Communication method, communication device, communication system, storage medium and program product

Abstract

The embodiments of the present disclosure relate to a communication method, a communication device, a communication system, a storage medium and a program product. The communication method may be applied to a first node. The method comprises: receiving first information from a second node, the first information being used for requesting a first AI inference service; on the basis of the first information, determining a third node, the third node being used for providing the first AI inference service; sending second information, the second information being used for instructing the third node to provide the first AI inference service; and receiving an inference result associated with the first AI inference service and sent by the third node. In this way, in the embodiments of the present disclosure, since the third node is determined on the basis of the first information, that is, the third node is matched with a requirement of the first information, the third node providing the AI inference service and sending the inference result can reduce not only the demands of the AI inference service for the computing capability and resources of a terminal, but also the delay of providing the inference service for the terminal.

Description

Communication method, communication device, communication system, storage medium and program product TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of communication, and in particular, to a communication method, a communication device, a communication system, a storage medium and a program product. BACKGROUND

[0002] With the progress of communication technology, artificial intelligence (AI) models, machine learning (ML) models or other models are introduced in communication systems. Through these models, data reasoning or prediction can be performed in scenarios such as autonomous driving, medical assistance, cross-device computing, digital twinning, etc.

[0003] SUMMARY

[0004] In the process of model reasoning, on the one hand, due to the limited computing capability of the terminal, the terminal cannot support reasoning services that exceed its own computing capability; on the other hand, although a third-party server can be used to provide reasoning services for the terminal, this will also bring additional delay, affecting the user experience.

[0005] Embodiments of the present disclosure provide a communication method, a communication device, a communication system, a storage medium and a program product, which can reduce the demand of AI reasoning service on terminal computing capability and resources, and reduce the delay of providing reasoning service for the terminal.

[0006] According to a first aspect of embodiments of the present disclosure, a communication method is provided, which is performed by a first node, and the method comprises: receiving first information from a second node, the first information being used to request a first AI reasoning service; determining a third node based on the first information, the third node being used to provide the first AI reasoning service; sending second information, the second information being used to instruct the third node to provide the first AI reasoning service; receiving a reasoning result associated with the first AI reasoning service sent by the third node.

[0007] According to a second aspect of embodiments of the present disclosure, a communication method is provided, which is performed by a second node, and the method comprises: sending first information to a first node, the first information being used to request a first AI reasoning service, the first AI reasoning service being provided by a third node, the third node being determined by the first node according to the first information.

[0008] According to a third aspect of embodiments of the present disclosure, a communication method is provided, which is performed by a third node, and the method comprises: receiving second information sent by a first node, the second information being used to instruct the third node to provide a first AI reasoning service; sending a reasoning result associated with the first AI reasoning service to the first node.

[0009] According to a fourth aspect of embodiments of the present disclosure, a first node is provided, comprising: a first transceiver configured to receive first information from a second node, the first information being used to request a first AI inference service; a first processing module configured to determine, based on the first information, a third node, the third node being used to provide the first AI inference service; the first transceiver configured to send second information, the second information being used to instruct the third node to provide the first AI inference service; and the first transceiver configured to receive an inference result associated with the first AI inference service sent by the third node.

[0010] According to a fifth aspect of embodiments of the present disclosure, a second node is provided, comprising: a second transceiver configured to send first information to a first node, the first information being used to request a first AI inference service, the first AI inference service being provided by a third node, the third node being determined by the first node according to the first information.

[0011] According to a sixth aspect of embodiments of the present disclosure, a third node is provided, comprising: a third transceiver configured to receive second information sent by a first node, the second information being used to instruct the third node to provide a first AI inference service; and the third transceiver configured to send an inference result associated with the first AI inference service to the first node.

[0012] According to a seventh aspect of embodiments of the present disclosure, a communication device is provided, comprising: one or more processors; and wherein the communication device is configured to perform the communication method according to any one of the first aspect to the third aspect.

[0013] According to an eighth aspect of embodiments of the present disclosure, a communication system is provided, comprising a first node, a second node and a third node; the first node is configured to implement the communication method according to the first aspect; the second node is configured to implement the communication method according to the second aspect; and the third node is configured to implement the communication method according to the third aspect.

[0014] According to a ninth aspect of embodiments of the present disclosure, a storage medium is provided, the storage medium storing instructions, when the instructions are executed on a communication device, causing the communication device to perform the communication method according to any one of the first aspect to the third aspect.

[0015] According to a tenth aspect of embodiments of the present disclosure, a computer program product is provided, comprising a computer program, when the computer program is executed on a communication device, causing the communication device to perform the communication method according to any one of the first aspect to the third aspect.

[0016] According to an eleventh aspect of embodiments of the present disclosure, a computer program is provided, the computer program comprising code, when the code is executed by a processor, implementing the communication method according to any one of the first aspect to the third aspect.

[0017] According to a twelfth aspect of the embodiments of the present disclosure, a chip or chip system is provided, which includes processing circuitry configured to perform the communication method according to any one of the first aspect to the third aspect.

[0018] According to the embodiments of the present disclosure, the first node determines the third node for providing the first AI inference service according to the first information for requesting the first AI inference service from the second node, then sends the second information to indicate the third node to provide the first AI inference service, and finally receives the inference result associated with the first AI inference service sent by the third node. Since the third node is determined based on the first information, that is, the third node matches the requirements of the first information, the third node provides the AI inference service and sends the inference result, which can not only reduce the demand of the AI inference service on the terminal computing capability and resources, but also reduce the delay of providing the inference service for the terminal.

[0019] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not constitute a limitation on the embodiments of the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following describes the drawings required for the embodiment description. The following drawings are only some embodiments of the present disclosure, and do not specifically limit the protection scope of the present disclosure.

[0021] FIG. 1A is a schematic diagram of an architecture of a communication system according to an embodiment of the present disclosure.

[0022] FIG. 1B is a schematic diagram of a service architecture of an AI inference service facing 6G according to an embodiment of the present disclosure.

[0023] FIG. 1C is a schematic diagram of a data connection structure between an AiISF and a UE and an inference node according to an embodiment of the present disclosure.

[0024] FIG. 2 is a schematic diagram of an interaction of a communication method according to an embodiment of the present disclosure.

[0025] FIG. 3A is a schematic diagram of a flow of an AiISF performing a communication method according to an embodiment of the present disclosure.

[0026] FIG. 3B is a schematic diagram of a flow of a terminal A performing a communication method according to an embodiment of the present disclosure.

[0027] FIG. 3C is a schematic diagram of a flow of an inference node A performing a communication method according to an embodiment of the present disclosure.

[0028] FIG. 4A is a schematic diagram of a flow of an AiISF performing a communication method according to an embodiment of the present disclosure.

[0029] FIG. 4B is a flow diagram illustrating a method for performing communication by a terminal A, according to an embodiment of the present disclosure.

[0030] FIG. 4C is a flow diagram illustrating a method for performing communication by an inference node A, according to an embodiment of the present disclosure.

[0031] FIG. 5 is a flow diagram illustrating an exemplary embodiment of a method for performing communication, according to an embodiment of the present disclosure.

[0032] FIG. 6A is a structural diagram illustrating a first node, according to an embodiment of the present disclosure.

[0033] FIG. 6B is a structural diagram illustrating a second node, according to an embodiment of the present disclosure.

[0034] FIG. 6C is a structural diagram illustrating a third node, according to an embodiment of the present disclosure.

[0035] FIG. 7A is a structural diagram illustrating a communication device, according to an embodiment of the present disclosure.

[0036] FIG. 7B is a structural diagram illustrating a chip, according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0037] The embodiments of the present disclosure provide a communication method, a communication device, a communication system, a storage medium and a program product.

[0038] In a first aspect, the embodiments of the present disclosure provide a communication method performed by a first node, the method comprising: receiving first information from a second node, the first information being used to request a first AI inference service; determining a third node based on the first information, the third node being used to provide the first AI inference service; sending second information, the second information being used to instruct the third node to provide the first AI inference service; and receiving an inference result associated with the first AI inference service sent by the third node.

[0039] In the embodiments of the present disclosure, since the third node is determined based on the first information, i.e., the third node matches the requirements of the first information, the third node provides the AI inference service and sends the inference result, which can not only reduce the demand of the AI inference service on the terminal computing capability and resources, but also reduce the delay of providing the inference service for the terminal.

[0040] In some embodiments, there are multiple third nodes, and the method further comprises: merging the inference results sent by the multiple third nodes to obtain a first inference result of the first AI inference service; and sending the first inference result to the second node.

[0041] In some embodiments, the first information comprises at least one of: first indication information indicating a second node requesting the first AI inference service; second indication information indicating a quality requirement of the first AI inference service; third indication information indicating a model requirement of the first AI inference service; fourth indication information indicating a type of the first AI inference service; fifth indication information indicating a location of the second node; and sixth indication information indicating a service area of the first AI inference service.

[0042] In some embodiments, determining the third node based on the first information comprises: determining the third node from a plurality of fourth nodes according to the first information and third information associated with the plurality of fourth nodes.

[0043] In some embodiments, the third information comprises at least one of: seventh indication information indicating a support capability of the fourth node for different types of AI inference services; eighth indication information indicating a support capability of the fourth node for AI models; ninth indication information indicating resources of the fourth node for processing AI inference services; and tenth indication information indicating a location of the fourth node.

[0044] In some embodiments, determining the third node based on the first information comprises: determining the third node according to the first information and preconfigured eleventh indication information, wherein the eleventh indication information indicates a mapping relationship between a type of the AI inference service and the third node.

[0045] In some embodiments, the method further comprises one of: determining, according to the first information, a first inference method adopted by the first AI inference service; and determining, according to the first information and preconfigured twelfth indication information, the first inference method adopted by the first AI inference service, wherein the twelfth indication information indicates a mapping relationship between a type of the AI inference service and the inference method.

[0046] In some embodiments, the second information further indicates that the third node provides the first AI inference service by using the first inference method.

[0047] In some embodiments, the method further comprises: receiving fourth information from the second node, the fourth information indicating inference data associated with the first AI inference service; and sending the inference data to the third node.

[0048] In some embodiments, the third node is a plurality of nodes, and the method further comprises: splitting the inference data to obtain a plurality of split data; and sending the plurality of split data to the plurality of third nodes respectively, the split data being used by the third node to perform an inference task associated with the first AI inference service.

[0049] In some embodiments, the number of third nodes is a plurality, and the method further includes: determining a plurality of inference tasks associated with the first AI inference service; and assigning the plurality of inference tasks to the plurality of third nodes.

[0050] In some embodiments, the second information includes at least one of the following: the first information; thirteenth indication information indicating the first AI inference service; fourteenth indication information indicating a plurality of inference tasks associated with the first AI inference service; and fifteenth indication information indicating a first inference method adopted by the first AI inference service.

[0051] In some embodiments, the second information is further used to request the establishment of a data connection between the third node and the first node.

[0052] In some embodiments, the method further includes: receiving fifth information sent by the third node, the fifth information being used to indicate an access address of the third node; and sending sixth information to the second node, the sixth information including the fifth information.

[0053] In some embodiments, the first node and the third node are deployed on the same device.

[0054] In a second aspect, the embodiments of the present disclosure provide a communication method, executed by a second node, the method including: sending first information to a first node, the first information being used to request a first AI inference service, the first AI inference service being provided by a third node, the third node being determined by the first node according to the first information.

[0055] In the embodiments of the present disclosure, the second node sends the first information to the first node, so that the first node determines the third node according to the first information. Since the third node is determined based on the first information, i.e., the third node matches the requirements of the first information, the third node provides the AI inference service and sends the inference result, which can not only reduce the demand of the AI inference service on the terminal computing capability and resources, but also reduce the delay of providing the inference service for the terminal.

[0056] In some embodiments, the number of third nodes is a plurality, and the method further includes: receiving first inference result sent by the first node, the first inference result being obtained by the first node merging inference results sent by the plurality of third nodes.

[0057] In some embodiments, the first information includes at least one of the following: first indication information indicating the second node requesting the first AI inference service; second indication information indicating a quality requirement of the first AI inference service; third indication information indicating a model requirement of the first AI inference service; fourth indication information indicating a type of the first AI inference service; fifth indication information indicating a location of the second node; and sixth indication information indicating a service area of the first AI inference service.

[0058] In some embodiments, the third node is determined by the first node from the plurality of fourth nodes according to the first information and third information associated with the plurality of fourth nodes.

[0059] In some embodiments, the third information includes at least one of: seventh indication information indicating support capabilities of the fourth node for different types of AI inference services; eighth indication information indicating support capabilities of the fourth node for AI models; ninth indication information indicating resources of the fourth node for processing AI inference services; and tenth indication information indicating a location of the fourth node.

[0060] In some embodiments, the third node is determined by the first node according to the first information and preconfigured eleventh indication information; and the eleventh indication information indicates a mapping relationship between types of AI inference services and the third node.

[0061] In some embodiments, the first inference method is obtained according to the first information; or the first inference method is obtained according to the first information and preconfigured twelfth indication information, and the twelfth indication information indicates a mapping relationship between types of AI inference services and inference methods.

[0062] In some embodiments, the first inference method is an inference method used by the third node to provide the first AI inference service.

[0063] In some embodiments, the method further includes: sending, to the first node, fourth information indicating inference data associated with the first AI inference service.

[0064] In some embodiments, the method further includes: receiving sixth information sent by the first node, the sixth information including fifth information received by the first node from the third node, and the fifth information indicating an access address of the third node.

[0065] In some embodiments, the first node and the third node are deployed on the same device.

[0066] In a third aspect, the embodiments of the present disclosure provide a communication method, executed by a third node, the method including: receiving second information sent by a first node, the second information indicating that the third node provides a first AI inference service; and sending, to the first node, inference results associated with the first AI inference service.

[0067] In the embodiments of the present disclosure, the third node provides the first AI inference service according to the second information sent by the first node, and sends the inference results associated with the first AI inference service to the first node, which can reduce the demand of AI inference services on terminal computing capability and resources, and can also reduce the delay of providing inference services for terminals.

[0068] In some embodiments, the second information is further used to instruct the third node to provide the first AI inference service by using the first inference method.

[0069] In some embodiments, the first inference method is obtained by the first node according to the first information; or, the first inference method is obtained by the first node according to the first information and preconfigured twelfth instruction information, the first information is information received by the first node from the second node, the first information is used to request the first AI inference service, and the twelfth instruction information is used to indicate a mapping relationship between a type of AI inference service and an inference method.

[0070] In some embodiments, the method further includes receiving inference data sent by the first node, the inference data being indicated by fourth information received by the first node from the second node, the second node being a node requesting the first AI inference service.

[0071] In some embodiments, the third node is multiple, and the method further includes receiving split data sent by the first node, the split data being obtained by the first node by splitting the inference data.

[0072] In some embodiments, the method further includes receiving inference tasks sent by the first node, the inference tasks being obtained by the first node by distributing a plurality of inference tasks associated with the first AI inference service.

[0073] In some embodiments, the second information includes at least one of the following: the first information; thirteenth instruction information used to indicate the first AI inference service; fourteenth instruction information used to indicate a plurality of inference tasks associated with the first AI inference service; and fifteenth instruction information used to indicate the first inference method used by the first AI inference service.

[0074] In some embodiments, the second information is further used to request the third node to establish a data connection with the first node.

[0075] In some embodiments, the method further includes sending fifth information to the first node, the fifth information being used to indicate an access address of the third node.

[0076] In some embodiments, the first node and the third node are deployed on the same device.

[0077] In a fourth aspect, the embodiments of the present disclosure provide a first node, comprising: a first receiving module configured to receive first information from a second node, the first information being used to request a first AI inference service; a first processing module configured to determine a third node based on the first information, the third node being used to provide the first AI inference service; the first receiving module is configured to send second information, the second information being used to instruct the third node to provide the first AI inference service; and the first receiving module is configured to receive an inference result associated with the first AI inference service sent by the third node.

[0078] In some embodiments, the third node is multiple, and the first processing module is further configured to: combine the inference results sent by the multiple third nodes to obtain a first inference result of the first AI inference service; and the first receiving module is further configured to: send the first inference result to the second node.

[0079] In some embodiments, the first information comprises at least one of: first indication information used to indicate the second node requesting the first AI inference service; second indication information used to indicate quality requirements of the first AI inference service; third indication information used to indicate model requirements of the first AI inference service; fourth indication information used to indicate a type of the first AI inference service; fifth indication information used to indicate a location of the second node; and sixth indication information used to indicate a service area of the first AI inference service.

[0080] In some embodiments, the first processing module is configured to: determine the third node from a plurality of fourth nodes according to the first information and third information associated with the plurality of fourth nodes.

[0081] In some embodiments, the third information comprises at least one of: seventh indication information used to indicate support capabilities of the fourth node for different types of AI inference services; eighth indication information used to indicate support capabilities of the fourth node for AI models; ninth indication information used to indicate resources of the fourth node for processing AI inference services; and tenth indication information used to indicate a location of the fourth node.

[0082] In some embodiments, the first processing module is configured to: determine the third node according to the first information and preconfigured eleventh indication information; and the eleventh indication information is used to indicate a mapping relationship between a type of the AI inference service and the third node.

[0083] In some embodiments, the first processing module is further configured to perform one of: determine a first inference method used by the first AI inference service according to the first information; and determine the first inference method used by the first AI inference service according to the first information and preconfigured twelfth indication information; and the twelfth indication information is used to indicate a mapping relationship between a type of the AI inference service and the inference method.

[0084] In some embodiments, the second information is further used to instruct the third node to provide the first AI inference service by using the first inference method.

[0085] In some embodiments, the first transceiver module is further configured to receive fourth information from the second node, the fourth information being used to indicate inference data associated with the first AI inference service; and the first transceiver module is further configured to send the inference data to the third node.

[0086] In some embodiments, there are multiple third nodes, and the first processing module is further configured to split the inference data to obtain multiple split data; and the first transceiver module is further configured to send the multiple split data to the multiple third nodes respectively, the split data being used by the third nodes to perform inference tasks associated with the first AI inference service.

[0087] In some embodiments, there are multiple third nodes, and the first processing module is further configured to determine multiple inference tasks associated with the first AI inference service; and the first transceiver module is further configured to assign the multiple inference tasks to the multiple third nodes.

[0088] In some embodiments, the second information includes at least one of the following: the first information; thirteenth indication information used to indicate the first AI inference service; fourteenth indication information used to indicate the multiple inference tasks associated with the first AI inference service; and fifteenth indication information used to indicate the first inference method used by the first AI inference service.

[0089] In some embodiments, the second information is further used to request the third node to establish a data connection with the first node.

[0090] In some embodiments, the first transceiver module is further configured to receive fifth information sent by the third node, the fifth information being used to indicate an access address of the third node; and the first transceiver module is further configured to send sixth information to the second node, the sixth information including the fifth information.

[0091] In some embodiments, the first node and the third node are deployed on the same device.

[0092] In some embodiments, there are multiple third nodes, and the second transceiver module is configured to receive first inference results sent by the first node, the first inference results being obtained by the first node by merging inference results sent by the multiple third nodes.

[0093] In some embodiments, there are multiple third nodes, and the second transceiver module is configured to receive first inference results sent by the first node, the first inference results being obtained by the first node by merging inference results sent by the multiple third nodes.

[0094] In some embodiments, the first information comprises at least one of: first indication information for indicating a second node requesting the first AI inference service; second indication information for indicating a quality requirement of the first AI inference service; third indication information for indicating a model requirement of the first AI inference service; fourth indication information for indicating a type of the first AI inference service; fifth indication information for indicating a location of the second node; and sixth indication information for indicating a service area of the first AI inference service.

[0095] In some embodiments, the third node is determined by the first node from a plurality of fourth nodes according to the first information and third information associated with the plurality of fourth nodes.

[0096] In some embodiments, the third information comprises at least one of: seventh indication information for indicating a support capability of the fourth node for different types of AI inference services; eighth indication information for indicating a support capability of the fourth node for AI models; ninth indication information for indicating resources of the fourth node for processing AI inference services; and tenth indication information for indicating a location of the fourth node.

[0097] In some embodiments, the third node is determined by the first node according to the first information and preconfigured eleventh indication information; and the eleventh indication information is used to indicate a mapping relationship between a type of the AI inference service and the third node.

[0098] In some embodiments, the first inference method is obtained according to the first information; or the first inference method is obtained according to the first information and preconfigured twelfth indication information, and the twelfth indication information is used to indicate a mapping relationship between a type of the AI inference service and the inference method.

[0099] In some embodiments, the first inference method is an inference method adopted by the third node when providing the first AI inference service.

[0100] In some embodiments, the second transceiver is further configured to: send, to the first node, fourth information used to indicate inference data associated with the first AI inference service.

[0101] In some embodiments, the second transceiver is further configured to: receive sixth information sent by the first node, the sixth information comprising fifth information received by the first node from the third node, and the fifth information being used to indicate an access address of the third node.

[0102] In some embodiments, the first node and the third node are deployed on a same device.

[0103] In a sixth aspect, the embodiments of the present disclosure provide a third node, comprising: a third transceiver configured to receive second information sent by a first node, the second information being used to indicate that the third node provides a first AI inference service; and the third transceiver is configured to send an inference result associated with the first AI inference service to the first node.

[0104] In some embodiments, the second information is further used to indicate that the third node provides the first AI inference service by using a first inference method.

[0105] In some embodiments, the first inference method is obtained by the first node according to the first information; or, the first inference method is obtained by the first node according to the first information and preconfigured twelfth indication information, the first information being information received by the first node from a second node, the first information being used to request the first AI inference service, and the twelfth indication information being used to indicate a mapping relationship between a type of AI inference service and an inference method.

[0106] In some embodiments, the third transceiver is further configured to receive inference data sent by the first node, the inference data being indicated by fourth information received by the first node from a second node, the second node being a node requesting the first AI inference service.

[0107] In some embodiments, there are multiple third nodes, and the third transceiver is further configured to receive split data sent by the first node, the split data being obtained by the first node by splitting the inference data.

[0108] In some embodiments, the third transceiver is further configured to receive an inference task sent by the first node, the inference task being obtained by the first node by distributing a plurality of inference tasks associated with the first AI inference service.

[0109] In some embodiments, the second information comprises at least one of the following: the first information; thirteenth indication information used to indicate the first AI inference service; fourteenth indication information used to indicate a plurality of inference tasks associated with the first AI inference service; and fifteenth indication information used to indicate the first inference method used by the first AI inference service.

[0110] In some embodiments, the second information is further used to request the third node to establish a data connection with the first node.

[0111] In some embodiments, the third transceiver is further configured to send fifth information to the first node, the fifth information being used to indicate an access address of the third node.

[0112] In some embodiments, the first node and the third node are deployed on the same device.

[0113] In a seventh aspect, the embodiments of the present disclosure provide a communication device. The communication device comprises one or more processors. The communication device is configured to perform the communication method in the first aspect and possible implementation manners thereof.

[0114] In an eighth aspect, the embodiments of the present disclosure provide a communication system comprising a first node, a second node and a third node. The first node is configured to implement the communication method in the first aspect and possible implementation manners thereof. The second node is configured to implement the communication method in the second aspect and possible implementation manners thereof. The third node is configured to implement the communication method in the third aspect and possible implementation manners thereof.

[0115] In a ninth aspect, the embodiments of the present disclosure provide a storage medium. The storage medium stores instructions. When the instructions are executed on a communication device, the communication device performs the communication method in the first aspect to the third aspect and possible implementation manners thereof.

[0116] In a tenth aspect, the embodiments of the present disclosure provide a computer program product comprising a computer program. When the computer program is executed on a communication device, the communication device performs the communication method in the first aspect to the third aspect and possible implementation manners thereof.

[0117] In an eleventh aspect, the embodiments of the present disclosure provide a computer program comprising code. When the code is executed on a processor, the code implements the communication method in the first aspect to the third aspect and possible implementation manners thereof.

[0118] In a twelfth aspect, the embodiments of the present disclosure provide a chip or chip system. The chip or chip system comprises processing circuitry. The processing circuitry is configured to perform the communication method in the first aspect to the third aspect and possible implementation manners thereof.

[0119] It can be understood that the above-mentioned first node, second node, third node, communication device, communication system, storage medium, computer program product, computer program, chip or chip system are all used to perform the method proposed in the embodiments of the present disclosure. Therefore, the beneficial effects they can achieve can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0120] The embodiments of the present disclosure provide a communication method, a communication device, a communication system, a storage medium and a program product. In some embodiments, the terms of the communication method, the method of providing AI model inference service, the information processing method, the information transmission method, etc. can be replaced with each other. The terms of the first node, the second node, the third node, the communication device, the device providing the AI model inference service, the information processing device, the information transmission device, etc. can be replaced with each other. The terms of the information processing system, the communication system, the system providing the AI model inference service, etc. can be replaced with each other.

[0121] The embodiments of the present disclosure are not exhaustive, but only illustrate some embodiments, and are not specific limitations on the protection scope of the present disclosure. In the case of no contradiction, each step in an embodiment can be implemented as an independent embodiment, and the steps can be combined arbitrarily, for example, the scheme after removing part of the steps in an embodiment can also be implemented as an independent embodiment, and the order of the steps in an embodiment can be exchanged arbitrarily, in addition, the optional implementation in an embodiment can be combined arbitrarily; in addition, the embodiments can be combined arbitrarily, for example, part or all of the steps of different embodiments can be combined arbitrarily, an embodiment can be combined with the optional implementation of other embodiments arbitrarily.

[0122] In each embodiment of the present disclosure, the terms and / or descriptions between the embodiments are consistent if there is no special description and logical conflict, and can be referred to each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0123] The terms used in the embodiments of the present disclosure are only for the purpose of describing the specific embodiments, and not as a limitation on the present disclosure.

[0124] In the embodiments of the present disclosure, unless otherwise specified, the elements expressed in singular form, such as "one", "a", "the", "above", "said", "preceding", "this" and the like, can represent "one and only one", and can also represent "one or more", "at least one" and the like. For example, in the case of using articles such as "a", "an", "the" and the like in English, the noun after the article can be understood as singular expression, and can also be understood as plural expression.

[0125] In the embodiments of the present disclosure, "plurality" means two or more.

[0126] In some embodiments, the terms "at least one of", "one or more", "a plurality of", "multiple" and the like can be replaced with each other.

[0127] In some embodiments, "at least one of A, B", "A and / or B", "in one case A, in another case B", "responsive to case A, responsive to case B" and the like, can be used to represent one or more of the following technical solutions: in some embodiments, A (A is executed regardless of B); in some embodiments, B (B is executed regardless of A); in some embodiments, A and B are selected from (A and B are selectively executed); in some embodiments, A and B (A and B are executed). When there are more branches such as A, B, C, and the like, the above is similar.

[0128] In some embodiments, "A or B" and the like can include the following technical solutions according to the case: in some embodiments, A (A is executed regardless of B); in some embodiments, B (B is executed regardless of A); in some embodiments, A and B are selected from (A and B are selectively executed). When there are more branches such as A, B, C, and the like, the above is similar.

[0129] In the embodiments of the present disclosure, the prefix words "first", "second" and the like are only used to distinguish different description objects, and do not constitute a limitation on the position, order, priority, quantity or content of the description objects. The description of the description objects should be referred to the description in the context of the claims or embodiments, and should not be limited by the prefix words. For example, the description object is "field", and the ordinal words before "field" in "first field" and "second field" do not limit the position or order between "fields", and "first" and "second" do not limit whether the "fields" modified thereby are in the same message or not, nor limit the order of "first field" and "second field". For another example, the description object is "level", and the ordinal words before "level" in "first level" and "second level" do not limit the priority between "levels". For another example, the quantity of the description object is not limited by the ordinal words, and can be one or more. For example, "first device", wherein the quantity of "device" can be one or more. In addition, the objects modified by different prefix words can be the same or different, for example, the description object is "device", and "first device" and "second device" can be the same device or different devices, and the types thereof can be the same or different; for another example, the description object is "information", and "first information" and "second information" can be the same information or different information, and the content thereof can be the same or different.

[0130] In some embodiments, "including A", "containing A", "for indicating A", "carrying A" can be interpreted as directly carrying A, or indirectly indicating A.

[0131] In some embodiments, the terms "in response to", "in response to determining", "in the case of", "when", "when", "if", "if" and the like can be replaced with each other.

[0132] In some embodiments, the terms "greater than", "greater than or equal to", "not less than", "more than", "more than or equal to", "not less than", "higher than", "higher than or equal to", "not lower than", "above", and the like can be replaced with each other, and the terms "less than", "less than or equal to", "not greater than", "fewer than", "fewer than or equal to", "not more than", "lower than", "lower than or equal to", "not higher than", "below", and the like can be replaced with each other.

[0133] In some embodiments, an apparatus and the like can be interpreted as an entity, and can also be interpreted as virtual, and the name thereof is not limited to the name recited in the embodiments, and the terms "apparatus", "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "subject", and the like can be replaced with each other.

[0134] In some embodiments, "network" can be interpreted as an apparatus (for example, an access network device, a core network device, and the like) included in the network.

[0135] In some embodiments, the terms “network devices,” “access network devices (AN devices),” “radio access network devices (RAN devices),” “base stations (BSs),” “radio base stations,” “fixed stations,” “nodes,” “access network nodes,” “access points,” “transmission points (TPs),” “reception points (RPs),” “transmission / reception points (TRPs),” “panels,” “antenna panels,” “antenna arrays,” “cells,” “macro cells,” “small cells,” “femtocells,” “picocells,” “sectors,” “cell groups,” “serving cells,” “carriers,” “component carriers,” “bandwidth parts (BWPs),” and the like can be used interchangeably.

[0136] In some embodiments, the terms "terminal," "terminal device," "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," and so on can be replaced with each other.

[0137] In some embodiments, the access network device, the core network device, or the network device can be replaced with a terminal. For example, the embodiments of the present disclosure can also be applied to a structure in which communication between the access network device, the core network device, or the network device and the terminal is replaced with communication between a plurality of terminals (e.g., device-to-device (D2D), vehicle-to-everything (V2X), etc.). In this case, the terminal can also be configured to have all or part of the functions of the access network device. In addition, the terms "uplink," "downlink," and the like can also be replaced with terms corresponding to the inter-terminal communication (e.g., "side"). For example, the uplink channel, the downlink channel, and the like can be replaced with the side channel, and the uplink, the downlink, and the like can be replaced with the sidelink.

[0138] In some embodiments, the terminal can be replaced with the access network device, the core network device, or the network device. In this case, the access network device, the core network device, or the network device can also be configured to have all or part of the functions of the terminal.

[0139] In some embodiments, the data, information, etc. can be obtained in compliance with the laws and regulations of the country where the location is situated.

[0140] In some embodiments, the data, information, etc. can be obtained after obtaining the consent of the user.

[0141] In addition, each element, each row, or each column in the table of the embodiments of the present disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0142] FIG. 1A is a schematic diagram of an architecture of a communication system according to an embodiment of the present disclosure. As shown in FIG. 1A, the communication system 100 includes a first node 101, a second node 102, and a fourth node 103.

[0143] In some embodiments, the first node is configured to manage an AI model inference service function (AiISF).

[0144] In some embodiments, the first node is configured to obtain an inference service request.

[0145] In some embodiments, the first node is configured to determine an inference method based on the inference service request.

[0146] In some embodiments, the first node is configured to determine a node that provides an inference service based on the inference service request.

[0147] In some embodiments, the first node is configured to determine a plurality of inference tasks associated with the inference service.

[0148] In some embodiments, the first node is configured to assign the plurality of inference tasks to a plurality of nodes that provide the inference service.

[0149] In some embodiments, the first node is configured to initiate a request to establish a data connection between the first node and the node that provides the inference service.

[0150] In some embodiments, the name of the first node is not limited, and it is, for example, “AiISF”, “inference service management node”, “AI inference management node”, “model inference management node”, “network data analytics function (NWDAF)”, etc.

[0151] In some embodiments, the second node is configured to request an inference service.

[0152] In some embodiments, the second node is configured to establish a data connection with a node that manages the AiISF.

[0153] In some embodiments, the second node is configured to send inference data to a node managing the AiISF.

[0154] In some embodiments, the second node is configured to obtain inference result of the AI inference service.

[0155] In some embodiments, the second node is not limited in name, for example, it can be a “request node”, “initiation node”, “inference request node”, “AI inference request node”, “model inference request node”, etc.

[0156] In some embodiments, the fourth node is configured to provide inference service.

[0157] In some embodiments, the fourth node can be dedicated to providing inference service. In an example, the fourth node is an operator server dedicated to providing inference service.

[0158] In some embodiments, the fourth node can provide other services in addition to inference service. In an example, the fourth node is a network element providing location management function (LMF), and the LMF can also provide AI inference service.

[0159] In some embodiments, the fourth node is not limited in name, for example, it can be an “inference node”, “AI inference node”, “AI model inference node”, etc.

[0160] In some embodiments, all fourth nodes can be third nodes, or a part of the fourth nodes are third nodes.

[0161] In some embodiments, the third node is configured to establish a data connection with a node managing the AiISF.

[0162] In some embodiments, the third node is configured to provide inference result to a node managing the AiISF.

[0163] In some embodiments, the third node has the same function as the fourth node, which will not be repeated here. The third node can be understood as a target inference node, and the fourth node can be understood as a candidate inference node.

[0164] In some embodiments, the fourth node is not limited in name, for example, it can be an “AiISF”, “inference node”, “AI inference node”, “AI model inference node”, “inference execution node”, etc. In some embodiments, the above-mentioned first node can be a core network device, an access network device (such as RAN), etc.

[0165] In some embodiments, the second node, the third node, and the fourth node described above can be a terminal, an access network device, a core network device, etc. In some embodiments, the first node, the second node, the third node, and the fourth node described above can be deployed in one device, or can be respectively deployed in multiple devices, each of which has the function of one or more nodes described above.

[0166] In some embodiments, the first node and the second node are deployed in the same device.

[0167] In an example, the first node and the third node are AISF.

[0168] In an example, the first node and the third node are NWDAF.

[0169] In some embodiments, the terminal includes at least one of a mobile phone, a wearable device, an Internet of Things device, a car with communication function, a smart car, a Pad, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, a wireless terminal device in smart home, etc., but is not limited thereto.

[0170] In some embodiments, the access network device is at least one of a node or device that accesses a terminal to a wireless network, and the network device can include at least one of an evolved NodeB (eNB), a next generation eNB (ng-eNB), a next generation NodeB (gNB), a node B (NB), a home node B (HNB), a home evolved node B (HeNB), a wireless backhaul device, a radio network controller (RNC), a base station controller (BSC), a base transceiver station (BTS), a base band unit (BBU), a mobile switching center, a base station in a communication system, an open base station (Open RAN), a cloud base station (Cloud RAN), a base station in other communication systems, an access node in a Wi-Fi system, but is not limited thereto.

[0171] In some embodiments, the technical solutions of the present disclosure can be applied to the Open RAN architecture, at this time, the interfaces between or within the network devices involved in the embodiments of the present disclosure can become internal interfaces of the Open RAN, and the processes and information interactions between these internal interfaces can be realized by software or programs.

[0172] In some embodiments, the access network device can be composed of a central unit (CU) and a distributed unit (DU), wherein the CU can also be referred to as a control unit. The CU-DU structure can split the protocol layers of the network device, and part of the protocol layer functions are controlled by the CU, and the remaining part or all of the protocol layer functions are distributed in the DU and controlled by the CU, but is not limited thereto.

[0173] In some embodiments, the core network device can be one device including one or more network elements, or can be multiple devices or device groups including one or more network elements respectively. The network element can be virtual or physical. The core network includes at least one of an evolved packet core (EPC), a 5G core network (5GCN), and a next generation core (NGC).

[0174] It can be understood that the communication system described in the embodiments of the present disclosure is for more clearly illustrating the technical solutions of the embodiments of the present disclosure, and does not constitute a limitation on the technical solutions proposed by the embodiments of the present disclosure. Those skilled in the art can know that, with the evolution of system architecture and the appearance of new business scenarios, the technical solutions proposed by the embodiments of the present disclosure are also applicable to similar technical problems.

[0175] The following embodiments of the present disclosure can be applied to the communication system 100 shown in FIG. 1A or part of the subject, but are not limited thereto. The subjects shown in FIG. 1A are illustrative, and the communication system can include all or part of the subjects in FIG. 1A, or other subjects other than FIG. 1A. The number and form of each subject is arbitrary, each subject can be real or virtual, the connection relationship between each subject is illustrative, each subject can not be connected or can be connected, the connection can be in any way, can be direct connection or indirect connection, can be wired connection or wireless connection.

[0176] Embodiments of the present disclosure can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), 6th generation mobile communication system (6G), Future Radio Access (FRA), New-Radio Access Technology (RAT), new radio (NR), new radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-Wide Band (UWB), Bluetooth (Bluetooth (registered trademark)), Public Land Mobile Network (PLMN) network, Device-to-Device (D2D) system, Machine to Machine (M2M) system, Internet of Things (IoT) system, Vehicle-to-Everything (V2X), system using other communication methods, next-generation system expanded based thereon, and the like. Further, a plurality of systems can be combined (for example, combination of LTE or LTE-A and 5G, and the like).

[0177] Hereinafter, terms related to the present disclosure are explained and interpreted.

[0178] I. Introduction to international mobile telecommunications (IMT) 2030

[0179] The usage scenarios for IMT-2030 are envisioned to be extended from the usage scenarios for IMT-2020, i.e., enhanced mobile broadband (eMBB), ultra-reliable and low latency communication (URLLC), and massive machine type communication (mMTC) introduced in the Recommendation International Telecommunication Union - Radio communication (ITU-R M).2083, to a wider range of usage that requires evolution and new capabilities. In addition to the extended IMT-2020 usage scenarios, IMT-2030 also envisions new usage scenarios resulting from functionalities such as artificial intelligence and sensing that were not supported by the previous generations of IMT design.

[0180] In some embodiments, the usage scenarios for IMT-2030 can include artificial intelligence and communications.

[0181] In some embodiments, the artificial intelligence and communications usage scenario will support distributed computing and artificial intelligence applications. Typical application examples can include IMT-2030 assisted autonomous driving, autonomous collaboration among devices for medical assistance applications, offloading heavy computing operations across devices and networks, creation and prediction of digital twins, etc.

[0182] In some embodiments, the artificial intelligence and communications usage scenario will support high regional traffic capacity and user experience data rates, low latency, and high reliability, depending on the specific application scenario. In addition to the communications aspect, this usage scenario is expected to include a set of new functionalities related to the integration of artificial intelligence and computing functionalities into IMT-2030, including data collection, preparation, and processing from different sources, distributed artificial intelligence model training, model sharing, and distributed inference across IMT systems, and computing resource orchestration and linking, etc.

[0183] In some embodiments, compared to the capabilities described in the ITU-R M.2083 Recommendation for IMT-2020, IMT-2030 is expected to provide enhanced capabilities, as well as new capabilities to support the extended usage scenarios for IMT-2030. In addition, each capability can have different relevance and applicability in different usage scenarios.

[0184] In some embodiments, the functions of IMT-2030 can include:

[0185] 1. Applicable AI-related capabilities

[0186] Applicable AI-related capabilities can refer to the capabilities of providing certain functions throughout IMT-2030 to support artificial intelligence applications. These functions include distributed data processing, distributed learning, AI computing, AI model execution, and AI model inference, etc.

[0187] 2. Interoperability

[0188] Interoperability can refer to the radio interface based on member inclusivity and transparency in order to achieve functions between different entities of the system.

[0189] II. Introduction to service architecture for AI inference service

[0190] FIG. 1B is a schematic diagram of a service architecture for AI inference service oriented to 6G according to an embodiment of the present disclosure. As shown in FIG. 1B, the service architecture includes a UE of 6G, an access network device (RAN) of 6G, an access and mobility management function (AMF) of 6G, a session management function (SMF) of 6G, an AiISF, an inference network function (inference NF) of 6G, a network function 1 with inference capability (NF1 with inference capability) of 6G, a network function 2 with inference capability (NF2 with inference capability) of 6G, and a network function 3 with inference capability (NF3 with inference capability) of 6G.

[0191] In some embodiments, the 6G inference NF can be a 6G network element dedicated to providing AI model inference service.

[0192] In some embodiments, the 6G NF1 with inference capability, the 6G NF2 with inference capability, and the 6G NF3 with inference capability can be 6G network elements with other functions and also with AI model inference capabilities.

[0193] In an example, the 6G NF1 with inference capability is an LMF with inference capability.

[0194] In an example, the 6G NF2 with inference capability is an analytics data repository function (ADRF) with inference capability.

[0195] FIG. 1C is a schematic diagram of a data connection structure between an AiISF, a UE, and inference nodes according to an embodiment of the present disclosure. As shown in FIG. 1C, the AiISF supports establishing a data connection between the UE and the AiISF, the AiISF supports establishing a data connection between the AiISF and an inference node 1, and the AiISF supports establishing a data connection between the AiISF and an inference node 2.

[0196] In some embodiments, the inference node 1 and the inference node 2 are configured to provide inference services.

[0197] In some embodiments, the inference node 1 and the inference node 2 can be dedicated to providing inference services. In an example, the inference node 1 and the inference node 2 are operator servers dedicated to providing inference services.

[0198] In some embodiments, the inference node 1 and the inference node 2 can provide other services in addition to providing inference services. In an example, the inference node 1 and the inference node 2 are network elements providing a location management function (LMF).

[0199] In some embodiments, AI model inference of a terminal can be performed locally at the terminal. However, due to limited computing capability and computing resources of the terminal, the terminal cannot support inference services beyond the computing capability and computing resources of the terminal. In this case, although a third-party server (e.g., a non-operator server) can be used to provide inference services for the terminal, this will cause delay and affect user experience.

[0200] The embodiment of the disclosure provides a communication method, a communication device, a communication system, a storage medium and a program product. A first node determines a third node for providing a first AI inference service according to first information from a second node for requesting the first AI inference service, then sends second information to indicate that the third node provides the first AI inference service, and finally receives an inference result associated with the first AI inference service sent by the third node. Since the third node is determined based on the first information, that is, the third node matches the requirements of the first information, the third node provides the AI inference service and sends the inference result, which can not only reduce the demand of the AI inference service on the terminal computing capability and resources, but also reduce the delay of providing the inference service for the terminal.

[0201] FIG. 2 is an interaction schematic diagram of a communication method provided by an embodiment of the disclosure. As shown in FIG. 2, the embodiment of the disclosure relates to a communication method, which is executed by the communication system 100 and includes steps S2101 to S2112.

[0202] In the embodiment of the disclosure, taking the first node as the AiISF, the second node as the terminal A, the third node as the inference node A, and the fourth node as the inference node as an example, the process of providing the AI inference service and obtaining the inference result is described.

[0203] In step S2101, the terminal A sends first information.

[0204] In some embodiments, the terminal A can be one terminal, or one or more terminals in a terminal group.

[0205] In some embodiments, the AiISF receives the first information.

[0206] In some embodiments, the first information is used to request the first AI inference service.

[0207] In some embodiments, the first AI inference service can be a positioning service, for example, inferring the moving track of the object A at a future time based on the moving track and the current position of the object A at a historical time.

[0208] In some embodiments, the first AI inference service can be a measurement service, for example, inferring the measurement result of the cell A at a current time based on the measurement result of the cell A at a historical time. For example, inferring the measurement result of the cell A based on the measurement result of the neighboring cell of the cell A. For example, inferring the channel state information (CSI) of the channel A at a current time based on the CSI of the channel A at a historical time.

[0209] In some embodiments, the first AI inference service can also be an image recognition service, a voice recognition service, a user behavior prediction service, a data repair service, etc.

[0210] In some embodiments, the name of the first information is not limited, which is, for example, "first AI inference service request information", "information requesting a first AI inference service", "AI inference service request information", etc.

[0211] In some embodiments, the number of first AI inference services can be one or more.

[0212] In some embodiments, in the case that the terminal A requests multiple first AI inference services, the terminal A can send one first information to request multiple first AI inference services, or the terminal A can send multiple first information, each of which is used to request one first AI inference service.

[0213] In some embodiments, the first information is carried in a first message, and the first message is an inference service request message.

[0214] In some embodiments, the first information is used to indicate at least one of the following: a terminal requesting a first AI inference service, a quality requirement of the first AI inference service, a model requirement of the first AI inference service, a type of the first AI inference service, a location of the terminal requesting the first AI inference service, and a service area of the first AI inference service.

[0215] In some embodiments, the terminal A requesting the first AI inference service can be indicated by first indication information. In an example, the first indication information is identification information (e.g., UE ID) of the terminal A. In an example, the first indication information is identification information (e.g., UE group ID) of a terminal group to which the terminal A belongs.

[0216] In some embodiments, the name of the first indication information is not limited, which is, for example, "requester identification information", "device identification information", "identification information", etc.

[0217] In some embodiments, the quality requirement of the first AI inference service can be indicated by second indication information. In some embodiments, the quality of service (QoS) of the first AI inference service can be indicated by the second indication information.

[0218] In some embodiments, the name of the second indication information is not limited, which is, for example, "quality requirement information", "quality information", "service quality information", etc.

[0219] In an embodiment, the second indication information includes at least one of: latency information, bandwidth information, transmission rate information, packet loss rate information, and data privacy information.

[0220] In some embodiments, the model requirement of the first AI inference service can be indicated by the third indication information.

[0221] In some embodiments, the name of the third indication information is not limited, which is, for example, "model requirement information", "model information", "model indication information", etc.

[0222] In some embodiments, the third indication information includes at least one of: model identification information, model version information, model function information, model structure information, model size information, inference accuracy information, and inference speed information.

[0223] In some embodiments, the type of the first AI inference service can be indicated by the fourth indication information.

[0224] In some embodiments, the name of the fourth indication information is not limited, which is, for example, "service type information", "type information", "type indication information", etc.

[0225] In some embodiments, the type of the first AI inference service can be associated with the quality requirement of the first AI inference service. The type of the first AI inference service can be determined by the quality requirement of the first AI inference service, and the quality requirement of the first AI inference service can be determined by the type of the first AI inference service. It can be understood that different types of AI inference services correspond to different quality requirements. In other words, the fourth indication information is associated with the second indication information.

[0226] In some embodiments, the location of the terminal A requesting the first AI inference service can be indicated by the fifth indication information.

[0227] In some embodiments, the name of the fifth indication information is not limited, which is, for example, "requester location information", "device location information", "location information", etc.

[0228] In some embodiments, the service area of the first AI inference service can be indicated by the sixth indication information.

[0229] In some embodiments, the name of the sixth indication information is not limited, which is, for example, "service area information", "area information", "service range information", "area identification information", etc.

[0230] In some embodiments, the first information can comprise at least one of: first indication information (denoted as requester identification information), second indication information (denoted as quality requirement information), third indication information (denoted as model requirement information), fourth indication information (denoted as service type information), fifth indication information (denoted as requester location information), and sixth indication information (service area information).

[0231] In some embodiments, when the first information comprises the service type information, the first information can not carry the quality requirement information.

[0232] In some embodiments, when the first information does not comprise the quality requirement information, the AiISF can determine the quality requirement information according to the service type information.

[0233] In some embodiments, when the first information comprises the requester identification information, the first information can not carry the requester location information.

[0234] In some embodiments, when the first information does not comprise the requester location information, the AiISF can determine the requester location information according to the requester identification information.

[0235] In some embodiments, when the second node and the first node are deployed on the same device, step S2101 can be omitted.

[0236] In an example, the second node and the first node are both AiISFs, in which case, the AiISF requests the first AI inference service, determines the first information, and performs step S2102 according to the first information.

[0237] In step S2102, the AiISF determines the inference node A.

[0238] In some embodiments, the inference node A is used to provide the first AI inference service, and the number of inference node A can be one or more.

[0239] In some embodiments, the AiISF can determine the inference node A that meets the requirements of the first information according to the first information.

[0240] In some embodiments, when the first information comprises the requester identification information or the requester location information, the AiISF can determine the inference node closest to the location of the terminal A as the inference node A, and the AiISF can also determine a first area with the location of the terminal A as the center and a first value as the radius according to the location of the terminal A, and determine the inference nodes within the first area as the inference node A.

[0241] In some embodiments, the number of inference nodes A is determined when the first information contains quality requirement information. The number of inference nodes A is determined when the first information contains quality requirement information and requester location information.

[0242] In some embodiments, the number of inference nodes A is determined when the first information contains service type information. The number of inference nodes A is determined when the first information contains service type information and requester location information.

[0243] In some embodiments, the AiISF can determine the inference node A according to the first information and the eleventh indication information.

[0244] In some embodiments, the eleventh indication information can be configured by the network device or specified by the protocol.

[0245] In some embodiments, the eleventh indication information can be used to indicate the mapping relationship between the type of AI inference service and the inference node, in which case the AiISF can determine the inference node A corresponding to the type of the first AI inference service from the above mapping relationship.

[0246] In an example, assuming that there are 4 types of AI inference services (for example, type 1, type 2, type 3, and type 4) and a total of 4 inference nodes (for example, inference node A1, inference node A2, inference node A3, and inference node A4), the mapping relationship between the type of AI inference service and the inference node is shown in Table 1 below:

[0247] Table 1

[0248] In some embodiments, the eleventh indication information can be used to indicate the mapping relationship between the type of model adopted by the AI inference service and the inference node, in which case the AiISF can determine the inference node A corresponding to the type of model adopted by the first AI inference service from the above mapping relationship.

[0249] In some embodiments, the eleventh indication information can be used to indicate the mapping relationship between the service area of the AI inference service and the inference node, in which case the AiISF can determine the inference node A corresponding to the service area of the first AI inference service from the above mapping relationship.

[0250] In some embodiments, the eleventh indication information can be used to indicate the mapping relationship between the preset area associated with the location of the device requesting the AI inference service and the inference node, in which case the AiISF can determine the inference node A corresponding to the preset area A associated with the location of the terminal from the above mapping relationship.

[0251] In some embodiments, the AiISF can determine the inference node A from the plurality of inference nodes according to the first information and third information associated with the plurality of inference nodes.

[0252] In some embodiments, the AiISF can request the plurality of inference nodes to obtain the third information.

[0253] In some embodiments, the AiISF can further request a network repository function (NRF) network element to obtain the third information of the inference node.

[0254] In some embodiments, the third information is used to indicate at least one of: support capabilities of the inference node for different types of AI inference services, support capabilities of the inference node for AI models, resources of the inference node for processing AI inference services, a location of the inference node, the inference node, and support capabilities of the inference node for inference methods.

[0255] In some embodiments, the support capabilities of the inference node for different types of AI inference services can be indicated by seventh indication information.

[0256] In some embodiments, the name of the seventh indication information is not limited, and is, for example, “type support information”, “capability information”, “service support information”, etc.

[0257] In an example, assuming that there are three types (e.g., type 1, type 2, and type 3) of AI inference services, the seventh indication information indicates that the inference node A1 supports type 1 and type 2 AI inference services, in which case it can be considered that the inference node A1 does not support type 3 AI inference services. In an example, the seventh indication information further indicates that the inference node A2 does not support type 1 AI inference services, in which case it can be considered that the inference node A2 supports type 2 and type 3 AI inference services.

[0258] In some embodiments, the support capabilities of the inference node for AI models can be indicated by eighth indication information.

[0259] In some embodiments, the name of the eighth indication information is not limited, and is, for example, “model support information”, “model capability information”, “capability information”, etc.

[0260] In an example, the eighth indication information indicates that the inference node A1 supports AI model A and AI model B. In an example, the eighth indication information further indicates that the inference node A2 does not support AI model A and AI model C.

[0261] In some embodiments, the eighth indication information can also be used to indicate the support capability of the inference node for different types of AI models. In some embodiments, the different types of AI models can be classified based on model structure, can be classified based on model function, or can be classified based on the data type processed by the model, and the present disclosure does not make any limitation thereto.

[0262] In an example, the eighth indication information indicates that the inference node A1 supports AI models of type 1, type 2 and type 3. In an example, the model support information further indicates that the inference node A2 does not support AI models of type 2.

[0263] In some embodiments, the resources of the inference node processing the AI inference service can be indicated by the ninth indication information.

[0264] In some embodiments, the name of the ninth indication information is not limited, which is, for example, “resource information”, “AI resource information”, “service resource information” and the like.

[0265] In some embodiments, the ninth indication information includes at least one of the following: storage resource information, computing resource information.

[0266] In some embodiments, the ninth indication information is also used to indicate the idle resources in the resources of the inference node processing the AI inference service.

[0267] In some embodiments, the location of the inference node can be indicated by the tenth indication information.

[0268] In some embodiments, the name of the tenth indication information is not limited, which is, for example, “inference party location information”, “inference location information”, “device location information”, “location information” and the like.

[0269] In some embodiments, the inference node can be indicated by inference party identification information.

[0270] In some embodiments, the name of the inference party identification information is not limited, which is, for example, “inference indication information”, “device identification information”, “inference device identification information”, “device identification information” and the like.

[0271] In an example, the inference party identification information is the ID of the inference node.

[0272] In some embodiments, the support capability of the inference node for the inference method can be indicated by inference method support information.

[0273] In some embodiments, the name of the inference method support information is not limited, which is, for example, “method support information”, “inference capability information”, “capability information” and the like.

[0274] In an example, the inference method support information is used to indicate that the inference node A1 supports inference method 1 and inference method 2. In an example, the inference method support information is also used to indicate that the inference node A2 does not support inference method 3.

[0275] In some embodiments, the third information comprises at least one of the following: seventh indication information (denoted as type support information), eighth indication information (denoted as model support information), ninth indication information (denoted as resource information), tenth indication information (denoted as inference party location information), inference party identification information, inference method support information.

[0276] In some embodiments, in the case that the third information does not contain inference party location information, the AiISF can determine the inference party location information according to the inference party identification information.

[0277] In some embodiments, in the case that the first information comprises quality requirement information and requestor location information, and the third information comprises inference party location information and resource information, the AiISF can determine, according to the third information, an inference node A that meets the requirements of the first information from the plurality of inference nodes.

[0278] In an example, assuming that the quality requirement information indicates that the time delay of the first AI inference service is less than a first threshold value, the AiISF can calculate the distance between each inference node and the terminal A according to the inference party location information and the requestor location information, determine the idle resources of each inference node that can currently handle the AI inference service according to the resource information, and select, based on this, an inference node A from the plurality of inference nodes that has a distance less than a second threshold value and an idle resource greater than a third threshold value.

[0279] In some embodiments, in the case that the first information comprises model requirement information, and the third information comprises model support information, the AiISF can determine, according to the third information, an inference node A that meets the requirements of the first information from the plurality of inference nodes.

[0280] In an example, assuming that the model requirement information indicates that the model of the first AI inference service is AI model C, the AiISF can determine, according to the model support information, an inference node A that supports AI model C from the plurality of inference nodes.

[0281] In some embodiments, in the case that the first information comprises service type information, and the third information comprises type support information, the AiISF can determine, according to the third information, an inference node A that meets the requirements of the first information from the plurality of inference nodes.

[0282] In an example, assuming that the service type information indicates that the type of the first AI inference service is type 1, the AiISF can determine, according to the type support information, an inference node A that supports processing AI inference services of type 1 from the plurality of inference nodes.

[0283] In some embodiments, the AiISF can determine, according to the first information, a first inference method adopted by the first AI inference service.

[0284] In some embodiments, the first inference method can be used to determine the inference node A.

[0285] In some embodiments, the first inference method can be used to determine the number of inference nodes A.

[0286] In some embodiments, the AiISF determines the inference node A according to the first inference method, the first information, and the third information. In an example, the AiISF can determine, according to the third information, the inference node A that meets the requirements of the first information and supports the first inference method from the plurality of inference nodes.

[0287] In some embodiments, the first inference method is one of the following: a central inference method, a distributed inference method, and a federated learning inference method. In some embodiments, the central inference method is to concentrate the AI inference service in one inference node for execution. The distributed inference method is to distribute the plurality of inference tasks associated with the AI inference service to the plurality of nodes for execution. The federated learning inference method is a special distributed inference method that allows inference through federated learning while keeping user data localized, ensuring the privacy of user data.

[0288] In an example, in a case where the AiISF determines, according to the first information, that the data volume of the first AI inference service is small, the requirement for real-time performance is low, and the model adopted is small, the first inference method can be determined as the central inference method.

[0289] In an example, in a case where the AiISF determines, according to the first information, that the data volume of the first AI inference service is large, the requirement for real-time performance is high, and the model adopted is large, the first inference method can be determined as the distributed inference method.

[0290] In an example, in a case where the AiISF determines, according to the first information, that the first AI inference service requires data privacy protection, the first inference method can be determined as the federated learning inference method.

[0291] In an example, in a case where the first inference method is determined as the central inference method, the AiISF can determine the number of inference nodes A as one.

[0292] In an example, in a case where the first inference method is determined as the distributed inference method or the federated learning inference method, the AiISF can determine the number of inference nodes A as a plurality.

[0293] In some embodiments, the AiISF can determine the first inference method adopted by the first AI inference service according to the first information and the preconfigured twelfth indication information.

[0294] In some embodiments, the twelfth indication information is used to indicate the mapping relationship between the type of the AI inference service and the inference method.

[0295] In some embodiments, the twelfth indication information is used to indicate the mapping relationship between the type of the model adopted by the AI inference service and the inference method.

[0296] In some embodiments, in the case that there are multiple inference methods determined by the AiISF according to the first information, the AiISF can determine one inference method from the multiple inference methods as the first inference method based on the eleventh indication information.

[0297] In step S2103, the AiISF allocates the first AI inference service.

[0298] In some embodiments, the inference node A receives the allocated first AI inference service.

[0299] In some embodiments, in the case that the number of the first AI inference service is one and the number of the inference node A is one, the inference node A can independently complete the first AI inference service.

[0300] In some embodiments, in the case that the number of the first AI inference service is one and the number of the inference node A is multiple, the AiISF determines multiple inference tasks associated with the first AI inference service and allocates the multiple inference tasks to the multiple inference nodes A, at this time, each inference node A only processes a part of the inference tasks.

[0301] In some embodiments, in the case that the number of the first AI inference service is multiple and the number of the inference node A is one, the inference node A can independently complete the multiple first AI inference services.

[0302] In some embodiments, in the case that the number of the first AI inference service is multiple and the number of the inference node A is multiple, the AiISF can allocate the multiple first AI inference services to the multiple inference nodes A according to the inference node A corresponding to each first AI inference service determined in step S2102. In the case that one of the multiple first AI inference services corresponds to multiple inference nodes A, the AiISF determines multiple inference tasks associated with the first AI inference service and allocates the multiple inference tasks to the multiple inference nodes A.

[0303] In an example, case 1, the AiISF determines that the first AI inference service 1 corresponds to the inference node A1, the first AI inference service 2 corresponds to the inference node A1, and the first AI inference service 3 corresponds to the inference node A2. In this case, the AiISF directly allocates the plurality of first AI inference services according to the determined inference nodes A.

[0304] Case 2, the AiISF determines that the first AI inference service 1 corresponds to the inference node A1, the first AI inference service 2 corresponds to the inference node A1 and the inference node A2, and the first AI inference service 3 corresponds to the inference node A2. In this case, the AiISF allocates the first AI inference service 1 to the inference node A1, allocates the first AI inference service 2 to the inference node A2, and determines that the first AI inference service 2 is associated with 3 tasks (for example, task 1, task 2, and task 3), allocates the task 1 and the task 2 to the inference node A1, and allocates the task 3 to the inference node A2.

[0305] In some embodiments, the determination of the plurality of inference tasks associated with the first AI inference service in the above-mentioned embodiments can be understood as splitting the first AI inference service into a plurality of inference tasks, or decomposing the first AI inference service into a plurality of inference tasks.

[0306] In some embodiments, the allocation of the first AI inference service to the inference node A by the AiISF in the above-mentioned embodiments can be understood as establishing a mapping relationship between the first AI inference service and the inference node A, or establishing a mapping relationship between the plurality of inference tasks associated with the first AI inference service and the inference node A.

[0307] In step S2104, the AiISF sends second information.

[0308] In some embodiments, in the case where the number of inference nodes A is a plurality, the AiISF can send one second information to each of the plurality of inference nodes A, or the AiISF can send a plurality of second information to one inference node A1 of the plurality of inference nodes A, and then the inference node A1 sends the second information to other inference nodes A.

[0309] In some embodiments, the inference node A receives the second information.

[0310] In some embodiments, the second information can be used to instruct the inference node A to provide the first AI inference service.

[0311] In some embodiments, the second information can be used to instruct the inference node A to provide the first AI inference service using the first inference method.

[0312] In some embodiments, the second information can be used to request the inference node A to establish a data connection with the AiISF.

[0313] In some embodiments, the second information is carried in a second message, for example, a data connection establishment request message.

[0314] In some embodiments, the second information comprises one of: the first information, thirteenth indication information (denoted as service identification information), fourteenth indication information (denoted as task identification information), and fifteenth indication information (denoted as method identification information).

[0315] In some embodiments, the service identification information is used to indicate the first AI inference service. In an example, the service identification information is an ID of the first AI inference service. In an example, the service identification information is an ID of a terminal A that requests the first AI inference service.

[0316] In some embodiments, the task identification information is used to indicate a plurality of inference tasks associated with the first AI inference service. In an example, the task identification information is an ID of an inference task.

[0317] In some embodiments, the method identification information is used to indicate the first inference method. In an example, the method identification information is an ID of the first inference method.

[0318] In some embodiments, the service identification information, the task identification information, and the method identification information are configured by the AiISF.

[0319] In some embodiments, step S2104 can be omitted in the case that the first node and the third node are deployed on the same device. In an example, the first node and the third node are both AiISFs, and step S2104 is omitted.

[0320] In step S2105, the inference node A sends fifth information.

[0321] In some embodiments, the AiISF receives the fifth information.

[0322] In some embodiments, the fifth information is used to indicate that the inference node A agrees to provide the first AI inference service.

[0323] In some embodiments, the fifth information is used to indicate that the inference node A agrees to provide the first AI inference service using the first inference method.

[0324] In some embodiments, the fifth information is used to indicate that the inference node A agrees to establish a data connection with the AiISF.

[0325] In some embodiments, the fifth information is used to indicate the access address of the inference node A. In some embodiments, the access address of the inference node A comprises at least one of: an IP address, a port.

[0326] In some embodiments, the fifth information is carried in a third message, the third message being a response message of the second message, for example, a data connection establishment response message.

[0327] In some embodiments, in the case that the third node and the first node are deployed in the same device, step S2105 can be omitted. In an example, the first node and the third node are both AiISFs, and step S2105 is omitted.

[0328] In step S2106, the AiISF sends sixth information.

[0329] In some embodiments, the terminal A receives the sixth information. In some embodiments, the terminal A is the terminal that requests the first AI inference service.

[0330] In some embodiments, the sixth information is used to indicate that the inference node A agrees to provide the first AI inference service.

[0331] In some embodiments, the sixth information is used to indicate the access address of the inference node A.

[0332] In some embodiments, the sixth information is used for the terminal A to send inference data associated with the first AI inference service to the AiISF. In some embodiments, the inference data is input data of a model associated with the first AI inference service.

[0333] In some embodiments, the sixth information is used to indicate a mapping relationship between the first AI inference service and the inference node A.

[0334] In some embodiments, the sixth information is used to indicate a mapping relationship between a plurality of inference tasks associated with the first AI inference service and the inference node A.

[0335] In some embodiments, the sixth information comprises the fifth information, in which case, the fifth information is used to indicate the access address of the inference node A.

[0336] In some embodiments, in the case that the number of inference nodes A is multiple, the sixth information further comprises sixteenth indication information.

[0337] In some embodiments, the sixteenth indication information can be used to indicate the mapping relationship between the first AI inference service and the inference node A. In some embodiments, the sixteenth indication information can also be used to indicate the mapping relationship between the plurality of inference tasks associated with the first AI inference service and the inference node A.

[0338] In some embodiments, the sixth information is carried in the fourth message, for example, an inference service response message.

[0339] In some embodiments, if the second node and the first node are deployed on the same device, step S2106 can be omitted. In an example, the second node and the first node are both AiISFs, and step S2106 is omitted.

[0340] In step S2107, the terminal A sends the fourth information.

[0341] In some embodiments, the AiISF receives the fourth information. In some embodiments, the fourth information is used to indicate the inference data associated with the first AI inference service. In some embodiments, the AiISF receives the inference data.

[0342] In some embodiments, the AiISF receives the inference data through the data connection with the terminal A.

[0343] In some embodiments, the name of the fourth information is not limited, for example, it is “inference data information”, “information of inference data associated with the first AI inference service”, “data information”, etc.

[0344] In some embodiments, if the second node and the first node are deployed on the same device, for example, the second node is an AiISF, at this time, the execution subject of step S2107 and step S2108 is the same, and step S2107 can be omitted.

[0345] In step S2108, the AiISF sends the inference data.

[0346] In some embodiments, the inference node A receives the inference data.

[0347] In some embodiments, the AiISF sends the inference data based on the data connection between the AiISF and the inference node A.

[0348] In some embodiments, the inference node A receives the inference data. In some embodiments, the inference node A receives the inference data through the data connection with the AiISF.

[0349] In some embodiments, the AiISF sends the inference data to the inference node A indicated by the access address of the inference node A indicated by the fifth information based on the access address of the inference node A.

[0350] In some embodiments, the AiISF distributes each inference data A of the first AI inference service to the corresponding access address or splits and distributes the inference data A of one first AI inference service to the corresponding access address according to the indication of the sixteenth indication information.

[0351] In some embodiments, when the number of inference nodes A is one, the AiISF sends the inference data to the inference node A.

[0352] In some embodiments, when the number of inference nodes A is multiple and the number of first AI inference services is one, the AiISF splits the inference data A according to the mapping relationship between the multiple inference tasks associated with the first AI inference service and the multiple inference nodes A to obtain multiple split data, and sends the split data to the corresponding inference node A.

[0353] In an example, assuming that the AiISF determines that the number of inference nodes A is 3 (for example, inference node A1, inference node A2, and inference node A3), and the AiISF determines that the first AI inference service is associated with 4 inference tasks (for example, task 1, task 2, task 3, and task 4). The mapping relationship between the inference tasks and the 3 inference nodes is shown in Table 2 below, inference node A1 is assigned to process task 1 and task 2, inference node A2 is assigned to process task 3, and inference node A3 is assigned to process task 4. In this case, the AiISF splits the inference data into inference data a1 of task 1, inference data a2 of task 2, inference data a3 of task 3, and inference data a4 of task 4, and then sends inference data a1 and a2 to inference node A1, inference data a3 to inference node A2, and inference data a4 to inference node A3.

[0354] Table 2

[0355] In some embodiments, when the number of inference nodes A is multiple and the number of first AI inference services is multiple, if each inference node A is assigned a complete first AI inference service or multiple first AI inference services, the AiISF distributes each inference data to the corresponding inference node A according to the mapping relationship between the multiple first AI inference services and the multiple inference nodes A. If one of the multiple first AI inference services is assigned to multiple inference nodes A for completion, the AiISF splits the inference data associated with the first AI inference service to obtain multiple split data, and sends the split data to the corresponding inference node A.

[0356] In step S2109, the inference node A performs AI inference based on the inference data to obtain an inference result.

[0357] In some embodiments, the inference node A employs the first inference method to perform AI inference on the inference data to obtain the inference result.

[0358] In an example, in the case that the number of inference nodes A is one, the inference node A employs the first inference method to perform AI inference on the inference data to obtain the inference result.

[0359] In an example, in the case that the number of inference nodes A is multiple, the multiple inference nodes A respectively employ the first inference method to perform AI inference on the respective received split data to obtain the inference result.

[0360] In some embodiments, the inference node A employs the first inference method and inputs the inference data into the model A to obtain the inference result output by the model A.

[0361] In an example, in the case that the number of inference nodes A is one, the inference node A employs the first inference method and inputs the inference data into the model A to obtain the inference result output by the model A.

[0362] In an example, in the case that the number of inference nodes A is multiple, the multiple inference nodes A respectively employ the first inference method and input the respective received split data into the model A to obtain the inference result output by the model A.

[0363] In some embodiments, the model A is a model associated with the first AI inference service. In some embodiments, the model A can be an AI model, can be an ML model, or can be another model, which is not limited in the present disclosure.

[0364] In some embodiments, in the case that the third node and the first node are deployed on the same device, for example, the third node is an AISF, at this time, the execution subject of step S2109 can be replaced by the AISF.

[0365] In step S2110, the inference node A sends the inference result.

[0366] In some embodiments, the inference node A directly sends the inference result to the AISF through a data connection with the AISF.

[0367] In some embodiments, the AISF receives the inference result. In some embodiments, the AISF receives the inference result through a data connection with the inference node A.

[0368] In some embodiments, in the case that the number of inference nodes A is one, the inference result obtained by the inference node A is the first inference result of the first AI inference service. In this case, the inference result received by the AISF is the first inference result of the first AI inference service.

[0369] In some embodiments, in the case that the number of inference nodes A is multiple and the number of the first AI inference services is one, the inference result obtained by each inference node A is used to determine the first inference result of the first AI inference service. In this case, the inference results received by the AiISF from the multiple inference nodes A are used to determine the first inference result of the first AI inference service.

[0370] In some embodiments, in the case that the number of inference nodes A is multiple and the number of the first AI inference services is multiple, if each inference node A is assigned a complete one or more first AI inference services, the inference result obtained by each inference node A is the first inference result of the one or more first AI inference services, in which case the inference results received by the AiISF from each inference node A are the first inference result of the one or more first AI inference services; if one of the multiple first AI inference services is assigned to multiple inference nodes A for completion, the inference results obtained by the multiple inference nodes A are used to determine the first inference result of the first AI inference service, in which case the inference results received by the AiISF from the multiple inference nodes A are used to determine the first inference result of the first AI inference service.

[0371] In some embodiments, in the case that the third node and the first node are deployed on the same device, for example, the third node is the AiISF, the execution subject of step S2110 can be replaced by the AiISF.

[0372] In step S2111, the AiISF determines the first inference result of the first AI inference service based on the inference results sent by the multiple third nodes.

[0373] In some embodiments, if the multiple inference tasks associated with the first AI inference service are assigned to multiple inference nodes A for execution, and each inference node A only processes a part of the first AI inference service, the AiISF receives the inference results sent by the multiple inference nodes A, and in this case, the AiISF combines the inference results sent by the multiple inference nodes A to obtain the first inference result of the first AI inference service.

[0374] In an example, the number of first AI inference services is 3 (e.g., service 1, service 2 and service 3), and the number of inference nodes A is 2 (e.g., inference node A1 and inference node A2). If the AiISF allocates service 1 to inference node A1, service 3 to inference node A2, splits service 2 into task 1, task 2 and task 3, and allocates task 1 and task 2 to inference node A1, and task 3 to inference node A2. In this case, inference node A1 sends the inference result of service 1, the inference result of task 1 and the inference result of task 2 to the AiISF, and inference node A2 sends the inference result of service 3 and the inference result of task 3 to the AiISF. The AiISF receives the inference result of service 1 and service 3, and merges the inference result of task 1, the inference result of task 2 and the inference result of task 3 to obtain the inference result of service 2.

[0375] In some embodiments, in the case where the number of inference nodes A is one, step S2111 can be omitted, in which case the inference result sent by the inference node A is the first inference result of the first AI inference service.

[0376] In some embodiments, in the case where each of the plurality of inference nodes A is allocated a complete one or more first AI inference services, step S2111 can be omitted, in which case the inference result sent by the plurality of inference nodes A is the first inference result of the plurality of first AI inference services.

[0377] In step S2112, the AiISF sends the first inference result.

[0378] In some embodiments, the terminal A receives the first inference result.

[0379] In some embodiments, in the case where the plurality of inference tasks associated with the first AI inference service are allocated to the plurality of inference nodes A for execution, and each inference node A only processes a part of the first AI inference service, the AiISF merges the inference results sent by the plurality of inference nodes A to obtain the first inference result of the first AI inference service. In this case, the first inference result received by the terminal A is the first inference result obtained by the AiISF through merging.

[0380] In some embodiments, in the case where step S2111 is omitted, the first inference result sent by the AiISF is the inference result sent by the inference node A. In this case, the first inference result received by the terminal A is the inference result sent by the AiISF.

[0381] In some embodiments, in the case where the second node and the first node are deployed on the same device, for example, the second node is the AiISF, step S2112 can be omitted.

[0382] The communication method related to the embodiments of the present disclosure can include at least one of steps S2101 to S2112. For example, step S2101 can be implemented as an independent embodiment. For example, step S2102 can be implemented as an independent embodiment. For example, step S2103 can be implemented as an independent embodiment. For example, step S2104 can be implemented as an independent embodiment. For example, step S2105 can be implemented as an independent embodiment. For example, step S2106 can be implemented as an independent embodiment. For example, step S2107 can be implemented as an independent embodiment. For example, step S2108 can be implemented as an independent embodiment. For example, step S2109 can be implemented as an independent embodiment. For example, step S2110 can be implemented as an independent embodiment. For example, step S2111 can be implemented as an independent embodiment. For example, step S2112 can be implemented as an independent embodiment. For example, step S2101 and step S2104 can be combined to be implemented as an independent embodiment. For example, step S2101, step S2102, step S2104, and step S2110 can be combined to be implemented as an independent embodiment. For example, step S2104, step S2105, and step S2106 can be combined to be implemented as an independent embodiment. For example, step S2104, step S2109, and step S2110 can be combined to be implemented as an independent embodiment, but not limited thereto.

[0383] In some embodiments, steps S2102 to S2112 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0384] In some embodiments, steps S2103, S2105 to S2109, S2111, and S2112 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0385] In some embodiments, steps S2101 to S2103, S2105 to S2108, S2111, and S2112 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0386] In some embodiments, the terms "AI inference service", "AI model inference service", "model inference service", "inference service", and the like can be replaced with each other.

[0387] In some embodiments, the names of information and the like are not limited to the names described in the embodiments, and terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "codebook", "codeword", "codepoint", "bit", "data", "program", "chip", and the like can be replaced with each other.

[0388] In some embodiments, terms such as "carrying", "including", "containing", and the like can be replaced with each other.

[0389] In some embodiments, terms such as "first type of mobility", "second type of mobility", "first mobility", "second mobility", and the like can be replaced with each other, terms such as "downlink", "physical downlink", and the like can be replaced with each other, terms such as "side", "sidelink", "sidelink communication", "direct", "direct link", "direct communication", and the like can be replaced with each other.

[0390] In some embodiments, terms such as "acquiring", "obtaining", "receiving", "transmitting", "bidirectional transmission", "sending and / or receiving" can be replaced with each other, which can be interpreted as receiving from other subjects, acquiring from protocols, acquiring from higher layers, obtaining by processing oneself, and the like.

[0391] In some embodiments, terms such as "sending", "transmitting", "reporting", "issuing", "transmitting", "requesting", "bidirectional transmission", "sending and / or receiving", and the like can be replaced with each other.

[0392] In some embodiments, the terms "certain", "preset", "pre-set", "set", "indicated", "any", "first", and the like can be replaced with each other, and "certain A", "preset A", "pre-set A", "set A", "indicated A", "any A", "first A" can be interpreted as A predetermined in a protocol or the like, or A obtained by setting, configuring, or indicating, or a specific A, any A, or first A, but are not limited thereto.

[0393] In some embodiments, the determination or judgment can be made by a value represented by 1 bit (0 or 1), or by a true or false value (Boolean value) represented by true or false, or by comparison of numerical values (for example, comparison with a predetermined value), but is not limited thereto.

[0394] FIG. 3A is a flow diagram of a method of performing communication by an AiISF according to an embodiment of the present disclosure. As shown in FIG. 3A, the present embodiment relates to a method of performing communication by an AiISF, and the method includes steps S3101 to S3111.

[0395] In step S3101, first information is acquired.

[0396] The optional implementation of step S3101 can be seen in step S2101 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0397] In step S3102, a reasoning node A is determined.

[0398] The optional implementation of step S3102 can be seen in step S2102 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0399] In step S3103, a first AI reasoning service is assigned.

[0400] The optional implementation of step S3103 can be seen in step S2103 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0401] In step S3104, second information is sent.

[0402] The optional implementation of step S3104 can be seen in step S2104 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0403] In step S3105, fifth information is acquired.

[0404] The optional implementation of step S3105 can refer to step S2105 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0405] In step S3106, the sixth information is sent.

[0406] The optional implementation of step S3106 can refer to step S2106 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0407] In step S3107, the fourth information is acquired.

[0408] The optional implementation of step S3107 can refer to step S2107 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0409] In step S3108, inference data is sent.

[0410] The optional implementation of step S3108 can refer to step S2108 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0411] In step S3109, inference results are acquired.

[0412] The optional implementation of step S3109 can refer to step S2110 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0413] In step S3110, based on the inference results sent by the plurality of inference nodes A, a first inference result of the first AI inference service is determined.

[0414] The optional implementation of step S3110 can refer to step S2111 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0415] In step S3111, the first inference result is sent.

[0416] The optional implementation of step S3111 can refer to step S2112 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0417] The communication method related to the embodiments of the present disclosure can include at least one of steps S3101 to S3111. For example, step S3101 can be implemented as an independent embodiment. For example, step S3102 can be implemented as an independent embodiment. For example, a combination of step S3101, step S3102, step S3104, and step S3109 can be implemented as an independent embodiment, but is not limited thereto.

[0418] In some embodiments, steps S3102 to S3111 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0419] In some embodiments, steps S3101, steps S3102 to S3111 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0420] In some embodiments, steps S3103, steps S3105 to S3108, step S3110, step S3111 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0421] FIG. 3B is a flow diagram of a communication method performed by terminal A according to an embodiment of the present disclosure. As shown in FIG. 3B, the embodiments of the present disclosure relate to a communication method performed by terminal A, and the above method includes steps S3201 to S3204.

[0422] In step S3201, first information is transmitted.

[0423] Optional implementation of step S3201 can be referred to step S2101 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0424] In step S3202, sixth information is acquired.

[0425] Optional implementation of step S3202 can be referred to step S2106 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0426] In step S3203, fourth information is transmitted.

[0427] Optional implementation of step S3203 can be referred to step S2107 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0428] In step S3204, a first inference result is acquired.

[0429] The optional implementation of step S3204 can be referred to step S2112 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0430] The communication method related to the embodiments of the present disclosure can include at least one of steps S3201 to S3204. For example, step S3201 can be implemented as an independent embodiment. For example, step S3202 can be implemented as an independent embodiment. For example, the combination of step S3201, step S3203 and step S3204 can be implemented as an independent embodiment, but is not limited thereto.

[0431] In some embodiments, steps S3202, S3203 and S3204 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0432] In some embodiments, steps S3201, S3203 and S3204 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0433] In some embodiments, step S3202 is optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0434] FIG. 3C is a flow diagram of a communication method performed by an inference node A according to an embodiment of the present disclosure. As shown in FIG. 3C, the communication method related to the embodiments of the present disclosure is performed by the inference node A, and the above method includes steps S3301 to S3306.

[0435] In step S3301, a first AI inference service is obtained.

[0436] The optional implementation of step S3301 can be referred to step S2103 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0437] In step S3302, second information is obtained.

[0438] The optional implementation of step S3302 can be referred to step S2104 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0439] In step S3303, fifth information is sent.

[0440] The optional implementation of step S3303 can be referred to step S2105 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0441] In step S3304, inference data is acquired.

[0442] Optional implementation of step S3304 can be referred to step S2108 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0443] In step S3305, AI inference is performed based on the inference data, and inference result is obtained.

[0444] Optional implementation of step S3305 can be referred to step S2109 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0445] In step S3306, the inference result is sent.

[0446] Optional implementation of step S3306 can be referred to step S2110 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0447] The communication method related to the embodiments of the present disclosure can include at least one of steps S3301 to S3306. For example, step S3301 can be implemented as an independent embodiment. For example, step S3302 can be implemented as an independent embodiment. For example, the combination of step S3302, step S3305 and step S3306 can be implemented as an independent embodiment, but not limited thereto.

[0448] In some embodiments, steps S3302 to S3305 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0449] In some embodiments, steps S3301, S3303 to S3306 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0450] In some embodiments, steps S3301, S3303 to S3305 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0451] FIG. 4A is a flow diagram of a communication method performed by an AiISF according to an embodiment of the present disclosure. As shown in FIG. 4A, the communication method related to the embodiments of the present disclosure is performed by an AiISF, and the above method includes steps S4101 to S4104.

[0452] In step S4101, first information is acquired.

[0453] The optional implementation of step S4101 can be referred to step S2101 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, and details are not described herein.

[0454] In step S4102, the inference node A is determined based on the first information.

[0455] The optional implementation of step S4102 can be referred to step S2102 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, and details are not described herein.

[0456] In step S4103, the second information is sent.

[0457] The optional implementation of step S4103 can be referred to step S2104 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, and details are not described herein.

[0458] In step S4104, the inference result is obtained.

[0459] The optional implementation of step S4101 can be referred to step S2110 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, and details are not described herein.

[0460] FIG. 4B is a flow diagram of a method for performing communication by a terminal A according to an embodiment of the present disclosure. As shown in FIG. 4B, the embodiment of the present disclosure relates to a method for performing communication, which is performed by a terminal A, and the above method includes step S4201.

[0461] In step S4201, the first information is sent.

[0462] The optional implementation of step S4201 can be referred to step S2101 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, and details are not described herein.

[0463] FIG. 4C is a flow diagram of a method for performing communication by an inference node A according to an embodiment of the present disclosure. As shown in FIG. 4C, the embodiment of the present disclosure relates to a method for performing communication, which is performed by an inference node A, and the above method includes step S4301 to step S4302.

[0464] In step S4301, the second information is obtained.

[0465] The optional implementation of step S4301 can be referred to step S2104 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, and details are not described herein.

[0466] In step S4302, the inference result is sent.

[0467] An optional implementation of step S4302 can be referred to step S2110 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be described herein again.

[0468] Hereinafter, the technical solutions of the embodiments of the present disclosure are exemplarily described through specific implementations.

[0469] FIG. 5 is a flow diagram of an exemplary implementation of a communication method according to an embodiment of the present disclosure. As shown in FIG. 5, the communication method can exemplarily include steps S501 to S513.

[0470] In step S501, a UE (such as a second node) sends an inference service request to an AiISF (such as a first node) through a 6G RAN.

[0471] In some embodiments, the inference service request can include a UE ID, inference service requirements (such as latency, data rate, etc.), AI model metadata.

[0472] In some embodiments, the inference service requirements can include latency, data rate, etc.

[0473] In some embodiments, the AI model metadata can include an AI model ID, version, size, etc.

[0474] In step S502, the AiISF determines to accept the request and determines an inference method.

[0475] In some embodiments, the inference method can include a central inference method, a FL inference method, a distributed inference method, or other pre-configured / defined methods.

[0476] In some embodiments, the AiISF selects inference nodes based on the selected inference method and assigns a task ID to each inference node.

[0477] In step S503, the AiISF requests to establish a data connection between the AiISF and the inference nodes.

[0478] In some embodiments, the data connection request can include an inference task ID, AI model metadata, QoS requirements, and IP address / port information of the AiISF.

[0479] In step S504, a data connection establishment response is received.

[0480] In some embodiments, the data connection establishment response can include IP address / port information of the inference nodes.

[0481] In step S505, the AiISF sends an inference service response message to the 6G RAN.

[0482] In some embodiments, the inference service corresponding information can comprise QOS information.

[0483] In step S506, based on the message from the AiISF, specific resources are established between the UE and the RAN.

[0484] In some embodiments, a data connection between the UE and the AiISF has already been established.

[0485] In step S507, the UE sends inference input data to the AiISF through the established data connection.

[0486] In step S508, the AiISF decides to split the inference input data to the correct inference node (e.g., the third node).

[0487] In an example, the inference node can comprise an inference node 1 and an inference node 2.

[0488] In step S509, the AiISF sends the inference data to the correct inference node.

[0489] In step S510, the inference node obtains inference results by performing inference on the AI model.

[0490] In step S511, the inference node sends the inference results to the AiISF.

[0491] In step S512, the AiISF merges the inference results of the plurality of inference nodes.

[0492] In step S513, the AiISF sends the final inference results to the UE.

[0493] The embodiments of the present disclosure also propose an apparatus for implementing any of the above methods, for example, an apparatus comprising units or modules for implementing the steps performed by the first node in any of the above methods. For another example, another apparatus is also proposed, comprising units or modules for implementing the steps performed by the second node in any of the above methods. For another example, another apparatus is also proposed, comprising units or modules for implementing the steps performed by the third node in any of the above methods.

[0494] It should be understood that the division of each unit or module in the above apparatus is only a logical function division, and all or part of them can be integrated into a physical entity or physically separated in actual implementation. In addition, the units or modules in the apparatus can be implemented in the form of processor calling software: for example, the apparatus includes a processor connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of each unit or module of the above apparatus, wherein the processor is, for example, a general processor such as a central processing unit (CPU) or a microprocessor, and the memory is a memory within the apparatus or a memory outside the apparatus. Alternatively, the units or modules in the apparatus can be implemented in the form of hardware circuit, and the functions of part or all of the units or modules can be realized by the design of the hardware circuit. The above hardware circuit can be understood as one or more processors; for example, in one implementation, the above hardware circuit is an application-specific integrated circuit (ASIC), and the functions of part or all of the above units or modules are realized by the design of the logical relationship of the elements in the circuit; for example, in another implementation, the above hardware circuit is a programmable logic device (PLD), and a field programmable gate array (FPGA) is taken as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of part or all of the above units or modules. All units or modules of the above apparatus can be implemented in the form of processor calling software, or all units or modules can be implemented in the form of hardware circuit, or part of the units or modules are implemented in the form of processor calling software, and the remaining part is implemented in the form of hardware circuit.

[0495] In embodiments of the present disclosure, the processor is a circuit with signal processing capability. In one implementation, the processor can be a circuit with instruction reading and running capability, such as a CPU, a microprocessor, a graphics processing unit (GPU) (which can also be understood as a microprocessor), a digital signal processor (DSP), or the like. In another implementation, the processor can implement certain functions through a logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or can be reconfigured. For example, the processor is an ASIC or a PLD implemented hardware circuit, such as an FPGA. In the reconfigurable hardware circuit, the processor loads a configuration document to implement the hardware circuit configuration. It can be understood that the processor loads instructions to implement the functions of the above part or all units or modules. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), and the like.

[0496] As shown in FIG. 6A, FIG. 6A is a structural schematic diagram of a first node according to embodiments of the present disclosure. The structure of the first node 6100 can be as shown in FIG. 6A. The first node 6100 includes a first transceiver module 6101 and a first processing module 6102. In some embodiments, the first transceiver module 6101 is configured to receive first information from a second node, the first information being used to request a first AI inference service; the first transceiver module 6101 is further configured to send second information, the second information being used to instruct a third node to provide the first AI inference service; and the first transceiver module 6101 is further configured to receive an inference result associated with the first AI inference service sent by the third node. In some embodiments, the first processing module 6102 is configured to determine the third node based on the first information, the third node being used to provide the first AI inference service.

[0497] In some embodiments, the first transceiver module 6101 described above is configured to perform at least one of the processing steps performed by the AiISF in any of the above methods (for example, steps S3101, steps S3103 to S3109, and step S3111), which will not be described here again.

[0498] In some embodiments, the first transceiver module 6101 described above can include a sending module and / or a receiving module, which can be separate or integrated together. Optionally, the first transceiver module 6101 can be mutually replaced with a transceiver.

[0499] In some embodiments, the first processing module 6102 can be one module, or can include a plurality of sub-modules. Optionally, the plurality of sub-modules respectively perform all or part of the steps required to be performed by the processing module. Optionally, the first processing module 6102 can be mutually replaced with a processor.

[0500] In some embodiments, the first node 6100 can be a core network device, an access network device, etc. For example, the first node 6100 can be an AISF.

[0501] As shown in FIG. 6B, FIG. 6B is a structural schematic diagram of a second node according to an embodiment of the present disclosure. The structure of the second node 6200 can be as shown in FIG. 6B. The second node 6200 includes a second transceiver module 6201. In some embodiments, the second transceiver module 6201 is configured to send first information to a first node, the first information being used to request a first AI inference service, the first AI inference service being provided by a third node, the third node being determined by the first node according to the first information. In some embodiments, the second transceiver module 6201 is configured to perform at least one of the communication steps (for example, steps S3201 to S3204) performed by the network device in any one of the above methods, and details are not repeated here.

[0502] In some embodiments, the second transceiver module 6201 can include a sending module and / or a receiving module, which can be separate or integrated together. Optionally, the second transceiver module 6201 can be mutually replaced with a transceiver.

[0503] In some embodiments, the second node 6200 can be a terminal, an access network device, a core network device, etc.

[0504] As shown in FIG. 6C, FIG. 6C is a structural schematic diagram of a third node according to an embodiment of the present disclosure. The structure of the third node 6300 can be as shown in FIG. 6C. The third node 6300 includes a third transceiver module 6301. In some embodiments, the third transceiver module 6301 is configured to receive second information sent by a first node, the second information being used to indicate that the third node provides a first AI inference service. In some embodiments, the third transceiver module 6301 is configured to perform at least one of the communication steps (for example, steps S3301 to S3304, S3306) performed by the network device in any one of the above methods, and details are not repeated here.

[0505] In some embodiments, the third transceiver module 6301 can include a transmitting module and / or a receiving module, which can be separate or integrated together. Optionally, the third transceiver module 6301 can be mutually replaced with a transceiver.

[0506] In some embodiments, the third node 6300 can be a terminal, an access network device, a core network device, etc.

[0507] FIG. 7A is a structural schematic diagram of a communication device according to an embodiment of the present disclosure. The communication device 7100 can be an access network device, a terminal (e.g., a user equipment, etc.), a core network device, a chip, a chip system, or a processor supporting the communication device to implement any of the above methods, or a chip, a chip system, or a processor supporting the terminal to implement any of the above methods. The communication device 7100 can be used to implement the methods described in the above method embodiments, and details can be referred to the descriptions in the above method embodiments.

[0508] As shown in FIG. 7A, the communication device 7100 includes one or more processors 7101. The processor 7101 can be a general-purpose processor or a special-purpose processor, for example, a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control a network node (e.g., a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.), execute programs, and process data of the programs. Optionally, the communication device 7100 is configured to perform any of the above methods. Optionally, the one or more processors 7101 are configured to invoke instructions to cause the communication device 7100 to perform any of the above methods.

[0509] In some embodiments, the communication device 7100 further includes one or more transceivers 7102. When the communication device 7100 includes the one or more transceivers 7102, the transceiver 7102 performs at least one of the communication steps (e.g., steps S2101, steps S2103 to S2108, step S2110, step S2101, step S2112, but not limited to) in the above methods, and the processor 7101 performs at least one of the other steps (e.g., steps S2102, step S2109, step S2111, but not limited to). In an optional embodiment, the transceiver can include a receiver and / or a transmitter, which can be separate or integrated together. Optionally, the terms of transceiver, transceiving unit, transceiver, transceiving circuit, interface circuit, interface, etc. can be mutually replaced, and the terms of transmitter, transmitting unit, transmitter, transmitting circuit, etc. can be mutually replaced, and the terms of receiver, receiving unit, receiver, receiving circuit, etc. can be mutually replaced.

[0510] In some embodiments, the communication device 7100 further includes one or more memories 7103 for storing data. Alternatively, all or part of the memories 7103 can be external to the communication device 7100. In optional embodiments, the communication device 7100 can include one or more interface circuits 7104. Optionally, the interface circuit 7104 is connected with the memory 7103, and the interface circuit 7104 can be used to receive data from the memory 7103 or other devices, and can be used to send data to the memory 7103 or other devices. For example, the interface circuit 7104 can read data stored in the memory 7103 and send the data to the processor 7101.

[0511] The communication device 7100 in the above embodiments can be a terminal, an access network device, a core network device, etc., but the scope of the communication device 7100 described in the present disclosure is not limited thereto, and the structure of the communication device 7100 can not be limited by FIG. 7A. The access network device can be a standalone device or can be part of a larger device. For example, the terminal can be: (1) a standalone integrated circuit (IC), or a chip, or a chip system or subsystem; (2) a set of one or more ICs, which can optionally include a storage component for storing data, programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, a terminal device, a smart terminal device, a cellular phone, a wireless device, a handset, a mobile unit, a vehicle-mounted device, a network device, a cloud device, an artificial intelligence device, etc.; (6) other devices, etc.

[0512] FIG. 7B is a structural schematic diagram of a chip according to an embodiment of the present disclosure. For the case where the communication device 7100 is a chip or a chip system, the structure of the chip 7200 can be as shown in FIG. 7B, but is not limited thereto.

[0513] The chip 7200 includes one or more processors 7201. The chip 7200 is configured to execute any of the above methods.

[0514] In some embodiments, the chip 7200 further includes one or more interface circuits 7202. Optionally, the terms interface circuit, interface, and transceiver pin can be replaced with each other. In some embodiments, the chip 7200 further includes one or more memories 7203 for storing data. Optionally, all or part of the memories 7203 can be external to the chip 7200. Optionally, the interface circuit 7202 is connected with the memory 7203, and the interface circuit 7202 can be used to receive data from the memory 7203 or other devices, and can be used to send data to the memory 7203 or other devices. For example, the interface circuit 7202 can read data stored in the memory 7203 and send the data to the processor 7201.

[0515] In some embodiments, the interface circuit 7202 performs at least one of the communication steps such as transmitting and / or receiving in the above method. The interface circuit 7202 performing the communication steps such as transmitting and / or receiving in the above method refers to, for example, the interface circuit 7202 performing data interaction between the processor 7201, the chip 7200, the memory 7203, or a transceiver device.

[0516] The modules and / or devices described in each of the embodiments of the virtual device, the physical device, the chip, etc. can be combined or separated according to actual conditions. Alternatively, part or all of the steps can also be executed by a plurality of modules and / or devices in cooperation, which is not limited here.

[0517] The embodiments of the present disclosure further provide a storage medium having instructions stored thereon, which, when executed on the communication device 7100, cause the communication device 7100 to perform any of the above methods. Alternatively, the storage medium is an electronic storage medium. Alternatively, the storage medium is a computer-readable storage medium, but is not limited to this, and it can also be a storage medium readable by other devices. Alternatively, the storage medium can be a non-transitory storage medium, but is not limited to this, and it can also be a transitory storage medium.

[0518] The embodiments of the present disclosure further provide a computer program product, which, when executed by the communication device 7100, causes the communication device 7100 to perform any of the above methods.

[0519] The embodiments of the present disclosure further provide a computer program, which, when executed on a computer, causes the computer to perform any of the above methods.

[0520] Other embodiments of the present disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the embodiments disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure following, in general, the principles of the present disclosure and including such features that are evident to those skilled in the art or are known in the art and can be used in combination with the present disclosure. The specification and examples are to be considered exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0521] It should be understood that the present disclosure is not limited to the precise structures herein described and illustrated above and that various modifications and changes in the embodiments can be made by those skilled in the art without departing from the scope of the present disclosure. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A communication method, performed by a first node, comprising: receiving first information from a second node, the first information being used for requesting a first artificial intelligence (AI) inference service; determining, based on the first information, a third node, the third node being used for providing the first AI inference service; sending second information, the second information being used for indicating that the third node provides the first AI inference service; receiving an inference result associated with the first AI inference service and sent by the third node.

2. The method of claim 1, wherein, The third node is multiple, and the method further comprises: merging inference results sent by multiple third nodes to obtain a first inference result of the first AI inference service; and sending the first inference result to the second node.

3. The method of claim 1 or 2, wherein, The first information comprises at least one of: first indication information used for indicating the second node requesting the first AI inference service; second indication information used for indicating a quality requirement of the first AI inference service; third indication information used for indicating a model requirement of the first AI inference service; fourth indication information used for indicating a type of the first AI inference service; fifth indication information used for indicating a location of the second node; sixth indication information used for indicating a service area of the first AI inference service.

4. The method according to any one of claims 1 to 3, wherein, The determining, based on the first information, of the third node comprises: determining the third node from a plurality of fourth nodes according to the first information and third information associated with the plurality of fourth nodes.

5. The method of claim 4, wherein, The third information comprises at least one of: seventh indication information used for indicating support capabilities of the fourth node for different types of AI inference services; eighth indication information used for indicating support capabilities of the fourth node for AI models; ninth indication information used for indicating resources of the fourth node for processing the AI inference service; tenth indication information used for indicating a location of the fourth node.

6. The method according to any one of claims 1 to 3, wherein, The determining, based on the first information, of the third node comprises: determining the third node according to the first information and preconfigured eleventh indication information; wherein the eleventh indication information is used for indicating a mapping relationship between a type of AI inference service and a third node.

7. The method according to any one of claims 1 to 6, wherein, The method further comprises one of: determining, according to the first information, a first inference method adopted by the first AI inference service; and determining, according to the first information and preconfigured twelfth indication information, a first inference method adopted by the first AI inference service; wherein the twelfth indication information is used for indicating a mapping relationship between a type of AI inference service and an inference method.

8. The method of claim 7, wherein, The second information is further used for indicating that the third node provides the first AI inference service by adopting the first inference method.

9. The method according to any one of claims 1 to 8, wherein, The method further comprises: receiving fourth information from the second node, the fourth information being used for indicating inference data associated with the first AI inference service; and sending the inference data to the third node.

10. The method of claim 9, wherein, The third node is multiple, and the method further comprises: splitting the inference data to obtain a plurality of split data; and The plurality of split data are respectively sent to a plurality of third nodes, and the split data are used for the third nodes to perform an inference task associated with the first AI inference service.

11. The method according to any one of claims 1 to 10, wherein, The number of the third nodes is a plurality, and the method further comprises: determining a plurality of inference tasks associated with the first AI inference service; allocating the plurality of inference tasks to the plurality of third nodes.

12. The method of any one of claims 1 to 11, wherein the second information comprises at least one of: the first information; thirteenth indication information indicating the first AI inference service; fourteenth indication information indicating a plurality of inference tasks associated with the first AI inference service; fifteenth indication information indicating a first inference method adopted by the first AI inference service.

13. The method of claim 12, wherein, The second information is further used to request the third nodes to establish a data connection with the first node.

14. The method of claim 13, wherein, The method further comprises: receiving fifth information sent by the third nodes, the fifth information being used to indicate an access address of the third nodes; sending sixth information to the second nodes, the sixth information comprising the fifth information.

15. The method according to any one of claims 1 to 14, wherein, The first node and the third nodes are deployed on a same device.

16. A communication method, performed by a second node, the method comprising: sending first information to a first node, the first information being used to request a first artificial intelligence (AI) inference service, the first AI inference service being provided by third nodes, the third nodes being determined by the first node according to the first information.

17. The method of claim 16, wherein, The third nodes are a plurality, and the method comprises: receiving first inference results sent by the first node, the first inference results being obtained by the first node merging inference results sent by a plurality of third nodes.

18. The method of claim 16 or 17, wherein, The first information comprises at least one of: first indication information indicating a second node requesting the first AI inference service; second indication information indicating a quality requirement of the first AI inference service; third indication information indicating a model requirement of the first AI inference service; fourth indication information indicating a type of the first AI inference service; fifth indication information indicating a location of the second node; sixth indication information indicating a service area of the first AI inference service.

19. The method of any one of claims 16 to 18, wherein, The third nodes are determined by the first node from a plurality of fourth nodes according to the first information and third information associated with the fourth nodes.

20. The method of claim 19, wherein, The third information comprises at least one of: seventh indication information indicating support capabilities of the fourth nodes for different types of AI inference services; eighth indication information indicating support capabilities of the fourth nodes for AI models; ninth indication information indicating resources of the fourth nodes for processing the AI inference services; tenth indication information indicating locations of the fourth nodes.

21. The method of any one of claims 16 to 20, wherein, The third nodes are determined by the first node according to the first information and preconfigured eleventh indication information; The eleventh indication information is used to indicate a mapping relationship between a type of AI inference service and a third node.

22. The method of any one of claims 16 to 21, wherein, The first inference method is obtained according to the first information, or the first inference method is obtained according to the first information and preconfigured twelfth indication information, the twelfth indication information being used to indicate a mapping relationship between a type of the AI inference service and an inference method.

23. The method of claim 22, wherein, The first inference method is an inference method used by the third node when providing the first AI inference service.

24. The method of any one of claims 16 to 23, wherein, The method further includes: sending fourth information to the first node, the fourth information being used to indicate inference data associated with the first AI inference service.

25. The method of any one of claims 16 to 24, wherein, The method further includes: receiving sixth information sent by the first node, the sixth information including fifth information received by the first node from the third node, the fifth information being used to indicate an access address of the third node.

26. The method of any one of claims 16 to 25, wherein, The first node and the third node are deployed on a same device. 27.A communication method, performed by a third node, the method comprising: receiving second information sent by a first node, the second information being used to indicate that the third node provides a first artificial intelligence (AI) inference service; sending, to the first node, an inference result associated with the first AI inference service.

28. The method of claim 27, wherein, The second information is further used to indicate that the third node provides the first AI inference service by using a first inference method.

29. The method of claim 28, wherein, The first inference method is obtained by the first node according to first information, or the first inference method is obtained by the first node according to the first information and preconfigured twelfth indication information, the first information being information received by the first node from a second node, the first information being used to request the first AI inference service, and the twelfth indication information being used to indicate a mapping relationship between a type of the AI inference service and an inference method. The method further includes:

30. The method of any one of claims 27 to 29, wherein, receiving inference data sent by the first node, the inference data being indicated by fourth information received by the first node from a second node, the second node being a node requesting a first AI inference service. The third node is multiple, and the method further includes:

31. The method of claim 30, wherein, receiving split data sent by the first node, the split data being obtained by splitting, by the first node, the inference data. The method further includes: receiving an inference task sent by the first node, the inference task being obtained by the first node after distributing multiple inference tasks associated with the first AI inference service.

32. The method of any one of claims 27 to 31, wherein, 33.The method of any one of claims 27-32, the second information including at least one of: the first information; thirteenth indication information used to indicate the first AI inference service; fourteenth indication information used to indicate multiple inference tasks associated with the first AI inference service; fifteenth indication information used to indicate a first inference method used by the first AI inference service. The second information is further used to request that a data connection be established between the third node and the first node. The method further includes:

34. The method of any one of claims 27 to 33, wherein, sending fifth information to the first node, the fifth information being used to indicate an access address of the third node.

35. The method of any one of claims 27 to 34, wherein, ​ ​ 36. The method of claim 29 or 30, wherein, The first node and the third node are deployed in a same device.

37. A first node, comprising: a first receiving module configured to receive first information from a second node, the first information being used to request a first artificial intelligence (AI) inference service; a first processing module configured to determine, based on the first information, a third node, the third node being used to provide the first AI inference service; a first sending module configured to send second information, the second information being used to instruct the third node to provide the first AI inference service; a first receiving module configured to receive an inference result associated with the first AI inference service and sent by the third node.

38. A second node, comprising: a second sending module configured to send first information to a first node, the first information being used to request a first artificial intelligence (AI) inference service, the first AI inference service being provided by a third node, the third node being determined by the first node according to the first information.

39. A third node, comprising: a third receiving module configured to receive second information sent by a first node, the second information being used to instruct the third node to provide the first artificial intelligence (AI) inference service; a third sending module configured to send an inference result associated with the first AI inference service to the first node.

40. A communication device, comprising: at least one processor; wherein the communication device is configured to perform the communication method of any one of claims 1 to 36.

41. A communication system, comprising a first node, a second node and a third node; the first node is configured to implement the communication method of any one of claims 1 to 15; the second node is configured to implement the communication method of any one of claims 16 to 26; and the third node is configured to implement the communication method of any one of claims 27 to 36.

42. A storage medium storing instructions, wherein, the instructions, when executed on the communication device, cause the communication device to perform the communication method of any one of claims 1 to 36.

43. A computer program product, comprising a computer program which, when executed on a communication device, causes the communication device to perform the communication method of any one of claims 1 to 36.

Images