CN117997738A

CN117997738A - AI model updating method and communication device

Info

Publication number: CN117997738A
Application number: CN202211372492.8A
Authority: CN
Inventors: 陈晓宇
Original assignee: Spreadtrum Communications Shanghai Co Ltd
Current assignee: Spreadtrum Communications Shanghai Co Ltd
Priority date: 2022-11-03
Filing date: 2022-11-03
Publication date: 2024-05-07
Also published as: WO2024094157A1

Abstract

The application provides an AI model updating method and a communication device, wherein the method can comprise the following steps: receiving a first signaling from a network device, wherein the first signaling is used for indicating first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the first model association information is different from the second model association information currently used, and the second AI model currently used is updated to the first AI model. By adopting the application, the network equipment can trigger the terminal equipment to update the AI model through signaling, so that the terminal equipment can update the AI model rapidly, thereby being beneficial to improving the communication efficiency.

Description

AI model updating method and communication device

Technical Field

The application relates to the technical field of communication, in particular to a method for updating an artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) model and a communication device.

Background

With the development of scientific technology, AI technology has been developed. AI technology can be widely used in various fields. The AI technology is applied to the technical field of wireless communication, so that the advantages of improving the throughput of a wireless communication system, effectively shortening the processing time delay, enhancing the user capacity and the like can be achieved, and the AI technology is an important task of the future wireless communication technology.

For a device, its AI functionality can be implemented by a deployed AI model. Currently, in the third generation partnership project (3 ^rd -generation partnership project,3 GPP) standardization, the deployment of AI models on terminal devices is supported to enable AI models to be applied in wireless communication systems. How to trigger the AI model update of the terminal device is a technical problem to be solved.

Disclosure of Invention

The embodiment of the application provides an AI model updating method which can quickly update an AI model of terminal equipment, thereby being beneficial to improving communication efficiency.

In a first aspect, an embodiment of the present application provides an AI model updating method, which may be performed by a terminal device, or by a device matched with the terminal device, for example, a chip module, or a processor. The method may include: receiving a first signaling from a network device, wherein the first signaling is used for indicating first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the first model association information is different from the second model association information which is currently used, and the second AI model which is currently used is updated to be the first AI model; wherein the second model association information has an association relationship with attribute information of the second AI model.

It can be seen that when the terminal device receives the first signaling indicating the first model association information, the second AI model currently used can be updated to the first AI model having an association relationship with the first model association information, so that the terminal device can quickly update the AI model, thereby contributing to improving communication efficiency.

In one possible implementation, when updating the currently used second AI model to the first AI model, the currently used second AI model may be updated to the first AI model, in particular, in a case where the first AI model is different from the second AI model and the terminal device supports the first AI model. Therefore, in a mobile scene or a network switching scene, the second AI model currently used by the terminal equipment can be updated to be the first AI model which has an association relation with the association information of the first model, and the communication quality and the communication efficiency can be improved.

Further, in the case that the first AI model is different from the second AI model, and the terminal device supports the first AI model, and the state of the first AI model is not the deactivated state, the second AI model currently used is updated to the first AI model. That is, in the case where the state of the first AI model is the deactivated state, the second AI model that is currently being used is not updated to the first AI model.

In one possible implementation manner, in a case that the first AI model is different from the second AI model and the terminal device does not support the first AI model, selecting a target AI model from AI models supported by the terminal device based on the first model association information, the target AI model satisfying the similarity condition; updating the first AI model or the second AI model to the target AI model. In the case where the terminal device does not support the first AI model, a target AI model closest to attribute information of the first AI model may be selected, and communication efficiency may be improved to some extent.

In one possible implementation, in a case where the first AI model is different from the second AI model and the terminal device does not support the first AI model, a first request message is sent to the network device, the first request message being for indicating attribute information of the third AI model. That is, in the case where the terminal device does not support the first AI model, the terminal device requests the network device to use the third AI model. A first response message is received from the network device, and the first AI model or the second AI model is updated to a third AI model if the first response message indicates acknowledgement information. Conversely, in the case where the first response message indicates rejection information, the AI model update may be stopped, or the first request message may be sent again to the network device, at which time the first request message may indicate that the terminal device requests attribute information using the additional AI model (that is, the additional AI model is different from the third AI model), waiting for receipt of acknowledgement information or rejection information from the network device.

In one possible implementation, in the case that the first AI model is the same as the second AI model, the association relationship between the current model association information and the attribute information of the second AI model is updated to be the association relationship between the first model association information and the attribute information of the second AI model. Thereby, the association relation between the model association information and the attribute information of the AI model can be updated in real time.

In one possible implementation, after updating the currently used second AI model to the first AI model, the second AI model may be deactivated to avoid using 2 or more AI models simultaneously, reducing the processing load of the terminal device. Updating the second AI model to the first AI model implicitly represents activating and using the first AI model.

In one possible implementation manner, the method further includes: sending a second request message to the network device, the second request message being used to determine the first information; a second response message is received from the network device, the second response message including the first information.

In one possible implementation, the first information includes attribute information of a fourth AI model; the method further comprises the following steps: the first AI model or the second AI model is updated to the fourth AI model.

In one possible implementation, the first information includes first reference signal information; the method further comprises the following steps: obtaining channel quality information according to the first reference signal information; updating the first AI model or the second AI model to an AI model with an association relationship with the channel quality information according to the channel quality information, and/or updating attribute information of the AI model with the association relationship with the channel quality information. Therefore, the terminal equipment retrains the AI model to ensure that the updated AI model is matched with the channel quality information.

In one possible implementation, the first information further includes a reporting threshold; the method further comprises the following steps: and selecting target channel quality information which is larger than or equal to a reporting threshold value from the channel quality information, and determining reference signal information corresponding to the target channel quality information.

In one possible implementation manner, the method further includes: and transmitting the target channel quality information and the reference signal information corresponding to the target channel quality information to the network equipment. So that the network device knows the reference signal information used by the terminal device to retrain the AI model.

In one possible implementation manner, the method further includes: and updating the attribute information of the AI model with the association relation with the channel quality information according to the target channel quality information and the reference signal information corresponding to the target channel quality information. Therefore, the terminal equipment retrains the AI model to ensure that the updated AI model is matched with the channel quality information.

In one possible implementation manner, the method further includes: and if the output result of the first AI model does not meet the preset condition, selecting a target AI model from AI models supported by the terminal equipment based on the first model association information, and updating the first AI model into the target AI model. Wherein the target AI model satisfies a similarity condition. That is, when the output result of the first AI model does not meet the performance requirement, the first AI model may be updated to the target AI model.

In one possible implementation manner, the method further includes: and when the state of the first AI model is the deactivation state, selecting a target AI model from AI models supported by the terminal equipment based on the first model association information, and updating the first AI model or the second AI model to the target AI model. Wherein the target AI model satisfies a similarity condition. That is, when the first AI model is deactivated, either the first AI model or the second AI model may be updated to the target AI model.

In one possible implementation manner, the method further includes: the first AI model is different from the AI model indicated by the AI model indication information, a target AI model is selected from AI models supported by the terminal device based on the first model association information, and the first AI model or the second AI model is updated to the target AI model. Wherein the target AI model satisfies a similarity condition. That is, when the first AI model is different from the AI model indicated by the network device, the first AI model or the second AI model may be updated to the target AI model.

In one possible implementation manner, the target AI model satisfies a similarity condition, including at least one of the following:

the difference between the model identification of the first AI model and the model identification of the target AI model is less than or equal to a first threshold;

The first model association information is the same as or similar to the model association information associated with the target AI model and is larger than a second threshold;

The target AI model is an AI model associated with preset cell information; the preset cell information is contained in the configured or activated cell information; or, presetting cell information as cell information associated with a target transmission configuration indication TCI state, wherein the target TCI state is indicated or configured or activated by a second signaling; optionally, a time difference between the time of receipt of the second signaling and the current system time is less than a third threshold;

the target AI model is the AI model most frequently used or the AI model last used in a second preset period of time before updating the second AI model.

In one possible implementation manner, the method further includes: third signaling is received from the network device, the third signaling being used to indicate an association between the first model association information and attribute information of the first AI model. That is, the network device may indicate the association relationship between the respective model association information and the attribute information of the respective AI models so that the network device and the terminal device understand the agreement for the association relationship between the AI models of the terminal device and the model association information.

In one possible implementation manner, the method further includes: and determining the association relation between the first model association information and the attribute information of the first AI model. That is, the terminal device maintains the association relationship between the respective model association information and the attribute information of the respective AI models, and the mobility of the terminal device can be improved.

In one possible implementation, the association relationship between the first model association information and the attribute information of the first AI model is protocol predefined. That is, the protocol predefines the association relationship between the respective model association information and the attribute information of the respective AI models so that the network device and the terminal device understand the agreement for the association relationship between the AI models of the terminal device and the model association information.

In one possible implementation, the first model association information may include, but is not limited to, at least one of the following: first operator identification information, first public land mobile network (public land mobile network, PLMN) information, first network device identification information, first tracking area identification (TRACKING AREA IDENTIFIER, TAI) information, first roaming guide (steering of roaming, SOR) information, first terminal routing policy (user equipment route selection policy, URSP) information, first operating frequency information, first cell information-associated reference signal information, first neighbor list, first timestamp information, first terminal device information;

Wherein the first cell information includes, but is not limited to, at least one of the following: first cell group information, first serving cell identification information, first physical cell identification information, first transmission reception point (transmission and reception point, TRP) identification information, and first bandwidth part (BWP) identification.

In one possible implementation, the attribute information of the first AI model includes, but is not limited to, at least one of: the model identification of the first AI model, the model structure information of the first AI model, the model parameters of the first AI model.

In a second aspect, an embodiment of the present application provides an AI model updating method, which may be performed by a network device, or by an apparatus matched with the network device, such as a chip, a chip module, or a processor. The method may include: and sending a first signaling to the terminal equipment, wherein the first signaling is used for indicating first model association information, the first model association information has association relation with attribute information of a first AI model, and the first model association information is different from second model association information which is currently used.

In one possible implementation manner, the method further includes: receiving a first request message from a terminal device, wherein the first request message is used for indicating attribute information of a third AI model; and the first response message is sent to the terminal equipment and is used for indicating the confirmation information.

In one possible implementation manner, the method further includes: receiving a second request message from the terminal equipment, and determining first information according to the second request message; sending a second response message to the terminal equipment, wherein the second response message comprises the first information;

Or determining first information according to the capability information of the terminal equipment; and sending the first information to the terminal equipment.

In one possible implementation, the first information includes attribute information of the fourth AI model.

In one possible implementation, the first information includes first reference signal information, where the first reference signal information is reference signal information of a reference signal configured or transmitted by the network device, or reference signal information matched with a purpose of an AI model proposed by the terminal device, or reference signal information having an association relationship with attribute information of an AI model expected by the network device.

In one possible implementation, the first information further includes a reporting threshold; the method further comprises the following steps: and receiving the target channel quality information from the terminal equipment and the reference signal information corresponding to the target channel quality information, so that the network equipment can acquire the reference signal information used by the terminal equipment for retraining the AI model. Optionally, according to the target channel quality information and the reference signal information corresponding to the target channel quality information, updating the attribute information of the AI model with the association relation with the channel quality information, and implementing retraining of the AI model by the network equipment so as to ensure that the updated AI model is matched with the channel quality information.

In one possible implementation manner, the method further includes: and sending a third signaling to the terminal equipment, wherein the third signaling is used for indicating the association relation between the first model association information and the attribute information of the first AI model. That is, the network device may indicate the association relationship between the respective model association information and the attribute information of the respective AI models so that the network device and the terminal device understand the agreement for the association relationship between the AI models of the terminal device and the model association information.

In one possible implementation, the association of the first model association information with the attribute information of the first AI model is protocol predefined. That is, the protocol predefines the association relationship between the respective model association information and the attribute information of the respective AI models so that the network device and the terminal device understand the agreement for the association relationship between the AI models of the terminal device and the model association information.

In one possible implementation, the first model association information may include, but is not limited to, at least one of the following: first operator identification information, first PLMN information, first network equipment identification information, first TAI information, first SOR information, first URSP information, first operating frequency information, first cell information-associated reference signal information, first neighbor list, first timestamp information, first terminal equipment information;

Wherein the first cell information includes, but is not limited to, at least one of the following: first cell group information, first serving cell identification information, first physical cell identification information, first TRP identification information, first BWP identification.

In a third aspect, an embodiment of the present application provides an AI model updating method, which may be performed by a terminal device, or by an apparatus matched with the terminal device, for example, a chip module, or a processor. The method may include: receiving first signaling from a network device, the first signaling being used to indicate attribute information of a first AI model; the attribute information of the first AI model is different from the attribute information of the second AI model currently used, and the second AI model is updated to the first AI model.

It can be seen that when the terminal device receives the first signaling indicating the attribute information of the first AI model, the second AI model currently used can be updated to be the same as the first AI model, so that the terminal device can quickly update the AI model, thereby contributing to improvement of communication efficiency.

In one possible implementation, the attribute information of the first AI model is attribute information of an AI model desired by the network device.

In one possible implementation, the attribute information of the first AI model is different from the attribute information of the second AI model currently in use.

In one possible implementation, the attribute information of the first AI model has an association with terminal device information of the terminal device.

In one possible implementation, when updating the second AI model to the first AI model, the second AI model that is currently in use may be updated to the first AI model, in particular if the terminal device supports the first AI model. Therefore, in a mobile scene or a network switching scene, the second AI model currently used by the terminal equipment can be updated to be the first AI model indicated by the first signaling, and the communication quality and the communication efficiency can be improved.

Further, in the case that the terminal device supports the first AI model and the state of the first AI model is not the deactivated state, the second AI model is updated to the first AI model. That is, in the case where the state of the first AI model is the deactivated state, the second AI model is not updated to the first AI model.

In one possible implementation manner, in a case that the terminal device does not support the first AI model, selecting a target AI model from AI models supported by the terminal device, where the target AI model satisfies a similarity condition; updating the first AI model or the second AI model to the target AI model. In the case where the terminal device does not support the first AI model, a target AI model closest to attribute information of the first AI model may be selected, and communication efficiency may be improved to some extent.

In one possible implementation, in a case where the terminal device does not support the first AI model, a first request message is sent to the network device, where the first request message is used to indicate attribute information of the third AI model. That is, in the case where the terminal device does not support the first AI model, the terminal device requests the network device to use the third AI model. A first response message is received from the network device, and the first AI model or the second AI model is updated to a third AI model if the first response message indicates acknowledgement information. Conversely, in the case where the first response message indicates rejection information, the AI model update may be stopped, or the first request message may be sent again to the network device, at which time the first request message may indicate that the terminal device requests attribute information using the additional AI model (that is, the additional AI model is different from the third AI model), waiting for receipt of acknowledgement information or rejection information from the network device.

In one possible implementation, after updating the second AI model to the first AI model, the second AI model may be deactivated to avoid using 2 or more AI models simultaneously, reducing the processing load of the terminal device. Updating the second AI model to the first AI model implicitly represents activating and using the first AI model.

In a fourth aspect, an embodiment of the present application provides an AI model updating method, which may be performed by a network device, or by an apparatus matched with the network device, such as a chip, a chip module, or a processor. The method may include: and sending a first signaling to the terminal equipment, wherein the first signaling is used for indicating attribute information of the first AI model.

In a fifth aspect, an embodiment of the present application provides a communication apparatus including a communication unit and a processing unit. In one mode, the communication unit is configured to receive a first signaling from a network device, where the first signaling is configured to indicate first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the processing unit is used for updating the second AI model which is currently used into a first AI model according to the first model association information and the second model association information which is currently used; wherein the second model association information has an association relationship with attribute information of the second AI model. In another mode, the communication unit is configured to send a first signaling to the terminal device, where the first signaling is used to indicate first model association information, the first model association information has an association relationship with attribute information of the first AI model, and the first model association information is different from second model association information that is currently used.

In a sixth aspect, the application provides a communications apparatus comprising a processor, a memory and a computer program or instructions stored on the memory, characterised in that the processor executes the computer program or instructions to carry out the method as provided in any one of the first to fourth aspects.

In a seventh aspect, the present application provides a chip. In one implementation, the chip is configured to receive a first signaling from a network device, where the first signaling is configured to indicate first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the first model association information is different from the second model association information which is currently used, and the second AI model which is currently used is updated to be the first AI model; wherein the second model association information has an association relationship with attribute information of the second AI model. In another implementation manner, the chip is configured to send a first signaling to the terminal device, where the first signaling is configured to indicate first model association information, the first model association information has an association relationship with attribute information of the first AI model, and the first model association information is different from second model association information that is currently used.

In an eighth aspect, the present application provides a computer readable storage medium having stored therein computer readable instructions which, when run on a computer, cause the communication device to perform the method provided in any one of the first to fourth aspects above.

In a ninth aspect, the present application provides a computer program or computer program product comprising code or instructions which, when run on a computer, cause the computer to perform the method as provided in any of the first to fourth aspects.

In a tenth aspect, the present application provides a chip module, the chip module including a communication module, a power module, a memory module, and a chip, wherein: the power supply module is used for providing electric energy for the chip module; the storage module is used for storing data and instructions; the communication module is used for carrying out internal communication of the chip module or carrying out communication between the chip module and external equipment; the chip is for performing the method as provided in any one of the first to fourth aspects.

Drawings

FIG. 1 is a schematic diagram of the architecture of a communication system to which the present application is applied;

FIG. 2 is a schematic flow chart of an AI model updating method according to an embodiment of the application;

FIG. 3 is a flowchart illustrating another AI model update method according to an embodiment of the present application;

FIG. 4 is a flowchart of another AI model update method according to an embodiment of the application;

Fig. 5 is a schematic structural diagram of a communication device according to the present application;

fig. 6 is a schematic structural diagram of another communication device provided by the present application;

fig. 7 is a schematic structural diagram of a chip module according to an embodiment of the present application.

Detailed Description

In the present application, the words "first", "second", etc. are used to distinguish between identical items or similar items that have substantially the same function and effect. It will be appreciated by those of skill in the art that the words "first," "second," and the like do not limit the amount and order of execution, and that the words "first," "second," and the like do not necessarily differ. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship.

It should be understood that in the present application, "at least one" means one or more; "plurality" means two or more. In addition, "equal" of the present application may be used in conjunction with "greater than" or "less than". Under the condition of being equal to and being greater than, adopting a technical scheme of being greater than; under the condition of being used together with 'equal to' and 'less than', the technical scheme of 'less than' is adopted.

First, a system architecture to which the present application relates will be explained.

The present application is applicable to fifth generation (5th generation,5G) systems, which may also be referred to as New Radio (NR) systems; or may be applied to sixth generation (6th generation,6G) systems, or seventh generation (7th generation,7G) systems, or other future communication systems; or may also be used in device-to-device (D2D) systems, machine-to-machine (machine to machine, M2M) systems, internet of vehicles (vehicle to everything, V2X), and the like.

The application is applicable to the system architecture shown in fig. 1. The communication system 10 shown in fig. 1 may include, but is not limited to: network device 110 and terminal device 120. The number and form of the devices in fig. 1 are used for illustration, and are not limited to the embodiments of the present application, and for example, a plurality of terminal devices may be included in practical applications.

A terminal device, also called User Equipment (UE), mobile Station (MS), mobile Terminal (MT), etc., is a device that provides voice and/or data connectivity to a user. Such as a handheld device, an in-vehicle device, etc., having a wireless connection function. Currently, examples of some terminal devices are: a mobile phone), a tablet, a notebook, a palm, a mobile internet device (mobile INTERNET DEVICE, MID), a wearable device, a Virtual Reality (VR) device, an augmented reality (augmented reality, AR) device, a wireless terminal in industrial control (industrial control), a wireless terminal in unmanned (SELF DRIVING), a wireless terminal in teleoperation (remote medical surgery), a wireless terminal in smart grid (SMART GRID), a wireless terminal in transportation security (transportation safety), a wireless terminal in smart city (SMART CITY), a wireless terminal in smart home (smart home), and the like.

In the present application, the means for realizing the functions of the terminal device may be the terminal device; or a device, such as a chip or a chip module, which can be installed in or matched with the terminal device, capable of supporting the terminal device to implement the function. In the technical solution provided in the present application, the device for implementing the function of the terminal device is an example of the terminal device, and the technical solution provided in the present application is described. The terminal device according to the embodiment of the present application has an AI model or is described as being deployed with an AI model. The number of AI models supported by the terminal device is not limited in the present application, and depends on factory settings of the terminal device or depends on actual conditions.

In one implementation, the network device may be an access network device, which refers to a radio access network (radio access network, RAN) node (or device), also referred to as a base station, that accesses the terminal device to the wireless network. Currently, some examples of RAN nodes are: a Node B (gNB) that continues to evolve, a transmit-receive point (transmission reception point, TRP), an evolved Node B (eNB), a radio network controller (radio network controller, RNC), a Node B (Node B, NB), a base station controller (base station controller, BSC), a base transceiver station (base transceiver station, BTS), a home base station (e.g., home evolved NodeB, or home Node B, HNB), a baseband unit (BBU), or a wireless fidelity (WIRELESS FIDELITY, wifi) access point (access point, AP), etc. In addition, in one network architecture, the network device may include a centralized unit (centralized unit, CU) node, or a Distributed Unit (DU) node, or a RAN device including a CU node and a DU node. It should be noted that, the centralized unit node and the distributed unit node may also use other names, and the present application is not limited thereto.

In another implementation, the network device may be a core network device. The core network device is mainly responsible for managing and controlling the whole network, sorting the received data and transmitting the data to the corresponding terminal device or Data Network (DN). The core network device may also be referred to as a core network node. In the present application, the core network device may be, for example, a network DATA ANALYTICS function (NWDAF) node or a location management function (location management function, LMF) node. And the NWDAF node is used for providing network analysis service according to the request data of the network service. The LMF node is used for overall coordination and scheduling of resources required by the position of the terminal equipment and is also used for providing positioning services.

In yet another implementation, the network device may be a neural network processing node, meaning a node that provides services to the AI model, such as providing model parameters for the AI model.

In the present application, the means for realizing the functions of the network device may be the network device; or may be a device, such as a chip or a chip module, capable of supporting the network device to implement the function, and the device may be installed in the network device or used in cooperation with the network device. In the technical solution provided in the present application, the device for implementing the function of the network device is exemplified by the network device, and the technical solution provided in the present application is described.

It may be understood that, the communication system described in the embodiment of the present application is for more clearly describing the technical solution of the embodiment of the present application, and does not constitute a limitation on the technical solution provided by the embodiment of the present application, and those skilled in the art can know that, with the evolution of the system architecture and the appearance of a new service scenario, the technical solution provided by the embodiment of the present application is equally applicable to similar technical problems.

Next, related names or terms related to the present application are set forth to facilitate understanding by those skilled in the art.

1. AI model

AI models may be used to implement AI functionality, and may also be described as machine learning models. The AI model may be implemented by at least one of a neural network, a decision tree, a support vector machine, a bayesian classifier, and the like.

Taking a neural network as an example, the neural network is composed of neurons, including: input parameters, multiplicative coefficients, additive coefficients, activation functions, etc. Common activation functions include: sigmoid function, hyperbolic function (tanh function), linear rectifying unit (RECTIFIED LINEAR unit, reLU) function, and the like.

The parameters of the neural network are optimized by a gradient optimization algorithm, which is a class of algorithms that minimizes or maximizes an objective function (also known as a loss function), which 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, with the model, the output information can be predicted from the input, and the gap between the predicted value and the true value can be calculated, which is the loss function. The purpose of building a neural network model is to minimize the value of the loss function as much as possible. The smaller the loss value, the closer the neural network model is to the real situation.

The error back propagation (errorBackPropagation, BP) algorithm is a common optimization algorithm used to optimize the AI model. The basic idea of the BP algorithm is that the learning process consists of two processes, forward propagation of the signal and backward propagation of the error. In forward propagation, input samples are transmitted from the input layer, processed layer by each hidden layer, and then transmitted to the output layer. If the actual output of the output layer does not match the desired output, the back propagation phase of the error is shifted. For convenience of description, the backward propagation phase of the error will be simply referred to as error back propagation. The error back transmission is to make the output error pass through hidden layer to input layer in a certain form and to distribute the error to all units of each layer, so as to obtain the error signal of each layer unit, which is used as the basis for correcting the weight of each unit. The weight adjustment process of each layer of forward propagation and error feedback of the signal is carried out repeatedly. The constant weight adjustment process is the learning and training process of the AI model. This process is continued until the error in the AI model output is reduced to an acceptable level or until a preset number of learning times.

The optimization algorithm that optimizes the AI model may also be a gradient descent (GRADIENT DESCENT) algorithm, a random gradient descent (Stochatic GRADIENT DESCENT, SGD) algorithm, a small-batch gradient descent (mini-batch GRADIENT DESCENT) algorithm, a Momentum method (Momentum), a random gradient descent with Momentum (Nesteroy) algorithm, an adaptive gradient descent (ADAPTIVE GRADIENT DESCENT, adagrad) algorithm, a root mean square error descent (Root Mean Square Prop) algorithm, an adaptive Momentum estimation (Adaptive Moment Estimation, adam) algorithm, and the like.

When the error is transmitted reversely, the optimization algorithm obtains the gradient by calculating the derivative/partial derivative of the current neuron according to the error obtained by the loss function and adding the influence of the learning rate, the previous gradient/derivative/partial derivative and the like, and transmits the gradient to the upper layer.

The AI model related to the embodiment of the application refers to an AI model applied to the technical field of wireless communication, and can be used for improving the throughput of a wireless communication system, shortening the processing time delay, enhancing the user capacity and the like. The AI model related to the embodiment of the application refers to the AI model of the terminal equipment, so that the communication capacity of the terminal equipment can be improved.

(1) Attribute information of AI model

The attribute information of the AI model is used to describe at least one of identification information of the AI model, model structure information of the AI model, and model parameters of the AI model. In other words, the attribute information of the AI model may include at least one of: identification information of the AI model, model structure information of the AI model, and model parameters of the AI model.

The identification information is used to identify AI models, for example, may be a model Identifier (ID), and different model IDs identify different AI models, and different AI models have different model IDs. The model structure information is used to describe the model structure, and may be, for example, structural information such as a neural network, a decision tree, or neurons and layers in the model. Different model structures may be considered as different AI models. The model parameters may include at least one parameter including, for example, an activation function, multiplicative coefficients, additive coefficients, and the like. Different model parameters may be considered different AI models. For example, two AI models, except for the activation function, are different, and the other model parameters are the same, and the two AI models are considered to be different AI models.

In the embodiment of the present application, the name of the attribute information of the AI model is used as an example, and may be described as model information of the AI model, feature information of the AI model, basic information of the AI model, or the like.

2. Model association information

In the embodiment of the application, the model association information is used for describing information with association relation with the AI model or information with association relation with attribute information of the AI model. Changes in the model association information may result in changes in the AI model. The association relationship between the model association information and the AI model may be a one-to-one relationship, that is, one type of model association information associates one type of AI model; the relation of many to one can be adopted, namely, multiple model association information can be associated with one AI model; one-to-many relationship is also possible, i.e. one model association information may associate a plurality of AI models.

The model association information may include, but is not limited to, at least one of the following: PLMN information, operator identification information, network equipment identification information, TAI information, SOR information, URSP information, operating frequency information, cell information, reference signal information associated with cell information, neighbor list, time stamp information, terminal equipment information.

1) PLMN information is used to describe a cellular mobile communication network of a certain standard of a certain operator of a certain region. The PLMN information is used, for example, to describe the 5G-mode network of the operator 1 of the home location of the terminal equipment. As another example, PLMN information is used to describe the 5G-mode network of operator 2 of the visited place of the terminal equipment.

2) The operator identification information is used to identify an operator, for example, an operator to which the terminal device signs up, or an operator to which the terminal device currently belongs, or the like.

3) The network device identification information is used to identify the network device manufacturer, and may be, for example, the manufacturer of the base station or the core network device manufacturer, etc.

4) The TAI information is used to identify the TAI, and may be expressed as tai=plmn+tracking area code (TRACKING AREA code, TAC). The tracking area (TRACKING AREA, TA) is a concept set up for location management of the terminal device. When the terminal equipment is in an idle state, the core network equipment can know the TA where the terminal equipment is located, and paging is carried out on all cells of the TA registered by the terminal equipment when the terminal equipment is in the idle state. The TA is a configuration at a cell level, a plurality of cells may configure the same TA, and one cell belongs to one TA.

5) The SOR information is used for the home PLMN to be able to direct the terminal equipment from one network to another. SOR is a technique that encourages roaming of roaming terminal devices to preferred roaming networks by the home PLMN. For example, the terminal device is registered on one PLMN, and for some reason the home PLMN of the terminal device wishes the terminal device to register on another PLMN.

6) URSP information is used to describe the routing policy of the terminal device. The policy describes a correspondence between applications on the terminal device and network slices, which the terminal device may select for the applications according to URSP.

7) The operating frequency information is used to describe a frequency that the terminal device will use or is using or is available to use, and may include, but is not limited to, at least one of a frequency number, a frequency band number, etc.

8) The cell information is used to describe serving cell information of the terminal device, and may include, but is not limited to, at least one of the following information: cell group information, serving cell identification information, physical cell identification (PHYSICAL CELL IDENTIFIER, PCI), TRP identification information, BWP identification. The cell group information may include a cell group identification (cell group ID) for identifying a cell group to which the serving cell belongs. The serving cell identification information may include a serving cell identification (SERVING CELL ID) for identifying the serving cell. PCI, used to identify the logical cell of the service cell. The TRP identification information may include, but is not limited to, at least one of: control-resource set (CORESET) identity, control-resource set pool (CORESET pool) identity, group identity of control resource set, transmission configuration indication (transmission configuration indication, TCI) status pool identity, TCI status identity, reference signal resource set identity. BWP identification for identifying BWP.

9) Reference signal (REFERENCE SIGNAL, RS) information associated with the cell information, which is used to describe the reference signal associated with the cell information, may include, but is not limited to, the number of reference signal resources, a set of reference signal resources, an index of reference signal resources, a pattern of reference signals, a repetition parameter value of the reference signal (e.g., repetition off or repetition on, repetition off indicating that the repetition of the reference signal is not turned on, repetition on indicating that the repetition of the reference signal is turned on). The reference signal refers to a downlink reference signal, and may be, for example, a synchronization signal block (synchronization signal block, SSB) or a channel state information reference signal (CHANNEL STATE information-REFERENCE SIGNAL, CSI-RS).

10A neighbor list describing neighbor cells of the serving cell. The neighbor list may be configured by the network device through higher layer signaling.

11 Timestamp information is used to describe the time interval of the AI model, such as the time interval of training or the time interval of application, etc. It is understood that the time stamp information is the effective time of the AI model. The network device may broadcast the timestamp information of the AI model.

12 Terminal device information for describing identification information of the terminal device, may include, but is not limited to, at least one of device identification information of the terminal device, and the like. Wherein the device identification information of the terminal device is used for identifying the manufacturer of the terminal device, such as the brand of the mobile phone, etc. The identification information of the terminal device is used to uniquely identify the terminal device and may include, but is not limited to, an international mobile subscriber identity (international mobile subscriber identity, IMSI), a subscriber permanent identifier (subscription PERMANENT IDENTIFIER, SUPI), a subscriber hidden identifier (subscription concealed identifier, SUCI), etc.

Further, the equipment manufacturer identification information of the terminal equipment and the identification information of the terminal equipment have an association relationship with the AI model. For example, AI model 1 is applicable to a cell phone with cell phone brand a, and AI model 2 is applicable to a cell phone with cell phone brand B. The network device may broadcast the device identification information of the terminal device to which the AI model applies, or broadcast the identification information of the terminal device to which the AI model applies. For example, the network device may broadcast that AI model 1 is applicable to a cell phone with cell phone brand A.

The above listed model association information is used for example, and is not meant to limit the embodiments of the present application, and the model association information may also include other types of information. The information included in the model association information may vary with changes in the application scenario, such as increasing or decreasing certain information.

3. The attribute information of the AI model and the model association information have an association relation

The attribute information of the AI model has an association relationship with the model association information, and in fact, the AI model has an association relationship with the model association information. Specifically, the association relationship between the attribute information of each AI model in at least one AI model and the model association information under different environments or different conditions. The association relationship may be a one-to-one relationship, a many-to-one relationship, or a one-to-many relationship.

For example, the model association information is operator identification information, and the association relationship between the attribute information of 2 AI models and 2 pieces of operator identification information can be seen in table 1 below.

TABLE 1

Model ID of AI model	Operator identification information	Model structure information/model parameters
			1	Operator identity A	Model parameter 1
2	Operator identity B	Model parameters 2

In table 1, the model IDs of the 2 AI models are 1 and 2, respectively, the model ID 1 and the model parameter 1 have an association relationship with the operator identifier a, and the model ID 2 and the model parameter 2 have an association relationship with the operator identifier B.

In one implementation, the association between the attribute information of the AI model and the model association information is indicated by the network device, e.g., the network device may indicate the association between the attribute information of the AI model and the model association information via higher layer signaling or downlink control information (downlink control information, DCI) or a broadcast message. For example, the terminal device supports 2 AI models, and the network device may indicate an association relationship between each of the 2 AI models and the respective associated model association information. It is understood that when indicating the association relationship, the network device may implicitly indicate attribute information and model association information of the AI model having the association relationship. I.e., the network device may indicate attribute information of the AI model, as well as model association information to which the attribute information is associated. The network equipment indicates the association relation, so that the network equipment and the terminal equipment understand the association relation between the AI model of the terminal equipment and the model association information consistently.

In another implementation, the association relationship between the attribute information of the AI model and the model association information is determined by the terminal device. That is, the terminal device maintains the association relationship. Optionally, the terminal device may report the determined association relationship to the network device, so that the network device and the terminal device understand and agree on the association relationship between the AI model of the terminal device and the model association information.

In yet another implementation, the association between the attribute information of the AI model and the model association information is predefined by the protocol. The association relation is predefined by the protocol, so that the network equipment and the terminal equipment understand and agree with each other on the association relation between the AI model of the terminal equipment and the model association information.

In the embodiment of the present application, updating a certain AI model can also be understood as at least one of the following: updating (or switching) the AI model to another AI model; deactivating the AI model and activating a new AI model; updating attribute information of the AI model; training the AI model; fine tuning the AI model; adjusting the AI model; testing the AI model; supervising the AI model; check/verify the AI model, etc.

The method for updating the AI model provided by the embodiment of the application is explained in detail below.

Fig. 2 is a schematic flow chart of an AI model updating method according to an embodiment of the present application. Taking the example that the terminal device executes the flow shown in fig. 2, the flow shown in fig. 2 may specifically include the following steps:

The network device sends 201 a first signaling to the terminal device. Correspondingly, the terminal device receives the first signaling from the network device. The first signaling is used for indicating first model association information, and the first model association information has association relation with attribute information of the first AI model.

The first model association information may include, but is not limited to, at least one of the following: the method comprises the steps of first PLMN information, first operator identification information, first network equipment identification information, first TAI information, first SOR information, first URSP information, first operating frequency information, first cell information, reference signal information associated with the first cell information, a first neighbor list, first timestamp information and first terminal equipment information. Wherein the first cell information includes, but is not limited to, at least one of the following: first cell group information, first serving cell identification information, first PCI, first TRP identification information, first BWP identification. The information included in the first model association information may refer to the specific description of the information included in the model association information, which is not described herein.

The first signaling is different according to different contents included in the first model association information.

For example, the first model association information is a first neighbor list, and the first signaling may be radio resource control (radio resource control, RRC) signaling used to configure the first neighbor list. That is, the first signaling may be RRC signaling for configuring the first neighbor list.

For example, the first model association information is a first PCI, and the first signaling may be a medium access control-control element (MAC-CE) signaling for indicating activation of the first PCI. That is, the first signaling may be MAC-CE signaling indicating activation of the first PCI.

For example, the first model association information is first cell information, and the first signaling may be RRC signaling, where the RRC signaling is used to instruct cell handover, and specifically instruct handover to the first cell. The RRC signaling may directly or implicitly indicate the first cell information. The RRC signaling may be, for example, RRC reconfiguration (RRCReconfiguration) signaling. That is, the first signaling may be RRC signaling indicating a cell handover. Or the first signaling may be MAC-CE signaling indicating activation of the first cell. The MAC-CE signaling may directly or implicitly indicate the first cell information. That is, the first signaling may be MAC-CE signaling indicating activation of the first cell.

For example, the first model association information is a first BWP identification, and the first signaling may be DCI, where the DCI is used to indicate a BWP handover, and specifically indicates a handover to the first BWP. The DCI may directly or implicitly indicate the first BWP identity. That is, the first signaling may be DCI indicating BWP handover.

The first model association information is first SOR information or first URSP information, and the first signaling may be a signaling that the core network device directly sends to the terminal device, or may be a signaling that the core network device sends to the terminal device through the base station.

The first model association information has an association relationship with attribute information of the first AI model. In one implementation, the network device sends the third signaling to the terminal device, and the terminal device receives the third signaling from the network device, accordingly. The third signaling is used for indicating the association relation between the first model association information and the attribute information of the first AI model. The association relation between the AI model of the terminal equipment and the model association information is understood to be consistent by the network equipment and the terminal equipment. In another implementation, the terminal device determines or maintains an association relationship between the first model association information and attribute information of the first AI model. In a further implementation, the association between the first model association information and the attribute information of the first AI model is pre-defined by a protocol such that the network device and the terminal device understand the agreement for the association between the AI model of the terminal device and the model association information.

202, The first model association information is different from the second model association information currently used, and the second AI model currently used is updated to the first AI model. Wherein the second model association information and the attribute information of the second AI model have an association relationship.

The second model association information currently used, i.e., the current model association information, can be understood as the model association information currently being used. For example, URSP information currently in use is second URSP information, or working frequency information currently in use is second working frequency information, etc. The information included in the second model association information may refer to the specific description of the information included in the foregoing model association information, which is not described herein.

When the terminal equipment receives the first signaling, the terminal equipment can judge whether the first model association information indicated by the first signaling is the same as the second model association information currently used. The first model association information and the second model association information are different in at least one type of model association information, and the first model association information and the second model association information can be judged to be different. For example, the first PCI in the first model association information is different from the second PCI in the second model association information, and it may be determined that the first model association information is different from the second model association information.

When the first model association information is judged to be different from the second model association information, the terminal device may update the second AI model currently used to the first AI model. That is, the terminal device updates the second AI model currently used to the first AI model having an association relationship with the first model association information.

When the terminal device updates the currently used second AI model to the first AI model, it may be updated in the following two cases.

In case 1, in the case where the first AI model is different from the second AI model and the terminal device supports the first AI model, the second AI model is updated to the first AI model. That is, when the terminal device updates the AI model, not only the AI model is considered, but also whether the terminal device supports a new AI model needs to be considered, so as to avoid the situation that the updated AI model is not available.

In case 2, in a case where the first AI model is different from the second AI model and the terminal device supports the first AI model and the state of the first AI model is not the deactivated state, the second AI model is updated to the first AI model. That is, the terminal device also considers whether the new AI model is in a deactivated state when considering that the AI model is different, whether the terminal device supports the new AI model. If the state is not the deactivation state, updating to a new AI model; if in the deactivated state, the new AI model cannot be updated. Whether the state of the first AI model is a deactivated state may be indicated by the network device or autonomously determined by the terminal device. For example, the terminal device deactivates the AI model 1 after updating the AI model 2 from the AI model 1, and determines that the AI model 2 can be updated to the AI model 1 when the first model-related information is detected, but the state of the AI model 1 is the deactivated state, so that the AI model 2 is not updated to the AI model 1. The change of state of the AI model may be determined by the terminal device or indicated by the network device.

Case 1 and case 2 described above support the first AI model for the terminal device, and if the terminal device does not support the first AI model, the following manner 1 or manner 2 may be adopted.

In mode 1, when the first AI model is different from the second AI model and the terminal device does not support the first AI model, the target AI model is selected from the AI models supported by the terminal device based on the first model association information, and the first AI model or the second AI model is updated to the target AI model. That is, in the case that the terminal device does not support the first AI model, the terminal device selects a suboptimal AI model, which is the target AI model. That is, the terminal device may update the first AI model or the second AI model to the target AI model when it finds that the first AI model is not supported.

For example, the terminal device supports AI model 1 and AI model 2, the terminal device moves from the network of operator 1 to the network of operator 3 (i.e., AI model having an association with the current model association information is model 1), AI model having an association with the first model association information is AI model 3, but the terminal device does not support AI model 3. The terminal device then selects AI model 2 as the target AI model. Only AI model 2 is associated with the network of operator 2 and not with the network of operator 3. Further, the terminal device may establish an association relationship between the attribute information of the AI model 2 and the first model association information after updating the AI model 1 to the AI model 2.

Wherein the target AI model satisfies the similarity condition, may include at least one of:

1) The difference between the model identification of the first AI model and the model identification of the target AI model is less than or equal to a first threshold. That is, the model ID of the first AI model is closest to the model ID of the target AI model. For example, the first threshold is 1, the model ID of the first AI model is 1, and the range of values of the model IDs of the AI models supported by the terminal device is {2,3}, then the AI model with the model ID of 2 is the target AI model. Further, if the model ID of the second AI model is 2, the second AI model may not be updated, or attribute information of the second AI model may be updated to better accommodate the change of the communication environment.

2) The first model association information is the same as or similar to the model association information associated with the target AI model by more than a second threshold. The greater the similarity of the first model-associated information and the model-associated information associated with the target AI model, the higher the degree of matching of the first AI model and the target AI model, and the closer the two AI models are. It should be noted that the second threshold may be different for different types of model association information. For example, the model association information is TAI information, and the second threshold may be 80%; the model association information is USRP information and the second threshold may be 90%. For the model association information including multiple types of information, comparing the similarity of one model association information to be greater than the corresponding second threshold value can be regarded as the overall similarity to be greater than the second threshold value. For example, the first model association information includes first TAI information and first USRP information, and if the similarity of the first TAI information and the TAI information associated with the target AI model is greater than a second threshold, the similarity of the first model association information and the model association information associated with the target AI model is considered to be greater than the second threshold.

3) The target AI model is an AI model associated with preset cell information. Optionally, the preset cell information is included in the configured or activated cell information, e.g. the target AI model is the AI model associated with the default BWP in the network device-activated BWP. Optionally, the preset cell information is cell information associated with a target TCI state, where the target TCI state is indicated by the second signaling or configured or activated. Wherein a time difference between the reception time of the second signaling and the current system time is less than a third threshold. That is, the target TCI state is the TCI state indicated or configured or activated by the second signaling sent by the nearest network device. For example, the second signaling is higher layer signaling, and the target TCI state is a TCI state configured by the higher layer signaling. For another example, the second signaling is MAC-CE signaling and the target TCI state is the TCI state activated by the MAC-CE signaling. The target TCI state may also be understood as the most recently used TCI state. So that the cell information associated with the target AI model can be understood as the cell information associated with the target TCI state. Optionally, the preset cell information includes a default cell identifier in the neighbor cell list, so that the target AI model is an AI model associated with the default cell identifier in the neighbor cell list configured by the network device for the terminal device.

4) The target AI model is the AI model most frequently used or the AI model last used in a second preset period of time before updating the second AI model. The specific value range of the second preset time period is not limited in the embodiment of the present application. The AI model last used before the second AI model is updated refers to the AI model used before the second AI model is used.

In mode 2, in a case where the first AI model is different from the second AI model and the terminal device does not support the first AI model, a first request message is sent to the network device, the first request message being used to indicate attribute information of the third AI model. That is, in the case where the terminal device does not support the first AI model, the terminal device transmits a first request message to the network device indicating attribute information of the third AI model, which may be different from the first AI model, indicating that the terminal device requests use of the third AI model. Alternatively, the third AI model may be the target AI model, or may be another AI model. The terminal device receives a first response message from the network device, and updates the first AI model or the second AI model to a third AI model if the first response message indicates acknowledgement information. Conversely, in the case where the first response message indicates rejection information, the AI model update may be stopped, or the first request message may be sent again to the network device, at which time the first request message may indicate that the terminal device requests attribute information using the additional AI model (that is, the additional AI model is different from the third AI model), waiting for receipt of acknowledgement information or rejection information from the network device.

Optionally, the first request message does not specify attribute information of any AI model, and the first request message is used for requesting the network device to indicate attribute information of one AI model. The network device sends a first response message to the terminal device in case of receiving the first request message, where the first response message is used to indicate attribute information of the new AI model. The terminal device may update the second AI model to a new AI model upon receipt of the first response message. Further, the terminal device may establish an association relationship between the new AI model and the first model association information.

It will be appreciated that the terminal device finds that the first AI model is not supported before or after updating to the first AI model, and the terminal device in turn sends a first request message to the network device for requesting use of the third AI model or the target AI model. The terminal device updates the first AI model or the second AI model to the third AI model or the target AI model, upon receiving a first response message indicating the acknowledgement information. Before updating to the first AI model, the terminal equipment finds that the first AI model is not supported, and updates the second AI model to a third AI model or a target AI model under the condition that a first response message indicating acknowledgement information is received; after updating to the first AI model, the terminal device finds that the first AI model is not supported, and updates the first AI model to a third AI model or a target AI model in the case that a first response message indicating acknowledgement information is received.

Case 1 and case 2 described above, and modes 1 and 2, are different for the first AI model and the second AI model. If the first AI model is the same as the second AI model, the terminal device updates the association relationship between the current model association information and the attribute information of the second AI model to the association relationship between the first model association information and the attribute information of the second AI model. Thereby, the association relation between the model association information and the attribute information of the AI model can be updated in real time.

The above-described modes 1 and 2 are directed to the terminal device not supporting the first AI model. If the terminal device finds that at least one of the following defects 1 to 3 exists before or after updating the second AI model to the first AI model, the terminal device may select the target AI model from the AI models supported by the terminal device based on the first model association information.

Defect 1, the output result of the first AI model does not meet the preset condition, that is, the output result of the first AI model does not meet the performance requirement, that is, when the output result of the first AI model does not meet the performance requirement, the first AI model may be updated to the target AI model. The similarity condition satisfied by the target AI model may be referred to in the specific description of mode 1, and will not be described herein.

The defect 2, the state of the first AI model is the deactivated state, that is, the first AI model is indicated by the network device as the deactivated state or the terminal device determines that the state of the first AI model is the deactivated state, and the first AI model or the second AI model may be updated to be the target AI model. If the terminal device has found the state of the first AI model to be a deactivated state before updating the second AI model to the first AI model, the second AI model is updated to the target AI model. If the terminal device finds that the state of the first AI model is a deactivated state after updating the second AI model to the first AI model, the first AI model is updated to the target AI model.

Defect 3, the first AI model is different from the AI model indicated by the AI model indication information. Wherein the AI model indication information is sent by the network device, different from the first signaling and the second signaling. Alternatively, the network device may send AI model indication information to the terminal device when the terminal device accesses. If the terminal device finds that the network device-indicated AI model is not the first AI model before updating the second AI model to the first AI model, the second AI model may be updated to the target AI model. If, after updating the second AI model to the first AI model, the terminal device finds that the AI model indicated by the network device is not the first AI model, the first AI model may be updated to the target AI model.

For the above-mentioned defects 1 to 3, the terminal device may employ mode 2 to send, to the network device, a first request message indicating attribute information of the third AI model, and update the first AI model or the second AI model to the third AI model according to the acknowledgement information when receiving the first response message indicating the acknowledgement information from the network device, in addition to selecting the target AI model. Or the first request message is used for requesting the network device to indicate the attribute information of one AI model, and when the first response message is received and indicates the attribute information of a new AI model, the first AI model or the second AI model is updated to the new AI model.

As is clear from the above-described modes 1, 2 and defects 1 to 3, in the case where the first AI model is not available, the first AI model may be updated to the third AI model or the target AI model; or the second AI model may be updated to a third AI model or a target AI model. The cases where the first AI model is not available may include, but are not limited to, at least one of:

(1) The terminal equipment does not support the first AI model;

(2) The output result of the first AI model does not meet the preset condition;

(3) The state of the first AI model is a deactivated state;

(4) The first AI model is different from the AI model indicated by the AI model indication information.

Alternatively, for cases (1) to (4), the terminal device may also update the second AI model to an AI model determined by measuring a reference signal transmitted by the network device. For example, the terminal device obtains a measurement result (including, for example, channel quality information) by measuring a reference signal transmitted by the network device in the first cell, determines an AI model according to the measurement result, and updates the second AI model to the AI model. Or the terminal device may also update the first AI model to an AI model determined by measuring a reference signal transmitted by the network device.

Optionally, after updating the second AI model, the terminal device may deactivate the second AI model, so as to avoid using 2 or more AI models simultaneously, thereby reducing the processing load of the terminal device.

In the embodiment shown in fig. 2, when the terminal device receives the first signaling indicating the first model association information, the second AI model currently used may be updated to the first AI model having an association relationship with the first model association information, so that the terminal device may quickly update the AI model, thereby contributing to improving communication efficiency.

Please refer to fig. 3, which is a flowchart illustrating another AI model updating method according to an embodiment of the present application, the method specifically includes the following steps:

The network device sends 301 a first signaling to the terminal device. Correspondingly, the terminal device receives the first signaling from the network device. The first signaling is used for indicating first model association information, and the first model association information has association relation with attribute information of the first AI model.

The implementation process of step 301 may refer to the specific description of step 201 in the embodiment shown in fig. 2, which is not repeated here. The terminal device may update the second AI model directly to the first AI model upon receiving the first signaling, or perform steps 302-304 if the first AI model is not available, or perform steps 302-304 if the terminal device does not accept the use of the first AI model.

302, The terminal device sends a second request message to the network device. Accordingly, the network device receives the second request message from the terminal device.

Wherein the second request message may include, but is not limited to, at least one of the following information:

And A, the application of the AI model suggested by the terminal equipment. The use of the AI model proposed by the terminal device may be used for the network device to provide the terminal device with corresponding reference signal information, e.g. to provide corresponding reference signal resources, based on the use. Uses for AI models may include, but are not limited to: training, reasoning, updating, switching, activating, deactivating, monitoring, etc. Training may be classified as first training or retraining, and training may be training model parameters of an AI model, and the like.

And B, attribute information of the AI model suggested by the terminal equipment. The attribute information of the AI model suggested by the terminal device may be used by the network device to determine whether to grant the terminal device to use the AI model. Attribute information of the terminal device suggested AI model, for example, at least one of a model ID of the terminal device suggested AI model, suggested model structure information, or suggested model parameters.

And C, the AI model updating time suggested by the terminal equipment is used for indicating the time required for updating the AI model determined by the terminal equipment. Alternatively, the suggested AI model update time period may include at least one of a network switch time period, a cell switch time period, a BWP switch time period, and a beam switch time period.

And D, the reference signal information recommended by the terminal equipment is used for indicating the terminal equipment to recommend how the network equipment transmits the reference signal. The reference signal information suggested by the terminal device may include, but is not limited to: at least one of a number of reference signal resources, a pattern of reference signals, a repetition parameter value of the reference signals (e.g., repetition off or repetition on, repetition off indicating that no repetition of the reference signals is enabled, repetition on indicating that repetition of the reference signals is enabled), a reporting threshold of the reference signals.

E, the number of reception beams expected by the terminal device is used to instruct the terminal device how many transmission beams the network device is expected to use to transmit the reference signal, so as to match the number of reception beams of the terminal device.

Further, the uplink resource corresponding to the second request message may include, but is not limited to, at least one of the following resources:

1) A pre-configured periodic physical uplink channel resource, e.g., a pre-configured periodic physical uplink control channel (physical uplink control channel, PUCCH) resource.

2) The uplink grant resource with the highest frequency is used in a first preset time period before the second request message is sent. The specific value range of the first preset time period is not limited in the embodiment of the present application. The first preset time period and the second preset time period may be the same or different.

3) Uplink grant resource used last time.

4) The next available uplink grant resource. The next available uplink grant resource (next available uplink grant) may also be described as a last available uplink grant resource or an uplink grant resource to be used, where "available" refers to that a last-arrived uplink grant resource can carry the second request message. The next available uplink grant resource may be a dynamic grant resource (DYNAMIC GRANT) or a configured grant resource (configured grant).

5) Random Access (RACH) resources associated with the first model association information. Optionally, the network device pre-configures an association relationship between the random access resource and the model association information, and the UE determines, according to the detected first model association information, the random access resource associated by using the first model association information, and sends the second request message.

6) Random access resources corresponding to the reference signal information associated with the first model association information. For example, the terminal device determines to use the RACH resource corresponding to the SSB of the first cell according to detecting the RACH resource corresponding to the SSB of the first cell, and sends the second request message.

303, The network device sends a second response message to the terminal device. Correspondingly, the terminal device receives a second response message from the network device.

Wherein the second response message is for responding to the second request message. The network device determines the first information according to the second request message. The second response message may include first information, which may be at least one of the following:

A1, reference signal information matched with the purpose of the AI model suggested by the terminal equipment. The A1 information corresponds to the a information in the second request message. That is, the terminal device requests which use of the AI model, and the network device provides the terminal device with the corresponding reference signal information.

B1, confirmation information or rejection information. The B1 information corresponds to the B information in the second request message. And B information is attribute information of the AI model suggested by the terminal equipment. For example, the terminal device suggests using the fourth AI model, i.e. the second request message includes attribute information of the fourth AI model suggested by the terminal device, and if the network device agrees to use the fourth AI model by the terminal device, the first information is acknowledgement information, where the acknowledgement information indicates that the network device agrees to use the requested AI model by the terminal device; if the network device does not agree that the terminal device uses the fourth AI model, the first information is reject information, which indicates that the network device does not agree that the terminal device uses the requested AI model. When the terminal device receives the rejection information, the AI model update may be stopped, or a second request message may be sent to the network device again.

Optionally, B1, attribute information of the AI model suggested by the terminal device. That is, the first information includes attribute information of the AI model suggested by the terminal device in the second request message. It is understood that the attribute information of the AI model in the second request message is the same as the attribute information of the AI model in the first information, i.e., the network device agrees to the terminal device to use the requested AI model. For example, the terminal device suggests using the fourth AI model, i.e. the second request message includes attribute information of the fourth AI model suggested by the terminal device, and if the network device agrees to use the fourth AI model by the terminal device, the first information includes attribute information of the fourth AI model; if the network device does not agree that the terminal device uses the fourth AI model, the first information includes attribute information of another AI model, or the first information includes D1 information. Wherein the fourth AI model may be the same as or different from the first AI model.

Optionally, B1, attribute information of the AI model desired by the network device. The attribute information of the AI model desired by the network device is independent of the B information in the second request message. That is, the network device directly feeds back to the terminal device attribute information of the AI model that the network device expects the terminal device to use, regardless of which AI model the terminal device requests to use. When the terminal equipment receives the attribute information of the AI model expected by the network equipment, if the terminal equipment agrees to use the AI model, the terminal equipment can update the second AI model into the AI model; if the terminal device does not agree to use the AI model, the AI model update may be stopped or a second request message may be sent again to the network device.

C1, the appointed AI model updating time length is the time length required by the AI model updating appointed by the network equipment. Optionally, the specified AI model update time period may include at least one of a network switch time period, a cell switch time period, a BWP switch time period, and a beam switch time period. For example, when the terminal device detects the first operator network and attribute information of the proposed AI model corresponding to the first operator network is included in the second request message, the network device determines a specified AI model update duration according to the second request message, and the duration may be greater than or equal to a duration required for the terminal device to complete the handover to the first operator network. The C1 information may or may not correspond to the C information in the second request message. That is, the network device may determine the specified AI model update time period according to the AI model update time period suggested by the terminal device; the network device may also determine the specified AI model update time period based on other information in the second request message.

D1, reference signal information of reference signals configured or transmitted by the network device. The reference signal configured or transmitted by the network device can be matched with the reference signal information suggested by the terminal device, namely, how the terminal device suggests the network device to configure or transmit, and the network device configures or transmits according to the suggestion of the terminal device; or may be mismatched, i.e. the network device does not configure or transmit the reference signal as suggested by the terminal device.

And E1, the number of the transmission beams supported by the network equipment is matched with the number of the receiving beams of the terminal equipment. That is, the E1 information matches the E information.

And 304, the terminal equipment updates the second AI model according to the first information.

In one implementation, in the event that the first information includes acknowledgement information, the terminal device may update the second AI model to the AI model requested for use. In the case where the first information includes attribute information of the terminal device-suggested AI model, the terminal device may update the second AI model to the terminal device-suggested AI model. For example, the first information includes attribute information of the fourth AI model, and the terminal device may update the second AI model to the fourth AI model. In the case where the first information includes attribute information of an AI model desired by the network device, the terminal device may update the second AI model to the AI model desired by the network device. For example, the first information includes attribute information of the fourth AI model, and the terminal device may update the second AI model to the fourth AI model.

In another implementation manner, the first information includes first reference signal information, where the first reference signal information may be reference signal information of a reference signal configured or transmitted by the network device, may also be reference signal information matched with a purpose of an AI model proposed by the terminal device, or may also be reference signal information having an association relationship with attribute information of an AI model expected by the network device. Further, the terminal device may measure the received reference signal based on the first reference signal information to obtain channel quality information. Wherein the channel quality information may include, but is not limited to, at least one of: signal-to-interference-and-noise ratio (signal to interference plus noise ratio, SINR), reference signal received power (REFERENCE SIGNAL RECEIVING power, RSRP), strength indication of received signal (RECEIVED SIGNAL STRENGTH indicator, RSSI), reference signal received quality (REFERENCE SIGNAL RECEIVING quality, RSRQ), channel quality indication (channel quality indication, CQI).

Optionally, the terminal device further updates the second AI model to an AI model having an association relationship with the channel quality information according to the channel quality information. Or the terminal equipment updates the attribute information of the AI model with the association relation with the channel quality information according to the channel quality information. It can be understood that the terminal device adjusts or fine-tunes the attribute information of the AI model having an association relationship with the channel quality information according to the channel quality information. Therefore, the terminal equipment retrains the AI model to ensure that the updated AI model is matched with the channel quality information.

Optionally, when the first information further includes a reporting threshold of the reference signal, the terminal device further selects target channel quality information greater than or equal to the reporting threshold from the channel quality information, and determines reference signal information corresponding to the target channel quality information. The terminal device may update attribute information of the AI model having an association relationship with the channel quality information according to the target channel quality information. Or the terminal device may report the target channel quality information and the reference signal information corresponding to the target channel quality information to the network device, so that the network device can obtain the reference signal information used by the terminal device to retrain the AI model. Optionally, the network device may update attribute information of the AI model having an association relationship with the channel quality information according to the target channel quality information, so as to implement retraining of the AI model by the network device, so as to ensure that the updated AI model matches with the channel quality information.

In the embodiment shown in fig. 3, the terminal device may negotiate with the network device to update the AI model of the terminal device when receiving the first signaling indicating the first model association information, thereby contributing to an improvement in communication efficiency.

As an alternative embodiment, the terminal device may also perform step 302 and step 303 after updating to the first AI model, and update the first AI model according to the first information. The process of updating the first AI model by the terminal device according to the first information is similar to the process of updating the second AI model according to the first information, and will not be described herein.

Referring to fig. 4, a flowchart of another AI model updating method according to an embodiment of the application may specifically include the following steps:

The network device sends 401 a first signaling to the terminal device. Correspondingly, the terminal device receives the first signaling from the network device. Wherein the first signaling is used to indicate attribute information of the first AI model.

In one implementation, the attribute information of the first AI model indicated by the first signaling may be attribute information of an AI model desired by the network device, i.e., attribute information of an AI model desired by the network device for use by the terminal device. For example, the attribute information of the first AI model may include all model parameters of the AI model desired by the network device, or part of model parameters of the AI model desired by the network device.

Optionally, the model parameters of the first AI model include model parameter values arranged in a predetermined order, for example, neuron parameters arranged in the order of neurons of each layer.

In another implementation, the attribute information of the first AI model indicated by the first signaling is different from the attribute information of the second AI model currently in use. That is, the network device transmits the first signaling to the terminal device while knowing attribute information of the second AI model currently used by the terminal. For example, the model structure of the first AI model is different from the model structure of the second AI model.

In yet another implementation, attribute information of the first AI model indicated by the first signaling has an association with terminal device information of the terminal device. For example, the first signaling is a broadcast message including attribute information of an AI model having an association relationship with equipment manufacturer identification information of the terminal equipment.

Step 401 differs from step 301 in that the first signaling in step 401 indicates attribute information of the first AI model; and the first signaling in step 301 indicates first model association information.

402, The attribute information of the first AI model is different from the attribute information of the second AI model currently used, and the second AI model is updated to the first AI model.

When the terminal device receives the first signaling, it can determine whether the attribute information of the first AI model is the same as the attribute information of the second AI model currently used. The attribute information of the first AI model is different from at least one type of attribute information in the attribute information of the second AI model, and it can be determined that the attribute information of the first AI model is different from the attribute information of the second AI model.

The process of updating the second AI model to the first AI model in step 402 may refer to the detailed description of step 202, and will not be described in detail herein.

In the embodiment shown in fig. 4, when the terminal device receives the first signaling indicating the attribute information of the first AI model, the second AI model currently used may be updated to be the same as the first AI model, so that the terminal device may quickly update the AI model, thereby contributing to an improvement in communication efficiency.

As an alternative embodiment a, this embodiment a may include:

(1) The network device sends a first signaling to the terminal device. Correspondingly, the terminal device receives the first signaling from the network device. Wherein the first signaling is used to indicate attribute information of the first AI model.

(2) The terminal device sends a second request message to the network device. Accordingly, the network device receives the second request message from the terminal device.

(3) The network device sends a second response message to the terminal device. Correspondingly, the terminal device receives a second response message from the network device.

This embodiment a differs from the embodiment shown in fig. 3 in that the first signaling is used to indicate attribute information of the first AI model in this embodiment, and in that the first signaling is used to indicate first model association information in the embodiment shown in fig. 3.

In the embodiment shown in fig. 3 and embodiment a, the network device determines the first information according to the second request message. As an alternative embodiment, the network device determines the first information based on capability information of the terminal device. The capability information of the terminal device includes attribute information of each of a plurality of AI models supported by the terminal device, and the attribute information can be reported to the network device when the terminal device is accessed. In this embodiment, the first information may include at least one of the following: attribute information of an AI model supported by the terminal device (for example, attribute information of a certain AI model in a plurality of AI models), attribute information of an AI model expected by the network device, a specified AI model update time period, reference signal information of a reference signal configured or transmitted by the network device, and a number of transmission beams supported by the network device. Wherein the attribute information of the AI model supported by the terminal device is similar to the attribute information of the AI model suggested by the terminal device. The remaining information in the first information may refer to the specific description of these information in step 303, which is not described herein.

In the embodiment shown in fig. 3 and embodiment a, the terminal device sends a second request message to the network device after receiving the first signaling. As an alternative embodiment, the terminal device sends a third request message to the network device before receiving the first signaling, and receives a third response message from the network device. The third request message is similar to the second request message and the third response message is similar to the second response message. This embodiment differs from the embodiment shown in fig. 3 in that the second request message is sent after the first signaling is received and the third request message is sent before the first signaling is received. For example, the terminal device may predict in advance that it will move to the first cell, and then the terminal device may send a third request message to the network device before receiving the first signaling to switch to the first cell.

Referring to fig. 5, fig. 5 is a schematic structural diagram of a communication device according to an embodiment of the application. The communication device 50 may be a terminal device or a device matching with a terminal device. As shown in fig. 5, the communication device 50 includes a processing unit 501 and a communication unit 502.

In one implementation, the communication unit 502 is configured to receive a first signaling from a network device, where the first signaling is configured to indicate first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the processing unit 501 updates the second AI model currently used to the first AI model, where the first model association information is different from the second model association information currently used; wherein the second model association information has an association relationship with attribute information of the second AI model.

In another implementation manner, the communication unit 502 is configured to send a first signaling to the terminal device, where the first signaling is used to indicate first model association information, and the first model association information has an association relationship with attribute information of the first AI model, and the first model association information is different from second model association information that is currently used.

Based on the same inventive concept, the principle and beneficial effects of the communication device 50 according to the embodiment of the present application are similar to those of the embodiments of fig. 2 to 4, and may be referred to as the principle and beneficial effects of implementation of the method, which are not described herein for brevity.

Referring to fig. 6, fig. 6 is a schematic structural diagram of another communication device according to an embodiment of the application. The communication device 60 may be a terminal device or a device matching with a terminal device. Optionally, the communication device may further comprise a memory 603. Wherein the transceiver 601, processor 602, memory 603 may be connected by a bus 604 or otherwise. The bus is shown in bold lines in fig. 6, and the manner in which other components are connected is merely illustrative and not limiting. The buses may be divided into address buses, data buses, control buses, etc. For ease of illustration, only one thick line is shown in fig. 6, but not only one bus or one type of bus.

The coupling in the embodiments of the present application is an indirect coupling or communication connection between devices, units, or modules, which may be in electrical, mechanical, or other forms for information interaction between the devices, units, or modules. The specific connection medium between the transceiver 601, the processor 602, and the memory 603 is not limited in the embodiment of the present application.

Memory 603 may include read only memory and random access memory and provide instructions and data to processor 602. A portion of memory 603 may also include non-volatile random access memory.

The Processor 602 may be a central processing unit (Central Processing Unit, CPU), the Processor 602 may also be other general purpose processors, digital signal processors (DIGITAL SIGNAL processors, DSPs), application SPECIFIC INTEGRATED Circuits (ASICs), field-Programmable gate arrays (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor, but in the alternative, the processor 602 may be any conventional processor or the like.

In an alternative embodiment, memory 603 is used for storing program instructions; a processor 602 for invoking program instructions stored in the memory 603 for performing the steps performed by the terminal device in the corresponding embodiment of fig. 2-4.

In the embodiments of the present application, the methods provided by the embodiments of the present application may be implemented by running a computer program (including program code) capable of executing the steps involved in the above-described methods on a general-purpose computing device such as a computer, including a processing element such as a CPU, a random access storage medium (Random Access Memory, RAM), a Read-Only Memory (ROM), or the like, and a storage element. The computer program may be recorded on, for example, a computer-readable recording medium, and loaded into and run in the above-described computing device through the computer-readable recording medium.

Based on the same inventive concept, the principle and beneficial effects of the communication device 60 according to the embodiment of the present application are similar to those of the embodiments of fig. 2 to 4, and may be referred to as the principle and beneficial effects of implementation of the method, which are not described herein for brevity.

The communication device may be, for example: a chip, or a chip module.

The embodiment of the application also provides a chip, which comprises a processor, wherein the processor can execute the relevant steps of the terminal equipment in the embodiment of the method.

In one implementation, the chip is to: receiving a first signaling from a network device, wherein the first signaling is used for indicating first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the first model association information is different from the second model association information which is currently used, and the second AI model which is currently used is updated to be the first AI model; wherein the second model association information has an association relationship with attribute information of the second AI model.

Optionally, the chip is specifically configured to update the second AI model to the first AI model, where the first AI model is different from the second AI model and the terminal device supports the first AI model.

Optionally, the chip is further configured to select the target AI model from AI models supported by the terminal device based on the first model association information, wherein the first AI model is different from the second AI model, and the terminal device does not support the first AI model; the target AI model satisfies the similarity condition; updating the first AI model or the second AI model to the target AI model.

Optionally, the chip is further configured to send a first request message to the network device, where the first AI model is different from the second AI model, and the terminal device does not support the first AI model, and the first request message is used to indicate attribute information of the third AI model; receiving a first response message from the network device; in the case where the first response message indicates acknowledgement information, the first AI model or the second AI model is updated to the third AI model.

Optionally, the chip is further configured to update an association relationship between the second model association information and attribute information of the second AI model to an association relationship between the first model association information and attribute information of the second AI model, where the first AI model is the same as the second AI model.

Optionally, the chip is further configured to deactivate the second AI model.

Optionally, the chip is further configured to send a second request message to the network device, where the second request message is used to determine the first information; a second response message is received from the network device, the second response message including the first information.

Optionally, the chip is further configured to include attribute information of a fourth AI model; the first AI model or the second AI model is updated to the fourth AI model.

Optionally, the chip is further configured to include first reference signal information; obtaining channel quality information according to the first reference signal information; updating the first AI model or the second AI model into an AI model with an association relation with the channel quality information according to the channel quality information; and/or updating attribute information of the AI model with an association relationship with the channel quality information.

Optionally, the chip is further configured to report a threshold to the first information; and selecting target channel quality information which is larger than or equal to a reporting threshold value from the channel quality information, and determining reference signal information corresponding to the target channel quality information.

Optionally, the chip is further configured to send the target channel quality information and reference signal information corresponding to the target channel quality information to the network device.

Optionally, the chip is further configured to update attribute information of an AI model having an association relationship with the channel quality information according to the target channel quality information and reference signal information corresponding to the target channel quality information.

Optionally, the chip is further configured to select a target AI model from AI models supported by the terminal device based on the first model association information, and update the first AI model to the target AI model, where the output result of the first AI model does not meet a preset condition;

Or, the state of the first AI model is a deactivation state, a target AI model is selected from AI models supported by the terminal equipment based on the first model association information, and the first AI model or the second AI model is updated to the target AI model;

or, the first AI model is different from the AI model indicated by the AI model indication information, a target AI model is selected from AI models supported by the terminal equipment based on the first model association information, and the first AI model or the second AI model is updated to the target AI model;

Wherein the target AI model satisfies a similarity condition.

Optionally, the target AI model satisfies a similarity condition, including at least one of:

the target AI model is an AI model associated with preset cell information; the preset cell information is contained in the configured or activated cell information; or, presetting the cell information as the cell information associated with the target TCI state, wherein the target TCI state is indicated by a second signaling or is configured or activated;

Optionally, the chip is further configured to receive a third signaling from the network device, where the second signaling is configured to indicate an association relationship between the first model association information and the attribute information of the first AI model.

Optionally, the chip is further configured to determine an association relationship between the first model association information and attribute information of the first AI model.

Optionally, the association relationship between the first model association information and the attribute information of the first AI model is predefined by a protocol.

Optionally, the first model association information includes at least one of: first PLMN information, first operator information, first network equipment identity information, first TAI information, first SOR information, first URSP information, first operating frequency information, first cell information-associated reference signal information, a first neighbor list, first timestamp information, first terminal equipment information;

Wherein the first cell information comprises at least one of: first cell group information, first serving cell identification information, first physical cell identification, first TRP identification information, first BWP identification.

Optionally, the attribute information of the first AI model includes at least one of: the model identification of the first AI model, the model structure information of the first AI model, the model parameters of the first AI model.

In one implementation, the chip is to: and sending a first signaling to the terminal equipment, wherein the first signaling is used for indicating first model association information, the first model association information has association relation with attribute information of a first AI model, and the first model association information is different from second model association information which is currently used.

Optionally, the chip is further configured to receive a first request message from the terminal device, where the first request message is used to indicate attribute information of the third AI model; and the first response message is sent to the terminal equipment and is used for indicating the confirmation information.

Optionally, the chip is further configured to receive a second request message from the terminal device, and determine the first information according to the second request message; sending a second response message to the terminal equipment, wherein the second response message comprises the first information; or determining first information according to the capability information of the terminal equipment; and sending the first information to the terminal equipment.

Optionally, the first information includes attribute information of the fourth AI model.

Optionally, the first information includes first reference signal information, where the first reference signal information is reference signal information of a reference signal configured or transmitted by the network device, or reference signal information matched with a purpose of an AI model proposed by the terminal device, or reference signal information having an association relationship with attribute information of an AI model expected by the network device.

Optionally, the chip is further configured to report a threshold to the first information; receiving target channel quality information from terminal equipment and reference signal information corresponding to the target channel quality information; and updating the attribute information of the AI model with the association relation with the channel quality information according to the target channel quality information and the reference signal information corresponding to the target channel quality information.

Optionally, the chip is further configured to send a third signaling to the terminal device, where the third signaling is used to indicate an association relationship between the first model association information and the attribute information of the first AI model.

Optionally, the first model association information includes at least one of: the method comprises the steps of first Public Land Mobile Network (PLMN) information, first operator information, first network equipment identity information, first Tracking Area Identity (TAI) information, first roaming guide (SOR) information, first terminal routing policy URSP information, first operating frequency information, first cell information, reference signal information associated with the first cell information, a first neighbor list, first timestamp information and first terminal equipment information;

wherein the first cell information comprises at least one of: first cell group information, first serving cell identification information, first physical cell identification, first transmission and reception point TRP identification information, first bandwidth part BWP identification.

Referring to fig. 7, fig. 7 is a schematic structural diagram of a chip module according to an embodiment of the application. The chip module 70 may perform the steps related to the terminal device in the foregoing method embodiment, where the chip module 70 includes: a communication interface 701 and a chip 702.

The communication interface is used for carrying out internal communication of the chip module or carrying out communication between the chip module and external equipment. The communication interface may also be described as a communication module. Chip 702 is used to implement the functions of the terminal device in the embodiment of the present application.

For example, the chip 702 is configured to receive a first signaling from a network device, where the first signaling is configured to indicate first model association information, and the first model association information has an association relationship with attribute information of a first AI model; the first model association information is different from the second model association information which is currently used, and the second AI model which is currently used is updated to be the first AI model; wherein the second model association information has an association relationship with attribute information of the second AI model.

As another example, the chip 702 is configured to receive first signaling from a network device, where the first signaling is configured to indicate attribute information of a first AI model; the attribute information of the first AI model is different from the attribute information of the second AI model currently used, and the second AI model is updated to the first AI model.

Optionally, the chip module 70 may further include a memory module 703 and a power module 704. The memory module 703 is used for storing data and instructions. The power module 704 is used for providing power to the chip module.

For each device and product applied to or integrated in the chip module, each module included in the device and product may be implemented by hardware such as a circuit, and different modules may be located in the same component (e.g. a chip, a circuit module, etc.) of the chip module or different components, or at least some modules may be implemented by software programs running on a processor integrated inside the chip module, where the remaining (if any) modules may be implemented by hardware such as a circuit.

Embodiments of the present application also provide a computer readable storage medium having one or more instructions stored therein, the one or more instructions being adapted to be loaded by a processor and to perform the method provided by the above-described method embodiments.

The present application also provides a computer program product comprising a computer program or instructions which, when run on a computer, cause the computer to perform the method provided by the method embodiments described above.

For the above embodiments, for simplicity of description, the same is denoted as a series of combinations of actions. It will be appreciated by persons skilled in the art that the application is not limited by the order of acts described, as some steps in embodiments of the application may be performed in other orders or concurrently. In addition, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts, steps, modules, or units, etc. that are described are not necessarily required by the embodiments of the application.

In the foregoing embodiments, the descriptions of the embodiments of the present application are emphasized, and in part, not described in detail in one embodiment, reference may be made to related descriptions of other embodiments.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, or may be embodied in software instructions executed by a processor. The software instructions may be comprised of corresponding software modules that may be stored in RAM, flash memory, ROM, erasable programmable read-only memory (erasable programmable ROM, EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disk, a removable disk, a compact disk read-only memory (CD-ROM), or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. In addition, the ASIC may be located in a terminal device or a management device. The processor and the storage medium may reside as discrete components in a terminal device or management device.

Those skilled in the art will appreciate that in one or more of the examples described above, the functions described in the embodiments of the present application may be implemented, in whole or in part, in software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a Solid State Drive (SSD)), or the like.

With respect to each of the apparatuses and each of the modules/units included in the products described in the above embodiments, it may be a software module/unit, a hardware module/unit, or a software module/unit, and a hardware module/unit. For example, for each device or product applied to or integrated on a chip, each module/unit included in the device or product may be implemented in hardware such as a circuit, or at least some modules/units may be implemented in software program, where the software program runs on a processor integrated inside the chip, and the remaining (if any) part of modules/units may be implemented in hardware such as a circuit; for each device and product applied to or integrated in the chip module, each module/unit contained in the device and product can be realized in a hardware manner such as a circuit, different modules/units can be located in the same component (such as a chip, a circuit module and the like) or different components of the chip module, or at least part of the modules/units can be realized in a software program, the software program runs on a processor integrated in the chip module, and the rest (if any) of the modules/units can be realized in a hardware manner such as a circuit; for each device, product, or application to or integrated with the terminal, each module/unit included in the device, product, or application may be implemented in hardware such as a circuit, where different modules/units may be located in the same component (e.g., a chip, a circuit module, etc.) or different components in the terminal, or at least some modules/units may be implemented in a software program, where the software program runs on a processor integrated within the terminal, and the remaining (if any) some modules/units may be implemented in hardware such as a circuit.

The foregoing detailed description of the embodiments of the present application further illustrates the purposes, technical solutions and advantageous effects of the embodiments of the present application, and it should be understood that the foregoing description is only a specific implementation of the embodiments of the present application, and is not intended to limit the scope of the embodiments of the present application, and any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application should be included in the scope of the embodiments of the present application.

Claims

1. An artificial intelligence AI model updating method, comprising:

Receiving first signaling from network equipment, wherein the first signaling is used for indicating first model association information, and the first model association information has association relation with attribute information of a first AI model;

the first model association information is different from the second model association information which is currently used, and the second AI model which is currently used is updated to be the first AI model; the second model association information and the attribute information of the second AI model have an association relationship.

2. The method of claim 1, wherein the updating the currently used second AI model to the first AI model comprises:

The first AI model is different from the second AI model, and terminal equipment supports the first AI model, updating the second AI model to the first AI model.

3. The method according to claim 2, wherein the method further comprises:

The first AI model is different from the second AI model, and the terminal device does not support the first AI model, and a target AI model is selected from AI models supported by the terminal device based on the first model association information; the target AI model meets a similarity condition;

updating the first AI model or the second AI model to the target AI model.

4. The method according to claim 2, wherein the method further comprises:

The first AI model is different from the second AI model, and the terminal equipment does not support the first AI model, and a first request message is sent to the network equipment, wherein the first request message is used for indicating attribute information of a third AI model;

receiving a first response message from the network device;

In the event that the first response message indicates acknowledgement information, the first AI model or the second AI model is updated to the third AI model.

5. The method according to claim 2, wherein the method further comprises:

And the first AI model is the same as the second AI model, and the association relation between the second model association information and the attribute information of the second AI model is updated to be the association relation between the first model association information and the attribute information of the second AI model.

6. The method according to any one of claims 1 to 5, further comprising:

deactivating the second AI model.

7. The method according to claim 1, wherein the method further comprises:

sending a second request message to the network device, wherein the second request message is used for determining first information;

a second response message is received from the network device, the second response message including the first information.

8. The method of claim 7, wherein the first information comprises attribute information of a fourth AI model;

The method further comprises the steps of:

updating the first AI model or the second AI model to the fourth AI model.

9. The method of claim 7, wherein the first information comprises first reference signal information;

The method further comprises the steps of:

obtaining channel quality information according to the first reference signal information;

updating the first AI model or the second AI model to an AI model with an association relation with the channel quality information according to the channel quality information; and/or updating attribute information of an AI model with an association relation with the channel quality information.

10. The method of claim 9, wherein the first information further comprises a reporting threshold;

The method further comprises the steps of:

And selecting target channel quality information which is larger than or equal to the reporting threshold value from the channel quality information, and determining reference signal information corresponding to the target channel quality information.

11. The method according to claim 10, wherein the method further comprises:

And transmitting the target channel quality information and the reference signal information corresponding to the target channel quality information to the network equipment.

12. The method according to claim 10, wherein the method further comprises:

And updating attribute information of an AI model with an association relation with the channel quality information according to the target channel quality information and the reference signal information corresponding to the target channel quality information.

13. The method according to claim 1, wherein the method further comprises:

The output result of the first AI model does not meet a preset condition, a target AI model is selected from AI models supported by the terminal equipment based on the first model association information, and the first AI model is updated to be the target AI model;

Or, the state of the first AI model is a deactivation state, a target AI model is selected from AI models supported by the terminal equipment based on the first model association information, and the first AI model or the second AI model is updated to be the target AI model;

Or, the first AI model is different from the AI model indicated by the AI model indication information, a target AI model is selected from the AI models supported by the terminal equipment based on the first model association information, and the first AI model or the second AI model is updated to be the target AI model;

wherein the target AI model satisfies a similarity condition.

14. The method of claim 3 or 13, wherein the target AI model satisfies the similarity condition, including at least one of:

the first model association information is the same as or similar to the model association information associated with the target AI model, and is larger than a second threshold;

The target AI model is an AI model associated with preset cell information; the preset cell information is contained in the configured or activated cell information; or, the preset cell information is cell information associated with a target transmission configuration indication TCI state, and the target TCI state is indicated by a second signaling or configured or activated;

the target AI model is the AI model with the largest use times or the AI model used last time in a second preset time period before the second AI model is updated.

15. The method according to any one of claims 1 to 5, further comprising:

And receiving third signaling from the network equipment, wherein the third signaling is used for indicating the association relation between the first model association information and the attribute information of the first AI model.

16. The method according to any one of claims 1 to 5, further comprising:

And determining the association relation between the first model association information and the attribute information of the first AI model.

17. The method according to any one of claims 1 to 5, wherein the association relationship of the first model association information with the attribute information of the first AI model is protocol predefined.

18. The method according to any one of claims 1 to 5, wherein the first model association information comprises at least one of: the method comprises the steps of first Public Land Mobile Network (PLMN) information, first operator information, first network equipment identity information, first Tracking Area Identity (TAI) information, first roaming guide (SOR) information, first terminal routing policy URSP information, first operating frequency information, first cell information, reference signal information associated with the first cell information, a first neighbor list, first timestamp information and first terminal equipment information;

wherein the first cell information includes at least one of: first cell group information, first serving cell identification information, first physical cell identification, first transmission and reception point TRP identification information, first bandwidth part BWP identification.

19. The method of any one of claims 1 to 5, wherein the attribute information of the first AI model includes at least one of: the model identification of the first AI model, the model structure information of the first AI model, the model parameters of the first AI model.

20. An artificial intelligence AI model updating method, comprising:

Receiving first signaling from a network device, the first signaling being used to indicate attribute information of a first AI model;

The attribute information of the first AI model is different from the attribute information of a second AI model currently used, and the second AI model is updated to the first AI model.

21. The method of claim 20, wherein the updating the second AI model to the first AI model comprises:

And the terminal equipment supports the first AI model and updates the second AI model to the first AI model.

22. The method of claim 21, wherein the method further comprises:

the terminal equipment does not support the first AI model, and a target AI model is selected from AI models supported by the terminal equipment; the target AI model meets a similarity condition;

updating the first AI model or the second AI model to the target AI model.

23. The method of claim 21, wherein the method further comprises:

the terminal equipment does not support the first AI model, and sends a first request message to the network equipment, wherein the first request message is used for indicating attribute information of a third AI model;

receiving a first response message from the network device;

24. The method of claim 20, wherein the method further comprises:

25. The method of claim 24, wherein the first information comprises attribute information of a fourth AI model;

The method further comprises the steps of:

updating the first AI model or the second AI model to the fourth AI model.

26. The method of claim 24, wherein the first information comprises first reference signal information;

The method further comprises the steps of:

27. An artificial intelligence AI model updating method, the method comprising:

And sending a first signaling to the terminal equipment, wherein the first signaling is used for indicating first model association information, the first model association information has an association relationship with attribute information of a first AI model, and the first model association information is different from second model association information which is currently used.

28. The method of claim 27, wherein the method further comprises:

Receiving a first request message from the terminal equipment, wherein the first request message is used for indicating attribute information of a third AI model;

and the first response message is sent to the terminal equipment and is used for indicating the confirmation information.

29. The method of claim 27, wherein the method further comprises:

receiving a second request message from the terminal equipment, and determining first information according to the second request message; sending a second response message to the terminal equipment, wherein the second response message comprises first information;

30. The method of claim 29, wherein the first information comprises attribute information of a fourth AI model.

31. The method of claim 30, wherein the first information comprises first reference signal information, the first reference signal information being reference signal information of a reference signal configured or transmitted by the network device, or reference signal information matching a purpose of an AI model suggested by the terminal device, or reference signal information having an association with attribute information of an AI model desired by the network device.

32. The method of claim 31, wherein the first information further comprises a reporting threshold;

The method further comprises the steps of:

And receiving target channel quality information from the terminal equipment and reference signal information corresponding to the target channel quality information.

33. The method according to any one of claims 27 to 32, further comprising:

And sending a third signaling to the terminal equipment, wherein the third signaling is used for indicating the association relation between the first model association information and the attribute information of the first AI model.

34. The method of any one of claims 27 to 32, wherein the association of the first model association information with the attribute information of the first AI model is protocol predefined.

35. The method according to any one of claims 27 to 32, wherein the first model association information comprises at least one of: the method comprises the steps of first Public Land Mobile Network (PLMN) information, first operator information, first network equipment identity information, first Tracking Area Identity (TAI) information, first roaming guide (SOR) information, first terminal routing policy URSP information, first operating frequency information, first cell information, reference signal information associated with the first cell information, a first neighbor list, first timestamp information and first terminal equipment information;

36. The method of any of claims 27-32, wherein the attribute information of the first AI model includes at least one of: the model identification of the first AI model, the model structure information of the first AI model, the model parameters of the first AI model.

37. A communication device, the device comprising:

A communication unit, configured to receive a first signaling from a network device, where the first signaling is used to indicate first model association information, and the first model association information has an association relationship with attribute information of a first AI model;

and the processing unit is used for updating the second AI model which is currently used as the first AI model, wherein the first model association information is different from the second model association information which is currently used.

38. A communication device, the device comprising:

The communication unit is used for sending a first signaling to the terminal equipment, wherein the first signaling is used for indicating first model association information, the first model association information has an association relationship with attribute information of a first AI model, the first model association information is different from second model association information which is currently used, and the first model association information is used for triggering updating of the second AI model which is currently used into the first AI model.

39. A communication device comprising a processor, a memory and a computer program or instructions stored on the memory, wherein the processor executes the computer program or instructions to implement the steps of the method of any one of claims 1-26 or to implement the steps of the method of any one of claims 27-36.

40. A chip comprising a processor, wherein the processor performs the steps of the method of any one of claims 1-26, or performs the steps of the method of any one of claims 27-36.

41. The utility model provides a chip module, its characterized in that, chip module includes communication module, power module, storage module and chip, wherein: the power supply module is used for providing electric energy for the chip module; the storage module is used for storing data and instructions; the communication module is used for carrying out internal communication of the chip module or carrying out communication between the chip module and external equipment; the chip being adapted to perform the steps of the method of any one of claims 1-26 or the steps of the method of any one of claims 27-36.

42. A computer readable storage medium, characterized in that it stores a computer program or instructions which, when executed, implement the steps of the method of any one of claims 1-26, or the steps of the method of any one of claims 27-36.