Model monitoring method and communication equipment

Abstract

The invention provides a model monitoring method and communication equipment. The method comprises the following steps: a first device executes target model monitoring on a first model, wherein the target model monitoring comprises one of the following: cycle-based model monitoring; model monitoring based on event triggering; model monitoring triggered based on the second device indication; model monitoring based on prediction data, the prediction data being predicted based on measurement data. In the embodiment of the invention, target model monitoring is introduced to perform model monitoring on the model (also known as the first model), and compared with a scheme of continuously performing model monitoring on the model in a traditional scheme, the execution rationality of model monitoring in an actual process can be improved.

Description

Model monitoring method and communication equipment Technical Field

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

[0002] With the widespread application of models in communication systems, model performance has a significant impact on the communication quality of these systems. For models with good performance, communication based on this model can significantly improve communication quality. For models with poor performance, communication based on this model may lead to a decrease in communication quality or even cause communication failure. Therefore, model monitoring of model performance has become extremely important. Currently, model monitoring mainly relies on continuous model performance monitoring, which may make model monitoring ineffective in practice.

[0003] Summary of the Invention

[0004] The present application provides a model monitoring method and communication device. The following introduces various aspects involved in the present application.

[0005] In a first aspect, a model monitoring method is provided, comprising: a first device performs target model monitoring on a first model, wherein the target model monitoring comprises one of the following: period-based model monitoring; event-triggered model monitoring; model monitoring triggered by a second device indication; and model monitoring based on predicted data, wherein the predicted data is predicted based on measurement data.

[0006] In a second aspect, a model monitoring method is provided, comprising: a second device sends first information to a first device, the first information being used to perform target model monitoring on the first model, wherein the target model monitoring comprises one of the following: period-based model monitoring; event-triggered model monitoring; model monitoring triggered by a second device indication; model monitoring based on predicted data, wherein the predicted data is predicted based on measurement data.

[0007] In a third aspect, a communication device is provided, which is a first device and includes: a processing unit for performing target model monitoring on a first model, wherein the target model monitoring includes one of the following: period-based model monitoring; event-triggered model monitoring; model monitoring triggered by a second device indication; model monitoring based on predicted data, wherein the predicted data is predicted based on measurement data.

[0008] In a fourth aspect, a communication device is provided, which is a second device and includes: a sending unit, sending first information to the first device, wherein the first information is used to perform target model monitoring on the first model, wherein the target model monitoring includes one of the following: period-based model monitoring; event-triggered model monitoring; model monitoring based on second device indication triggering; model monitoring based on predicted data, wherein the predicted data is predicted based on measurement data.

[0009] In a fifth aspect, a communication device is provided, comprising a processor, a memory, and a communication interface, wherein the memory is used to store one or more computer programs, and the processor is used to call the computer program in the memory so that the communication device executes part or all of the steps in the methods of the above aspects.

[0010] In a sixth aspect, an embodiment of the present application provides a communication system, which includes the first device and / or the second device described above. In another possible design, the system may also include other devices that interact with the first device and / or the second device in the solution provided in the embodiment of the present application.

[0011] In the seventh aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and the computer program enables a communication device (for example, a first device or a second device) to perform some or all of the steps in the methods of the above aspects.

[0012] In an eighth aspect, embodiments of the present application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a communication device (e.g., a first device and / or a second device) to perform some or all of the steps of the methods described in each of the above aspects. In some implementations, the computer program product may be a software installation package.

[0013] In a ninth aspect, an embodiment of the present application provides a chip comprising a memory and a processor, wherein the processor can call and run a computer program from the memory to implement some or all of the steps described in the methods of the above aspects.

[0014] In an embodiment of the present application, target model monitoring is introduced to perform model monitoring on the model (also called the first model). Compared with the traditional solution of continuously performing model monitoring on the model, this helps to improve the rationality of the execution of model monitoring in the actual process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] FIG1 is a wireless communication system 100 used in an embodiment of the present application.

[0016] FIG2 is a schematic diagram of channel estimation and signal recovery applicable to an embodiment of the present application.

[0017] FIG3 is a schematic diagram of a channel-state information (CSI) feedback system based on an artificial intelligence (AI) model applicable to an embodiment of the present application.

[0018] FIG4 shows a schematic diagram of an AI model-based positioning solution applicable to an embodiment of the present application.

[0019] Figure 5 shows a schematic diagram of AI model-based beam management applicable to an embodiment of the present application.

[0020] FIG6 is a schematic diagram of a neural network applicable to an embodiment of the present application.

[0021] FIG7 is a schematic diagram of a convolutional neural network (CNN) applicable to an embodiment of the present application.

[0022] FIG8 is a flowchart of a method for model monitoring triggered by a network device applicable to an embodiment of the present application.

[0023] FIG9 is a flowchart of a method for model monitoring triggered by a terminal device applicable to an embodiment of the present application.

[0024] FIG10 is a flowchart of a method for event-triggered model monitoring applicable to an embodiment of the present application.

[0025] FIG11 is a flowchart of a method for performing model monitoring on a terminal side model applicable to an embodiment of the present application.

[0026] FIG12 is a flowchart of a model monitoring method according to an embodiment of the present application.

[0027] FIG13 is a schematic diagram of a cycle-based model monitoring solution in an embodiment of the present application.

[0028] FIG14 is a schematic diagram of a period-based model monitoring solution combined with a first timer in an embodiment of the present application.

[0029] FIG15 is a schematic diagram of a communication device according to an embodiment of the present application.

[0030] FIG16 is a schematic diagram of a communication device according to another embodiment of the present application.

[0031] FIG17 is a schematic structural diagram of a communication device according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] The technical solution in this application will be described below with reference to the accompanying drawings.

[0033] Figure 1 illustrates a wireless communication system 100 used in an embodiment of the present application. The wireless communication system 100 may include a network device 110 and a terminal device 120. The network device 110 may be a device that communicates with the terminal device 120. The network device 110 may provide communication coverage for a specific geographic area and may communicate with the terminal device 120 within the coverage area.

[0034] FIG1 exemplarily shows a network device and two terminals. Optionally, the wireless communication system 100 may include multiple network devices and each network device may include other numbers of terminal devices within its coverage area, which is not limited in the embodiments of the present application.

[0035] Optionally, the wireless communication system 100 may further include other network entities such as a network controller and a mobility management entity, which is not limited in the embodiment of the present application.

[0036] It should be understood that the technical solutions of the embodiments of the present application can be applied to various communication systems, such as: fifth generation (5G) system or new radio (NR), long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD), etc. The technical solutions provided in this application can also be applied to future communication systems, such as the sixth generation mobile communication system, satellite communication system, etc.

[0037] The terminal device in the embodiments of the present application may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station (MS), mobile terminal (MT), remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device. The terminal device in the embodiments of the present application may refer to a device that provides voice and / or data connectivity to a user and can be used to connect people, objects and machines, such as a handheld device with wireless connection function, a vehicle-mounted device, etc. The terminal device in the embodiments of the present application can be a mobile phone, a tablet computer, a laptop computer, a PDA, a mobile internet device (MID), a wearable device, a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical surgery, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, etc. Optionally, the UE can be used to act as a base station. For example, the UE can act as a scheduling entity that provides sidelink signals between UEs in V2X or D2D, etc. For example, a cellular phone and a car communicate with each other using sidelink signals. The cellular phone and smart home devices communicate without relaying the communication signal through the base station.

[0038] The network device in the embodiments of the present application may be a device for communicating with a terminal device, and may also be referred to as an access network device or a radio access network device. For example, the network device may be a base station. The network device in the embodiments of the present application may refer to a radio access network (RAN) node (or device) that connects a terminal device to a wireless network. A base station can broadly cover various names as follows, or be replaced with the following names, such as: NodeB, evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmission point (TRP), transmission point (TP), master station MeNB, secondary station SeNB, multi-standard radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), positioning node, etc. A base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof. A base station can also refer to a communication module, a modem or a chip used to be set in the aforementioned device or apparatus. The base station can also be a mobile switching center and a device that performs base station functions in device-to-device D2D, vehicle-to-everything (V2X), and machine-to-machine (M2M) communications, a network-side device in a 6G network, or a device that performs base station functions in future communication systems. The base station can support networks with the same or different access technologies. The embodiments of this application do not limit the specific technology and specific device form used by the network device.

[0039] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move based on the location of the mobile base station. In other examples, a helicopter or drone can be configured to act as a device that communicates with another base station.

[0040] In some deployments, the network device in the embodiments of the present application may refer to a CU or a DU, or the network device may include a CU and a DU. The gNB may also include an AAU.

[0041] The network equipment and terminal devices can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; they can also be deployed in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the scenarios in which the network equipment and terminal devices are located.

[0042] It should be understood that all or part of the functions of the communication device in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (such as a cloud platform).

[0043] Channel estimation and signal recovery based on AI model

[0044] Due to the complexity and time-varying nature of wireless channel environments, in wireless communication systems (e.g., the wireless communication systems described above), a receiver needs to recover received signals based on channel estimation results. Figure 2 is a schematic diagram of channel estimation and signal recovery applicable to embodiments of the present application.

[0045] As shown in FIG2 , in step S210 , the transmitter transmits, in addition to the data signal, a series of pilot signals known to the receiver on the time-frequency resources, such as the channel state information-reference signal (CSI-RS) and the demodulation reference signal (DMRS).

[0046] In step S211, the transmitter transmits the above data signal and pilot signal to the transmitter through the channel.

[0047] In step S212, after receiving the pilot signal, the receiver may perform channel estimation. In one possible implementation, the receiver may estimate channel information of the channel transmitting the pilot signal based on a pre-stored pilot sequence and the received pilot sequence using a channel estimation algorithm (e.g., a least squares (LS) channel estimation method).

[0048] In step S213, the receiver may recover the channel information on all time-frequency resources using an interpolation algorithm based on the channel information of the channel transmitting the pilot sequence, for subsequent channel state information (CSI) feedback or data recovery.

[0049] CSI feedback system based on AI model

[0050] In wireless communication systems, codebook-based solutions are primarily used to extract and provide feedback on channel characteristics. This means that after the receiver performs channel estimation, it selects the precoding matrix that best matches the current channel from a pre-set precoding codebook based on the estimation results and an optimization criterion. The receiver then feeds the precoding matrix index (PMI) back to the transmitter via an air interface feedback link for precoding. In some implementations, the receiver can also provide the transmitter with a measured channel quality indicator (CQI) to facilitate adaptive modulation and coding.

[0051] Figure 3 is a schematic diagram of a CSI feedback system based on an AI model applicable to an embodiment of the present application. As shown in Figure 3, the entire feedback system includes an AI encoder 311 and an AI decoder 321 part of the autoencoder, wherein the AI ​​encoder 311 is deployed at the transmitter 310 and the AI ​​decoder 321 is deployed at the receiver 320. The transmitter 310 compresses and encodes the CSI to be transmitted through the AI ​​encoder 311 to obtain compressed CSI. The compressed CSI is then fed back to the receiver 320 through the feedback link, and the receiver 320 decodes the compressed CSI through the AI ​​decoder 321 to obtain the recovered CSI. In this way, the communication overhead of feedback CSI can be saved without affecting the accuracy of CSI transmission.

[0052] Positioning based on AI models

[0053] In cellular wireless positioning, the straight-line propagation of electromagnetic waves between network devices and terminal devices is called line-of-sight (LOS) wireless propagation. In some cases, electromagnetic wave signals cannot propagate in a straight line due to obstruction by buildings or trees, which is usually called non-line-of-sight (NLOS) wireless propagation. Traditional positioning algorithms such as time difference of arrival (TDOA) and angle-of-arrival (AOA) are based on LOS channels and are no longer applicable in environments where NLOS is predominant. In most scenarios, the number of network devices with LOS channels to terminal devices is often small, resulting in the inability of traditional positioning algorithms to meet the requirements of high-precision positioning. In addition, there may be some non-ideal factors in actual systems, which can lead to reduced positioning accuracy.

[0054] Therefore, a high-precision positioning method based on AI models has been proposed for scenarios where LOS / NLOS channels coexist. Existing research results have shown that by using machine learning methods to train models based on large amounts of channel data and to explore the mapping relationship between channel responses and location coordinates, it is possible to address the limitations of traditional positioning algorithms in LOS / NLOS channel coexistence scenarios and improve positioning accuracy.

[0055] FIG4 shows a schematic diagram of an AI model-based positioning solution applicable to an embodiment of the present application. Referring to FIG4 , in a positioning solution based on an AI model 410 in a LOS / NLOS channel coexistence scenario, the channel response can be used as the input of the AI ​​model 410, and the location coordinates can be used as the output of the AI ​​model 410. The AI ​​model 410 learns the intrinsic relationship between the wireless channel and the location of the terminal device. In this way, even in a scenario where there are not enough LOS channels and / or in a scenario where there are non-ideal conditions, the positioning solution based on the AI ​​model 410 can also output the location coordinates of the terminal device with higher accuracy, which helps to meet the needs of high-precision positioning.

[0056] The above introduces several communication processes applicable to the AI ​​model. The following introduces the AI ​​model applicable to the embodiments of the present application. It should be noted that the AI ​​model applicable to the embodiments of the present application is not limited to the several AI models introduced below.

[0057] AI-based beam management

[0058] In the traditional beam selection process, it is usually necessary to traverse all combinations of receive beams and transmit beams to select the appropriate beam. However, traversing all combinations takes a long time, resulting in low beam selection efficiency.

[0059] For example, suppose the network equipment deploys 64 different downlink transmission directions in FR2 (carried by up to 64 synchronization signals and physical broadcast channel blocks (SSB)). Accordingly, the terminal device uses one or more antenna panels to simultaneously scan the receiving beams when receiving, and each antenna panel has 4 receiving beams. Then the terminal device needs to measure at least 256 beam pairs, which means that 256 resources of downlink resource overhead are required. From a time perspective, each SSB cycle is approximately 20ms, and 4 SSB cycles are required to complete the measurement of 4 receiving beams. Assuming that multiple receiving antenna panels can perform beam scanning simultaneously, it will take at least 80ms.

[0060] As the number of beams in future massive multiple-input, multiple-output (MIMO) systems increases, using beam scanning-based beam management solutions to match optimal beam pairs will only result in increased reference signal transmission overhead and beam scanning latency. Therefore, to avoid these issues, AI-based beam management was proposed in Release 18. The following describes this AI-based beam management solution, combining the training and prediction processes of the AI ​​model.

[0061] Assume that the AI ​​model is used to predict the available beams in beam set A. Accordingly, during the training phase, the beam measurement results of beam set B can be used as AI model training data. That is, the AI ​​model is trained based on the beam measurement results of beam set B so that the AI ​​model can predict the available beams from beam set A.

[0062] It should be noted that the beam measurement results of the above-mentioned beam set B may include the measurement results corresponding to the layer 1 (layer1, L1) measurement quantity, and / or the indication information of the selected beam in beam set B (for example, the transmitting beam identifier, the receiving beam identifier or the beam pair identifier, etc.).

[0063] In some implementations, the training data may also include label information of beam set A, and the label information is used to indicate one or more of the following beams in beam set A: optimal transmit beam, optimal receive beam, optimal beam pair, better multiple transmit beams, better multiple receive beams, better beam pair, etc.

[0064] As shown in Figure 5, in the prediction stage, the input of the AI ​​model 510 may include the link quality measurement results (for example, L1 measurement quantity) corresponding to the beams in the beam set A, and the prediction results output by the AI ​​model 510 may include the target beam selected from the beam set A, and the link quality corresponding to the target beam.

[0065] In some implementations, the target beam may be one or more beams. For example, if the target beam is a single beam, the target beam may be the optimal beam or a relatively optimal beam in beam set A. For example, if the target beam is multiple beams, the target beam may be multiple beams in beam set A that meet the requirements. "Meeting the requirements" may be understood as meaning that the link quality corresponding to the beam meets the requirements, for example, the link quality corresponding to the beam is greater than or equal to a threshold.

[0066] In other implementations, the target beam may refer to one or more beam pairs, each of which may include a receive beam or a transmit beam. For example, if the target beam is a single beam pair, the target beam may be the optimal beam pair or a relatively optimal beam pair in beam set A. For example, if the target beam is multiple beam pairs, the target beam may be multiple beam pairs in beam set A that meet the requirements. "Meeting the requirements" may be understood as meaning that the link quality corresponding to the beam pair meets the requirements, for example, the link quality corresponding to the beam pair is greater than or equal to a threshold.

[0067] It should be noted that the link quality in the embodiment of the present application can be determined by one or more measurement quantities described above. Of course, the link quality in the embodiment of the present application can also be determined based on other measurement quantities in future communication systems, and the embodiment of the present application is not limited to this.

[0068] In addition, the link quality is determined based on one or more measurement quantities, which can be understood as the link quality being obtained by processing one or more measurement quantities. Of course, the link quality can also be a measurement quantity, which is not limited in the present embodiment.

[0069] It should also be noted that if the prediction result only indicates one beam in the beam pair, the other beam in the beam pair can be determined by other means. For example, it can be determined by one or some of the processes P1 to P3 in the traditional beam selection process. Of course, it can also be determined by one or some of the processes U1 to U3 in the traditional beam selection process. The embodiments of the present application are not limited to this.

[0070] In some implementations, the beam set B may be a different beam set from the beam set A. In some implementations, the beam set B may be a subset of the beam set A. Accordingly, by measuring fewer beams (beams in the beam set B), predictions for more beams (beams in the beam set A) may be achieved. Compared with the above-mentioned scheme of selecting beams based on traversing all combinations, it helps to reduce the time of executing the beam selection process. Of course, in the embodiment of the present application, the beams in the beam set B and the beams in the beam set A may be completely different beams. For example, there is no intersection between the beam set B and the beam set A, but the beam direction corresponding to the beam set B may be similar to the beam direction corresponding to the beam set A.

[0071] In some other implementations, the beam set B may be exactly the same as the beam set A.

[0072] The above describes the communication process applicable to the AI ​​model in conjunction with Figures 2 to 5. The embodiments of the present application are not limited to this. The AI ​​model can also be applied to other communication processes specified in future communication protocols. For example, the AI ​​model can also be used to select the target cell in cell switching. For another example, the AI ​​model can also be used to select the target cell during cell reselection. For another example, the AI ​​model can also be used to predict communication link failure. For another example, the AI ​​model can also be used for communication link recovery. The following describes the AI ​​model applicable to the embodiments of the present application in conjunction with Figures 6 to 7.

[0073] AI models

[0074] In recent years, artificial intelligence research, exemplified by neural networks, has achieved remarkable success in many fields, and will continue to play a vital role in people's lives and production for a long time to come. A neural network can be understood as a computational model consisting of multiple interconnected neuron nodes. The connections between these nodes represent the weighted values ​​from input signals to output signals, often referred to as weights. Each node performs a weighted summation of different input signals and outputs the result through a specific activation function.

[0075] Common neural networks include CNN, recurrent neural network (RNN), deep neural network (DNN), etc.

[0076] The following describes a neural network applicable to embodiments of the present application in conjunction with FIG6 . The neural network shown in FIG6 can be divided into three categories based on the location of different layers: input layer 610, hidden layer 620, and output layer 630. Generally speaking, the first layer is the input layer 610, the last layer is the output layer 630, and the intermediate layers between the first and last layers are all hidden layers 620.

[0077] The input layer 610 is used to input data, where the input data can be, for example, a received signal received by a receiver. The hidden layer 620 is used to process the input data, for example, decompress the received signal. The output layer 630 is used to output processed output data, for example, a decompressed signal.

[0078] As shown in Figure 6, a neural network consists of multiple layers, each of which contains multiple neurons. The neurons between layers can be fully connected or partially connected. For connected neurons, the output of the neurons in the previous layer can serve as the input of the neurons in the next layer.

[0079] With the continuous advancement of neural network research, deep learning algorithms have been proposed in recent years. These algorithms introduce a large number of hidden layers into neural networks, forming DNNs. More hidden layers allow DNNs to better capture complex real-world situations. Theoretically, a model with more parameters has higher complexity and a greater "capacity," meaning it can handle more complex learning tasks. These neural network models are widely used in pattern recognition, signal processing, optimization and combination, anomaly detection, and other fields.

[0080] CNN is a deep neural network with a convolutional structure, and its structure is shown in FIG7 , which may include an input layer 710 , a convolutional layer 720 , a pooling layer 730 , a fully connected layer 740 , and an output layer 750 .

[0081] Each convolution layer 720 may include a plurality of convolution operators, which are also called kernels. The convolution operator can be regarded as a filter for extracting specific information from the input signal. The convolution operator can essentially be a weight matrix, which is usually predefined.

[0082] The weight values ​​in these weight matrices need to be obtained through a lot of training in practical applications. The weight matrices formed by the weight values ​​obtained through training can extract information from the input signal, thereby helping CNN to make correct predictions.

[0083] When CNN has multiple convolutional layers, the initial convolutional layer tends to extract more general features, which can also be called low-level features. As the depth of CNN increases, the features extracted by the subsequent convolutional layers become more and more complex.

[0084] Pooling layers 730 are often used periodically after convolutional layers to reduce the number of training parameters. For example, a single convolutional layer can be followed by a pooling layer, as shown in Figure 7, or multiple convolutional layers can be followed by one or more pooling layers. In signal processing, the sole purpose of a pooling layer is to reduce the spatial size of the extracted information.

[0085] The fully connected layer 740, after being processed by the convolution layer 720 and the pooling layer 730, is not sufficient for CNN to output the required output information. Because as mentioned above, the convolution layer 720 and the pooling layer 730 only extract features and reduce the parameters brought by the input data. However, in order to generate the final output information (for example, the bit stream of the original information transmitted by the transmitter), CNN also needs to use the fully connected layer 740. Generally, the fully connected layer 740 may include multiple hidden layers, and the parameters contained in the multiple hidden layers can be pre-trained based on the relevant training data of the specific task type. For example, the task type may include decoding the data signal received by the receiver. For example, the task type may also include channel estimation based on the pilot signal received by the receiver.

[0086] Following the multiple hidden layers in the fully connected layer 740, the final layer of the CNN is the output layer 750, which is used to output the results. Typically, this output layer 750 is configured with a loss function (e.g., a loss function similar to categorical cross entropy) to calculate the prediction error, or to evaluate the degree of difference between the output of the CNN model (also known as the predicted value) and the ideal result (also known as the true value).

[0087] To minimize the loss function, the CNN model needs to be trained. In some implementations, the CNN model can be trained using the backpropagation algorithm (BP). The BP training process consists of a forward propagation process and a backward propagation process. During the forward propagation process (e.g., the propagation from 710 to 750 in Figure 7 is forward propagation), the input data is fed into the aforementioned layers of the CNN model, processed layer by layer, and transmitted to the output layer. If the output result of the output layer differs significantly from the ideal result, the aforementioned loss function is minimized as the optimization goal, and the backward propagation process is switched to (e.g., the propagation from 750 to 710 in Figure 7 is backward propagation). The partial derivatives of the optimization goal with respect to each neuron weight are calculated layer by layer, forming the gradient of the optimization goal with respect to the weight vector, which serves as the basis for modifying the model weights. The CNN training process is completed during the weight modification process. When the aforementioned error reaches the desired value, the CNN training process ends.

[0088] It should be noted that the CNN shown in Figure 7 is only an example of a convolutional neural network. In specific applications, the convolutional neural network can also exist in the form of other network models, and the embodiments of the present application are not limited to this.

[0089] RNNs are designed to process sequential data. In traditional neural network models (for example, CNN models), the layers are fully connected, from the input layer to the hidden layer to the output layer, and the nodes within each layer are disconnected. However, these ordinary neural networks are inadequate for many problems. For example, if you want to predict the next word in a sentence, you generally need to use the previous word, because the previous and next words in a sentence are not independent. RNNs are called recurrent neural networks because the current output of a sequence is also related to the previous output. Specifically, the network remembers the previous information and applies it to the calculation of the current output. That is, the nodes between hidden layers are no longer disconnected but connected, and the input of the hidden layer includes not only the output of the input layer but also the output of the hidden layer at the previous moment. In theory, RNNs can process sequence data of any length.

[0090] Training an RNN is similar to training a traditional ANN (artificial neural network). The same backpropagation error algorithm is used, but there is a slight difference. If the RNN is expanded, the parameters W, U, and V are shared, while traditional neural networks are not. Furthermore, when using the gradient descent algorithm, the output of each step depends not only on the network state at the current step, but also on the state of the network at the previous steps. For example, at t = 4, three steps need to be propagated backward, and various gradients need to be added to the three subsequent steps. This learning algorithm is called backpropagation through time (BPTT).

[0091] Given the existence of artificial neural networks and convolutional neural networks, why do we still need recurrent neural networks? The reason is simple. Both convolutional and artificial neural networks assume that elements are independent of each other, and that inputs and outputs are also independent, like cats and dogs. However, in the real world, many elements are interconnected, such as the changes in stock prices over time. For example, someone said, "I love traveling, and my favorite place is Yunnan. I must visit __ someday." Everyone knows to fill in the blank with "Yunnan." This is because we infer this information based on the context, but achieving this is quite difficult. Therefore, recurrent neural networks were developed. Their essence is that they possess memory, just like humans. Therefore, their output depends on the current input and memory.

[0092] Model monitoring

[0093] Model monitoring, also known as "model supervision," is used to monitor the model's inference process, or in other words, to monitor the model's model inference performance. Model monitoring can be used to test indicators such as model accuracy, stability, robustness, and efficiency to ensure the model's reliability and effectiveness in practical applications. Currently, the model monitoring process mainly serves processes such as model activation, model selection, model switching, model rollback, and model update (including retraining). Model rollback can be understood as falling back from model-based communication to traditional communication methods (i.e., non-model-based communication methods).

[0094] In some implementations, the model can be deployed in multiple ways. For example, the model can be deployed on the UE side, in which case the model is also called a "UE-side model" or a "terminal-side model." For another example, the model can be deployed on the network side, in which case the model is also called a "network-side model." For another example, the model can be deployed on devices at both ends of a communication link (e.g., a terminal device and a network device, or a terminal device and a terminal device), in which case the model is also called a "two-sided model."

[0095] In some implementations, different model monitoring triggering methods may correspond to models with different deployment methods. The following describes the model monitoring triggering methods corresponding to different deployment methods in conjunction with Figures 8 to 10. It should be understood that in the embodiments of the present application, the correspondence between the model deployment method and the model monitoring triggering method is not limited. The following is merely an example.

[0096] Figure 8 is a flow chart of a method for network device-triggered model monitoring applicable to an embodiment of the present application. Assuming that the model to be monitored is a network-side model, the method shown in Figure 8 includes steps S810 to S840.

[0097] In step S810, the network device triggers model monitoring of the network side model.

[0098] In step S820, the network device sends a performance feedback request to the terminal device to request the terminal device to feed back the measured performance data of the network side model.

[0099] In step S830, the terminal device sends a performance feedback report to the network device to carry the performance data of the network side model.

[0100] In step S840, the network device performs model management on the network side model based on the performance data of the network side model.

[0101] In some implementations, model management may include one or more of the following: controlling the network side model to be in an activation state (activation); controlling the network side model to be in a deactivation state (deactivation); controlling the network side model to perform model rollback; performing a model selection process; and performing a model switching process.

[0102] In step S850, the network device sends indication information to the terminal device, where the indication information is used to indicate the model management result of the model management.

[0103] Taking the example of model management including controlling the network-side model to be in an activated state, the model management result includes whether the network-side model is in an activated state. Taking the example of model management including controlling the network-side model to be in an inactivated state, the model management result includes whether the network-side model is in an inactivated state. Taking the example of model management including controlling the network-side model to execute model rollback, the model management result includes whether the network-side model successfully executes model rollback. Taking the example of model management including executing the model selection process, the model management result includes the target model selected by the model selection process. Taking the example of model management including executing the model switching process, the model management result includes the target model switched by the model switching process.

[0104] FIG9 is a flow chart of a method for model monitoring triggered by a terminal device applicable to an embodiment of the present application. Assuming that the model to be monitored is a terminal-side model, the method shown in FIG9 includes steps S910 to S940.

[0105] In step S910, the terminal device triggers model monitoring of the terminal side model.

[0106] In step S920, the terminal device performs model management on the terminal side model based on the model monitoring result of the terminal side model.

[0107] In some implementations, model management includes one of the following: controlling the network side model to be in an activated state; controlling the network side model to be in a deactivated state; controlling the network side model to perform model rollback; performing a model selection process; and performing a model switching process.

[0108] In step S930 , the terminal device sends a request message to the network device to request a model management result allowing model management.

[0109] Taking the example of model management including controlling the network-side model to be in an activated state, the request message is used to request permission for the network-side model to be in an activated state. Taking the example of model management including controlling the network-side model to be in an inactivated state, the request message is used to request permission for the network-side model to be in an inactivated state. Taking the example of model management including controlling the network-side model to perform model rollback, the request message is used to request permission for the network-side model to perform model rollback. Taking the example of model management including performing a model selection process, the request message is used to request permission to use the target model selected by the model selection process. Taking the example of model management including performing a model switching process, the request message is used to request permission to use the target model switched by the model switching process.

[0110] In step S940, the network device sends a response message to the terminal device in response to the request message. The response message is used to indicate whether the model management result is allowed. For example, if the response message is ACK, it indicates that the model management result is allowed. If the response message is NACK, it indicates that the model management result is not allowed.

[0111] Figure 10 is a flow chart of a method for event-triggered model monitoring applicable to an embodiment of the present application. Assuming that the model to be monitored is a network-side model, the method shown in Figure 10 includes steps S1010 to S1030.

[0112] In step S1010 , the network device sends configuration information to the terminal device, where the configuration information is used to configure an event for triggering model management.

[0113] In some implementations, model management includes one of the following: controlling the network side model to be in an activated state; controlling the network side model to be in a deactivated state; controlling the network side model to perform model rollback; performing a model selection process; and performing a model switching process.

[0114] In step S1020, the terminal device triggers model monitoring of the terminal side model.

[0115] In step S1030, the terminal device performs model management on the terminal side model based on the model monitoring result of the terminal side model.

[0116] In some scenarios, the device performing model monitoring can indicate the performance of the model and / or the model management results by sending a model inference result report. Figure 11 is a flowchart of a method for performing model monitoring on a terminal-side model applicable to an embodiment of the present application. The method shown in Figure 11 includes steps S1110 to S1130.

[0117] In step S1110, the terminal device triggers model monitoring of the terminal side model.

[0118] In step S1120, the terminal device performs model management on the terminal side model based on the model monitoring result of the terminal side model.

[0119] In some implementations, model management includes one or more of the following: controlling the network side model to be in an activated state; controlling the network side model to be in a deactivated state; controlling the network side model to perform model rollback; performing a model selection process; and performing a model switching process.

[0120] In step S1130, the terminal device sends a model inference result report to the network device.

[0121] In some implementations, model monitoring can be performed on the model to obtain model monitoring information (also known as model monitoring results). Generally, model monitoring information can be divided into two categories. The first category of model monitoring information is information related to the wireless communication system, such as throughput and bit error rate. The second category of model monitoring information is information related to model performance, such as data offset, confidence level, and confidence probability.

[0122] In other implementations, model monitoring may be performed based on a model identification of the model. For example, the model to be monitored by the model monitoring may be determined based on the model identification.

[0123] With the widespread application of models in communication systems, model performance has a significant impact on the communication quality of these systems. For models with good performance, communication based on this model can significantly improve communication quality. For models with poor performance, communication based on this model may lead to a decrease in communication quality or even cause communication failure. Therefore, model monitoring of model performance has become extremely important. Currently, model monitoring mainly relies on continuous model performance monitoring, which may make model monitoring ineffective in practice.

[0124] For example, the communication quality based on the first model is relatively stable. In this case, if the first device continuously monitors the first model, it is unnecessary and may also increase the power consumption of the first device.

[0125] For another example, in order to save energy consumption of the first device, the first device may periodically enter a silent period of radio resource management (RRM) measurement. During the silent period, the first device does not perform RRM measurement, and RRM measurement is the basis of model monitoring, that is, model monitoring needs to be based on the measurement results of RRM. If the first device enters the silent period of RRM measurement, continuous model monitoring cannot be performed on the model. If the first device still performs RRM measurement after entering the silent period of RRM measurement, the power consumption of the first device will increase significantly.

[0126] Therefore, to address the above issues, an embodiment of the present application provides a model monitoring method. In this method, by introducing target model monitoring, a model (also called a first model) is monitored, which helps to improve the rationality of model monitoring performed in actual processes.

[0127] In the embodiments of the present application, the first model is not limited. In some implementations, the first model may be an AI model. For example, the first model may be any of the AI ​​models described above, or may be a new AI model introduced in the future. In other implementations, the first model may be a machine learning (ML) model.

[0128] The following describes a model monitoring method according to an embodiment of the present application in conjunction with Figure 12. The method shown in Figure 12 includes step S1210. In step S1210, the first device performs target model monitoring on the first model.

[0129] In some implementations, target model monitoring may include a discontinuous model monitoring method, where discontinuous can be understood as discontinuous monitoring in time. That is, target model monitoring may include model monitoring that is discontinuous in time. For example, target model monitoring may include periodic model monitoring. For another example, target model monitoring may include event-triggered model monitoring. For another example, target model monitoring may include model monitoring triggered by a second device indication.

[0130] In an embodiment of the present application, the introduction of a discontinuous model monitoring method is helpful to reduce the power consumption required for model monitoring compared to the continuous model monitoring method. Taking the cell switching scenario as an example, after the terminal device switches, it can be determined whether the model-based communication method is improved relative to the traditional communication method (i.e., the non-model-based communication method). At this time, if a continuous model monitoring method is adopted, the terminal device needs to continuously perform model monitoring for a period of time after the switch. For example, the terminal device continuously measures the signal during this time period, which means that in addition to monitoring the reference signal receiving power (RSRP) of the current serving cell, the terminal device also needs to continuously measure the RSRP of the cell determined based on the traditional communication method, resulting in a large power consumption of the terminal device. If a discontinuous model monitoring method is adopted, the terminal device only needs to perform discontinuous model monitoring, for example, the terminal device performs discontinuous measurements during this time period, which helps to reduce the power consumption of the terminal device.

[0131] In other implementations, the aforementioned discontinuity may be understood as discontinuous measurement in time. That is, target model monitoring may include model monitoring with discontinuous measurement in time, or in other words, the measurement of the signal during the model monitoring period is discontinuous. For example, target model monitoring may include model monitoring based on predicted data. In this type of model monitoring solution, the first device may not perform RRM measurements for a period of time, but instead perform model monitoring based on predicted data. In this case, RRM measurements may be performed discontinuously while implementing continuous model monitoring.

[0132] In an embodiment of the present application, the introduction of a discontinuous measurement model monitoring method helps to reduce the power consumption required for model monitoring compared to the continuous measurement model monitoring method. Taking the RRM measurement silence mechanism adopted by the terminal device as an example, the terminal device usually does not perform signal measurement during the RRM measurement silence period to reduce the power consumption of the terminal device. In this scenario, if the terminal device adopts a continuous model monitoring method, it goes against the original design intention of the RRM measurement silence mechanism, making the introduction of the RRM measurement silence mechanism meaningless. If the discontinuous measurement model monitoring method introduced above is adopted, the terminal device can predict the model performance during the silence period based on the prediction data. Since the prediction data is obtained based on the measurement data, at this time, the terminal device does not need to perform RRM measurement during the silence period, which matches the RRM measurement silence mechanism and helps to reduce the power consumption of the terminal device.

[0133] For ease of understanding, the following describes the implementation methods of target model monitoring in the embodiments of the present application in conjunction with Examples 1 to 4. It should be noted that the embodiments of the present application are not limited to the several model monitoring methods described below. The target model monitoring methods in the embodiments of the present application may also include other non-continuous measurement model monitoring methods and / or non-continuous monitoring model monitoring methods.

[0134] Example 1: Cycle-based model monitoring.

[0135] In some implementations, the above-mentioned periodic model monitoring can be understood as model monitoring that is performed periodically. Therefore, this period can also be called a model monitoring interval period.

[0136] In some implementations, the cycle-based model monitoring may be performed based on a target parameter, where the target parameter includes a first parameter and / or a second parameter.

[0137] Taking the example where the target parameter includes the first parameter, the first parameter is used to determine the cycle length of the cycle, or in other words, the first parameter is used to determine the duration of the cycle. The first parameter of the embodiment of the present application is described below in conjunction with Example 1 and Example 2.

[0138] In Example 1, the first parameter is used to indicate a first time interval between two adjacent cycles, wherein the two adjacent cycles may refer to two cycles that are adjacent in time.

[0139] In the embodiment of the present application, there is no limitation on the method for determining the first time interval. In some implementations, the first time interval may be the time interval between the start times of two adjacent cycles. Referring to FIG13 , two adjacent cycles may include cycle 1 and cycle 2, and the first time interval may be the time interval between the start time of cycle 1 and the start time of cycle 2. In other implementations, the first time interval may be the time interval between the end times of two adjacent cycles. Referring to FIG13 , two adjacent cycles may include cycle 1 and cycle 2, and the first time interval may be the time interval between the end time of cycle 1 and the end time of cycle 2. Of course, in the embodiment of the present application, the first time interval may also be the time interval between the middle moments of two adjacent cycles.

[0140] In an embodiment of the present application, the first time interval may belong to a time interval set, which may include one or more candidate values ​​for the first time interval. In this case, a suitable candidate value may be selected from the time interval set as the first time interval. Taking a network device maintenance time interval set as an example, the network device may select a suitable candidate value from the time interval set as the first time interval and send it to a terminal device (as an example of the first device).

[0141] In some scenarios, the network device can configure a first time interval for multiple first devices (e.g., terminal devices), and the first time intervals configured for the multiple terminal devices can all be from a time interval set. Alternatively, some of the first time intervals configured for the multiple terminal devices can be from a time interval set. Of course, in an embodiment of the present application, the first time intervals configured for the multiple terminal devices can also be directly determined by the network device, rather than selected from the time interval set.

[0142] In the embodiment of the present application, the representation of the time interval set is not limited. For example, the time interval set can be represented in the form of a list. In addition, in the embodiment of the present application, the above time interval set can be predefined and / or preconfigured.

[0143] Assume that the time interval set pre-stored in the second device is {20ms, 40ms, 60ms}. In this case, the second device may indicate to the first device that the value of the first time interval is 40ms.

[0144] In some other implementations, two adjacent periods (or, two periods adjacent in time) include a first period and a second period, and the first time interval is the time interval between a first duration during which model monitoring can be performed in the first period and a first duration during which model monitoring can be performed in the second period. For ease of description, the first duration during which model monitoring can be performed in the first period is referred to as "Duration 1," and the first duration during which model monitoring can be performed in the second period is referred to as "Duration 2."

[0145] It should be noted that the first duration is a time period in which model monitoring can be performed, that is, model monitoring can be performed within the first duration period. However, the duration of performing model monitoring may only occupy part of the duration of the first duration period. For example, the duration of the first duration period is 10ms, and accordingly, the duration of model monitoring performed within the first duration period may be 8ms. Of course, in an embodiment of the present application, the duration of performing model monitoring may occupy the entire duration of the first duration period. For example, the duration of the first duration period is 10ms, and accordingly, the duration of model monitoring performed within the first duration period may be 10ms.

[0146] In the embodiment of the present application, the method for determining the length of the first duration is not limited. In some implementations, the length of the first duration can be determined based on the length of time required for the first model to perform the model reasoning process. For example, the length of the first duration can be greater than or equal to the length of time required for the first model to perform the model reasoning process. In other implementations, the length of the first duration can be determined based on the preparation time required for the first model to prepare to enter the model reasoning process. For example, the length of the first duration can be greater than or equal to the preparation time required for the first model to prepare to enter the model reasoning process.

[0147] In the embodiments of the present application, the method for determining the first time interval is not limited. In some implementations, the first time interval may be the time interval between the start time of duration 1 and the start time of duration 2. Referring to FIG13 , two adjacent cycles may include cycle 1 and cycle 2, wherein the duration of model monitoring in cycle 1 is duration 1, and the duration of model monitoring in cycle 2 is duration 2. In this case, the first time interval may be the time interval between the start time of duration 1 and the start time of duration 2.

[0148] In some other implementations, the first time interval may be the time interval between the end time of Duration 1 and the end time of Duration 2. Continuing with FIG13 , the first time interval may be the time interval between the end time of Duration 1 and the end time of Duration 2. Of course, in the embodiment of the present application, the first time interval may also be the time interval between the middle moment of Duration 1 and the middle moment of Duration 2.

[0149] In an embodiment of the present application, the first time interval may belong to a time interval set, which may include one or more candidate values ​​for the first time interval. In this case, a suitable candidate value may be selected from the time interval set as the first time interval. Taking a network device maintenance time interval set as an example, the network device may select a suitable candidate value from the time interval set as the first time interval and send it to the first device (as an example of the first device).

[0150] In some scenarios, the network device can configure a first time interval for multiple first devices (e.g., terminal devices), and the first time intervals configured for the multiple terminal devices can all be from a time interval set. Alternatively, some of the first time intervals configured for the multiple terminal devices can be from a time interval set. Of course, in an embodiment of the present application, the first time intervals configured for the multiple terminal devices can also be directly determined by the network device, rather than selected from the time interval set.

[0151] In the embodiment of the present application, the representation of the time interval set is not limited. For example, the time interval set can be represented in the form of a list. In addition, in the embodiment of the present application, the above time interval set can be predefined and / or preconfigured.

[0152] Assume that the time interval set pre-stored in the second device is {20ms, 40ms, 60ms}. In this case, the second device may indicate to the first device that the value of the first time interval is 60ms.

[0153] In Example 2, the first parameter is used to indicate a basic time interval, and the basic time interval is used to determine the first time interval.

[0154] In the embodiments of the present application, there is no limitation on the method for determining the first time interval. For example, the first time interval may be an integer multiple of the base time interval, or in other words, the first time interval may be N times the base time interval, where N is a positive integer. For another example, the first time interval may be obtained by adjusting (e.g., increasing or decreasing) the time offset value with the base time interval as the initial value.

[0155] In an embodiment of the present application, the parameter N may belong to a parameter set, which may include one or more candidate values ​​for the parameter N. In this case, a suitable candidate value may be selected from the parameter set as the parameter N to determine the first time interval. Taking a network device maintenance parameter set as an example, the network device may select a suitable candidate value from the parameter set as the parameter N and send it to the first device, so that the first device determines the first time interval based on the parameter N and the base time interval.

[0156] In some scenarios, the network device can configure parameters N for multiple first devices (e.g., terminal devices), and the parameters N configured for the multiple terminal devices can all come from a parameter set. Alternatively, some of the parameters N configured for the multiple terminal devices can come from a parameter set. Of course, in the embodiment of the present application, the parameters N configured for the multiple terminal devices can also be directly determined by the network device, rather than selected from a parameter set.

[0157] In the embodiments of the present application, the representation of the parameter set is not limited. For example, the parameter set may be presented in the form of a list. In addition, in the embodiments of the present application, the parameter set may be predefined and / or preconfigured. In addition, in the embodiments of the present application, the basic time interval may be determined in one or more of the following ways: predefined, preconfigured, or configured by a network device.

[0158] Assume that the base time interval 1 is 10 ms and the second device stores a parameter set of {1, 2, ... 31}. The second device can indicate to the first device that the value of parameter N is 2. Accordingly, the first device can determine the value of the first time interval T to be T = 2 × N = 20 ms based on parameter N and the base time interval.

[0159] Taking the example where the target parameter includes the second parameter, in some implementations, the second parameter is used to determine the first duration. The first duration can be described above and will not be described here for brevity.

[0160] In some implementations, the second parameter is used to indicate the length of the first duration and / or the start time of the first duration. Of course, in embodiments of the present application, the second parameter can be used to indicate other parameters used to determine the first duration, for example, the second parameter can be used to indicate the end time of the first duration.

[0161] Assume that the second parameter indicates that the duration of the first duration is duration 1, the start time of the first duration is moment 1, and the first duration is a period of time starting at moment 1 and lasting for duration 1.

[0162] Assume that the second parameter indicates that the duration of the first duration is duration 2, and the end time of the first duration is time 1, the first duration is a period of time starting from time 1 and lasting for duration 2.

[0163] It should be noted that, as described above, the duration of the first duration may be indicated by the second parameter. Of course, in the embodiment of the present application, the duration of the first duration may also be determined based on one or more of the following methods: a predefined method; a preconfigured method; or a method based on network device configuration.

[0164] In some implementations, if the second parameter is used to indicate the start time of the first duration, the second parameter includes a time offset between the start time of the first duration and the start time of the period to which the first duration belongs. That is, the start time of the period is used as the starting position, and the time offset is offset to the start time of the first duration. This method of setting the first duration helps to improve the flexibility of the first duration.

[0165] Continuing to refer to FIG. 13 , assuming that the time offset included in the second parameter is Δt, for period 1, the start time of period 1 is taken as the initial position, and the time after the offset time is shifted by Δt is the start time of duration 1.

[0166] In some implementations, the time offset can be determined based on the inference response time of the first model. The time offset can be greater than or equal to the inference response time of the first model, where the inference response time indicates the time required for the first model to respond to input data and enter the inference process. This approach to setting the time offset value helps reserve sufficient time for the first model to enter the model inference phase, so that when the first duration arrives, model monitoring can be performed on the first model within the first duration.

[0167] Of course, in the embodiment of the present application, the start time of the first duration may overlap with the start time of the period, or in other words, the start time of the first duration may be the same as the start time of the period. In other words, the time offset value may be 0, or the second parameter may no longer indicate the time offset value.

[0168] In addition, in the embodiment of the present application, the time domain position of the first duration within the cycle is not limited. For example, the end time of the first duration can be the end time of the cycle. For another example, the first duration can overlap with the cycle, that is, the start time of the first duration can be the start time of the cycle, and the end time of the first duration can be the end time of the cycle. In this case, the meaning of the cycle is similar to the first duration, that is, all time within the cycle can be used to perform model monitoring, or, in this case, the first duration can no longer be introduced additionally.

[0169] In this embodiment of the present application, the second parameter may include one or more time offset values. If the second parameter includes one time offset value, then the corresponding period may include only one first duration. If the second parameter includes multiple time offset values, then the corresponding period may include multiple first durations.

[0170] For example, the duration of the cycle is 60ms, the duration of the first duration is 5ms, the start time of the first segment of the first duration is the same as the start time of the cycle, and the time offset contained in the second parameter is {10ms, 40ms}. Then, one cycle can include two segments of the first duration: 10ms~15ms and 40ms~45ms, and model monitoring can be performed for 10ms in each of these two durations.

[0171] In the embodiment of the present application, the method for determining the first duration is not limited. In some implementations, the second parameter can be used to indicate the time interval between multiple first durations included in the cycle, and the number of first durations in the cycle. For example, assuming that the duration of the cycle is 60ms, the duration of the first duration is 5ms, the start time of the first first duration is the same as the start time of the cycle, the time interval between the multiple first durations included in the second parameter is 10ms, and the second parameter indicates that the number of first durations included in a cycle is 2, then two first durations can be included in a cycle: 0ms~5ms and 15ms~20ms.

[0172] In some implementations, the first device may include multiple models, and the multiple models may include the first model. In this case, the target parameters associated with some or all of the multiple models may be different, that is, the cycles for model monitoring associated with some or all of the multiple models are different, which helps to improve the rationality of model monitoring performed on different models. Of course, in an embodiment of the present application, the target parameters associated with multiple models may be the same, that is, the cycles for model monitoring associated with multiple models are the same, so as to reduce the complexity of performing model monitoring on multiple models.

[0173] In the embodiments of the present application, the association relationship between the model and the target parameter is not limited. In some implementations, models for different functions can be associated with different target parameters. For example, model 1 is used for CSI encoding and model 2 is used for beam management. In this case, model 1 and model 2 can be associated with different target parameters. In other implementations, models for the same function can be associated with the same target parameter. For example, model 1 and model 2 are both used for beam management. In this case, model 1 and model 2 can be associated with the same target parameter. In other implementations, models for different functions can be associated with the same target parameter. For example, model 1 is used for CSI encoding and model 2 is used for beam management. In this case, model 1 and model 2 can be associated with the same target parameter. For another example, models for the same function can be associated with different target parameters. For example, model 1 and model 2 are both used for beam management. In this case, model 1 and model 2 can be associated with different target parameters.

[0174] In some implementations, the first device may request the second device to adjust the target parameters of the first model, thereby improving the rationality of the target parameters. Specifically, the method further includes: the first device sending a first request to the second device, the first request being for adjusting the target parameters associated with the first model; and the first device receiving a response message from the second device to the first request.

[0175] In some implementations, the first request may include one or more of the following parameters: a model identifier of the first model; and information indicating that the request is for adjusting a target parameter associated with the first model.

[0176] In some implementations, the first request may include requirement information for performing target model monitoring on the first model, so that the second device determines the adjusted target parameters based on the requirement information. For example, the requirement information may indicate the inference response time of the first model, wherein the inference response time is used to indicate the time required for the first model to respond to input data and enter the inference process. For another example, the requirement information may indicate the inference time of the first model, wherein the inference time is used to indicate the time required for the first model to perform the inference process. In an embodiment of the present application, for a scheme for determining the first duration based on the inference response time and / or the inference time, please refer to the above introduction to the first duration. For the sake of brevity, it will not be repeated here.

[0177] In other implementations, the first device may directly send the proposed candidate value of the target parameter to the second device, so that the second device can adjust the target parameter based on the candidate value of the target parameter. In other words, the first request may include the candidate value of the target parameter proposed by the first device.

[0178] In some implementations, the first request may include one or more of the following parameters: a model identifier of the first model; and information of candidate values ​​for the target parameter.

[0179] In the embodiment of the present application, the first request is not limited. Taking the first device as a UE as an example, the first request can be carried in UE assistance information (UEAssistanceInformation).

[0180] In the embodiment of the present application, the first request may include a candidate value of the target parameter, which helps reduce the overhead required to transmit the first request. Of course, in the embodiment of the present application, the first request may include multiple candidate values ​​of the target parameter for the second device to select, which helps improve the rationality of the target parameter selection.

[0181] In some implementations, the candidate value in the above-mentioned first request may be generated by the first device. In the embodiment of the present application, the manner in which the first device determines the candidate value is not limited. Taking the target parameter as the second parameter for determining the duration of the first duration as an example, in some implementations, the second parameter may be determined based on the duration required for the first model to perform the model reasoning process. For example, the duration of the first duration indicated by the second parameter may be greater than or equal to the duration required for the first model to perform the model reasoning process. In other implementations, the second parameter may be determined based on the preparation time required for the first model to prepare to enter the model reasoning process. For example, the duration of the first duration indicated by the second parameter may be greater than or equal to the preparation time required for the first model to prepare to enter the model reasoning process. Taking the target parameter as the first parameter for determining the cycle duration as an example, in some implementations, the first parameter may be determined based on the duration required for the first model to perform the model reasoning process. For example, the cycle duration indicated by the first parameter may be greater than or equal to the duration required for the first model to perform the model reasoning process. In other implementations, the first parameter may be determined based on the duration required for the first model to prepare to enter the model reasoning process. For example, the cycle duration indicated by the first parameter may be greater than or equal to the duration required for the first model to prepare to enter the model inference process.

[0182] In some implementations, the response message includes the adjusted target parameter. In other words, the second device may indicate the adjusted target parameter to the first device via the response message.

[0183] In an embodiment of the present application, there is no limitation on the manner in which the response message indicates the target parameter. The target parameter can be directly carried in the response message described above. In other implementations, assuming that the first request carries multiple candidate values ​​for the target parameter, the response message can also indicate the target parameter by carrying the arrangement sequence number of the candidate value in the first request. For example, the candidate values ​​of the target parameter carried by the first request are {10ms, 20ms, 30ms}, wherein the arrangement sequence number corresponding to the candidate value 10ms is 1, the arrangement sequence number corresponding to the candidate value 20ms is 2, and the arrangement sequence number corresponding to the candidate value 30ms is 3. At this time, the response message can indicate that the target parameter is 20ms by carrying the arrangement sequence number 2, or the response message can indicate that the target parameter is 10ms by carrying the arrangement sequence number 1, or the response message can indicate that the target parameter is 30ms by carrying the arrangement sequence number 3.

[0184] In some other implementations, the response message may include information indicating whether the target parameter is allowed to be adjusted based on the first request. The embodiments of the present application do not limit the response message. For example, the response message may be ACK / NACK, where ACK is used to indicate that the target parameter is allowed to be adjusted based on the first request, and NACK is used to indicate that the target parameter is not allowed (or rejected) to be adjusted based on the first request.

[0185] In an embodiment of the present application, any of the first requests described above may be used in conjunction with a response message. In some implementations, the first request may include candidate values ​​for the target parameter, and accordingly, the response message may include information indicating whether adjustment of the target parameter based on the first request is permitted.

[0186] For example, the target parameters include a second parameter, and the value of the second parameter before adjustment is 10 ms. The first device sends a first request to the second device, and the first request carries a candidate value of 20 ms for the second parameter. Accordingly, in response to receiving the first request, the second device sends a response message to the first device, where the response message is used to indicate whether adjustment of the second parameter based on the candidate value of the second parameter is permitted. If the response message indicates that adjustment of the second parameter based on the candidate value of the second parameter is permitted, the value of the second parameter is adjusted to 20 ms. If the response message indicates that adjustment of the second parameter based on the candidate value of the second parameter is not permitted, the value of the second parameter is the value before adjustment, i.e., 10 ms.

[0187] Of course, in an embodiment of the present application, the first request may include information about the requirement to perform target model monitoring on the first model, and accordingly, the response message may include the target parameters. Alternatively, the first request may include information about the requirement to perform target model monitoring on the first model, as well as candidate values ​​for the target parameters. Accordingly, the response message may include the target parameters and information indicating whether adjustment of the target parameters based on the first request is permitted.

[0188] Example 2: Model monitoring based on event triggering.

[0189] In some implementations, the target model monitoring is event-triggered model monitoring. Therefore, the event used to trigger the model monitoring may also be referred to as a “triggering event”.

[0190] In some implementations, the above event is associated with one or more of the following: the service type of the service to be transmitted; the model state of the first model; the device state of the first device; the wireless communication environment of the first device; and the success of the model switching operation to switch to the first model.

[0191] Taking the association of events with the business type of the business to be transmitted as an example, in some implementations, the above-mentioned events may include a change in the business type of the business to be transmitted. That is to say, if the business type of the business to be transmitted changes, target model monitoring may be triggered. Generally, if the business type changes, the input and output information of the first model may be affected, which may cause the model performance of the first model to degrade. At this time, executing target model monitoring helps to promptly discover the problem of mismatch between the first model and the business type. Of course, in an embodiment of the present application, the above-mentioned event may also include that the business type of the business to be transmitted is a target business type. The target business type may, for example, be a business type that does not match the first model in the historical communication process. The embodiment of the present application does not limit the specific form of the event.

[0192] For example, when the service type of the service to be transmitted on the first device changes from the File Transfer Protocol (FTP) service type to the Hypertext Transfer Protocol (HTTP) service type, the probability of packet loss during the service data transmission process increases, resulting in a large number of zero values ​​or missing values ​​in the data set (sparseness is generally used to describe the presence of a large number of zero values ​​or missing values ​​in a data set). Therefore, if the service data is received based on the first model, the first model is required to be able to make predictions based on input information with higher sparsity (i.e., the above-mentioned data set). In this case, the model performance requirements for the first model have changed.

[0193] Accordingly, in the above scenario, if the service type of the service to be transmitted changes from the FTP service type to the HTTP service type, that is, an event occurs, the first device performs target model monitoring on the first model.

[0194] For another example, when the service type of the service to be transmitted in the first device is the HTTP service type, that is, the event occurs, the first device performs target model monitoring on the first model.

[0195] Taking the association of an event with the model state of the first model as an example, in some implementations, the model state of the first model includes a transmission state of the first model, wherein the transmission state is used to indicate whether the transmission of the first model is successful.

[0196] Typically, the performance of the first model is unstable immediately after it is transferred to the first device. Therefore, target model monitoring can be performed on the first model. In other words, the aforementioned event includes the first model's transfer status being a successful transfer, or the first model's transfer completion.

[0197] In the embodiments of the present application, the transmission of the first model is not specifically limited. Taking the first device as a terminal device as an example, the transmission of the first model may include transmitting the first model from a network device to the terminal device, where the network device may be a core network device and / or an access network device. For details, please refer to the description of the first device and the second device below.

[0198] In some other implementations, if the model state of the first model includes the running state of the first model, wherein the running state is used to indicate whether the first model is activated. If the first model is activated, the running state of the first model is the activated state. If the first model is not activated, the running state of the first model is the inactivated state (also referred to as the "deactivated state"). Generally, a model in an activated state can execute a model reasoning process, while a model in a deactivated state cannot execute a model reasoning process.

[0199] Typically, immediately after the first model is activated, its performance is not stable. Therefore, the first model can be monitored by executing the target model monitoring described above. In other words, the event includes the activation state of the first model being successfully activated, or the event includes the first model being activated. In the embodiments of the present application, the specific content of the event is not limited. For example, the event may include the operating state of the first model being converted from a deactivated state to an activated state.

[0200] For example, the first model is converted from a deactivated state to an activated state, that is, an event occurs, and the first device performs target model monitoring on the first model.

[0201] Taking the example of an event associated with a successful model switch operation to the first model, in some implementations, if the first model switch is successful, the operating state of the first model is activated. As described above, at this point, the performance of the first model is not stable. Therefore, target model monitoring can be performed on the first model. In other words, the above event includes the successful first model switch, or the successful model switch operation to the first model.

[0202] For example, after the first device switches from cell 1 to cell 2, it needs to switch from other models to the first model. If the first model is switched successfully, that is, the event occurs, the first device performs target model monitoring on the first model.

[0203] An event is associated with the device state of the first device. In some implementations, the device state of the first device is used to indicate the movement state of the first device, wherein the movement state of the first device can be used to indicate one or more of the following: whether the first device is in a moving state; the movement speed of the first device; the change in the movement speed of the first device, etc.

[0204] The embodiments of the present application do not limit the specific content of the above-mentioned events. In some implementations, the above-mentioned event includes a change in the mobile state of the first device. For example, the above-mentioned event includes a change in the moving speed of the first device. Taking the change in the moving speed as an example, it may include the moving speed changing from high speed to low speed, and / or the moving speed changing from low speed to high speed. For another example, the above-mentioned event includes a change in the mobile state of the first device. Taking the change in the moving state as an example, it may include the moving state changing from a stationary state to a moving state, and / or the moving state changing from a moving state to a stationary state.

[0205] For example, if the moving speed of the first device changes from a low speed to a high speed, that is, an event occurs, the first device performs target model monitoring on the first model.

[0206] For another example, if the movement state of the first device changes from a stationary state to a moving state, that is, an event occurs, the first device performs target model monitoring on the first model.

[0207] In some other implementations, the device status of the first device is used to indicate the resource size of the available resources of the first device. In an embodiment of the present application, the available resources of the first device are not limited. In some implementations, the available resources of the first device may be available resources associated with the operation of the first model, or in other words, the available resources of the first device are resources that can be used to run the first model. Of course, in an embodiment of the present application, the available resources of the first device may also be the remaining resources of the first device, wherein some or all of the remaining resources can be used for the operation of the first model, or some or all of the remaining resources are used for other operations besides the operation of the first model.

[0208] In the embodiments of the present application, the specific content of the available resources is not limited. For example, the available resources of the first device may include the computing resources of the first device. For another example, the available resources of the first device may include the storage resources of the first device, wherein the storage resources may include the memory resources of the first device and / or the external storage resources of the first device. For another example, the available resources of the first device may include the energy consumption resources of the first device, wherein the energy consumption resources may include the power consumption of the first device.

[0209] In the embodiments of the present application, the specific content of the above-mentioned events is not limited. In some implementations, the above-mentioned events may include a change in the amount of available resources of the first device. For example, the amount of available resources of the first device decreases. For another example, the amount of available resources of the first device decreases below threshold 1. For another example, the amount of decrease in the amount of available resources of the first device is greater than threshold 2. Threshold 1 and / or threshold 2 may be determined based on one or more of the following: predefined information; preconfigured information; or configuration information of the network device.

[0210] For example, the computing resources of the first device decrease from resource amount 1 to resource amount 2, and the decrease Δ between resource amount 1 and resource amount 2 is greater than threshold 2, that is, an event occurs, and the first device performs target model monitoring on the first model.

[0211] For another example, the storage resources of the first device decrease from resource amount 1 to resource amount 2, and resource amount 2 is less than threshold 1, that is, an event occurs, and the first device performs target model monitoring on the first model.

[0212] Taking the association of an event with the wireless communication environment of the first device as an example, in some implementations, the wireless environment of the first device includes a wireless communication area of ​​the first device. For example, the wireless communication area may include one or more of the following: a cell, a RAN-based notification area (RNA), a tracking area (TA), an indoor wireless communication environment, an outdoor wireless communication environment, a hotspot coverage area, and a non-hotspot coverage area.

[0213] In some implementations, if the wireless communication environment of the first device includes a wireless communication area of ​​the first device, the event includes a change in the communication area of ​​the first device. Taking the wireless communication area as an example, the event may include a change in the cell of the first device, or the event may include a cell handover of the serving cell of the first device. Taking the wireless communication area as an example, the event may include a change in the RNA of the first device. Taking the wireless communication area as an example, the event may include a change in the TA of the first device, or in other words, an update of the TA of the first device. Taking the wireless communication area as an example, the event may include a change in the wireless communication environment of the first device from an indoor wireless communication environment to an outdoor wireless communication environment, or the event may include a change in the wireless communication environment of the first device from an outdoor wireless communication environment to an indoor wireless communication environment. Taking the wireless communication area as an example, the event may include a change in the wireless communication environment of the first device from a hotspot coverage area to a non-hotspot coverage area, or the event may include a change in the wireless communication environment of the first device from a non-hotspot coverage area to a hotspot coverage area.

[0214] In some other implementations, the wireless environment of the first device includes a wireless communication frequency band of the first device, where the wireless communication frequency band can be understood as a frequency band used by the first device for wireless communication.

[0215] In some other implementations, if the wireless communication environment of the first device includes a wireless communication frequency band of the first device, the event may include a change in the frequency band used by the first device. For example, the event may include a switch in the frequency band used by the first device from frequency band 1 to frequency band 2. For another example, the wireless communication frequency band may include a bandwidth part (BWP), and accordingly, the event may include a BWP switch occurring on the first device.

[0216] In other implementations, associating the wireless communication environment of the first device with the event may include associating the wireless communication environment type of the wireless communication environment of the first device with the event. The wireless communication environment type may include a line-of-sight communication type and a non-line-of-sight communication type.

[0217] In some implementations, the event may include a change in the wireless communication type of the first device. For example, the event may include a change in the wireless communication type of the first device from a line-of-sight communication type to a non-line-of-sight communication type. For another example, the event may include a change in the wireless communication type of the first device from a non-line-of-sight communication type to a line-of-sight communication type.

[0218] In the embodiments of the present application, the above-mentioned different events can be used independently of each other or in combination with each other. For example, if the event includes the successful transmission of the first model and the change of the business type of the business to be transmitted, if the first model is successfully transmitted and the business type of the business to be transmitted on the first device changes, the first device can perform target model monitoring on the first model.

[0219] In some scenarios, the above event may be configured by the second device. That is, the above method further includes: the second device sending first configuration information to the first device, where the first configuration information is used to configure the event for the first device.

[0220] In some implementations, the first configuration information may also be used to configure the duration of the target model monitoring process, or in other words, the first configuration information may also be used to configure the duration of the target model monitoring performed on the first model. Of course, in the embodiments of the present application, the duration may also be predefined or preconfigured. Alternatively, the duration may also be configured using other information in addition to the first configuration information.

[0221] In some implementations, the start time of the time period for executing target model monitoring can be determined based on the time of occurrence of the event. For example, the start time of the time period for executing target model monitoring can be the time of occurrence of the event. For another example, the start time of the time period for executing target model monitoring can be the time of occurrence of the event, offset by a time offset value of 1, where the time offset value of 1 can be predefined, preconfigured, or configured by the network device.

[0222] In some implementations, the duration of the target model monitoring can be maintained by timer 1, that is, the first device can perform target model monitoring on the first model during the operation of timer 1. In other implementations, the start condition of timer 1 can include the occurrence of the above event.

[0223] As mentioned above, in some scenarios, it may be possible to jointly determine whether to perform target model monitoring based on multiple events. At this time, multiple events can all be configured through the first configuration information. For example, the first configuration information may include information for configuring each of the multiple events, and these multiple information are transmitted simultaneously, that is, the first configuration information may include multiple information transmitted simultaneously to configure the multiple events respectively. For another example, the first configuration information may include information for configuring each of the multiple events, and these multiple information are transmitted independently, that is, the first configuration information may include multiple independently transmitted information to configure the multiple events respectively.

[0224] In the embodiment of the present application, the first configuration information is not limited. For example, the first configuration information may be a radio resource control (RRC) message.

[0225] Example 3: Model monitoring based on predicted data.

[0226] As previously mentioned, during the RRM measurement quiet period, the first device typically remains in a quiet state and does not perform RRM measurements. However, RRM measurements are the basis for model monitoring. If the first device does not perform RRM measurements, model monitoring cannot be performed, potentially reducing the accuracy of model monitoring. If the first device continues to perform measurements during the RRM measurement quiet period, this defeats the purpose of the quiet period and results in excessive power consumption by the first device.

[0227] Therefore, to address the above issues, embodiments of the present application propose model monitoring based on predicted data. In some implementations, the predicted data is used to perform target model monitoring on the first model during the RRM measurement quiet period. This allows the first device to refrain from performing RRM measurements during the RRM measurement quiet period, thereby reducing the first device's power consumption. Furthermore, during the RRM measurement quiet period, the first device can continuously monitor the first model based on the predicted data, thereby enabling continuous monitoring of the first model and improving the accuracy of model monitoring.

[0228] In some implementations, the predicted data is predicted based on measurement data obtained by RRM measurements performed by the first device during a non-silent period. Assuming that the measurement data is obtained based on RRM measurements during a first time period, where the first time period is a period before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period, this helps improve the accuracy of the predicted data. Of course, in embodiments of the present application, the first time period may not be adjacent to the RRM measurement silent period.

[0229] In the embodiments of the present application, the measurement data is not limited. In some implementations, the measurement data can be understood as the measurement results of the RRM measurement. For example, the measurement data may include an RSRP value. For another example, the measurement data may include a reference signal receiving quality (RSRQ) value. For another example, the measurement data may include a received signal strength indication (RSSI).

[0230] For example, in RRM measurement, the first device measures for 5 seconds and then remains silent for 5 seconds, and this process is repeated. If the first device performs RMM measurement within the period of 1 to 5 seconds and obtains measurement data 1. Thereafter, during the silent period of 6 to 10 seconds, the first device can perform target model monitoring on the first model based on predicted data 1 obtained from measurement data 1, where predicted data 1 is the predicted RRM measurement result within 6 to 10 seconds. Thereafter, the first device can input predicted data 1 into the first model to obtain predicted data 2, where predicted data 2 is the predicted RRM measurement result within 11 to 15 seconds. Thereafter, the first device can continue to perform RMM measurement within 11 to 15 seconds and obtain measurement data 2. At this time, the predicted data 2 can be compared with the measurement data 2 to determine the model monitoring result of the first model.

[0231] In some implementations, the second device may indicate to the first device whether to perform model monitoring based on the predicted data, that is, the second device sends a first indication message to the first device, wherein the first indication message may be used to indicate whether the first device performs model monitoring based on the predicted data. Of course, in an embodiment of the present application, the first device may also determine whether to perform model monitoring based on the predicted data through predefined information or preconfiguration information. For example, in a case where the preconfiguration information or predefined information indicates that the first device does not perform model monitoring based on the predicted data, if the first indication message is absent, the first device may determine not to perform model monitoring based on the predicted data based on the preconfiguration information or predefined information. On the contrary, if the first indication message indicates to perform model monitoring based on the predicted data, the first device may determine to perform model monitoring based on the predicted data based on the first indication message.

[0232] In the embodiment of the present application, the specific form of the first indication information is not limited. For example, the first indication information can occupy one bit, which helps to reduce the transmission resources for transmitting the first indication information. When the value of the bit is the first value, it can be used to instruct the first device to perform model monitoring based on predicted data. When the value of the bit is the second value, it can be used to instruct the first device not to perform model monitoring based on predicted data. The first value and the second value are two different values. For example, the value of the first value can be 0, and the value of the second value can be 1. For another example, the value of the first value can be 1, and the value of the second value can be 0. Of course, in the embodiment of the present application, the first indication information can also be transmitted through multiple bits.

[0233] Embodiment 4: Model monitoring based on triggering of a second device indication.

[0234] In some implementations, the second device may instruct the first device to perform target model monitoring on the first model by sending the first information to the first device. Therefore, the first information may also be referred to as a "model monitoring start instruction."

[0235] In the embodiment of the present application, the specific form of the first information is not limited. For example, the first information can occupy one bit, which helps to reduce the transmission resources for transmitting the first information. When the value of the bit is the first value, it can be used to instruct the first device to perform target model monitoring on the first model. When the value of the bit is the second value, it can be used to instruct the first device not to perform target model monitoring. The first value and the second value are two different values. For example, the value of the first value can be 0, and the value of the second value can be 1. For another example, the value of the first value can be 1, and the value of the second value can be 0. Of course, in the embodiment of the present application, the first information can also be transmitted through multiple bits.

[0236] In addition, in an embodiment of the present application, in a scenario where the first device does not perform target model monitoring, the second device may also not send the first indication information to the first device. In this way, if the first device does not receive the first information, the first device does not perform target model monitoring.

[0237] In the embodiment of the present application, the conditions for the second device to send the first information are not limited. In some implementations, the second device can determine whether to instruct the first device to perform target model monitoring through the first information based on one or more events described above.

[0238] In some implementations, if the second device detects a change in wireless communication quality, it may instruct the first device to perform target model monitoring. For example, if the second device detects that the wireless communication quality is less than a threshold value 1, it may instruct the first device to perform target model monitoring. In some implementations, if the second device detects a change in the wireless communication environment, it may instruct the first device to perform target model monitoring. For example, if the second device detects a change in the cell of the first device, it may instruct the first device to perform target model monitoring. In some implementations, if the second device detects a change in the service type of the service data transmitted by the first device, it may instruct the first device to perform target model monitoring. For example, if the second device detects that the service type of the service data transmitted by the first device changes from an FTP service to an HTTP service type, it may instruct the first device to perform target model monitoring.

[0239] In some scenarios, the first device may detect an anomaly in the first model while performing target model monitoring on the first model. Immediately stopping the first model and sending information indicating the anomaly to the second device may reduce the accuracy of model monitoring. This is because the first model may experience an anomaly for a short period of time due to various reasons and then resume normal operation. Therefore, immediately stopping the first model or immediately indicating the anomaly is unnecessary.

[0240] In response to the above-mentioned problem, a first timer is introduced in an embodiment of the present application so that after an abnormality is detected in the first model, the first model is continuously monitored within the duration corresponding to the first timer (also known as the second duration). Therefore, in an embodiment of the present application, the first timer is also called a "model monitoring abnormality timer". This scheme of performing model monitoring based on the first timer when an abnormality occurs in the first model helps to improve the accuracy of model monitoring compared to the scheme of immediately deactivating the first model or immediately indicating the abnormality of the first model introduced above.

[0241] That is to say, the target model monitoring is associated with the first timer, and the duration of monitoring the first model is determined based on the first timer. The first timer is used to indicate the second duration of model monitoring after the first moment, and the first moment is determined based on the moment when the abnormality occurs in the first model.

[0242] In some implementations, the first moment is determined based on the moment when the first model anomaly occurs, and may include the first moment being the moment when the first model anomaly occurs. Alternatively, the first moment is determined based on the moment when the first model anomaly occurs, and may include a moment offset by a time offset value, starting from the first moment being the moment when the first model anomaly occurs. This embodiment of the present application is not limited to this.

[0243] In some implementations, the start condition of the first timer may include an abnormality in the first model. In the embodiment of the present application, there is no limitation on the abnormality in the first model. In some implementations, the abnormality in the first model may include an abnormality in the model monitoring result of the first model. For example, if the model monitoring result of the first model is RSRP = -200dBm, and the model monitoring result seriously deviates from the historical model monitoring result range of the first model [-135dBm, -75dBm], then it can be determined that the first model is abnormal. Of course, in the embodiment of the present application, it can also be judged based on other model indicator information introduced below, and the judgment method is similar to the judgment method based on RSRP. For the sake of brevity, they are not listed one by one here.

[0244] In some implementations, the stop condition of the first timer may include that the first model returns to normal, at which point the first timer may be stopped. In an embodiment of the present application, the method for determining that the first model returns to normal is not limited. In some implementations, the first model returning to normal may include that the model monitoring result of the first model returns to normal. For example, if the model monitoring result of the first model is RSRP = -100dBm, and the model monitoring result belongs to the historical model monitoring result range of the first model [-135dBm, -75dBm], then it can be determined that the first model has returned to normal.

[0245] In other implementations, the first model returning to normal may include the model monitoring results of the first model returning to normal for P consecutive times, where P is a positive integer greater than or equal to 1. Taking P=2 as an example, the model monitoring results for two consecutive times are RSRP1=-100dBm, and RSRP2=-90dBm. RSRP1 and RSRP2 both belong to the historical model monitoring result range of the first model [-135dBm, -75dBm]. At this time, it can be determined that the first model has returned to normal. In an embodiment of the present application, the above-mentioned parameter P can be determined based on one or more of the following methods: pre-defined; pre-configured and network device configuration.

[0246] In some implementations, if the first device deploys multiple models, some or all of the different models in the multiple models are associated with different first timers. Of course, in the embodiment of the present application, the first timers associated with different models in the multiple models can be the same.

[0247] In the embodiments of the present application, the first timer can be used in combination with any of the target model monitoring methods described above. Of course, in the embodiments of the present application, the first timer can also be used in combination with traditional model monitoring. For ease of understanding, the following description uses the combination of the first timer and the periodic model monitoring method described in Example 1 as an example.

[0248] In some scenarios, the end time of the first duration for performing target model monitoring on the first model may be earlier than the end time of the second duration. In other words, the first duration may end while the first timer is running. At this time, the first device can continue to perform target model monitoring on the first model until the first timer times out or the first timer stops, which helps improve the accuracy of model monitoring performed on the first model.

[0249] Figure 14 shows a schematic diagram of combining the first timer with cycle-based model monitoring in an embodiment of the present application. Referring to Figure 14, in cycle 1, the first duration is a period of time from moment t1 to moment t2. At this time, model monitoring can be performed on the first model from moment t1 to moment t2. Assume that the first model is abnormal at moment t3 within the first duration, and the first timer is started, wherein the second duration indicated by the first timer is a period of time from moment t3 to moment t4, wherein moment t4 is after moment t2. Based on the first duration introduced above, it can be seen that the first device can only perform model monitoring on the first model from moment t1 to moment t2, but since the first model is abnormal, at this time, the first device can continue to perform model monitoring on the first model from moment t3 to moment t4, that is, the first device can continue to perform model monitoring on the first model from moment t2 to moment t4.

[0250] In some implementations, the first timer may be configured by the second device. That is, the method further includes: the first device receiving second configuration information sent by the second device, where the second configuration information is used to configure the first timer. Of course, in the embodiments of the present application, the first timer may also be predefined or preconfigured.

[0251] The above describes the target model monitoring of the embodiment of the present application in conjunction with Figures 1 to 14. The following describes the model indicators monitored by the target model monitoring.

[0252] In some implementations, the information of the model indicators of the first model includes one or more of the following: model information of the first model; model call information of the first model; model performance statistical information of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; and model input data quality information of the first model.

[0253] For example, the model information of the first model includes the model information of the first model. The model information of the first model is used to describe basic information of the first model. In some implementations, the model information of the first model is used to indicate one or more of the following: a model ID of the first model; a model update time of the first model; a model version of the first model; a model size of the first model; historical model versions of the first model; a model authorization entity of the first model; and a licensed manufacturer of the first model.

[0254] Taking the example of the model indicator information of the first model including the model call information of the first model, the model call information of the first model may include the model call information of the first model, wherein the model call information of the first model may be used to indicate the situation in which the first model is called. In some implementations, the call information of the first model is used to indicate one or more of the following: the number of calls of the first model; the call time of the first model; the call frequency of the first model; and the number of deactivations of the first model.

[0255] For example, the model indicator of the first model includes model performance statistics of the first model, where the model performance statistics of the first model is used to analyze the model inference performance of the first model. In some implementations, the model performance statistics of the first model are used to indicate one or more of the following: mean average precision; maximum value of the model inference result; minimum value of the model inference result; median of the model inference result; quartiles of the model inference result; accuracy of the model inference; recall of the model inference; precision of the model inference; and model drift information.

[0256] For example, the model metrics of the first model include model stability information of the first model, which is used to evaluate the deviation between multiple inference results of the first model. In some implementations, the model score information of the first model is used to indicate one or more of the following: mean square error, root mean square error, mean absolute error, coefficient of determination (R2_score), confidence level, model performance continuity information, and model performance significant degradation information.

[0257] For example, the model indicators of the first model include the model scoring information of the first model. The model scoring information of the first model is used to evaluate the model performance of the first model when performing a certain task. Generally, the model scoring information is a comprehensive consideration of information such as model performance statistics and model stability information. Different models can be compared on a unified dimension using the model scoring information. For example, when the model scoring information can be a percentage score, for example, the model scoring information of the first model can be 95 points.

[0258] Taking the example of a first model's model indicator including the model operation information of the first model, the model operation information of the first model includes device-level or resource-level information during the model inference process. For example, the model operation information of the first model may include one or more of the following: the model response speed of the first model; the number of response failures of the first model; and the resource utilization of the first model, where resources may include computing resources and / or storage resources. Computing resources may include, for example, CPU resources and / or GPU resources, and storage resources may include memory resources and / or external storage resources.

[0259] For example, if the model indicator of the first model includes model input data quality information of the first model, the model input data quality information is used to indicate the quality of the model input data. In some implementations, the model input data quality information may include one or more of the following: completeness of the model input data; sparsity of the model input data; and measurement accuracy of the model input data.

[0260] In some implementations, the second device may indicate to the first device the model metrics of the first model being monitored by the target model, thereby improving the rationality of the model monitoring. Specifically, the method includes: the second device sending first information to the first device, wherein the first information includes information indicating the model metrics of the first model being monitored by the target model.

[0261] Of course, in the embodiment of the present application, the above-mentioned first information is used to perform target model monitoring on the first model. The first information may also include model information of the first model; and information used to indicate whether to perform target model monitoring.

[0262] In an embodiment of the present application, any information involved above (for example, the first information, the first request, the first configuration information, and the second configuration information, etc.) can be carried by one or more of the following message types: NR Positioning Protocol A (NRPPa) message, long term evolution positioning protocol (LPP) message, non-access stratum (NAS) message, radio resource control (RRC) message, media access control control element (MAC CE), downlink control information (DCI), uplink control information (UCI), physical uplink control channel (PUCCH), physical uplink shared channel (PUSCH), inter-node message, Xn interface message, F1 interface message, E1 interface message, NG interface message, core network service architecture-based message or AI dedicated message.

[0263] In an embodiment of the present application, any of the information mentioned above can be carried by one or more of unicast messages, multicast messages and broadcast messages.

[0264] Unicast messages can be understood as one-to-one transmissions, meaning they are sent from one sender to one receiver. The source delivers unicast messages over a unicast channel, and only devices or network equipment allocated the corresponding unicast resources can attempt to receive them. Unicast messages are also called dedicated signaling.

[0265] The multicast message described above can be understood as a one-to-many transmission, meaning it's sent from a single sender to multiple receivers. The source transmits the message via a multicast channel. Terminal devices or network devices within the multicast signal's coverage area and that are members of the group can attempt to receive the message. When a terminal device or network device joins a group, it acquires multicast channel resources.

[0266] The broadcast message can be understood as a message transmitted from one sender to any receiver. The source transmits the message via a broadcast channel, and any terminal or network device within the coverage area of ​​the broadcast signal can attempt to receive it.

[0267] In the embodiments of the present application, the first device and / or the second device are not limited. For example, the first device is a terminal device, and the second device is a network device. In another example, the first device is a network device, and the second device is a terminal device. In another example, the first device is a first terminal device, and the second device is a second terminal device. In another example, the first device is a first network device, and the second device is a second network device.

[0268] The network device is an access network device, a core network device, an AI / ML model-related information management device, or an operation administration and maintenance (OAM) device. Exemplarily, the access network device is any one of the following: a gNB, a centralized unit (CU), a distributed unit (DU), a centralized unit-control plane (CU-CP), or a centralized unit-user plane (CU-UP).

[0269] Exemplarily, the core network device is any one of the following: location management function (LMF) network element, network slice selection function (NSSF), authentication server function (AUSF), unified data management (UDM), access and mobility management function (AMF), session management function (SMF), policy control function (PCF), user plane function (UPF), sensing control function (SF), network data analysis (NWDAF) network element.

[0270] The method embodiment of the present application is described in detail above in conjunction with Figures 1 to 14 . The device embodiment of the present application is described in detail below in conjunction with Figures 15 to 17 . It should be understood that the description of the method embodiment corresponds to the description of the device embodiment. Therefore, for portions not described in detail, reference can be made to the above method embodiment.

[0271] FIG15 is a schematic diagram of a communication device according to an embodiment of the present application. The communication device 1500 shown in FIG15 is a first device, and the communication device 1500 includes a processing unit 1510 .

[0272] The processing unit 1510 is used to perform target model monitoring on the first model, and the target model monitoring includes one of the following: cycle-based model monitoring; event-triggered model monitoring; second device indication-triggered model monitoring; and predicted data-based model monitoring, where the predicted data is predicted based on measurement data.

[0273] In some implementations, the target model monitoring is the cycle-based model monitoring, and the cycle-based model monitoring is performed based on target parameters, and the target parameters include one or more of the following: a first parameter, the first parameter is used to determine the cycle length of the cycle; a second parameter, the second parameter is used to determine a first duration of executable model monitoring within the cycle.

[0274] In some implementations, the target parameter includes the first parameter, which is used to indicate a first time interval between two adjacent cycles; or the first parameter is used to indicate a basic time interval, which is used to determine the first time interval.

[0275] In some implementations, the first parameter is used to indicate the first time interval, and the two adjacent cycles include a first cycle and a second cycle, and the first time interval is the time interval between the start time of the first duration of executable model monitoring in the first cycle and the start time of the first duration of executable model monitoring in the second cycle.

[0276] In some implementations, the first parameter is used to indicate the basic time interval, and the first time interval is an integer multiple of the basic time interval.

[0277] In some implementations, the first time interval is the time interval between the start times of two adjacent cycles.

[0278] In some implementations, the target parameter includes the second parameter, where the second parameter is used to indicate the length of the first duration and / or the start time of the first duration.

[0279] In some implementations, the second parameter is used to indicate the start time of the first duration, and the second parameter includes a time offset between the start time of the first duration and the start time of a period to which the first duration belongs.

[0280] In some implementations, the first model is one of multiple models deployed in the first device, and the target parameters associated with some or all of the multiple models are different; or the target parameters associated with different models in the multiple models are the same.

[0281] In some implementations, the communication device further includes: a sending unit for sending a first request to a second device, wherein the first request is used to request adjustment of the target parameters associated with the first model; a first receiving unit for receiving a response message sent by the second device to the first request, wherein the response message includes the adjusted target parameters and / or information indicating whether to accept adjustment of the target parameters based on the first request.

[0282] In some implementations, the first request includes one or more of the following: requirement information for performing the target model monitoring on the first model; candidate values ​​of the target parameters recommended by the first device, and the candidate values ​​of the target parameters are used to determine the adjusted target parameters.

[0283] In some implementations, the target model monitoring is the event-triggered model monitoring, and the event is associated with one or more of the following: the service type of the service to be transmitted; the model status of the first model; the device status of the first device; the wireless communication environment of the first device; and the success of the model switching operation to switch to the first model.

[0284] In some implementations, the event is associated with the model state of the first model. If the model state of the first model includes the transmission state of the first model, the event includes the transmission state of the first model being successful transmission; if the model state of the first model includes the activation state of the first model, the event includes the activation state of the first model being successful activation.

[0285] In some implementations, the event is associated with the device status of the first device. If the device status of the first device is used to indicate the mobile status of the first device, the event includes a change in the mobile status of the first device; if the device status of the first device is used to indicate the resource size of the available resources of the first device, the event includes a change in the resource size of the available resources of the first device.

[0286] In some implementations, the event is associated with the wireless communication environment of the first device. If the wireless communication environment of the first device includes the wireless communication area of ​​the first device, the event includes a change in the communication area where the first device is located; if the wireless communication environment of the first device includes the wireless communication frequency band of the first device, the event includes a change in the wireless communication frequency band used by the first device; the event includes a change in the wireless communication type of the first device, and the wireless communication type is associated with the wireless communication environment of the first device, and the wireless communication type includes a line-of-sight communication type and a non-line-of-sight communication type.

[0287] In some implementations, the event is associated with a service type of the service to be transmitted, and the event includes a change in the service type of the service to be transmitted.

[0288] In some implementations, the communication device further includes: a second receiving unit, configured to receive first configuration information sent by a second device, where the first configuration information is used to configure the event for the first device.

[0289] In some implementations, the target model monitoring is the model monitoring based on prediction data, and the prediction data is used to perform the target model monitoring on the first model during a silent period of radio resource management RRM measurements.

[0290] In some implementations, the measurement data is obtained based on RRM measurements in a first time period, where the first time period is a period before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period.

[0291] In some implementations, the target model monitoring is associated with a first timer, and the duration of monitoring the first model is determined based on the first timer, and the first timer is used to indicate a second duration of model monitoring after a first moment, and the first moment is determined based on the moment when an abnormality occurs in the first model.

[0292] In some implementations, the first model is one of multiple models deployed by the first device, and the first timers associated with some or all of the different models in the multiple models are different; or the first timers associated with different models in the multiple models are the same.

[0293] In some implementations, the communication device further includes: a third receiving unit, configured to receive second configuration information sent by a second device, where the second configuration information is used to configure the first timer.

[0294] In some implementations, the communication device further includes: a fourth receiving unit, configured to receive first information sent by a second device, where the first information is used to perform the target model monitoring on the first model.

[0295] In some implementations, the first information includes one or more of the following: model information of the first model; model information of the first model; information for indicating whether to perform the target model monitoring; information for indicating the model indicators of the first model monitored by the target model monitoring.

[0296] In some implementations, the first information includes information indicating model indicators of the first model, and the information of the model indicators of the first model includes one or more of the following: model call information of the first model; model performance statistical information of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; and model input data quality information of the first model.

[0297] FIG16 is a schematic diagram of a communication device according to an embodiment of the present application. The communication device 1600 shown in FIG16 is a second device, and the communication device 1600 includes a sending unit 1610 .

[0298] The sending unit 1610 sends first information to the first device, where the first information is used to perform target model monitoring on the first model, wherein the target model monitoring includes one of the following: period-based model monitoring; event-triggered model monitoring; model monitoring triggered by a second device indication; model monitoring based on predicted data, where the predicted data is predicted based on measurement data.

[0299] In some implementations, the target model monitoring is the cycle-based model monitoring, and the cycle-based model monitoring is performed based on target parameters, and the target parameters include one or more of the following: a first parameter, the first parameter is used to determine the cycle length of the cycle; a second parameter, the second parameter is used to determine a first duration of executable model monitoring within the cycle.

[0300] In some implementations, the target parameter includes the first parameter, which is used to indicate a first time interval between two adjacent cycles; or the first parameter is used to indicate a basic time interval, which is used to determine the first time interval.

[0301] In some implementations, the first parameter is used to indicate the first time interval, and the two adjacent cycles include a first cycle and a second cycle, and the first time interval is the time interval between the start time of the first duration of executable model monitoring in the first cycle and the start time of the first duration of executable model monitoring in the second cycle.

[0302] In some implementations, the first parameter is used to indicate the basic time interval, and the first time interval is an integer multiple of the basic time interval.

[0303] In some implementations, the first time interval is the time interval between the start times of two adjacent cycles.

[0304] In some implementations, the target parameter includes the second parameter, where the second parameter is used to indicate the length of the first duration and / or the start time of the first duration.

[0305] In some implementations, the second parameter is used to indicate the start time of the first duration, and the second parameter includes a time offset between the start time of the first duration and the start time of a period to which the first duration belongs.

[0306] In some implementations, the first model is one of multiple models deployed in the first device, and the target parameters associated with some or all of the multiple models are different; or the target parameters associated with different models in the multiple models are the same.

[0307] In some implementations, the communication device further includes: a receiving unit for receiving a first request sent by a first device, wherein the first request is used to request adjustment of the target parameters associated with the first model; and a sending unit for sending a response message to the first request to the first device, wherein the response message includes the adjusted target parameters and / or information indicating whether to accept adjustment of the target parameters based on the first request.

[0308] In some implementations, the first request includes one or more of the following: requirement information for performing the target model monitoring on the first model; candidate values ​​of the target parameters recommended by the first device, and the candidate values ​​of the target parameters are used to determine the adjusted target parameters.

[0309] In some implementations, the target model monitoring is the event-triggered model monitoring, and the event is associated with one or more of the following: the service type of the service to be transmitted; the model status of the first model; the device status of the first device; the wireless communication environment of the first device; and the success of the model switching operation to switch to the first model.

[0310] In some implementations, the event is associated with the model state of the first model. If the model state of the first model includes the transmission state of the first model, the event includes the transmission state of the first model being successful transmission; if the model state of the first model includes the activation state of the first model, the event includes the activation state of the first model being successful activation.

[0311] In some implementations, the event is associated with the device status of the first device. If the device status of the first device is used to indicate the mobile status of the first device, the event includes a change in the mobile status of the first device; if the device status of the first device is used to indicate the resource size of the available resources of the first device, the event includes a change in the resource size of the available resources of the first device.

[0312] In some implementations, the event is associated with the wireless communication environment of the first device. If the wireless communication environment of the first device includes the wireless communication area of ​​the first device, the event includes a change in the communication area where the first device is located; if the wireless communication environment of the first device includes the wireless communication frequency band of the first device, the event includes a change in the wireless communication frequency band used by the first device; the event includes a change in the wireless communication type of the first device, and the wireless communication type is associated with the wireless communication environment of the first device, and the wireless communication type includes a line-of-sight communication type and a non-line-of-sight communication type.

[0313] In some implementations, the event is associated with a service type of the service to be transmitted, and the event includes a change in the service type of the service to be transmitted.

[0314] In some implementations, the sending unit is further configured to: send first configuration information to the first device, where the first configuration information is used to configure the event for the first device.

[0315] In some implementations, the target model monitoring is the model monitoring based on prediction data, and the prediction data is used to perform the target model monitoring on the first model during a silent period of radio resource management RRM measurements.

[0316] In some implementations, the measurement data is obtained based on RRM measurements in a first time period, where the first time period is a period before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period.

[0317] In some implementations, the sending unit is further used to: send second configuration information to the first device, the second configuration information is used to configure a first timer associated with the target model monitoring, the first timer is used to indicate a second duration of model monitoring after a first moment, and the first moment is determined based on the moment when an abnormality occurs in the first model.

[0318] In some implementations, the first information includes one or more of the following: model information of the first model; model information of the first model; information for indicating whether to perform the target model monitoring; information for indicating the model indicators of the first model monitored by the target model monitoring.

[0319] In some implementations, the first information includes information indicating model indicators of the first model, and the information of the model indicators of the first model includes one or more of the following: model call information of the first model; model performance statistical information of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; and model input data quality information of the first model.

[0320] In an optional embodiment, the processing unit 1510 may be a processor 1710. The communication device 1500 may further include a transceiver 1730 and a memory 1720, as specifically shown in FIG17 .

[0321] In an optional embodiment, the sending unit 1610 may be a transceiver 1730. The communication device 1600 may further include a processor 1710 and a memory 1720, as specifically shown in FIG17 .

[0322] Figure 17 is a schematic block diagram of a communication device according to an embodiment of the present application. The dashed lines in Figure 17 indicate that the unit or module is optional. Device 1700 may be used to implement the method described in the above method embodiment. Device 1700 may be a chip, a terminal device, or a network device.

[0323] The device 1700 may include one or more processors 1717. The processor 1717 may support the device 1700 to implement the method described in the above method embodiment. The processor 1717 may be a general-purpose processor or a special-purpose processor. For example, the processor may be a central processing unit (CPU). Alternatively, the processor may be another general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0324] The apparatus 1700 may further include one or more memories 1720. The memories 1720 may store programs that can be executed by the processor 1717, causing the processor 1717 to perform the methods described in the above method embodiments. The memories 1720 may be independent of the processor 1717 or integrated into the processor 1717.

[0325] The apparatus 1700 may further include a transceiver 1730. The processor 1717 may communicate with other devices or chips via the transceiver 1730. For example, the processor 1717 may transmit and receive data with other devices or chips via the transceiver 1730.

[0326] The present application also provides a computer-readable storage medium for storing a program. The computer-readable storage medium can be applied to a terminal or network device provided in the present application, and the program enables a computer to execute the method performed by the terminal or network device in each embodiment of the present application.

[0327] The present application also provides a computer program product. The computer program product includes a program. The computer program product can be applied to a terminal or network device provided in the present application, and the program causes a computer to execute the method performed by the terminal or network device in each embodiment of the present application.

[0328] The embodiments of the present application also provide a computer program. The computer program can be applied to the terminal or network device provided in the embodiments of the present application, and the computer program enables a computer to execute the method performed by the terminal or network device in each embodiment of the present application.

[0329] It should be understood that the terms "system" and "network" in this application can be used interchangeably. In addition, the terms used in this application are only used to explain the specific embodiments of this application and are not intended to limit this application. The terms "first", "second", "third", and "fourth" in the specification and claims of this application and the accompanying drawings are used to distinguish different objects rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions.

[0330] In the embodiments of this application, the term "indication" may refer to a direct indication, an indirect indication, or an indication of an association. For example, "A indicates B" may refer to a direct indication of B, e.g., B can obtain information through A; it may refer to an indirect indication of B, e.g., A indicates C, e.g., B can obtain information through C; or it may refer to an association between A and B.

[0331] In the embodiment of the present application, "B corresponding to A" means that B is associated with A and B can be determined based on A. However, it should be understood that determining B based on A does not mean determining B based solely on A, but B can also be determined based on A and / or other information.

[0332] In the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship between indication and indication, configuration and configuration, etc.

[0333] In the embodiments of the present application, "pre-definition" or "pre-configuration" may be implemented by pre-storing corresponding codes, tables, or other methods that can be used to indicate relevant information in a device (e.g., a terminal device and a network device). The present application does not limit the specific implementation method. For example, pre-definition may refer to information defined in a protocol.

[0334] In the embodiments of the present application, the “protocol” may refer to a standard protocol in the communications field, for example, it may include an LTE protocol, an NR protocol, and related protocols used in future communication systems, and the present application does not limit this.

[0335] In the embodiments of this application, the term "and / or" is simply a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0336] In various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0337] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0338] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0339] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0340] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be read by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0341] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A model monitoring method, characterized in that: include: The first device performs target model monitoring on the first model, where the target model monitoring includes one of the following: Cycle-based model monitoring; Model monitoring based on event triggering; Model monitoring triggered based on a second device indication; Model monitoring based on predicted data, wherein the predicted data is predicted based on the measured data.

2. The method according to claim 1, characterized in that The target model monitoring is the cycle-based model monitoring, and the cycle-based model monitoring is performed based on target parameters, and the target parameters include one or more of the following: A first parameter, the first parameter is used to determine the cycle length of the cycle; A second parameter is used to determine a first duration during which the model monitoring can be performed within the period.

3. The method according to claim 2, characterized in that The target parameter includes the first parameter, where the first parameter is used to indicate a first time interval between two adjacent cycles; or The first parameter is used to indicate a basic time interval, and the basic time interval is used to determine the first time interval.

4. The method according to claim 3, characterized in that The first parameter is used to indicate the first time interval, and the two adjacent cycles include a first cycle and a second cycle. The first time interval is the time interval between the start time of the first duration of executable model monitoring in the first cycle and the start time of the first duration of executable model monitoring in the second cycle.

5. The method according to claim 4, characterized in that The first parameter is used to indicate the basic time interval, and the first time interval is an integer multiple of the basic time interval.

6. The method according to any one of claims 3 to 5, characterized in that The first time interval is the time interval between the start times of two adjacent cycles.

7. The method according to any one of claims 2 to 6, characterized in that The target parameters include the second parameter, where the second parameter is used to indicate the length of the first duration and / or the start time of the first duration.

8. The method according to claim 7, characterized in that The second parameter is used to indicate the start time of the first duration, and the second parameter includes a time offset between the start time of the first duration and the start time of a cycle to which the first duration belongs.

9. The method according to any one of claims 2 to 8, characterized in that The first model is one of a plurality of models deployed in the first device, and the target parameters associated with some or all of the plurality of models are different; or The target parameters associated with different models among the multiple models are the same.

10. The method according to any one of claims 2 to 9, characterized in that The method further comprises: The first device sends a first request to the second device, where the first request is used to request adjustment of the target parameter associated with the first model; The first device receives a response message to the first request sent by the second device, where the response message includes the adjusted target parameter; and / or information indicating whether adjustment of the target parameter based on the first request is allowed.

11. The method according to claim 10, characterized in that The first request includes one or more of the following: Requirement information for executing the target model monitoring on the first model; The candidate value of the target parameter suggested by the first device, and the candidate value of the target parameter is used to determine the adjusted target parameter.

12. The method according to claim 1, characterized in that The target model monitoring is the event-triggered model monitoring, and the event is associated with one or more of the following: The service type of the service to be transmitted; a model state of the first model; a device status of the first device; The wireless communication environment of the first device; The model switching operation to the first model is successful.

13. The method according to claim 12, characterized in that The event is associated with a model state of the first model, If the model state of the first model includes a transmission state of the first model, the event includes that the transmission state of the first model is successfully transmitted; If the model state of the first model includes an activation state of the first model, the event includes that the activation state of the first model is successfully activated.

14. The method according to claim 12 or 13, characterized in that The event is associated with a device state of the first device, If the device state of the first device is used to indicate a movement state of the first device, the event includes a change in the movement state of the first device; If the device state of the first device is used to indicate the resource size of the available resources of the first device, the event includes a change in the resource size of the available resources of the first device.

15. The method according to any one of claims 12 to 14, characterized in that The event is associated with a wireless communication environment of the first device, If the wireless communication environment of the first device includes a wireless communication area of ​​the first device, the event includes a change in the communication area where the first device is located; If the wireless communication environment of the first device includes a wireless communication frequency band of the first device, the event includes a change in the wireless communication frequency band used by the first device; The event includes a change in a wireless communication type of the first device, where the wireless communication type is associated with a wireless communication environment of the first device, and the wireless communication type includes a line-of-sight communication type and a non-line-of-sight communication type.

16. The method according to any one of claims 12 to 15, characterized in that The event is associated with the service type of the service to be transmitted, and the event includes a change in the service type of the service to be transmitted.

17. The method according to any one of claims 12 to 16, characterized in that The method further comprises: The first device receives first configuration information sent by the second device, where the first configuration information is used to configure the event for the first device.

18. The method of claim 1, wherein: The target model monitoring is the model monitoring based on prediction data, and the prediction data is used to perform the target model monitoring on the first model during a silent period of radio resource management RRM measurement.

19. The method according to claim 18, characterized in that The measurement data is obtained based on RRM measurement in a first time period, where the first time period is a period of time before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period.

20. The method according to any one of claims 1 to 19, characterized in that The target model monitoring is associated with a first timer, and the duration of monitoring the first model is determined based on the first timer. The first timer is used to indicate a second duration of model monitoring after a first moment, and the first moment is determined based on the moment when an abnormality occurs in the first model.

21. The method of claim 20, wherein: The first model is one of multiple models deployed by the first device, and different first timers are associated with some or all different models among the multiple models; or The first timer associated with different models among the multiple models is the same.

22. The method according to claim 20 or 21, characterized in that The method further comprises: The first device receives second configuration information sent by the second device, where the second configuration information is used to configure the first timer.

23. The method according to any one of claims 1 to 22, characterized in that The method further comprises: The first device receives first information sent by the second device, where the first information is used to perform the target model monitoring on the first model.

24. The method of claim 23, wherein: The first information includes one or more of the following: model information of the first model; Information for indicating whether to perform the target model monitoring; Information used to instruct the target model to monitor the model index of the monitored first model.

25. The method of claim 24, wherein: The first information includes information for indicating a model index of the first model, and the information of the model index of the first model includes one or more of the following: model information of the first model; Model calling information of the first model; model performance statistics of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; Model input data quality information of the first model.

26. A method for model monitoring, characterized in that: include: The second device sends first information to the first device, where the first information is used to perform target model monitoring on the first model, wherein the target model monitoring includes one of the following: Cycle-based model monitoring; Model monitoring based on event triggering; Model monitoring triggered based on a second device indication; Model monitoring based on predicted data, wherein the predicted data is predicted based on the measured data.

27. The method of claim 26, wherein: The target model monitoring is the cycle-based model monitoring, and the cycle-based model monitoring is performed based on target parameters, and the target parameters include one or more of the following: A first parameter, the first parameter is used to determine the cycle length of the cycle; A second parameter is used to determine a first duration during which the model monitoring can be performed within the period.

28. The method of claim 27, wherein: The target parameter includes the first parameter, where the first parameter is used to indicate a first time interval between two adjacent cycles; or The first parameter is used to indicate a basic time interval, and the basic time interval is used to determine the first time interval.

29. The method of claim 28, wherein: The first parameter is used to indicate the first time interval, and the two adjacent cycles include a first cycle and a second cycle. The first time interval is the time interval between the start time of the first duration of executable model monitoring in the first cycle and the start time of the first duration of executable model monitoring in the second cycle.

30. The method of claim 29, wherein: The first parameter is used to indicate the basic time interval, and the first time interval is an integer multiple of the basic time interval.

31. The method according to any one of claims 27 to 30, characterized in that The first time interval is the time interval between the start times of two adjacent cycles.

32. The method according to any one of claims 27 to 31, characterized in that The target parameters include the second parameter, where the second parameter is used to indicate the length of the first duration and / or the start time of the first duration.

33. The method of claim 32, wherein: The second parameter is used to indicate the start time of the first duration, and the second parameter includes a time offset between the start time of the first duration and the start time of a cycle to which the first duration belongs.

34. The method according to any one of claims 27 to 33, characterized in that The first model is one of a plurality of models deployed in the first device, and the target parameters associated with some or all of the plurality of models are different; or The target parameters associated with different models among the multiple models are the same.

35. The method according to any one of claims 27 to 34, characterized in that The method further comprises: The second device receives a first request sent by the first device, where the first request is used to request adjustment of the target parameter associated with the first model; The second device sends a response message to the first request to the first device, where the response message includes the adjusted target parameter and / or information indicating whether to accept adjustment of the target parameter based on the first request.

36. The method of claim 35, wherein: The first request includes one or more of the following: Requirement information for executing the target model monitoring on the first model; The candidate value of the target parameter suggested by the first device, and the candidate value of the target parameter is used to determine the adjusted target parameter.

37. The method of claim 36, wherein: The target model monitoring is the event-triggered model monitoring, and the event is associated with one or more of the following: The service type of the service to be transmitted; a model state of the first model; a device status of the first device; The wireless communication environment of the first device; The model switching operation to the first model is successful.

38. The method of claim 37, wherein: The event is associated with a model state of the first model, If the model state of the first model includes a transmission state of the first model, the event includes that the transmission state of the first model is successfully transmitted; If the model state of the first model includes an activation state of the first model, the event includes that the activation state of the first model is successfully activated.

39. The method according to claim 37 or 38, characterized in that The event is associated with a device state of the first device, If the device state of the first device is used to indicate a movement state of the first device, the event includes a change in the movement state of the first device; If the device state of the first device is used to indicate the resource size of the available resources of the first device, the event includes a change in the resource size of the available resources of the first device.

40. The method according to any one of claims 37 to 39, characterized in that The event is associated with a wireless communication environment of the first device, If the wireless communication environment of the first device includes a wireless communication area of ​​the first device, the event includes a change in the communication area where the first device is located; If the wireless communication environment of the first device includes a wireless communication frequency band of the first device, the event includes a change in the wireless communication frequency band used by the first device; The event includes a change in a wireless communication type of the first device, where the wireless communication type is associated with a wireless communication environment of the first device, and the wireless communication type includes a line-of-sight communication type and a non-line-of-sight communication type.

41. The method according to any one of claims 37 to 40, characterized in that The event is associated with the service type of the service to be transmitted, and the event includes a change in the service type of the service to be transmitted.

42. The method according to any one of claims 37 to 41, characterized in that The method further comprises: The second device sends first configuration information to the first device, where the first configuration information is used to configure the event for the first device.

43. The method of claim 26, wherein: The target model monitoring is the model monitoring based on prediction data, and the prediction data is used to perform the target model monitoring on the first model during a silent period of radio resource management RRM measurement.

44. The method of claim 43, wherein: The measurement data is obtained based on RRM measurement in a first time period, where the first time period is a period of time before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period.

45. The method according to any one of claims 26 to 44, characterized in that The method further comprises: The second device sends second configuration information to the first device, the second configuration information is used to configure a first timer associated with the target model monitoring, the first timer is used to indicate a second duration of model monitoring after a first moment, and the first moment is determined based on the moment when an abnormality occurs in the first model.

46. ​​The method according to any one of claims 26 to 45, characterized in that The first information includes one or more of the following: model information of the first model; Information for indicating whether to perform the target model monitoring; Information used to instruct the target model to monitor the model index of the monitored first model.

47. The method of claim 46, wherein: The first information includes information for indicating a model index of the first model, and the information of the model index of the first model includes one or more of the following: model information of the first model; Model calling information of the first model; model performance statistics of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; Model input data quality information of the first model.

48. A communication device, characterized in that: The communication device is a first device, comprising: A processing unit is configured to perform target model monitoring on the first model, wherein the target model monitoring includes one of the following: Cycle-based model monitoring; Model monitoring based on event triggering; Model monitoring triggered based on a second device indication; Model monitoring based on predicted data, wherein the predicted data is predicted based on the measured data.

49. The communication device according to claim 48, characterized in that The target model monitoring is the cycle-based model monitoring, and the cycle-based model monitoring is performed based on target parameters, and the target parameters include one or more of the following: A first parameter, the first parameter is used to determine the cycle length of the cycle; A second parameter is used to determine a first duration during which the model monitoring can be performed within the period.

50. The communication device according to claim 49, characterized in that The target parameter includes the first parameter, where the first parameter is used to indicate a first time interval between two adjacent cycles; or The first parameter is used to indicate a basic time interval, and the basic time interval is used to determine the first time interval.

51. The communication device according to claim 50, characterized in that The first parameter is used to indicate the first time interval, and the two adjacent cycles include a first cycle and a second cycle. The first time interval is the time interval between the start time of the first duration of executable model monitoring in the first cycle and the start time of the first duration of executable model monitoring in the second cycle.

52. The communication device according to claim 51, characterized in that The first parameter is used to indicate the basic time interval, and the first time interval is an integer multiple of the basic time interval.

53. The communication device according to any one of claims 50 to 52, characterized in that: The first time interval is the time interval between the start times of two adjacent cycles.

54. The communication device according to any one of claims 49 to 53, characterized in that: The target parameters include the second parameter, where the second parameter is used to indicate the length of the first duration and / or the start time of the first duration.

55. The communication device according to claim 54, characterized in that The second parameter is used to indicate the start time of the first duration, and the second parameter includes a time offset between the start time of the first duration and the start time of a cycle to which the first duration belongs.

56. The communication device according to any one of claims 49 to 55, characterized in that: The first model is one of a plurality of models deployed in the first device, and the target parameters associated with some or all of the plurality of models are different; or The target parameters associated with different models among the multiple models are the same.

57. The communication device according to any one of claims 49 to 56, characterized in that: The communication device further comprises: A sending unit, configured to send a first request to a second device, wherein the first request is used to request adjustment of the target parameter associated with the first model; The first receiving unit is used to receive a response message sent by the second device to the first request, wherein the response message includes the adjusted target parameter and / or information indicating whether to accept the adjustment of the target parameter based on the first request.

58. The communication device according to claim 57, characterized in that The first request includes one or more of the following: Requirement information for executing the target model monitoring on the first model; The candidate value of the target parameter suggested by the first device, and the candidate value of the target parameter is used to determine the adjusted target parameter.

59. The communication device according to claim 48, characterized in that The target model monitoring is the event-triggered model monitoring, and the event is associated with one or more of the following: The service type of the service to be transmitted; a model state of the first model; a device status of the first device; The wireless communication environment of the first device; The model switching operation to the first model is successful.

60. The communication device according to claim 59, characterized in that The event is associated with a model state of the first model, If the model state of the first model includes a transmission state of the first model, the event includes that the transmission state of the first model is successfully transmitted; If the model state of the first model includes an activation state of the first model, the event includes that the activation state of the first model is successfully activated.

61. The communication device according to claim 59 or 60, characterized in that: The event is associated with a device state of the first device, If the device state of the first device is used to indicate a movement state of the first device, the event includes a change in the movement state of the first device; If the device state of the first device is used to indicate the resource size of the available resources of the first device, the event includes a change in the resource size of the available resources of the first device.

62. The communication device according to any one of claims 59 to 61, characterized in that: The event is associated with a wireless communication environment of the first device, If the wireless communication environment of the first device includes a wireless communication area of ​​the first device, the event includes a change in the communication area where the first device is located; If the wireless communication environment of the first device includes a wireless communication frequency band of the first device, the event includes a change in the wireless communication frequency band used by the first device; The event includes a change in a wireless communication type of the first device, where the wireless communication type is associated with a wireless communication environment of the first device, and the wireless communication type includes a line-of-sight communication type and a non-line-of-sight communication type.

63. The communication device according to any one of claims 59 to 62, characterized in that: The event is associated with the service type of the service to be transmitted, and the event includes a change in the service type of the service to be transmitted.

64. The communication device according to any one of claims 59 to 63, characterized in that: The communication device further comprises: The second receiving unit is configured to receive first configuration information sent by a second device, where the first configuration information is used to configure the event for the first device.

65. The communication device of claim 48, wherein: The target model monitoring is the model monitoring based on prediction data, and the prediction data is used to perform the target model monitoring on the first model during a silent period of radio resource management RRM measurement.

66. The communication device according to claim 65, characterized in that The measurement data is obtained based on RRM measurement in a first time period, where the first time period is a period of time before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period.

67. The communication device according to any one of claims 48 to 66, characterized in that: The target model monitoring is associated with a first timer, and the duration of monitoring the first model is determined based on the first timer. The first timer is used to indicate a second duration of model monitoring after a first moment, and the first moment is determined based on the moment when an abnormality occurs in the first model.

68. The communication device according to claim 67, characterized in that The first model is one of multiple models deployed by the first device, and different first timers are associated with some or all different models among the multiple models; or The first timer associated with different models among the multiple models is the same.

69. The communication device according to claim 67 or 68, characterized in that The communication device further comprises: The third receiving unit is used to receive second configuration information sent by a second device, where the second configuration information is used to configure the first timer.

70. The communication device according to any one of claims 48 to 69, characterized in that: The communication device further comprises: The fourth receiving unit is used to receive first information sent by the second device, where the first information is used to perform the target model monitoring on the first model.

71. The communication device according to claim 70, characterized in that The first information includes one or more of the following: model information of the first model; Information for indicating whether to perform the target model monitoring; Information used to instruct the target model to monitor the model index of the monitored first model.

72. The communication device according to claim 71, characterized in that The first information includes information for indicating a model index of the first model, and the information of the model index of the first model includes one or more of the following: model information of the first model; Model calling information of the first model; model performance statistics of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; Model input data quality information of the first model.

73. A communication device, characterized in that: The communication device is a second device, comprising: A sending unit sends first information to a first device, where the first information is used to perform target model monitoring on the first model, wherein the target model monitoring includes one of the following: Cycle-based model monitoring; Model monitoring based on event triggering; Model monitoring triggered based on a second device indication; Model monitoring based on predicted data, wherein the predicted data is predicted based on the measured data.

74. The communication device according to claim 73, characterized in that The target model monitoring is the cycle-based model monitoring, and the cycle-based model monitoring is performed based on target parameters, and the target parameters include one or more of the following: A first parameter, the first parameter is used to determine the cycle length of the cycle; A second parameter is used to determine a first duration during which the model monitoring can be performed within the period.

75. The communication device according to claim 74, characterized in that The target parameter includes the first parameter, where the first parameter is used to indicate a first time interval between two adjacent cycles; or The first parameter is used to indicate a basic time interval, and the basic time interval is used to determine the first time interval.

76. The communication device according to claim 75, characterized in that The first parameter is used to indicate the first time interval, and the two adjacent cycles include a first cycle and a second cycle. The first time interval is the time interval between the start time of the first duration of executable model monitoring in the first cycle and the start time of the first duration of executable model monitoring in the second cycle.

77. The communication device according to claim 76, characterized in that The first parameter is used to indicate the basic time interval, and the first time interval is an integer multiple of the basic time interval.

78. The communication device according to any one of claims 74 to 77, characterized in that: The first time interval is the time interval between the start times of two adjacent cycles.

79. The communication device according to any one of claims 74 to 78, characterized in that: The target parameter includes the second parameter, and the second parameter is used to indicate the length of the first duration and / or the start time of the first duration.

80. The communication device according to claim 79, characterized in that The second parameter is used to indicate the start time of the first duration, and the second parameter includes a time offset between the start time of the first duration and the start time of a cycle to which the first duration belongs.

81. The communication device according to any one of claims 74 to 80, characterized in that: The first model is one of a plurality of models deployed in the first device, and the target parameters associated with some or all of the plurality of models are different; or The target parameters associated with different models among the multiple models are the same.

82. The communication device according to any one of claims 74 to 81, characterized in that: The communication device further comprises: A receiving unit, configured to receive a first request sent by a first device, wherein the first request is used to request adjustment of the target parameter associated with the first model; The sending unit is used to send a response message to the first request to the first device, where the response message includes the adjusted target parameter and / or information indicating whether to accept the adjustment of the target parameter based on the first request.

83. The communication device according to claim 82, characterized in that The first request includes one or more of the following: Requirement information for executing the target model monitoring on the first model; The candidate value of the target parameter suggested by the first device, and the candidate value of the target parameter is used to determine the adjusted target parameter.

84. The communication device according to claim 83, characterized in that The target model monitoring is the event-triggered model monitoring, and the event is associated with one or more of the following: The service type of the service to be transmitted; a model state of the first model; a device status of the first device; The wireless communication environment of the first device; The model switching operation to the first model is successful.

85. The communication device according to claim 84, characterized in that The event is associated with a model state of the first model, If the model state of the first model includes a transmission state of the first model, the event includes that the transmission state of the first model is successfully transmitted; If the model state of the first model includes an activation state of the first model, the event includes that the activation state of the first model is successfully activated.

86. The communication device according to claim 84 or 85, characterized in that The event is associated with a device state of the first device, If the device state of the first device is used to indicate a movement state of the first device, the event includes a change in the movement state of the first device; If the device state of the first device is used to indicate the resource size of the available resources of the first device, the event includes a change in the resource size of the available resources of the first device.

87. The communication device according to any one of claims 84 to 86, characterized in that: The event is associated with a wireless communication environment of the first device, If the wireless communication environment of the first device includes a wireless communication area of ​​the first device, the event includes a change in the communication area where the first device is located; If the wireless communication environment of the first device includes a wireless communication frequency band of the first device, the event includes a change in the wireless communication frequency band used by the first device; The event includes a change in a wireless communication type of the first device, where the wireless communication type is associated with a wireless communication environment of the first device, and the wireless communication type includes a line-of-sight communication type and a non-line-of-sight communication type.

88. The communication device according to any one of claims 84 to 87, characterized in that: The event is associated with the service type of the service to be transmitted, and the event includes a change in the service type of the service to be transmitted.

89. The communication device according to any one of claims 84 to 88, characterized in that: The sending unit is further used to: send first configuration information to the first device, where the first configuration information is used to configure the event for the first device.

90. The communication device of claim 73, wherein: The target model monitoring is the model monitoring based on prediction data, and the prediction data is used to perform the target model monitoring on the first model during a silent period of radio resource management RRM measurement.

91. The communication device according to claim 90, characterized in that The measurement data is obtained based on RRM measurement in a first time period, where the first time period is a period of time before the RRM measurement silent period, and the first time period is temporally adjacent to the RRM measurement silent period.

92. The communication device according to any one of claims 73 to 91, characterized in that: The sending unit is further used for: Send second configuration information to the first device, the second configuration information is used to configure a first timer associated with the target model monitoring, the first timer is used to indicate a second duration of model monitoring after a first moment, and the first moment is determined based on the moment when an abnormality occurs in the first model.

93. The communication device according to any one of claims 73 to 92, characterized in that: The first information includes one or more of the following: model information of the first model; Information for indicating whether to perform the target model monitoring; Information used to instruct the target model to monitor the model index of the monitored first model.

94. The communication device according to claim 93, characterized in that The first information includes information for indicating a model index of the first model, and the information of the model index of the first model includes one or more of the following: model information of the first model; Model calling information of the first model; model performance statistics of the first model; model stability information of the first model; model scoring information of the first model; model operation information of the first model; Model input data quality information of the first model.

95. A communication device, characterized in that: It includes a transceiver, a memory and a processor, wherein the memory is used to store programs, and the processor is used to call the programs in the memory and control the transceiver to receive or send signals so that the terminal device executes the method as described in any one of claims 1-47.

96. A device, characterized in that It comprises a processor, which is used to call a program from a memory so that the device executes the method as described in any one of claims 1-47.

97. A chip, characterized in that: It comprises a processor, which is used to call a program from a memory so that a device equipped with the chip executes a method as described in any one of claims 1 to 47.

98. A computer-readable storage medium, characterized in that A program is stored thereon, the program causing a computer to execute the method according to any one of claims 1 to 47.

99. A computer program product, characterized in that A program is included, which causes a computer to execute the method as claimed in any one of claims 1 to 47.

100. A computer program, characterized in that The computer program causes a computer to execute the method according to any one of claims 1 to 47.

Images