Artificial intelligence (AI) task processing method and device

Abstract

The embodiment of the invention discloses an AI task processing method and device, and the method comprises the steps that a first AI network element receives an AI service request message sent by an AMF network element, and the AI service request message is used for indicating an AI service needing to be provided; determining at least one AI task according to the AI service request message; determining a first processing parameter of the first AI network element and a second processing parameter of the second AI network element; and determining a first task executed by the first AI network element and / or a second task executed by the second AI network element in the AI task according to the AI task, the first processing parameter and the second processing parameter. Therefore, the first AI network element determines the first task executed by the first AI network element and / or the second task executed by the second AI network element in the AI tasks, classification scheduling can be carried out on the AI tasks, resource allocation is carried out according to scheduling, expenses can be reduced, resources are reasonably allocated, and AI services can be carried out more efficiently and flexibly.

Description

Artificial intelligence (AI) task processing method and device Technical Field

[0001] The present disclosure relates to the field of communication technologies, and in particular to an AI task processing method and device. Background Art

[0002] Among related technologies, networks have adopted many automated means to improve operation and maintenance efficiency. Among them, AI (Artificial Intelligence) can help networks achieve a higher level of autonomy and has become a core technology for future communications.

[0003] However, due to the relatively late application of AI technology in communication networks, the AI ​​functions in the network are simply superimposed on the network process and are plug-in applications. As AI functions continue to increase, the overhead required for the network to implement different AI functions is very high, which is a problem that needs to be solved urgently.

[0004] Summary of the Invention

[0005] Embodiments of the present disclosure provide an AI task processing method and apparatus. In this method, a first AI network element determines a first task to be executed by the first AI network element and / or a second task to be executed by a second AI network element. This method enables categorized scheduling of AI tasks and resource allocation based on the scheduling. This method reduces overhead and rationally allocates resources, enabling more efficient and flexible AI services.

[0006] In a first aspect, an embodiment of the present disclosure provides an AI task processing method, which is executed by a first AI network element, including: receiving an AI service request message sent by an access and mobility management function (AMF) network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided; determining at least one AI task based on the AI ​​service request message; determining a first processing parameter of the first AI network element and a second processing parameter of the second AI network element; and determining, based on the AI ​​task, the first processing parameter and the second processing parameter, a first task executed by the first AI network element and / or a second task executed by the second AI network element in the AI ​​task.

[0007] In this technical solution, a first AI network element receives an AI service request message sent by an AMF network element, where the AI ​​service request message indicates the AI ​​service to be provided; determines at least one AI task based on the AI ​​service request message; determines a first processing parameter for the first AI network element and a second processing parameter for the second AI network element; and determines, based on the AI ​​task, the first processing parameter, and the second processing parameter, which of the first and second AI network elements in the AI ​​task should be executed by the first AI network element and / or the second task should be executed by the second AI network element. Thus, the first AI network element determines the first task to be executed by the first AI network element and / or the second task to be executed by the second AI network element in the AI ​​task, enabling classified scheduling of AI tasks and resource allocation based on the scheduling. This reduces overhead, rationalizes resource allocation, and enables more efficient and flexible AI services.

[0008] In a second aspect, an embodiment of the present disclosure provides another AI task processing method, which is executed by an AMF network element, including: receiving an AI service establishment request message sent by a terminal device, wherein the AI ​​service request message is used to indicate the AI ​​service required by the terminal device; sending an AI service request message to a first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service that needs to be provided, and the AI ​​service request message is used by the first AI network element to determine at least one AI task based on the AI ​​service request message, determine a first processing parameter of the first AI network element, and a second processing parameter of the second AI network element, and determine, based on the AI ​​task, the first processing parameter and the second processing parameter, the first task executed by the first AI network element and / or the second task executed by the second AI network element in the AI ​​task.

[0009] In a third aspect, an embodiment of the present disclosure provides another AI task processing method, which is executed by a second AI network element, including: receiving a second task sent by a first AI network element, wherein the second AI task is determined by the first AI network element based on the AI ​​task, the determined first processing parameter of the first AI network element, and the determined second processing parameter of the second AI network element, and is executed by the second AI network element and sent to the second AI network element. The AI ​​task is determined by the first AI network element based on an AI service request message sent by an AMF network element, and the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0010] In a fourth aspect, an embodiment of the present disclosure provides a communication device that implements some or all of the functions of the first AI network element in the method described in the first aspect above. For example, the functions of the communication device may include the functions of some or all of the embodiments of the present disclosure, or may include the functions of implementing any one of the embodiments of the present disclosure separately. The functions may be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more units or modules corresponding to the above functions.

[0011] In one implementation, the communication device may include a transceiver module and a processing module, wherein the processing module is configured to support the communication device in performing the corresponding functions of the above-mentioned method. The transceiver module is used to support communication between the communication device and other devices. The communication device may also include a storage module, which is coupled to the transceiver module and the processing module and stores computer programs and data necessary for the communication device.

[0012] As an example, the processing module may be a processor, the transceiver module may be a transceiver or a communication interface, and the storage module may be a memory.

[0013] In one implementation, the communication device includes: a transceiver module configured to receive an AI service request message sent by an access and mobility management function (AMF) network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided; a processing module configured to determine at least one AI task based on the AI ​​service request message; the processing module is further configured to determine a first processing parameter of a first AI network element and a second processing parameter of a second AI network element; the processing module is further configured to determine, in the AI ​​task, a first task performed by the first AI network element and / or a second task performed by the second AI network element based on the AI ​​task, the first processing parameter, and the second processing parameter.

[0014] In a fifth aspect, an embodiment of the present disclosure provides another communication device, which has some or all of the functions of the AMF network element in the method example described in the second aspect above. For example, the functions of the communication device may have some or all of the functions in the embodiments of the present disclosure, or may have the functions of implementing any one of the embodiments of the present disclosure separately. The functions can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more units or modules corresponding to the above functions.

[0015] In one implementation, the communication device may include a transceiver module and a processing module, the processing module being configured to support the communication device in performing the corresponding functions of the above-described method. The transceiver module is configured to support communication between the communication device and other devices. The communication device may also include a storage module, coupled to the transceiver module and the processing module, which stores computer programs and data necessary for the communication device.

[0016] In one implementation, the communication device includes: a transceiver module configured to receive an AI service establishment request message sent by a terminal device, wherein the AI ​​service request message is used to indicate the AI ​​service required by the terminal device; the transceiver module is further configured to send an AI service request message to a first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided, and the AI ​​service request message is used by the first AI network element to determine at least one AI task based on the AI ​​service request message, determine a first processing parameter of the first AI network element, and a second processing parameter of a second AI network element, and determine, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task.

[0017] In a sixth aspect, an embodiment of the present disclosure provides another communication device, which has some or all of the functions of the second AI network element in the method example described in the second aspect above. For example, the functions of the communication device may have the functions of some or all of the embodiments in this disclosure, or may have the functions of implementing any one of the embodiments in this disclosure separately. The functions can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more units or modules corresponding to the above functions.

[0018] In one implementation, the communication device may include a transceiver module and a processing module, the processing module being configured to support the communication device in performing the corresponding functions of the above-described method. The transceiver module is configured to support communication between the communication device and other devices. The communication device may also include a storage module, coupled to the transceiver module and the processing module, which stores computer programs and data necessary for the communication device.

[0019] In one implementation, the communication device includes: a transceiver module configured to receive a second task sent by a first AI network element, wherein the second AI task is determined by the first AI network element based on the AI ​​task, the determined first processing parameter of the first AI network element, and the determined second processing parameter of the second AI network element, and is executed by the second AI network element and sent to the second AI network element. The AI ​​task is determined by the first AI network element based on the AI ​​service request message sent by the AMF network element, and the AI ​​service request message is used to indicate the AI ​​service to be provided.

[0020] In a seventh aspect, an embodiment of the present disclosure provides a communication device, which includes a processor. When the processor calls a computer program in a memory, it executes the method described in the first aspect above.

[0021] In an eighth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor. When the processor calls a computer program in a memory, the method described in the second aspect is executed.

[0022] In a ninth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor. When the processor calls a computer program in a memory, the method described in the third aspect is executed.

[0023] In the tenth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and a memory, in which a computer program is stored; the processor executes the computer program stored in the memory so that the communication device executes the method described in the first aspect above.

[0024] In the eleventh aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and a memory, in which a computer program is stored; the processor executes the computer program stored in the memory so that the communication device executes the method described in the second aspect above.

[0025] In the twelfth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and a memory, in which a computer program is stored; the processor executes the computer program stored in the memory so that the communication device executes the method described in the third aspect above.

[0026] In a thirteenth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and an interface circuit, wherein the interface circuit is used to receive code instructions and transmit them to the processor, and the processor is used to run the code instructions to enable the device to execute the method described in the first aspect above.

[0027] In a fourteenth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and an interface circuit, wherein the interface circuit is used to receive code instructions and transmit them to the processor, and the processor is used to run the code instructions to enable the device to execute the method described in the second aspect above.

[0028] In a fifteenth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and an interface circuit, wherein the interface circuit is used to receive code instructions and transmit them to the processor, and the processor is used to run the code instructions to enable the device to execute the method described in the third aspect above.

[0029] In the sixteenth aspect, an embodiment of the present disclosure provides a communication system, which includes the communication device described in the fourth aspect, the communication device described in the fifth aspect, and the communication device described in the sixth aspect; or, the system includes the communication device described in the seventh aspect, the communication device described in the eighth aspect, and the communication device described in the ninth aspect; or, the system includes the communication device described in the tenth aspect, the communication device described in the eleventh aspect, and the communication device described in the twelfth aspect; or, the system includes the communication device described in the thirteenth aspect, the communication device described in the fourteenth aspect, and the communication device described in the fifteenth aspect.

[0030] In a seventeenth aspect, an embodiment of the present invention provides a computer-readable storage medium for storing instructions for the first AI network element. When the instructions are executed, the first AI network element executes the method described in the first aspect.

[0031] In the eighteenth aspect, an embodiment of the present invention provides a readable storage medium for storing instructions used for the above-mentioned AMF network element. When the instructions are executed, the AMF network element executes the method described in the above-mentioned second aspect.

[0032] In a nineteenth aspect, an embodiment of the present invention provides a readable storage medium for storing instructions for the second AI network element. When the instructions are executed, the second AI network element executes the method described in the third aspect.

[0033] In a twentieth aspect, the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method described in the first aspect above.

[0034] In a twenty-first aspect, the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method described in the second aspect above.

[0035] In a twenty-second aspect, the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method described in the third aspect above.

[0036] In a twenty-third aspect, the present disclosure provides a chip system comprising at least one processor and an interface for supporting a first AI network element in implementing the functions described in the first aspect, such as determining or processing at least one of the data and information described in the aforementioned method. In one possible design, the chip system further comprises a memory for storing computer programs and data necessary for the first AI network element. The chip system may consist of a chip or may include a chip and other discrete components.

[0037] In a twenty-fourth aspect, the present disclosure provides a chip system comprising at least one processor and an interface for supporting an AMF network element in implementing the functions described in the second aspect, such as determining or processing at least one of the data and information described in the aforementioned method. In one possible design, the chip system further comprises a memory for storing computer programs and data necessary for the AMF network element. The chip system may consist of a chip alone or may include a chip and other discrete components.

[0038] In aspect 25, the present disclosure provides a chip system comprising at least one processor and an interface for supporting a second AI network element in implementing the functions described in aspect 2, such as determining or processing at least one of the data and information described in the aforementioned method. In one possible design, the chip system further comprises a memory for storing computer programs and data necessary for the second AI network element. The chip system may consist of a chip or may include a chip and other discrete components.

[0039] In a twenty-sixth aspect, the present disclosure provides a computer program which, when executed on a computer, enables the computer to execute the method described in the first aspect above.

[0040] In aspect 27, the present disclosure provides a computer program which, when executed on a computer, enables the computer to execute the method described in aspect 2 above.

[0041] In a twenty-eighth aspect, the present disclosure provides a computer program which, when executed on a computer, enables the computer to execute the method described in the third aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the background technology, the drawings required for use in the embodiments of the present disclosure or the background technology will be described below.

[0043] FIG1 is an architecture diagram of a communication system provided by an embodiment of the present disclosure;

[0044] FIG2 is a schematic diagram of a system architecture provided by an embodiment of the present disclosure;

[0045] FIG3 is a flowchart of an AI task processing method provided by an embodiment of the present disclosure;

[0046] FIG4 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0047] FIG5 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0048] FIG6 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0049] FIG7 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0050] FIG8 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0051] FIG9 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0052] FIG10 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0053] FIG11 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0054] FIG12 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0055] FIG13 is a flowchart of another AI task processing method provided by an embodiment of the present disclosure;

[0056] FIG14 is a structural diagram of a communication device provided by an embodiment of the present disclosure;

[0057] FIG15 is a structural diagram of another communication system provided by an embodiment of the present disclosure;

[0058] FIG16 is a structural diagram of another communication device provided in an embodiment of the present disclosure;

[0059] FIG17 is a structural diagram of a chip provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] In order to better understand the AI ​​task processing method and device disclosed in the embodiments of the present disclosure, the communication system to which the embodiments of the present disclosure are applicable is first described below.

[0061] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, unless otherwise indicated, like numbers in different figures represent like or similar elements. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0062] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a," "an," "the," and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0063] It should be understood that although the terms first, second, third, etc. may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, for example, the term "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining."

[0064] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0065] Please refer to Figure 1, which is a schematic diagram of a communication system provided by an embodiment of the present disclosure. As shown in Figure 1, the communication system may include but is not limited to a (radio) access network ((radio) access network, (R)AN), a terminal device and a core network device. Access network devices communicate with each other via wired or wireless means, for example, communicating with each other through the Xn interface in Figure 1. The access network device may cover one or more cells, such as: access network device 1 covers cell 1.1 and cell 1.2, and access network device 2 covers cell 2.1. The terminal device may reside in the access network device in one of the cells and be in a connected state. Furthermore, the terminal device may be converted from a connected state to an inactive state, that is, to a non-connected state, through an RRC release process. The terminal device in a non-connected state may reside in the original cell and perform uplink transmission and / or downlink transmission with the access network device in the original cell according to the transmission parameters of the terminal device in the original cell. A terminal device in a non-connected state may also move to a new cell and perform uplink transmission and / or downlink transmission with an access network device of the new cell according to the transmission parameters of the terminal device in the new cell.

[0066] It should be noted that Figure 1 is only an exemplary framework diagram. The number of nodes, number of cells, and terminal device states included in Figure 1 are not limited. In addition to the functional nodes shown in Figure 1, other nodes may also be included, such as gateway devices, application servers, etc., without limitation. Access network devices communicate with core network devices via wired or wireless means, such as through next-generation (NG) interfaces.

[0067] Among them, the terminal device is an entity on the user side for receiving or transmitting signals, such as a mobile phone. The terminal device can also be called a terminal device (terminal), user equipment (UE), mobile station (MS), mobile terminal device (MT), etc. The terminal device can be a car with communication function, a smart car, a mobile phone, a wearable device, a tablet computer (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 (industrial control), a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid (smart grid), a wireless terminal device in transportation safety (transportation safety), a wireless terminal device in smart city (smart city), a wireless terminal device in smart home (smart home), etc. The embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the terminal device.

[0068] A (radio) access network (R)AN) is used to provide network access for authorized terminal devices in a specific area and can use transmission tunnels of varying quality based on the terminal device's level and service requirements. For example, the (R)AN manages wireless resources, provides access services to terminal devices, and forwards control information and / or data between the terminal device and the core network (CN). The access network device in the embodiments of the present disclosure is a device that provides wireless communication capabilities for terminal devices and can also be referred to as a network device. For example, the access network equipment may include: the next generation node basestation (gNB) in the 5G system, the evolved node B (eNB) in the long term evolution (LTE), the radio network controller (RNC), the node B (NB), the base station controller (BSC), the base transceiver station (BTS), the home base station (for example, home evolved nodeB, or home node B, HNB), the base band unit (BBU), the transmission point (TRP), the transmitting point (TP), the small base station device (pico), the mobile switching center, or the network equipment in the future network. It is understandable that the embodiments of the present disclosure do not limit the specific type of access network equipment. In systems with different wireless access technologies, the names of devices with access network device functions may be different.

[0069] The core network device may include an AMF and / or a location management function network element. Optionally, the location management function network element includes a location server, which may be implemented as any one of the following: LMF (Location Management Function, location management network element), E-SMLC (Enhanced Serving Mobile Location Centre, enhanced serving mobile location centre), SUPL (Secure User Plane Location, secure user plane location), SUPL SLP (SUPL Location Platform, secure user plane location platform).

[0070] To facilitate understanding of the network architecture of the communication system, please refer to Figure 2, which is a schematic diagram of a network architecture provided by an embodiment of the present disclosure. As shown in Figure 2, the network architecture includes AMF network elements, UDM network elements, AUSF network elements, UPF network elements, UDR network elements, PCF network elements, NRF network elements, AI0 network elements, AI1 network elements...AIN network elements.

[0071] Among them, the access and mobility management function (AMF) network element is mainly used for mobility management and access management, etc., and can be used to implement other functions of the mobility management entity (MME) function except session management, such as legal monitoring and access authorization / authentication. It can be understood that the AMF network function is hereinafter referred to as AMF. In the embodiment of the present disclosure, the AMF may include an initial AMF (initialAMF), an old AMF (oldAMF) and a target AMF (targetAMF). For example, the initial AMF can be understood as the first AMF to process the UE registration request in this registration. The initial AMF is selected by the (R)AN, but the initial AMF may not necessarily serve the UE. The old AMF can be understood as the AMF that served the UE when the UE last registered with the network. The target AMF can be understood as the AMF that serves the UE after the UE re-registers.

[0072] Session management function (SMF) network element: mainly used for session management, UE Internet Protocol (IP) address allocation and management, etc.

[0073] User Plane Function (UPF) network element: Also known as the user plane gateway. It can be used for packet routing and forwarding, or for quality of service (QoS) processing of user plane data. User data can be connected to the data network (DN) through this network element.

[0074] Data Network (DN): A network used to transmit data, such as a carrier's service network, the Internet, or a third-party service network.

[0075] Authentication server function (AUSF) network element: mainly used for user authentication, etc.

[0076] Network exposure function (NEF) network element: used to securely expose services and capabilities provided by 3GPP network functions to the outside world.

[0077] Network function (NF) repository function (NRF) network element: used to store descriptions of network function entities and the services they provide, and to support service discovery, network element entity discovery, etc.

[0078] Policy control function (PCF) network element: A unified policy framework used to guide network behavior and provide policy rule information to control plane function network elements (such as AMF, SMF network elements, etc.).

[0079] Unified data management (UDM) network element: used to process user identification, access authentication, registration, or mobility management.

[0080] In this network architecture, the N1 interface is the interface between the terminal device and the AMF network element. The N2 interface is the interface between the RAN and the AMF network element, used for sending non-access stratum (NAS) messages, etc. The N3 interface is the interface between the (R)AN and the UPF entity, used for transmitting user plane data, etc. The N4 interface is the interface between the SMF entity and the UPF entity, used to transmit information such as tunnel identification information of the N3 connection, data cache indication information, and downlink data notification messages. The N6 interface is the interface between the UPF entity and the DN, used for transmitting user plane data, etc.

[0081] It is understood that the terms introduced above may have different names in different fields or different standards, and therefore the names shown above should not be understood as limiting the embodiments of the present disclosure. The above-mentioned network functions or functions can be network elements in hardware devices, software functions running on dedicated hardware, or virtualized functions instantiated on a platform (e.g., a cloud platform).

[0082] It should be noted that the network elements involved in the embodiments of the present disclosure may also be referred to as functional devices or functions or entities or functional entities. For example, the access and mobility management network element may also be referred to as access and mobility management functional devices or access and mobility management functional entities or access and mobility management functional entities. The names of the various functional devices are not limited in this disclosure. Those skilled in the art may replace the names of the above functional devices with other names while performing the same functions, which all fall within the scope of protection of this disclosure. The above functional devices may be network elements in hardware devices, software functions running on dedicated hardware, or virtualized functions instantiated on a platform (e.g., a cloud platform).

[0083] It can be understood that the communication system and network architecture described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Ordinary technicians in this field can know that with the evolution of the system architecture and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0084] The following is a detailed introduction to the AI ​​task processing method and device provided by the present disclosure with reference to the accompanying drawings.

[0085] Among related technologies, AI will become one of the core technologies of future communications. Typical application scenarios for 6G (6th Generation) and AI overlap by over 80%, demonstrating a deep integration of the two. Furthermore, the massive coverage of 6G networks will provide a ubiquitous carrier space for AI, resolving the significant pain point of a lack of carriers and channels for AI implementation and significantly promoting the development and prosperity of the AI ​​industry.

[0086] Relevant technologies have already adopted numerous automated approaches to improve operational efficiency in network planning, construction, maintenance, and optimization. However, the overall level of network autonomy is still low, leaving significant room for improvement. Software-Defined Network (SDN) and Network Function Virtualization (NFV) architectures have enabled networks to achieve both high flexibility and increased complexity. This requires more intelligent approaches to factors such as network resource allocation, transmission paths, and optimization algorithm design. AI technology can help networks achieve a higher level of autonomy, reducing costs and increasing efficiency. Due to the relatively recent application of AI technology in communications networks, existing intelligent network applications optimize and transform traditional network architectures, generally acting as add-on applications. The lack of a common AI workflow and unified technical framework has led to fragmented network AI application scenarios and siloed R&D. Network AI functions are simply overlaid on existing network processes, and cross-domain and cross-layer intelligent application collaboration is difficult. The Network Data Analytics Function (NWDAF) network function collects data, performs analysis, and provides the results to other network functions. However, there is no detailed breakdown of data analysis types, no classification of specific AI algorithms, and no grading of AI tasks based on task-level computational complexity. Therefore, as AI capabilities continue to increase, the network overhead required to implement different AI functions is significant, a problem that needs to be addressed urgently.

[0087] Based on this, the disclosed embodiments consider refining AI network functions, introducing a hierarchical relationship, and subdividing AI network elements according to specific algorithms and task types, dividing the hierarchical relationship into an AI management-level network element (AI0) and several equal-level sub-AIi network elements (AI1\AI2\...\AIN). AI0 is responsible for signaling analysis, resource allocation, and distribution deployment of AI services. It is closely integrated with other NFs (Network Functions) such as UDM and AMF. It can analyze the input information from the UE end, determine the specific AI task type, and then select the corresponding sub-AIi network element to provide services, including classification, regression, clustering, etc. At the same time, it has strong computing and storage resources and can handle computing-intensive tasks. The entire AI network function service process is implemented through the combined arrangement of AI0 and several sub-AIi network elements.

[0088] During this process, the scheduling of each task dispatch becomes the key point. In the embodiment of the present disclosure, the AI0 network element determines the processing parameters of the AI0 network element and each AIi network element, determines the task offloading strategy, and determines the AI ​​tasks executed at the AI0 network element and / or the AI ​​tasks executed at the AIi network element, which can reduce overhead; and based on the task offloading strategy, corresponding resource allocation can be performed, which can enable reasonable allocation of resources and enable AI services to be carried out efficiently and flexibly.

[0089] In addition, in order to facilitate understanding of the embodiments of the present disclosure, the following points are explained.

[0090] First, in the embodiments of the present disclosure, "used to indicate" may include being used to indicate directly or indirectly. When describing that a certain information is used to indicate A, it may include that the information directly indicates A or indirectly indicates A, but it does not necessarily mean that the information carries A.

[0091] The information indicated by the information is called the information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated. For example, but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated. The information to be indicated can also be indirectly indicated by indicating other information, where there is an association between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while the other parts of the information to be indicated are known or agreed in advance. For example, the indication of specific information can be achieved by using the arrangement order of each piece of information that is pre-agreed (for example, specified by the protocol), thereby reducing the indication overhead to a certain extent.

[0092] The information to be indicated can be sent as a whole or as multiple sub-information messages that are sent separately. The transmission periods and / or timings of these sub-information messages can be the same or different. The specific transmission method is not limited in this disclosure. The transmission periods and / or timings of these sub-information messages can be predefined, for example, according to a protocol.

[0093] Second, the first, second, and various numerical numbers in this disclosure are merely distinctions for ease of description and are not intended to limit the scope of the embodiments of this disclosure, for example, to distinguish different information, different AI network elements, and so on.

[0094] Third, the terminal device has completed the initial registration process and is connected to the network.

[0095] Fourth, the first AI network element has completed registration with the NRF function and can be accessed and work normally in the core network architecture.

[0096] Fifth, the core network has authenticated each first AI network element and at least one second AI network element to ensure their secure access.

[0097] Sixth, the first AI network element and at least one second AI network element trust each other and transmit authentic communication information. The second AI network element can be a sub-network function such as AI1, AI2, ..., AIN as shown in Figure 2. The second AI network element is parallel to other NFs such as PCF and UDR.

[0098] Seventh, the communication quality (channel quality, bandwidth) between different second AI network elements and the first AI network element can be the same or different.

[0099] Eighth, each time the first AI network element performs task allocation, the first AI network element and / or the second AI network element jointly complete the overall task, each responsible for the subtask portion thereof. In each round of tasks, some second AI network elements can participate in computing and communication.

[0100] Ninth, the "protocol" involved in the embodiments of the present disclosure may refer to a standard protocol in the communication field, for example, it may include an LTE protocol, an NR protocol, and related protocols used in future communication systems, and the present disclosure does not limit this.

[0101] Tenth, the embodiments of the present disclosure list multiple implementation methods to clearly illustrate the technical solutions of the embodiments of the present disclosure. Of course, those skilled in the art will understand that the multiple embodiments provided in the embodiments of the present disclosure can be implemented individually, combined with the methods of other embodiments of the embodiments of the present disclosure, or implemented individually or in combination with some methods in other related technologies; the embodiments of the present disclosure are not limited to this.

[0102] Please refer to Figure 3, which is a flow chart of an AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 3, the method may include but is not limited to the following steps:

[0103] In the embodiment of the present disclosure, the AMF network element can receive an AI service establishment request message (AI Service Establishment Request) sent by a terminal device (such as transparent transmission) through an access network device. The AI ​​service establishment request message is used to indicate the AI ​​service required by the terminal device, and then the AI ​​service required by the terminal device can be determined based on the AI ​​service establishment request message.

[0104] The AI ​​service establishment request message includes information such as the AI ​​service type (AI Service Type), the AI ​​service identifier (AI Service ID), etc.

[0105] Among them, the AMF network element can execute S31 when determining the AI ​​service required by the terminal device:

[0106] S31: Send an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0107] In the embodiment of the present disclosure, the AMF network element sends an AI service request message (CreateAI0Context_Request) to the first AI network element, indicating the AI ​​service that needs to be provided.

[0108] The AI ​​service request message includes information such as AI service type (AI Service Type), AI service identifier (AI Service ID), and terminal device information (User information).

[0109] In the disclosed embodiment, the first AI network element may be a management-level network element, responsible for signaling analysis, resource allocation, and distribution deployment of the AI ​​service. The first AI network element receives an AI service request message sent by the AMF network element and may execute S32 to S34 according to the AI ​​service request message.

[0110] S32: Determine at least one AI task according to the AI ​​service request message.

[0111] Among them, the first AI network element receives the AI ​​service request message sent by the AMF and can determine the AI ​​service that needs to be provided. The first AI network element can analyze the AI ​​service and determine at least one AI task that needs to be provided.

[0112] It is understandable that the first AI network element can analyze the AI ​​service, determine the AI ​​algorithm that needs to be provided, split the tasks according to the AI ​​algorithm, and determine at least one AI task.

[0113] Exemplarily, based on the AI ​​service request message, at least one classification AI task is determined, or at least one regression AI task is determined, or at least one clustering AI task is determined, or a classification AI task and a regression AI task are determined, and so on.

[0114] It should be noted that the above examples are for illustration only and are not intended to be specific limitations on the embodiments of the present disclosure. The determined AI tasks may also be types other than the above examples, or other methods may be used to determine the AI ​​tasks. For example, the first AI network element may determine in advance which method to use to determine the AI ​​task based on the AI ​​model functions locally deployed by the first AI network element and the AI ​​model functions locally deployed by each second AI network element, and this may be pre-set.

[0115] S33: Determine a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0116] In the disclosed embodiment, the first AI network element may determine the first processing parameter independently. The first AI network element may determine the second processing parameter of the second AI network element according to a protocol agreement, an instruction from a network device, or an instruction from the second AI network element.

[0117] Exemplarily, the first AI network element determines the second processing parameter of the second AI network element based on an instruction from the second AI network element. The second AI network element may report instruction information to the first AI network element, where the instruction information is used to indicate the second processing parameter of the second AI network element. Thus, the first AI network element may determine the second processing parameter of the second AI network element.

[0118] In some possible implementations, the first processing parameter may include a first task category supported by the first AI network element, and the second processing parameter may include a second task category supported by the second AI network element.

[0119] In some possible implementations, the first processing parameter may include the computing rate of the first AI network element processing the AI ​​task, and the second processing parameter may include the computing rate of the second AI network element processing the AI ​​task. In the case where there are multiple AI tasks, the first processing parameter may include the computing rate of the first AI network element processing each AI task, and the second processing parameter may include the computing rate of the second AI network element processing each AI task.

[0120] In some possible implementations, the first processing parameter may include a specific parameter for determining whether the first AI network element performs the AI ​​task, and the second processing parameter may include a specific parameter for determining whether the first AI network element performs the AI ​​task.

[0121] S34: Determine, according to the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task.

[0122] In an embodiment of the present disclosure, when a first AI network element determines at least one AI task based on an AI service request message, and determines a first processing parameter of the first AI network element and a second processing parameter of a second AI network element, the first task to be performed by the first AI network element and / or the second task to be performed by the second AI network element in the AI ​​task can be determined based on the AI ​​task, the first processing parameter, and the second processing parameter.

[0123] Exemplarily, when the first processing parameter includes a first task category supported by the first AI network element and the second processing parameter includes a second task category supported by the second AI network element, the first AI network element may determine a target task category for the AI ​​task, and then, if it is determined that the first task category supported by the first AI network element is the same as the target task category, determine that the first AI network element executes the AI ​​task; conversely, if it is determined that the first task category supported by the first AI network element is different from the target task category, determine that the AI ​​task is not executed at the first AI network element; or, if it is determined that the second task category supported by the second AI network element is the same as the target task category, determine that the second AI network element executes the AI ​​task; conversely, if it is determined that the second task category supported by the second AI network element is different from the target task category, determine that the AI ​​task is not executed at the second AI network element.

[0124] It is understandable that the above examples are for illustration only, and the first processing parameter and the second processing parameter may also be other parameters other than the above examples, or may also include other parameters including the above examples, and the embodiments of the present disclosure do not impose specific limitations on this.

[0125] By implementing the embodiments of the present disclosure, a first AI network element receives an AI service request message sent by an AMF network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided; determines at least one AI task based on the AI ​​service request message; determines a first processing parameter for the first AI network element and a second processing parameter for the second AI network element; and determines, based on the AI ​​task, the first processing parameter, and the second processing parameter, the first task to be executed by the first AI network element and / or the second task to be executed by the second AI network element in the AI ​​task. Thus, the first AI network element determines the first task to be executed by the first AI network element and / or the second task to be executed by the second AI network element in the AI ​​task, can classify and schedule the AI ​​tasks, and allocate resources based on the scheduling, which can reduce overhead and rationally allocate resources, allowing AI services to be performed more efficiently and flexibly.

[0126] In some embodiments, a method for determining, based on an AI task, a first processing parameter, and a second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by a second AI network element in the AI ​​task is performed by the first AI network element, as shown in FIG4 . The method is performed by the first AI network element and includes, but is not limited to, the following steps:

[0127] S41: Determine the target task category of the AI ​​task.

[0128] S42: Determine a first task to be performed by a first AI network element and / or a second task to be performed by a second AI network element in the AI ​​task based on a target task category, a first processing parameter, and a second processing parameter, wherein the first processing parameter includes a first task category supported by the first AI network element and the second processing parameter includes a second task category supported by the second AI network element.

[0129] In the embodiment of the present disclosure, the first AI network element may determine a target task category of the AI ​​task, such as a classification task, a regression task, and the like.

[0130] In the embodiment of the present disclosure, the first AI network element determines the first processing parameter of the first AI network element, and can determine the first task category supported by the first AI network element for processing, for example: the first task category supported by the AI ​​service function locally stored in the first AI network element for processing. It can be understood that the first AI network element can support processing multiple task categories, and the first task category can include multiple task categories.

[0131] In the embodiment of the present disclosure, the first AI network element determines the second processing parameter of the second AI network element, and can determine the second task category supported by the second AI network element for processing, for example: the second task category supported by the AI ​​service function locally stored in the second AI network element for processing. It can be understood that the second AI network element can support processing multiple task categories, and the second task category can include multiple task categories.

[0132] In one possible implementation, a first AI network element determines a target task category of an AI task and a first task category supported by the first AI network element for processing. When it is determined that the first task category supported by the first AI network element for processing is the same as the target task category, the first AI network element is determined to execute the AI ​​task. Conversely, when it is determined that the first task category supported by the first AI network element for processing is different from the target task category, the AI ​​task is determined not to be executed at the first AI network element.

[0133] In one possible implementation, a first AI network element determines a target task category of an AI task and a second task category supported and processed by a second AI network element. When it is determined that the second task category supported and processed by the second AI network element is the same as the target task category, the second AI network element is determined to execute the AI ​​task. Conversely, when it is determined that the second task category supported and processed by the second AI network element is different from the target task category, the AI ​​task is determined not to be executed at the second AI network element.

[0134] For example, the first AI network element determines that the target task category of the kth task in the AI ​​tasks is a classification task. The first AI network element determines that the first task category supported by the first AI network element includes classification tasks, and the first task category supported by the first AI network element includes regression tasks. Based on this, the first AI network element determines that the kth task in the AI ​​tasks is to be executed at the first AI network element, where k is a positive integer.

[0135] For example, a first AI network element determines that the target task category of the kth task in the AI ​​tasks is a classification task. The first AI network element determines that the first task category supported by the first AI network element includes regression tasks, and the first task category supported by the first AI network element includes classification tasks. Based on this, the first AI network element determines that the kth task in the AI ​​tasks is to be executed at the second AI network element, where k is a positive integer.

[0136] Exemplarily, a first AI network element determines that a target task category of the kth task in an AI task is a classification task. The first AI network element determines that a first task category supported by the first AI network element includes a classification task. Based on this, the first AI network element determines that the kth task in the AI ​​task is to be executed simultaneously at the first AI network element and the second AI network element, where k is a positive integer.

[0137] By implementing the disclosed embodiments, a first AI network element determines a target task category for an AI task, and based on the target task category, a first processing parameter, and a second processing parameter, determines a first task to be executed by the first AI network element and / or a second task to be executed by a second AI network element within the AI ​​task, wherein the first processing parameter includes a first task category supported by the first AI network element, and the second processing parameter includes a second task category supported by the second AI network element. Thus, the first AI network element determines the first task to be executed by the first AI network element and / or the second task to be executed by the second AI network element within the AI ​​task, enabling categorized scheduling of AI tasks and resource allocation based on the scheduling. This reduces overhead, rationalizes resource allocation, and enables more efficient and flexible AI services.

[0138] In some embodiments, the AI ​​service request message is further used to indicate a time threshold for obtaining a processing result. The first AI network element determines, based on the AI ​​task, the first processing parameter, and the second processing parameter, a method for determining a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task. As shown in FIG. 5 , the method is performed by the first AI network element and includes, but is not limited to, the following steps:

[0139] S51: Determine a first time duration required to obtain a first processing result based on the AI ​​task and the first processing parameter, wherein the first processing result is obtained by the first AI network element processing the first task.

[0140] In an embodiment of the present disclosure, the first AI network element can determine the first time required to obtain a first processing result based on the AI ​​task and the first processing parameter, wherein the first processing result is obtained by the first AI network element processing the first task.

[0141] Exemplarily, the first processing parameter may include the computing rate of the first AI network element processing the AI ​​task, and the first AI network element may determine the data volume of the first task, thereby determining the first time required to obtain the first processing result based on the computing rate of the first processing parameter and the data volume of the first task.

[0142] S52: Determine a second time duration required to obtain a second processing result based on the AI ​​task and the second processing parameter, wherein the second processing result is obtained by the second AI network element processing the second task.

[0143] In the embodiment of the present disclosure, the first AI network element can determine the second time required to obtain the second processing result based on the AI ​​task and the second processing parameters, wherein the second processing result is obtained by the second AI network element processing the second task.

[0144] Exemplarily, the second processing parameters may include a computing rate at which the second AI network element processes the second task, an upload rate at which the second AI network element uploads the processing result of the second task, and a waiting delay. The first AI network element may determine the data volume of the second task. Thus, the second time required to obtain the second processing result may be determined based on the computing rate at which the second AI network element processes the second task, the upload rate at which the second AI network element uploads the processing result of the second task, the waiting delay, and the data volume of the second task.

[0145] S53: Determine, according to the time threshold, the first duration, and the second duration, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task.

[0146] In the embodiment of the present disclosure, the AI ​​service request message is also used to indicate the time threshold for obtaining the processing results, for example, the time threshold is 5 minutes, 1 minute, etc.

[0147] In one possible implementation, if the first AI network element determines that the first duration is less than or equal to a time threshold, it can be determined that the AI ​​task can be executed at the first AI network element. Conversely, if it is determined that the first duration is greater than the time threshold, it can be determined that the AI ​​task is not executed at the first AI network element.

[0148] In one possible implementation, if the first AI network element determines that the second duration is less than or equal to the time threshold, it can be determined that the AI ​​task can be executed at the second AI network element. Conversely, if it is determined that the second duration is greater than the time threshold, it can be determined that the AI ​​task is not executed at the second AI network element.

[0149] Exemplarily, the time threshold is 5 minutes, and the first AI network element determines that the first time duration required to obtain the first processing result obtained by the first AI network element processing the kth task in the AI ​​task is 4 minutes. The first time duration of 4 minutes is less than the time threshold of 5 minutes. It can be determined that the kth task in the AI ​​task can be executed at the first AI network element, where k is a positive integer.

[0150] Exemplarily, the time threshold is 5 minutes, and the first AI network element determines that a first time duration required to obtain a first processing result obtained by the first AI network element processing the kth task in the AI ​​task is 6 minutes. The first time duration of 6 minutes is greater than the time threshold of 5 minutes. It can be determined that the kth task in the AI ​​task is not executed at the first AI network element, where k is a positive integer.

[0151] Exemplarily, the time threshold is 5 minutes, and the first AI network element determines that the second time duration required to obtain the second processing result obtained by the second AI network element processing the kth task in the AI ​​task is 3 minutes. The second time duration of 3 minutes is less than the time threshold of 5 minutes. It can be determined that the kth task in the AI ​​task can be executed at the second AI network element, where k is a positive integer.

[0152] Exemplarily, the time threshold is 5 minutes, and the first AI network element determines that the second time duration required to obtain the second processing result obtained by the second AI network element processing the kth task in the AI ​​task is 6 minutes. The second time duration of 6 minutes is greater than the time threshold of 5 minutes. It can be determined that the kth task in the AI ​​task is not executed at the second AI network element, where k is a positive integer.

[0153] It should be noted that the above examples are for illustration only, and the values ​​of the time threshold, the first duration and the second duration may also be other values, and the embodiments of the present disclosure do not impose any specific limitation on this.

[0154] In some embodiments, the first AI network element determines a first processing parameter of the first AI network element, including: determining a computing rate r of the first AI network element processing the kth AI task 0,k .

[0155] in, f0 is the computing frequency of the first AI network element, and M is the number of CPU cycles required for the first AI network element to process one bit of task data.

[0156] In some embodiments, the first AI network element determines the second processing parameter of the second AI network element, including: determining the computing rate of the i-th second AI network element processing the k-th AI task The upload rate of the processing result of the kth AI task And the waiting delay T i,k .

[0157] in, B is bandwidth, P is power, N0 is Gaussian white noise, h i is the wireless channel gain between the i-th second AI network element and the first AI network element, f i is the calculation frequency of the i-th second AI network element, M i The number of CPU cycles required to process one bit of task data for the i-th second AI network element.

[0158] In some embodiments, the first AI network element determines the first task to be performed by the first AI network element and / or the second task to be performed by the second AI network element according to the time threshold, the first duration, and the second duration, including: in response to satisfying t 0,k ≤T max , determine that the first AI network element performs the kth AI task; and / or in response to satisfying Determine the i-th second AI network element to perform the k-th AI task.

[0159] Among them, T max is the time threshold;

[0160] t 0,k The first duration of the first AI network element processing the kth AI task, where D k is the data volume of the kth AI task, r 0,k The computing rate of the first AI network element processing the kth AI task;

[0161] The second duration for the i-th second AI network element to process the k-th AI task,

[0162] in, The computing time required for the i-th second AI network element to process the k-th AI task,

[0163] is the computing rate of the i-th second AI network element processing the k-th AI task, T i,k For the waiting delay,

[0164] The upload time for the i-th second AI network element to upload the processing result of the k-th AI task,

[0165] The upload rate of the processing result of the kth AI task for the i-th second AI network element;

[0166] Wherein, i and k are both integers.

[0167] By implementing the embodiments of the present disclosure, the first AI network element determines the first time required to obtain the first processing result based on the AI ​​task and the first processing parameter, wherein the first processing result is obtained by the first AI network element processing the first task; determines the second time required to obtain the second processing result based on the AI ​​task and the second processing parameter, wherein the second processing result is obtained by the second AI network element processing the second task; and determines the first task executed by the first AI network element and / or the second task executed by the second AI network element in the AI ​​task based on the time threshold, the first time, and the second time. Thus, the first AI network element determines the first task executed by the first AI network element and / or the second task executed by the second AI network element in the AI ​​task, can classify and schedule AI tasks, and allocate resources based on the scheduling, which can reduce overhead and reasonably allocate resources, so that AI services can be carried out more efficiently and flexibly.

[0168] In some embodiments, a method for determining, based on an AI task, a first processing parameter, and a second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by a second AI network element in the AI ​​task is performed by the first AI network element, as shown in FIG6 . The method is performed by the first AI network element and includes, but is not limited to, the following steps:

[0169] S61: Determine a task offloading strategy generation model.

[0170] S62: Inputting the computing frequencies of the first AI network element and the second AI network element, and the wireless channel gain between the second AI network element and the first AI network element into a task offloading strategy generation model to generate a target task offloading strategy, wherein the target task offloading strategy includes the first task performed by the first AI network element and / or the second task performed by the second AI network element in the AI ​​task, the first processing parameter includes the computing frequency of the first AI network element, and the second processing parameter includes the computing frequency of the second AI network element and the wireless channel gain between the second AI network element and the first AI network element.

[0171] In an embodiment of the present disclosure, a first AI network element determines, based on the AI ​​task, a first processing parameter, and a second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task. A task offloading strategy generation model may be predetermined, and the computing frequencies of the first AI network element and the second AI network element, as well as the wireless channel gain between the second AI network element and the first AI network element, may be input into the task offloading strategy generation model to generate a target task offloading strategy.

[0172] The target task offloading strategy includes a first task performed by a first AI network element and / or a second task performed by a second AI network element in the AI ​​task; the first processing parameter includes a computing frequency of the first AI network element; and the second processing parameter includes a computing frequency of the second AI network element and a wireless channel gain between the second AI network element and the first AI network element.

[0173] By implementing the disclosed embodiments, a first AI network element determines a task offloading policy generation model and inputs the computation frequencies of the first and second AI network elements, as well as the radio channel gain between the second AI network element and the first AI network element, into the task offloading policy generation model to generate a target task offloading policy. The target task offloading policy includes the first task executed by the first AI network element and / or the second task executed by the second AI network element within the AI ​​task. The first processing parameter includes the computation frequency of the first AI network element, and the second processing parameter includes the computation frequency of the second AI network element and the radio channel gain between the second AI network element and the first AI network element. Thus, the first AI network element determines the first task executed by the first AI network element and / or the second task executed by the second AI network element within the AI ​​task. This allows for categorized scheduling of AI tasks and resource allocation based on the scheduling, reducing overhead and ensuring rational resource allocation, enabling more efficient and flexible AI services.

[0174] In some embodiments, a method for a first AI network element to determine a task offloading strategy generation model is shown in FIG7 . The method is performed by the first AI network element and includes but is not limited to the following steps:

[0175] S71: Initialize model parameters and determine the initial task offloading strategy generation model.

[0176] In the disclosed embodiment, the initial task offloading strategy generation model based on DRL (Deep Reinforcement Learning) can use a DNN (Deep Neural Network) model to initialize the model parameters of the DNN model, such as the number of layers and neurons.

[0177] Of course, other models can also be used as the initial task offloading strategy generation model. As long as the first AI network element in this solution can determine the task offloading strategy generation model, the initial task offloading strategy generation model can be set arbitrarily, and the embodiments of the present disclosure do not impose specific restrictions on this.

[0178] S72: Determine initial calculation frequencies of the first AI network element and the second AI network element, and an initial wireless channel gain between the second AI network element and the first AI network element.

[0179] In the embodiments of the present disclosure, the first AI network element may determine its own initial calculation frequency. The first AI network element may determine the initial calculation frequency of the second AI network element based on a protocol agreement, a network-side instruction, or a second AI network element instruction. This embodiment of the present disclosure does not impose any specific restrictions on this.

[0180] The first AI network element may determine the initial wireless channel gain between the second AI network element and the first AI network element based on a protocol agreement, or based on a network-side instruction, or based on an instruction from the second AI network element. This embodiment of the present disclosure does not impose any specific restrictions on this.

[0181] S73: Based on the initial calculation frequency and the initial wireless channel gain, jointly train the initial task offloading strategy generation model and the initial local model of the first AI network element and / or the second AI network element to generate the task offloading strategy generation model and the local model of the first AI network element and / or the second AI network element.

[0182] In an embodiment of the present disclosure, after determining the initial computing frequencies of the first AI network element and the second AI network element, and the initial wireless channel gain between the second AI network element and the first AI network element, the first AI network element may jointly train the initial task offloading policy generation model and the initial local models of the first AI network element and / or the second AI network element based on the initial computing frequencies and the initial wireless channel gain to generate the task offloading policy generation model and the local models of the first AI network element and / or the second AI network element.

[0183] In some embodiments, the first AI network element generates a model for the initial task offloading strategy based on the initial computing frequency and the initial wireless channel gain, and a method for jointly training the initial local models of the first AI network element and / or the second AI network element includes but is not limited to the following steps:

[0184] Step 1: Determine the number of iterations T, where T is a positive integer.

[0185] Step 2: Determine the first round of input model data as the initial calculation frequency and the initial wireless channel gain.

[0186] Step 3: Determine the input model data for round t as the updated calculation frequency and initial wireless channel gain for round t-1 of the first AI network element and / or the second AI network element after updating the initial local model of the first AI network element and / or the second AI network element based on the input model data for round t-1, where 2≤t≤T.

[0187] Step 4: Generate a model for the initial task offloading strategy based on each round of input model data, and jointly train the initial local model of the first AI network element and / or the second AI network element.

[0188] Step 5: Jointly train the initial task offloading strategy generation model and the initial local models of the first AI network element and / or the second AI network element based on the T-th round of input model data to generate the task offloading strategy generation model and the local models of the first AI network element and / or the second AI network element.

[0189] In the embodiment of the present disclosure, the first AI network element determines the number of iteration rounds T. The first AI network element may determine the number of iteration rounds T based on a protocol agreement, or based on an instruction from a network-side device, or based on implementation. The embodiment of the present disclosure does not impose any specific restrictions on this.

[0190] Exemplarily, the first AI network element determines that the number of iteration rounds T may be 100 rounds, 200 rounds, 500 rounds, and so on.

[0191] In the embodiment of the present disclosure, the first AI network element determines that the first round of input model data is the initial calculation frequency and the initial wireless channel gain.

[0192] In one possible implementation, a first AI network element determines an initial calculation frequency of a second AI network element that inputs model data in a first round, where the initial calculation frequency of the second AI network element is the frequency reported by the second AI network element to the first AI network element. The first AI network element determines an initial wireless channel gain between the second AI network element and the first AI network element, where the initial wireless channel gain between the second AI network element and the first AI network element is the frequency reported by the second AI network element to the first AI network element.

[0193] In an embodiment of the present disclosure, the first AI network element determines the input model data for the t-th round as the updated calculation frequency and initial wireless channel gain for the t-1th round of the first AI network element and / or the second AI network element after updating the initial local model of the first AI network element and / or the second AI network element based on the input model data for the t-1th round.

[0194] In the embodiments of the present disclosure, the first AI network element may determine the update calculation frequency of the t-th round of input model data of the first AI network element to be the t-1th round. The first AI network element may determine the update calculation frequency of the t-th round of input model data of the second AI network element to be the t-1th round based on a protocol agreement, or based on a network-side instruction, or based on an instruction from the second AI network element. The embodiments of the present disclosure do not impose specific limitations on this.

[0195] In one possible implementation, the first AI network element determines that the t-th round input model data is the t-1-th round update calculation frequency of the second AI network element, where the t-1-th round update calculation frequency of the second AI network element is reported by the second AI network element to the first AI network element.

[0196] In an embodiment of the present disclosure, the first AI network element determines that the first round of input model data is the initial calculation frequency and the initial wireless channel gain, and based on the first round of input model data, generates a model for the initial task offloading strategy, and jointly trains the initial local models of the first AI network element and / or the second AI network element.

[0197] On this basis, based on the determined input model data of each iteration, the initial task offloading policy generation model and the initial local models of the first AI network element and / or the second AI network element are jointly trained until the initial task offloading policy generation model and the initial local models of the first AI network element and / or the second AI network element are jointly trained based on the T-th round of input model data to generate the task offloading policy generation model and the local models of the first AI network element and / or the second AI network element.

[0198] In some embodiments, a method in which a first AI network element generates a model for an initial task offloading strategy based on a first round of input model data, and a method in which an initial local model of the first AI network element and / or a second AI network element is jointly trained includes:

[0199] 1. Input the initial calculation frequency and the initial wireless channel gain into an initial task offloading strategy generation model to generate an initial task offloading strategy, where the initial task offloading strategy includes an initial AI task executed by the first AI network element and / or the second AI network element.

[0200] 2. Determine a processing result of the first AI network element and / or the second AI network element performing the initial AI task, and generate model update parameters, where the model update parameters include update parameters of the first AI network element and / or the second AI network element.

[0201] 3. In response to the model update parameter including the first update parameter of the first AI network element, the initial task offloading strategy generation model and / or the initial local model of the first AI network element are updated according to the first update parameter.

[0202] 4. In response to the model update parameter including the second update parameter of the second AI network element, distribute the second update parameter to the second AI network element.

[0203] In some embodiments, the second AI network element receives a second update parameter sent by the first AI network element; and updates the initial local model of the second AI network element according to the second update parameter.

[0204] In the disclosed embodiment, the first AI network element jointly trains the initial task offloading policy generation model and the initial local models of the first AI network element and / or the second AI network element based on the initial calculation frequency and the initial wireless channel gain to generate the task offloading policy generation model and the local models of the first AI network element and / or the second AI network element.

[0205] This involves multiple iterations, focusing on three key steps: i) the first AI network element makes the task offloading decision; ii) the first and second AI network elements perform local training and computation; iii) the first AI network element aggregates the output results and performs a weighted average; iv) the first AI network element distributes the aggregated model (model parameters) to each second AI network element. These steps are repeated multiple times throughout the training process. The two main goals of this process are: i) maximizing the computation rate; and ii) satisfying execution latency constraints. This analysis will be focused on below.

[0206] During the entire AI task processing process, no matter whether it is executed at the first AI network element or the second AI network element, a single task must be completed before the deadline Tmax. i,t ∈{0,1} is an integer variable, where x i,t =1 means that the i-th task is executed at the first AI network element, x i,t =0 means that the i-th task is offloaded to the second AI network element.

[0207] Step 1: The first AI network element performs local computing:

[0208] The first AI network element has more powerful computing resources than the second AI network element. Therefore, when the first AI network element receives a task request, it first analyzes which tasks must be calculated locally on the first AI network element and which tasks can be sent to the second AI network element for calculation. Let f0 represent the calculation frequency (cycles / s) of the first AI network element, t k,t represents the computation time of task k in the tth round of training, satisfying 0≤t k,t ≤T. Then the total number of bits processed by the first AI network element is Where M represents the number of CPU cycles required to process one bit of task data. Therefore, the computing rate of the first AI network element is:

[0209] Step 2: Offload to the Second AI NE

[0210] Since the uplink communication rate is much lower than the downlink rate, that is, the upload rate is much slower than the download rate, the computing rate of a single task here is equal to the computing rate of the second AI network element plus the data upload rate from the second AI network element to the first AI network element, that is:

[0211]

[0212] Where hi represents the wireless channel gain between the first AI network element and the second AI network element, which is a dynamically changing variable. The weighted comprehensive calculation rate of the entire system is:

[0213]

[0214] Where the wireless channel gain h={h1,h2,...,h i |i∈N}, the calculation frequency of each second AI network element f={f1,f2,...,f i |i∈N}, different tasks k have different computational loads and different requirements for computing resources and computing frequencies.

[0215] Step 3: Determine time constraints

[0216] During the entire process, because the second AI network element processes tasks in parallel, the latency refers to the slowest of the local model training time and the parameter result upload time of the second AI network element sub-function. Since the downlink communication rate is much higher than the uplink rate, the time it takes for the first AI network element to send instructions to the second AI network element can be ignored. At the same time, since the first AI network element has much stronger computing resources than the second AI network element, the first AI network element can always complete the computing task first compared to the second AI network element. Assume that each task is divided into multiple tasks when it is sent, let and They represent the local model training time and result upload time of the second AI network element sub-function when executing task k in round t, Depends on both: i) Computation time ii) Waiting time T in the task queue of the second AI network element i,wait . Among them D k,t is the data volume of subtask k in round t, and the latter reflects the queuing time of the remaining workload ongoing on the second AI network element.

[0217] therefore, It can be expressed as:

[0218] The time required for the second AI network element to upload model parameters is:

[0219] Then the time for the second AI network element to complete the subtask must meet the following requirements:

[0220] Step 4: Optimization method modeling

[0221] Therefore, in summary, the task offloading problem is modeled as follows:

[0222] P1:

[0223] subject to:

[0224] x i ∈{0,1}

[0225] Problem P1 is a mixed-integer non-convex optimization problem with exponential complexity, making it difficult to solve in a finite time. Here, a deep reinforcement learning (DRL) approach is used to solve the offloading decision and allocation problem. This approach dynamically updates offloading decisions based on task type and channel state changes.

[0226] Before the task is dispatched, the first AI network element will sink the DNN model to the second AI network element for training. Within the time constraint, the first AI network element obtains the offloading decision through the DRL model and sends it to each second AI network element. Each second AI network element will calculate the channel gain h i,t , calculate the frequency f i,t Input to DNN. In the tth round, DNN passes the parameter h i,t and f i,t Get uninstall decision where θ i,t Represents, for example, the number of neurons and the number of neural network layers. The offloading decision is then uploaded to the first AI network element. After all subtask offloading decisions are finalized, the offloading action of the first AI network element is represented as:

[0227] Then, we use the threshold quantization method to quantize the slack offloading action Quantized into m binary offloading action combinations. The order-preserving quantization method follows the following rules:

[0228] After the first AI network element obtains all m offloading action combinations, we solve problem P2 for each offloading decision separately to obtain the weighted sum calculation rate Q(h,f).

[0229] P2:

[0230] subject to:

[0231] Finally, we select the best Q*(h,f) corresponding to As the final uninstallation result. After the first AI network element obtains the uninstallation action, the newly obtained state-action pair (h i,t ,f i,t, ) is added to memory.

[0232] The DRL offloading decision is updated once in each round of training. In the strategy update phase of the t-th round of tasks, each second AI network element selects the latest state-action pair (h i,t ,f i,t, ) to train the DNN. After training, the DNN will change its parameters from θ t Update to θ t+1 , the parameter update method is SGD algorithm. The new unloading strategy πθ generated t+1 It will be used in the next round of tasks based on the observed new channel state h i,t+1 and the new calculation frequency f i,t+1 Generating uninstall decisions Thereafter, once the channel state and task information change, this DRL method will continue to iterate, and the DNN will continue to improve its strategy. To improve the final training results.

[0233] Algorithm 1: DRL-based dynamic offloading decision solution algorithm

[0234] Input: Task category, wireless channel gain h in each round t , calculate the frequency f t .

[0235] Output: The optimal offloading decision for all second AI network elements

[0236] 1. Initialize the DNN model parameters θ and clear the memory R;

[0237] 2. Set the number of iterations T;

[0238] 3. For the number of iterations t = 1, 2, ..., T

[0239] 4. Generate a slack offloading action for each second AI network element And upload it to the first AI network element for decision-making;

[0240] 5. Quantized into m binomial action combinations

[0241] 6. For all {x k}Calculate Q*(h,f);

[0242] 7. Choose the best action

[0243] 8. For each second AI network element

[0244] 9. Update the memory and change (hi,t ,f i,t ,x i,t ) is added to memory R;

[0245] 10. Select the latest state pair from memory to train DNN and use SGD to update θ i,t →θ i,t+1 ;

[0246] 11. End

[0247] 12. Set the parameters θ of all AIi i,t Send to the first AI network element;

[0248] 13. Weighted average to obtain the global model parameter θ g,t ;

[0249] 14. Then send the global parameters to each second AI network element;

[0250] 15. End.

[0251] Please refer to Figure 8, which is a flowchart of another AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 8, the method may include but is not limited to the following steps:

[0252] S81: The AMF network element sends an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0253] S82: The first AI network element determines at least one AI task according to the AI ​​service request message.

[0254] S83: The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0255] S84: The first AI network element determines a first task to be performed by the first AI network element in the AI ​​task according to the AI ​​task, the first processing parameter, and the second processing parameter.

[0256] Among them, the relevant descriptions of S81 to S84 can be found in the relevant descriptions in the above embodiments, and will not be repeated here.

[0257] S85: The first AI network element executes the first task and generates a first processing result.

[0258] S86: The first AI network element sends the first processing result to the AMF network element.

[0259] In the embodiment of the present disclosure, the first AI network element determines the first task to be performed by the first AI network element in the AI ​​task based on the AI ​​task, the first processing parameter and the second processing parameter. In this case, the first AI network element performs the first task, generates a first processing result, and sends the first processing result to the AMF network element.

[0260] It can be understood that the AMF network element can send the first processing result sent by the first AI network element through RAN (such as transparent transmission) to the terminal device to feed back the processing result of the AI ​​service requested by the terminal device to the terminal device, so as to provide AI services to the terminal device.

[0261] Upon receiving the first processing result sent by the AMF, the terminal device may respond with an acknowledgement of receipt by sending an indication message to the AMF, indicating receipt of the first processing result. The indication message may also indicate whether the terminal device is satisfied with the first processing result, for example, indicating that the first processing result is accurate or inaccurate.

[0262] By implementing the embodiments of the present disclosure, the AMF network element sends an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The first AI network element determines at least one AI task based on the AI ​​service request message, determines a first processing parameter of the first AI network element, and a second processing parameter of the second AI network element. The first AI network element determines a first task to be performed by the first AI network element in the AI ​​task based on the AI ​​task, the first processing parameter, and the second processing parameter. The first AI network element performs the first task and generates a first processing result, which the first AI network element sends to the AMF network element. Thus, the first AI network element can classify and schedule AI tasks and allocate resources based on the scheduling, thereby reducing overhead and allocating resources reasonably, enabling AI services to be performed more efficiently and flexibly, and enabling AI tasks to be performed quickly and efficiently, providing users with satisfactory AI services.

[0263] Please refer to Figure 9, which is a flowchart of another AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 9, the method may include but is not limited to the following steps:

[0264] S91: The AMF network element sends an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0265] S92: The first AI network element determines at least one AI task according to the AI ​​service request message.

[0266] S93: The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0267] S94: The first AI network element determines a second task in the AI ​​task to be executed by the second AI network element according to the AI ​​task, the first processing parameter, and the second processing parameter.

[0268] For the relevant descriptions of S91 to S94, reference can be made to the relevant descriptions in the above embodiments, which will not be repeated here.

[0269] S95: The first AI network element sends the second task to the second AI network element.

[0270] S96: The second AI network element performs the second task and generates a preliminary processing result.

[0271] S97: The second AI network element sends the preliminary processing result to the first AI network element.

[0272] S98: The first AI network element generates a second processing result based on the preliminary processing result.

[0273] S99: The first AI network element sends a second processing result to the AMF network element, where the second processing result is determined by the first AI network element based on the preliminary processing result.

[0274] In an embodiment of the present disclosure, a first AI network element determines a second task to be performed by a second AI network element within the AI ​​task based on the AI ​​task, the first processing parameter, and the second processing parameter. In this case, the first AI network element sends the second task to the second AI network element, and the second AI network element performs the second task and generates a preliminary processing result. Further, the second AI network element may send the preliminary processing result to the first AI network element.

[0275] Among them, the first AI network element receives the preliminary processing result sent by the second AI network element, can process the preliminary processing result to generate a second processing result, and send the second processing result to the AMF network element.

[0276] It can be understood that the AMF network element can send the second processing result sent by the first AI network element through RAN (such as transparent transmission) to the terminal device to feed back the processing result of the AI ​​service requested by the terminal device to the terminal device, so as to provide AI services to the terminal device.

[0277] Upon receiving the second processing result sent by the AMF, the terminal device may respond with an acknowledgement of receipt and send an indication to the AMF indicating receipt of the second processing result. The indication may also indicate whether the terminal device is satisfied with the second processing result, for example, indicating that the second processing result is accurate or inaccurate.

[0278] By implementing the embodiments of the present disclosure, the AMF network element sends an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The first AI network element determines at least one AI task based on the AI ​​service request message, determines a first processing parameter for the first AI network element, and a second processing parameter for the second AI network element. The first AI network element determines a second task within the AI ​​task to be performed by the second AI network element based on the AI ​​task, the first processing parameter, and the second processing parameter. The first AI network element sends the second task to the second AI network element, which performs the second task and generates a preliminary processing result. The second AI network element sends the preliminary processing result to the first AI network element, which generates a second processing result based on the preliminary processing result. The first AI network element sends the second processing result to the AMF network element, wherein the second processing result is determined by the first AI network element based on the preliminary processing result. As a result, the first AI network element can classify and schedule AI tasks and allocate resources based on the scheduling, thereby reducing overhead and allocating resources rationally. This allows AI services to be performed more efficiently and flexibly, and allows AI tasks to be performed quickly and efficiently, providing users with satisfactory AI services.

[0279] Please refer to Figure 10, which is a flowchart of another AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 9, the method may include but is not limited to the following steps:

[0280] S101: The AMF network element sends an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0281] S102: The first AI network element determines at least one AI task according to the AI ​​service request message.

[0282] S103: The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0283] S104: The first AI network element determines, according to the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and a second task to be performed by the second AI network element in the AI ​​task.

[0284] For the relevant descriptions of S101 to S104 , reference can be made to the relevant descriptions in the above embodiments, which will not be repeated here.

[0285] S105: The first AI network element executes the first task and generates a first processing result.

[0286] S106: The first AI network element sends the second task to the second AI network element.

[0287] S107: The second AI network element performs the second task and generates a preliminary processing result.

[0288] S108: The second AI network element sends the preliminary processing result to the first AI network element.

[0289] S109: The first AI network element generates a target processing result according to the first processing result and the preliminary processing result.

[0290] S100: The first AI network element sends the target processing result to the AMF network element.

[0291] In the embodiment of the present disclosure, the first AI network element determines the first task to be performed by the first AI network element in the AI ​​task based on the AI ​​task, the first processing parameter and the second processing parameter. In this case, the first AI network element performs the first task and generates a first processing result.

[0292] In an embodiment of the present disclosure, a first AI network element determines a second task to be performed by a second AI network element within the AI ​​task based on the AI ​​task, the first processing parameter, and the second processing parameter. In this case, the first AI network element sends the second task to the second AI network element, and the second AI network element performs the second task and generates a preliminary processing result. Further, the second AI network element may send the preliminary processing result to the first AI network element.

[0293] Among them, the first AI network element receives the preliminary processing result sent by the second AI network element, and can generate a target processing result based on the first processing result and the preliminary processing result, and send the target processing result to the AMF network element.

[0294] It can be understood that the AMF network element can send the target processing result sent by the first AI network element through RAN (such as transparent transmission) to the terminal device to feed back the processing result of the AI ​​service requested by the terminal device to the terminal device, so as to provide AI services to the terminal device.

[0295] Upon receiving the target processing result sent by the AMF, the terminal device may respond with an indication message indicating receipt of the result. The indication message may also indicate whether the terminal is satisfied with the target processing result. For example, the indication message may indicate that the target processing result is accurate, or that the target processing result is inaccurate.

[0296] By implementing the embodiments of the present disclosure, the AMF network element sends an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The first AI network element determines, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and a second task to be performed by the second AI network element in the AI ​​task. The first AI network element performs the first task and generates a first processing result. The first AI network element sends the second task to the second AI network element. The second AI network element performs the second task and generates a preliminary processing result. The second AI network element sends the preliminary processing result to the first AI network element. The first AI network element generates a target processing result based on the first processing result and the preliminary processing result. The first AI network element sends the target processing result to the AMF network element. As a result, the first AI network element can classify and schedule AI tasks and allocate resources based on the scheduling, thereby reducing overhead and allocating resources reasonably, enabling AI services to be performed more efficiently and flexibly, and enabling AI tasks to be performed quickly and efficiently, providing users with satisfactory AI services.

[0297] Please refer to Figure 11, which is a flow chart of another AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 8, the method may include but is not limited to the following steps:

[0298] S111: The AMF network element sends an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0299] S112: The first AI network element determines at least one AI task according to the AI ​​service request message.

[0300] S113: The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0301] S114: The first AI network element determines a first task to be performed by the first AI network element in the AI ​​task according to the AI ​​task, the first processing parameter, and the second processing parameter.

[0302] For the relevant descriptions of S111 to S114 , reference can be made to the relevant descriptions in the above embodiments, which will not be repeated here.

[0303] S115: The first AI network element receives a first data set sent by the network function NF network element.

[0304] S116: The first AI network element executes the first task according to the first data set and generates a first processing result.

[0305] S117: The first AI network element sends the first processing result to the AMF network element.

[0306] In the embodiment of the present disclosure, the first AI network element determines the first task to be performed by the first AI network element in the AI ​​task based on the AI ​​task, the first processing parameter and the second processing parameter. In this case, the first AI network element receives the first data set sent by the network function NF network element, executes the first task based on the first data set, generates a first processing result, and sends the first processing result to the AMF network element.

[0307] Among them, the NF network element can be a UDR (unified data repository) network element and / or a UDSF (unstructured data storage function) network element, the first data set can include structured data and / or unstructured data, and the data source of the first data set is stored in the UDR network element and / or UDSF network element when the terminal device registers and makes a service request.

[0308] It can be understood that the AMF network element can send the first processing result sent by the first AI network element through RAN (such as transparent transmission) to the terminal device to feed back the processing result of the AI ​​service requested by the terminal device to the terminal device, so as to provide AI services to the terminal device.

[0309] Upon receiving the first processing result sent by the AMF, the terminal device may respond with an acknowledgement of receipt by sending an indication message to the AMF, indicating receipt of the first processing result. The indication message may also indicate whether the terminal device is satisfied with the first processing result, for example, indicating that the first processing result is accurate or inaccurate.

[0310] By implementing the embodiments of the present disclosure, an AMF network element sends an AI service request message to a first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The first AI network element determines at least one AI task based on the AI ​​service request message, determines a first processing parameter for the first AI network element, and a second processing parameter for the second AI network element. The first AI network element determines a first task to be performed by the first AI network element within the AI ​​task based on the AI ​​task, the first processing parameter, and the second processing parameter. The first AI network element receives a first data set sent by a network function NF network element, executes the first task based on the first data set, generates a first processing result, and sends the first processing result to the AMF network element. Thus, the first AI network element can classify and schedule AI tasks and allocate resources based on the scheduling, thereby reducing overhead and allocating resources rationally. This allows AI services to be performed more efficiently and flexibly, and allows AI tasks to be executed quickly and efficiently, providing users with satisfactory AI services.

[0311] Please refer to Figure 12, which is a flowchart of another AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 9, the method may include but is not limited to the following steps:

[0312] S121: The AMF network element sends an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0313] S122: The first AI network element determines at least one AI task according to the AI ​​service request message.

[0314] S123: The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0315] S124: The first AI network element determines a second task in the AI ​​task to be executed by the second AI network element according to the AI ​​task, the first processing parameter, and the second processing parameter.

[0316] For the relevant descriptions of S121 to S124, reference can be made to the relevant descriptions in the above embodiments, which will not be repeated here.

[0317] S125: The first AI network element sends the second task to the second AI network element.

[0318] S126: The second AI network element receives the second data set sent by the network function NF network element.

[0319] S127: The second AI network element performs the second task according to the second data set and generates a preliminary processing result.

[0320] S128: The second AI network element sends the preliminary processing result to the first AI network element.

[0321] S129: The first AI network element generates a second processing result based on the preliminary processing result.

[0322] S120: The first AI network element sends a second processing result to the AMF network element, where the second processing result is determined by the first AI network element based on the preliminary processing result.

[0323] In an embodiment of the present disclosure, the first AI network element determines a second task to be performed by the second AI network element in the AI ​​task based on the AI ​​task, the first processing parameter, and the second processing parameter. In this case, the first AI network element sends the second task to the second AI network element. The second AI network element receives the second data set sent by the network function NF network element, executes the second task based on the second data set, and generates a preliminary processing result. Furthermore, the second AI network element can send the preliminary processing result to the first AI network element.

[0324] Among them, the NF network element can be a UDR (unified data repository) network element and / or a UDSF (unstructured data storage function) network element, the first data set can include structured data and / or unstructured data, and the data source of the first data set is stored in the UDR network element and / or UDSF network element when the terminal device registers and makes a service request.

[0325] Among them, the first AI network element receives the preliminary processing result sent by the second AI network element, can process the preliminary processing result to generate a second processing result, and send the second processing result to the AMF network element.

[0326] It can be understood that the AMF network element can send the second processing result sent by the first AI network element through RAN (such as transparent transmission) to the terminal device to feed back the processing result of the AI ​​service requested by the terminal device to the terminal device, so as to provide AI services to the terminal device.

[0327] Upon receiving the second processing result sent by the AMF, the terminal device may respond with an acknowledgement of receipt and send an indication to the AMF indicating receipt of the second processing result. The indication may also indicate whether the terminal device is satisfied with the second processing result, for example, indicating that the second processing result is accurate or inaccurate.

[0328] By implementing the embodiment of the present disclosure, the AMF network element sends an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service that needs to be provided, the first AI network element determines at least one AI task based on the AI ​​service request message, the first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element, the first AI network element determines a second task performed by the second AI network element in the AI ​​task based on the AI ​​task, the first processing parameter and the second processing parameter, the first AI network element sends the second task to the second AI network element, the second AI network element receives a second data set sent by the network function NF network element, the second AI network element executes the second task based on the second data set and generates a preliminary processing result, the second AI network element sends the preliminary processing result to the first AI network element, the first AI network element generates a second processing result based on the preliminary processing result, and the first AI network element sends the second processing result to the AMF network element, wherein the second processing result is determined by the first AI network element based on the preliminary processing result. As a result, the first AI network element can classify and schedule AI tasks and allocate resources according to the scheduling, which can reduce overhead and allocate resources reasonably, so that AI services can be carried out more efficiently and flexibly, and can execute AI tasks quickly and efficiently, providing users with satisfactory AI services.

[0329] Please refer to Figure 13, which is a flow chart of another AI task processing method provided by an embodiment of the present disclosure. As shown in Figure 9, the method may include but is not limited to the following steps:

[0330] S131: The AMF network element sends an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service that needs to be provided.

[0331] S132: The first AI network element determines at least one AI task according to the AI ​​service request message.

[0332] S133: The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element.

[0333] S134: The first AI network element determines a second task in the AI ​​task to be executed by the second AI network element according to the AI ​​task, the first processing parameter, and the second processing parameter.

[0334] S135: The first AI network element sends the second task to the second AI network element.

[0335] S136: The second AI network element performs the second task and generates a preliminary processing result.

[0336] S137: The second AI network element sends the preliminary processing result to the first AI network element.

[0337] For the relevant descriptions of S131 to S137, reference can be made to the relevant descriptions in the above embodiments, which will not be repeated here.

[0338] S138: The first AI network element sends a response message to the second AI network element, where the response message is used to indicate that the first AI network element has received the preliminary processing result.

[0339] In the embodiment of the present disclosure, upon receiving the preliminary processing result sent by the second AI network element, the first AI network element may send a response message to the second AI network element to inform the second AI network element that the preliminary processing result sent by the second AI network element has been sent to the first AI network element.

[0340] S139: The first AI network element generates a second processing result based on the preliminary processing result.

[0341] S130: The first AI network element sends a second processing result to the AMF network element, where the second processing result is determined by the first AI network element based on the preliminary processing result.

[0342] For the relevant descriptions of S139 to S130, reference can be made to the relevant descriptions in the above embodiments, which will not be repeated here.

[0343] By implementing the embodiments of the present disclosure, the AMF network element sends an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The first AI network element determines at least one AI task based on the AI ​​service request message. The first AI network element determines a first processing parameter of the first AI network element and a second processing parameter of the second AI network element. The first AI network element determines a second task to be performed by the second AI network element in the AI ​​task based on the AI ​​task, the first processing parameter, and the second processing parameter. The first AI network element sends the second task to the second AI network element. The second AI network element performs the second task and generates a preliminary processing result. The second AI network element sends the preliminary processing result to the first AI network element. The first AI network element sends a response message to the second AI network element, wherein the response message is used to indicate that the first AI network element has received the preliminary processing result. The first AI network element generates a second processing result based on the preliminary processing result. The first AI network element sends the second processing result to the AMF network element, wherein the second processing result is determined by the first AI network element based on the preliminary processing result. As a result, the first AI network element can classify and schedule AI tasks and allocate resources according to the scheduling, which can reduce overhead and allocate resources reasonably, so that AI services can be carried out more efficiently and flexibly, and can execute AI tasks quickly and efficiently, providing users with satisfactory AI services.

[0344] In the embodiments provided by the present disclosure above, the solutions provided by the embodiments of the present disclosure are mainly introduced from the perspective of interaction between devices. It is understandable that, in order to implement the above functions, each device includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily appreciate that, in combination with the algorithm steps of each example described in the embodiments disclosed herein, the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present disclosure.

[0345] Please refer to Figure 14, which is a schematic diagram of the structure of a communication device 1 provided in an embodiment of the present disclosure. The communication device 1 shown in Figure 14 may include a transceiver module 11 and a processing module 12. The transceiver module 11 may include a sending module and / or a receiving module. The sending module is used to implement a sending function, and the receiving module is used to implement a receiving function. The transceiver module 11 can implement the sending function and / or the receiving function.

[0346] The communication device 1 is provided at the first AI network element side and includes a transceiver module 11 and a processing module 12.

[0347] The transceiver module 11 is configured to receive an AI service request message sent by an access and mobility management function AMF network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided;

[0348] The processing module 12 is configured to determine at least one AI task according to the AI ​​service request message;

[0349] The processing module 12 is further configured to determine a first processing parameter of the first AI network element and a second processing parameter of the second AI network element;

[0350] The processing module 12 is further configured to determine, according to the AI ​​task, the first processing parameter and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task.

[0351] In some embodiments, the processing module 12 is further configured to determine a target task category of the AI ​​task; and determine the first task performed by the first AI network element and / or the second task performed by the second AI network element in the AI ​​task based on the target task category, the first processing parameter, and the second processing parameter, wherein the first processing parameter includes the first task category supported by the first AI network element for processing, and the second processing parameter includes the second task category supported by the second AI network element for processing.

[0352] In some embodiments, the AI ​​service request message is also used to indicate a time threshold for obtaining a processing result. The processing module 12 is also configured to determine a first time duration required to obtain a first processing result based on the AI ​​task and the first processing parameter, wherein the first processing result is obtained by the first AI network element processing the first task.

[0353] The processing module 12 is further configured to determine a second time duration required to obtain a second processing result based on the AI ​​task and the second processing parameter, wherein the second processing result is obtained by the second AI network element processing the second task.

[0354] The processing module 12 is further configured to determine, according to the time threshold, the first duration, and the second duration, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task.

[0355] In some embodiments, the processing module 12 is further configured to, in response to satisfying t0, k ≤T max , determine that the first AI network element performs the kth AI task; and / or in response to satisfying Determine the i-th second AI network element to perform the k-th AI task;

[0356] Among them, T max is the time threshold;

[0357] t 0,k The first duration of the first AI network element processing the kth AI task, where D k is the data volume of the kth AI task, r 0,k The computing rate of the first AI network element processing the kth AI task;

[0358] The second duration for the i-th second AI network element to process the k-th AI task,

[0359] in, The computing time required for the i-th second AI network element to process the k-th AI task,

[0360] is the computing rate of the i-th second AI network element processing the k-th AI task, T i,k For the waiting delay,

[0361] The upload time for the i-th second AI network element to upload the processing result of the k-th AI task,

[0362] The upload rate of the processing result of the kth AI task for the i-th second AI network element;

[0363] Wherein, i and k are both integers.

[0364] In some embodiments, the processing module 12 is further configured to determine the computing rate r of the first AI network element processing the kth AI task. 0,k ;

[0365] in, f0 is the computing frequency of the first AI network element, and M is the number of CPU cycles required for the first AI network element to process one bit of task data.

[0366] In some embodiments, the processing module 12 is further configured to determine the computing rate of the i-th second AI network element processing the k-th AI task The upload rate of the processing results of the kth AI task And the waiting delay T i,k ;

[0367] in, B is bandwidth, P is power, N0 is Gaussian white noise, h i is the wireless channel gain between the i-th second AI network element and the first AI network element, f i is the calculation frequency of the i-th second AI network element, M i The number of CPU cycles required to process one bit of task data for the i-th second AI network element.

[0368] In some embodiments, the processing module 12 is further configured to determine a task offloading strategy generation model; input the computing frequencies of the first AI network element and the second AI network element, and the wireless channel gain between the second AI network element and the first AI network element into the task offloading strategy generation model to generate a target task offloading strategy, wherein the target task offloading strategy includes the first task performed by the first AI network element and / or the second task performed by the second AI network element in the AI ​​task, the first processing parameter includes the computing frequency of the first AI network element, and the second processing parameter includes the computing frequency of the second AI network element and the wireless channel gain between the second AI network element and the first AI network element.

[0369] In some embodiments, the processing module 12 is further configured to initialize model parameters and determine an initial task offloading strategy generation model.

[0370] The processing module 12 is further configured to determine initial calculation frequencies of the first AI network element and the second AI network element, and an initial wireless channel gain between the second AI network element and the first AI network element.

[0371] The processing module 12 is further configured to jointly train the initial task offloading strategy generation model and the initial local model of the first AI network element and / or the second AI network element based on the initial calculation frequency and the initial wireless channel gain to generate the task offloading strategy generation model and the local model of the first AI network element and / or the second AI network element.

[0372] In some embodiments, the processing module 12 is further configured to determine the number of iterations T, where T is a positive integer.

[0373] The processing module 12 is further configured to determine that the first round of input model data is an initial calculation frequency and an initial wireless channel gain.

[0374] The processing module 12 is further configured to determine the input model data for the t-th round as the updated calculation frequency and initial wireless channel gain of the first AI network element and / or the second AI network element for the t-1th round after updating the initial local model of the first AI network element and / or the second AI network element based on the input model data for the t-1th round, where 2≤t≤T.

[0375] The processing module 12 is further configured to generate a model for the initial task offloading strategy and jointly train the initial local model of the first AI network element and / or the second AI network element according to each round of input model data.

[0376] The processing module 12 is further configured to jointly train the initial task offloading strategy generation model and the initial local model of the first AI network element and / or the second AI network element according to the T-th round of input model data to generate the task offloading strategy generation model and the local model of the first AI network element and / or the second AI network element.

[0377] In some embodiments, the processing module 12 is further configured to input the initial calculation frequency and the initial wireless channel gain into an initial task offloading strategy generation model to generate an initial task offloading strategy, wherein the initial task offloading strategy includes an initial AI task executed by the first AI network element and / or the second AI network element.

[0378] The processing module 12 is further configured to determine a processing result of the first AI network element and / or the second AI network element performing the initial AI task, and generate a model update parameter, wherein the model update parameter includes an update parameter of the first AI network element and / or the second AI network element.

[0379] The processing module 12 is further configured to update the initial task offloading strategy generation model and / or the initial local model of the first AI network element according to the first update parameter in response to the model update parameter including the first update parameter of the first AI network element.

[0380] The processing module 12 is further configured to distribute the second update parameter to the second AI network element in response to the model update parameter including the second update parameter of the second AI network element.

[0381] In some embodiments, the processing module 12 is further configured to, in response to determining the first task to be performed by the first AI network element, execute the first task and generate a first processing result.

[0382] In some embodiments, the transceiver module 11 is further configured to receive a first data set sent by a network function NF network element in response to determining the first task performed by the first AI network element.

[0383] The processing module 12 is further configured to execute a first task according to the first data set and generate a first processing result.

[0384] In some embodiments, the transceiver module 11 is further configured to send the first processing result to the AMF network element.

[0385] In some embodiments, the transceiver module 11 is further configured to send the second task to the second AI network element in response to determining that the second AI network element performs the second task.

[0386] The transceiver module 11 is further configured to receive a preliminary processing result sent by the second AI network element, wherein the preliminary processing result is generated by the second AI network element when performing the second task.

[0387] In some embodiments, the transceiver module 11 is further configured to send a response message to the second AI network element, wherein the response message is used to indicate that the first AI network element has received the preliminary processing result.

[0388] In some embodiments, the transceiver module 11 is further configured to send a second processing result to the AMF network element, wherein the second processing result is determined by the first AI network element based on the preliminary processing result.

[0389] In some embodiments, the processing module 12 is further configured to, in response to determining the first processing result and the preliminary processing result, process the first processing result and the preliminary processing result to generate a target processing result.

[0390] In some embodiments, the transceiver module 11 is further configured to send the target processing result to the AMF network element.

[0391] The communication device 1 is arranged on the AMF network element side: including: a transceiver module 11.

[0392] The transceiver module 11 is configured to receive an AI service establishment request message sent by a terminal device, wherein the AI ​​service request message is used to indicate the AI ​​service required by the terminal device.

[0393] The transceiver module 11 is further configured to send an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The AI ​​service request message is used by the first AI network element to determine at least one AI task based on the AI ​​service request message, determine a first processing parameter of the first AI network element, and a second processing parameter of the second AI network element, and determine, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element.

[0394] In some embodiments, the transceiver module 11 is further configured to receive a first processing result sent by the first AI network element, wherein the first processing result is generated by the first AI network element performing the first task.

[0395] In some embodiments, the transceiver module 11 is further configured to receive a second processing result sent by the first AI network element, wherein the second processing result is determined by the first AI network element based on the preliminary processing result, and the preliminary processing result is generated by the second AI network element performing the second task.

[0396] In some embodiments, the transceiver module 11 is further configured to receive a target processing result sent by the first AI network element, where the target processing result is generated by the first AI network element processing the first processing result and the preliminary processing result when the first AI network element determines the first processing result and the preliminary processing result; the first processing result is generated by the first AI network element performing an AI task, and the preliminary processing result is generated by the second AI network element performing an AI task.

[0397] The communication device 1 is provided at the second network element side and includes a transceiver module 11 and a processing module 12 .

[0398] The transceiver module 11 is configured to receive a second task sent by the first AI network element, wherein the second AI task is determined by the first AI network element to be executed by the second AI network element and sent to the second AI network element based on the AI ​​task, the determined first processing parameter of the first AI network element, and the determined second processing parameter of the second AI network element. The AI ​​task is determined by the first AI network element based on the AI ​​service request message sent by the AMF network element, and the AI ​​service request message is used to indicate the AI ​​service to be provided.

[0399] In some embodiments, the processing module 12 is configured to perform a second task and generate a preliminary processing result.

[0400] In some embodiments, the transceiver module 11 is further configured to receive a second data set sent by the network function NF network element.

[0401] The processing module 12 is further configured to perform a second task according to the second data set and generate a preliminary processing result.

[0402] In some embodiments, the transceiver module 11 is further configured to receive a second update parameter sent by the first AI network element.

[0403] The processing module 12 is further configured to update the initial local model of the second AI network element according to the second update parameter.

[0404] In some embodiments, the transceiver module 11 is further configured to send the preliminary processing result to the first AI network element.

[0405] In some embodiments, the transceiver module 11 is further configured to receive a response message sent by the first AI network element, wherein the response message is used to indicate that the first AI network element has received the preliminary processing result.

[0406] Regarding the communication device 1 in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated on here.

[0407] The communication device 1 provided in the above embodiments of the present disclosure achieves the same or similar beneficial effects as the AI ​​task processing methods provided in some of the above embodiments, and will not be described in detail here.

[0408] Please refer to Figure 15, which is a structural diagram of a communication system provided by an embodiment of the present disclosure. As shown in Figure 15, the communication system 10 includes an AMF network element 101, a first AI network element 102, and a second AI network element 103.

[0409] The AMF network element 101 is configured to receive an AI service establishment request message sent by a terminal device, where the AI ​​service request message is used to indicate the AI ​​service required by the terminal device; and send an AI service request message to the first AI network element, where the AI ​​service request message is used to indicate the AI ​​service to be provided.

[0410] The first AI network element 102 is configured to receive an AI service request message sent by an AMF network element, where the AI ​​service request message is used to indicate an AI service to be provided; determine at least one AI task based on the AI ​​service request message; determine a first processing parameter of the first AI network element and a second processing parameter of the second AI network element; and determine, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​service.

[0411] The first AI network element 102 is further configured to, in response to determining that the second AI network element performs the second task, send the second task to the second AI network element.

[0412] The second AI network element 103 is configured to receive the second task sent by the first AI network element.

[0413] In the embodiments of the present disclosure, the AMF network element 101, the first AI network element 102, and the second AI network element 103 can implement the AI ​​task processing method provided in the above embodiments. The specific methods of the operations respectively performed by the AMF network element 101, the first AI network element 102, and the second AI network element 103 have been described in detail in the embodiments of the method and will not be elaborated here.

[0414] The communication system 10 provided in the above embodiments of the present disclosure achieves the same or similar beneficial effects as the AI ​​task processing methods provided in some of the above embodiments, and will not be described in detail here.

[0415] Please refer to Figure 16, which is a structural diagram of another communication device 1000 provided in an embodiment of the present disclosure. Communication device 1000 can be an AMF network element, a first AI network element, or a second AI network element. This device can be used to implement the method described in the above method embodiment. For details, please refer to the description of the above method embodiment.

[0416] The communication device 1000 may include one or more processors 1001. The processor 1001 may be a general-purpose processor or a dedicated processor. For example, it may be a baseband processor or a central processing unit. The baseband processor may be used to process communication protocols and communication data, and the central processing unit may be used to control the communication device (e.g., a base station, a baseband chip, a terminal device, a terminal device chip, a DU or CU, etc.), execute computer programs, and process computer program data.

[0417] Optionally, the communication device 1000 may further include one or more memories 1002, on which a computer program 1004 may be stored. The memory 1002 executes the computer program 1004 to enable the communication device 1000 to perform the method described in the above method embodiment. Optionally, the memory 1002 may also store data. The communication device 1000 and the memory 1002 may be provided separately or integrated together.

[0418] Optionally, the communication device 1000 may further include a transceiver 1005 and an antenna 1006. The transceiver 1005 may be referred to as a transceiver unit, a transceiver, or a transceiver circuit, etc., and is configured to implement transceiver functions. The transceiver 1005 may include a receiver and a transmitter. The receiver may be referred to as a receiver or a receiving circuit, etc., and is configured to implement a receiving function; the transmitter may be referred to as a transmitter or a transmitting circuit, etc., and is configured to implement a transmitting function.

[0419] Optionally, the communication device 1000 may further include one or more interface circuits 1007. The interface circuit 1007 is configured to receive code instructions and transmit the code instructions to the processor 1001. The processor 1001 executes the code instructions to enable the communication device 1000 to perform the method described in the above method embodiment.

[0420] The communication device 1000 is a first AI network element, and the transceiver 1005 is configured to execute S31 in FIG3; S81 and S86 in FIG8; S91, S95, S97, and S99 in FIG9; S101, S106, S108, and S100 in FIG10; S111, S115, and S117 in FIG11; S121, S125, S128, and S120 in FIG12; S131, S135, S137, S138, and S130 in FIG13; and the processor 100. 1 is used to execute S32 to S34 in Figure 3; S41 to S42 in Figure 4; S51 to S53 in Figure 5; S61 to S62 in Figure 6; S71 to S73 in Figure 7; S82 to S85 in Figure 8; S92 to S94 and S98 in Figure 9; S102 to S105 and S109 in Figure 10; S112 to S114 and S116 in Figure 11; S122 to S124 and S129 in Figure 12; and S132 to S134 and S139 in Figure 13.

[0421] The communication device 1000 is an AMF network element: the transceiver 1005 is used to execute S31 in Figure 3; S81 and S86 in Figure 8; S91 and S99 in Figure 9; S101 and S100 in Figure 10; S111 and S117 in Figure 11; S121 and S120 in Figure 12; and S131 and S130 in Figure 13.

[0422] The communication device 1000 is a second AI network element: the transceiver 1005 is used to execute S95 and S97 in Figure 9; S106 and S108 in Figure 10; S115 in Figure 11; S125, S126 and S128 in Figure 12; and S135, S137 and S138 in Figure 13.

[0423] Processor 1001 is used to execute S96 in Figure 9; S107 in Figure 10; S127 in Figure 12; and S136 in Figure 13.

[0424] In one implementation, processor 1001 may include a transceiver for implementing receiving and transmitting functions. For example, the transceiver may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuit, interface, or interface circuit for implementing the receiving and transmitting functions may be separate or integrated. The transceiver circuit, interface, or interface circuit may be used for reading and writing code / data, or may be used for transmitting or delivering signals.

[0425] In one implementation, processor 1001 may store a computer program 1003. Computer program 1003, when executed on processor 1001, enables communication device 1000 to perform the method described in the above method embodiment. Computer program 1003 may be embedded in processor 1001, in which case processor 1001 may be implemented by hardware.

[0426] In one implementation, the communication device 1000 may include a circuit that can implement the functions of sending, receiving, or communicating in the aforementioned method embodiments. The processor and transceiver described in the present disclosure can be implemented on an integrated circuit (IC), an analog IC, a radio frequency integrated circuit RFIC, a mixed signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (nMetal-oxide-semiconductor, NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (bipolar junction transistor, BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

[0427] The communication device described in the above embodiment may be an AMF network element, a first AI network element, or a second AI network element, but the scope of the communication device described in this disclosure is not limited thereto, and the structure of the communication device may not be limited to FIG. 16. The communication device may be an independent device or may be part of a larger device. For example, the communication device may be:

[0428] (1) An independent integrated circuit (IC), or chip, or chip system or subsystem;

[0429] (2) a collection of one or more ICs, optionally including a storage component for storing data and computer programs;

[0430] (3) ASIC, such as modem;

[0431] (4) Modules that can be embedded in other devices;

[0432] (5) Receivers, terminal devices, intelligent terminal devices, cellular phones, wireless devices, handheld devices, mobile units, vehicle-mounted devices, network devices, cloud devices, artificial intelligence devices, etc.;

[0433] (6)Others, etc.

[0434] For the case where the communication device can be a chip or a chip system, please refer to Figure 17, which is a structural diagram of a chip provided in an embodiment of the present disclosure.

[0435] As shown in Figure 17, the chip 1100 includes a processor 1101 and an interface 1103. There may be one or more processors 1101, and there may be more than one interface 1103.

[0436] For the case where the chip is used to implement the function of the first AI network element in the embodiment of this disclosure:

[0437] The interface 1103 is used to receive code instructions and transmit them to the processor.

[0438] The processor 1101 is configured to execute code instructions to perform the AI ​​task processing method described in some of the above embodiments.

[0439] For the case where the chip is used to implement the functions of the AMF network element in the embodiments of the present disclosure:

[0440] The interface 1103 is used to receive code instructions and transmit them to the processor.

[0441] The processor 1101 is configured to execute code instructions to perform the AI ​​task processing method described in some of the above embodiments.

[0442] For the case where the chip is used to implement the function of the second AI network element in the embodiments of this disclosure:

[0443] The interface 1103 is used to receive code instructions and transmit them to the processor.

[0444] The processor 1101 is configured to execute code instructions to perform the AI ​​task processing method described in some of the above embodiments.

[0445] Optionally, the chip 1100 further includes a memory 1102 , which is used to store necessary computer programs and data.

[0446] Those skilled in the art will also appreciate that the various illustrative logical blocks and steps listed in the embodiments of the present disclosure may be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented by hardware or software depends on the specific application and the design requirements of the entire system. Those skilled in the art may use various methods to implement the described functionality for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of the present disclosure.

[0447] The present disclosure also provides a readable storage medium having instructions stored thereon, which implement the functions of any of the above method embodiments when executed by a computer.

[0448] The present disclosure also provides a computer program product, which implements the functions of any of the above method embodiments when executed by a computer.

[0449] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer programs. When the computer program is loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0450] Unless the context requires otherwise, throughout the specification and claims, the term "comprise" and its other forms, such as the third person singular form "comprises" and the present participle form "comprising", are to be interpreted as open and inclusive, that is, "including, but not limited to". In the description of the specification, the terms "some embodiments", "exemplary embodiments", etc. are intended to indicate that the particular features, structures, materials or characteristics associated with the embodiment or example are included in at least one embodiment or example of the present disclosure. The schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, the particular features, structures, materials or characteristics may be included in any one or more embodiments or examples in any appropriate manner.

[0451] Those skilled in the art will understand that the various numerical numbers such as first and second involved in the present disclosure are only for the convenience of description and are not used to limit the scope of the embodiments of the present disclosure, and also indicate the order of precedence.

[0452] The at least one in the present disclosure can also be described as one or more, and the multiple can be two, three, four or more, which is not limited in the present disclosure. In the embodiments of the present disclosure, for a technical feature, the technical features in the technical feature are distinguished by "first", "second", "third", "A", "B", "C" and "D", and there is no order of precedence or size between the technical features described by the "first", "second", "third", "A", "B", "C" and "D". "A and / or B" includes the following three combinations: only A, only B, and a combination of A and B.

[0453] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0454] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0455] The above description is merely a specific embodiment of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.

Claims

1. An artificial intelligence (AI) task processing method, characterized in that: The method is performed by a first AI network element and includes: Receive an AI service request message sent by an access and mobility management function (AMF) network element, where the AI ​​service request message is used to indicate the AI ​​service to be provided; Determining at least one AI task according to the AI ​​service request message; Determining a first processing parameter of the first AI network element and a second processing parameter of the second AI network element; A first task executed by the first AI network element and / or a second task executed by the second AI network element in the AI ​​task are determined according to the AI ​​task, the first processing parameter, and the second processing parameter.

2. The method according to claim 1, wherein The determining, according to the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task includes: Determining a target task category for the AI ​​task; Determining, in the AI ​​task, a first task performed by the first AI network element and / or a second task performed by the second AI network element based on the target task category, the first processing parameter, and the second processing parameter, wherein the first processing parameter includes a first task category supported by the first AI network element, and the second processing parameter includes a second task category supported by the second AI network element.

3. The method according to claim 1 or 2, wherein: The AI ​​service request message is further used to indicate a time threshold for obtaining a processing result. The determining, based on the AI ​​task, the first processing parameter, and the second processing parameter, of the first task to be performed by the first AI network element and / or the second task to be performed by the second AI network element in the AI ​​task includes: Determining a first duration required to obtain a first processing result based on the AI ​​task and the first processing parameter, wherein the first processing result is obtained by the first AI network element processing the first task; Determining a second duration required to obtain a second processing result based on the AI ​​task and the second processing parameter, wherein the second processing result is obtained by the second AI network element processing the second task; A first task performed by the first AI network element and / or a second task performed by the second AI network element in the AI ​​task are determined according to the time threshold, the first duration, and the second duration.

4. The method according to claim 3, wherein The determining, based on the time threshold, the first duration, and the second duration, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element includes: In response to satisfying t 0,k ≤T max , determining that the first AI network element performs the kth AI task; and / or Responding to Satisfaction Determining that the i-th second AI network element performs the k-th AI task; Among them, T max is the time threshold; t 0,k The first duration for the first AI network element to process the kth AI task, wherein: D k is the data volume of the kth AI task, r 0,k A computing rate at which the first AI network element processes the kth AI task; The second duration for the i-th second AI network element to process the k-th AI task, in, The computing time required for the i-th second AI network element to process the k-th AI task, is the computing rate of the i-th second AI network element processing the k-th AI task, T i,k For the waiting delay, The upload time of the processing result of the kth AI task uploaded by the i-th second AI network element, The upload rate of the processing result of the kth AI task uploaded by the i-th second AI network element; Wherein, i and k are both integers.

5. The method according to claim 4, wherein The determining the first processing parameter of the first AI network element includes: Determine the computing rate r of the first AI network element processing the kth AI task 0,k ; in, f0 is the computing frequency of the first AI network element, and M is the number of CPU cycles required for the first AI network element to process one bit of task data.

6. The method according to claim 4 or 5, characterized in that The determining the second processing parameter of the second AI network element includes: Determine the computing rate of the i-th second AI network element processing the k-th AI task The upload rate of the processing result of the kth AI task And the waiting delay T i,k ; in, B is bandwidth, P is power, N0 is Gaussian white noise, h i is the wireless channel gain between the i-th second AI network element and the first AI network element, f i is the calculation frequency of the second AI network element of the i-th node, M i The number of CPU cycles required for the i-th second AI network element to process one bit of task data.

7. The method according to any one of claims 1 to 6, characterized in that The determining, according to the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​task includes: Determine the task offloading strategy generation model; The computing frequencies of the first AI network element and the second AI network element, and the wireless channel gain between the second AI network element and the first AI network element, are input into the task offloading policy generation model to generate a target task offloading policy. The target task offloading policy includes, among the AI ​​tasks, a first task performed by the first AI network element and / or a second task performed by the second AI network element. The first processing parameter includes the computing frequency of the first AI network element, and the second processing parameter includes the computing frequency of the second AI network element and the wireless channel gain between the second AI network element and the first AI network element.

8. The method according to claim 7, wherein The determining of the task offloading strategy generation model includes: Initialize model parameters and determine the initial task offloading strategy to generate the model; Determining initial calculation frequencies of the first AI network element and the second AI network element, and an initial wireless channel gain between the second AI network element and the first AI network element; The initial task offloading policy generation model and the initial local model of the first AI network element and / or the second AI network element are jointly trained according to the initial calculation frequency and the initial wireless channel gain to generate the task offloading policy generation model and the local model of the first AI network element and / or the second AI network element.

9. The method according to claim 8, wherein The jointly training the initial task offloading strategy generation model and the initial local model of the first AI network element and / or the second AI network element according to the initial calculation frequency and the initial wireless channel gain includes: Determine the number of iterations T, where T is a positive integer; Determining the first round of input model data as the initial calculation frequency and the initial wireless channel gain; Determining the input model data for round t as the updated calculation frequency and initial wireless channel gain for the first AI network element and / or the second AI network element for round t-1 after updating the initial local model of the first AI network element and / or the second AI network element based on the input model data for round t-1, where 2≤t≤T; Jointly training the initial task offloading strategy generation model and the initial local model of the first AI network element and / or the second AI network element based on each round of input model data; Until the initial task offloading strategy generation model and the initial local models of the first AI network element and / or the second AI network element are jointly trained according to the T-th round of input model data to generate the task offloading strategy generation model and the local models of the first AI network element and / or the second AI network element.

10. The method according to claim 9, wherein The jointly training the initial task offloading strategy generation model and the initial local model of the first AI network element and / or the second AI network element according to the first round of input model data includes: Inputting the initial calculation frequency and the initial wireless channel gain into the initial task offloading strategy generation model to generate an initial task offloading strategy, wherein the initial task offloading strategy includes an initial AI task executed by the first AI network element and / or the second AI network element; determining a processing result of the first AI network element and / or the second AI network element performing the initial AI task, and generating model update parameters, wherein the model update parameters include update parameters of the first AI network element and / or the second AI network element; In response to the model update parameter including a first update parameter of the first AI network element, updating the initial task offloading strategy generation model and / or the initial local model of the first AI network element according to the first update parameter; In response to the model update parameter including the second update parameter of the second AI network element, the second update parameter is distributed to the second AI network element.

11. The method according to any one of claims 1 to 10, characterized in that The method further comprises: In response to determining a first task to be performed by the first AI network element, the first task is performed and a first processing result is generated.

12. The method according to claim 11, wherein In response to determining that the first AI network element performs a first task, executing the first task and generating a first processing result includes: In response to determining that the first task is performed by the first AI network element, receiving a first data set sent by a network function NF network element; The first task is executed according to the first data set to generate a first processing result.

13. The method according to claim 11 or 12, wherein: The method further comprises: Send the first processing result to the AMF network element.

14. The method according to any one of claims 1 to 12, characterized in that The method further comprises: In response to determining the second task to be performed by the second AI network element, sending the second task to the second AI network element; Receive a preliminary processing result sent by the second AI network element, where the preliminary processing result is generated by the second AI network element when performing the second task.

15. The method according to claim 14, wherein The method further comprises: Send a response message to the second AI network element, where the response message is used to indicate that the first AI network element has received the preliminary processing result.

16. The method according to claim 14 or 15, characterized in that The method further comprises: Send the second processing result to the AMF network element, wherein the second processing result is determined by the first AI network element based on the preliminary processing result.

17. The method according to claim 14 or 15, wherein: The method further comprises: In response to determining the first processing result and the preliminary processing result, the first processing result and the preliminary processing result are processed to generate a target processing result.

18. The method according to claim 17, wherein The method further comprises: Send the target processing result to the AMF network element.

19. An artificial intelligence (AI) task processing method, characterized in that: The method is performed by an AMF network element, including: Receiving an AI service establishment request message sent by a terminal device, wherein the AI ​​service request message is used to indicate the AI ​​service required by the terminal device; Sending an AI service request message to a first AI network element, wherein the AI ​​service request message is used to indicate an AI service to be provided, and the AI ​​service request message is used by the first AI network element to determine at least one AI task based on the AI ​​service request message, determine a first processing parameter of the first AI network element, and a second processing parameter of a second AI network element, and determine, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element.

20. The method according to claim 19, wherein The method further comprises: Receive a first processing result sent by the first AI network element, where the first processing result is generated by the first AI network element performing the first task.

21. The method according to claim 19, wherein The method further comprises: Receive a second processing result sent by the first AI network element, wherein the second processing result is determined by the first AI network element based on a preliminary processing result, and the preliminary processing result is generated by the second AI network element performing the second task.

22. The method of claim 19, wherein: The method further comprises: receiving a target processing result sent by the first AI network element, where the target processing result is generated by the first AI network element processing the first processing result and the preliminary processing result after determining the first processing result and the preliminary processing result, the first processing result is generated by the first AI network element performing the AI ​​task, and the preliminary processing result is generated by the second AI network element performing the AI ​​task.

23. An artificial intelligence (AI) task processing method, characterized in that: The method is performed by a second AI network element and includes: Receive a second task sent by the first AI network element, wherein the second AI task is determined by the first AI network element according to the AI ​​task, the determined first processing parameter of the first AI network element, and the determined second processing parameter of the second AI network element, and is executed by the second AI network element and sent to the second AI network element. The AI ​​task is determined by the first AI network element according to the AI ​​service request message sent by the AMF network element, and the AI ​​service request message is used to indicate the AI ​​service to be provided.

24. The method according to claim 23, wherein The method further comprises: Execute the second task and generate a preliminary processing result.

25. The method of claim 24, wherein: The performing of the second task and generating a preliminary processing result includes: receiving a second data set sent by a network function NF network element; The second task is performed according to the second data set to generate the preliminary processing result.

26. The method according to any one of claims 23 to 25, characterized in that The method further comprises: receiving a second update parameter sent by the first AI network element; Update the initial local model of the second AI network element according to the second update parameter.

27. The method according to claim 24 or 25, wherein: The method further comprises: Sending the preliminary processing result to the first AI network element.

28. The method of claim 27, wherein: The method further comprises: Receive a response message sent by the first AI network element, where the response message is used to indicate that the first AI network element has received the preliminary processing result.

29. A communication device, provided at a first AI network element side, characterized in that: include: A transceiver module is configured to receive an AI service request message sent by an access and mobility management function (AMF) network element, wherein the AI ​​service request message is used to indicate an AI service to be provided; a processing module, configured to determine at least one AI task according to the AI ​​service request message; The processing module is further configured to determine a first processing parameter of the first AI network element and a second processing parameter of the second AI network element; The processing module is further configured to determine, in the AI ​​task, a first task performed by the first AI network element and / or a second task performed by the second AI network element based on the AI ​​task, the first processing parameter, and the second processing parameter.

30. A communication device, provided at the AMF network element side, characterized in that: include: a transceiver module, configured to receive an AI service establishment request message sent by a terminal device, wherein the AI ​​service request message is used to indicate the AI ​​service required by the terminal device; The transceiver module is further configured to send an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided. The AI ​​service request message is used by the first AI network element to determine at least one AI task based on the AI ​​service request message, determine a first processing parameter of the first AI network element, and a second processing parameter of the second AI network element, and determine, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element.

31. A communication device, provided at a second AI network element side, characterized in that: include: The transceiver module is configured to receive a second task sent by the first AI network element, wherein the second AI task is determined by the first AI network element to be executed by the second AI network element and sent to the second AI network element based on the AI ​​task, the determined first processing parameter of the first AI network element, and the determined second processing parameter of the second AI network element. The AI ​​task is determined by the first AI network element based on the AI ​​service request message sent by the AMF network element, and the AI ​​service request message is used to indicate the AI ​​service to be provided.

32. A communication device, characterized in that: The device comprises: a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the computer program stored in the memory so that the device performs the method as described in any one of claims 1 to 18; or the processor executes the computer program stored in the memory so that the device performs the method as described in any one of claims 19 to 22; or the processor executes the computer program stored in the memory so that the device performs the method as described in any one of claims 23 to 28.

33. A communication device, characterized in that: include: processor and interface circuits; The interface circuit is used to receive code instructions and transmit them to the processor; The processor is configured to execute the code instructions to perform the method according to any one of claims 1 to 18; or execute the code instructions to execute the method according to any one of claims 19 to 22; or execute the code instructions to execute the method according to any one of claims 23 to 28.

34. A communication system, characterized in that include: AMF network element, a first AI network element, and a second AI network element; The AMF network element is configured to receive an AI service establishment request message sent by a terminal device, wherein the AI ​​service request message is used to indicate the AI ​​service required by the terminal device; and send an AI service request message to the first AI network element, wherein the AI ​​service request message is used to indicate the AI ​​service to be provided; The first AI network element is configured to receive an AI service request message sent by the AMF network element, wherein the AI ​​service request message is used to indicate an AI service to be provided; determine at least one AI task according to the AI ​​service request message; determine a first processing parameter of the first AI network element and a second processing parameter of the second AI network element; and determine, based on the AI ​​task, the first processing parameter, and the second processing parameter, a first task to be performed by the first AI network element and / or a second task to be performed by the second AI network element in the AI ​​service; The first AI network element is further configured to, in response to determining that the second task is performed by the second AI network element, send the second task to the second AI network element; The second AI network element is configured to receive the second task sent by the first AI network element.

35. A computer-readable storage medium for storing instructions, which, when executed, enable the method according to any one of claims 1 to 18 to be implemented; or, when executed, enable the method according to any one of claims 19 to 22 to be implemented; or, when executed, enable the method according to any one of claims 23 to 28 to be implemented.

Images