Method for transmitting artificial intelligence model and communication device

Abstract

The invention provides a method for transmitting an AI model and a communication device, in the method, a first node can request a second node for pre-configuration information of the AI model which can be provided by the second node and a manufacturer identifier corresponding to the second node, and the first node carries the manufacturer identifier corresponding to the first node in a request message; the first node can determine the AI model needing to be requested from the second node based on the pre-configuration information of the AI model, and the second node can determine whether the second node and the first node belong to the same manufacturer or not based on the manufacturer identifier corresponding to the first node, so as to determine how to send the AI model required by the first node subsequently. According to the method provided by the invention, cross-network entity transmission of the AI model can be realized, and the privacy problem of the AI model is avoided.

Description

Technical Field

[0001] The present application relates to the field of communications, and more specifically, to a method and a communication device for transmitting an artificial intelligence model. Background Art

[0002] At present, the RAN3 standard of release 18 (R18) specifies that artificial intelligence (AI) or machine learning (ML) models are trained and inferred separately in different network entities (e.g., gNB, operation, maintenance and management (OAM) network element). For example, the specific scenarios include the following two: AI training and AI inference in gNB, or AI training in OAM and AI inference in gNB.

[0003] Currently, R18 does not involve a mechanism for transmitting AI models across network entities. However, transmitting models across nodes has the following benefits: (1) Improving the accuracy of AI prediction information or decisions. For example, if the accuracy of AI models on some nodes is too low, high-precision reasoning information can be obtained by transmitting high-precision AI models on other nodes across nodes. (2) Nodes without AI model training resources can obtain AI models. For example, if the computing power and / or storage resources of some nodes are insufficient and AI models cannot be trained, trained AI models can be obtained by transmitting AI models on other nodes across nodes. (3) Reducing node resource consumption. For example, in a scenario where the same manufacturer (or manufacturer that has signed an agreement) can share models, it is not necessary for all nodes to train AI models. Some nodes with low computing power can obtain AI models of a specific accuracy by transmitting AI models on other nodes across nodes.

[0004] Furthermore, in the application scenario of transmitting AI models across network entities, when different network entities belong to different manufacturers, transmitting AI models across nodes may involve privacy issues related to AI models. For example, AI models trained by each manufacturer cannot be exposed to other manufacturers; for another example, the training data involved in each AI model may also contain the user's private data and cannot be exposed to other manufacturers.

[0005] Therefore, how to achieve cross-network entity transmission of AI models has become a technical problem that needs to be solved. Summary of the invention

[0006] The present application provides a method and communication device for transmitting an artificial intelligence model, which can realize the cross-network entity transmission of the AI ​​model and avoid the privacy issues of the AI ​​model.

[0007] In a first aspect, a method for transmitting an artificial intelligence model is provided, which can be executed by a device corresponding to a first node, or can also be executed by a component (such as a chip or circuit) of the device corresponding to the first node, without limitation. For example, the first node can be any one of the following nodes: gNB, gNB-CU, gNB-DU, UE, 5GC, OAM, etc.

[0008] The method includes: a first node sends a first request message to a second node, the first request message is used to request pre-configuration information of an artificial intelligence (AI) model that can be provided by the second node and a manufacturer identifier corresponding to the second node, and the first request message carries the manufacturer identifier corresponding to the first node; the first node receives a first response message from the second node, the first response message carries the pre-configuration information of the AI ​​model that can be provided by the second node and the manufacturer identifier corresponding to the second node, the pre-configuration information of the AI ​​model includes a first association relationship, the first association relationship is an association relationship between at least two of the following: a model identifier corresponding to the AI ​​model, a model algorithm corresponding to the AI ​​model, a model algorithm identifier corresponding to the AI ​​model algorithm, a business corresponding to the AI ​​model, a business identifier corresponding to the AI ​​model business, a purpose corresponding to the AI ​​model, a purpose identifier corresponding to the purpose, reasoning information corresponding to the AI ​​model, and model information corresponding to the AI ​​model, the first association relationship is used by the first node to determine a first AI model to be requested, wherein the first AI model is one or more of the AI ​​models that can be provided by the second node.

[0009] In this application, the second node can be any one of the following nodes: gNB, gNB-CU, gNB-DU, UE, 5GC, OAM, etc.

[0010] In one possible implementation, the first node and the second node may be different nodes of the same manufacturer; in another possible implementation, the first node and the second node are nodes of different manufacturers. It can also be understood that in this application, the first node and the second node are not limited.

[0011] In this application, for example, when the first node is operating manufacturer #1, the manufacturer ID corresponding to the first node can be understood as the vendor ID of operating manufacturer #1, and the manufacturer ID can be used by the second node to determine whether it is the same manufacturer or a different manufacturer as the first node. In addition, the first node also requests the manufacturer ID corresponding to the second node from the second node, and the pre-configuration information of the AI ​​model provided by the second node can be associated with the manufacturer ID of the second node later. Alternatively, it can also be understood that the first node will determine that the pre-configuration information of the AI ​​model received is sent by the manufacturer corresponding to the second node.

[0012] In this application, model information is used to characterize the characteristics of the AI ​​model, and may include one or more of the following: the input of the AI ​​model, the output of the AI ​​model, the number of layers of the AI ​​model, and the dimension of the AI ​​model.

[0013] Based on the above technical solution, in this application, the first node can request the second node for the pre-configuration information of the AI ​​model that the second node can provide and the manufacturer identifier corresponding to the second node, and the first node carries its own corresponding manufacturer identifier in the request message. The first node can determine the AI ​​model that needs to be requested from the second node based on the pre-configuration information of the AI ​​model, and the second node can determine whether it belongs to the same manufacturer as the first node based on the manufacturer identifier corresponding to the first node, thereby determining how to send the AI ​​model required by the first node in the future. The method provided in this application can realize the transmission of AI models across network entities and avoid the privacy issues of AI models.

[0014] In one possible implementation, the pre-configuration information of the AI ​​model includes one or more of the following: information on the AI ​​model scrambling method supported by the second node, wherein the information on the AI ​​model scrambling method includes: the association between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; information on the AI ​​model updating method supported by the second node, wherein the information on the model updating method includes: the association between the AI ​​model update method identifier and the corresponding update method; information on the AI ​​model transmission method supported by the second node, wherein the information on the AI ​​model transmission method includes: the association between the AI ​​model transmission method identifier and the corresponding transmission method.

[0015] In a possible implementation, the first request message also carries pre-configuration information of the AI ​​model provided by the first node to the second node, and the pre-configuration information of the AI ​​model provided by the first node to the second node includes one or more of the following: information on the AI ​​model scrambling method supported by the first node, wherein the information on the AI ​​model scrambling method includes: the association between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; information on the AI ​​model update method supported by the first node, wherein the information on the AI ​​model update method includes: the association between the AI ​​model update method identifier and the corresponding update method; information on the AI ​​model transmission method supported by the first node, wherein the information on the AI ​​model transmission method includes: the association between the AI ​​model transmission method identifier and the corresponding transmission method.

[0016] Based on the above implementation method, in the present application, the second node and the first node can exchange AI model encryption method information, AI model update method information, AI model update method information, and AI model transmission method information, thereby ensuring that the AI ​​model is transmitted in an appropriate manner in the subsequent transmission of the AI ​​model.

[0017] In a possible implementation, the method also includes: the first node sends a second request message to the second node, the second request message is used to request the first AI model, the second request message carries first indication information, and the first indication information is used to indicate pre-configuration information of the first AI model; the first node receives a second response message from the second node, the second response message carries second indication information, and the second indication information is used to indicate pre-configuration information of the second AI model, wherein the pre-configuration information of the second AI model is information in the pre-configuration information of the first AI model that the second node supports sending to the first node, and the second AI model is one or more AI models in the first AI model.

[0018] In a possible implementation, the pre-configuration information of the first AI model includes one or more of the following: a model identifier corresponding to the first AI model, model information corresponding to the first AI model, a model algorithm identifier corresponding to the first AI model, a service identifier corresponding to the first AI model, a purpose identifier corresponding to the first AI model, an AI model scrambling method identifier supported by the first node, an AI model update method identifier supported by the first node, an AI model transmission method identifier supported by the first node, a reason for requesting the first AI model, and a model accuracy corresponding to the first AI model.

[0019] Based on the above implementation, in the present application, the first node can request the required AI model from the second node based on the received first association relationship. Specifically, the pre-configuration information of the AI ​​model transmitted by the first node and the second node includes the first association relationship, and the first node can subsequently transmit only the corresponding identifier to the second node through the first association relationship, thereby indicating to the second node the AI ​​model that needs to be requested, which can reduce signaling overhead.

[0020] Optionally, the second request message also carries one or more of the following: a period for the second node to report the AI ​​model; a message identifier corresponding to the second request message, wherein the message identifier is used to associate the second response message; and a message type corresponding to the second request message.

[0021] In the present application, the "message type corresponding to the second request message" may, for example, include the request message type for starting to send the requested AI model (start type), the request message type for stopping sending the requested AI model (stop type), the request message type for adding the requested AI model (add type), the request message type for deleting the requested AI model (del type), and the request message type for updating the requested AI model (revise type).

[0022] In one possible implementation, the pre-configuration information of the second AI model includes one or more of the following: a model identifier corresponding to the second AI model, model information corresponding to the second AI model, a model algorithm identifier corresponding to the second AI model, a service identifier corresponding to the second AI model, and model accuracy feedback information corresponding to the second AI model, wherein the model accuracy feedback information is used to indicate that the second node supports sending the model accuracy corresponding to the second AI model to the first node.

[0023] In one possible implementation, the second response message also carries one or more of the following: reporting cycle feedback information, the reporting cycle feedback information is used to indicate the AI ​​model reporting cycle supported by the second node; a message identifier; failure cause indication information, the failure cause indication information is used to indicate that the second node does not support the identifier and reason of the first AI model sent to the first node.

[0024] In the present application, based on the pre-configuration information of the first AI model required as indicated by the first node, the second node can determine the pre-configuration information of the AI ​​model that can be provided to the first node, thereby achieving negotiation between the two nodes.

[0025] In a possible implementation, the method further includes: the first node receiving an AI model update message from the second node, the AI ​​model update message carrying the encrypted second AI model, or the AI ​​model update message carrying an address corresponding to the encrypted second AI model, or the AI ​​model update message carrying the second AI model, or the AI ​​model update message carrying the address corresponding to the second AI model.

[0026] In the present application, if the first node and the second node belong to the same manufacturer, the second node can send the real AI model or the address of the AI ​​model to the first node. If the first node and the second node belong to different manufacturers, the second node can send the scrambled AI model or the address of the scrambled AI model to the first node, thereby avoiding privacy issues of the AI ​​model.

[0027] Optionally, the method also includes: the first node determines fourth indication information based on the second indication information, the fourth indication information is used to indicate pre-configuration information of the third AI model, wherein the pre-configuration information of the third AI model is information determined by the first node to request from the second node in the pre-configuration information of the second AI model, and the third AI model is one or more AI models in the second AI model; the first node sends an AI model confirmation message to the second node, and the AI ​​model confirmation message carries the fourth indication information.

[0028] Optionally, the method also includes: the first node receives an AI model update message from the second node, the AI ​​model update message carries the encrypted third AI model, or the AI ​​model update message carries the address corresponding to the encrypted third AI model, or the AI ​​model update message carries the third AI model, or the AI ​​model update message carries the address corresponding to the third AI model.

[0029] In a second aspect, a method for transmitting an artificial intelligence model is provided, which can be executed by a device corresponding to a second node, or can also be executed by a component (such as a chip or circuit) of the device corresponding to the second node, without limitation. For example, the second node can be any one of the following nodes: gNB, gNB-CU, gNB-DU, UE, 5GC, OAM, etc.

[0030] It should be noted that the beneficial effects achieved by the various methods in the second aspect that are the same as those achieved in the first aspect will not be described again, and can be understood by referring to the beneficial effects in the first aspect.

[0031] The method includes: a second node receives a first request message from a first node, the first request message is used to request pre-configuration information of an artificial intelligence (AI) model that can be provided by the second node and a manufacturer identifier corresponding to the second node, and the first request message carries the manufacturer identifier corresponding to the first node; the second node sends a first response message to the first node, the first response message carries the pre-configuration information of the AI ​​model that can be provided by the second node and the manufacturer identifier corresponding to the second node, the pre-configuration information of the AI ​​model includes a first association relationship, the first association relationship is an association relationship between at least two of the following: a model identifier corresponding to the AI ​​model, a model algorithm corresponding to the AI ​​model, a model algorithm identifier corresponding to the AI ​​model algorithm, a business corresponding to the AI ​​model, a business identifier corresponding to the AI ​​model business, a purpose corresponding to the AI ​​model, a purpose identifier corresponding to the purpose, reasoning information corresponding to the AI ​​model, and model information corresponding to the AI ​​model, the first association relationship is used by the first node to determine the first AI model to be requested, wherein the first AI model is one or more of the AI ​​models that can be provided by the second node.

[0032] In one possible implementation, the pre-configuration information of the AI ​​model includes one or more of the following: information on the AI ​​model scrambling method supported by the second node, wherein the information on the AI ​​model scrambling method includes: the association between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; information on the AI ​​model updating method supported by the second node, wherein the information on the model updating method includes: the association between the AI ​​model update method identifier and the corresponding update method; information on the AI ​​model transmission method supported by the second node, wherein the information on the AI ​​model transmission method includes: the association between the AI ​​model transmission method identifier and the corresponding transmission method.

[0033] In a possible implementation, the method also includes: the second node receives a second request message from the first node, the second request message is used to request the first AI model, the second request message carries first indication information, and the first indication information is used to indicate pre-configuration information of the first AI model; the second node sends a second response message to the first node, the second response message carries second indication information, and the second indication information is used to indicate pre-configuration information of the second AI model, wherein the pre-configuration information of the second AI model is information in the pre-configuration information of the first AI model that the second node supports sending to the first node, and the second AI model is one or more AI models in the first AI model.

[0034] In a possible implementation, the first request message also carries pre-configuration information of the AI ​​model provided by the first node to the second node, and the pre-configuration information of the AI ​​model provided by the first node to the second node includes one or more of the following: information on the AI ​​model scrambling method supported by the first node, wherein the information on the AI ​​model scrambling method includes: the association between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; information on the AI ​​model update method supported by the first node, wherein the information on the AI ​​model update method includes: the association between the AI ​​model update method identifier and the corresponding update method; information on the AI ​​model transmission method supported by the first node, wherein the information on the AI ​​model transmission method includes: the association between the AI ​​model transmission method identifier and the corresponding transmission method.

[0035] In a possible implementation, the pre-configuration information of the first AI model includes one or more of the following: a model identifier corresponding to the first AI model, model information corresponding to the first AI model, a model algorithm identifier corresponding to the first AI model, a service identifier corresponding to the first AI model, a purpose identifier corresponding to the first AI model, an AI model scrambling method identifier supported by the first node, an AI model update method identifier supported by the first node, an AI model transmission method identifier supported by the first node, a reason for requesting the first AI model, and a model accuracy corresponding to the first AI model.

[0036] In one possible implementation, the second request message also carries one or more of the following: a period for the second node to report the AI ​​model; a message identifier corresponding to the second request message, wherein the message identifier is used to associate the second response message; and a message type corresponding to the second request message.

[0037] In one possible implementation, the pre-configuration information of the second AI model includes one or more of the following: a model identifier corresponding to the second AI model, model information corresponding to the second AI model, a model algorithm identifier corresponding to the second AI model, a service identifier corresponding to the second AI model, and model accuracy feedback information corresponding to the second AI model, wherein the model accuracy feedback information is used to indicate that the second node supports the model accuracy corresponding to the second AI model sent to the first node.

[0038] In one possible implementation, the second response message also carries one or more of the following: reporting cycle feedback information, the reporting cycle feedback information is used to indicate the AI ​​model reporting cycle supported by the second node; a message identifier; failure cause indication information, the failure cause indication information is used to indicate that the second node does not support the identifier and reason of the first AI model sent to the first node.

[0039] In a possible implementation manner, the method further includes: the second node sending an AI model update message to the first node, where the AI ​​model update message carries the encrypted second AI model, or the AI ​​model update message carries an address corresponding to the encrypted second AI model, or the AI ​​model update message carries the second AI model, or the AI ​​model update message carries the address corresponding to the second AI model.

[0040] In a possible implementation, the method also includes: the second node receives an AI model confirmation message from the first node, the AI ​​model confirmation message carries fourth indication information, and the fourth indication information is used to indicate pre-configuration information of the third AI model, wherein the pre-configuration information of the third AI model is information determined by the first node to request from the second node in the pre-configuration information of the second AI model, and the third AI model is one or more AI models in the second AI model.

[0041] In a possible implementation manner, the method further includes: the AI ​​of the second node receives a model update message from the first node, the AI ​​model update message carries the encrypted third AI model, or the AI ​​model update message carries the address corresponding to the encrypted third AI model, or the AI ​​model update message carries the third AI model, or the AI ​​model update message carries the address corresponding to the third AI model.

[0042] In a third aspect, a method for transmitting training data is provided, which can be executed by a device corresponding to a first node, or can also be executed by a component (such as a chip or circuit) of the device corresponding to the first node, without limitation. For example, the first node can be any one of the following nodes: gNB, gNB-CU, gNB-DU, UE, 5GC, OAM, etc.

[0043] The method includes: a first node sends a first request message to a second node, the first request message is used to request pre-configuration information of training data that can be provided by the second node and a manufacturer identifier corresponding to the second node, wherein the first request message carries the manufacturer identifier corresponding to the first node; the first node receives a first response message from the second node, the first response message carries the pre-configuration information of training data that can be provided by the second node and the manufacturer identifier corresponding to the second node, the pre-configuration information of the training data includes a second association relationship, the second association relationship is an association relationship between at least two of the following: a data identifier corresponding to the training data, a model algorithm corresponding to the training data, a model algorithm identifier corresponding to the model algorithm, a business corresponding to the training data, a business identifier corresponding to the business, a purpose corresponding to the training data, and a purpose identifier corresponding to the purpose, the pre-configuration information of the training data is used by the first node to determine the first training data to be requested, and the first training data is part or all of the training data that can be provided by the second node.

[0044] Based on the above technical solution, in this application, the first node can request the second node for the pre-configuration information of the training data that the second node can provide and the manufacturer identifier corresponding to the second node, and the first node carries its own corresponding manufacturer identifier in the request message. The first node can determine the training data that needs to be requested from the second node based on the pre-configuration information of the training data, and the second node can determine whether it belongs to the same manufacturer as the first node based on the manufacturer identifier corresponding to the first node, thereby determining how to send the training data required by the first node in the future. The method provided by this application can realize the transmission of training data across network entities and avoid the privacy issues of training data.

[0045] The beneficial effects achieved by the following implementations can refer to the beneficial effects achieved by the implementations in the first aspect above. The only difference is that the "AI model" in the beneficial effects of the first aspect can be replaced with "training data" for understanding.

[0046] In a possible implementation, the pre-configuration information of the training data that can be provided by the second node includes one or more of the following: information on the training data scrambling method supported by the second node, wherein the information on the training data scrambling method includes: an association between a training data scrambling method identifier and a corresponding scrambling method and / or scrambling level; information on the training data updating method supported by the second node, wherein the information on the training data updating method includes: an association between an updating method identifier of the training data and a corresponding updating method; information on the training data transmission method supported by the second node, wherein the information on the training data transmission method includes: an association between a transmission method identifier of the training data and a corresponding transmission method.

[0047] In a possible implementation, the method also includes: the first node sends a second request message to the second node, the second request message is used to request the first training data, the second request message carries first indication information, and the first indication information is used to indicate the pre-configuration information of the first training data; the first node receives a second response message from the second node, the second response message carries second indication information, and the second indication information is used to indicate the pre-configuration information of the second training data, wherein the pre-configuration information of the second training data is information in the pre-configuration information of the first training data that the second node supports sending to the first node, and the second training data is part or all of the first training data.

[0048] In a possible implementation, the first request carries pre-configuration information of training data provided by the first node to the second node, and the pre-configuration information of training data provided by the first node to the second node includes one or more of the following: information on training data scrambling methods supported by the first node, wherein the information on the training data scrambling methods includes: an association relationship between a training data scrambling method identifier and a corresponding scrambling method and / or scrambling level; information on training data update methods supported by the first node, wherein the information on the training data update methods includes: an association relationship between a training data update method identifier and a corresponding update method; information on training data transmission methods supported by the first node, wherein the information on training data transmission methods includes: an association relationship between a training data transmission method identifier and a corresponding transmission method.

[0049] In a possible implementation, the pre-configuration information of the first training data includes one or more of the following: a data identifier corresponding to the first training data, a model algorithm identifier corresponding to the first training data, a service identifier corresponding to the first training data, a purpose identifier corresponding to the first training data, a training data scrambling method identifier supported by the first node, a training data update method identifier supported by the first node, a training data transmission method identifier supported by the first node, a reason for requesting the first training data, and a model accuracy corresponding to the first training data.

[0050] In a possible implementation, the second request message also carries one or more of the following: a period for the second node to report training data; a message identifier corresponding to the second request message, wherein the message identifier is used to associate the second response message; and a message type corresponding to the second request message.

[0051] In one possible implementation, the pre-configuration information of the second training data includes one or more of the following: a model identifier corresponding to the second training data, a model algorithm identifier corresponding to the second training data, a service identifier corresponding to the second training data, and model accuracy feedback information corresponding to the second training data, wherein the model accuracy feedback information is used to indicate that the second node supports the model accuracy corresponding to the second training data sent to the first node.

[0052] In one possible implementation, the second response message also carries one or more of the following: reporting cycle feedback information, the reporting cycle feedback information is used to indicate the reporting cycle of the second response message supported by the second node; a message identifier; failure cause indication information, the failure cause indication information is used to indicate the identifier and reason why the second node does not support the first training data sent to the first node.

[0053] In a possible implementation, the method further includes: the first node receiving a training data update message from the second node, the training data update message carrying the scrambled second training data, or the training data update message carrying an address corresponding to the scrambled second training data.

[0054] In a possible implementation, the method further includes: the first node determines fourth indication information based on the second indication information, the fourth indication information is used to indicate pre-configuration information of third training data, wherein the pre-configuration information of the third training data is information determined by the first node to request from the second node in the pre-configuration information of the second training data, and the third training data is part or all of the second training data; the first node sends a training data confirmation message to the second node, and the training data confirmation message carries the fourth indication information.

[0055] In a possible implementation, the method also includes: the first node receives a training data update message from the second node, the training data update message carries the encrypted third training data AI model, or the training data update message carries the address corresponding to the encrypted third training data.

[0056] In a fourth aspect, a method for transmitting training data is provided, which can be executed by a device corresponding to a second node, or can also be executed by a component (such as a chip or circuit) of the device corresponding to the second node, without limitation. For example, the second node can be any one of the following nodes: gNB, gNB-CU, gNB-DU, UE, 5GC, OAM, etc.

[0057] The method includes: a second node receives a first request message from a first node, the first request message is used to request pre-configuration information of training data that can be provided by the second node and a manufacturer identifier corresponding to the second node, wherein the first request message carries the manufacturer identifier corresponding to the first node; the second node sends a first response message to the first node, the first response message carries the pre-configuration information of training data that can be provided by the second node and the manufacturer identifier corresponding to the second node, the pre-configuration information of the training data includes a second association relationship, the second association relationship is an association relationship between at least two of the following: a data identifier corresponding to the training data, a model algorithm corresponding to the training data, a model algorithm identifier corresponding to the model algorithm, a business corresponding to the training data, a business identifier corresponding to the business, a purpose corresponding to the training data, and a purpose identifier corresponding to the purpose, the pre-configuration information of the training data is used by the first node to determine the first training data to be requested, and the first training data is part or all of the training data that can be provided by the second node.

[0058] In a possible implementation, the pre-configuration information of the training data that can be provided by the second node includes one or more of the following: information on the training data scrambling method supported by the second node, wherein the information on the training data scrambling method includes: an association between a training data scrambling method identifier and a corresponding scrambling method and / or scrambling level; information on the training data updating method supported by the second node, wherein the information on the training data updating method includes: an association between an updating method identifier of the training data and a corresponding updating method; information on the training data transmission method supported by the second node, wherein the information on the training data transmission method includes: an association between a transmission method identifier of the training data and a corresponding transmission method.

[0059] In a possible implementation, the method also includes: the second node receives a second request message from the first node, the second request message is used to request the first training data, the second request message carries first indication information, and the first indication information is used to indicate pre-configuration information of the first training data; the second node sends a second response message to the first node, the second response message carries second indication information, and the second indication information is used to indicate pre-configuration information of the second training data, wherein the pre-configuration information of the second training data is information in the pre-configuration information of the first training data that the second node supports sending to the first node, and the second training data is part or all of the first training data.

[0060] In a possible implementation, the first request carries pre-configuration information of training data provided by the first node to the second node, and the pre-configuration information of training data provided by the first node to the second node includes one or more of the following: information on training data scrambling methods supported by the first node, wherein the information on the training data scrambling methods includes: an association relationship between a training data scrambling method identifier and a corresponding scrambling method and / or scrambling level; information on training data update methods supported by the first node, wherein the information on the training data update methods includes: an association relationship between a training data update method identifier and a corresponding update method; information on training data transmission methods supported by the first node, wherein the information on training data transmission methods includes: an association relationship between a training data transmission method identifier and a corresponding transmission method.

[0061] In a possible implementation, the pre-configuration information of the first training data includes one or more of the following: a data identifier corresponding to the first training data, a model algorithm identifier corresponding to the first training data, a service identifier corresponding to the first training data, a purpose identifier corresponding to the first training data, a training data scrambling method identifier supported by the first node, a training data update method identifier supported by the first node, a training data transmission method identifier supported by the first node, a reason for requesting the first training data, and a model accuracy corresponding to the first training data.

[0062] In a possible implementation, the second request message also carries one or more of the following: a period for the second node to report training data; a message identifier corresponding to the second request message, wherein the message identifier is used to associate the second response message; and a message type corresponding to the second request message.

[0063] In one possible implementation, the pre-configuration information of the second training data includes one or more of the following: a model identifier corresponding to the second training data, a model algorithm identifier corresponding to the second training data, a service identifier corresponding to the second training data, and model accuracy feedback information corresponding to the second training data, wherein the model accuracy feedback information is used to indicate that the second node supports the model accuracy corresponding to the second training data sent to the first node.

[0064] In one possible implementation, the second response message also carries one or more of the following: reporting cycle feedback information, the reporting cycle feedback information is used to indicate the reporting cycle of the training data supported by the second node; a message identifier; failure cause indication information, the failure cause indication information is used to indicate the identifier and reason why the second node does not support the first training data sent to the first node.

[0065] In a possible implementation, the method further includes: the second node sending a training data update message to the first node, the training data update message carrying the scrambled second training data, or the training data update message carrying an address corresponding to the scrambled second training data.

[0066] In a possible implementation, the method further includes: the second node sends a training data confirmation message to the first node, the training data confirmation message carries the fourth indication information, the fourth indication information is used to indicate pre-configuration information of the third training data, wherein the pre-configuration information of the third training data is information determined by the first node to request from the second node in the pre-configuration information of the second training data, and the third training data is part or all of the second training data.

[0067] In a possible implementation, the method also includes: the second node sends a training data update message to the first node, the training data update message carries the encrypted third training data AI model, or the training data update message carries the address corresponding to the encrypted third training data.

[0068] In a fifth aspect, a communication device is provided, which is used to execute the method in any possible implementation of the first aspect or the third aspect. Specifically, the device may include a unit and / or module, such as a transceiver unit and / or a processing unit, for executing the method in any possible implementation of the first aspect or the third aspect.

[0069] In one implementation, the device is a first node. When the device is a communication device, the communication unit may be a transceiver, or an input / output interface; the processing unit may be at least one processor. Optionally, the transceiver may be a transceiver circuit. Optionally, the input / output interface may be an input / output circuit.

[0070] In another implementation, the device is a chip, a chip system or a circuit for the first node. When the device is a chip, a chip system or a circuit for a communication device, the communication unit may be an input / output interface, an interface circuit, an output circuit, an input circuit, a pin or a related circuit on the chip, the chip system or the circuit; the processing unit may be at least one processor, a processing circuit or a logic circuit.

[0071] In a sixth aspect, a communication device is provided, which is used to execute the method in any possible implementation of the second aspect or the fourth aspect. Specifically, the device may include a unit and / or module, such as a transceiver unit and / or a processing unit, for executing the method in any possible implementation of the second aspect or the fourth aspect.

[0072] In one implementation, the device is a second node. When the device is a communication device, the communication unit may be a transceiver, or an input / output interface; the processing unit may be at least one processor. Optionally, the transceiver may be a transceiver circuit. Optionally, the input / output interface may be an input / output circuit.

[0073] In another implementation, the device is a chip, a chip system or a circuit for the second node. When the device is a chip, a chip system or a circuit for a communication device, the communication unit may be an input / output interface, an interface circuit, an output circuit, an input circuit, a pin or a related circuit on the chip, the chip system or the circuit; the processing unit may be at least one processor, a processing circuit or a logic circuit.

[0074] In a seventh aspect, a communication device is provided, the device comprising: at least one processor, configured to execute a computer program or instruction stored in a memory, so as to execute the method in any possible implementation of any one of the first aspect and the third aspect. Optionally, the device further comprises a memory, configured to store a computer program or instruction. Optionally, the device further comprises a communication interface, and the processor reads the computer program or instruction stored in the memory through the communication interface.

[0075] In one implementation, the device is a first node.

[0076] In another implementation, the device is a chip, a chip system, or a circuit for the first node.

[0077] On the eighth aspect, a communication device is provided, the device comprising: at least one processor, configured to execute a computer program or instruction stored in a memory, so as to execute the method in any possible implementation of any of the second aspect and the fourth aspect. Optionally, the device further comprises a memory, configured to store a computer program or instruction. Optionally, the device further comprises a communication interface, and the processor reads the computer program or instruction stored in the memory through the communication interface.

[0078] In one implementation, the device is a second node.

[0079] In another implementation, the device is a chip, a chip system, or a circuit for the second node.

[0080] In a ninth aspect, the present application provides a processor, comprising: an input circuit, an output circuit, and a processing circuit. The processing circuit is used to receive a signal through the input circuit and transmit a signal through the output circuit, so that the processor executes the method in any possible implementation of any aspect from the first aspect to the fourth aspect.

[0081] In the specific implementation process, the processor can be one or more chips, the input circuit can be an input pin, the output circuit can be an output pin, and the processing circuit can be a transistor, a gate circuit, a trigger, and various logic circuits. The input signal received by the input circuit can be received and input by, for example, but not limited to, a transceiver, and the signal output by the output circuit can be, for example, but not limited to, output to a transmitter and transmitted by the transmitter, and the input circuit and the output circuit can be the same circuit, which is used as an input circuit and an output circuit at different times. The embodiments of the present application do not limit the specific implementation methods of the processor and various circuits.

[0082] For the operations such as sending and acquiring / receiving involved in the processor, unless otherwise specified, or unless they conflict with their actual function or internal logic in the relevant description, they can be understood as operations such as processor output, reception, input, etc., or as sending and receiving operations performed by the radio frequency circuit and antenna, and this application does not limit this.

[0083] In a tenth aspect, a processing device is provided, comprising a processor and a memory. The processor is used to read instructions stored in the memory, and can receive signals through a transceiver and transmit signals through a transmitter to execute the method in any possible implementation of any aspect from the first aspect to the fourth aspect.

[0084] Optionally, the number of the processors is one or more, and the number of the memories is one or more.

[0085] Optionally, the memory may be integrated with the processor, or the memory may be provided separately from the processor.

[0086] In the specific implementation process, the memory can be a non-transitory memory, such as a read-only memory (ROM), which can be integrated with the processor on the same chip or can be set on different chips respectively. The embodiments of the present application do not limit the type of memory and the setting method of the memory and the processor.

[0087] It should be understood that the related data interaction process, such as sending indication information, can be a process of outputting indication information from the processor, and receiving capability information can be a process of receiving input capability information by the processor. Specifically, the data output by the processor can be output to the transmitter, and the input data received by the processor can come from the transceiver. Among them, the transmitter and the transceiver can be collectively referred to as a transceiver.

[0088] The processing device in the ninth aspect may be one or more chips. The processor in the processing device may be implemented by hardware or by software. When implemented by hardware, the processor may be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor may be a general-purpose processor implemented by reading software code stored in a memory, which may be integrated in the processor or located outside the processor and exist independently.

[0089] In an eleventh aspect, a computer-readable storage medium is provided, which stores a program code for execution by a device, wherein the program code includes a method for executing any possible implementation of the first to fourth aspects above.

[0090] In a twelfth aspect, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to execute a method in any possible implementation of the first to fourth aspects above.

[0091] In the thirteenth aspect, a chip system is provided, comprising a processor for calling and running a computer program from a memory, so that a device equipped with the chip system executes the methods in each implementation manner in any one of the first to fourth aspects above.

[0092] In a fourteenth aspect, a communication system is provided, the communication system comprising the first node and the second node. The first node is used to execute any possible implementation method in any of the first and third aspects, and the second node is used to execute any possible implementation method in any of the second and fourth aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 A schematic diagram of the structure of a communication system.

[0094] Figure 2 This is a schematic diagram of a CU-DU architecture provided in this application.

[0095] Figure 3 A schematic diagram of the structure of a neuron.

[0096] Figure 4 A schematic diagram of the layer relationship of a neural network.

[0097] Figure 5 A schematic diagram of the training and reasoning framework of an AI model provided for this application.

[0098] Figure 6 This is a schematic diagram of an AI application framework in NR applicable to this application.

[0099] Figure 7 It is a schematic flowchart of a method 700 for transmitting an artificial intelligence model provided by the present application.

[0100] Figure 8 It is a schematic flowchart of a method 800 for transmitting training data provided by the present application.

[0101] Fig. 9 It is a schematic block diagram of the communication device 900 provided in this application.

[0102] Fig.10 It is a schematic block diagram of the communication device 1000 provided in this application. DETAILED DESCRIPTION

[0103] The technical solution in this application will be described below in conjunction with the accompanying drawings.

[0104] The technology provided in this application can be applied to various communication systems, for example, the communication system can be a fourth generation (4G) communication system (such as a long term evolution (LTE) system), a fifth generation (5G) communication system, a worldwide interoperability for microwave access (WiMAX) or a wireless local area network (WLAN) system, a satellite communication system, a future communication system, such as a sixth generation (6G) mobile communication system, or a fusion system of multiple systems. Among them, the 5G communication system can also be called a new radio (NR) system. Satellite communication system, future communication system, such as a sixth generation (6G) mobile communication system, or a fusion system of multiple systems.

[0105] A device in a communication system can send a signal to another device or receive a signal from another device, wherein the signal may include information, signaling or data, etc. Among them, the device can also be replaced by an entity, a network entity, a communication device, a communication module, a node, a communication node, etc. For example, a communication system may include at least one terminal device and at least one network device. For another example, a communication system may include a training device and at least one network device. The network device can send a downlink signal to the terminal device, and / or the terminal device can send an uplink signal to the access network device. In addition, it can be understood that if the communication system includes multiple terminal devices, multiple terminal devices can also send signals to each other, that is, the signal sending network element and the signal receiving network element can both be terminal devices. It can be understood that the terminal device in the present application can be replaced by the first device, and the network device can be replaced by the second device, and the two perform the corresponding communication method in the present disclosure.

[0106] The method provided in the embodiment of the present application can be applied to wireless communication systems such as 5G, 6G, satellite communication, etc. Figure 1 , Figure 1 is a simplified schematic diagram of a wireless communication system provided by an embodiment of the present application. Figure 1 As shown, the wireless communication system includes a wireless access network 100 (an example of a network device). The wireless access network 100 may be a next generation (e.g., 6G or higher) wireless access network, or a traditional (e.g., 5G, 4G, 3G, or 2G) wireless access network. One or more communication devices (120a-120j, collectively referred to as 120) may be connected to each other or to one or more network devices (110a, 110b, collectively referred to as 110) in the wireless access network 100. Optionally, Figure 1 This is just a schematic diagram. The wireless communication system may also include other devices, such as core network devices, wireless relay devices and / or wireless backhaul devices. Figure 1 Not drawn in.

[0107] Optionally, in practical applications, the wireless communication system may include multiple network devices (e.g., access network devices) at the same time, and may also include multiple communication devices at the same time. A network device may serve one or more communication devices at the same time. A communication device may also access one or more network devices at the same time. The embodiment of the present application does not limit the number of communication devices and network devices included in the wireless communication system.

[0108] The network device may be an entity on the network side for transmitting or receiving signals. The network device may be an access device for a communication device to access the wireless communication system in a wireless manner, for example, the network device may be a base station. The base station can broadly cover the following various names, or be replaced with the following names, such as: NodeB, evolved NodeB (eNB), next generation NodeB (gNB), access network equipment in open radio access network (O-RAN), relay station, access point, transmission point (transmitting and receiving point, TRP), transmission point (transmitting point, TP), master station MeNB, secondary station SeNB, multi-standard radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), radio head (RRH), central unit (CU), distributed unit (DU), radio unit (radio unit, RU), centralized unit control plane (CUcontrol plane, CU-CP) node, centralized unit user plane (CU user plane, CU-UP) node, positioning node, etc. The base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof. The network device may also refer to a communication module, a modem or a chip used to be arranged in the aforementioned device or apparatus. The network device may also be a mobile switching center and a device to device (Device-to-Device, D2D), vehicle external connection (vehicle-to-everything, V2X), a device that performs the base station function in machine-to-machine (machine-to-machine, M2M) communication, a network side device in a 6G network, and a device that performs the base station function in a future communication system. The network device may support networks with the same or different access technologies. The embodiments of the present application do not limit the specific technology and specific device form adopted by the network device.

[0109] In a network structure, the network device may also refer to a centralized unit (central unit, CU) or a distributed unit (distributed unit, DU), or the network device may also be composed of a CU and a DU. CU and DU can be understood as the division of the base station from the perspective of logical functions. Among them, the CU and DU can be physically separated or deployed together, and the embodiment of the present application does not specifically limit this. A CU can be connected to a DU, or multiple DUs can share a CU, which can save costs and facilitate network expansion. The division of CU and DU can be divided according to the protocol stack, and one possible way is to deploy the Radio Resource Control (RRC), Service Data Adaptation Protocol Stack (SDAP) and Packet Data Convergence Protocol (PDCP) layer in the CU, and the remaining Radio Link Control (RLC) layer, Media Access Control (MAC) layer and physical layer in the DU. The present application does not completely limit the above-mentioned protocol stack segmentation method, and there may be other segmentation methods. For details, please refer to the technical research report (technical report, TR) 38.801v14.0.0.

[0110] Figure 2 A schematic diagram of a CU-DU architecture is shown, Figure 2 As shown, CU and DU are connected via the F1 interface. CU represents gNB and is connected to the core network via the Ng interface. The network device in the embodiment of the present application may also refer to a centralized unit control plane (CU-CP) node or a centralized unit user plane (CU-UP) node, or the network device may also be CU-CP and CU-UP. CU-CP is responsible for control plane functions, mainly including RRC and PDCP-C. PDCP-C is mainly responsible for encryption and decryption, integrity protection, data transmission, etc. of control plane data. CU-UP is responsible for user plane functions, mainly including SDAP and PDCP-U. SDAP is mainly responsible for processing the data of the core network and mapping the flow to the bearer. PDCP-U is mainly responsible for encryption and decryption, integrity protection, header compression, sequence number maintenance, data transmission, etc. of the data plane. CU-CP and CU-UP are connected via the E1 interface. CU-CP represents gNB and is connected to the core network via the Ng interface. Connected to DU via F1-C (control plane). CU-UP is connected to DU via F1-U (user plane). Of course, another possible implementation is that PDCP-C is also in CU-UP.

[0111] The access network device mentioned in the embodiments of the present application may be a device including a CU, or a DU, or a device including a CU and a DU, or a device including a control plane CU node (CU-CP node) and a user plane CU node (CU-UP node) and a DU node.

[0112] The network equipment may be fixed or mobile. For example, base stations 110a and 110b (examples of network equipment) are stationary and are responsible for wireless transmission and reception in one or more cells from the communication device 120. Figure 1 The helicopter or drone 120i shown in the figure can be configured to act as a mobile base station, and one or more cells can move according to the location of the mobile base station 120i. In other examples, the helicopter or drone (120i) can be configured to be used as a communication device for communicating with the base station 110b.

[0113] In the present application, the communication device used to implement the above network functions may be, for example, an access network device, or a network device having some functions of accessing the network, or a device capable of supporting the implementation of the access network function, such as a chip system, a hardware circuit, a software module, or a hardware circuit plus a software module. The device may be installed in the access network device or used in combination with the access network device.

[0114] In the present application, a communication device may be an entity on the user side for receiving or transmitting signals, such as a mobile phone. The communication device may be used to connect people, objects and machines. The communication device may communicate with one or more core networks through a network device. The communication device includes a handheld device with a wireless connection function, other processing devices connected to a wireless modem, or a vehicle-mounted device. The communication device may be a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device. The communication device 120 may be widely used in various scenarios, such as cellular communication, device-to-device D2D, vehicle-to-all V2X, end-to-end P2P, machine-to-machine M2M, machine type communication MTC, Internet of Things IOT, virtual reality VR, augmented reality AR, industrial control, autonomous driving, telemedicine, smart grid, smart furniture, smart office, smart wear, smart transportation, smart city, drone, robot, remote sensing, passive sensing, positioning, navigation and tracking, autonomous delivery and mobility, etc. Some examples of the communication device 120 include: 3GPP standard user equipment (UE), fixed equipment, mobile equipment, handheld equipment, wearable equipment, cellular phone, smart phone, session initiation protocol (SIP) phone, laptop, personal computer, smart book, vehicle, satellite, global positioning system (GPS) equipment, target tracking equipment, drone, helicopter, aircraft, ship, remote control equipment, smart home equipment, industrial equipment, personal communication service (PCS) phone, wireless local loop (WLL) station, personal digital assistant (PDA), wireless network camera, tablet computer, PDA, mobile internet device (MID), wearable equipment such as smart watch, virtual reality (VR) equipment, augmented reality (AR) equipment, wireless terminal in industrial control, terminal in vehicle networking system, wireless terminal in self driving, wireless terminal in smart grid, wireless terminal in transportation safety, smart city, etc. The communication device 120 may be a wireless device in the above scenarios or a device used to be set in a wireless device, for example, a communication module, a modem or a chip in the above device.The communication device may also be referred to as a terminal, terminal equipment, user equipment (UE), mobile station (MS), mobile terminal (MT), etc. The communication device may also be a communication device in a future wireless communication system. The communication device may be used in a dedicated network device or a general device. The embodiments of the present application do not limit the specific technology and specific device form used by the communication device.

[0115] Optionally, the communication device may be used to act as a base station. For example, the UE may act as a scheduling entity that provides sidelink signals between UEs in V2X, D2D, or P2P, etc. Figure 1 As shown, the cell phone 120a and the car 120b communicate with each other using a sidelink signal. The cell phone 120a and the smart home device 120e communicate with each other without relaying the communication signal through the base station 110b.

[0116] In the present application, the communication device used to implement the functions of the communication device may be a terminal device, or a terminal device having some of the functions of the above communication device, or a device capable of supporting the functions of the above communication device, such as a chip system, which may be installed in the terminal device or used in combination with the terminal device. In the present application, the chip system may be composed of a chip, or may include a chip and other discrete devices.

[0117] Optionally, a wireless communication system is usually composed of cells, and the base station provides management of the cell. The base station provides communication services to multiple mobile stations (MS) in the cell. The base station includes a baseband unit (BBU) and a remote radio unit (RRU). The BBU and RRU can be placed in different places, for example: the RRU is remote and placed in an area with high traffic volume, and the BBU is placed in a central computer room. The BBU and RRU can also be placed in the same computer room. The BBU and RRU can also be different components under one rack. Optionally, a cell can correspond to a carrier or a component carrier.

[0118] It should be understood that Figure 1 The number and type of each device in the communication system shown are for illustration only, and the present application is not limited thereto. In actual applications, the communication system may also include more terminal devices, more network devices, and other network elements, such as core network devices, and / or network elements for implementing artificial intelligence functions.

[0119] The wireless access network equipment and terminal equipment can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on the water surface; they can also be deployed on aircraft, balloons and satellites in the air. The embodiments of the present application do not limit the application scenarios of the wireless access network equipment and terminal equipment.

[0120] The embodiments of the present application may be applicable to downlink signal transmission, uplink signal transmission, and device-to-device (D2D) signal transmission. For downlink signal transmission, the transmitting device is a wireless access network device, and the corresponding receiving device is a terminal device. For uplink signal transmission, the transmitting device is a terminal device, and the corresponding receiving device is a wireless access network device. For D2D signal transmission, the transmitting device is a terminal device, and the corresponding receiving device is also a terminal device. The embodiments of the present application do not limit the transmission direction of the signal.

[0121] The wireless access network equipment and the terminal equipment, as well as the terminal equipment and the terminal equipment, can communicate through the licensed spectrum (licensed spectrum), can also communicate through the unlicensed spectrum (unlicensed spectrum), or can also communicate through the licensed spectrum and the unlicensed spectrum at the same time. The wireless access network equipment and the terminal equipment, as well as the terminal equipment and the terminal equipment, can communicate through the spectrum below 6 gigahertz (GHz), can also communicate through the spectrum above 6G, and can also communicate using the spectrum below 6G and the spectrum above 6G at the same time. The embodiments of the present application do not limit the spectrum resources used between the wireless access network equipment and the terminal equipment.

[0122] In order to facilitate understanding of the technical solution provided by the present application, the following first briefly introduces the professional terms involved in the present application. It should be understood that the introduction is not intended to limit the present application.

[0123] 1. AI Model

[0124] The AI ​​model is the specific implementation of the AI ​​technology function. The AI ​​model represents the mapping relationship between the model's input and output. The type of AI model can be a neural network, linear regression model, decision tree model, support vector machine (SVM), Bayesian network, Q learning model or other machine learning (ML) model.

[0125] Depending on the specific methods and / or technologies used to implement artificial intelligence, AI models can also be specifically referred to as machine learning models, deep learning models, or reinforcement learning models. Machine learning is a method for implementing artificial intelligence. The goal of this method is to design and analyze some algorithms (also known as "models") that allow computers to automatically "learn". The designed algorithms are called "machine learning models". Machine learning models are a type of algorithm that automatically analyzes data to obtain patterns and uses the patterns to predict unknown data. There are many types of machine learning models. Depending on whether the model training needs to rely on the labels corresponding to the training data, machine learning models can be divided into supervised learning models and unsupervised learning models. The following mainly introduces "supervised learning models".

[0126] 2. Deep neural network (DNN)

[0127] DNN is a specific implementation of machine learning. According to the universal approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling neural networks to learn any mapping. Traditional communication systems require extensive expert knowledge to design communication modules, while DNN-based deep learning communication systems can automatically discover implicit pattern structures from large data sets, establish mapping relationships between data, and achieve performance that is superior to traditional modeling methods.

[0128] The idea of ​​DNN comes from the neuron structure of the brain. Each neuron performs a weighted sum operation on its input value and generates an output through a nonlinear function. For example, Figure 3 Specifically, assume that the neural input is x = [x0,…,x n ], and the weight corresponding to the input is d=[d0,…,d n ], the bias of the weighted sum is b, and the form of the nonlinear function can be diversified. For example, if the nonlinear function is max{0,x}, the effect of the execution of a neuron can be At this time, the weight d=[d0,…,d n ] and bias b can be understood as the parameters of the model.

[0129] DNN generally has a multi-layer structure. Each layer of DNN can contain multiple neurons. The input layer passes the received values ​​to the middle hidden layer after being processed by neurons. Similarly, the hidden layer passes the calculation results to the final output layer to generate the final output of DNN, such as Figure 4As shown in the figure. DNN generally has more than one hidden layer, and the hidden layer often directly affects the ability to extract information and fit functions. Increasing the number of hidden layers of DNN or expanding the width of each layer can improve the function fitting ability of DNN. The weighted value in each neuron is the parameter of the DNN network model. The model parameters are optimized through the training process, so that the DNN network has the ability to extract data features and express mapping relationships.

[0130] 3. AI model training and reasoning

[0131] Before any AI model can be used to solve a specific technical problem, it needs to be trained. Figure 5 The training of the AI ​​model shown here refers to the process of calculating the training data using the specified initial model, and adjusting the parameters in the initial model using a certain method based on the calculation results, so that the model gradually learns certain rules and has specific functions. After training, the AI ​​model with stable functions can be used for reasoning. The reasoning of the AI ​​model is the process of using the trained AI model to calculate the input data and obtain the predicted reasoning results.

[0132] During the training phase, you first need to build a training set for the deep learning model based on the goal. The training set includes multiple training data, each of which is assigned a label. The label of the training data is the correct answer to a specific question. The label can represent the goal of using the training data to train the deep learning model.

[0133] When training a deep learning model, the training data can be input into the deep learning model after parameter initialization in batches. The deep learning model calculates the training data (i.e., "inference") to obtain prediction results for the training data. The prediction results obtained through inference and the labels corresponding to the training data are used as data for calculating the loss according to the loss function. The loss function is a function used to calculate the gap between the model's prediction results for the training data and the labels of the training data (i.e., "loss value") during the model training phase. The loss function can be implemented using different mathematical functions. Commonly used expressions of loss functions include: mean square error loss function, logarithmic loss function, least squares method, etc. Model training is a repeated iterative process. Each iteration infers different training data and calculates the loss value. The goal of multiple iterations is to continuously update the parameters of the deep learning model and find the parameter configuration that minimizes or stabilizes the loss value of the loss function.

[0134] 4. Training data and inference data

[0135] "Training data" is used for the training of AI models. The training data may include the input of the AI ​​model, or the input and target output of the AI ​​model. Among them, the training data includes one or more training data. The training data may be a training sample input to the AI ​​model, or it may be the target output of the AI ​​model. Among them, the target output may also be referred to as a "label" or a "label sample". Training data is one of the important parts of machine learning. Model training is essentially to learn certain features of the training data from the training data so that the output of the AI ​​model is as close to the target output as possible, for example, to make the difference between the output of the AI ​​model and the target output as small as possible. The composition and selection of the training data set can determine the performance of the trained AI model to a certain extent. Among them, the performance of the model can be measured, for example, by "loss value", "inference accuracy" and the like.

[0136] In addition, during the training process of an AI model (such as a neural network), a loss function can be defined. The loss function describes the gap or difference between the output value of the AI ​​model and the target output value. This application does not limit the specific form of the loss function. The training process of the AI ​​model is to adjust the model parameters of the AI ​​model so that the value of the loss function is less than the threshold, or the value of the loss function meets the target requirements. For example, the AI ​​model is a neural network, and adjusting the model parameters of the neural network includes adjusting at least one of the following parameters: the number of layers, width, weights of neurons, or parameters in the activation function of neurons.

[0137] "Inference data" can be used as the input of a trained AI model for the inference of the AI ​​model. During the model inference process, the inference data is input into the AI ​​model, and the corresponding output is the inference result.

[0138] Figure 6 This is an AI application framework diagram in NR applicable to this application, such as Figure 6 As shown, data collection (data collection) 610 can store data inputs from gNB, gNB-CU, gNB-DU, UE or other management entities as a database for AI model training and data analysis and reasoning. Model training (model training) 620 analyzes the training data (training data) provided by data collection 610 to provide the optimal AI model. Model inference (model inference) 630 uses the AI ​​model to provide reasonable AI-based predictions for network operation based on the data provided by data collection 610, or guide the network to make policy adjustments. The relevant policy adjustments are uniformly planned by the action entity 640 (actor) and sent to multiple network entities for operation. At the same time, after the relevant policies are applied, the specific performance of the network will be input into the database again for storage.

[0139] At present, the RAN3 standard of release 18 (R18) specifies that artificial intelligence (AI) or machine learning (ML) models are trained and inferred separately in different network entities (e.g., gNB, operation, maintenance and management (OAM) network element). For example, the specific scenarios include the following two: AI training and AI inference in gNB, or AI training in OAM and AI inference in gNB. Currently, R18 does not involve the transmission of AI models across network entities. However, the transmission of models across nodes has the following benefits: (1) Improving the accuracy of AI prediction information or decisions. For example, if the accuracy of AI models of some nodes is too low, high-precision reasoning information can be obtained by transmitting high-precision AI models of other nodes across nodes. (2) Nodes without AI model training resources can obtain AI models. For example, if the computing power and / or storage resources of some nodes are insufficient and AI models cannot be trained, the trained AI models can be obtained by transmitting the AI ​​models of other nodes across nodes. (3) Reducing the resource consumption of nodes. For example, in the scenario where the same manufacturer (or manufacturers that have signed an agreement) can share models, it is not necessary for all nodes to train AI models. Some nodes with low computing power can obtain AI models of specific accuracy by transmitting the AI ​​models of other nodes across nodes. However, in the application scenario of transmitting AI models across network entities, when different network entities belong to different manufacturers, the transmission of AI models across nodes may involve privacy issues related to AI models. For example, the AI ​​models trained by each manufacturer cannot be exposed to other manufacturers; for another example, the training data involved in each AI model may also contain the user's private data and cannot be exposed to other manufacturers. Therefore, how to achieve cross-network entity transmission of AI models has become a technical problem that needs to be solved.

[0140] In view of this, the present application provides a method for transmitting an AI model, in which the first node can request the second node for the pre-configuration information of the AI ​​model that the second node can provide and the manufacturer identifier corresponding to the second node, and the first node carries its own corresponding manufacturer identifier in the request message, the first node can determine the AI ​​model that needs to be requested from the second node based on the pre-configuration information of the AI ​​model, and the second node can determine whether it belongs to the same manufacturer as the first node based on the manufacturer identifier corresponding to the first node, thereby determining how to send the AI ​​model required by the first node in the future. The method provided by the present application can realize the transmission of AI models across network entities and avoid the privacy issues of AI models.

[0141] It should be noted that the “AI model” in this application can also be replaced by the “ML model”. For the convenience of description, the following unified description is “AI model”.

[0142] Figure 7 is a schematic flow chart of a method 700 for transmitting an AI model provided by the present application, such as Figure 7 As shown, the method includes:

[0143] 710. The first node sends a first request message to the second node. The first request message is used to request the pre-configuration information of the AI ​​model that the second node can provide and the manufacturer identifier corresponding to the second node. The first request message carries the manufacturer identifier corresponding to the first node.

[0144] Correspondingly, the second node receives the first request message from the first node.

[0145] In the present application, the first node and the second node can be any one of the following nodes: gNB, gNB-CU, gNB-DU, UE, 5GC, OAM, etc.

[0146] In one possible implementation, the first node and the second node may be different nodes of the same manufacturer; in another possible implementation, the first node and the second node are nodes of different manufacturers. It can also be understood that in this application, the first node and the second node are not limited.

[0147] In this application, for example, when the first node is operating manufacturer #1, the manufacturer ID corresponding to the first node can be understood as the vendor ID of operating manufacturer #1, and the manufacturer ID can be used by the second node to determine whether it is the same manufacturer or a different manufacturer as the first node. In addition, the first node also requests the manufacturer ID corresponding to the second node from the second node, and the pre-configuration information of the AI ​​model provided by the second node can be associated with the manufacturer ID of the second node later. Alternatively, it can also be understood that the first node will determine that the pre-configuration information of the AI ​​model received is sent by the manufacturer corresponding to the second node.

[0148] In one possible implementation, the first request message also carries pre-configuration information of an AI model that the first node can provide to the second node, and the pre-configuration information of the AI ​​model is used by the second node to subsequently determine the AI ​​model to be requested, wherein the AI ​​model to be requested is one or more of the AI ​​models that the first node can provide.

[0149] In one possible implementation, the pre-configuration information of the AI ​​model includes a third association relationship. The "third association relationship" can also be understood as an association relationship between at least the following two items: a model identifier corresponding to the AI ​​model, a model algorithm corresponding to the AI ​​model, a model algorithm identifier corresponding to the AI ​​model algorithm, a service corresponding to the AI ​​model, a service identifier corresponding to the AI ​​model service, a purpose corresponding to the AI ​​model, a purpose identifier corresponding to the purpose, reasoning information corresponding to the AI ​​model, model information corresponding to the AI ​​model, and a cell identifier (cell ID).

[0150] Exemplarily, the third association relationship includes one or more of the following association relationships: the association relationship between the model identifier, model algorithm, and model algorithm identifier corresponding to the AI ​​model; the association relationship between the model identifier, business, and business identifier corresponding to the AI ​​model; the association relationship between the model identifier, purpose, and purpose identifier corresponding to the AI ​​model; the association relationship between the model identifier, model information, and model reasoning information corresponding to the AI ​​model; the association relationship between the model identifier corresponding to the AI ​​model and the reasoning information corresponding to the AI ​​model, and the association relationship between the model identifier corresponding to the AI ​​model and the model information.

[0151] In this application, model information is used to characterize one or more of the following aspects of the AI ​​model: the input of the AI ​​model, the output of the AI ​​model, the number of layers of the AI ​​model, and the dimension of the AI ​​model.

[0152] The “AI model algorithm” mentioned in this application may include, for example: decision tree algorithm, random forest algorithm, logistic regression algorithm, naive Bayes algorithm, K nearest neighbor algorithm, K means algorithm, Adaboost algorithm, neural network algorithm, Markov algorithm, etc.

[0153] The "AI model services" mentioned in this application can be, for example, application scenarios designed in the 3rd Generation Partnership Project (3GPP), such as: channel state information reference signal (CSI-RS) feedback enhanced services, beam scanning enhanced services, positioning enhanced services, network energy-saving services, load balancing services, mobility optimization services, and any other use cases that may be expanded in the future, etc.

[0154] The "purpose corresponding to the AI ​​model" mentioned in this application can also be understood as the "business corresponding to the AI ​​model". In other words, the "purpose", "business" and "scenario" corresponding to the AI ​​model in this application can be described interchangeably.

[0155] The "inference information corresponding to the AI ​​model" mentioned in this application can be understood as the reasoning result of the model, which can be prediction information, prediction strategy, etc.

[0156] Exemplarily, each of the above association relationships may be presented in the form of a table, as shown in Table 1 to Table 5 below.

[0157] Table 1

[0158] Model Identification Model algorithm identification Model Algorithm Model ID #1 <![CDATA[Model algorithm identifier #m1]]> Model Algorithm #M1 Model ID #2 <![CDATA[Model algorithm identifier #m2]]> Model Algorithm #M2 Model ID #3 <![CDATA[Model algorithm identifier #m3]]> Model Algorithm #M3

[0159] Table 2

[0160] Model Identification Model Business Identity business Model ID #1 <![CDATA[Model business identifier #s1]]> Business #S1 Model ID #2 <![CDATA[Model business identifier #s2]]> Business #S2 Model ID #3 <![CDATA[Model business identifier #s3]]> Business#S3

[0161] Table 3

[0162] Model Identification Model usage identification use Model ID #1 <![CDATA[Model usage identifier #u1]]> Usage#U1 Model ID #2 <![CDATA[Model usage identifier #u2]]> Usage#U2 Model ID #3 <![CDATA[Model usage identifier #u3]]> Usage#U3

[0163] Table 4

[0164] Model Identification Model inference information Model ID #1 Model inference information #Z1 Model ID #2 Model inference information #Z2 Model ID #3 Model reasoning information Z3

[0165] Table 5

[0166]

[0167]

[0168] Exemplarily, the pre-configuration information of the AI ​​model supported by the first node to be provided to the second node may also include multiple association relationships, as shown in Tables 6 and 7 below.

[0169] Table 6

[0170] Model Identification Model Business Identity business Model algorithm identification Model Algorithm Model ID #1 Model Business Identity #s1 Business #S1 <![CDATA[Model algorithm identifier #m1]]> Model Algorithm #M1 Model ID #2 Model Business Logo #s2 Business #S2 <![CDATA[Model algorithm identifier #m2]]> Model Algorithm #M2 Model ID #3 Model Business Identity #s3 Business#S3 <![CDATA[Model algorithm identifier #m3]]> Model Algorithm #M3

[0171] Table 7

[0172]

[0173] It can be seen that in the present application, the first node sends the pre-configuration information of its own AI model to the second node, so that the second node can determine the AI ​​model that needs to be requested from the first node subsequently based on the association relationship in the pre-configuration information of the AI ​​model.

[0174] It should be noted that, in the present application, the above-mentioned "association relationship" can also be replaced by "correspondence relationship", "mapping relationship", etc.

[0175] Optionally, the pre-configuration information of the AI ​​model provided by the first node to the second node may also include one or more of the following: information on the AI ​​model scrambling method supported by the first node, information on the AI ​​model update method supported by the first node, and information on the AI ​​model transmission method supported by the first node. Among them, the information on the AI ​​model scrambling method includes: the association between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; the information on the AI ​​model update method includes: the association between the AI ​​model update method identifier and the corresponding update method; the information on the AI ​​model transmission method includes: the association between the AI ​​model transmission method identifier and the corresponding transmission method. The above-mentioned information is introduced below.

[0176] Taking the above-mentioned AI model scrambling method information as an example, for example, the scrambling method may include one or more of the following: differential privacy based on input perturbation / intermediate parameter perturbation, based on distributed multi-party secure computing, based on homomorphic encryption, based on noise power; for example, taking differential privacy as an example, it may be the standard deviation σ of the Gaussian distribution when adding Gaussian noise; for another example, the scrambling level may include level one, level two, and level three. Exemplarily, assuming that the scrambling method associated with the scrambling method identifier #1 of AI model #1 is a homomorphic encryption method, and the scrambling level is level one; for example, the scrambling method associated with the scrambling method identifier #2 of AI model #1 is a differential privacy method based on input perturbation, and the scrambling level is level two; for another example, the scrambling method associated with the scrambling method identifier #3 of the AI ​​model is a method based on noise power. For another example, the scrambling level associated with the scrambling method identifier #4 of the AI ​​model is level three. In this application, the specific indication method of the scrambling method and / or scrambling level associated with the scrambling method identifier is not limited, and those skilled in the art can flexibly design the indication method.

[0177] In the present application, "AI model update mode" is used to indicate the update mode of the received AI model that the first node can support. For example, the model update mode can be a "model retransmission" mode or a "weight update" mode. Among them, "model retransmission" means that the requested node re-sends the updated AI model, and "weight update" means that the requested node only needs to transmit the weight parameters that need to be updated in the model. Exemplarily, it is assumed that the AI ​​model update mode corresponding to the AI ​​model update mode identifier #1 is the model retransmission mode, and it is assumed that the AI ​​model update mode corresponding to the AI ​​model update mode identifier #2 is the weight update mode. Specifically, the specific implementation method of the "AI model update mode" can be understood with reference to the above-mentioned "AI model scrambling mode information", which will not be repeated here.

[0178] In the present application, "AI model transmission mode" is used to indicate the AI ​​model transmission mode that the first node can support. Depending on the different entities of the requesting node and the requested node, possible transmission modes are: UP-based transmission, CP-based transmission, Internet protocol channel (internet protocol tunnel, IP tunnel)-based transmission, forwarding by 5GC, etc. Optionally, the decision conditions for AI model transmission can also be carried. For example, when the AI ​​model is small, you can choose to transmit the AI ​​model based on CP, when the AI ​​model is large, you can choose to transmit the AI ​​model based on UP, when the AI ​​model is non-air interface transmission, the AI ​​model is transmitted based on IP tunnel, and when the AI ​​model is large and 5GC has the latest AI model, the AI ​​model is forwarded based on 5GC. Specifically, the specific implementation method of the "AI model transmission mode" can be understood with reference to the above-mentioned "AI model scrambling method information", which will not be repeated.

[0179] 720. The second node sends a first response message to the first node. The first response message carries pre-configuration information of the AI ​​model that the second node can provide and the manufacturer identifier corresponding to the second node.

[0180] Correspondingly, the first node receives the first response message.

[0181] The pre-configuration information of the AI ​​model is used by the first node to determine the first AI model to be requested, and the first AI model is one or more of the AI ​​models that the second node can provide.

[0182] In one possible implementation, the pre-configuration information of the AI ​​model that the second node can provide includes a first association relationship, where the first association relationship is an association relationship between at least two of the following: a model identifier corresponding to the AI ​​model, a model algorithm corresponding to the AI ​​model, a model algorithm identifier corresponding to the AI ​​model algorithm, a business corresponding to the AI ​​model, a business identifier corresponding to the AI ​​model business, a purpose corresponding to the AI ​​model, a purpose identifier corresponding to the purpose, reasoning information corresponding to the AI ​​model, and model information corresponding to the AI ​​model.

[0183] Exemplarily, the first association relationship includes one or more of the following association relationships: the association relationship between the model identifier, model algorithm, and model algorithm identifier corresponding to the AI ​​model; the association relationship between the model identifier, business, and business identifier corresponding to the AI ​​model; the association relationship between the model identifier, purpose, and purpose identifier corresponding to the AI ​​model; the association relationship between the model identifier, reasoning information, and model information corresponding to the AI ​​model; the association relationship between the model identifier corresponding to the AI ​​model and the reasoning information corresponding to the AI ​​model; the association relationship between the model identifier corresponding to the AI ​​model and the model information.

[0184] Among them, the model information can be used to characterize the characteristics of the AI ​​model, for example, it can be one or more of the following: the input of the AI ​​model, the output of the AI ​​model, the number of layers of the AI ​​model, and the dimension of the AI ​​model.

[0185] Exemplarily, the above-mentioned various association relationships can be understood by referring to Tables 1 to 7 above, and no further examples are given one by one.

[0186] Optionally, the pre-configuration information of the AI ​​model that the second node can provide also includes one or more of the following: information on the AI ​​model scrambling method supported by the second node, information on the AI ​​model updating method supported by the second node, and information on the AI ​​model transmission method supported by the second node.

[0187] Assume that the second node is a node of vendor #2, then the vendor identifier corresponding to the second node is vendor ID #2.

[0188] In the present application, for AI pre-configuration information that the second node cannot transmit to the first node, the second node can feedback indication information and its reason to the first node. For example, the second node does not agree to transmit the AI ​​model, or the second node does not accept the scrambling method of the AI ​​model, the second node does not support the model transmission method, etc.

[0189] In the present application, the specific signaling of step 710 and step 720 may be a new process or a reused legacy process. In the present application, the name and implementation method of the specific signaling of step 710 and step 720 are not limited. For example, the specific signaling of step 710 may be an "XN SETUP" signaling, and the specific signaling of step 720 may be an "NG-RAN NODECONFIGURATION" signaling.

[0190] Based on the above steps 710 to 720, in this application, the first node can request the pre-configuration information of the AI ​​model of the second node through the request-response mechanism, so that the first node determines the AI ​​model that needs to be requested from the second node in the future according to the pre-configuration information of the AI ​​model that the second node can provide. In addition, in this application, the first node carries its own manufacturer identification in the request message, so that the second node can determine whether it belongs to the same manufacturer as the first node through the manufacturer identification of the first node, thereby determining how to send the AI ​​model when sending the AI ​​model to the first node in the future, thereby ensuring that there will be no privacy issues with the AI ​​model.

[0191] 730. The first node sends a second request message to the second node, where the second request message is used to request the first AI model. The second request message carries first indication information, where the first indication information is used to indicate pre-configuration information of the first AI model.

[0192] The corresponding second node receives the second request message from the first node.

[0193] In the present application, the pre-configuration information of the first AI model may indicate the requested AI model to the second node.

[0194] Exemplarily, the second request message may be an AI model request message.

[0195] In a possible implementation manner, the pre-configuration information of the first AI model includes one or more of the following: a model identifier corresponding to the first AI model, model information corresponding to the first AI model, a model algorithm identifier corresponding to the first AI model, a service identifier corresponding to the first AI model, a purpose identifier corresponding to the first AI model, an AI model scrambling method identifier supported by the first node, an AI model update method identifier supported by the first node, an AI model transmission method identifier supported by the first node, a cell identifier, a reason for requesting the first AI model, and a model accuracy corresponding to the first AI model.

[0196] The “reason for requesting the first AI model” may be, for example, that the first node requires a high-precision AI model, or that certain AI models cannot be trained on the first node, etc. “Model accuracy corresponding to the first AI model” indicates the accuracy requirement of the first node for the requested model, such as the model confidence of the requested model > 0.9, or the number of training steps of the requested AI model > 1000, or the requested AI model accuracy > 0.95.

[0197] Exemplarily, the second request message may carry AI model identifier #1 and AI model identifier #2. Exemplarily, the second request message may carry model algorithm identifier #m1, model algorithm identifier #m2, service #S1, service #S2, and AI model identifier #1, AI model identifier #2.

[0198] Optionally, the second request message also carries one or more of the following: a period for the second node to report the first AI model, a message identifier corresponding to the second request message, and a message type corresponding to the second request message.

[0199] Among them, "the reporting period of the second response message" means that the first node instructs the second node to regularly report the updated AI model to the first node. The "message identifier corresponding to the second request message" can be used to associate subsequent steps. For example, after the first node subsequently receives a response message from the second node, if the response message also carries the message identifier, the first node can determine that the response message is a reply message to the second request message. For example, the message identifier can be Measurement ID#1. In the present application, the "message identifier corresponding to the second request message" can also be replaced with the identifier corresponding to the process of steps 730 to 770.

[0200] "The message type corresponding to the second request message" may, for example, include the request message type for starting to send the requested AI model (start type), the request message type for stopping sending the requested AI model (stop type), the request message type for adding the requested AI model (add type), the request message type for deleting the requested AI model (del type), and the request message type for updating the requested AI model (revise type).

[0201] 740. The second node sends a second response message to the first node, where the second response message carries second indication information, and the second indication information is used to indicate pre-configuration information of the second AI model.

[0202] The pre-configuration information of the second AI model is information that the second node supports sending to the first node in the pre-configuration information of the first AI model, and the second AI model is one or more AI models in the first AI model.

[0203] Correspondingly, the first node receives a second response message from the second node.

[0204] Exemplarily, the second response message may be an AI model response message (AI model request response).

[0205] In the present application, the pre-configuration information of the second AI model includes one or more of the following: a model identifier corresponding to the second AI model, model information corresponding to the second AI model, a model algorithm identifier corresponding to the second AI model, a service identifier corresponding to the second AI model, and model accuracy feedback information corresponding to the second AI model. Among them, the "model accuracy feedback information" is used to indicate that the second node supports the model accuracy corresponding to the second AI model sent to the first node.

[0206] Optionally, the second response message also carries one or more of the following: reporting cycle feedback information, a message identifier corresponding to the second request message (for example, Measurement ID#1), and failure cause indication information. The "reporting cycle feedback information" is used to indicate the AI ​​model reporting cycle supported by the second node. For example, the second node may indicate a new reporting cycle for the first node to select. The specific implementation form may be a time list, a time range, a time identifier, etc., without limitation. The "failure cause indication information" is used to indicate the identifier and reason why the second node does not support the first AI model sent to the first node. For example, it may be because the model accuracy of the AI ​​model that the second node can provide to the first node does not meet the requirements of the first node, or because privacy is not allowed.

[0207] In one possible implementation, if the second node cannot send any AI model requested by the first node to the first node, the second node may send an AI model failure message to the first node in response to the second request message sent by the first node. For example, the AI ​​model failure message may include a message identifier and a failure cause. The AI ​​model failure message may be used to indicate the identifier and cause of some AI models that cannot be provided, for example, the accuracy does not meet the requirements due to privacy.

[0208] Optionally, the method further includes step 750, where the first node determines fourth indication information based on the second indication information.

[0209] The fourth indication information is used to indicate pre-configuration information of the third AI model, wherein the pre-configuration information of the third AI model is information determined by the first node to request from the second node in the pre-configuration information of the second AI model, and the third AI model is one or more AI models in the second AI model.

[0210] Exemplarily, the first node may confirm, based on the second indication information sent by the second node, the pre-configuration information of the AI ​​model that ultimately needs to be requested from the second node.

[0211] Exemplarily, the first node may determine a new reporting period according to the second indication information of the second node. If the second node does not modify the reporting period, the first node may not report, or use 1 bit to indicate that the reporting period has not been modified.

[0212] Optionally, the method further includes step 760, in which the first node sends an AI model confirmation message to the second node, and the AI ​​model confirmation message carries fourth indication information.

[0213] Exemplarily, the AI ​​model confirmation message may be an AI model request acknowledgement.

[0214] Correspondingly, the second node receives the AI ​​model confirmation message.

[0215] Optionally, the AI ​​model confirmation message also includes a message identifier (eg, Measurement ID#1).

[0216] At 770 , the first node receives an AI model update message from the second node.

[0217] Exemplarily, the AI ​​model update message may be AI model update.

[0218] In a possible implementation, if step 750 and step 760 are not performed, the AI ​​model update message carries the encrypted second AI model, or the AI ​​model update message carries the address corresponding to the encrypted second AI model, or the AI ​​model update message carries the second AI model, or the AI ​​model update message carries the address corresponding to the second AI model.

[0219] In another possible implementation, if step 750 and step 760 are performed, the AI ​​model update message carries the encrypted third AI model, or the AI ​​model update message carries the address corresponding to the encrypted third AI model, or the AI ​​model update message carries the third AI model, or the AI ​​model update message carries the address corresponding to the third AI model.

[0220] Exemplarily, if the transmission mode of the AI ​​model is IP tunnel, the second node may only transmit the address of the AI ​​model.

[0221] Optionally, the AI ​​model update message also carries a model encryption method identifier and a message identifier (for example, Measurement ID#1).

[0222] It should be noted that the above steps 730 to 770 are mainly for the first node to request the second node for the AI ​​model required by the first node itself. Optionally, since the first node can also provide the pre-configuration information of its AI model to the second node in step 710, in subsequent steps, the second node can also request the AI ​​model required by itself from the first node. The specific implementation method can be understood with reference to steps 730 to 770. It is only necessary to regard the first node in steps 730 to 770 as the "second node" and the second node as the "first node" for understanding.

[0223] Based on the above steps 730 to 770, in the present application, the first node and the second node can first negotiate the pre-configuration information of the AI ​​model to be requested based on the pre-configuration information of the AI ​​model previously interacted, so that the second node can subsequently transmit the AI ​​model that meets the needs of the first node itself to the first node in an appropriate manner, and the privacy issues of the AI ​​model can be avoided.

[0224] In addition, the present application also provides a method 800 for transmitting training data. The description of each information in the steps of the method 800 in the present application can refer to Figure 7 The corresponding steps of method 700 in the embodiment are understood, and the differences between method 800 and method 700 are mainly described below. Figure 8 As shown, the method 800 includes:

[0225] 810. The first node sends a first request message to the second node. The first request message is used to request pre-configuration information of training data that can be provided by the second node and a manufacturer identifier corresponding to the second node. The first request message carries the manufacturer identifier corresponding to the first node.

[0226] Correspondingly, the second node receives the first request message from the first node.

[0227] In a possible implementation, the first request message also carries pre-configuration information of the training data provided by the first node, wherein the pre-configuration information of the training data includes a fourth association relationship, and the fourth association relationship is an association relationship between at least two of the following: a data identifier corresponding to the training data, a model algorithm corresponding to the training data, a model algorithm identifier corresponding to the model algorithm, a service corresponding to the training data, a service identifier corresponding to the training data service, a purpose corresponding to the training data, and a purpose identifier corresponding to the purpose. Exemplarily, the fourth association relationship includes at least one of the following association relationships: an association relationship between the data identifier corresponding to the training data, the model algorithm, and the algorithm identifier; an association relationship between the data identifier corresponding to the training data, the service, and the service identifier; an association relationship between the data identifier corresponding to the training data, the purpose, and the purpose identifier.

[0228] Optionally, the first request message also carries one or more of the following: information about the training data scrambling method supported by the first node, information about the training data updating method supported by the first node, and information about the training data transmission method supported by the first node. The information about the training data scrambling method includes: the association between the training data scrambling method identifier and the corresponding scrambling method and / or scrambling level; the information about the training data updating method includes: the association between the training data updating method identifier and the corresponding updating method; the information about the training data transmission method includes: the association between the training data transmission method identifier and the corresponding transmission method.

[0229] Among them, the "scrambling method of training data" can be understood with reference to the "scrambling method of AI model" in step 710. The "updating method of training data" may include, for example, resending the updated training data, or only needing to send the updated training data which is different from the original training data. The "transmission method of training data" may include: UP-based transmission, CP-based transmission, IP tunnel-based transmission, forwarding by 5GC, etc. Optionally, the decision conditions for the transmission of training data may also be carried, such as transmitting training data based on CP when the training data is small, transmitting training data based on UP when the training data is large, transmitting training data based on IP tunnel when the training data is not air interfaced, and obtaining training data based on 5GC forwarding when the training data is large and 5GC has the latest training data.

[0230] Specifically, the meaning of each information in 810 and the implementation method of the associated information can be understood by referring to step 710 in the above method 700, and will not be described in detail again.

[0231] 820. The second node sends a first response message to the first node. The first response message carries pre-configuration information of training data that can be provided by the second node and a manufacturer identifier corresponding to the second node.

[0232] The pre-configuration information of the training data is used by the first node to determine the first training data to be requested, and the first training data is part or all of the training data that can be provided by the second node.

[0233] Correspondingly, the first node receives a first response message from the second node.

[0234] In the present application, the pre-configuration information of the training data that the second node can provide may include a second association relationship. The second association relationship is an association relationship between at least two of the following: a data identifier corresponding to the training data, a model algorithm corresponding to the training data, a model algorithm identifier corresponding to the model algorithm, a service corresponding to the training data, a service identifier corresponding to the training data service, a purpose corresponding to the training data, and a purpose identifier corresponding to the purpose. Exemplarily, the second association relationship includes at least one of the following association relationships: an association relationship between the data identifier, the model algorithm, and the algorithm identifier corresponding to the training data; an association relationship between the data identifier, the service, and the service identifier corresponding to the training data; an association relationship between the data identifier, the purpose, and the purpose identifier corresponding to the training data.

[0235] Optionally, the first response message also carries one or more of the following: information on the training data scrambling method supported by the second node, information on the training data updating method supported by the second node, and information on the training data transmission method supported by the second node. The information on the training data scrambling method includes: the association between the training data scrambling method identifier and the corresponding scrambling method and / or scrambling level; the information on the training data updating method includes: the association between the training data updating method identifier and the corresponding updating method; the information on the training data transmission method includes: the association between the training data transmission method identifier and the corresponding transmission method.

[0236] Specifically, the meaning of each information in 820 and the implementation method of the associated information can be understood by referring to step 720 in the above method 700, and will not be described in detail.

[0237] 830. The first node sends a second request message to the second node, where the second request message is used to request the first training data. The second request message carries first indication information, where the first indication information is used to indicate pre-configuration information of the first training data.

[0238] Correspondingly, the second node receives the second request message from the first node.

[0239] Exemplarily, the second request message may be an AI data request (AI data request) message.

[0240] In a possible implementation, the pre-configuration information of the first training data includes one or more of the following: a model identifier corresponding to the first training data, a model algorithm identifier corresponding to the first training data, a service identifier corresponding to the first training data, a purpose identifier corresponding to the first training data, a reason for requesting the first training data, a model accuracy corresponding to the first training data, a training data scrambling method identifier supported by the first node, a training data update method identifier supported by the first node, and a training data transmission method identifier supported by the first node.

[0241] The “reason for requesting the first AI training data” may be, for example, that the first node requires high-precision AI training data, or that the AI ​​training data cannot be generated on the first node, etc. “Model accuracy corresponding to the first training data” indicates the accuracy requirement of the first node for the requested training data, such as training data with a model confidence (confidence) > 0.9 corresponding to the requested training data, or training data with a training step number > 1000 corresponding to the requested training data, or training data with an AI use case accuracy > 0.95 corresponding to the requested training data.

[0242] Optionally, the second request message further carries one or more of the following: a reporting period of training data, a message identifier corresponding to the second request message, and a message type corresponding to the second request message.

[0243] Among them, "the reporting period of the second response message" means that the first node instructs the second node to periodically report the updated training data to the first node. The "message identifier corresponding to the second request message" can be used to associate subsequent steps. For example, after the first node subsequently receives a response message from the second node, if the response message also carries the message identifier, the first node can determine that the response message is a reply message to the second request message. For example, the message identifier can be Measurement ID#2. In the present application, the "message identifier corresponding to the second request message" can also be replaced with the identifier corresponding to the process of steps 830 to 870.

[0244] The “message type corresponding to the second request message” may include, for example, a request message type for starting to send request training data (start type), a request message type for stopping sending request training data (stop type), a request message type for adding requested training data (add type), a request message type for deleting requested training data (del type), and a request message type for updating requested training data (revise type). Specifically, the meaning and implementation of each information in step 830 can be understood with reference to step 730 in method 700, and will not be described in detail.

[0245] 840. The first node receives a second response message from the second node, where the second response message carries second indication information, and the second indication information is used to indicate pre-configuration information of second training data.

[0246] The pre-configuration information of the second training data is information in the pre-configuration information of the first training data that the second node supports sending to the first node, and the second training data is part or all of the first training data.

[0247] Exemplarily, the second response message may be an AI data response message (AI data request response).

[0248] In one possible implementation, the pre-configuration information of the second training data includes one or more of the following: a model identifier corresponding to the second training data, a model algorithm identifier corresponding to the second training data, a service identifier corresponding to the second training data, and model accuracy feedback information corresponding to the second training data, wherein the model accuracy feedback information is used to indicate that the second node supports the model accuracy corresponding to the second training data sent to the first node.

[0249] Optionally, the second response message also carries one or more of the following: reporting cycle feedback information, a message identifier corresponding to the second response message, and failure cause indication information. Among them, the reporting cycle feedback information is used to indicate the reporting cycle of the second response message supported by the second node; the message identifier is used to associate the first request message; the failure cause indication information is used to indicate the identifier and reason why the second node does not support the first training data sent to the first node, for example, because the accuracy of the training data does not meet the requirements, or because privacy is not allowed.

[0250] Specifically, the meaning and implementation of each information in step 840 can be understood by referring to step 740 in method 700, and will not be described in detail.

[0251] Optionally, the method further includes step 850, where the first node determines fourth indication information according to the second indication information, where the fourth indication information is used to indicate pre-configuration information of the third training data.

[0252] The pre-configuration information of the third training data is information determined by the first node to request from the second node in the pre-configuration information of the second training data, and the third training data is part or all of the second training data.

[0253] Optionally, the method further includes step 860, in which the first node sends a training data confirmation message to the second node, where the training data confirmation message carries fourth indication information.

[0254] Exemplarily, the “training data confirmation message” may be an AI data request confirmation message.

[0255] Step 870: The first node receives a training data update message from the second node.

[0256] Exemplarily, the “training data update message” may be an AI data update message.

[0257] In a possible implementation, if steps 850 and 860 are not performed, the training data update message carries the scrambled second training data, or the training data update message carries the address corresponding to the scrambled second training data.

[0258] In a possible implementation, if steps 850 and 860 are performed, the training data update message carries the scrambled third training data AI model, or the training data update message carries the address corresponding to the scrambled third training data.

[0259] It should be noted that the above steps 830 to 870 are mainly for the first node to request the second node for the training data required by the first node itself. Optionally, since the first node can also provide the pre-configuration information of its training data to the second node in step 810, in subsequent steps, the second node can also request the first node for the training data required by itself. The specific implementation method can be understood with reference to steps 830 to 870. It is only necessary to regard the first node in steps 830 to 870 as the "second node" and the second node as the "first node" for understanding.

[0260] Based on the above steps 830 to 870, in the present application, the first node and the second node can first negotiate the pre-configuration information of the training data to be requested based on the pre-configuration information of the training data previously interacted, so that the second node can subsequently transmit the training data that meets the needs of the first node itself to the first node in an appropriate manner, and the privacy issues of the training data can be avoided.

[0261] It is understood that the examples in method 700 and method 800 in the embodiments of the present application are only for the convenience of those skilled in the art to understand the embodiments of the present application, and are not intended to limit the embodiments of the present application to the specific scenarios illustrated. Those skilled in the art can obviously make various equivalent modifications or changes based on the examples in method 700 and method 800, and such modifications or changes also fall within the scope of the embodiments of the present application.

[0262] It can also be understood that some optional features in the embodiments of the present application may not depend on other features in some scenarios, or may be combined with other features in some scenarios, without limitation.

[0263] It is also understood that the various embodiments described in this application may be independent solutions or may be combined according to internal logic, and these solutions all fall within the protection scope of this application. In addition, the explanations or descriptions of various terms appearing in the embodiments may be mutually referenced or explained in various embodiments, without limitation thereto.

[0264] It can be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0265] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of interaction between each node. It is understandable that each node, such as a terminal device, a network device, includes a hardware structure and / or software module corresponding to each function in order to realize the above functions. Those skilled in the art should be aware that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0266] The embodiment of the present application can divide the functional modules of the terminal device and the network device according to the above method example. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. The following is an example of dividing each functional module corresponding to each function.

[0267] Fig. 9 1 is a schematic block diagram of a communication device 900 provided in an embodiment of the present application. As shown in the figure, the device 900 may include: a transceiver unit 910 and a processing unit 920.

[0268] In one possible design, the device 900 may be the first node in the above method embodiment, or may be a chip for implementing the function of the first node in the above method embodiment. It should be understood that the device 900 may correspond to the first node in the method 700 and the method 800 according to the embodiment of the present application, and the device 900 may perform the steps corresponding to the first node in the method 700 and the method 800 according to the embodiment of the present application.

[0269] In a possible implementation manner, the transceiver unit is used to send a first request message to the second node, and the transceiver unit is used to receive a first response message from the second node.

[0270] In a possible implementation manner, the transceiver unit is used to send a second request message to the second node, and the transceiver unit is used to receive a second response message from the second node.

[0271] In one possible implementation, the processing unit is used to determine fourth indication information based on the second indication information; the transceiver unit is used to send an AI model confirmation message to the second node, and the AI ​​model confirmation message carries the fourth indication information, or the transceiver unit is used to send an AI training data confirmation message to the second node, and the training data confirmation message carries the fourth indication information.

[0272] In one possible implementation, the transceiver unit is used to receive an AI model update message or a training data update message from the second node.

[0273] In one possible design, the device 900 may be the second node in the above method embodiment, or may be a chip for implementing the function of the second node in the above method embodiment. It should be understood that the device 900 may correspond to the second node in the method 700 and the method 800 according to the embodiment of the present application, and the device 900 may perform the steps corresponding to the second node in the method 700 and the method 800 according to the embodiment of the present application.

[0274] In a possible implementation manner, the transceiver unit is used to receive a first request message from a first node; the transceiver unit is also used to send a first response message to the first node.

[0275] In a possible implementation manner, the transceiver unit is used to receive the second request message from the first node, and the transceiver unit is also used to send a second response message to the first node.

[0276] In one possible implementation, the transceiver unit is used to receive an AI model confirmation message from the first node, or a training data confirmation message.

[0277] In one possible implementation, the transceiver unit is used to send an AI model update request message, or a training data update request message, to the first node.

[0278] It should also be understood that the device 900 here is embodied in the form of a functional unit. The term "unit" here may refer to an application specific integrated circuit (ASIC), an electronic circuit, a processor (e.g., a shared processor, a proprietary processor, or a group processor, etc.) and a memory for executing one or more software or firmware programs, a merged logic circuit, and / or other suitable components that support the described functions. In an optional example, those skilled in the art will understand that the device 900 may be specifically the first node or the second node in the above-mentioned embodiment, and may be used to execute the various processes and / or steps corresponding to the first node or the second node in the above-mentioned method embodiments, and to avoid repetition, it will not be repeated here.

[0279] The device 900 of each of the above schemes has the function of implementing the corresponding steps performed by the first node or the second node in the above method. The function can be implemented by hardware, or by hardware executing the corresponding software implementation. The hardware or software includes one or more modules corresponding to the above functions; for example, the transceiver unit can be replaced by a transceiver (for example, the sending unit in the transceiver unit can be replaced by a transmitter, and the receiving unit in the transceiver unit can be replaced by a receiver), and other units, such as the processing unit, can be replaced by a processor to respectively perform the transceiver operations and related processing operations in each method embodiment.

[0280] In addition, the transceiver unit 910 may also be a transceiver circuit (for example, may include a receiving circuit and a sending circuit), and the processing unit may be a processing circuit.

[0281] It should be pointed out that Fig. 9 The device in the embodiment may be the first node or the second node in the foregoing embodiment, or may be a chip or a chip system, such as a system on chip (SoC). The transceiver unit may be an input / output circuit or a communication interface; the processing unit may be a processor or a microprocessor or an integrated circuit integrated on the chip. This is not limited here.

[0282] Fig.10 1 is a schematic block diagram of a communication device 1000 provided in an embodiment of the present application. As shown in the figure, the device 1000 includes: at least one processor 1020. The processor 1020 is coupled to the memory and is used to execute instructions stored in the memory to send signals and / or receive signals. Optionally, the device 1000 also includes a memory 1030 for storing instructions. Optionally, the device 1000 also includes a transceiver 1010, and the processor 1020 controls the transceiver 1010 to send signals and / or receive signals.

[0283] It should be understood that the processor 1020 and the memory 1030 may be combined into one processing device, and the processor 1020 is used to execute the program code stored in the memory 1030 to implement the above functions. In specific implementation, the memory 1030 may also be integrated into the processor 1020, or independent of the processor 1020.

[0284] It should also be understood that the transceiver 1010 may include a transceiver (or receiver) and a transmitter (or transmitter). The transceiver may further include an antenna, and the number of antennas may be one or more. The transceiver 1010 may also be a communication interface or an interface circuit.

[0285] Specifically, the transceiver 1010 in the device 1000 may correspond to the transceiver unit 110 in the device 1000 , and the processor 220 in the device 1000 may correspond to the processing unit 120 in the device 1000 .

[0286] As a solution, the device 1000 is used to implement the operations performed by the first node in each of the above method embodiments.

[0287] For example, the processor 1020 is used to execute the computer program or instructions stored in the memory 1030 to implement the relevant operations of the first node in each method embodiment above. For example, the method performed by the first node in any of the embodiments shown in method 700 and method 800.

[0288] As another solution, the device 1000 is used to implement the operations performed by the second node in each of the above method embodiments. For example, the processor 1020 is used to execute the computer program or instructions stored in the memory 1030 to implement the relevant operations of the second node in each of the above method embodiments. For example, the method performed by the second node in any of the embodiments shown in method 700 and method 800.

[0289] It should be understood that the specific process of each transceiver and processor executing the above corresponding steps has been described in detail in the above method embodiment, and for the sake of brevity, it will not be repeated here.

[0290] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0291] It should be noted that the processor in the embodiment of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by the hardware integrated logic circuit or software instructions in the processor. The above processor can be a general processor, a digital signal processor (digitalsignal processor, DSP), an application-specific integrated circuit (application-specific integrated circuit, ASIC), a field-programmable gate array (field-programmable gate array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor can be combined and performed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0292] It can be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0293] According to the method provided in the embodiments of the present application, the present application also provides a computer program product having a computer program code stored thereon. When the computer program code runs on a computer, the computer executes the method executed by a terminal device or a network device in any one of the embodiments of method 700 and method 800.

[0294] According to the method provided in the embodiments of the present application, the present application also provides a computer-readable medium, which stores a program code. When the program code runs on a computer, the computer executes the method performed by the first node or the second node in the above embodiment.

[0295] According to the method provided in the embodiment of the present application, the present application also provides a communication system, which includes a first node and a second node. The terminal device is used to execute the steps corresponding to the first node in the above method 700 and method 800, and the second node is used to execute the steps corresponding to the second node in the above method 700 and method 800.

[0296] The explanation of the relevant contents and beneficial effects of any of the above-mentioned devices can be referred to the corresponding method embodiments provided above, which will not be repeated here.

[0297] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disc (SSD)).

[0298] In the above-mentioned various device embodiments, the corresponding steps are performed by the corresponding modules or units. For example, the transceiver unit (transceiver) performs the steps of receiving or sending in the method embodiment, and other steps except sending and receiving can be performed by the processing unit (processor). The functions of the specific units can refer to the corresponding method embodiments. Among them, there can be one or more processors.

[0299] The terms "component", "module", "system", etc. used in this specification are used to represent computer-related entities, hardware, firmware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program and / or a computer. By way of illustration, both applications running on a computing device and a computing device can be components. One or more components may reside in a process and / or an execution thread, and a component may be located on a computer and / or distributed between two or more computers. In addition, these components may be executed from various computer-readable media having various data structures stored thereon. Components may, for example, communicate through local and / or remote processes according to signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system and / or a network, such as the Internet interacting with other systems through signals).

[0300] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0301] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0302] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0303] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0304] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0305] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can essentially or in other words, the part that contributes or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0306] It should be understood that the "embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, the various embodiments in the entire specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner.

[0307] It should also be understood that the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish multiple objects, and are not used to limit the size, content, order, timing, priority or importance of the multiple objects. For example, the first PDSCH and the second PDSCH can be the same physical channel or different physical channels, and such a name does not indicate the difference in the amount of information, content, priority or importance of the two physical channels.

[0308] It should also be understood that in this application, "at least one" means one or more, and "more than one" means two or more. "At least one item" or similar expressions means one or more items, that is, any combination of these items, including any combination of single items or plural items. For example, at least one item of a, b, or c means: a, b, c, a and b, a and c, b and c, or a and b and c.

[0309] It should also be understood that in each embodiment of the present application, "A corresponds to B" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.

[0310] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for transmitting an artificial intelligence model, characterized in that: include: The first node sends a first request message to the second node, where the first request message is used to request pre-configuration information of an artificial intelligence AI model that can be provided by the second node and a manufacturer identifier corresponding to the second node, and the first request message carries the manufacturer identifier corresponding to the first node; The first node receives a first response message from the second node, the first response message carrying pre-configuration information of the AI ​​model that can be provided by the second node and the manufacturer identifier corresponding to the second node, the pre-configuration information of the AI ​​model includes a first association relationship, the first association relationship is an association relationship between at least two of the following: a model identifier corresponding to the AI ​​model, a model algorithm corresponding to the AI ​​model, a model algorithm identifier corresponding to the AI ​​model algorithm, a business corresponding to the AI ​​model, a business identifier corresponding to the AI ​​model business, a purpose corresponding to the AI ​​model, a purpose identifier corresponding to the purpose, reasoning information corresponding to the AI ​​model, and model information corresponding to the AI ​​model, the first association relationship is used by the first node to determine a first AI model to be requested, wherein the first AI model is one or more of the AI ​​models that can be provided by the second node.

2. The method according to claim 1, characterized in that The pre-configuration information of the AI ​​model includes one or more of the following: Information about the AI ​​model scrambling method supported by the second node, wherein the information about the AI ​​model scrambling method includes: an association relationship between the AI ​​model scrambling method identifier and a corresponding scrambling method and / or scrambling level; Information about the AI ​​model update method supported by the second node, wherein the information about the model update method includes: an association relationship between an update method identifier of the AI ​​model and a corresponding update method; Information on the AI ​​model transmission mode supported by the second node, wherein the information on the AI ​​model transmission mode includes: an association relationship between the transmission mode identifier of the AI ​​model and the corresponding transmission mode.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: The first node sends a second request message to the second node, where the second request message is used to request the first AI model, and the second request message carries first indication information, where the first indication information is used to indicate pre-configuration information of the first AI model; The first node receives a second response message from the second node, where the second response message carries second indication information, where the second indication information is used to indicate pre-configuration information of a second AI model, wherein the pre-configuration information of the second AI model is information in the pre-configuration information of the first AI model that the second node supports sending to the first node, and the second AI model is one or more AI models in the first AI model.

4. The method according to claim 3, characterized in that The first request message also carries pre-configuration information of the AI ​​model provided by the first node to the second node, and the pre-configuration information of the AI ​​model provided by the first node to the second node includes one or more of the following: Information about the AI ​​model scrambling method supported by the first node, wherein the information about the AI ​​model scrambling method includes: an association relationship between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; Information about the AI ​​model update method supported by the first node, wherein the information about the AI ​​model update method includes: an association relationship between an update method identifier of the AI ​​model and a corresponding update method; Information about the AI ​​model transmission mode supported by the first node, wherein the information about the AI ​​model transmission mode includes: an association relationship between the transmission mode identifier of the AI ​​model and the corresponding transmission mode.

5. The method according to claim 4, characterized in that The pre-configuration information of the first AI model includes one or more of the following: a model identifier corresponding to the first AI model, model information corresponding to the first AI model, a model algorithm identifier corresponding to the first AI model, a service identifier corresponding to the first AI model, a usage identifier corresponding to the first AI model, an AI model scrambling method identifier supported by the first node, an AI model update method identifier supported by the first node, an AI model transmission method identifier supported by the first node, a reason for requesting the first AI model, and a model accuracy corresponding to the first AI model.

6. The method according to any one of claims 3 to 5, characterized in that The second request message further carries one or more of the following: The period for the second node to report the AI ​​model; a message identifier corresponding to the second request message, wherein the message identifier is used to associate with the second response message; The message type corresponding to the second request message.

7. The method according to any one of claims 3 to 6, characterized in that The pre-configuration information of the second AI model includes one or more of the following: the model identifier corresponding to the second AI model, the model information corresponding to the second AI model, the model algorithm identifier corresponding to the second AI model, the service identifier corresponding to the second AI model, and the model accuracy feedback information corresponding to the second AI model, wherein the model accuracy feedback information is used to indicate that the second node supports the model accuracy corresponding to the second AI model sent to the first node.

8. The method according to any one of claims 3 to 7, characterized in that The second response message further carries one or more of the following: Reporting cycle feedback information, where the reporting cycle feedback information is used to indicate an AI model reporting cycle supported by the second node; the message identifier; Failure reason indication information, where the failure reason indication information is used to indicate an identifier and a reason why the second node does not support the first AI model sent to the first node.

9. The method according to any one of claims 3 to 8, characterized in that The method further comprises: The first node receives an AI model update message from the second node, where the AI ​​model update message carries the encrypted second AI model, or the AI ​​model update message carries an address corresponding to the encrypted second AI model, or the AI ​​model update message carries the second AI model, or the AI ​​model update message carries the address corresponding to the second AI model.

10. The method according to any one of claims 3 to 8, characterized in that The method further comprises: The first node determines fourth indication information according to the second indication information, where the fourth indication information is used to indicate pre-configuration information of a third AI model, wherein the pre-configuration information of the third AI model is information determined by the first node to request from the second node in the pre-configuration information of the second AI model, and the third AI model is one or more AI models in the second AI model; The first node sends an AI model confirmation message to the second node, where the AI ​​model confirmation message carries the fourth indication information.

11. The method according to claim 10, characterized in that The method further comprises: The first node receives an AI model update message from the second node, where the AI ​​model update message carries the encrypted third AI model, or the AI ​​model update message carries an address corresponding to the encrypted third AI model, or the AI ​​model update message carries the third AI model, or the AI ​​model update message carries the address corresponding to the third AI model.

12. A method for transmitting an artificial intelligence model, characterized in that: include: The second node receives a first request message from the first node, where the first request message is used to request pre-configuration information of an artificial intelligence AI model that can be provided by the second node and a manufacturer identifier corresponding to the second node, and the first request message carries the manufacturer identifier corresponding to the first node; The second node sends a first response message to the first node, the first response message carrying pre-configuration information of the AI ​​model that can be provided by the second node and the manufacturer identifier corresponding to the second node, the pre-configuration information of the AI ​​model includes a first association relationship, the first association relationship is an association relationship between at least two of the following: a model identifier corresponding to the AI ​​model, a model algorithm corresponding to the AI ​​model, a model algorithm identifier corresponding to the AI ​​model algorithm, a business corresponding to the AI ​​model, a business identifier corresponding to the AI ​​model business, a purpose corresponding to the AI ​​model, a purpose identifier corresponding to the purpose, reasoning information corresponding to the AI ​​model, and model information corresponding to the AI ​​model, the first association relationship is used by the first node to determine the first AI model to be requested, wherein the first AI model is one or more of the AI ​​models that can be provided by the second node.

13. The method according to claim 12, characterized in that The pre-configuration information of the AI ​​model includes one or more of the following: Information about the AI ​​model scrambling method supported by the second node, wherein the information about the AI ​​model scrambling method includes: an association relationship between the AI ​​model scrambling method identifier and a corresponding scrambling method and / or scrambling level; Information about the AI ​​model update method supported by the second node, wherein the information about the model update method includes: an association relationship between an update method identifier of the AI ​​model and a corresponding update method; Information on the AI ​​model transmission mode supported by the second node, wherein the information on the AI ​​model transmission mode includes: an association relationship between the transmission mode identifier of the AI ​​model and the corresponding transmission mode.

14. The method according to claim 12 or 13, characterized in that The method further comprises: The second node receives a second request message from the first node, where the second request message is used to request the first AI model, and the second request message carries first indication information, where the first indication information is used to indicate pre-configuration information of the first AI model; The second node sends a second response message to the first node, the second response message carrying second indication information, the second indication information being used to indicate pre-configuration information of a second AI model, wherein the pre-configuration information of the second AI model is information in the pre-configuration information of the first AI model that the second node supports sending to the first node, and the second AI model is one or more AI models in the first AI model.

15. The method according to claim 14, characterized in that The first request message also carries pre-configuration information of the AI ​​model provided by the first node to the second node, and the pre-configuration information of the AI ​​model provided by the first node to the second node includes one or more of the following: Information about the AI ​​model scrambling method supported by the first node, wherein the information about the AI ​​model scrambling method includes: an association relationship between the AI ​​model scrambling method identifier and the corresponding scrambling method and / or scrambling level; Information about the AI ​​model update method supported by the first node, wherein the information about the AI ​​model update method includes: an association relationship between an update method identifier of the AI ​​model and a corresponding update method; Information about the AI ​​model transmission mode supported by the first node, wherein the information about the AI ​​model transmission mode includes: an association relationship between the transmission mode identifier of the AI ​​model and the corresponding transmission mode.

16. The method according to claim 15, characterized in that The pre-configuration information of the first AI model includes one or more of the following: a model identifier corresponding to the first AI model, model information corresponding to the first AI model, a model algorithm identifier corresponding to the first AI model, a service identifier corresponding to the first AI model, a usage identifier corresponding to the first AI model, an AI model scrambling method identifier supported by the first node, an AI model update method identifier supported by the first node, an AI model transmission method identifier supported by the first node, a reason for requesting the first AI model, and a model accuracy corresponding to the first AI model.

17. The method according to any one of claims 14 to 16, characterized in that The second request message further carries one or more of the following: The period for the second node to report the AI ​​model; a message identifier corresponding to the second request message, wherein the message identifier is used to associate with the second response message; The message type corresponding to the second request message.

18. The method according to any one of claims 14 to 17, characterized in that The pre-configuration information of the second AI model includes one or more of the following: the model identifier corresponding to the second AI model, the model information corresponding to the second AI model, the model algorithm identifier corresponding to the second AI model, the service identifier corresponding to the second AI model, and the model accuracy feedback information corresponding to the second AI model, wherein the model accuracy feedback information is used to indicate that the second node supports the model accuracy corresponding to the second AI model sent to the first node.

19. The method according to any one of claims 14 to 18, characterized in that The second response message further carries one or more of the following: Reporting cycle feedback information, where the reporting cycle feedback information is used to indicate an AI model reporting cycle supported by the second node; the message identifier; Failure reason indication information, where the failure reason indication information is used to indicate an identifier and a reason why the second node does not support the first AI model sent to the first node.

20. The method according to any one of claims 14 to 19, characterized in that The method further comprises: The second node sends an AI model update message to the first node, where the AI ​​model update message carries the encrypted second AI model, or the AI ​​model update message carries the encrypted address corresponding to the second AI model, or the AI ​​model update message carries the second AI model, or the AI ​​model update message carries the address corresponding to the second AI model.

21. The method according to any one of claims 14 to 19, characterized in that The method further comprises: The second node receives an AI model confirmation message from the first node, where the AI ​​model confirmation message carries the fourth indication information, where the fourth indication information is used to indicate pre-configuration information of a third AI model, wherein the pre-configuration information of the third AI model is information determined by the first node to request from the second node in the pre-configuration information of the second AI model, and the third AI model is one or more AI models in the second AI model.

22. The method according to claim 21, characterized in that The method further comprises: The AI ​​of the second node receives a model update message from the first node, where the AI ​​model update message carries the encrypted third AI model, or the AI ​​model update message carries the encrypted address corresponding to the third AI model, or the AI ​​model update message carries the third AI model, or the AI ​​model update message carries the address corresponding to the third AI model.

23. A communication device, characterized in that: Used to implement the method according to any one of claims 1 to 11.

24. A communication device, characterized in that: The communication device comprises a processor and a memory, wherein the memory is used to store a computer program or instructions, and the processor is used to execute the computer program or instructions in the memory, so that the method according to any one of claims 1 to 11 is executed.

25. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 11.

26. A computer program product, characterized in that The computer program product comprises means for executing the method according to any one of claims 1 to 11.

Images