Model monitoring method, device and equipment and readable storage medium

Abstract

The invention discloses a model monitoring method, device and equipment and a readable storage medium, and belongs to the field of communication, and the method comprises the steps that a terminal obtains first information; performing model monitoring on a first artificial intelligence (AI) model according to the first information, the first AI model being used for executing a first prediction function; wherein the first information comprises at least one of the following items: one or more prediction precision requirements; configuration information corresponding to the first prediction function; the first parameter is used for determining a prediction precision requirement associated with the first prediction function; wherein the one or more prediction precision requirements are associated with at least one of the following items: a prediction window, a measurement reduction ratio (MRR), a frequency point, a cell, a beam, an AI function and an AI model.

Description

Technical Field

[0001] This application belongs to the field of communication technology, specifically relating to a model monitoring method, apparatus, device, and readable storage medium. Background Technology

[0002] In related technologies, terminals can manage models through model monitoring results. For example, when model performance fails to meet accuracy requirements, the model can be deactivated, switched, or rolled back. However, excessively high accuracy requirements may cause the model to malfunction, while excessively low accuracy requirements may affect the model's predictive performance. Therefore, how to select appropriate accuracy requirements for model monitoring to balance ensuring normal model operation and predictive performance is a problem that urgently needs to be solved. Summary of the Invention

[0003] This application provides a model monitoring method, apparatus, device, and readable storage medium that can ensure both the normal operation of the model and its predictive performance.

[0004] Firstly, a model monitoring method is provided, which includes:

[0005] The terminal obtains the first information;

[0006] Based on the first information, the first artificial intelligence (AI) model is monitored, wherein the first AI model is used to perform a first prediction function;

[0007] The first information includes at least one of the following:

[0008] One or more prediction accuracy requirements;

[0009] Configuration information corresponding to the first prediction function;

[0010] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0011] The one or more prediction accuracy requirements are associated with at least one of the following:

[0012] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0013] Secondly, a model monitoring method is provided, which includes:

[0014] The network-side device sends first information to the terminal, the first information being used by the terminal to monitor the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function;

[0015] The first information includes at least one of the following:

[0016] One or more prediction accuracy requirements;

[0017] Configuration information corresponding to the first prediction function;

[0018] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0019] The one or more prediction accuracy requirements are associated with at least one of the following:

[0020] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0021] Thirdly, a wireless communication device is provided, comprising:

[0022] The processing module is used to obtain the first information;

[0023] Based on the first information, the first artificial intelligence (AI) model is monitored, wherein the first AI model is used to perform a first prediction function;

[0024] The first information includes at least one of the following:

[0025] One or more prediction accuracy requirements;

[0026] Configuration information corresponding to the first prediction function;

[0027] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0028] The one or more prediction accuracy requirements are associated with at least one of the following:

[0029] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0030] Fourthly, a wireless communication device is provided, comprising:

[0031] The sending module is used to send first information to the terminal, the first information being used by the terminal to monitor the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function;

[0032] The first information includes at least one of the following:

[0033] One or more prediction accuracy requirements;

[0034] Configuration information corresponding to the first prediction function;

[0035] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0036] The one or more prediction accuracy requirements are associated with at least one of the following:

[0037] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0038] Fifthly, a wireless communication device is provided, the device being configured to perform the steps of the method described in the first aspect, or to implement the steps of the method described in the second aspect.

[0039] In a sixth aspect, a terminal is provided, the terminal including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the first aspect.

[0040] In a seventh aspect, a terminal is provided, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the method as described in the first aspect.

[0041] Eighthly, a network-side device is provided, the network-side device including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the second aspect.

[0042] In a ninth aspect, a network-side device is provided, including a processor and a communication interface, wherein the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the method as described in the second aspect.

[0043] In a tenth aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or implement the steps of the method described in the second aspect.

[0044] Eleventhly, a wireless communication system is provided, comprising: a terminal and a network-side device, wherein the terminal can be used to perform the steps of the method as described in the first aspect, and the network-side device can be used to perform the steps of the method as described in the second aspect.

[0045] In a twelfth aspect, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being configured to run programs or instructions to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0046] In a thirteenth aspect, a computer program / program product is provided, which is stored in a storage medium and is executed by at least one processor to implement the steps of the method as described in the first aspect, or to implement the steps of the method as described in the second aspect.

[0047] In this embodiment, the terminal can select at least one of the prediction accuracy requirements associated with the prediction window, MRR, frequency point, cell, AI model, and AI function to perform model monitoring on the first AI model based on the first information. Alternatively, the terminal can determine the target prediction accuracy requirement used to perform model monitoring on the first AI model based on the configuration information and / or first parameters corresponding to the first prediction function. This is beneficial for the terminal to select appropriate prediction accuracy requirements to perform model monitoring on the first AI model, ensuring the normal operation of the AI ​​model and taking into account the prediction performance of the AI ​​model. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of a communication system architecture provided in an embodiment of this application.

[0049] Figure 2 This is a schematic diagram of a neuron structure.

[0050] Figure 3 This is a schematic diagram of a typical neural network.

[0051] Figure 4 This is a functional framework diagram of AI / ML for air interface provided in an embodiment of this application.

[0052] Figure 5 This is a schematic diagram of a model monitoring method provided in an embodiment of this application.

[0053] Figure 6 These are schematic diagrams illustrating other model monitoring methods provided in the embodiments of this application.

[0054] Figure 7 This is a schematic diagram of another model monitoring method provided in the embodiments of this application.

[0055] Figure 8 This is a schematic diagram of another model monitoring method provided in the embodiments of this application.

[0056] Figure 9 This is a schematic diagram of another model monitoring method provided in the embodiments of this application.

[0057] Figure 10 This is a schematic block diagram of a wireless communication device provided according to an embodiment of this application.

[0058] Figure 11This is a schematic block diagram of a wireless communication device provided according to an embodiment of this application.

[0059] Figure 12 A schematic block diagram of a communication device provided according to an embodiment of this application.

[0060] Figure 13 This is a schematic diagram of the hardware structure of a terminal according to an embodiment of this application.

[0061] Figure 14 This is a schematic block diagram of a network-side device provided according to an embodiment of this application.

[0062] Figure 15 This is a schematic block diagram of another network-side device provided according to an embodiment of this application. Detailed Implementation

[0063] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0064] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, the scope of protection for "A or B" covers at least three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. In addition, the terms "A and / or B," "at least one of A and B," and "at least one of A or B" also cover at least the above three scenarios. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0065] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc., in the instruction sent. An indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0066] It is worth noting that the technologies described in this application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA), or other systems. The terms "system" and "network" in this application are often used interchangeably, and the described technologies can be used with the systems and radio technologies mentioned above, as well as with other systems and radio technologies. The following description describes a New Radio (NR) system for illustrative purposes, and the term NR is used in most of the following description; however, these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) radio systems. th Generation 6G communication system.

[0067] Figure 1This diagram illustrates a block diagram of a wireless communication system applicable to embodiments of this application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can be a mobile phone, tablet computer, laptop computer, notebook computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR), virtual reality (VR) device, robot, wearable device, flight vehicle, vehicle user equipment (VUE), shipboard equipment, pedestrian user equipment (PUE), smart home devices (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game consoles, personal computers (PCs), ATMs, or self-service machines, etc. Wearable devices include: smartwatches, smart bracelets, smart earphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among these, in-vehicle devices can also be referred to as in-vehicle terminals, in-vehicle controllers, in-vehicle modules, in-vehicle components, in-vehicle chips, or in-vehicle units, etc. It should be noted that the specific type of terminal 11 is not limited in the embodiments of this application.

[0068] In the embodiments of this application, the terminal may also be referred to as user equipment (UE), terminal equipment, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication equipment, user agent, or user device, etc.

[0069] Network-side equipment 12 may include access network equipment or core network equipment. Access network equipment may also be referred to as Radio Access Network (RAN) equipment, radio access network function, or radio access network unit. Access network equipment may include base stations, wireless local area network (WLAN) access points (APs), or wireless Fidelity (WiFi) nodes, etc. The term "base station" can be referred to as Node B (NB), Evolved Node B (eNB), Next Generation Node B (gNB), New Radio Node B (NR Node B), Access Point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), Radio Base Station, Radio Transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B (HNB), Home Evolved Node B, Transmit / Receive Point (TRP), or any other suitable term in the relevant field, as long as the same technical effect is achieved. The term "base station" is not limited to any specific technical terminology. It should be noted that this application embodiment only uses a base station in an NR system as an example for description and does not limit the specific type of base station.

[0070] Core network equipment, also known as core network nodes, core network functions, or core network elements, includes, but is not limited to, at least one of the following: Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized Network Configuration (CNC), and Network Storage Function (Network). The core network functions include Repository Function (NRF), Network Exposure Function (NEF), Local NEF (or L-NEF), Binding Support Function (BSF), Application Function (AF), Location Management Function (LMF), Gateway Mobile Location Centre (GMLC), and Network Data Analytics Function (NWDAF). It should be noted that this application only uses core network equipment in the NR system as an example and does not limit the specific type of core network equipment. If the name of the core network equipment mentioned in this application changes in subsequent protocol versions (e.g., 6G), it will still be within the scope of protection of this application.

[0071] Optionally, the core network equipment can be implemented by one or more functional modules in a single device, or by multiple devices working together; this application does not specifically limit this. It is understood that the aforementioned functional modules can be network elements in hardware devices, software functional modules running on dedicated hardware, or virtualized functional modules instantiated on a platform (e.g., a cloud platform).

[0072] To facilitate understanding of the embodiments of this application, the artificial intelligence (AI) technology related to this application will be described.

[0073] AI technology has been widely applied in various fields. Integrating AI technology into wireless communication networks to improve technical indicators such as throughput, latency, and user capacity is an important task for future wireless communication networks. AI models can be implemented in various ways, such as neural networks, decision trees, support vector machines, and Bayesian classifiers. The following explanation uses neural networks as an example, but does not limit the specific type of AI module.

[0074] A neural network is a computational model consisting of multiple interconnected neurons. The connections between nodes represent weights, which are weights used to calculate the weighted sum of different input signals. Each node performs a weighted summation (SUM) on different input signals and outputs the sum through a specific activation function (f). Figure 2 This is a schematic diagram of a neuron structure, where a1, a2, ..., an represent the inputs, w1, w2, ..., wn represent the weights (multiplicative coefficients), b represents the bias (additive coefficient), σ represents the activation function, and y represents the output, where z = a1*w1 + a2*w2 + ... + an*wn + b. Common activation functions include Sigmoid, tanh, ReLU (Rectified Linear Unit), etc.

[0075] A typical neural network is as follows Figure 3 As shown, this neural network comprises an input layer, hidden layers, and an output layer. Through different connection methods, weights, and activation functions of multiple neurons, it can produce different outputs, thereby fitting the mapping relationship from input to output. Each node in the higher-level hierarchy is connected to all its lower-level nodes. This neural network is a fully connected neural network, also known as a deep neural network (DNN).

[0076] Deep learning employs deep neural networks with multiple hidden layers, greatly enhancing the network's ability to learn features and fit complex nonlinear mappings from input to output. Therefore, it has found widespread application in speech and image processing. Besides deep neural networks, deep learning also includes commonly used basic structures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for different tasks.

[0077] The parameters of a neural network are optimized using gradient optimization algorithms. Gradient optimization algorithms are a class of algorithms that minimize or maximize an objective function (or loss function), which is often a mathematical combination of model parameters and data. For example, given data X and its corresponding label Y, a neural network model f(.) can be constructed. With the model, the output f(x) can be predicted based on the input x, and the difference between the predicted value and the true value (f(x) - Y) can be calculated; this is the loss function. Our goal is to find suitable W and b that minimize the value of the above loss function. The smaller the loss value, the closer our model is to the reality.

[0078] Common optimization algorithms include those based on error back propagation (BP). The basic idea of ​​BP is that the learning process consists of two parts: forward propagation of the signal and backward propagation of the error. During forward propagation, input samples are introduced from the input layer, processed layer by layer by the hidden layers, and then transmitted to the output layer. If the actual output of the output layer does not match the expected output, the process transitions to the error back propagation stage. Error back propagation involves propagating the output error back to the input layer layer by layer through the hidden layers, distributing the error to all units in each layer, thus obtaining the error signal for each unit. This error signal can be used as the basis for adjusting the weights of each unit. This process of adjusting the weights of each layer through forward and backward propagation is repeated continuously. This continuous adjustment of weights is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until a predetermined number of learning iterations are completed.

[0079] Common optimization algorithms may include, but are not limited to: Gradient Descent, Stochastic Gradient Descent (SGD), mini-batch gradient descent, Momentum, Nesterov, Adagrad (Adaptive Gradient Descent), Adadelta, RMSprop (root mean square prop), and Adam (Adaptive Moment Estimation).

[0080] During error backpropagation, these optimization algorithms calculate the gradient by taking the derivative / partial derivative of the error / loss obtained from the loss function with respect to the current neuron, adding the influence of the learning rate, previous gradients / derivatives / partial derivatives, etc., and then pass the gradient to the previous layer.

[0081] To facilitate understanding of the embodiments of this application, the radio resource management (RRM) related to this application will be described.

[0082] In related technologies, measurement configuration mainly consists of Measurement Object (MO), Report Configuration, and Measurement Identifier (measId).

[0083] The measurement object can refer to the frequency point to be measured. Reporting configurations can include: reporting accuracy (e.g., periodic or event-triggered); reference signal type, such as Synchronization Signal Block (SSB) or Channel State Information Reference Signal (CSI-RS); whether to report beam measurement results; and the maximum number of beams that can be reported. A measurement identifier is associated with a measurement object and a reporting configuration; a measurement object can be associated with multiple reporting configurations, and a reporting configuration can be associated with multiple measurement objects.

[0084] In a communication system, the measurement object, measurement identifier, and device configuration can be associated together in the following way:

[0085]

[0086] Among them, measObjectId represents the measurement object identifier, and reportConfigId represents the reporting configuration identifier.

[0087] In related technologies, the reporting configuration can be configured to trigger reporting based on events. For example, events defined in a communication system can include the events listed in Table 1 below.

[0088] Table 1

[0089]

[0090]

[0091] For event A3, the meanings of the parameters involved in the entry and exit conditions are as follows:

[0092] Mn: Neighboring cell measurement results, without considering any offset;

[0093] Ofn: Specific offset of the neighboring cell measurement object;

[0094] Ocn: Neighboring cell-level specific offset;

[0095] Mp: Measurement results for a special cell (SpCell), without considering any offset;

[0096] Ofp: SpCell measures a specific offset of an object;

[0097] Ocp: SpCell cell-level specific offset;

[0098] Hys: The hysteresis parameter of the event;

[0099] Off: The offset parameter for the event.

[0100] If the reporting type is event-triggered, in order to avoid frequent reporting or ping-pong handover, the network-side device configures a trigger time (timeToTrigger) parameter for each event. If the Layer 3 (L3) filtered signal quality of one or more candidate cells meets the event entry conditions within the timeToTrigger time, then the measurement reporting is triggered.

[0101] In some scenarios, AI / ML functional frameworks for air interfaces have been introduced, such as... Figure 4 As shown, it can include the following modules:

[0102] Data Collection module: Used to provide input data to the Model Training module, Management module, and Interference module.

[0103] The Model Training module is responsible for training, validating, and testing AI / ML models. It can also handle data preparation, such as data preprocessing, including converting data to a specific format.

[0104] The Management module is responsible for model selection, activation, deactivation, switching, or rollback (such as rolling back to a non-AI-based method).

[0105] The Inference module assists in providing the output after applying AI / ML models or AI / ML functions.

[0106] Model Storage module: helps save trained or updated models.

[0107] Model Transfer / Delivery module: Assists in delivering AI / ML models to inference function nodes.

[0108] In some scenarios, consider introducing AI-based mobility enhancements, such as the following use cases:

[0109] 1. RRM measurement and prediction.

[0110] For example, future RRM measurement predictions can be made in the time domain.

[0111] For example, in the frequency domain, it is possible to predict the current or future RRM measurement results at other frequency points;

[0112] For example, in the spatial domain, the RRM measurement prediction results of other beams in this cell can be predicted.

[0113] 2. Measurement event prediction: Predicting whether a measurement event will be satisfied in the future. This can include the following two types of methods:

[0114] Method 1: Directly measure event prediction, such as the output of an AI model being a flag or probability indicating whether an event meets certain criteria;

[0115] Method 2: Indirect measurement event prediction: For example, the output of the AI ​​model is the RRM prediction result, and the measurement event will be determined based on the RRM prediction result.

[0116] 3. Radio Link Failure (RLF) prediction: Predicting whether an RLF will occur in the future. This can include methods such as the following two categories:

[0117] Method 1: Direct RLF prediction, such as the output of the AI ​​model being a flag or probability of whether an RLF will occur;

[0118] Method 2: Indirect RLF prediction: If the output of the AI ​​model is the signal-to-interference-plus-noise ratio (SINR), determine whether RLF will occur based on the SINR.

[0119] In related technologies, network-side devices or terminals can manage models based on model monitoring results. For example, when model performance fails to meet accuracy requirements, the model can be deactivated, switched, or rolled back. However, excessively high accuracy requirements may cause the model to malfunction, while low accuracy requirements may affect its predictive performance. Therefore, determining the appropriate accuracy requirements for model monitoring to balance ensuring normal model operation and predictive performance is a pressing issue that needs to be addressed.

[0120] The wireless communication method provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0121] Figure 5 This is a schematic diagram of a wireless communication method 200 provided in an embodiment of this application. Figure 5 As shown, the method 200 includes at least the following:

[0122] S210, the terminal obtains the first information;

[0123] S220, based on the first information, perform model monitoring on the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function.

[0124] It should be noted that in the embodiments of this application, AI model may also be referred to as AI unit, machine learning (ML) model, ML unit, AI structure, AI function, AI characteristic, machine learning model, neural network, neural network function, neural network function, etc. Alternatively, AI model may refer to a processing unit that can implement specific algorithms, formulas, processing flows, capabilities, etc. related to AI. Alternatively, AI model may be a processing method, algorithm, function, module, or unit for a specific dataset. Alternatively, AI model may be a processing method, algorithm, function, module, or unit running on AI / ML related hardware such as Graphics Processing Unit (GPU), Neural Processing Unit (NPU), Tensor Processing Unit (TPU), and Application Specific Integrated Circuit (ASIC). This application does not make specific limitations in this regard.

[0125] It should be noted that, in the embodiments of this application, the identification information of the AI ​​model may be, for example, an AI model ID, an AI structure ID, an AI algorithm ID, or an ID of a specific dataset associated with the AI ​​model, or an ID of a specific scenario, environment, channel characteristics, or device related to AI / ML, or an ID of a function, feature, capability, or module related to AI / ML. This application does not make any specific limitations on this.

[0126] In this embodiment of the application, a first AI model is deployed on the terminal, the terminal can perform a first prediction function based on the first AI model, and the terminal can monitor the first AI model based on the method 200.

[0127] It should be understood that in the embodiments of this application, a second AI model can be deployed on the network-side device, and the network-side device can perform a second prediction function based on the second AI model. The network-side device can also perform model monitoring on the second AI model in a manner similar to method 200. This application does not limit this. The following uses the terminal performing model monitoring on the first AI model as an example to illustrate the model monitoring method provided in the embodiments of this application.

[0128] In some embodiments of this application, the first prediction function includes, but is not limited to, at least one of the following:

[0129] RRM measurement prediction;

[0130] Measurement event prediction;

[0131] RLF forecast;

[0132] HOF (Handover Failure) prediction;

[0133] Target cell prediction.

[0134] It should be understood that the prediction functions supported by the first AI model described above are merely examples, and the first AI model may also support other prediction functions, which are not limited in this application.

[0135] In some embodiments, RRM measurement prediction may include, but is not limited to, at least one of the following:

[0136] In the time domain, predict future RRM measurement results;

[0137] In the frequency domain, predict current or future RRM measurement results for other frequency points / cells;

[0138] In the airspace, predict the RRM measurement results of other beams within the serving cell or neighboring cells.

[0139] Optionally, the RRM measurement prediction result in the embodiments of this application can be the predicted signal quality, such as Reference Signal Receiving Power (RSRP), Reference Signal Receiving Quality (RSRQ), Signal to Interference plus Noise Ratio (SINR), etc.

[0140] In some embodiments, the input information of the first AI model may be the RRM measurement results or RRM measurement prediction results of cells within the second frequency point. The other frequency points may be frequency points other than the second frequency point, such as the first frequency point. The RRM measurement results of cells within the second frequency point may be obtained by measurement, and the RRM measurement prediction results of cells within the second frequency point may be obtained by prediction.

[0141] In some embodiments, the input information of the first AI model may be the RRM measurement result or RRM measurement prediction result of the second cell, and the other cell may be a cell other than the second cell, such as the first cell, wherein the RRM measurement result of the second cell may be obtained by measurement, and the RRM measurement prediction result of the second cell may be obtained by prediction.

[0142] In some embodiments, the input information of the first AI model may be the RRM measurement result or RRM measurement prediction result of the second beam, and the other beam may be a beam other than the second beam in the serving cell or neighboring cells, such as the first beam, wherein the RRM measurement result of the second beam may be obtained by measurement, and the RRM measurement prediction result of the second beam may be obtained by prediction.

[0143] In some embodiments, measurement event prediction may include at least one of the following:

[0144] Direct measurement event prediction, for example, the output of the first AI model can be a flag indicating whether the measurement event is met, or the probability of the measurement event being met;

[0145] Indirect measurement event prediction, for example, the output of an AI model is an RRM measurement prediction result, which is used to determine whether a measurement event is met.

[0146] In some embodiments, RLF prediction may include at least one of the following:

[0147] Direct RLF prediction, for example, the output of the first AI model is a flag indicating whether an RLF will occur, or the probability of an RLF occurring;

[0148] Indirect RLF prediction, for example, can be achieved by the output of the first AI model, which can be SINR, used to determine whether an RLF will occur.

[0149] In some embodiments, HOF prediction may refer to predicting whether an HOF will occur.

[0150] In some embodiments, target cell prediction may refer to predicting the target cell for the terminal to perform cell handover.

[0151] In some embodiments, a prediction window can be provided for RRM measurement prediction, measurement event prediction, and RLF prediction. This prediction window corresponds to a future moment or time period. For example, the prediction window can be 400ms in the future, or 1s or 2s in the future.

[0152] In some embodiments, for AI model-based predictions, a Measurement Reduction Rate (MRR) can be defined, representing the proportion of measurements reduced through prediction.

[0153] For example, for time-domain RRM measurements, the measurement times include times 1, 2, 3, 4, 5, and 6. Among them, times 1, 3, and 5 use actual measured values, while times 2, 4, and 6 use predicted values. In this case, the measurement reduction rate in the temporal domain (MRRT) is 1 / 2.

[0154] For example, for spatial RRM measurements: the set of all transmitted beams (Tx beams) is {Tx beam1, Tx beam2, Tx beam3, Tx beam4}, where Tx beam1 and Tx beam2 are the actual measurement beams, and Tx beam3 and Tx beam4 are the predicted beams. In this case, the measurement reduction rate in the spatial domain (MRRS) is 1 / 2.

[0155] In some embodiments of this application, the terminal acquires first information, including:

[0156] The terminal obtains the first information through predefined information; or

[0157] The terminal obtains the first information from the network-side device.

[0158] In other words, the first piece of information can be predefined or configured by the network-side devices.

[0159] For example, the network-side device can send first information to the terminal through at least one of the following signaling methods:

[0160] Radio Resource Control (RRC) signaling;

[0161] Media Access Control (MAC) CE;

[0162] Downlink Control Information (DCI).

[0163] In some embodiments of this application, the first information includes at least one of the following:

[0164] One or more prediction accuracy requirements;

[0165] Configuration information corresponding to the first prediction function;

[0166] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function.

[0167] In some embodiments, the one or more prediction accuracy requirements are associated with at least one of the following:

[0168] Prediction window, Measurement Reduction Ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0169] In this application embodiment, prediction accuracy requirements can be designed at the granularity of prediction window, MRR, frequency point, cell, beam, AI function, or AI model. Thus, the terminal can select appropriate prediction accuracy requirements to monitor the first AI model based on the above parameters associated with the first AI model, which helps to ensure the normal operation of the first AI model while taking into account its prediction performance.

[0170] In some embodiments, since the configuration information corresponding to the first prediction function can be used to determine whether an event is triggered or the probability of the event occurring, when determining the target prediction accuracy requirement used to perform model monitoring on the first AI model, considering the configuration information corresponding to the first prediction function performed by the first AI model is beneficial for the terminal to select appropriate prediction accuracy requirements to monitor the first AI model, ensuring the normal operation of the AI ​​model, while also taking into account the prediction performance of the first AI model.

[0171] In some embodiments, when the terminal determines the target prediction accuracy requirement based on the configuration information corresponding to the first prediction function, the target prediction accuracy requirement can be determined based on the first parameter and the configuration information corresponding to the first prediction function. Optionally, the first parameter can be a scaling factor of the configuration information, or an adjustment amount, etc. Some methods for determining the target prediction accuracy requirement will be further described in the following embodiments.

[0172] Therefore, in this embodiment, the terminal can select at least one of the prediction accuracy requirements associated with the prediction window, MRR, frequency point, cell, AI model, and AI function to perform model monitoring on the first AI model based on the first information. Alternatively, the terminal can determine the target prediction accuracy requirement used to perform model monitoring on the first AI model based on the configuration information corresponding to the first prediction function and the first parameter. This is beneficial for the terminal to select appropriate prediction accuracy requirements to perform model monitoring on the first AI model, ensuring the normal operation of the AI ​​model and taking into account the prediction performance of the AI ​​model.

[0173] In some embodiments, the one or more prediction accuracy requirements include, but are not limited to, at least one of the following:

[0174] Prediction accuracy requirements for RRM measurement forecasting;

[0175] Prediction accuracy requirements for measuring event prediction;

[0176] Prediction accuracy requirements for RLF forecasting;

[0177] Prediction accuracy requirements for HOF forecasting;

[0178] Prediction accuracy requirements for target cell prediction.

[0179] Therefore, in this embodiment, the terminal can obtain the prediction accuracy requirements corresponding to different prediction functions, so as to select the appropriate prediction accuracy requirements to perform the prediction according to the prediction function to be executed, which can ensure the normal operation of the AI ​​model, while also taking into account the prediction performance of the AI ​​model in performing the prediction function.

[0180] For example, measurement event prediction is used to predict whether a measurement event will be triggered in the short term, while HOF prediction is used to predict whether a switching ping-pong effect will occur in the future. The events predicted by the two are different, and correspondingly, the prediction accuracy requirements should also be designed differently to ensure the normal operation of the AI ​​model, while also taking into account the model's prediction performance.

[0181] In some embodiments, one or more prediction accuracy requirements may include at least one of the following:

[0182] One or more prediction accuracy requirements associated with a prediction window;

[0183] One or more prediction accuracy requirements associated with MRR;

[0184] One or more prediction accuracy requirements associated with frequency points;

[0185] One or more prediction accuracy requirements associated with a cell;

[0186] One or more beam-related prediction requirements;

[0187] One or more prediction requirements associated with AI functionality;

[0188] One or more prediction requirements associated with an AI model.

[0189] In some embodiments, the prediction accuracy requirement can be associated with a prediction window one by one, or a prediction accuracy requirement can be associated with multiple prediction windows.

[0190] In some embodiments, the prediction accuracy requirement can be associated one-to-one with the MRR, or one prediction accuracy requirement can be associated with multiple MRRs.

[0191] In some embodiments, the prediction accuracy requirement may be associated one-to-one with a frequency point, cell, or beam, or a prediction accuracy requirement may be associated with multiple frequency points, cells, or beams.

[0192] In some embodiments, the prediction accuracy requirement may be associated one-to-one with an AI function or AI model, or one prediction accuracy requirement may be associated with multiple AI functions or AI models.

[0193] In some embodiments, when the first information includes a prediction accuracy requirement associated with at least one of frequency point, cell, and beam, the prediction accuracy requirement associated with at least one of frequency point, cell, and beam is configured by the network-side device in the first configuration information, which is used to configure the measurement object.

[0194] The prediction accuracy requirements for frequency points, cells, or beam associations can be configured in the measurement object. In this way, network-side equipment does not need to configure the prediction accuracy requirements for frequency points, cells, or beam associations for the terminal through additional signaling, which helps to reduce signaling interaction overhead.

[0195] In some embodiments, the first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window and the second prediction accuracy requirement is associated with a second prediction window, wherein the first prediction accuracy requirement is lower than the second prediction accuracy requirement when the length of the first prediction window is greater than the length of the second prediction window, or when the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time (e.g., the first prediction window is farther away from the current time, and correspondingly, the prediction difficulty is greater).

[0196] It is understandable that different prediction windows or MRRs have different prediction difficulties. For example, the longer the prediction window, the larger the MRR, and the greater the prediction difficulty. Correspondingly, the prediction accuracy requirement should also be set lower to ensure the normal operation of the AI ​​model.

[0197] Furthermore, different prediction windows or MRRs may have different purposes. For example, a short prediction window can be used to determine whether a measurement event will be triggered in the short term, while a long prediction window can be used to determine whether a switching ping-pong effect will occur in the future. Since their purposes are different, their accuracy requirements should also be designed differently. This application provides configuration of corresponding prediction accuracy requirements for different prediction windows or MRRs, which helps ensure the normal operation of the AI ​​model while also taking into account the model's prediction performance.

[0198] In some embodiments of this application, the step of monitoring the first AI model based on the first information includes:

[0199] Based on the second information, determine the target prediction accuracy requirement from among the one or more prediction accuracy requirements;

[0200] The first AI model is monitored according to the target prediction accuracy requirement;

[0201] The second information includes at least one of the following:

[0202] The prediction window associated with the first AI model;

[0203] The MRR associated with the first AI model;

[0204] The frequency points associated with the first AI model;

[0205] The cell associated with the first AI model;

[0206] The beam associated with the first AI model;

[0207] The AI ​​functions supported by the first AI model;

[0208] The first AI model.

[0209] In some embodiments, the target prediction accuracy requirement is associated with at least one of the following:

[0210] The first prediction window is the prediction window associated with the first AI model;

[0211] The first MRR is the MRR associated with the first AI model;

[0212] The first frequency point is the frequency point associated with the first AI model;

[0213] The first cell, the first frequency point is the cell associated with the first AI model;

[0214] The first beam is the beam associated with the first AI model;

[0215] The first AI function is the AI ​​function supported by the first AI model.

[0216] The first AI model.

[0217] In some embodiments, associating the first AI model with the first prediction window can be understood as the output of the first AI model being the prediction result within the first prediction window. For example, the prediction result within the first prediction window could be an RRM measurement prediction result within the first prediction window, such as the predicted signal quality. As another example, the prediction result within the first prediction window could be a flag indicating whether a measurement event / RLF / HOF will occur within the first prediction window, or the probability of a measurement event / RLF / HOF occurring within the first prediction window.

[0218] In some embodiments, the association of the first AI model with the first MRR can be understood as the percentage reduction in measurement that can be achieved by performing the first prediction function through the first AI model.

[0219] In some embodiments, associating the first AI model with the first frequency point can be understood as the output of the first AI model being a prediction result of a cell within the first frequency point, such as a prediction result of RRM measurement of a cell within the first frequency point (e.g., the signal quality of a cell within the first frequency point). Alternatively, the first AI model can be used to predict the RRM measurement result of a cell within the first frequency point. Optionally, the first frequency point can be a frequency point to be measured or predicted by the terminal. By measuring or predicting the corresponding frequency point, the terminal can obtain the signal quality of the cell within the corresponding frequency point, which can be used for mobility management or radio link monitoring.

[0220] In some embodiments, associating the first AI model with the first cell can be understood as the output of the first AI model being a prediction result of the first cell, such as the RRM measurement prediction result of the first cell; or, the first AI model can be used to predict the RRM measurement prediction result of the first cell. Optionally, the first cell can be the cell to be measured or predicted by the terminal. By measuring or predicting the corresponding cell, the terminal can obtain the signal quality of the corresponding cell, which can be used for mobility management or radio link monitoring.

[0221] In some embodiments, associating the first AI model with the first beam can be understood as the output of the first AI model being a prediction result of the first beam, such as an RRM measurement prediction result of the first beam; or, the first AI model can be used to predict the RRM measurement prediction result of the first beam. Optionally, the first beam can be a beam to be measured or predicted by the terminal. By measuring or predicting the corresponding beam, the terminal can obtain the signal quality of the corresponding beam, which can be used for mobility management or wireless link monitoring.

[0222] In some embodiments of this application, the step of monitoring the first AI model based on the first information includes:

[0223] Based on the configuration information corresponding to the first prediction function and the first parameter, the target prediction accuracy requirement is determined;

[0224] Based on the target prediction accuracy requirements, the first AI model is monitored.

[0225] Optionally, in this case, the first prediction function can be an indirect event prediction, such as an indirect measurement event prediction, or an indirect RLF prediction, etc.

[0226] For indirect event prediction, the AI ​​model outputs an intermediate prediction result, which is used to determine the final prediction result. If the prediction accuracy error of the intermediate prediction result is higher than the threshold corresponding to the event configuration, the accuracy of the event prediction cannot be guaranteed. For example, for A3 event prediction, if the event is configured to meet the event entry condition when the neighboring cell is 3dB higher than the serving cell, then if the accuracy error of the RRM measurement prediction result output by the AI ​​model is higher than 3dB, the accuracy of the measurement event prediction is likely to be unreliable.

[0227] Therefore, in this embodiment of the application, for indirect event prediction, the terminal can determine the target prediction accuracy requirement based on the configuration information corresponding to the event and the first parameter, which helps to ensure that the AI ​​model used for event prediction can run normally and also helps to ensure the accuracy of event prediction.

[0228] Optionally, the configuration information corresponding to the first prediction function may include the configuration information corresponding to the prediction of measurement events.

[0229] Optionally, the measurement event can be an A3 event, and the configuration information corresponding to the A3 event may include, but is not limited to, at least one of the following:

[0230] Mn: Neighbor cell measurement results, without considering any offset;

[0231] Ofn: Specific offset of the neighboring cell measurement object;

[0232] Ocn: Neighboring cell-level specific offset;

[0233] Mp: SpCell measurement result, without considering any offset;

[0234] Ofp: SpCell measures a specific offset of an object;

[0235] Ocp: SpCell cell-level specific offset;

[0236] Hys: The hysteresis parameter of the event;

[0237] Off: The offset parameter for the event.

[0238] In some embodiments, determining the target prediction accuracy requirement based on the configuration information corresponding to the first prediction function and the first parameter includes:

[0239] The product of the second parameter and the first parameter is determined as the target prediction accuracy requirement, wherein the second parameter is one or a combination of one or more parameters in the configuration information corresponding to the first prediction function (or, the second parameter is determined by the multiple parameters).

[0240] Optionally, the first parameter can be a value between 0 and 1.

[0241] By using the product of the second parameter and the first parameter as the target prediction accuracy requirement, the prediction accuracy requirement for the intermediate prediction results of the first AI model is improved, thereby improving the accuracy of event prediction based on the intermediate prediction results of the first AI model.

[0242] For example, the second parameter is the offset parameter Off = 3dB for event A3, and the value of the first parameter can be 0.5. Then the target prediction accuracy requirement can be 3 * 0.5 = 1.5dB. This can improve the prediction accuracy requirement of the intermediate prediction result of the first AI model from 3dB to 1.5dB. This is beneficial to improving the accuracy of event prediction when predicting event A3 based on the offset parameter 3dB.

[0243] For example, the second parameter is a combination of multiple parameters (such as Mp, Ofp, Ocp, Off, Hys, Ofn, Ocn) in the configuration information corresponding to the A3 event, or in other words, it is determined by multiple parameters. For example, the second parameter = Mp + Ofp + Ocp + Off + Hys - Ofn - Ocn. The terminal can determine the target prediction accuracy requirement by multiplying the second parameter and the first parameter.

[0244] In some embodiments of this application, the method 200 further includes:

[0245] If the monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, the terminal performs a first operation, wherein the first operation includes at least one of the following:

[0246] The terminal reports third information to the network-side device;

[0247] The terminal deactivates the first AI model;

[0248] The terminal switches the first AI model to the second AI model;

[0249] The terminal reverts to a non-AI model-based approach to implement the first prediction function.

[0250] If the monitoring results of the first AI model do not meet the target prediction accuracy requirements of the first AI model, the terminal can report third information to the network-side device, or perform model deactivation, switching, or rollback.

[0251] In some embodiments, the third information includes, but is not limited to, at least one of the following:

[0252] The model identifier of the first AI model;

[0253] Functional identifiers of the first AI model;

[0254] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0255] The model monitoring results of the first AI model, such as the prediction accuracy achieved by the first AI model when performing the first prediction function.

[0256] Therefore, by reporting third information to the network-side device, the terminal can learn that the first AI model used by the terminal cannot meet the prediction accuracy requirements. Thus, the network-side device can configure a new AI model for the terminal to meet the target prediction accuracy requirements, thereby balancing the normal operation of the AI ​​model and the prediction performance of the AI ​​model.

[0257] In some embodiments, the model monitoring result of the first AI model may refer to the prediction accuracy achieved by the first AI model in performing the first prediction function.

[0258] In some embodiments, the first prediction function is RRM measurement prediction. The prediction accuracy of the first AI model performing RRM measurement prediction can be determined based on the RRM measurement prediction result and the actual RRM measurement result. For example, the prediction accuracy requirement for RRM measurement prediction can be the difference between the actual RRM measurement result and the RRM measurement prediction result, such as the difference between the RRM measurement result of the second beam and the RRM measurement prediction result of the second beam, or the difference between the RRM measurement result of the cell in the second frequency point and the RRM measurement prediction result of the cell in the second frequency point.

[0259] In some embodiments, the first prediction function is measurement event prediction, and the prediction accuracy of the first AI model performing measurement event prediction may include, for example, at least one of the following:

[0260] The measurement event prediction accuracy, measurement event prediction precision, measurement event prediction recall, measurement event prediction sensitivity, measurement event prediction F1 score, and measurement event prediction specificity.

[0261] In some embodiments, the first prediction function is RLF prediction, and the prediction accuracy of the first AI model performing RLF prediction may include, for example, at least one of the following:

[0262] The accuracy, precision, recall, sensitivity, F1 score, and specificity of RLF predictions.

[0263] In some embodiments, the first prediction function is HOF prediction, and the prediction accuracy of the first AI model performing HOF prediction may include, for example, at least one of the following:

[0264] HOF prediction accuracy, HOF prediction precision, HOF prediction recall, HOF prediction sensitivity, HOF prediction F1 score, and HOF prediction specificity.

[0265] In some embodiments, the first prediction function is target cell prediction, and the prediction accuracy of the first AI model performing target cell prediction may include, for example, at least one of the following:

[0266] Accuracy of target cell prediction, precision of target cell prediction, recall of target cell prediction, sensitivity of target cell prediction, F1 score of target cell prediction, and specificity of target cell prediction.

[0267] In some embodiments, the second AI model supports the first prediction function, and the prediction accuracy of the second AI model meets the target prediction accuracy requirements.

[0268] The following, combined with Figures 6 to 9 This describes the implementation method of the model monitoring method provided in the embodiments of this application.

[0269] Example 1:

[0270] In this embodiment 1, the first information may include the prediction accuracy requirement associated with the prediction window or MRR. The terminal may select the target prediction accuracy requirement based on the prediction window or MRR associated with the first AI model, and perform model monitoring on the first AI model based on the target accuracy requirement.

[0271] like Figure 6 As shown, the model monitoring method may include the following steps:

[0272] S310, the terminal obtains first information, which includes the prediction accuracy requirement associated with the prediction window or MRR.

[0273] The prediction difficulty of an AI model varies depending on the prediction window or MRR. For example, the longer the prediction window and the larger the MRR, the greater the prediction difficulty for the AI ​​model, and the lower the corresponding prediction accuracy requirement (or prediction performance requirement) should be; otherwise, the normal operation of the AI ​​model cannot be guaranteed. Furthermore, different prediction windows or MRRs may have different uses. For instance, a short prediction window can be used to determine whether a measurement event will be triggered in the short term, while a long prediction window can be used to determine whether a ping-pong switch will occur in the future. Since their purposes are different, the corresponding prediction accuracy requirements should also be different to ensure the normal operation of the AI ​​model.

[0274] In Example 1, the corresponding prediction accuracy requirements can be configured for different prediction windows or MRR, which helps to ensure the normal operation of the AI ​​model and also helps to take into account the prediction performance of the AI ​​model.

[0275] In some embodiments, the first information includes at least one of the following:

[0276] At least one prediction window, and the prediction accuracy requirement associated with each prediction window;

[0277] At least one MRR, and the prediction accuracy requirement associated with each MRR.

[0278] In some embodiments, the prediction accuracy requirement can be associated with a prediction window one by one, or a prediction accuracy requirement can be associated with multiple prediction windows.

[0279] In some embodiments, the prediction accuracy requirement can be associated one-to-one with the MRR, or one prediction accuracy requirement can be associated with multiple MRRs.

[0280] Optionally, the first information can be configured by the network-side device, or it can be predefined.

[0281] Step 320: The terminal determines the target prediction accuracy requirement from the first information based on the first prediction window or the first MRR. Here, the first prediction window is the prediction window associated with the first AI model, and the first MRR is the MRR associated with the first AI model.

[0282] In some embodiments, the target prediction accuracy requirement is a prediction accuracy requirement associated with a first prediction window, or a prediction accuracy requirement associated with a first MRR.

[0283] S330: The terminal monitors the first AI model according to the target prediction accuracy requirements.

[0284] For example, a first AI model is used to perform a first prediction function, wherein the first prediction function includes at least one of the following: RRM measurement prediction, measurement event prediction, RLF prediction, HOF prediction, and target cell prediction.

[0285] S340, if the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, the terminal executes the first operation.

[0286] In some embodiments, the model monitoring result of the first AI model may refer to the prediction accuracy achieved by the first AI model in performing the first prediction function.

[0287] In some embodiments, the first operation includes at least one of the following:

[0288] The terminal reports third information to the network-side device;

[0289] The terminal deactivates the first AI model;

[0290] The terminal switches the first AI model to the second AI model;

[0291] The terminal reverts to a non-AI model-based approach to implement the first prediction function.

[0292] In some embodiments, the third information includes at least one of the following:

[0293] The model identifier of the first AI model;

[0294] Functional identifiers of the first AI model;

[0295] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0296] The model monitoring results of the first AI model, such as the prediction accuracy achieved by the first AI model when performing the first prediction function.

[0297] Therefore, in this embodiment of the application, by reporting third information to the network-side device, the terminal can learn that the first AI model used by the terminal cannot meet the prediction accuracy requirements. Thus, the network-side device can configure a new AI model for the terminal to meet the target prediction accuracy requirements, thereby taking into account both the normal operation of the AI ​​model and the prediction performance of the AI ​​model.

[0298] Example 2:

[0299] In this embodiment 2, the first information may include the prediction accuracy requirements associated with frequency points, cells, or beams. The terminal may select the target prediction accuracy requirements based on the frequency points, cells, or beams associated with the first AI model, and perform model monitoring on the first AI model based on the target accuracy requirements.

[0300] like Figure 7 As shown, the model monitoring method may include the following steps:

[0301] S410, the terminal obtains the first information, which includes the prediction accuracy requirements of frequency point, cell or beam association.

[0302] The AI ​​model's prediction difficulty varies depending on the frequency, cell, or beam (where "different" refers to the input frequency, cell, or beam to the AI ​​model differing from the frequency, cell, or beam predicted by the AI ​​model). For example, the AI ​​model performs well when the target cell and serving cell operate on the same frequency, but poorly when they operate on different frequencies. Similarly, the AI ​​model performs well when the target cell and serving cell are co-located, but poorly when they are not. Furthermore, the AI ​​model performs well when predicting a high-frequency cell using a low-frequency cell, but poorly when predicting a low-frequency cell using a high-frequency cell.

[0303] In Example 2, corresponding prediction accuracy requirements can be configured for different frequency points, cells, or beams, which helps to ensure the normal operation of the AI ​​model and also helps to take into account the prediction performance of the AI ​​model.

[0304] In some embodiments, the first information includes at least one of the following:

[0305] At least one frequency point, and the prediction accuracy requirement associated with each frequency point;

[0306] At least one cell, and the prediction accuracy requirements associated with each cell;

[0307] At least one beam, and the prediction accuracy requirements associated with each beam.

[0308] In some embodiments, the prediction accuracy requirement may be associated one-to-one with a frequency point, cell, or beam, or a prediction accuracy requirement may be associated with multiple frequency points, cells, or beams.

[0309] Optionally, in this embodiment 2, the prediction accuracy requirements for frequency points, cells, or beam associations can be configured in the measurement object.

[0310] Optionally, the first information can be configured by the network-side device, or it can be predefined.

[0311] Step 420: The terminal determines the target prediction accuracy requirement based on the first frequency point, the first cell, or the first beam from the first information. Here, the first frequency point is the frequency point associated with the first AI model, the first cell is the cell associated with the first AI model, and the first beam is the beam associated with the first AI model.

[0312] In some embodiments, the target prediction accuracy requirement is the prediction accuracy requirement associated with a first frequency point, a first cell, or a first beam.

[0313] In some embodiments, associating the first AI model with the first frequency point can be understood as the output of the first AI model being the prediction result of the first frequency point. Optionally, the first frequency point can be the frequency point to be measured or predicted by the terminal.

[0314] In some embodiments, associating the first AI model with the first cell can be understood as the output of the first AI model being a prediction result of the first cell, such as the RRM measurement prediction result of the first cell; or, the first AI model can be used to predict the RRM measurement prediction result of the first cell. Optionally, the first cell can be the cell to be measured or predicted by the terminal.

[0315] In some embodiments, associating the first AI model with the first beam can be understood as the output of the first AI model being a prediction result of the first beam, such as an RRM measurement prediction result of the first beam, or the first AI model being used to predict the RRM measurement prediction result of the first beam. Optionally, the first beam can be the beam to be measured or predicted at the terminal.

[0316] S430: The terminal monitors the first AI model according to the target prediction accuracy requirements.

[0317] For example, a first AI model is used to perform a first prediction function, wherein the first prediction function includes at least one of the following: RRM measurement prediction, measurement event prediction, RLF prediction, HOF prediction, and target cell prediction.

[0318] S440, if the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, the terminal executes the first operation.

[0319] In some embodiments, the model monitoring result of the first AI model may refer to the prediction accuracy achieved by the first AI model in performing the first prediction function.

[0320] In some embodiments, the first operation includes at least one of the following:

[0321] The terminal reports third information to the network-side device;

[0322] The terminal deactivates the first AI model;

[0323] The terminal switches the first AI model to the second AI model;

[0324] The terminal reverts to a non-AI model-based approach to implement the first prediction function.

[0325] In some embodiments, the third information includes at least one of the following:

[0326] The model identifier of the first AI model;

[0327] Functional identifiers of the first AI model;

[0328] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0329] The model monitoring results of the first AI model, such as the prediction accuracy achieved by the first AI model when performing the first prediction function.

[0330] Therefore, in this embodiment of the application, by reporting third information to the network-side device, the terminal can learn that the first AI model used by the terminal cannot meet the prediction accuracy requirements. Thus, the network-side device can configure a new AI model for the terminal to meet the target prediction accuracy requirements, thereby taking into account both the normal operation of the AI ​​model and the prediction performance of the AI ​​model.

[0331] Example 3:

[0332] In this embodiment 3, the first information may include the prediction accuracy requirement associated with the AI ​​function or AI model. The terminal may select the target prediction accuracy requirement based on the first AI model, or the AI ​​function supported by the first AI model, and perform model monitoring on the first AI model based on the target accuracy requirement.

[0333] like Figure 8 As shown, the model monitoring method in this application embodiment may include the following steps:

[0334] S510, the terminal obtains the first information, which includes the prediction accuracy requirements associated with the AI ​​function or AI model.

[0335] Optionally, when the prediction window, MRR, frequency, cell, and beam associated with an AI model or AI function are fixed, the prediction accuracy requirement can be associated with the AI ​​model or AI function, thereby indirectly associating the prediction accuracy requirement with the prediction window, MRR, frequency, cell, and beam.

[0336] In some embodiments, the first information includes at least one of the following:

[0337] At least one AI model, and the required prediction accuracy for each AI model;

[0338] At least one AI function, and the prediction accuracy requirements associated with each AI function.

[0339] In some embodiments, the prediction accuracy requirement can be associated with an AI model one-to-one. For example, when an AI model performs a fixed prediction function, the AI ​​model can be associated with a prediction accuracy requirement.

[0340] In some embodiments, an AI model may be associated with multiple prediction accuracy requirements. For example, when an AI model performs multiple prediction functions, each prediction function may be associated with a prediction accuracy requirement.

[0341] In some embodiments, prediction accuracy requirements can be associated one-to-one with AI functions, or one AI function can be associated with multiple prediction accuracy requirements, each associated with a different prediction window, MRR, frequency point, cell, or beam.

[0342] Optionally, when multiple AI functions or AI models have different functions, the required prediction accuracy of the associated multiple AI functions or multiple AI models may differ.

[0343] Optionally, when multiple AI functions or AI models have the same function, but the prediction windows, MRR, frequency points, cells or beams corresponding to the multiple AI functions or AI models are different, the prediction accuracy requirements associated with the multiple AI functions or AI models are different. This can ensure the normal operation of the AI ​​functions or AI models when the multiple AI functions or AI models perform different prediction functions, and take into account the prediction performance of the multiple AI functions or AI models.

[0344] Optionally, the first information can be configured by the network-side device, or it can be predefined.

[0345] Step 520: The terminal determines the target prediction accuracy requirement from the first information based on the first AI model or the first AI function. The first AI function is the AI ​​function supported by the first AI model.

[0346] In some embodiments, the target prediction accuracy requirement is the prediction accuracy requirement associated with the first AI model or the first AI function.

[0347] S530: The terminal monitors the first AI model according to the target prediction accuracy requirements.

[0348] For example, a first AI model is used to perform a first prediction function, wherein the first prediction function includes at least one of the following: RRM measurement prediction, measurement event prediction, RLF prediction, HOF prediction, and target cell prediction.

[0349] S540, if the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, the terminal executes the first operation.

[0350] In some embodiments, the model monitoring result of the first AI model may refer to the prediction accuracy achieved by the first AI model in performing the first prediction function.

[0351] In some embodiments, the first operation includes at least one of the following:

[0352] The terminal reports third information to the network-side device;

[0353] The terminal deactivates the first AI model;

[0354] The terminal switches the first AI model to the second AI model;

[0355] The terminal reverts to a non-AI model-based approach to implement the first prediction function.

[0356] In some embodiments, the third information includes at least one of the following:

[0357] The model identifier of the first AI model;

[0358] Functional identifiers of the first AI model;

[0359] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0360] The model monitoring results of the first AI model, such as the prediction accuracy achieved by the first AI model when performing the first prediction function.

[0361] Therefore, in this embodiment of the application, by reporting third information to the network-side device, the terminal can learn that the first AI model used by the terminal cannot meet the prediction accuracy requirements. Thus, the network-side device can configure a new AI model for the terminal to meet the target prediction accuracy requirements, thereby taking into account both the normal operation of the AI ​​model and the prediction performance of the AI ​​model.

[0362] Example 4:

[0363] In this embodiment 4, when the terminal performs indirect measurement event prediction, the prediction accuracy requirement can be determined based on the configuration information and the first parameter corresponding to the measurement event.

[0364] like Figure 9 As shown, the model monitoring method in this application embodiment may include the following steps:

[0365] S610, obtain first information, which includes configuration information and first parameters corresponding to the measurement event.

[0366] The first parameter is used to determine the required prediction accuracy for the association between the measurement event prediction and the prediction.

[0367] Optionally, the first parameter can be between 0 and 1.

[0368] Optionally, the configuration information corresponding to the measurement event includes at least one of the following parameters:

[0369] Mn: Neighboring cell measurement results, without considering any offset;

[0370] Ofn: Specific offset of the neighboring cell measurement object;

[0371] Ocn: Neighboring cell-level specific offset;

[0372] Mp: SpCell measurement result, without considering any offset;

[0373] Ofp: SpCell measures a specific offset of an object;

[0374] Ocp: SpCell cell-level specific offset;

[0375] Hys: The hysteresis parameter of the event;

[0376] Off: The offset parameter for the event.

[0377] S620 determines the target prediction accuracy requirement based on the configuration information and first parameter corresponding to the measurement event.

[0378] For example, the target prediction accuracy requirement is determined based on a first parameter and a second parameter, wherein the second parameter includes at least one of the following parameters:

[0379] Mn: Neighbor cell measurement results, without considering any offset;

[0380] Ofn: Specific offset of the neighboring cell measurement object;

[0381] Ocn: Neighboring cell-level specific offset;

[0382] Mp: SpCell measurement result, without considering any offset;

[0383] Ofp: SpCell measures a specific offset of an object;

[0384] Ocp: SpCell cell-level specific offset;

[0385] Hys: The hysteresis parameter of the event;

[0386] Off: The offset parameter for the event.

[0387] For example, the product of the first parameter and the second parameter can be used as the target prediction accuracy requirement.

[0388] For example, for event A3, assuming the event is configured to meet the entry condition when the neighboring cell is 3dB higher than the serving cell, if the accuracy error of the AI ​​model's RRM measurement prediction result is higher than 3dB, the accuracy of the measurement event prediction is likely not guaranteed. Therefore, the accuracy requirement for measurement event prediction is related to the event configuration. In embodiment 4, by configuring the first parameter, the terminal determines the prediction accuracy requirement used for indirect measurement event prediction based on the configuration information corresponding to the measurement event and the first parameter, ensuring that the AI ​​model used for measurement event prediction can operate normally and helping to ensure the accuracy of measurement event prediction.

[0389] S640, if the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, the terminal executes the first operation.

[0390] In some embodiments, the model monitoring result of the first AI model may refer to the prediction accuracy achieved by the first AI model in performing the first prediction function.

[0391] In some embodiments, the first operation includes at least one of the following:

[0392] The terminal reports third information to the network-side device;

[0393] The terminal deactivates the first AI model;

[0394] The terminal switches the first AI model to the second AI model;

[0395] The terminal reverts to a non-AI model-based approach to implement the first prediction function.

[0396] In some embodiments, the third information includes at least one of the following:

[0397] The model identifier of the first AI model;

[0398] Functional identifiers of the first AI model;

[0399] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0400] The model monitoring results of the first AI model, such as the prediction accuracy achieved by the first AI model when performing the first prediction function.

[0401] In this embodiment, the terminal reports third information to the network-side device, which can then know that the first AI model used by the terminal cannot meet the prediction accuracy requirements. The network-side device can then configure a new AI model for the terminal to meet the target prediction accuracy requirements, thereby balancing the normal operation of the AI ​​model and its prediction performance.

[0402] In summary, in the embodiments of this application, the terminal can select at least one of the prediction accuracy requirements associated with the prediction window, MRR, frequency point, cell, AI model, and AI function to perform model monitoring on the first AI model based on the first information. Alternatively, the terminal can determine the target prediction accuracy requirement used to perform model monitoring on the first AI model based on the configuration information and / or first parameters corresponding to the first prediction function. This is beneficial for the terminal to select appropriate prediction accuracy requirements to perform model monitoring on the first AI model, ensuring the normal operation of the AI ​​model and taking into account the prediction performance of the AI ​​model.

[0403] The model monitoring method provided in this application can be executed by a wireless communication device. This application uses a wireless communication device executing the model monitoring method as an example to illustrate the wireless communication device provided in this application.

[0404] This application provides a wireless communication device. As an example, the wireless communication device may be a communication equipment or a component within a communication equipment, such as a chip. The communication equipment may be a terminal, a network-side device, or a server, etc. Exemplarily, the terminal may include, but is not limited to, the type of terminal 11 listed above, and the network-side device may include, but is not limited to, the type of network-side device 12 listed above. This application does not impose specific limitations.

[0405] The wireless communication device includes a receiving module, a transmitting module, and a processing module. These modules can be implemented in software or hardware. When implemented in hardware, the processing module can be implemented by a processor. For example, the processor can include general-purpose processors, special-purpose processors, such as a Central Processing Unit (CPU), microprocessor, Digital Signal Processor (DSP), Artificial Intelligence (AI) processor, Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Network Processor (NP), Field Programmable Gate Array (FPGA), or other programmable logic devices, gate circuits, transistors, discrete hardware components, etc. The receiving and transmitting modules can be implemented by a communication interface, which can include one or more of the following: transceiver, pins, circuits, bus, radio frequency unit, etc.

[0406] Optionally, see Figure 10 When the wireless communication device is a terminal or a component within a terminal, the wireless communication device 1500 includes:

[0407] Processing module 1510 is used to acquire first information; and

[0408] Based on the first information, the first artificial intelligence (AI) model is monitored, wherein the first AI model is used to perform a first prediction function;

[0409] The first information includes at least one of the following:

[0410] One or more prediction accuracy requirements;

[0411] Configuration information corresponding to the first prediction function;

[0412] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0413] The one or more prediction accuracy requirements are associated with at least one of the following:

[0414] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0415] In some embodiments, the processing module 1510 is further configured to:

[0416] Based on the second information, determine the target prediction accuracy requirement from among the one or more prediction accuracy requirements;

[0417] The first AI model is monitored according to the target prediction accuracy requirement;

[0418] The second information includes at least one of the following:

[0419] The prediction window associated with the first AI model;

[0420] The MRR associated with the first AI model;

[0421] The frequency points associated with the first AI model;

[0422] The cell associated with the first AI model;

[0423] The beam associated with the first AI model;

[0424] The AI ​​functions supported by the first AI model;

[0425] The first AI model.

[0426] In some embodiments, the target prediction accuracy requirement is associated with at least one of the following:

[0427] The first prediction window is the prediction window associated with the first AI model;

[0428] The first MRR is the MRR associated with the first AI model;

[0429] The first frequency point is the frequency point associated with the first AI model;

[0430] The first cell, the first frequency point is the cell associated with the first AI model;

[0431] The first beam is the beam associated with the first AI model;

[0432] The first AI function is the AI ​​function supported by the first AI model.

[0433] The first AI model.

[0434] In some embodiments, the processing module 1510 is further configured to:

[0435] Based on the configuration information corresponding to the first prediction function and the first parameter, the target prediction accuracy requirement is determined;

[0436] Based on the target prediction accuracy requirements, the first AI model is monitored.

[0437] In some embodiments, the processing module 1510 is further configured to:

[0438] If the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, a first operation is performed, wherein the first operation includes at least one of the following:

[0439] Report third-party information to network-side devices;

[0440] Deactivate the first AI model;

[0441] Switch the first AI model to the second AI model;

[0442] The first prediction function is reverted to a non-AI model-based approach.

[0443] The third information includes at least one of the following:

[0444] The model identifier of the first AI model;

[0445] Functional identifiers of the first AI model;

[0446] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0447] The model monitoring results of the first AI model.

[0448] In some embodiments, the processing module 1510 is further configured to:

[0449] The first information is obtained by predefined information; or

[0450] Obtain the first information from the network-side device.

[0451] In some embodiments, the first information includes a prediction accuracy requirement associated with at least one of frequency point, cell, and beam, wherein the prediction accuracy requirement associated with at least one of frequency point, cell, and beam is configured by the network-side device in first configuration information, which is used to configure the measurement object.

[0452] In some embodiments, the first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window, the second prediction accuracy requirement is associated with a second prediction window, and the first prediction accuracy requirement is lower than the second prediction accuracy requirement when the length of the first prediction window is greater than the length of the second prediction window, or when the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time.

[0453] In some embodiments, the value of the first parameter is between 0 and 1.

[0454] In some embodiments, the one or more prediction accuracy requirements include at least one of the following:

[0455] Prediction accuracy requirements for RRM measurement prediction in wireless resource management;

[0456] Prediction accuracy requirements for measuring event prediction;

[0457] Prediction accuracy requirements for RLF prediction in wireless link failure;

[0458] Prediction accuracy requirements for switching failed HOF predictions;

[0459] Prediction accuracy requirements for target cell prediction.

[0460] In some embodiments, the first prediction function includes at least one of the following:

[0461] RRM measurement prediction;

[0462] Measurement event prediction;

[0463] RLF forecast;

[0464] HOF prediction;

[0465] Target cell prediction.

[0466] In some embodiments, the configuration information corresponding to the first prediction function includes the configuration information corresponding to the measurement event prediction.

[0467] Therefore, in this embodiment, the terminal can select at least one of the prediction accuracy requirements associated with the prediction window, MRR, frequency point, cell, AI model, and AI function to perform model monitoring on the first AI model based on the first information. Alternatively, the terminal can determine the target prediction accuracy requirement used to perform model monitoring on the first AI model based on the configuration information and / or first parameters corresponding to the first prediction function. This is beneficial for the terminal to select appropriate prediction accuracy requirements to perform model monitoring on the first AI model, ensuring the normal operation of the AI ​​model and taking into account the prediction performance of the AI ​​model.

[0468] See Figure 11 When the wireless communication device is a network-side device or a component within a network-side device, the wireless communication device 1600 includes:

[0469] The sending module 1610 is used to send first information to the terminal, the first information being used by the terminal to monitor the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function;

[0470] The first information includes at least one of the following:

[0471] One or more prediction accuracy requirements;

[0472] Configuration information corresponding to the first prediction function;

[0473] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0474] The one or more prediction accuracy requirements are associated with at least one of the following:

[0475] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0476] In some embodiments, the value of the first parameter is between 0 and 1.

[0477] In some embodiments, the first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window, the second prediction accuracy requirement is associated with a second prediction window, and the first prediction accuracy requirement is lower than the second prediction accuracy requirement when the length of the first prediction window is greater than the length of the second prediction window, or when the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time.

[0478] In some embodiments, the first information includes a prediction accuracy requirement associated with at least one of frequency point, cell, and beam, wherein the prediction accuracy requirement associated with at least one of frequency point, cell, and beam is configured by the network-side device in first configuration information, which is used to configure the measurement object.

[0479] In some embodiments, the one or more prediction accuracy requirements include at least one of the following:

[0480] Prediction accuracy requirements for RRM measurement prediction in wireless resource management;

[0481] Prediction accuracy requirements for measuring event prediction;

[0482] Prediction accuracy requirements for RLF prediction in wireless link failure;

[0483] Prediction accuracy requirements for switching failed HOF predictions;

[0484] Prediction accuracy requirements for target cell prediction.

[0485] In some embodiments, the first prediction function includes at least one of the following:

[0486] RRM measurement prediction;

[0487] Measurement event prediction;

[0488] RLF forecast;

[0489] HOF prediction;

[0490] Target cell prediction.

[0491] In some embodiments, the device 1600 further includes:

[0492] A receiving module is used to receive third information from the terminal;

[0493] The third information includes at least one of the following:

[0494] The model identifier of the first AI model;

[0495] Functional identifiers of the first AI model;

[0496] The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements;

[0497] The model monitoring results of the first AI model.

[0498] In some embodiments, the configuration information corresponding to the first prediction function includes the configuration information corresponding to the measurement event prediction.

[0499] Therefore, in this embodiment, the network-side device can provide the terminal with first information, so that the terminal can select at least one of the prediction accuracy requirements associated with the prediction window, MRR, frequency point, cell, AI model, and AI function to perform model monitoring on the first AI model, or determine the target prediction accuracy requirement used to perform model monitoring on the first AI model based on the configuration information and / or first parameters corresponding to the first prediction function. This is beneficial for the terminal to select appropriate prediction accuracy requirements to perform model monitoring on the first AI model, ensuring the normal operation of the AI ​​model, and taking into account the prediction performance of the AI ​​model.

[0500] The wireless communication device provided in this application embodiment can achieve... Figures 5 to 9 The various processes implemented in the method embodiments achieve the same technical effect, and will not be described again here to avoid repetition.

[0501] like Figure 12 As shown, this application embodiment also provides a communication device 800, including a processor 801 and a memory 802. The memory 802 stores a program or instructions that can run on the processor 801. For example, when the communication device 800 is a terminal, the program or instructions executed by the processor 801 implement the above-mentioned... Figures 5 to 9 The various steps executed by the terminal in the method embodiment can achieve the same technical effect. When the communication device 800 is a network-side device, the program or instruction executed by the processor 801 implements the above. Figures 5 to 9 The steps performed by the network-side device in the method embodiment can achieve the same technical effect, so they will not be described again here to avoid repetition.

[0502] This application embodiment also provides a terminal, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement, for example... Figures 5 to 9 The steps in the method embodiment shown are illustrated. This terminal embodiment corresponds to the above-described terminal-side method embodiment. All implementation processes and methods of the above-described method embodiments can be applied to this terminal embodiment and achieve the same technical effect. The terminal can be... Figure 10 The wireless communication device 1500 shown is included. Figure 13 A schematic diagram of the hardware structure of a terminal to implement an embodiment of this application.

[0503] The terminal 900 includes, but is not limited to, at least some of the following components: radio frequency unit 901, network module 902, audio output unit 903, input unit 904, sensor 905, display unit 906, user input unit 907, interface unit 908, memory 909, and processor 910.

[0504] Those skilled in the art will understand that the terminal 900 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 910 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 13 The terminal structure shown does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0505] It should be understood that, in this embodiment, the input unit 904 may include a graphics processor 9041 and a microphone 9042. The graphics processor 9041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 906 may include a display panel 9061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 907 includes at least one of a touch panel 9071 and other input devices 9072. The touch panel 9071 is also called a touch screen. The touch panel 9071 may include a touch detection device and a touch controller. Other input devices 9072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0506] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 901 can transmit it to the processor 910 for processing; in addition, the radio frequency unit 901 can send uplink data to the network-side device. Typically, the radio frequency unit 901 includes, but is not limited to, antennas, amplifiers, transceivers, couplers, low-noise amplifiers, duplexers, etc.

[0507] The memory 909 can be used to store software programs or instructions, as well as various data. The memory 909 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 909 may include volatile memory or non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 909 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0508] Processor 910 may include one or more processing units; optionally, processor 910 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 910.

[0509] Among them, processor 910 is used to acquire the first information;

[0510] Based on the first information, the first artificial intelligence (AI) model is monitored, wherein the first AI model is used to perform a first prediction function;

[0511] The first information includes at least one of the following:

[0512] One or more prediction accuracy requirements;

[0513] Configuration information corresponding to the first prediction function;

[0514] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0515] The one or more prediction accuracy requirements are associated with at least one of the following:

[0516] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0517] It is understood that the implementation process of each implementation method mentioned in this embodiment can be referred to the method embodiment. Figures 5 to 9 The relevant descriptions and the achievement of the same or corresponding technical effects will not be repeated here to avoid duplication.

[0518] This application embodiment also provides a network-side device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement, for example... Figure 14 The steps of the method embodiment shown are illustrated. This network-side device embodiment corresponds to the above-described network-side device method embodiment. All implementation processes and methods of the above-described method embodiments can be applied to this network-side device embodiment and can achieve the same technical effect.

[0519] In this application embodiment, a network-side device is also provided, which may be: Figure 11 The wireless communication device 1600 is shown. (For example...) Figure 14 As shown, the network-side device 1000 includes: an antenna 1001, a radio frequency (RF) device 1002, a baseband device 1003, a processor 1004, and a memory 1005. The antenna 1001 is connected to the RF device 1002. In the uplink direction, the RF device 1002 receives information through the antenna 1001 and transmits the received information to the baseband device 1003 for processing. In the downlink direction, the baseband device 1003 processes the information to be transmitted and sends it to the RF device 1002. The RF device 1002 processes the received information and transmits it through the antenna 1001.

[0520] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 1003, which includes a baseband processor.

[0521] The baseband device 1003 may include, for example, at least one baseband board on which multiple chips are disposed, such as... Figure 14 As shown, one of the chips is, for example, a baseband processor, which is connected to the memory 1005 via a bus interface to call the program in the memory 1005 and execute the network device operation shown in the above method embodiment.

[0522] The network-side device may also include a network interface 1006, such as a Common Public Radio Interface (CPRI).

[0523] The network-side device 1000 in this embodiment further includes: instructions or programs stored in memory 1005 and executable on processor 1004, wherein processor 1004 calls the instructions or programs in memory 1005 to execute. Figure 11 The methods executed by each module shown achieve the same technical effect, and to avoid repetition, they will not be described in detail here.

[0524] The radio frequency device 1002 is used to: send first information to the terminal, the first information being used by the terminal to monitor the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function;

[0525] The first information includes at least one of the following:

[0526] One or more prediction accuracy requirements;

[0527] Configuration information corresponding to the first prediction function;

[0528] The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function;

[0529] The one or more prediction accuracy requirements are associated with at least one of the following:

[0530] Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

[0531] This application also provides a network-side device. For example... Figure 15 As shown, the network-side device 1100 includes: a processor 1101, a network interface 1102, and a memory 1103. This network-side device can be... Figure 11 The wireless communication device shown. The network interface 1102 is, for example, a common public radio interface (CPRI).

[0532] The network-side device 1100 in this embodiment further includes: instructions or programs stored in memory 1103 and executable on processor 1101, wherein processor 1101 calls the instructions or programs in memory 1103 to execute... Figure 11 The methods executed by each module shown achieve the same technical effect, and to avoid repetition, they will not be described in detail here.

[0533] This application embodiment also provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the above-described functionality. Figures 5 to 9 The various processes in the method embodiments can achieve the same technical effect, and will not be described again here to avoid repetition.

[0534] The processor mentioned above is the processor in the terminal described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. In some examples, the readable storage medium may be a non-transient readable storage medium.

[0535] This application embodiment also provides a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the above. Figures 5 to 9 The various processes in the method embodiments can achieve the same technical effect, and will not be described again here to avoid repetition.

[0536] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0537] This application embodiment also provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the above. Figures 5 to 9 The various processes in the method embodiments can achieve the same technical effect, and will not be described again here to avoid repetition.

[0538] This application also provides a wireless communication system, including a terminal and a network-side device. The terminal can be used to execute the steps of the model monitoring method described above, and the network-side device can be used to execute the steps of the model monitoring method described above.

[0539] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0540] From the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of computer software products plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions to cause the terminal or network-side device to execute the methods described in the various embodiments of this application.

[0541] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other implementations under the guidance of this application without departing from the spirit and scope of the claims. All of these implementations are within the protection scope of this application.

Claims

1. A model monitoring method, characterized in that, include: The terminal obtains the first information; Based on the first information, the first artificial intelligence (AI) model is monitored, wherein the first AI model is used to perform a first prediction function; The first information includes at least one of the following: One or more prediction accuracy requirements; Configuration information corresponding to the first prediction function; The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function; The one or more prediction accuracy requirements are associated with at least one of the following: Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

2. The method according to claim 1, characterized in that, The step of monitoring the first artificial intelligence (AI) model based on the first information includes: Based on the second information, determine the target prediction accuracy requirement from among the one or more prediction accuracy requirements; The first AI model is monitored according to the target prediction accuracy requirement; The second information includes at least one of the following: The prediction window associated with the first AI model; The MRR associated with the first AI model; The frequency points associated with the first AI model; The cell associated with the first AI model; The beam associated with the first AI model; The AI ​​functions supported by the first AI model; The first AI model.

3. The method according to claim 2, characterized in that, The target prediction accuracy requirement is associated with at least one of the following: The first prediction window is the prediction window associated with the first AI model; The first MRR is the MRR associated with the first AI model; The first frequency point is the frequency point associated with the first AI model; The first cell, the first frequency point is the cell associated with the first AI model; The first beam is the beam associated with the first AI model; The first AI function is the AI ​​function supported by the first AI model. The first AI model.

4. The method according to claim 1, characterized in that, The step of monitoring the first artificial intelligence (AI) model based on the first information includes: Based on the configuration information corresponding to the first prediction function and the first parameter, the target prediction accuracy requirement is determined; Based on the target prediction accuracy requirements, the first AI model is monitored.

5. The method according to any one of claims 2-4, characterized in that, The method further includes: If the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, the terminal performs a first operation, wherein the first operation includes at least one of the following: The terminal reports third information to the network-side device; The terminal deactivates the first AI model; The terminal switches the first AI model to the second AI model; The terminal reverts to a non-AI model-based approach to perform the first prediction function. The third information includes at least one of the following: The model identifier of the first AI model; Functional identifiers of the first AI model; The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements; The model monitoring results of the first AI model.

6. The method according to any one of claims 1-5, characterized in that, The terminal acquires the first information, including: The terminal obtains the first information through predefined information; or The terminal obtains the first information from the network-side device.

7. The method according to claim 6, characterized in that, The first information includes a prediction accuracy requirement associated with at least one of frequency point, cell, and beam, and the prediction accuracy requirement associated with at least one of frequency point, cell, and beam is configured by the network-side device in the first configuration information, which is used to configure the measurement object.

8. The method according to any one of claims 1-7, characterized in that, The first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window and the second prediction accuracy requirement is associated with a second prediction window. If the length of the first prediction window is greater than the length of the second prediction window, or if the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time, the first prediction accuracy requirement is lower than the second prediction accuracy requirement.

9. The method according to any one of claims 1-8, characterized in that, The value of the first parameter is between 0 and 1.

10. The method according to any one of claims 1-9, characterized in that, The one or more prediction accuracy requirements include at least one of the following: Prediction accuracy requirements for RRM measurement prediction in wireless resource management; Prediction accuracy requirements for measuring event prediction; Prediction accuracy requirements for RLF prediction in wireless link failure; Prediction accuracy requirements for switching failed HOF predictions; Prediction accuracy requirements for target cell prediction.

11. The method according to any one of claims 1-10, characterized in that, The first prediction function includes at least one of the following: RRM measurement prediction; Measurement event prediction; RLF forecast; HOF prediction; Target cell prediction.

12. The method according to any one of claims 1-11, characterized in that, The configuration information corresponding to the first prediction function includes: configuration information corresponding to the prediction of measurement events.

13. A model monitoring method, characterized in that, include: The network-side device sends first information to the terminal, the first information being used by the terminal to monitor the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function; The first information includes at least one of the following: One or more prediction accuracy requirements; Configuration information corresponding to the first prediction function; The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function; The one or more prediction accuracy requirements are associated with at least one of the following: Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

14. The method according to claim 13, characterized in that, The value of the first parameter is between 0 and 1.

15. The method according to claim 13 or 14, characterized in that, The first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window and the second prediction accuracy requirement is associated with a second prediction window. If the length of the first prediction window is greater than the length of the second prediction window, or if the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time, the first prediction accuracy requirement is lower than the second prediction accuracy requirement.

16. The method according to any one of claims 13-15, characterized in that, The first information includes a prediction accuracy requirement associated with at least one of frequency point, cell, and beam, and the prediction accuracy requirement associated with at least one of frequency point, cell, and beam is configured by the network-side device in the first configuration information, which is used to configure the measurement object.

17. The method according to any one of claims 13-16, characterized in that, The one or more prediction accuracy requirements include at least one of the following: Prediction accuracy requirements for RRM measurement prediction in wireless resource management; Prediction accuracy requirements for measuring event prediction; Prediction accuracy requirements for RLF prediction in wireless link failure; Prediction accuracy requirements for switching failed HOF predictions; Prediction accuracy requirements for target cell prediction.

18. The method according to any one of claims 13-17, characterized in that, The first prediction function includes at least one of the following: RRM measurement prediction; Measurement event prediction; RLF forecast; HOF prediction; Target cell prediction.

19. The method according to any one of claims 13-18, characterized in that, The method further includes: The network-side device receives third information from the terminal; The third information includes at least one of the following: The model identifier of the first AI model; Functional identifiers of the first AI model; The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements; The model monitoring results of the first AI model.

20. The method according to any one of claims 13-19, characterized in that, The configuration information corresponding to the first prediction function includes: configuration information corresponding to the prediction of measurement events.

21. A wireless communication device, characterized in that, include: The processing module is used to obtain the first information; Based on the first information, the first artificial intelligence (AI) model is monitored, wherein the first AI model is used to perform the first prediction function; The first information includes at least one of the following: One or more prediction accuracy requirements; Configuration information corresponding to the first prediction function; The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function; The one or more prediction accuracy requirements are associated with at least one of the following: Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

22. The apparatus according to claim 21, characterized in that, The processing module is also used for: Based on the second information, determine the target prediction accuracy requirement from among the one or more prediction accuracy requirements; The first AI model is monitored according to the target prediction accuracy requirement; The second information includes at least one of the following: The prediction window associated with the first AI model; The MRR associated with the first AI model; The frequency points associated with the first AI model; The cell associated with the first AI model; The beam associated with the first AI model; The AI ​​functions supported by the first AI model; The first AI model.

23. The apparatus according to claim 21, characterized in that, The processing module is also used for: Based on the configuration information corresponding to the first prediction function and the first parameter, the target prediction accuracy requirement is determined; Based on the target prediction accuracy requirements, the first AI model is monitored.

24. The apparatus according to any one of claims 21-23, characterized in that, The processing module is also used for: If the model monitoring result of the first AI model is less than or equal to the target prediction accuracy requirement, a first operation is performed, wherein the first operation includes at least one of the following: Report third-party information to network-side devices; Deactivate the first AI model; Switch the first AI model to the second AI model; The first prediction function is reverted to a non-AI model-based approach. The third information includes at least one of the following: The model identifier of the first AI model; Functional identifiers of the first AI model; The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements; The model monitoring results of the first AI model.

25. The apparatus according to any one of claims 21-24, characterized in that, The first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window and the second prediction accuracy requirement is associated with a second prediction window. If the length of the first prediction window is greater than the length of the second prediction window, or if the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time, the first prediction accuracy requirement is lower than the second prediction accuracy requirement.

26. A wireless communication device, characterized in that, include: The sending module is used to send first information to the terminal, the first information being used by the terminal to monitor the first artificial intelligence (AI) model, wherein the first AI model is used to perform a first prediction function; The first information includes at least one of the following: One or more prediction accuracy requirements; Configuration information corresponding to the first prediction function; The first parameter is used to determine the prediction accuracy requirement associated with the first prediction function; The one or more prediction accuracy requirements are associated with at least one of the following: Prediction window, measurement reduction ratio (MRR), frequency point, cell, beam, AI function, AI model.

27. The apparatus according to claim 26, characterized in that, The first information includes a first prediction accuracy requirement and a second prediction accuracy requirement, wherein the first prediction accuracy requirement is associated with a first prediction window and the second prediction accuracy requirement is associated with a second prediction window. If the length of the first prediction window is greater than the length of the second prediction window, or if the time interval between the first prediction window and the current time is greater than the time interval between the second prediction window and the current time, the first prediction accuracy requirement is lower than the second prediction accuracy requirement.

28. The apparatus according to claim 26 or 27, characterized in that, The first information includes a prediction accuracy requirement associated with at least one of frequency point, cell, and beam, which is configured in the first configuration information used to configure the measurement object.

29. The apparatus according to any one of claims 26-28, characterized in that, The device further includes: A receiving module is used to receive third information from the terminal; The third information includes at least one of the following: The model identifier of the first AI model; Functional identifiers of the first AI model; The first indication is used to indicate that the first AI model does not meet the prediction accuracy requirements; The model monitoring results of the first AI model.

30. A terminal, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the model monitoring method as described in any one of claims 1 to 12.

31. A network-side device, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the model method as described in any one of claims 13 to 20.

32. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the model monitoring method as described in any one of claims 1-12, or implement the steps of the model monitoring method as described in any one of claims 13 to 20.

Images