. As an example, after receiving a prediction event report from a terminal device based solely on entry conditions and / or confidence thresholds, the network device can utilize the greater lead time provided by the report to prepare resources and plan mobility in advance.

[0258] For example, network devices can select the target cell for handover and configure relevant parameters before the event is expected to be triggered; or, they can dynamically adjust the measurement or prediction strategies of terminal devices; or, if a decrease in confidence is found in subsequent predictions, they can revoke or modify the previous decision.

[0259] Depend on Figure 13 As can be seen, Implementation Example 1 omits the step of checking the T condition on the prediction timeline, thus enabling the prediction event report to be triggered at an earlier time point, which is beneficial for providing a more forward-looking basis for mobility decision-making.

[0260] Example 2

[0261] In Embodiment 2, the triggering method for the predicted event is determined based on the entry condition, prediction time, and confidence threshold. That is, Embodiment 2 is a predicted event triggering method based on the entry condition, prediction time, and confidence threshold. Based on the state machine described above, the path S0→S1→S2 is enabled. Compared to Embodiment 1, the terminal device needs to simultaneously check the entry condition and TTT condition on the prediction time axis; candidate state S1 mainly serves as a buffer function for when the entry condition is met but the timing length + confidence level still needs to be confirmed.

[0262] Figure 14 The diagram illustrates the process by which the triggering method for a predicted event is determined based on the entry conditions, confidence threshold, and prediction time. Figure 14 The steps shown are performed by the terminal device; the steps performed by the network device are not shown, but will be explained below. The terminal device supports AI / ML-based RRM measurement prediction and triggers prediction event reporting when the prediction result meets the entry conditions for a measurement event and the confidence level is not lower than a threshold.

[0263] See Figure 14 In step S1410, the terminal device receives the measurement prediction configuration. The network device sends the measurement prediction configuration.

[0264] As an example, the terminal device receives first configuration information from the network device. This first configuration information can configure the triggering conditions for the predicted event. Unlike Embodiment 1, the first configuration information also includes timer information for the predicted event, such as T information.

[0265] In step S1420, the terminal device collects the measured RRM and updates the historical sequence.

[0266] As an example, the terminal device performs RRM measurements (e.g., RSRP, RSRQ, SINR, etc.) on the serving cell and one or more neighboring cells at consecutive times, and obtains the measurement results for each cell. The measurement results for each cell are written into the corresponding measurement value for each cell and placed in a buffer for subsequent prediction.

[0267] In step S1430, the terminal device performs RRM prediction based on the AI / ML model.

[0268] As an example, when the prediction period is reached or the preset triggering conditions are met, the terminal device can input the historical measurement values ​​and historical information (such as location, speed, beam information, etc.) into a pre-trained AI / ML model to obtain the prediction results for a period of time in the future.

[0269] In step S1440, the terminal device generates valid measurement values ​​for event determination.

[0270] As an example, the terminal device can determine the valid measurement value for event decision-making based on the network configuration. For instance, when only predicted values ​​are used, the predicted RRM measurement value is taken as the valid measurement value. Or, when it is necessary to merge measured values, the measured RRM measurement value and the predicted RRM measurement value (such as predicted RSRP / RSRQ / SINR) are combined according to a preset function.

[0271] In step S1450, the terminal device determines the event entry conditions on the prediction time axis.

[0272] As an example, the terminal device checks the entry conditions of an event one by one at a time on the prediction timeline for one or more sampling moments within a future time period (e.g., a time period of length T after the current time). Specifically, for each future sampling moment k, it determines whether the entry condition of the corresponding event is met based on the valid measurement value m(cell,t+kΔ).

[0273] For example, for event A3, it can be determined whether the measurement difference between the neighboring cell and the serving cell is greater than the offset. If at least one sampling moment within the time period satisfies the entry condition, the terminal device determines a predicted entry time index t. index , t index The sampling time that first meets the entry condition on the prediction timeline corresponds to the sampling time when the entry condition is met; if there is no sampling time that meets the entry condition, the terminal device remains in an idle state and does not trigger the prediction event.

[0274] In step S1460, the terminal device determines condition T on the prediction time axis.

[0275] As an example, the terminal device determines the predicted entry time index t index In this case, the terminal device further checks whether the pre-configured condition T can be satisfied on the prediction time axis. Specifically: based on the duration T, it is discretized into a corresponding number of prediction sampling points to determine the prediction sampling interval [t] corresponding to T. index ,t index +T]; The terminal device determines whether the event entry condition is continuously met at every or at least most sampling times within the sampling interval; if the event condition is continuously met within the T interval, the terminal device believes that the entry condition will be met in the future; otherwise, the terminal device believes that the T condition is not met and does not trigger the predicted event.

[0276] In step S1470, the terminal device calculates the confidence index for interval T.

[0277] As an example, for cases where "entry condition + T condition" are met on the prediction timeline, the terminal device can calculate the confidence index of the T interval based on one or more confidence values ​​within the T interval.

[0278] In step S1480, the terminal device makes a prediction event decision based on the confidence threshold.

[0279] As an example, the terminal device will use the confidence index Conf of the T interval. T Confidence threshold for network configuration Thr Comparison: If Conf T Greater than or equal to Conf Thr If the terminal device determines that the target measurement event will be reliably triggered in the future, it will switch the current predicted event state from idle state to predicted event active state; if Conf T Less than Conf Thr If the terminal device deems the prediction result unreliable, it will not trigger a prediction event, or it will only use it as a local reference and will not initiate a report.

[0280] In step S1490, the terminal device triggers a predictive measurement report for network-side decision-making. The predictive measurement report indicates the predicted attributes.

[0281] As an example, when a predicted event is determined to be active, the terminal device triggers the sending of a measurement report message, provided that the reporting cycle or number of reports is met. In other words, the terminal device determines to send a predicted measurement report. The predicted measurement report is essentially a predicted event report.

[0282] From the network side's perspective, mobility decisions can be made based on predicted event reports. As an example, after receiving a predicted measurement report from a terminal device, the network device can initiate the mobility decision-making process in advance based on the predicted measurement results, the type of measurement event, and confidence information. For instance, it can pre-configure the target cell for handover before the future time period corresponding to the predicted event; adjust measurement or prediction configuration parameters; or cancel or modify planned mobility operations when a decrease in confidence is detected.

[0283] Through Example 2, the terminal device simultaneously considers the entry conditions, T conditions, and confidence thresholds given by the AI / ML model on the prediction timeline, which can effectively avoid erroneous handovers caused by unreliable predictions, thereby improving mobility control based on predicted events.

[0284] The above text combined Figures 1 to 14 The method embodiments of this application are described in detail below, in conjunction with... Figures 15 to 17 The present application provides a detailed description of the apparatus embodiments. It should be understood that the descriptions of the method embodiments correspond to the descriptions of the apparatus embodiments; therefore, any parts not described in detail can be found in the foregoing method embodiments.

[0285] Figure 15 This is a schematic block diagram of another device for wireless communication according to an embodiment of this application. The device 1500 can be any of the terminal devices described above. Figure 15 The apparatus 1500 shown includes a receiving unit 1510, a first processing unit 1520, and a second processing unit 1530.

[0286] The receiving unit 1510 can be used to receive first configuration information, which is used to indicate inference configuration.

[0287] The first processing unit 1520 can be used to perform measurement prediction and / or measurement event prediction according to the inference configuration.

[0288] The second processing unit 1530 can be used to determine whether to send a predicted measurement report based on the results of the measurement prediction and / or the measurement event prediction; wherein, the inference configuration includes a triggering method for the predicted event, and the triggering method for the predicted event is determined according to the entry conditions and / or confidence threshold of the predicted event.

[0289] Optionally, the measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result, and the predicted measurement result and / or the measured measurement result is used to predict the measurement event.

[0290] Optionally, the first processing unit 1520 is further configured to perform measurement prediction on the second time at the first time to obtain the predicted measurement result at the second time; the predicted measurement result at the second time is used to evaluate the measurement event at the second time, the second time being later than the first time, and the duration between the first time and the second time being a first duration, the first duration being determined according to predefined information or network device configuration information.

[0291] Optionally, the first processing unit 1520 is further configured to perform measurement prediction on multiple times within the first window at a first time to obtain multiple predicted measurement results; the multiple predicted measurement results are used to determine one or more predicted times that trigger the target measurement event within the first window, or the distribution information of the target measurement event triggered within the first window.

[0292] Optionally, the length of the first window is a second duration, where the second duration is less than or equal to the first duration.

[0293] Optionally, the measurement prediction is an RRM measurement prediction, the inference configuration of the RRM measurement prediction is carried in the RRM measurement configuration, and the RRM measurement report indicated by the RRM measurement configuration includes the predicted measurement results and / or the measured measurement results.

[0294] Optionally, the predicted measurement result and the measured measurement result are associated based on at least one of the following IDs: measurement ID, measurement object ID, cell ID, and PCI.

[0295] Optionally, the measurement event prediction predicts one or more prediction events based on the triggering method of the prediction event, and the one or more prediction events are determined according to the RRM measurement events.

[0296] Optionally, the first configuration information includes one or more of the following: type information of the predicted event or the measurement event corresponding to the predicted event; information of the entry condition; parameter information of the measurement prediction and / or the measurement event prediction; information of the confidence threshold; and flag information of the triggering method of the predicted event.

[0297] Optionally, the triggering method of the predicted event is also determined based on the prediction time, and the first configuration information also includes timer information for the predicted event.

[0298] Optionally, the prediction measurement report is submitted when at least one of the following conditions is met: the change in the new trigger time of the predicted event relative to the most recently submitted trigger time is greater than a first threshold; the target cell or the type of the predicted event changes; the prediction result of the predicted event within the prediction window changes from triggered to not triggered, or from not triggered to triggered; or the confidence level of the prediction measurement result does not fall within a pre-configured confidence threshold range.

[0299] Optionally, the measurement event prediction is controlled based on the transition between multiple states, which include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state.

[0300] Optionally, the plurality of states correspond to a state machine, which controls the transitions between the plurality of states based on transition conditions and / or indications from network devices, and the state machine is used to maintain at least one target measurement event.

[0301] Optionally, the apparatus further includes a transmitting unit, which can be used to transmit first capability information; wherein the first capability information is used to determine the first configuration information, and the first capability information includes at least one of the following: capability information related to time-domain prediction; capability information related to frequency-domain prediction; and capability information related to measurement event prediction.

[0302] Optionally, the time-domain prediction includes multiple use case types, and the capability information related to the time-domain prediction includes multiple sub-information corresponding to the multiple use case types.

[0303] Optionally, the capability information related to the prediction of the measurement event includes multiple sub-information; the multiple sub-information corresponds one-to-one with multiple predicted events, or the multiple sub-information corresponds one-to-one with a combination of multiple predicted events.

[0304] Optionally, the measurement prediction and the measurement event prediction are implemented using the same or different artificial intelligence / machine learning models.

[0305] Optionally, the receiving unit 1510 in device 1500 can be a transceiver 1730, and the first processing unit 1520 and the second processing unit 1530 can be processors 1710. Device 1500 may also include a memory 1720. See details below. Figure 17 .

[0306] Figure 16 This is a schematic block diagram of another device for wireless communication according to an embodiment of this application. The device 1600 can be any of the network devices described above. Figure 16 The apparatus 1600 shown includes a transmitting unit 1610 and a processing unit 1620.

[0307] The sending unit 1610 can be used to send first configuration information to the terminal device.

[0308] The processing unit 1620 can be used to make mobility decisions based on the predicted measurement report sent by the terminal device; wherein, the first configuration information is used to indicate the inference configuration, the inference configuration is used by the terminal device to perform measurement prediction and / or measurement event prediction, the result of the measurement prediction and / or the measurement event prediction is used by the terminal device to determine whether to send a predicted measurement report, the inference configuration includes the triggering method of the predicted event, the triggering method of the predicted event is determined according to the entry condition and / or confidence threshold of the predicted event.

[0309] Optionally, the measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result, and the predicted measurement result and / or the measured measurement result is used to predict the measurement event.

[0310] Optionally, the predicted measurement result includes a predicted measurement result at a second time point. The predicted measurement result at the second time point is determined based on the measurement prediction made by the terminal device at a first time point. The predicted measurement result at the second time point is used to evaluate the measurement event at the second time point. The second time point is later than the first time point. The duration between the first time point and the second time point is a first duration. The first duration is determined based on predefined information or the configuration information of the network device.

[0311] Optionally, the predicted measurement results include multiple predicted measurement results at multiple times within the first window, and the multiple predicted measurement results are determined based on the measurement prediction made by the terminal device at the first time. The multiple predicted measurement results are used to determine one or more predicted times that trigger the target measurement event within the first window, or the distribution information of the target measurement event triggered within the first window.

[0312] Optionally, the length of the first window is a second duration, where the second duration is less than or equal to the first duration.

[0313] Optionally, the measurement prediction is an RRM measurement prediction, the inference configuration of the RRM measurement prediction is carried in the RRM measurement configuration, and the RRM measurement report indicated by the RRM measurement configuration includes the predicted measurement results and / or the measured measurement results.

[0314] Optionally, the predicted measurement result and the measured measurement result are associated based on at least one of the following identifiers: measurement ID, measurement object ID, cell ID, and PCI.

[0315] Optionally, the measurement event prediction predicts one or more prediction events based on the triggering method of the prediction event, and the one or more prediction events are determined according to the RRM measurement events.

[0316] Optionally, the first configuration information includes one or more of the following: type information of the predicted event or the measurement event corresponding to the predicted event; information of the entry condition; parameter information of the measurement prediction and / or the measurement event prediction; information of the confidence threshold; and flag information of the triggering method of the predicted event.

[0317] Optionally, the triggering method of the predicted event is also determined based on the prediction time, and the first configuration information also includes timer information for the predicted event.

[0318] Optionally, the prediction measurement report is submitted when at least one of the following conditions is met: the change in the new trigger time of the predicted event relative to the most recently submitted trigger time is greater than a first threshold; the target cell or the type of the predicted event changes; the prediction result of the predicted event within the prediction window changes from triggered to not triggered, or from not triggered to triggered; or the confidence level of the prediction measurement result does not fall within a pre-configured confidence threshold range.

[0319] Optionally, the measurement event prediction is controlled based on the transition between multiple states, which include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state.

[0320] Optionally, the plurality of states correspond to a state machine, which controls the transitions between the plurality of states based on transition conditions and / or indications from network devices, and the state machine is used to maintain at least one target measurement event.

[0321] Optionally, the apparatus further includes a receiving unit, which can be used to receive first capability information sent by the terminal device; wherein the first capability information is used to determine the first configuration information, and the first capability information includes at least one of the following: capability information related to time domain prediction; capability information related to frequency domain prediction; and capability information related to measurement event prediction.

[0322] Optionally, the time-domain prediction includes multiple use case types, and the capability information related to the time-domain prediction includes multiple sub-information corresponding to the multiple use case types.

[0323] Optionally, the capability information related to the prediction of the measurement event includes multiple sub-information; the multiple sub-information corresponds one-to-one with multiple predicted events, or the multiple sub-information corresponds one-to-one with a combination of multiple predicted events.

[0324] Optionally, the measurement prediction and the measurement event prediction are implemented using the same or different artificial intelligence / machine learning models.

[0325] Optionally, the transmitting unit 1610 in device 1600 can be a transceiver 1730, the processing unit 1620 can be a processor 1710, and device 1600 may also include a memory 1720, see details below. Figure 17 .

[0326] Figure 17 The diagram shown is a structural schematic of a communication device according to an embodiment of this application. Figure 17 The dashed lines indicate that the unit or module is optional. The device 1700 can be used to implement the methods described in the above method embodiments. The device 1700 can be a chip, a terminal device, or a network device.

[0327] Apparatus 1700 may include one or more processors 1710. The processor 1710 may support apparatus 1700 in implementing the methods described in the preceding method embodiments. The processor 1710 may be a general-purpose processor or a special-purpose processor. For example, the processor may be a central processing unit (CPU). Alternatively, the processor may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0328] The apparatus 1700 may further include one or more memories 1720. The memories 1720 store a program that can be executed by the processor 1710, causing the processor 1710 to perform the methods described in the preceding method embodiments. The memories 1720 may be independent of the processor 1710 or integrated within the processor 1710.

[0329] The device 1700 may also include a transceiver 1730. The processor 1710 can communicate with other devices or chips via the transceiver 1730. For example, the processor 1710 can send and receive data with other devices or chips via the transceiver 1730.

[0330] This application also provides a computer-readable storage medium for storing a program. This computer-readable storage medium can be applied to a terminal device or network device provided in this application embodiment, and the program causes a computer to execute the methods performed by the terminal device or network device in the various embodiments of this application.

[0331] The computer-readable storage medium can be any available medium that a computer can read, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs, DVDs), or semiconductor media (e.g., solid-state disks, SSDs), etc.

[0332] This application also provides a computer program product. The computer program product includes a program. This computer program product can be applied to a terminal device or network device provided in the embodiments of this application, and the program causes a computer to execute the methods performed by the terminal device or network device in the various embodiments of this application.

[0333] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0334] This application also provides a computer program. This computer program can be applied to a terminal device or network device provided in this application, and the computer program causes the computer to execute the methods performed by the terminal or network device in various embodiments of this application.

[0335] In this application, the terms "system" and "network" are used interchangeably. Furthermore, the terminology used in this application is only for explaining specific embodiments of the application and is not intended to limit the application. The terms "first," "second," "third," and "fourth," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. In addition, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0336] In the embodiments of this application, the term "instruction" can be a direct instruction, an indirect instruction, or an indication of a relationship. For example, A instructing B can mean that A directly instructs B, such as B being able to obtain information through A; it can also mean that A indirectly instructs B, such as A instructing C, so B can obtain information through C; or it can mean that there is a relationship between A and B.

[0337] In the embodiments of this application, the term "correspondence" may indicate a direct or indirect correspondence between two things, or an association between two things, or a relationship such as instruction and being instructed, configuration and being configured.

[0338] In the embodiments of this application, "predefined" or "preconfigured" can be implemented by pre-storing corresponding codes, tables, or other means that can be used to indicate relevant information in the device (e.g., including terminal devices and network devices). This application does not limit the specific implementation method. For example, predefined can refer to what is defined in the protocol.

[0339] In the embodiments of this application, the term "protocol" may refer to standard protocols in the field of communications, such as LTE protocols, NR protocols, and related protocols applied in future communication systems. This application does not limit the scope of these protocols.

[0340] In the embodiments of this application, determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0341] In the embodiments of this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0342] In the embodiments of this application, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0343] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0344] The units described as separate components may or may not be physically separate. 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 the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0345] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0346] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for wireless communication, characterized in that, include: The terminal device receives first configuration information, which is used to indicate inference configuration; The terminal device performs measurement prediction and / or measurement event prediction based on the inference configuration; Based on the results of the measurement prediction and / or the measurement event prediction, the terminal device determines whether to send a predicted measurement report; The inference configuration includes a triggering method for the predicted event, which is determined based on the entry conditions and / or confidence threshold of the predicted event.

2. The method according to claim 1, characterized in that, The measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result, and the predicted measurement result and / or the measured measurement result are used to predict the measurement event.

3. The method according to claim 2, characterized in that, The method further includes: The terminal device performs a measurement prediction at the first moment to obtain the predicted measurement result at the second moment; The predicted measurement result at the second moment is used to evaluate the measurement event at the second moment, which is later than the first moment. The duration between the first moment and the second moment is the first duration, which is determined based on predefined information or network device configuration information.

4. The method according to claim 2, characterized in that, The method further includes: The terminal device performs measurement and prediction on multiple times within the first window at the first moment to obtain multiple predicted measurement results; The multiple predicted measurement results are used to determine one or more predicted times that trigger the target measurement event within the first window, or the distribution information of the target measurement event triggered within the first window.

5. The method according to claim 4, characterized in that, The length of the first window is the second duration, which is less than or equal to the first duration.

6. The method according to any one of claims 2-5, characterized in that, The measurement prediction is a Radio Resource Management (RRM) measurement prediction. The inference configuration of the RRM measurement prediction is carried in the RRM measurement configuration. The RRM measurement report indicated by the RRM measurement configuration includes the predicted measurement results and / or the measured measurement results.

7. The method according to any one of claims 2-6, characterized in that, The predicted measurement results and the measured measurement results are associated based on at least one of the following identifiers: measurement ID, measurement object ID, cell ID, and physical cell identifier (PCI).

8. The method according to any one of claims 1-7, characterized in that, The measurement event prediction is based on the triggering method of the prediction event to predict one or more prediction events, which are determined according to RRM measurement events.

9. The method according to any one of claims 1-8, characterized in that, The first configuration information includes one or more of the following: The type information of the predicted event or the measurement event corresponding to the predicted event; Information regarding the entry conditions; The parameter information of the measurement prediction and / or the measurement event prediction; Information regarding the confidence threshold; The flag information indicating the triggering method of the predicted event.

10. The method according to claim 9, characterized in that, The triggering method of the predicted event is also determined based on the prediction time, and the first configuration information also includes the timer information of the predicted event.

11. The method according to any one of claims 1-10, characterized in that, The predictive measurement report is submitted when at least one of the following conditions is met: The change in the new trigger time of the predicted event relative to the most recently reported trigger time is greater than the first threshold. The target cell or the type of predicted event has changed; The prediction result of the predicted event within the prediction window changes from triggered to not triggered, or from not triggered to triggered; The confidence level of the predicted measurement result does not fall within the pre-configured confidence threshold interval.

12. The method according to any one of claims 1-11, characterized in that, The measurement event prediction is controlled based on the transitions between multiple states, which include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state.

13. The method according to claim 12, characterized in that, The plurality of states correspond to a state machine, which controls the transitions between the plurality of states based on transition conditions and / or indications from network devices, and the state machine is used to maintain at least one target measurement event.

14. The method according to any one of claims 1-13, characterized in that, The method further includes: The terminal device sends first capability information; Wherein, the first capability information is used to determine the first configuration information, and the first capability information includes at least one of the following: Capability information related to time-domain prediction; Capability information related to frequency domain prediction; Capability information related to the prediction of measurement events.

15. The method according to claim 14, characterized in that, The time-domain prediction includes multiple use case types, and the capability information related to the time-domain prediction includes multiple sub-information corresponding to the multiple use case types.

16. The method according to claim 14 or 15, characterized in that, The capability information related to the prediction of measurement events includes multiple sub-information; the multiple sub-information corresponds one-to-one with multiple predicted events, or the multiple sub-information corresponds one-to-one with a combination of multiple predicted events.

17. The method according to any one of claims 1-16, characterized in that, The measurement prediction and the measurement event prediction are achieved through the same or different artificial intelligence / machine learning models.

18. A method for wireless communication, characterized in that, include: The network device sends the first configuration information to the terminal device; Based on the predictive measurement report sent by the terminal device, the network device makes mobility decisions; Wherein, the first configuration information is used to indicate the inference configuration, the inference configuration is used by the terminal device to perform measurement prediction and / or measurement event prediction, the result of the measurement prediction and / or the measurement event prediction is used by the terminal device to determine whether to send a predicted measurement report, the inference configuration includes the triggering method of the predicted event, the triggering method of the predicted event is determined according to the entry condition and / or confidence threshold of the predicted event.

19. The method according to claim 18, characterized in that, The measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result, and the predicted measurement result and / or the measured measurement result are used to predict the measurement event.

20. The method according to claim 19, characterized in that, The predicted measurement results include the predicted measurement results at a second time point. The predicted measurement results at the second time point are determined based on the measurement prediction made by the terminal device at a first time point. The predicted measurement results at the second time point are used to evaluate the measurement event at the second time point. The second time point is later than the first time point. The duration between the first time point and the second time point is a first duration. The first duration is determined based on predefined information or the configuration information of the network device.

21. The method according to claim 19, characterized in that, The predicted measurement results include multiple predicted measurement results at multiple times within the first window, and the multiple predicted measurement results are determined based on the measurement prediction made by the terminal device at the first time. The multiple predicted measurement results are used to determine one or more predicted times that trigger the target measurement event within the first window, or the distribution information of the target measurement event triggered within the first window.

22. The method according to claim 21, characterized in that, The length of the first window is the second duration, which is less than or equal to the first duration.

23. The method according to any one of claims 19-22, characterized in that, The measurement prediction is a Radio Resource Management (RRM) measurement prediction. The inference configuration of the RRM measurement prediction is carried in the RRM measurement configuration. The RRM measurement report indicated by the RRM measurement configuration includes the predicted measurement results and / or the measured measurement results.

24. The method according to any one of claims 19-23, characterized in that, The predicted measurement results and the measured measurement results are associated based on at least one of the following identifiers: measurement ID, measurement object ID, cell ID, and physical cell identifier (PCI).

25. The method according to any one of claims 18-24, characterized in that, The measurement event prediction is based on the triggering method of the prediction event to predict one or more prediction events, which are determined according to RRM measurement events.

26. The method according to any one of claims 18-25, characterized in that, The first configuration information includes one or more of the following: The type information of the predicted event or the measurement event corresponding to the predicted event; Information regarding the entry conditions; The parameter information of the measurement prediction and / or the measurement event prediction; Information regarding the confidence threshold; The flag information indicating the triggering method of the predicted event.

27. The method according to claim 26, characterized in that, The triggering method of the predicted event is also determined based on the prediction time, and the first configuration information also includes the timer information of the predicted event.

28. The method according to any one of claims 18-27, characterized in that, The predictive measurement report is submitted when at least one of the following conditions is met: The change in the new trigger time of the predicted event relative to the most recently reported trigger time is greater than the first threshold. The target cell or the type of predicted event has changed; The prediction result of the predicted event within the prediction window changes from triggered to not triggered, or from not triggered to triggered; The confidence level of the predicted measurement result does not fall within the pre-configured confidence threshold interval.

29. The method according to any one of claims 18-28, characterized in that, The measurement event prediction is controlled based on the transitions between multiple states, which include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state.

30. The method according to claim 29, characterized in that, The plurality of states correspond to a state machine, which controls the transitions between the plurality of states based on transition conditions and / or indications from network devices, and the state machine is used to maintain at least one target measurement event.

31. The method according to any one of claims 18-30, characterized in that, The method further includes: The network device receives the first capability information sent by the terminal device; Wherein, the first capability information is used to determine the first configuration information, and the first capability information includes at least one of the following: Capability information related to time-domain prediction; Capability information related to frequency domain prediction; Capability information related to the prediction of measurement events.

32. The method according to claim 31, characterized in that, The time-domain prediction includes multiple use case types, and the capability information related to the time-domain prediction includes multiple sub-information corresponding to the multiple use case types.

33. The method according to claim 31 or 32, characterized in that, The capability information related to the prediction of measurement events includes multiple sub-information; the multiple sub-information corresponds one-to-one with multiple predicted events, or the multiple sub-information corresponds one-to-one with a combination of multiple predicted events.

34. The method according to any one of claims 18-33, characterized in that, The measurement prediction and the measurement event prediction are achieved through the same or different artificial intelligence / machine learning models.

35. A device for wireless communication, characterized in that, The device is a terminal device, and the device includes: A receiving unit is configured to receive first configuration information, wherein the first configuration information is used to indicate inference configuration; The first processing unit is configured to perform measurement prediction and / or measurement event prediction based on the inference configuration. The second processing unit is used to determine whether to send a predicted measurement report based on the results of the measurement prediction and / or the measurement event prediction. The inference configuration includes a triggering method for the predicted event, which is determined based on the entry conditions and / or confidence threshold of the predicted event.

36. The apparatus according to claim 35, characterized in that, The measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result, and the predicted measurement result and / or the measured measurement result are used to predict the measurement event.

37. The apparatus according to claim 36, characterized in that, The first processing unit is further configured to perform measurement prediction on the second time at the first time to obtain the predicted measurement result at the second time; the predicted measurement result at the second time is used to evaluate the measurement event at the second time, the second time being later than the first time, and the duration between the first time and the second time being a first duration, the first duration being determined according to predefined information or network device configuration information.

38. The apparatus according to claim 36, characterized in that, The first processing unit is further configured to perform measurement prediction on multiple times within the first window at a first time to obtain multiple predicted measurement results; the multiple predicted measurement results are used to determine one or more predicted times that trigger the target measurement event within the first window, or the distribution information of the target measurement event triggered within the first window.

39. The apparatus according to claim 38, characterized in that, The length of the first window is the second duration, which is less than or equal to the first duration.

40. The apparatus according to any one of claims 36-39, characterized in that, The measurement prediction is a Radio Resource Management (RRM) measurement prediction. The inference configuration of the RRM measurement prediction is carried in the RRM measurement configuration. The RRM measurement report indicated by the RRM measurement configuration includes the predicted measurement results and / or the measured measurement results.

41. The apparatus according to any one of claims 36-40, characterized in that, The predicted measurement results and the measured measurement results are associated based on at least one of the following identifiers: measurement ID, measurement object ID, cell ID, and physical cell identifier (PCI).

42. The apparatus according to any one of claims 35-41, characterized in that, The measurement event prediction is based on the triggering method of the prediction event to predict one or more prediction events, which are determined according to RRM measurement events.

43. The apparatus according to any one of claims 35-42, characterized in that, The first configuration information includes one or more of the following: The type information of the predicted event or the measurement event corresponding to the predicted event; Information regarding the entry conditions; The parameter information of the measurement prediction and / or the measurement event prediction; Information regarding the confidence threshold; The flag information indicating the triggering method of the predicted event.

44. The apparatus according to claim 43, characterized in that, The triggering method of the predicted event is also determined based on the prediction time, and the first configuration information also includes the timer information of the predicted event.

45. The apparatus according to any one of claims 35-44, characterized in that, The predictive measurement report is submitted when at least one of the following conditions is met: The change in the new trigger time of the predicted event relative to the most recently reported trigger time is greater than the first threshold. The target cell or the type of predicted event has changed; The prediction result of the predicted event within the prediction window changes from triggered to not triggered, or from not triggered to triggered; The confidence level of the predicted measurement result does not fall within the pre-configured confidence threshold interval.

46. ​​The apparatus according to any one of claims 35-45, characterized in that, The measurement event prediction is controlled based on the transitions between multiple states, which include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state.

47. The apparatus according to claim 46, characterized in that, The plurality of states correspond to a state machine, which controls the transitions between the plurality of states based on transition conditions and / or indications from network devices, and the state machine is used to maintain at least one target measurement event.

48. The apparatus according to any one of claims 35-47, characterized in that, The device further includes: A transmitting unit, used to transmit first capability information; Wherein, the first capability information is used to determine the first configuration information, and the first capability information includes at least one of the following: Capability information related to time-domain prediction; Capability information related to frequency domain prediction; Capability information related to the prediction of measurement events.

49. The apparatus according to claim 48, characterized in that, The time-domain prediction includes multiple use case types, and the capability information related to the time-domain prediction includes multiple sub-information corresponding to the multiple use case types.

50. The apparatus according to claim 48 or 49, characterized in that, The capability information related to the prediction of measurement events includes multiple sub-information; the multiple sub-information corresponds one-to-one with multiple predicted events, or the multiple sub-information corresponds one-to-one with a combination of multiple predicted events.

51. The apparatus according to any one of claims 35-50, characterized in that, The measurement prediction and the measurement event prediction are achieved through the same or different artificial intelligence / machine learning models.

52. A device for wireless communication, characterized in that, The device is a network device, and the device includes: The sending unit is used to send first configuration information to the terminal device; The processing unit is used to make mobility decisions based on the predictive measurement report sent by the terminal device. Wherein, the first configuration information is used to indicate the inference configuration, the inference configuration is used by the terminal device to perform measurement prediction and / or measurement event prediction, the result of the measurement prediction and / or the measurement event prediction is used by the terminal device to determine whether to send a predicted measurement report, the inference configuration includes the triggering method of the predicted event, the triggering method of the predicted event is determined according to the entry condition and / or confidence threshold of the predicted event.

53. The apparatus according to claim 52, characterized in that, The measurement prediction is used to determine the predicted measurement result and / or the confidence level corresponding to the predicted measurement result, and the predicted measurement result and / or the measured measurement result are used to predict the measurement event.

54. The apparatus according to claim 53, characterized in that, The predicted measurement results include the predicted measurement results at a second time point. The predicted measurement results at the second time point are determined based on the measurement prediction made by the terminal device at a first time point. The predicted measurement results at the second time point are used to evaluate the measurement event at the second time point. The second time point is later than the first time point. The duration between the first time point and the second time point is a first duration. The first duration is determined based on predefined information or the configuration information of the network device.

55. The apparatus according to claim 53, characterized in that, The predicted measurement results include multiple predicted measurement results at multiple times within the first window, and the multiple predicted measurement results are determined based on the measurement prediction made by the terminal device at the first time. The multiple predicted measurement results are used to determine one or more predicted times that trigger the target measurement event within the first window, or the distribution information of the target measurement event triggered within the first window.

56. The apparatus according to claim 55, characterized in that, The length of the first window is the second duration, which is less than or equal to the first duration.

57. The apparatus according to any one of claims 53-56, characterized in that, The measurement prediction is a Radio Resource Management (RRM) measurement prediction. The inference configuration of the RRM measurement prediction is carried in the RRM measurement configuration. The RRM measurement report indicated by the RRM measurement configuration includes the predicted measurement results and / or the measured measurement results.

58. The apparatus according to any one of claims 53-57, characterized in that, The predicted measurement results and the measured measurement results are associated based on at least one of the following identifiers: measurement ID, measurement object ID, cell ID, and physical cell identifier (PCI).

59. The apparatus according to any one of claims 52-58, characterized in that, The measurement event prediction is based on the triggering method of the prediction event to predict one or more prediction events, which are determined according to RRM measurement events.

60. The apparatus according to any one of claims 52-59, characterized in that, The first configuration information includes one or more of the following: The type information of the predicted event or the measurement event corresponding to the predicted event; Information regarding the entry conditions; The parameter information of the measurement prediction and / or the measurement event prediction; Information regarding the confidence threshold; The flag information indicating the triggering method of the predicted event.

61. The apparatus according to claim 60, characterized in that, The triggering method of the predicted event is also determined based on the prediction time, and the first configuration information also includes the timer information of the predicted event.

62. The apparatus according to any one of claims 52-61, characterized in that, The predictive measurement report is submitted when at least one of the following conditions is met: The change in the new trigger time of the predicted event relative to the most recently reported trigger time is greater than the first threshold. The target cell or the type of predicted event has changed; The prediction result of the predicted event within the prediction window changes from triggered to not triggered, or from not triggered to triggered; The confidence level of the predicted measurement result does not fall within the pre-configured confidence threshold interval.

63. The apparatus according to any one of claims 52-62, characterized in that, The measurement event prediction is controlled based on the transitions between multiple states, which include at least one of the following: idle state, candidate state, prediction active state, reported state, and cleared state.

64. The apparatus according to claim 63, characterized in that, The plurality of states correspond to a state machine, which controls the transitions between the plurality of states based on transition conditions and / or indications from network devices, and the state machine is used to maintain at least one target measurement event.

65. The apparatus according to any one of claims 52-64, characterized in that, The device further includes: The receiving unit is configured to receive the first capability information sent by the terminal device; Wherein, the first capability information is used to determine the first configuration information, and the first capability information includes at least one of the following: Capability information related to time-domain prediction; Capability information related to frequency domain prediction; Capability information related to the prediction of measurement events.

66. The apparatus according to claim 65, characterized in that, The time-domain prediction includes multiple use case types, and the capability information related to the time-domain prediction includes multiple sub-information corresponding to the multiple use case types.

67. The apparatus according to claim 65 or 66, characterized in that, The capability information related to the prediction of measurement events includes multiple sub-information; the multiple sub-information corresponds one-to-one with multiple predicted events, or the multiple sub-information corresponds one-to-one with a combination of multiple predicted events.

68. The apparatus according to any one of claims 52-67, characterized in that, The measurement prediction and the measurement event prediction are achieved through the same or different artificial intelligence / machine learning models.

69. A communication device, characterized in that, It includes a memory and a processor, the memory being used to store a program, and the processor being used to invoke the program in the memory to perform the method as described in any one of claims 1-17 or 18-34.

70. An apparatus, characterized in that, Includes a processor for calling a program from memory to perform the method as described in any one of claims 1-17 or 18-34.

71. A chip, characterized in that, Includes a processor for calling a program from memory, causing a device on which the chip is mounted to perform the method as described in any one of claims 1-17 or 18-34.

72. A computer-readable storage medium, characterized in that, It contains a program that causes a computer to perform the method as described in any one of claims 1-17 or 18-34.

73. A computer program product, characterized in that, Includes a program that causes a computer to perform the method as described in any one of claims 1-17 or 18-34.

74. A computer program, characterized in that, The computer program causes the computer to perform the method as described in any one of claims 1-17 or 18-34.

Images