AI object management method and device, source node, first node and terminal

Abstract

The invention discloses an AI object management method and device, a source node, a first node and a terminal, and belongs to the technical field of communication, and the AI object management method comprises the steps that in the switching preparation process, the source node sends first information to the first node; wherein the first node comprises a candidate node or a target node, and the first information comprises at least one of the following items: information of an AI object supported by the terminal; information of an AI object currently activated by the terminal; information of the first AI object; the source node collects first data which is not used for AI object training or AI object monitoring; the source node collects second data which are not used for AI object reasoning; the application range of AI object reasoning of the terminal for AI object reasoning; an RRM measurement result; a target prediction result; monitoring a result of the AI object; the AI object comprises an AI model or an AI function, and the first AI object is an AI object which is stored by the source node and is suitable for the terminal.

Description

Technical Field

[0001] This application belongs to the field of communication technologies, and particularly relates to an AI object management method, apparatus, source node, first node, and terminal. Background Art

[0002] Currently, in a mobile communication network, tasks can be performed or services can be provided based on Artificial Intelligence (AI). For example, mobility management-related processes such as radio resource management (RRM) measurement prediction and event prediction can be performed based on an AI model or AI function. However, in the related art, there is no corresponding solution for how to manage an AI model or AI function in the case of cell handover, which can easily lead to difficulties for the terminal to correctly and effectively perform inference based on the AI model or AI function after cell handover. Summary of the Invention

[0003] Embodiments of this application provide an AI object management method, apparatus, source node, first node, and terminal, which can interact information related to an AI object between the source node and the first node in the case of cell handover, so that the target cell can more accurately manage the AI object related to the terminal based on the information related to the AI object after cell handover.

[0004] In a first aspect, an AI object management method is provided. The method includes:

[0005] During the handover preparation process, the source node sends first information to the first node;

[0006] Wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0007] Information about the AI objects supported by the terminal;

[0008] Information about the AI objects currently activated by the terminal;

[0009] Information about the first AI object;

[0010] First data collected by the source node that has not been used for AI object training or AI object monitoring;

[0011] Second data collected by the source node that has not been used for AI object inference;

[0012] The applicable scope for the terminal to perform AI object inference;

[0013] Radio Resource Management (RRM) measurement results;

[0014] Target prediction results;

[0015] Monitoring results of the AI object;

[0016] The AI object includes an AI model or an AI function. The first AI object is the AI object applicable to the terminal saved by the source node. The target prediction result includes at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object.

[0017] In a second aspect, an AI object management device is provided, and the device includes:

[0018] A sending module, configured to send first information to a first node during a handover preparation process;

[0019] Wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0020] Information on AI objects supported by the terminal;

[0021] Information on the AI object currently activated by the terminal;

[0022] Information on the first AI object;

[0023] First data collected by the source node that has not been used for AI object training or AI object monitoring;

[0024] Second data collected by the source node that has not been used for AI object inference;

[0025] Scope of application for the terminal to perform AI object inference;

[0026] Radio Resource Management (RRM) measurement results;

[0027] Target prediction results;

[0028] Monitoring results of the AI object;

[0029] The AI object includes an AI model or an AI function. The first AI object is the AI object applicable to the terminal saved by the source node. The target prediction result includes at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object.

[0030] In a third aspect, an AI object management method is provided, and the method includes:

[0031] During a handover preparation process, a first node receives first information sent by a source node;

[0032] Wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0033] Information on AI objects supported by the terminal;

[0034] Information of the currently activated AI object of the terminal;

[0035] Information of the first AI object;

[0036] The first data collected by the source node that has not been used for AI object training or AI object monitoring;

[0037] The second data collected by the source node that has not been used for AI object inference;

[0038] The applicable scope of the terminal for AI object inference;

[0039] Radio Resource Management (RRM) measurement results;

[0040] Target prediction results;

[0041] Monitoring results of the AI object;

[0042] The AI object includes an AI model or an AI function. The first AI object is the AI object suitable for the terminal saved by the source node. The target prediction results include at least one of the RRM measurement prediction results based on the AI object and the event prediction results based on the AI object.

[0043] In a fourth aspect, an AI object management device is provided. The device includes:

[0044] A receiving module, configured to receive first information sent by a source node during a handover preparation process;

[0045] Wherein, the first information includes at least one of the following:

[0046] Information of the AI objects supported by the terminal;

[0047] Information of the currently activated AI object of the terminal;

[0048] Information of the first AI object;

[0049] The first data collected by the source node that has not been used for AI object training or AI object monitoring;

[0050] The second data collected by the source node that has not been used for AI object inference;

[0051] The applicable scope of the terminal for AI object inference;

[0052] Radio Resource Management (RRM) measurement results;

[0053] Target prediction results;

[0054] Monitoring results of the AI object;

[0055] The AI object includes an AI model or an AI function. The first AI object is the AI object applicable to the terminal saved by the source node. The target prediction result includes at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object.

[0056] In a fifth aspect, an AI object management method is provided. The method includes:

[0057] When a handover is performed or the handover is completed, the terminal performs a second operation;

[0058] Wherein, the second operation includes at least one of the following:

[0059] Deactivate the AI object activated in the source cell;

[0060] Release the AI object from the source cell;

[0061] Release the AI object only applicable to the source cell;

[0062] Release the AI object beyond the applicable scope;

[0063] Release the acquisition data corresponding to the second AI object, where the second AI object is the AI object deactivated or released by the terminal;

[0064] When the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is the AI object that remains activated when the terminal switches from the source cell to the target cell;

[0065] The AI object includes an AI model or an AI function.

[0066] In a sixth aspect, an AI object management device is provided. The device includes:

[0067] An execution module, configured to perform a second operation when a handover is performed or the handover is completed;

[0068] Wherein, the second operation includes at least one of the following:

[0069] Deactivate the AI object activated in the source cell;

[0070] Release the AI object from the source cell;

[0071] Release the AI object only applicable to the source cell;

[0072] Release the AI object beyond the applicable scope;

[0073] Release the collected data corresponding to the second AI object, where the second AI object is an AI object that has been deactivated or released by the terminal;

[0074] In the case where the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is an AI object that remains active when the terminal switches from the source cell to the target cell;

[0075] The AI object includes an AI model or an AI function.

[0076] In a seventh aspect, a source node is provided. The source node includes a processor and a memory. The memory stores a program or instructions that can be run on the processor. When the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.

[0077] In an eighth aspect, a source node is provided, including a processor and a communication interface. Among them, the communication interface is used to send first information to a first node during the handover preparation process; among them, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0078] Information about the AI objects supported by the terminal;

[0079] Information about the AI objects currently activated by the terminal;

[0080] Information about the first AI object;

[0081] The first data collected by the source node that has not been used for AI object training or AI object monitoring;

[0082] The second data collected by the source node that has not been used for AI object inference;

[0083] The applicable scope of the terminal for AI object inference;

[0084] Radio Resource Management (RRM) measurement results;

[0085] Target prediction results;

[0086] Monitoring results of AI objects;

[0087] The AI object includes an AI model or an AI function. The first AI object is an AI object suitable for the terminal saved by the source node. The target prediction results include at least one of the RRM measurement prediction result based on the AI object and the event prediction result based on the AI object.

[0088] In a ninth aspect, a first node is provided, which includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the programs or instructions are executed by the processor, the steps of the method described in the third aspect are implemented.

[0089] In a tenth aspect, a first node is provided, including a processor and a communication interface. Wherein, the communication interface is used to receive first information sent by a source node during a handover preparation process. Wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0090] Information on AI objects supported by the terminal;

[0091] Information on the currently activated AI objects of the terminal;

[0092] Information on the first AI object;

[0093] First data collected by the source node that has not been used for AI object training or AI object monitoring;

[0094] Second data collected by the source node that has not been used for AI object inference;

[0095] Scope of application for the terminal to perform AI object inference;

[0096] Radio Resource Management (RRM) measurement results;

[0097] Target prediction results;

[0098] Monitoring results of AI objects;

[0099] The AI object includes an AI model or an AI function. The first AI object is an AI object suitable for the terminal saved by the source node. The target prediction results include at least one of the RRM measurement prediction results based on the AI object and the event prediction results based on the AI object.

[0100] In an eleventh aspect, a terminal is provided, which includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the programs or instructions are executed by the processor, the steps of the method described in the fifth aspect are implemented.

[0101] In a twelfth aspect, a terminal is provided, including a processor and a communication interface. Wherein, the processor is used to perform a second operation when a handover or handover completion is executed. Wherein, the second operation includes at least one of the following:

[0102] Deactivate the AI objects activated in the source cell;

[0103] Release the AI objects from the source cell;

[0104] Release the AI objects applicable only to the source cell;

[0105] Release the AI objects beyond the applicable scope;

[0106] Release the acquisition data corresponding to the second AI object, where the second AI object is the AI object that has been deactivated or released by the terminal;

[0107] In the case where the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is the AI object that remains active when the terminal switches from the source cell to the target cell;

[0108] The AI object includes an AI model or an AI function.

[0109] In a thirteenth aspect, there is provided an AI object management system, including: a source node, a first node, and a terminal. The source node can be used to execute the steps of the AI object management method as described in the first aspect, the first node can be used to execute the steps of the AI object management method as described in the third aspect, and the terminal can be used to execute the steps of the AI object management method as described in the fifth aspect.

[0110] In a fourteenth aspect, there is provided a readable storage medium, on which a program or instructions are stored. When the program or instructions are executed by a processor, the steps of the method as described in the first aspect, or the steps of the method as described in the third aspect, or the steps of the method as described in the fifth aspect are implemented.

[0111] In a fifteenth aspect, there is provided a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run a program or instructions to implement the steps of the method as described in the first aspect, or the steps of the method as described in the third aspect, or the steps of the method as described in the fifth aspect.

[0112] In a sixteenth aspect, there is provided a computer program / program product. The computer program product includes a computer program or computer instructions. The computer program or computer instructions are executed by at least one processor to implement the steps of the method as described in the first aspect, or the steps of the method as described in the third aspect, or the steps of the method as described in the fifth aspect.

[0113] In an embodiment of the present application, during the handover preparation process, a source node sends first information to a first node; wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following: information on AI objects supported by a terminal; information on currently activated AI objects of the terminal; information on a first AI object; first data collected by the source node that has not been used for AI object training or AI object monitoring; second data collected by the source node that has not been used for AI object inference; the applicable scope of the terminal for performing AI object inference; RRM measurement results; target prediction results; monitoring results of AI objects; the AI objects include AI models or AI functions, the first AI object is an AI object suitable for the terminal saved by the source node, and the target prediction results include at least one of an RRM measurement prediction result based on an AI object and an event prediction result based on an AI object. That is, in an embodiment of the present application, relevant information of AI objects is exchanged between the source node and the first node in the case of cell handover, so that after the cell handover, the target cell can more accurately manage the AI objects related to the terminal, which is conducive to ensuring that the terminal can perform inference more accurately and effectively based on the AI objects. Description of the Drawings

[0114] Figure 1 is a block diagram of a wireless communication system to which an embodiment of the present application can be applied;

[0115] Figure 2a is a schematic structural diagram of a neural network provided by an embodiment of the present application;

[0116] Figure 2b is a schematic structural diagram of a neuron provided by an embodiment of the present application;

[0117] Figure 3 is a schematic diagram of an AI / ML functional architecture provided by an embodiment of the present application;

[0118] Figure 4 is a flowchart of cell handover provided by an embodiment of the present application;

[0119] Figure 5 is a flowchart of an AI object management method provided by an embodiment of the present application;

[0120] Figure 6 is a flowchart of another AI object management method provided by an embodiment of the present application;

[0121] Figure 7 is a flowchart of yet another AI object management method provided by an embodiment of the present application;

[0122] Figure 8 is a structural diagram of an AI object management device provided by an embodiment of the present application;

[0123] Figure 9 It is the structural diagram of another AI object management device provided by an embodiment of the present application;

[0124] Figure 10 It is the structural diagram of yet another AI object management device provided by an embodiment of the present application;

[0125] Figure 11 It is the structural diagram of a communication device provided by an embodiment of the present application;

[0126] Figure 12 It is the structural diagram of a source node provided by an embodiment of the present application.

[0127] Figure 13 It is the structural diagram of a first node provided by an embodiment of the present application.

[0128] Figure 14 It is the structural diagram of a terminal provided by an embodiment of the present application. Detailed implementation manners

[0129] Next, the technical solutions in the embodiments of the present application will be clearly described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0130] The terms "first", "second", etc. in the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are usually of the same type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "or" in the present application means at least one of the connected objects. For example, "A or B" covers three scenarios, namely, Scenario 1: including A and not including B; Scenario 2: including B and not including A; Scenario 3: including both A and B. The character " / " generally indicates an "or" relationship between the associated objects before and after.

[0131] The term "indicate" in the present application can be either a direct indication (or an explicit indication) or an indirect indication (or an implicit indication). Among them, a direct indication can be understood as that the sender clearly informs the receiver of specific information, operations to be performed, or request results, etc. in the sent indication; an indirect indication can be understood as that the receiver determines the corresponding information according to the indication sent by the sender, or makes a judgment and determines the operations to be performed or request results, etc. according to the judgment result.

[0132] It should be noted that the technology described in the embodiments of the present application is not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, and can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA), or other systems. The terms "system" and "network" in the embodiments of the present application are often used interchangeably, and the described technology can be used in the above-mentioned systems and radio technologies, as well as in other systems and radio technologies. The following description describes the New Radio (NR) system for example purposes, and the NR term is used in most of the following descriptions, but these technologies can also be applied to systems other than the NR system, such as the 6th Generation (6 th Generation, 6G) communication system.

[0133] Figure 1The block diagram of a wireless communication system to which the embodiments of the present application can be applied is shown. The wireless communication system includes a terminal 11 and a network-side device 12. Among them, the terminal 11 can be a mobile phone, a tablet personal computer, a laptop computer, a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR), a virtual reality (VR) device, a robot, a wearable device, a flight vehicle, a vehicle user equipment (VUE), a shipborne device, a pedestrian user equipment (PUE), a smart home (home appliances with wireless communication functions, such as refrigerators, TVs, washing machines, or furniture, etc.), a game console, a personal computer (PC), a teller machine, or a self-service machine, etc., which are terminal-side devices. Wearable devices include: smart watches, smart bracelets, smart earphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart ankle chains, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle user equipment can also be referred to as a vehicle terminal, a vehicle controller, a vehicle module, a vehicle component, a vehicle chip, or a vehicle unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiments of the present application.

[0134] The network-side device 12 may include an access network device or a core network device. Among them, the access network device may also be referred to as a radio access network (RAN) device, a radio access network function, or a radio access network unit. The access network device may include a base station, a wireless local area network (WLAN) access point (AP), or a wireless fidelity (WiFi) node, etc. Among them, the base station may be referred to as Node B (NB), Evolved Node B (eNB), next generation Node B (gNB), New Radio Node B (NR Node B), access point, relay base station (RBS), serving base station (SBS), base transceiver station (BTS), radio base station, radio transceiver, basic service set (BSS), extended service set (ESS), home Node B (HNB), home evolved Node B, transmission reception point (TRP), or some other suitable term in the art. As long as the same technical effect is achieved, the base station is not limited to a specific technical term. It should be noted that in the embodiments of this application, only the base station in the NR system is taken as an example for introduction, and the specific type of the base station is not limited.

[0135] The core network device may include but is not limited to at least one of the following: core network node, core network function, Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (L-NEF), Binding Support Function (BSF), Application Function (AF), etc. It should be noted that in the embodiments of this application, only the core network devices in the NR system are taken as examples for introduction, and the specific types of core network devices are not limited.

[0136] It should be noted that the source node provided in the embodiments of this application may be the above-mentioned access network device, and the first node provided in the embodiments of this application may also be the above-mentioned access network device.

[0137] For the convenience of understanding, some contents related to the embodiments of this application are described below:

[0138] I. Artificial Intelligence (AI)

[0139] Artificial intelligence (AI) has currently been widely applied in various fields. Incorporating artificial intelligence into wireless communication networks to significantly improve technical indicators such as throughput, latency, and user capacity is an important task for future wireless communication networks. There are various implementation methods for the AI module, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc. This application takes neural networks as an example for illustration, but does not limit the specific type of the AI module.

[0140] Exemplarily, a neural network can be as Figure 2a shown. The neural network is composed of neurons, and each neuron can be as Figure 2b shown. Among them, a1, a2, …, aK are inputs, w is the weight (multiplicative coefficient), b is the bias (additive coefficient), and σ(.) is the activation function. Common activation functions include Sigmoid, tanh, Rectified Linear Unit (ReLU), etc.

[0141] The parameters of the neural network are optimized through gradient optimization algorithms. Gradient optimization algorithms are a class of algorithms that minimize or maximize the objective function (also known as the loss function), and the objective function is often a mathematical combination of model parameters and data. For example, given data X and its corresponding label Y, we construct a neural network model f(.). After having the model, according to the input x, we can obtain the predicted output f(x), and we can calculate the gap between the predicted value and the true value (f(x) - Y), which is the loss function. The goal is to find the appropriate W and b to minimize the value of the above loss function. Among them, the smaller the loss value, the closer the model is to the real situation.

[0142] Currently, common optimization algorithms are basically based on the error Back Propagation (BP) algorithm. The basic idea of the BP algorithm is that the learning process consists of two processes: the forward propagation of signals and the backward propagation of errors. During forward propagation, the input samples are input from the input layer, and after being processed layer by layer through each hidden layer, they are transmitted to the output layer. If the actual output of the output layer does not match the expected output, it will enter the stage of backward propagation of errors. The error backpropagation is to transmit the output error back to the input layer layer by layer through the hidden layer in a certain form, and distribute the error to all units of each layer, so as to obtain the error signals of each layer of units. This error signal is used as the basis for correcting the weights of each unit. This process of adjusting the weights of each layer in the forward propagation of signals and the backward propagation of errors is carried out cyclically. The process of continuously adjusting the weights is also the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until a pre-set number of learning times is reached.

[0143] Common optimization algorithms include Gradient Descent, Stochastic Gradient Descent (SGD), mini-batch gradient descent, Momentum, Nesterov (specifically Stochastic Gradient Descent with Momentum), Adaptive Gradient descent (Adagrad), Adadelta, Root Mean Square Propagation (RMSprop), Adaptive Moment Estimation (Adam), etc.

[0144] When these optimization algorithms perform backpropagation of errors, they all obtain the gradient by taking the derivative / partial derivative of the current neuron with respect to the error / loss obtained from the loss function, adding the learning rate, previous gradients / derivatives / partial derivatives, etc., and then passing the gradient to the previous layer.

[0145] II. AI Unit / AI Model

[0146] The AI unit / AI model in the embodiments of the present application can also be referred to as a Machine Learning (ML) model, ML unit, AI structure, AI function, AI feature, machine learning model, neural network, neural network function, neural network capability, etc. Alternatively, the above AI unit / AI model can also refer to a processing unit that can implement specific algorithms, formulas, processing flows, capabilities, etc. related to AI, or the AI unit / AI model can be a processing method, algorithm, function, module, or unit for a specific data set, or the AI unit / AI model can be a processing method, algorithm, function, module, or unit running on AI / ML-related hardware such as GPUs, NPUs, TPUs, ASICs, etc. The embodiments of the present application do not make specific limitations in this regard. Optionally, the specific data set includes at least one of the input and output of the AI unit / AI model.

[0147] Optionally, the identifier of the AI unit / AI model can be an AI model identifier, an AI structure identifier, an AI algorithm identifier, or the identifier of a specific data set associated with the AI unit / AI model, or the identifier of a specific scenario, environment, channel feature, device related to AI / ML, or the identifier of a function, feature, capability, or module related to AI / ML. The embodiments of the present application do not make specific limitations in this regard.

[0148] AI functionality: That is, an AI algorithm function that can include multiple AI models.

[0149] III. AI / ML Architecture (Framework)

[0150] The air interface AI project of Release 18 studied AI / ML frameworks, such as Figure 3 shown below, and the main processes are as follows:

[0151] Data Collection: Responsible for providing input data to Model Training, Management, and Inference;

[0152] Model Training: Responsible for performing AI / ML model training, verification, and testing. Also responsible for data preparation, i.e., data preprocessing, conversion to a specific format, etc.;

[0153] Management: Responsible for model selection / activation / deactivation / switching / rollback, etc.;

[0154] Inference: Responsible for providing the output after applying the AI / ML model or AI / ML function;

[0155] Model Storage: Responsible for saving the trained / updated model

[0156] Model Transfer / Delivery: Responsible for delivering the AI / ML model to the inference function node.

[0157] IV. Layer 3 Handover Process

[0158] As Figure 4 shown below, the main steps of layer 3 handover include:

[0159] (1) The source base station sends a measurement configuration to the UE;

[0160] (2) The UE performs measurements according to the measurement configuration and reports a measurement report when the conditions are met;

[0161] (3) The source base station interacts with the target base station for handover preparation;

[0162] (4) The UE receives the handover command sent by the source base station, immediately disconnects from the source cell, and accesses the target cell.

[0163] V. RRM Measurement Report

[0164] The measurement configuration mainly consists of measurement objects, reporting configuration, and measurement identifier (ID);

[0165] Measurement Object: That is, the frequency point to be measured

[0166] ReportConfig: Includes reporting criteria (periodic / event-triggered); reference signal type (Synchronous Signal Block (SSB) / Channel State Information Reference Signal (CSI-RS)), measurement reporting quantities (any combination of Reference Signal Receiving Power (RSRP) / Reference Signal Received Quality (RSRQ) / Signal to Interference Plus Noise Ratio (SINR)); whether to report beam measurement results, the maximum number of beams that can be reported, etc.;

[0167] Measurement identity (measId): Used to associate a measurement object and a report configuration. One measurement object can be associated with multiple report configurations, and one report configuration can be associated with multiple measurement objects.

[0168] In NR, the three are associated together in the following way:

[0169]

[0170] The report configuration can include event-triggered reporting. The events defined in NR can be seen in Table 1, including the following events:

[0171] Table 1

[0172]

[0173]

[0174] Taking event A3 as an example, the meanings of the parameters for the entry condition and the departure condition are as follows:

[0175] Mn: Neighboring cell measurement result, without considering any offset;

[0176] Ofn: Neighboring cell measurement object specific offset;

[0177] Ocn: Neighboring cell specific offset at the cell level;

[0178] Mp: Serving cell (SpCell) measurement result, without considering any offset;

[0179] Ofp: Specific offset of the SpCell measurement object;

[0180] Ocp: Specific offset at the SpCell cell level;

[0181] Hys: Hysteresis parameter of the event;

[0182] Off: Offset parameter of the event.

[0183] If the reporting type is event-triggered reporting, in order to avoid frequent reporting or ping-pong handover, the base station configures the trigger time (timeToTrigger) parameter for each event. If the L3 filtered signal quality of one or more candidate cells meets the entry conditions of the event within the timeToTrigger time, a measurement report is triggered;

[0184] For conditional handover, the UE uses the cell that meets the conditions as the trigger cell and selects one to perform conditional reconfiguration in the trigger cell.

[0185] VI. Conditional Handover

[0186] Conditional handover means that the network pre-configures multiple candidate cells for the UE. The UE evaluates the execution conditions of the candidate cells and switches to the corresponding cell when the conditions are met. The execution conditions may include one or two trigger conditions.

[0187] Taking the parameters in NR as an example:

[0188] condReconfigId indicates the conditional reconfiguration ID; condExecutionCond is used to configure the execution conditions for the change of the Primary Cell (PCell), and condRRCReconfig is used to configure the configuration parameters of the target Master CellGroup (MCG). The three parameters correspond to a set of candidate cell configurations for conditional reconfiguration of a candidate PCell, as follows:

[0189]

[0190] The above execution conditions are the measurement events configured in the reporting configuration associated with MeasId. For conditional handover, the measurement events support conditional handover event A3 (condEventA3), conditional handover event A4 (condEventA4), or conditional handover event A5 (condEventA5), and their judgment conditions are the same as those of A3, A4, and A5 above. If the L3 filtered signal quality of one or more candidate cells meets the entry conditions of the event within the timeToTrigger time, the UE uses the cell that meets the conditions as the trigger cell and selects one to perform conditional reconfiguration in the trigger cell.

[0191] The AI object management method provided by the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings through some embodiments and their application scenarios.

[0192] Please refer to Figure 5 , Figure 5 which is a flowchart of an AI object management method provided by the embodiments of the present application. This method can be executed by a source node, as Figure 5 shown, and includes the following steps:

[0193] Step 501: During the handover preparation process, the source node sends first information to the first node;

[0194] Among them, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0195] Information about AI objects supported by the terminal;

[0196] Information about the currently activated AI object of the terminal;

[0197] Information about the first AI object;

[0198] The first data collected by the source node that has not been used for AI object training or AI object monitoring;

[0199] The second data collected by the source node that has not been used for AI object inference;

[0200] The applicable scope of the terminal for AI object inference;

[0201] Radio Resource Management (RRM) measurement results;

[0202] Target prediction results;

[0203] Monitoring results of AI objects;

[0204] The AI object includes an AI model or an AI function. The first AI object is an AI object suitable for the terminal saved by the source node. The target prediction results include at least one of the RRM measurement prediction result based on the AI object and the event prediction result based on the AI object.

[0205] In this embodiment, the above handover preparation process refers to the process in which the source node requests the target node or candidate node to provide the target cell or candidate cell configuration for the terminal. Exemplarily, the above first information may be carried in a handover request message, a UE context setup request message, a UE context modification request message, etc.

[0206] For the AI objects supported by the terminal, it can be understood that the terminal is capable of reasoning based on the AI object or the terminal has the ability to use the AI object for reasoning, etc. In some optional embodiments, the terminal may report the AI objects supported by the terminal to the source cell through UE capability information or UE assistance information.

[0207] For the AI object currently activated by the terminal, the terminal uses the AI object for reasoning. In some optional embodiments, the above AI object currently activated by the terminal may be indicated by the source cell for the terminal to activate, or determined by the terminal itself and then notified to the source cell after activation.

[0208] The above first AI object is the AI object applicable to the terminal saved by the source node. The AI object applicable to the terminal can be understood as the AI object applicable to RRM measurement prediction or event prediction for the terminal. Specifically, the source node transmits the information of the AI object applicable to the current terminal saved by it to the candidate node or target node, so that the target node can obtain the corresponding AI object.

[0209] It should be noted that the information of the above AI object may include, but is not limited to, at least one of the identification of the AI object, the description information of the AI object, the file information of the AI object, the input of the AI object, and the output of the AI object. Among them, for the identification of the above AI object, if the AI object is an AI model, the identification of the above AI object may include at least one of the identification of the AI model and the AI function identification corresponding to the AI model; if the above AI object is an AI function, the identification of the above AI object may be the identification of the AI function. The description information of the above AI object may include, but is not limited to, the usage information of the AI object (for example, the validity period, the effective range, etc. of the AI object), the size and other description information of the AI object. The file information of the above AI object may include, but is not limited to, the parameter information of the AI object, the structure or architecture information of the AI object, etc.

[0210] The first data collected by the above-mentioned source node that has not been used for AI object training or AI object monitoring. For example, the source node collects at least one of the input and output of the AI object. If it is determined that a cell handover is required before using at least one of the collected input and output of the AI object for AI object training or AI object monitoring, the source node can send at least one of the collected input and output of the AI object to the candidate node or the target node. In this way, after the cell handover, the target node can perform AI object training or AI object monitoring based on at least one of the input and output of the AI object collected by the source node.

[0211] The second data collected by the above-mentioned source node that has not been used for AI object inference. For example, the source node collects the input of the AI object. If it is determined that a cell handover is required before using the collected input of the AI object for AI object inference, the source node can send the collected input of the AI object to the candidate node or the target node. In this way, after the cell handover, the target node can perform AI object inference based on the input of the AI object collected by the source node. Herein, performing AI object inference based on the input of the AI object collected by the source node can be understood as the AI object performing inference based on the input of this AI object collected by the source node.

[0212] The applicable scope of the above-mentioned terminal for performing AI object inference can be understood as the applicable scope of the terminal for performing inference based on the AI object. For example, the frequency band scope, cell scope, Tracking Area (TA) scope, or slice scope, etc. of the above-mentioned terminal for performing AI object inference. Taking the cell scope as an example, the AI object activated by the terminal may be applicable to multiple cells, but the terminal only performs inference based on this AI object for some cells. Therefore, the candidate node or the target node can configure candidate cells or target cells within the applicable scope of the terminal for performing inference based on this AI object by obtaining the applicable scope of the above-mentioned terminal for performing AI object inference.

[0213] The above-mentioned target prediction result can include at least one of the RRM measurement prediction result based on the AI object and the event prediction result based on the AI object. Herein, the RRM measurement prediction result based on the AI object can be understood as the result obtained by performing RRM measurement prediction based on the AI object, and the event prediction result based on the AI object can be understood as the result obtained by performing event prediction based on the AI object. Exemplarily, the above-mentioned event prediction result can include at least one of the prediction results of measurement events, the prediction result of Radio Link Failure (RLF), the prediction result of Handover Failure (HOF), etc.

[0214] The monitoring result of the above AI object can be understood as the monitoring result obtained by monitoring the AI object. Exemplarily, the AI object can be monitored based on the monitoring metric parameters configured by the network-side device or predefined by the protocol. Among them, the above monitoring metric parameters include at least one of the RRM measurement prediction accuracy, the prediction accuracy of the optimal cell, the prediction accuracy of the optimal beam, the prediction accuracy of the HOF, the prediction accuracy of the RLF, the prediction accuracy of the measurement event, the number or probability of abnormal events occurring within the first time period, the handover delay or interruption duration within the second time period, the number of handovers per unit time or within the third time period, etc.

[0215] In this embodiment, the source node sends the first information to the first node. Then, the first node can configure the target cell or candidate cell for the terminal based on the first information, or the first node can perform operations such as AI object training, AI object monitoring, AI object inference, and AI object management based on the first information. The following is an example of this embodiment.

[0216] The source node indicates the AI object information supported by the terminal to the first node. In this way, the first node can provide indication information for AI object management operations in the candidate cell or target cell configuration based on the AI object information supported by the terminal. For example, the selection, activation, deactivation, handover, or fallback of the AI object, etc. Then, after the handover is completed, the terminal can quickly perform the AI object management operations indicated by the first node based on the indication information.

[0217] The source node indicates the currently activated AI object information of the terminal to the first node. In this way, the first node can provide indication information for AI object management operations in the candidate cell or target cell configuration based on the currently activated AI object information of the terminal. For example, the selection, activation, deactivation, handover, or fallback of the AI object, etc. Then, after the handover is completed, the terminal can quickly perform the AI object management operations indicated by the first node based on the indication information.

[0218] The source node indicates the information of the first AI object to the first node. In this way, after the terminal switches to the target cell, the first node can use the first AI object provided by the source node to perform AI object inference on the terminal to continue to support functions such as measurement prediction or event prediction.

[0219] The source node sends the first data collected that has not been used for AI object training or AI object monitoring to the first node. In this way, the first node can use the first data for AI object training or AI object monitoring, which is beneficial to making full use of the data collected at the source node and avoiding repeated collection. In addition, the first node using the first data that has not been used for AI object monitoring for AI object monitoring is also beneficial to ensuring the accuracy of the prediction results of the AI object.

[0220] The source node sends the second data collected that has not been used for AI object inference to the first node, so that the first node can use the second data for AI object inference to obtain a prediction result. That is, the first node can obtain a prediction result using the data collected at the source node, which is beneficial to optimizing the performance of the terminal in the target cell.

[0221] The source node sends the applicable range of the terminal for AI object inference to the first node, so that the first node can determine the applicable range of the terminal's AI object inference, and thus adjust the applicable range of the AI object according to its own needs. For example, if the original applicable range is large and the first node does not need to perform inference within such a large range, the first node can reduce the applicable range; or the first node instructs the terminal to delete / release the AI object. For example, if the first node is not within the applicable range, the first node can instruct the terminal to release the AI object.

[0222] The source node sends the RRM measurement result to the first node, so that the first node can judge the target cell / beam quality based on the RRM measurement result to provide a suitable configuration.

[0223] The source node sends the target prediction result to the first node, so that the first node can judge the target cell / beam quality based on the target prediction result to provide a suitable configuration.

[0224] The source node sends the monitoring result of the AI object to the first node, so that the first node can judge the effectiveness of the AI object according to the target prediction result and decide on AI object management operations, such as the selection, activation, deactivation, handover or fallback of the AI object.

[0225] In the embodiments of the present application, during the handover preparation process, the source node sends the first information to the first node; wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following: information on the AI objects supported by the terminal; information on the currently activated AI objects of the terminal; information on the first AI object; the first data collected by the source node that has not been used for AI object training or AI object monitoring; the second data collected by the source node that has not been used for AI object inference; the applicable range of the terminal for AI object inference; the RRM measurement result; the target prediction result; the monitoring result of the AI object; the AI object includes an AI model or an AI function, the first AI object is the AI object applicable to the terminal saved by the source node, and the target prediction result includes at least one of the RRM measurement prediction result based on the AI object and the event prediction result based on the AI object. That is, in the embodiments of the present application, during cell handover, relevant information of the AI object is exchanged between the source node and the first node, so that after cell handover, the target cell can more accurately manage the AI object related to the terminal, which is beneficial to ensuring that the terminal can perform inference more accurately and effectively based on the AI object.

[0226] Optionally, the information of the AI object supported by the terminal includes at least one of the following: the identifier of the AI object supported by the terminal, the file information of the AI object supported by the terminal, the input of the AI object supported by the terminal, the output of the AI object supported by the terminal; wherein, the file information of the AI object supported by the terminal includes at least one of the parameter information and the structure information of the AI object supported by the terminal;

[0227] Or,

[0228] The information of the currently activated AI object of the terminal includes at least one of the following: the identifier of the currently activated AI object of the terminal, the file information of the currently activated AI object of the terminal, the input of the currently activated AI object of the terminal, the output of the currently activated AI object of the terminal; wherein, the file information of the currently activated AI object of the terminal includes at least one of the parameter information and the structure information of the currently activated AI object of the terminal;

[0229] Or,

[0230] The information of the first AI object includes at least one of the following: the identifier of the first AI object, the file information of the first AI object, the input of the first AI object, the output of the first AI object.; wherein, the file information of the first AI object includes at least one of the parameter information and the structure information of the first AI object.

[0231] For the identifier of the AI object supported by the terminal above, if the AI object supported by the terminal above is an AI model, the identifier of the AI object supported by the terminal above may include at least one of the identifier of the AI model supported by the terminal and the AI function identifier corresponding to the AI model supported by the terminal; if the AI object supported by the terminal above is an AI function, the identifier of the AI object supported by the terminal above may include the identifier of the AI function supported by the terminal.

[0232] The file information of the AI object supported by the terminal above may include at least one of the parameter information of the AI object supported by the terminal, the structure or architecture information of the AI object supported by the terminal, etc.

[0233] The input of the AI object supported by the terminal above may include, but is not limited to, at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell.

[0234] The output of the AI object supported by the terminal above may include, but is not limited to, at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether RLF occurs, the flag indicating whether HOF occurs.

[0235] For the identifier of the currently activated AI object of the above terminal, if the currently activated AI object of the above terminal is an AI model, the identifier of the currently activated AI object of the above terminal may include at least one of the identifier of the currently activated AI model of the terminal and the AI function identifier corresponding to the currently activated AI model of the terminal; if the currently activated AI object of the above terminal is an AI function, the identifier of the currently activated AI object of the above terminal may include the identifier of the currently activated AI function of the terminal.

[0236] The file information of the currently activated AI object of the above terminal may include at least one of the parameter information of the currently activated AI object of the terminal, the structure or architecture information of the currently activated AI object of the terminal, etc.

[0237] The input of the currently activated AI object of the above terminal may include, but is not limited to, at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, and the beam signal quality of the neighboring cell.

[0238] The output of the currently activated AI object of the above terminal may include, but is not limited to, at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether RLF occurs, and the flag indicating whether HOF occurs.

[0239] For the identifier of the first AI object, if the first AI object is an AI model, the identifier of the first AI object may include at least one of the identifier of the first AI model and the AI function identifier corresponding to the first AI model; if the first AI object is an AI function, the identifier of the first AI object may include the identifier of the first AI function.

[0240] The file information of the first AI object may include at least one of the parameter information of the first AI object, the structure or architecture information of the first AI object, etc.

[0241] The input of the first AI object may include, but is not limited to, at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, and the beam signal quality of the neighboring cell.

[0242] The output of the first AI object may include, but is not limited to, at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether RLF occurs, and the flag indicating whether HOF occurs.

[0243] Optionally, the applicable scope of the terminal for AI object inference includes at least one of the following: the cell range for the terminal to perform AI object inference, and the frequency band range for the terminal to perform AI object inference.

[0244] The cell range for the above terminal to perform AI object inference can be understood as that the terminal only performs inference based on the AI object for the cells within this cell range. The frequency range for the above terminal to perform AI object inference can be understood as that the terminal only performs inference based on the AI object for the frequencies within this frequency range.

[0245] Optionally, the first data includes at least one of the input of the AI object and the output of the AI object;

[0246] Or,

[0247] The second data includes the input of the AI object.

[0248] Optionally, the input of the AI object includes at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell;

[0249] Or,

[0250] The output of the AI object includes at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether radio link failure (RLF) occurs, the flag indicating whether handover failure (HOF) occurs.

[0251] Optionally, the RRM measurement result includes at least one of the following:

[0252] The measured value of the cell signal quality of the serving cell;

[0253] The measured value of the beam signal quality of the serving cell;

[0254] The measured value of the cell signal quality of the candidate cell;

[0255] The measured value of the beam signal quality of the candidate cell;

[0256] The measured value of the cell signal quality of the target cell;

[0257] The measured value of the beam signal quality of the target cell;

[0258] The identifier of the optimal beam of the measured candidate cell;

[0259] The identifier of the optimal beam of the measured target cell.

[0260] It can be understood that the above RRM measurement result can be understood as the measurement result obtained by the terminal performing RRM measurement.

[0261] Optionally, the RRM measurement prediction result includes at least one of the following:

[0262] Predicted value of the cell signal quality of the serving cell;

[0263] Predicted value of the beam signal quality of the serving cell;

[0264] Predicted value of the cell signal quality of the candidate cell;

[0265] Predicted value of the beam signal quality of the candidate cell;

[0266] Predicted value of the cell signal quality of the target cell;

[0267] Predicted value of the beam signal quality of the target cell;

[0268] Identifier of the optimal beam of the predicted candidate cell;

[0269] Identifier of the optimal beam of the predicted target cell.

[0270] It can be understood that the above RRM measurement prediction results can be understood as the prediction results obtained by the terminal based on the AI object for RRM measurement prediction.

[0271] Optionally, the monitoring results of the AI object include at least one of the following:

[0272] RRM measurement prediction accuracy;

[0273] Prediction accuracy of the optimal cell;

[0274] Prediction accuracy of the optimal beam;

[0275] Prediction accuracy of HOF;

[0276] Prediction accuracy of RLF;

[0277] Prediction accuracy of measurement events;

[0278] Number of occurrences or probability of abnormal events within the first time period;

[0279] Handover delay or interruption duration within the second time period;

[0280] Number of handovers per unit time or within the third time period.

[0281] Exemplarily, the above RRM measurement prediction accuracy may include prediction errors of reference signal receiving power (RSRP) / reference signal received quality (RSRQ) / signal to interference plus noise ratio (SINR) at the cell level or beam level. Among them, the above prediction errors may include, but are not limited to, mean squared error (MSE), root mean square error (RMSE), normalized mean squared error (NMSE), etc.

[0282] The prediction of the above measurement event can be understood as predicting whether the measurement event will be satisfied.

[0283] Exemplarily, the above handover delay or interruption duration may include the average handover delay or average interruption duration within the second time period.

[0284] The above first time period, second time period, and third time period can all be configured by the network-side device or predefined by the protocol.

[0285] The determination methods of the above various monitoring results are illustrated by examples as follows:

[0286] I. RRM measurement prediction accuracy: determined according to the RRM measurement prediction result and the error between the RRM measurement result corresponding to the RRM measurement prediction result within the predicted time or time period; among them, the above RRM measurement prediction result can be understood as the result obtained by the terminal based on the AI object for RRM measurement prediction, and the above RRM measurement result can be understood as the result obtained by the terminal for RRM measurement.

[0287] II. The prediction accuracy of the optimal beam or optimal cell is determined according to the following steps:

[0288] Step S11: The terminal predicts the optimal beam or optimal cell at the second moment at the first moment;

[0289] Step S12: The terminal measures the signal quality of m beams or n cells at the second moment, where the m beams include the predicted optimal beam, or the n cells include the predicted optimal cell;

[0290] Step S13: The terminal sorts according to the signal quality from high to low and determines the actual optimal beam or optimal cell;

[0291] Step S14: The terminal determines whether the prediction is accurate based on the prediction result of step S11 and the actual result of step S13.

[0292] III. The prediction accuracy rate of measurement events is determined according to the following steps:

[0293] Step S21: The terminal predicts at the third moment that the measurement event associated with the first cell at the fourth moment will be satisfied;

[0294] Step S22: The terminal determines whether the measurement event at the fourth moment will be satisfied according to the measurement result of the first cell between the third moment and the fourth moment;

[0295] Step S23: The terminal determines whether the measurement event prediction is accurate based on the prediction result and the actual result of the measurement event.

[0296] IV. The HOF prediction accuracy rate is determined according to the following steps:

[0297] Step S31: The terminal predicts at the fifth moment that HOF will occur in the second cell at the sixth moment;

[0298] Step S32: The terminal determines whether HOF will occur at the sixth moment according to the measurement result of the second cell between the fifth moment and the sixth moment. How to determine HOF based on the measurement result can be implemented based on the UE or pre-configured by the network. For example, the network can configure a second threshold, and when the average signal quality of the second cell within a given time period is lower than the second threshold, the UE determines that HOF will occur;

[0299] Step S33: The terminal determines whether the HOF prediction is accurate based on the prediction result and the actual result of HOF.

[0300] V. The RLF prediction accuracy rate is determined according to the following steps:

[0301] Step S41: The terminal predicts at the seventh moment that RLF will occur in the third cell at the eighth moment;

[0302] Step S42: The terminal determines whether RLF will occur at the eighth moment according to the measurement result of the third cell between the seventh moment and the eighth moment. The terminal can reuse the existing Radio Link Monitor (RLM) process to determine whether RLF will occur based on the signal quality;

[0303] Step S43: The terminal determines whether the RLF prediction is accurate based on the prediction result and the actual result of RLF.

[0304] VI. The number of occurrences / probability of abnormal events within the first time period: The terminal counts the number of occurrences / probability of abnormal events within the first time period (network-configured or pre-defined by the protocol).

[0305] VII. Handoff latency / interruption duration during the second time period, or average handoff latency / average interruption duration during the second time period: The terminal counts the handoff latency / interruption duration during the second time period, or the terminal counts the average handoff latency / average interruption duration during the second time period.

[0306] VIII. Number of handoffs within a unit time or within a third time period: The terminal counts the number of handoffs within a unit time or within a third time period (configured by the network or predefined by the protocol).

[0307] Optionally, the method further includes:

[0308] The source node receives the capability information of the terminal reported by the terminal, where the capability information is used to indicate the AI objects supported by the terminal.

[0309] Exemplarily, the above capability information of the terminal may be carried in the capability reporting information (UECapabilityInformation) or the auxiliary information (UEAssistanceInformation) to indicate the AI objects supported by the terminal.

[0310] Please refer to Figure 6 , Figure 6 which is a flowchart of an AI object management method provided by an embodiment of the present application. This method may be executed by a first node. As Figure 6 shown, it includes the following steps:

[0311] Step 601: During the handoff preparation process, the first node receives the first information sent by the source node;

[0312] wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0313] Information about the AI objects supported by the terminal;

[0314] Information about the AI objects currently activated by the terminal;

[0315] Information about the first AI object;

[0316] The first data collected by the source node that has not been used for AI object training or AI object monitoring;

[0317] The second data collected by the source node that has not been used for AI object inference;

[0318] The applicable scope of the terminal for AI object inference;

[0319] Radio Resource Management (RRM) measurement results;

[0320] Target prediction results;

[0321] Monitoring results of the AI object

[0322] The AI object includes an AI model or an AI function. The first AI object is the AI object applicable to the terminal saved by the source node. The target prediction result includes at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object.

[0323] Optionally, the method further includes at least one of the following:

[0324] The first node sends the configuration information of the first cell to the terminal through the source node; wherein, the first cell includes a candidate cell or a target cell, and the configuration information of the first cell is determined according to the first information;

[0325] The first node performs a first operation according to the first information; wherein, the first operation includes at least one of the following: AI object training, AI object monitoring, AI object inference, AI object management operation; the AI object management operation includes activation, deactivation, switching, or fallback to a non-AI operation.

[0326] In one embodiment, the first node may determine the configuration information of the first cell according to the first information and send the configuration information of the first cell to the source node, so as to send the configuration information of the first cell to the terminal through the source node. Exemplarily, when the first information includes information about the AI object currently activated by the terminal, the first node may configure the first cell that supports the activated AI object for the terminal based on the information about the AI object currently activated by the terminal; or, when the first information includes information about the supported AI object, the first node may configure the first cell that supports the AI object supported by the terminal for the terminal based on the information about the AI object supported by the terminal.

[0327] In another embodiment, the first node may perform a first operation according to the first information. Among them, the above-mentioned fallback to a non-AI operation can be understood as falling back to an operation that does not use the AI object for inference. For example, when the terminal performs RRM measurement prediction based on the AI object, if it receives the indication information to fallback to a non-AI operation, the terminal stops performing RRM measurement prediction based on the AI object and reporting the RRM measurement prediction result, and the terminal performs RRM measurement to obtain the RRM measurement result and reports it.

[0328] Optionally, the configuration information of the first cell carries first indication information, and the first indication information is used to indicate the AI object management operation.

[0329] In this embodiment, when the first node performs the AI object management operation according to the first information, the above-mentioned AI object management operation can be indicated to the terminal through the configuration information of the first cell, which can save signaling while facilitating the terminal to quickly execute the corresponding operation.

[0330] Optionally, the method further includes:

[0331] When the handover is completed, the first node sends second indication information to the terminal; wherein, the second indication information is used to indicate the AI object management operation.

[0332] In this embodiment, when the first node performs the AI object management operation according to the first information, after the handover is completed, indication information indicating the AI object management operation is sent to the terminal so that the terminal can execute the corresponding operation.

[0333] Optionally, the information of the AI object supported by the terminal includes at least one of the following: the identifier of the AI object supported by the terminal, the file information of the AI object supported by the terminal, the input of the AI object supported by the terminal, the output of the AI object supported by the terminal; wherein, the file information of the AI object supported by the terminal includes at least one of the parameter information and the structure information of the AI object supported by the terminal;

[0334] Or,

[0335] The information of the currently active AI object of the terminal includes at least one of the following: the identifier of the currently active AI object of the terminal, the file information of the currently active AI object of the terminal, the input of the currently active AI object of the terminal, the output of the currently active AI object of the terminal; wherein, the file information of the currently active AI object of the terminal includes at least one of the parameter information and the structure information of the currently active AI object of the terminal;

[0336] Or,

[0337] The information of the first AI object includes at least one of the following: the identifier of the first AI object, the file information of the first AI object, the input of the first AI object, the output of the first AI object; wherein, the file information of the first AI object includes at least one of the parameter information and the structure information of the first AI object.

[0338] Optionally, the applicable scope of the AI object inference of the terminal includes at least one of the following: the cell range of the AI object inference of the terminal, the frequency band range of the AI object inference of the terminal.

[0339] Optionally, the first data includes at least one of the input of the AI object and the output of the AI object;

[0340] Or,

[0341] The second data includes the input of the AI object.

[0342] Optionally, the input of the AI object includes at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell;

[0343] Or,

[0344] The output of the AI object includes at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether a radio link failure (RLF) has occurred, the flag indicating whether a handover failure (HOF) has occurred.

[0345] Optionally, the RRM measurement result includes at least one of the following:

[0346] The measured value of the cell signal quality of the serving cell;

[0347] The measured value of the beam signal quality of the serving cell;

[0348] The measured value of the cell signal quality of the candidate cell;

[0349] The measured value of the beam signal quality of the candidate cell;

[0350] The measured value of the cell signal quality of the target cell;

[0351] The measured value of the beam signal quality of the target cell;

[0352] The identifier of the optimal beam of the candidate cell obtained by measurement;

[0353] The identifier of the optimal beam of the target cell obtained by measurement.

[0354] Optionally, the RRM measurement prediction result includes at least one of the following:

[0355] The predicted value of the cell signal quality of the serving cell;

[0356] The predicted value of the beam signal quality of the serving cell;

[0357] The predicted value of the cell signal quality of the candidate cell;

[0358] The predicted value of the beam signal quality of the candidate cell;

[0359] The predicted value of the cell signal quality of the target cell;

[0360] The predicted value of the beam signal quality of the target cell;

[0361] Identifier of the optimal beam of the predicted candidate cell

[0362] Identifier of the optimal beam of the predicted target cell

[0363] Optionally, the monitoring result of the AI object includes at least one of the following:

[0364] RRM measurement prediction accuracy

[0365] Prediction accuracy of the optimal cell

[0366] Prediction accuracy of the optimal beam

[0367] Prediction accuracy of HOF

[0368] Prediction accuracy of RLF

[0369] Prediction accuracy of measurement events

[0370] Number or probability of abnormal events occurring within the first time period

[0371] Handover delay or interruption duration within the second time period

[0372] Number of handovers per unit time or within the third time period

[0373] It should be noted that the implementation method of this embodiment can refer to the relevant description of the embodiment shown in Figure 6 and will not be elaborated here

[0374] Please refer to Figure 7 , Figure 7 which is a flowchart of an AI object management method provided by an embodiment of the present application. This method can be executed by a terminal, as shown in Figure 7 and includes the following steps:

[0375] Step 701: When a handover is executed or completed, the terminal executes a second operation

[0376] Among them, the second operation includes at least one of the following:

[0377] Deactivate the AI object activated in the source cell

[0378] Release the AI object from the source cell

[0379] Release the AI object that is only applicable to the source cell

[0380] Release the AI object that exceeds the applicable scope

[0381] Release the acquisition data corresponding to the second AI object, where the second AI object is the AI object that has been deactivated or released by the terminal

[0382] When the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is an AI object that remains active when the terminal switches from the source cell to the target cell;

[0383] The AI object includes an AI model or an AI function.

[0384] In this embodiment, the above AI object from the source cell can be understood as the AI object passed by the source cell to the terminal.

[0385] The above applicable scope may include but is not limited to the cell scope, frequency band scope, etc. The applicable scope of the above AI object can be understood as the scope in which reasoning can be performed based on the AI object. For the AI object beyond the applicable scope, for example, the target cell is not within the cell scope applicable to the AI object or the target cell frequency band is not within the frequency band scope applicable to the AI object. Optionally, releasing the AI object beyond the applicable scope can also be referred to as releasing the fourth AI object, where the applicable scope of the fourth AI object does not include the target cell or the target cell frequency band.

[0386] The following is an example of this embodiment:

[0387] When performing handover or after handover is completed, the terminal deactivates the AI object activated in the source cell to avoid the AI object activated in the source cell being inapplicable in the target cell, resulting in a decrease in system performance.

[0388] When performing handover or after handover is completed, the terminal releases the AI object from the source cell to avoid the AI object used in the source cell being inapplicable in the target cell, resulting in a decrease in system performance.

[0389] When performing handover or after handover is completed, the terminal releases the AI object that is only applicable to the source cell. Since the AI object that is only applicable to the source cell is inapplicable in the target cell, it will lead to a decrease in system performance. Therefore, releasing the AI object that is only applicable to the source cell is beneficial to reducing the decrease in system performance.

[0390] When performing handover or after handover is completed, the terminal releases the AI object beyond the applicable scope. Since the AI object beyond the applicable scope is inapplicable in the target cell, it will lead to a decrease in system performance. Therefore, releasing the AI object beyond the applicable scope is beneficial to reducing the decrease in system performance.

[0391] When a handover is being performed or has been completed, the terminal releases the acquisition data corresponding to the AI object that has been deactivated or released by the terminal. Since the acquisition data corresponding to the AI object that has been deactivated or released is no longer useful, storage space can be saved after the release.

[0392] When a handover is being performed or has been completed, if the actual result corresponding to the monitoring result of the third AI object has not been reported to the source cell, the terminal reports the actual result corresponding to the monitoring result of the third AI object to the target cell, so that the target cell can continue to monitor the AI object to ensure the accuracy of the prediction result of the AI object.

[0393] It should be noted that the nodes in the embodiments of the present application may include a Distributed Unit (DU), a Centralized Unit (CU), or a base station, etc. The handovers in the embodiments of the present application may include L1 / L2-triggered mobility (LTM) or Primary and Secondary Cell change (PSCell change), etc.; the conditional handovers in the embodiments of the present application may include Conditional PSCell addition / change or Conditional LTM (i.e., Condition L1 / L2-triggered mobility).

[0394] It should also be noted that for the AI object management method provided in the embodiments of the present application, the execution subject may be an AI object management device, or a control module in the AI object management device for executing the AI object management method. In the embodiments of the present application, the AI object management method is executed by the AI object management device as an example to illustrate the AI object management device provided in the embodiments of the present application.

[0395] Please refer to Figure 8 , Figure 8 which is a structural diagram of an AI object management device provided in the embodiments of the present application. As Figure 8 shown, the AI object management device 800 includes:

[0396] A sending module 801, configured to send first information to a first node during a handover preparation process;

[0397] wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following:

[0398] Information about the AI objects supported by the terminal;

[0399] Information about the AI objects currently activated by the terminal;

[0400] Information of the first AI object;

[0401] First data collected by the source node that has not been used for AI object training or AI object monitoring;

[0402] Second data collected by the source node that has not been used for AI object inference;

[0403] Scope of application for the terminal to perform AI object inference;

[0404] Radio Resource Management (RRM) measurement results;

[0405] Target prediction results;

[0406] Monitoring results of the AI object;

[0407] The AI object includes an AI model or an AI function. The first AI object is an AI object suitable for the terminal saved by the source node. The target prediction results include at least one of the RRM measurement prediction result based on the AI object and the event prediction result based on the AI object.

[0408] Optionally, the information of the AI object supported by the terminal includes at least one of the following: the identifier of the AI object supported by the terminal, the file information of the AI object supported by the terminal, the input of the AI object supported by the terminal, the output of the AI object supported by the terminal; wherein, the file information of the AI object supported by the terminal includes at least one of the parameter information and the structure information of the AI object supported by the terminal;

[0409] Or,

[0410] The information of the AI object currently activated by the terminal includes at least one of the following: the identifier of the AI object currently activated by the terminal, the file information of the AI object currently activated by the terminal, the input of the AI object currently activated by the terminal, the output of the AI object currently activated by the terminal; wherein, the file information of the AI object currently activated by the terminal includes at least one of the parameter information and the structure information of the AI object currently activated by the terminal;

[0411] Or,

[0412] The information of the first AI object includes at least one of the following: the identifier of the first AI object, the file information of the first AI object, the input of the first AI object, the output of the first AI object.; wherein, the file information of the first AI object includes at least one of the parameter information and the structure information of the first AI object.

[0413] Optionally, the applicable scope of the terminal for AI object inference includes at least one of the following: the cell range of the terminal for AI object inference, the frequency band range of the terminal for AI object inference.

[0414] Optionally, the first data includes at least one of the input of the AI object and the output of the AI object;

[0415] Or,

[0416] The second data includes the input of the AI object.

[0417] Optionally, the input of the AI object includes at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell;

[0418] Or,

[0419] The output of the AI object includes at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether radio link failure (RLF) occurs, the flag indicating whether handover failure (HOF) occurs.

[0420] Optionally, the RRM measurement result includes at least one of the following:

[0421] The measured value of the cell signal quality of the serving cell;

[0422] The measured value of the beam signal quality of the serving cell;

[0423] The measured value of the cell signal quality of the candidate cell;

[0424] The measured value of the beam signal quality of the candidate cell;

[0425] The measured value of the cell signal quality of the target cell;

[0426] The measured value of the beam signal quality of the target cell;

[0427] The identifier of the optimal beam of the candidate cell obtained by measurement;

[0428] The identifier of the optimal beam of the target cell obtained by measurement.

[0429] Optionally, the RRM measurement prediction result includes at least one of the following:

[0430] The predicted value of the cell signal quality of the serving cell;

[0431] The predicted value of the beam signal quality of the serving cell;

[0432] The predicted value of the cell signal quality of the candidate cell;

[0433] Predicted value of the beam signal quality of the candidate cell;

[0434] Predicted value of the cell signal quality of the target cell;

[0435] Predicted value of the beam signal quality of the target cell;

[0436] Identifier of the optimal beam of the predicted candidate cell;

[0437] Identifier of the optimal beam of the predicted target cell.

[0438] Optionally, the monitoring result of the AI object includes at least one of the following:

[0439] RRM measurement prediction accuracy;

[0440] Prediction accuracy of the optimal cell;

[0441] Prediction accuracy of the optimal beam;

[0442] Prediction accuracy of HOF;

[0443] Prediction accuracy of RLF;

[0444] Prediction accuracy of the measurement event;

[0445] Number or probability of abnormal events occurring within the first time period;

[0446] Handover delay or interruption duration within the second time period;

[0447] Number of handovers per unit time or within the third time period.

[0448] Optionally, the device further includes:

[0449] A receiving module, configured to receive the capability information of the terminal reported by the terminal, where the capability information is used to indicate the AI object supported by the terminal.

[0450] The AI object management device in the embodiments of the present application may be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be a network-side device or other devices other than network-side devices. Exemplarily, the network-side device may include, but is not limited to, the types of network-side devices 12 listed above, and other devices may be servers, Network Attached Storage (NAS), etc., which are not specifically limited in the embodiments of the present application.

[0451] The AI object management device provided by the embodiments of the present application can implement Figure 5The processes implemented by the method embodiments achieve the same technical effects. To avoid repetition, they will not be elaborated here.

[0452] Please refer to Figure 9 , Figure 9 FIG. is a structural diagram of an AI object management device provided by an embodiment of the present application. As Figure 9 shown, the AI object management device 900 includes:

[0453] A receiving module 901, configured to receive first information sent by a source node during a handover preparation process;

[0454] Wherein, the first information includes at least one of the following:

[0455] Information of AI objects supported by the terminal;

[0456] Information of the currently activated AI object of the terminal;

[0457] Information of the first AI object;

[0458] First data collected by the source node that has not been used for AI object training or AI object monitoring;

[0459] Second data collected by the source node that has not been used for AI object inference;

[0460] Scope of application for the terminal to perform AI object inference;

[0461] Radio Resource Management (RRM) measurement results;

[0462] Target prediction results;

[0463] Monitoring results of AI objects;

[0464] The AI object includes an AI model or an AI function. The first AI object is an AI object suitable for the terminal saved by the source node. The target prediction results include at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object.

[0465] Optionally, the device further includes at least one of the following:

[0466] A first sending module, configured to send configuration information of a first cell to the terminal through the source node; wherein, the first cell includes a candidate cell or a target cell, and the configuration information of the first cell is determined according to the first information;

[0467] An execution module, configured to perform a first operation according to the first information; wherein, the first operation includes at least one of the following: AI object training, AI object monitoring, AI object inference, and AI object management operation; the AI object management operation includes activation, deactivation, switching, or fallback to a non-AI operation.

[0468] Optionally, the configuration information of the first cell carries first indication information, and the first indication information is used to indicate the AI object management operation.

[0469] Optionally, the device further includes:

[0470] A second sending module, configured to send second indication information to the terminal when the handover is completed; wherein, the second indication information is used to indicate the AI object management operation.

[0471] Optionally, the information of the AI object supported by the terminal includes at least one of the following: the identifier of the AI object supported by the terminal, the file information of the AI object supported by the terminal, the input of the AI object supported by the terminal, and the output of the AI object supported by the terminal; wherein, the file information of the AI object supported by the terminal includes at least one of the parameter information and the structure information of the AI object supported by the terminal;

[0472] Or,

[0473] The information of the AI object currently activated by the terminal includes at least one of the following: the identifier of the AI object currently activated by the terminal, the file information of the AI object currently activated by the terminal, the input of the AI object currently activated by the terminal, and the output of the AI object currently activated by the terminal; wherein, the file information of the AI object currently activated by the terminal includes at least one of the parameter information and the structure information of the AI object currently activated by the terminal;

[0474] Or,

[0475] The information of the first AI object includes at least one of the following: the identifier of the first AI object, the file information of the first AI object, the input of the first AI object, and the output of the first AI object.; wherein, the file information of the first AI object includes at least one of the parameter information and the structure information of the first AI object.

[0476] Optionally, the applicable scope of the AI object inference performed by the terminal includes at least one of the following: the cell range for the AI object inference performed by the terminal, and the frequency band range for the AI object inference performed by the terminal.

[0477] Optionally, the first data includes at least one of the input of the AI object and the output of the AI object;

[0478] Alternatively,

[0479] The second data includes the input of the AI object.

[0480] Optionally, the input of the AI object includes at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell;

[0481] Alternatively,

[0482] The output of the AI object includes at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether a radio link failure (RLF) has occurred, the flag indicating whether a handover failure (HOF) has occurred.

[0483] Optionally, the RRM measurement result includes at least one of the following:

[0484] The measured value of the cell signal quality of the serving cell;

[0485] The measured value of the beam signal quality of the serving cell;

[0486] The measured value of the cell signal quality of the candidate cell;

[0487] The measured value of the beam signal quality of the candidate cell;

[0488] The measured value of the cell signal quality of the target cell;

[0489] The measured value of the beam signal quality of the target cell;

[0490] The identifier of the optimal beam of the measured candidate cell;

[0491] The identifier of the optimal beam of the measured target cell.

[0492] Optionally, the RRM measurement prediction result includes at least one of the following:

[0493] The predicted value of the cell signal quality of the serving cell;

[0494] The predicted value of the beam signal quality of the serving cell;

[0495] The predicted value of the cell signal quality of the candidate cell;

[0496] The predicted value of the beam signal quality of the candidate cell;

[0497] The predicted value of the cell signal quality of the target cell;

[0498] The predicted value of the beam signal quality of the target cell;

[0499] The identifier of the optimal beam of the predicted candidate cell;

[0500] The identifier of the optimal beam of the predicted target cell.

[0501] Optionally, the monitoring result of the AI object includes at least one of the following:

[0502] The prediction accuracy of RRM measurement;

[0503] The prediction accuracy of the optimal cell;

[0504] The prediction accuracy of the optimal beam;

[0505] The prediction accuracy of HOF;

[0506] The prediction accuracy of RLF;

[0507] The prediction accuracy of measurement events;

[0508] The number of occurrences or probability of abnormal events within the first time period;

[0509] The handover delay or interruption duration within the second time period;

[0510] The number of handovers per unit time or within the third time period.

[0511] The AI object management device in the embodiments of the present application may be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be a network-side device or other devices other than network-side devices. Exemplarily, the network-side device may include, but is not limited to, the types of network-side devices 12 listed above, and other devices may be a server, a Network Attached Storage (NAS), etc., which are not specifically limited in the embodiments of the present application.

[0512] The AI object management device provided in the embodiments of the present application can implement Figure 6 each process implemented by the method embodiment and achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0513] Please refer to Figure 10 , Figure 10 is the structural diagram of an AI object management device provided in the embodiments of the present application. As Figure 10 shown, the AI object management device 1000 includes:

[0514] An execution module 1001, configured to perform a second operation when a handover is executed or the handover is completed;

[0515] Wherein, the second operation includes at least one of the following:

[0516] Deactivate the AI object activated in the source cell;

[0517] Release the AI object from the source cell;

[0518] Release the AI object only applicable to the source cell;

[0519] Release the AI object beyond the applicable scope;

[0520] Release the acquisition data corresponding to the second AI object, where the second AI object is an AI object that has been deactivated or released by the terminal;

[0521] In the case where the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is an AI object that remains activated when the terminal switches from the source cell to the target cell;

[0522] The AI object includes an AI model or an AI function.

[0523] The AI object management device in the embodiments of the present application may be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be a terminal or other devices other than the terminal. Exemplarily, the terminal may include, but is not limited to, the types of the terminal 11 listed above, and other devices may be a server, a Network Attached Storage (NAS), etc., which are not specifically limited in the embodiments of the present application.

[0524] The AI object management device provided by the embodiments of the present application can implement Figure 7 each process implemented by the method embodiments and achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0525] Optionally, as Figure 11As shown in the figure, an embodiment of the present application further provides a communication device 1100, including a processor 1101 and a memory 1102. A program or instruction that can run on the processor 1101 is stored on the memory 1102. For example, when the communication device 1100 is a source node, when the program or instruction is executed by the processor 1101, each step of the above-mentioned embodiment of the AI object management method on the source node side is implemented, and the same technical effect can be achieved. When the communication device 1100 is a first node, when the program or instruction is executed by the processor 1101, each step of the above-mentioned embodiment of the AI object management method on the first node side is implemented, and the same technical effect can be achieved. When the communication device 1100 is a terminal, when the program or instruction is executed by the processor 1101, each step of the above-mentioned embodiment of the AI object management method on the terminal side is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be elaborated here.

[0526] An embodiment of the present application further provides a source node, including a processor and a communication interface. The communication interface is used to send first information to a first node during a handover preparation process. Among them, the first node includes a candidate node or a target node, and the first information includes at least one of the following: information on AI objects supported by the terminal; information on currently activated AI objects of the terminal; information on a first AI object; first data collected by the source node that has not been used for AI object training or AI object monitoring; second data collected by the source node that has not been used for AI object inference; applicable scope for the terminal to perform AI object inference; radio resource management (RRM) measurement results; target prediction results; monitoring results of AI objects. The AI object includes an AI model or an AI function, and the first AI object is an AI object suitable for the terminal saved by the source node. The target prediction results include at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object. This source node embodiment corresponds to the above-mentioned source node method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this source node embodiment, and the same technical effect can be achieved.

[0527] Specifically, an embodiment of the present application further provides a source node. As Figure 12 shown, the source node 1200 includes: an antenna 1201, a radio frequency device 1202, a baseband device 1203, a processor 1204, and a memory 1205. The antenna 1201 is connected to the radio frequency device 1202. In the uplink direction, the radio frequency device 1202 receives information through the antenna 1201 and sends the received information to the baseband device 1203 for processing. In the downlink direction, the baseband device 1203 processes the information to be sent and sends it to the radio frequency device 1202. After the radio frequency device 1202 processes the received information, it is sent out through the antenna 1201.

[0528] The method executed by the source node in the above embodiments may be implemented in the baseband device 1203, which includes a baseband processor.

[0529] The baseband device 1203 may, for example, include at least one baseband board, on which a plurality of chips are provided, such as Figure 12 shown, where one of the chips is, for example, a baseband processor, which is connected to the memory 1205 through a bus interface to call the program in the memory 1205 and execute the network device operations shown in the above method embodiments.

[0530] The source node may further include a network interface 1206, which is, for example, a Common Public Radio Interface (CPRI).

[0531] Specifically, the source node 1200 in the embodiments of the present application further includes: instructions or programs stored on the memory 1205 and executable on the processor 1204. The processor 1204 calls the instructions or programs in the memory 1205 to execute Figure 8 the methods executed by the respective modules shown, and achieves the same technical effects. To avoid repetition, they are not described herein again.

[0532] The embodiments of the present application further provide a first node, including a processor and a communication interface. The communication interface is used to receive first information sent by a source node during a handover preparation process. Among them, the first node includes a candidate node or a target node, and the first information includes at least one of the following: information on AI objects supported by a terminal; information on currently activated AI objects of the terminal; information on a first AI object; first data collected by the source node that has not been used for AI object training or AI object monitoring; second data collected by the source node that has not been used for AI object inference; applicable scope for the terminal to perform AI object inference; radio resource management (RRM) measurement results; target prediction results; monitoring results of AI objects. The AI object includes an AI model or an AI function, the first AI object is an AI object suitable for the terminal saved by the source node, and the target prediction results include at least one of an RRM measurement prediction result based on an AI object and an event prediction result based on an AI object. This first node embodiment corresponds to the above first node method embodiment. Each implementation process and implementation method of the above method embodiment can be applied to this first node embodiment and can achieve the same technical effects.

[0533] Specifically, the embodiments of the present application further provide a first node. As Figure 13As shown, the first node 1300 includes: an antenna 1301, a radio frequency device 1302, a baseband device 1303, a processor 1304, and a memory 1305. The antenna 1301 is connected to the radio frequency device 1302. In the uplink direction, the radio frequency device 1302 receives information through the antenna 1301 and sends the received information to the baseband device 1303 for processing. In the downlink direction, the baseband device 1303 processes the information to be sent and sends it to the radio frequency device 1302. After processing the received information, the radio frequency device 1302 sends it out through the antenna 1301.

[0534] In the above embodiments, the method executed by the first node can be implemented in the baseband device 1303, and the baseband device 1303 includes a baseband processor.

[0535] The baseband device 1303 may include, for example, at least one baseband board, and a plurality of chips are provided on the baseband board, such as Figure 13 As shown, one of the chips is, for example, a baseband processor, which is connected to the memory 1305 through a bus interface to call the program in the memory 1305 and execute the operations of the network device shown in the above method embodiments.

[0536] The first node may further include a network interface 1306, and the interface is, for example, a Common Public Radio Interface (CPRI).

[0537] Specifically, the first node 1300 in the embodiments of the present application further includes: instructions or programs stored on the memory 1305 and executable on the processor 1304. The processor 1304 calls the instructions or programs in the memory 1305 to execute Figure 9 the methods executed by the modules shown, and achieves the same technical effects. To avoid repetition, it will not be described in detail here.

[0538] An embodiment of the present application further provides a terminal, including a processor and a communication interface. The processor is configured to perform a second operation when a handover or handover completion is executed; wherein, the second operation includes at least one of the following: deactivating an AI object activated in a source cell; releasing an AI object from the source cell; releasing an AI object only applicable to the source cell; releasing an AI object beyond the applicable scope; releasing the acquisition data corresponding to a second AI object, where the second AI object is an AI object that has been deactivated or released by the terminal; when the actual result corresponding to the monitoring result of a third AI object has not been reported to the source cell, reporting the actual result corresponding to the monitoring result of the third AI object to a target cell, where the third AI object is an AI object that remains activated when the terminal handovers from the source cell to the target cell; the AI object includes an AI model or an AI function. This terminal embodiment corresponds to the above terminal-side method embodiment. Each implementation process and implementation manner of the above method embodiment can be applied to this terminal embodiment and can achieve the same technical effect. Specifically, Figure 14 Schematic diagram of the hardware structure of a terminal for implementing an embodiment of the present application.

[0539] The terminal 1400 includes, but is not limited to, at least some components such as a radio frequency unit 1401, a network module 1402, an audio output unit 1403, an input unit 1404, a sensor 1405, a display unit 1406, a user input unit 1407, an interface unit 1408, a memory 1409, and a processor 1410.

[0540] Those skilled in the art can understand that the terminal 1400 may further include a power source (such as a battery) for supplying power to each component. The power source can be logically connected to the processor 1410 through a power management system, so as to implement functions such as management of charging, discharging, and power consumption management through the power management system. Figure 14 The terminal structure shown does not limit the terminal. The terminal may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements, which will not be elaborated here.

[0541] It should be understood that in the embodiments of the present application, the input unit 1404 may include a Graphics Processing Unit (GPU) 14041 and a microphone 14042. The graphics processor 14041 processes the image data of static pictures or videos obtained by an image capture device (such as a camera) in the video capture mode or the image capture mode. The display unit 1406 may include a display panel 14061, and the display panel 14061 may be configured in the form of, for example, a liquid crystal display, an organic light emitting diode, etc. The user input unit 1407 includes at least one of a touch panel 14071 and other input devices 14072. The touch panel 14071 is also referred to as a touch screen. The touch panel 14071 may include two parts: a touch detection device and a touch controller. The other input devices 14072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, which will not be elaborated here.

[0542] In the embodiments of the present application, after receiving downlink data from a network-side device, the radio frequency unit 1401 may transmit it to the processor 1410 for processing; in addition, the radio frequency unit 1401 may send uplink data to the network-side device. Generally, the radio frequency unit 1401 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.

[0543] The memory 1409 can be used to store software programs or instructions as well as various data. The memory 1409 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data. Among them, the first storage area may store an operating system, application programs or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 1409 may include a volatile memory or a non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM), and a direct rambus random access memory (DRRAM). The memory 1409 in the embodiments of the present application includes but is not limited to these and any other suitable types of memories.

[0544] The processor 1410 may include one or more processing units; optionally, the processor 1410 integrates an application processor and a modem processor. Among them, the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above modem processor may not be integrated into the processor 1410 either.

[0545] Among them, the processor 1410 is used to perform a second operation when a handover or the handover is completed; among them, the second operation includes at least one of the following:

[0546] Deactivate the AI object activated in the source cell;

[0547] Release the AI object from the source cell;

[0548] Release the AI object only applicable to the source cell;

[0549] Release AI objects beyond the applicable scope;

[0550] Release the collected data corresponding to the second AI object, where the second AI object is an AI object that has been deactivated or released by the terminal;

[0551] In the case where the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is an AI object that remains active when the terminal switches from the source cell to the target cell;

[0552] The AI object includes an AI model or an AI function.

[0553] It can be understood that the implementation processes of the various implementation manners mentioned in this embodiment can refer to the relevant descriptions of the foregoing method embodiments, and achieve the same or corresponding technical effects. To avoid repetition, they will not be elaborated here.

[0554] The embodiment of the present application further provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, it implements each process of the above-mentioned AI object management method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0555] Wherein, the processor is the processor in the terminal described in the above embodiment. The readable storage medium includes computer-readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc. In some examples, the readable storage medium can be a non-transitory readable storage medium.

[0556] The embodiment of the present application further provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement each process of the above-mentioned AI object management method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0557] It should be understood that the chip mentioned in the embodiment of the present application can also be referred to as a system-on-chip, system chip, chip system or system-on-chip, etc.

[0558] The embodiment of the present application further provides a computer program / program product, which includes a computer program or computer instruction. The computer program or computer instruction is executed by at least one processor to implement each process of the above-mentioned AI object management method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0559] An embodiment of the present application further provides an AI object management system, including: a source node, a first node, and a terminal. The source node is used to execute as Figure 5 and each process of the above-mentioned method embodiments, and the first node is used to execute as Figure 6 and each process of the above-mentioned method embodiments, and the terminal is used to execute as Figure 7 and each process of the above-mentioned method embodiments, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0560] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including that element. In addition, it should be pointed out that the methods and devices in the embodiments of the present application are not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may also be added, omitted or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0561] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of a computer software product plus a necessary general hardware platform, and of course, they can also be implemented by hardware. This computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions for causing a terminal or a network-side device to execute the methods described in various embodiments of the present application.

[0562] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms of embodiments without departing from the purpose of the present application and the scope protected by the claims. These embodiments are all within the protection scope of the present application.

Claims

1. An AI object management method, characterized in that, Including: During the handover preparation process, the source node sends first information to the first node; Wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following: Information about the AI objects supported by the terminal; Information about the AI objects currently activated by the terminal; Information about the first AI object; First data collected by the source node that has not been used for AI object training or AI object monitoring; Second data collected by the source node that has not been used for AI object inference; The applicable scope of the terminal for AI object inference; Radio Resource Management (RRM) measurement results; Target prediction results; Monitoring results of AI objects; The AI object includes an AI model or an AI function, the first AI object is an AI object suitable for the terminal saved by the source node, and the target prediction results include at least one of the RRM measurement prediction results based on the AI object and the event prediction results based on the AI object.

2. The method according to claim 1, characterized in that The information about the AI objects supported by the terminal includes at least one of the following: the identifier of the AI objects supported by the terminal, the file information of the AI objects supported by the terminal, the input of the AI objects supported by the terminal, the output of the AI objects supported by the terminal; wherein, the file information of the AI objects supported by the terminal includes at least one of the parameter information and the structure information of the AI objects supported by the terminal; Or, The information about the AI objects currently activated by the terminal includes at least one of the following: the identifier of the AI objects currently activated by the terminal, the file information of the AI objects currently activated by the terminal, the input of the AI objects currently activated by the terminal, the output of the AI objects currently activated by the terminal; wherein, the file information of the AI objects currently activated by the terminal includes at least one of the parameter information and the structure information of the AI objects currently activated by the terminal; Or, The information about the first AI object includes at least one of the following: the identifier of the first AI object, the file information of the first AI object, the input of the first AI object, the output of the first AI object; wherein, the file information of the first AI object includes at least one of the parameter information and the structure information of the first AI object.

3. The method according to claim 1 or 2, characterized in that, The applicable scope of the terminal for AI object inference includes at least one of the following: the cell range of the terminal for AI object inference, the frequency band range of the terminal for AI object inference.

4. The method according to any one of claims 1 to 3, characterized in that The first data includes at least one of the input of the AI object and the output of the AI object; Or, The second data includes the input of the AI object.

5. The method according to claim 4, wherein The input of the AI object includes at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell; Or, The output of the AI object includes at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether a Radio Link Failure (RLF) occurs, the flag indicating whether a Handover Failure (HOF) occurs.

6. The method according to any one of claims 1 to 5, characterized in that The RRM measurement results include at least one of the following: The measured value of the cell signal quality of the serving cell; Measured value of the beam signal quality of the serving cell; Measured value of the cell signal quality of the candidate cell; Measured value of the beam signal quality of the candidate cell; Measured value of the cell signal quality of the target cell; Measured value of the beam signal quality of the target cell; Identifier of the optimal beam of the candidate cell obtained by measurement; Identifier of the optimal beam of the target cell obtained by measurement.

7. The method according to any one of claims 1 to 6, characterized in that The RRM measurement prediction result includes at least one of the following: Predicted value of the cell signal quality of the serving cell; Predicted value of the beam signal quality of the serving cell; Predicted value of the cell signal quality of the candidate cell; Predicted value of the beam signal quality of the candidate cell; Predicted value of the cell signal quality of the target cell; Predicted value of the beam signal quality of the target cell; Identifier of the optimal beam of the candidate cell obtained by prediction; Identifier of the optimal beam of the target cell obtained by prediction.

8. The method according to any one of claims 1 to 7, characterized in that, The monitoring result of the AI object includes at least one of the following: RRM measurement prediction accuracy; Prediction accuracy of the optimal cell; Prediction accuracy of the optimal beam; Prediction accuracy of the HOF; Prediction accuracy of the RLF; Prediction accuracy of the measurement event; Number or probability of abnormal events occurring within the first time period; Handover delay or interruption duration within the second time period; Number of handovers per unit time or within the third time period.

9. The method according to any one of claims 1 to 8, characterized in that The method further includes: The source node receives the capability information of the terminal reported by the terminal, where the capability information is used to indicate the AI objects supported by the terminal.

10. An AI object management method, characterized in that, Includes: During the handover preparation process, the first node receives the first information sent by the source node; Wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following: Information of the AI objects supported by the terminal; Information of the currently activated AI objects of the terminal; Information of the first AI object; First data collected by the source node that has not been used for AI object training or AI object monitoring; Second data collected by the source node that has not been used for AI object inference; Applicable scope for the terminal to perform AI object inference; Radio Resource Management (RRM) measurement results; Target prediction results; Monitoring results of AI objects; The AI object includes an AI model or an AI function, the first AI object is the AI object suitable for the terminal saved by the source node, and the target prediction result includes at least one of the RRM measurement prediction result based on the AI object and the event prediction result based on the AI object.

11. The method according to claim 10, wherein The method further includes at least one of the following: The first node sends the configuration information of the first cell to the terminal through the source node; wherein, the first cell includes a candidate cell or a target cell, and the configuration information of the first cell is determined according to the first information; The first node performs a first operation according to the first information; wherein, the first operation includes at least one of the following: AI object training, AI object monitoring, AI object inference, AI object management operation; the AI object management operation includes activation, deactivation, handover, or fallback to non-AI operation.

12. The method according to claim 11, wherein The configuration information of the first cell carries first indication information, and the first indication information is used to indicate the AI object management operation.

13. The method according to claim 11, characterized in that, The method further includes: In the case where the handover is completed, the first node sends second indication information to the terminal; wherein, the second indication information is used to indicate the AI object management operation.

14. The method according to any one of claims 1 to 13, characterized in that, The information of the AI objects supported by the terminal includes at least one of the following: the identifier of the AI objects supported by the terminal, the file information of the AI objects supported by the terminal, the input of the AI objects supported by the terminal, the output of the AI objects supported by the terminal; wherein, the file information of the AI objects supported by the terminal includes at least one of the parameter information and the structure information of the AI objects supported by the terminal; Or, The information of the currently active AI objects of the terminal includes at least one of the following: the identifier of the currently active AI objects of the terminal, the file information of the currently active AI objects of the terminal, the input of the currently active AI objects of the terminal, the output of the currently active AI objects of the terminal; wherein, the file information of the currently active AI objects of the terminal includes at least one of the parameter information and the structure information of the currently active AI objects of the terminal; Or, The information of the first AI object includes at least one of the following: the identifier of the first AI object, the file information of the first AI object, the input of the first AI object, the output of the first AI object; wherein, the file information of the first AI object includes at least one of the parameter information and the structure information of the first AI object.

15. The method according to any one of claims 10 to 14, characterized in that, The applicable scope of the terminal for performing AI object inference includes at least one of the following: the cell range of the terminal for performing AI object inference, the frequency band range of the terminal for performing AI object inference.

16. The method according to any one of claims 10 to 15, characterized in that, The first data includes at least one of the input of the AI object and the output of the AI object; Or, The second data includes the input of the AI object.

17. The method according to claim 16, wherein The input of the AI object includes at least one of the following: the cell signal quality of the serving cell, the beam signal quality of the serving cell, the cell signal quality of the neighboring cell, the beam signal quality of the neighboring cell; Or, The output of the AI object includes at least one of the following: the identifier of the optimal cell, the identifier of the optimal beam, the flag indicating whether the measurement event is satisfied, the flag indicating whether radio link failure (RLF) occurs, the flag indicating whether handover failure (HOF) occurs.

18. The method according to any one of claims 10 to 17, characterized in that The RRM measurement results include at least one of the following: The measured value of the cell signal quality of the serving cell; The measured value of the beam signal quality of the serving cell; The measured value of the cell signal quality of the candidate cell; The measured value of the beam signal quality of the candidate cell; The measured value of the cell signal quality of the target cell; The measured value of the beam signal quality of the target cell; The identifier of the optimal beam of the measured candidate cell; The identifier of the optimal beam of the measured target cell.

19. The method according to any one of claims 10 to 18, characterized in that, The RRM measurement prediction results include at least one of the following: The predicted value of the cell signal quality of the serving cell; The predicted value of the beam signal quality of the serving cell; The predicted value of the cell signal quality of the candidate cell; The predicted value of the beam signal quality of the candidate cell; The predicted value of the cell signal quality of the target cell; The predicted value of the beam signal quality of the target cell; The identifier of the optimal beam of the predicted candidate cell; The identifier of the optimal beam of the target cell obtained by prediction.

20. The method according to any one of claims 10 to 19, characterized in that The monitoring results of the AI object include at least one of the following: The prediction accuracy of RRM measurement; The prediction accuracy of the optimal cell; The prediction accuracy of the optimal beam; The prediction accuracy of HOF; The prediction accuracy of RLF; The prediction accuracy of measurement events; The number of occurrences or probability of abnormal events within the first time period; The handover delay or interruption duration within the second time period; The number of handovers per unit time or within the third time period.

21. An AI object management method, characterized in that, including: When a handover is executed or completed, the terminal executes a second operation; wherein, the second operation includes at least one of the following: Deactivate the AI object activated in the source cell; Release the AI object from the source cell; Release the AI object only applicable to the source cell; Release the AI object beyond the applicable scope; Release the collected data corresponding to the second AI object, where the second AI object is the AI object that has been deactivated or released by the terminal; When the actual result corresponding to the monitoring result of the third AI object has not been reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is the AI object that remains active when the terminal switches from the source cell to the target cell; The AI object includes an AI model or an AI function.

22. An AI object management device, characterized in that, including: A sending module, configured to send first information to a first node during the handover preparation process; wherein, the first node includes a candidate node or a target node, and the first information includes at least one of the following: Information about the AI objects supported by the terminal; Information about the AI objects currently activated by the terminal; Information about the first AI object; The first data collected by the source node that has not been used for AI object training or AI object monitoring; The second data collected by the source node that has not been used for AI object inference; The applicable scope of the terminal for AI object inference; The radio resource management (RRM) measurement results; The target prediction results; The monitoring results of the AI object; The AI object includes an AI model or an AI function, the first AI object is the AI object suitable for the terminal saved by the source node, and the target prediction results include at least one of the RRM measurement prediction results based on the AI object and the event prediction results based on the AI object.

23. An AI object management device, characterized in that, including: A receiving module, configured to receive the first information sent by the source node during the handover preparation process; wherein, the first information includes at least one of the following: Information about the AI objects supported by the terminal; Information about the AI objects currently activated by the terminal; Information about the first AI object; The first data collected by the source node that has not been used for AI object training or AI object monitoring; The second data collected by the source node that has not been used for AI object inference; The applicable scope of the terminal for AI object inference; The radio resource management (RRM) measurement results; The target prediction results; The monitoring results of the AI object; The AI object includes an AI model or an AI function. The first AI object is the AI object applicable to the terminal saved by the source node. The target prediction result includes at least one of an RRM measurement prediction result based on the AI object and an event prediction result based on the AI object.

24. An AI object management device, characterized in that, Comprising: An execution module, configured to perform a second operation when a handover is executed or the handover is completed; Wherein, the second operation includes at least one of the following: Deactivate the AI object activated in the source cell; Release the AI object from the source cell; Release the AI object only applicable to the source cell; Release the AI object beyond the applicable scope; Release the acquisition data corresponding to the second AI object, where the second AI object is the AI object deactivated or released by the terminal; When the actual result corresponding to the monitoring result of the third AI object is not reported to the source cell, report the actual result corresponding to the monitoring result of the third AI object to the target cell, where the third AI object is the AI object that remains activated when the terminal handovers from the source cell to the target cell; The AI object includes an AI model or an AI function.

25. A source node, characterized in that, Comprising a processor and a memory, where the memory stores a program or instruction that can run on the processor, and when the program or instruction is executed by the processor, the steps of the AI object management method according to any one of claims 1 to 9 are implemented.

26. A first node, characterized in that, Comprising a processor and a memory, where the memory stores a program or instruction that can run on the processor, and when the program or instruction is executed by the processor, the steps of the AI object management method according to any one of claims 10 to 20 are implemented.

27. A terminal, characterized in that, Comprising a processor and a memory, where the memory stores a program or instruction that can run on the processor, and when the program or instruction is executed by the processor, the steps of the AI object management method according to claim 21 are implemented.

28. A readable storage medium, characterized in that, A program or instruction is stored on the readable storage medium, and when the program or instruction is executed by the processor, the steps of the AI object management method according to any one of claims 1 to 9 are implemented, or the steps of the AI object management method according to any one of claims 10 to 20 are implemented, or the steps of the AI object management method according to claim 21 are implemented.

29. A computer program product, characterized in that, The computer program product includes a computer program or a computer instruction, and the computer program or the computer instruction is executed by at least one processor to implement the steps of the AI object management method according to any one of claims 1 to 9, or to implement the steps of the AI object management method according to any one of claims 10 to 20, or to implement the steps of the AI object management method according to claim 21.

Images