Service providing method and device, storage medium and electronic equipment

Abstract

The invention provides a service providing method and device, a storage medium and electronic equipment, and relates to the technical field of wireless communication. The method is applied to a wireless access network, and comprises the following steps: receiving a service request sent by a terminal; the service request comprises a data request and an artificial intelligence (AI) computing power request; and if the available computing resources of the wireless access network cannot meet the AI computing power request, sending the service request to an AI management and arrangement function, the AI management and arrangement function sends a data set corresponding to the data request and the AI computing power request to a cloud node matched with the AI computing power request for model training and service rendering, generates a service processing result and sends the service processing result to the wireless access network; the service processing result is used for providing AI-based service for the terminal in the wireless access network. The computing power distribution can be dynamically adjusted, and the computing demand is ensured to be always met, so that the service continuity and the user experience are improved.

Description

Background Art

[0002] At present, mobile communication systems are mainly based on the standard architecture developed by 3GPP (3rd Generation Partnership Project), and most of the implementation process relies on dedicated hardware equipment. This traditional architecture has significant advantages in performance and stability, but the specialization and closedness of hardware resources limit the flexibility and scalability of the system. In recent years, with the rise of artificial intelligence (AI) technology, 3GPP has also carried out a lot of research on the use of AI / ML technology to optimize communication systems. However, the current 3GPP AI is one-to-one oriented to application scenarios, and each working group independently designs AI. This approach leads to the problem of superposition of AI function plug-ins in the network, and AI capabilities such as computing power and models cannot be coordinated across domains.

[0003] In the current network, wireless access networks, core networks, network management systems, and the cloud all deploy computing resources with different capabilities. However, based on the current network architecture, these computing powers are difficult to collaborate across domains, resulting in poor computing resource availability and weak end-to-end intelligent service capabilities.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The present disclosure provides a service providing method and device, a storage medium and an electronic device, which at least to a certain extent overcome the problem that the computing power of related technologies is difficult to perform cross-domain collaboration.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a service providing method is provided. Applied to a wireless access network, comprising:

[0008] Receiving a service request sent by a terminal; the service request includes: a data request and an artificial intelligence (AI) computing power request;

[0009] If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

[0010] In some embodiments, it also includes:

[0011] If the available computing resources of the wireless access network can satisfy the AI ​​computing power request, sending the data request to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request to the wireless access network;

[0012] Receive a data set corresponding to the data request, schedule corresponding computing power according to the AI ​​computing power request to perform model training and business rendering, and generate business processing results;

[0013] According to the business processing result, provide AI-based business services for the terminal.

[0014] In some embodiments, the AI ​​management and orchestration function includes: a data management unit; the data management unit stores a data set; the data set includes: model training data and business rendering data;

[0015] If the available computing resources of the wireless access network can satisfy the AI ​​computing power request, sending the data request to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request to the wireless access network, including:

[0016] Comparing currently available computing resources of the wireless access network with the AI ​​computing power request;

[0017] If the available computing resources of the wireless access network can meet the AI ​​computing power request, the data request is sent to the data management unit of the AI ​​management and orchestration function, so that the data management unit sends the model training data and business rendering data corresponding to the data request to the wireless access network.

[0018] In some embodiments, receiving a data set corresponding to the data request, scheduling corresponding computing power according to the AI ​​computing power request to perform model training and business rendering, and generating business processing results, includes:

[0019] Receiving the model training data and business rendering data;

[0020] Performing AI model training according to the model training data to determine a trained AI model;

[0021] Use the trained AI model to complete model reasoning and determine the model reasoning results;

[0022] Performing service rendering according to the service rendering data to determine the service rendering function;

[0023] A business processing result is determined according to the model reasoning result and the business rendering function.

[0024] In some embodiments, the AI ​​management and orchestration function includes: a data management unit and an AI business orchestration unit; the data management unit stores a data set; the data set includes: model training data and business rendering data;

[0025] If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering and generates a service processing result and sends it to the wireless access network, including:

[0026] If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function uses the data management unit to match the model training data and service rendering data corresponding to the data request, and the AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notify the data management unit to send the model training data and service rendering data to the cloud node for model training and service rendering, and generate a service processing result and send it to the wireless access network.

[0027] In some embodiments, sending the service request to the AI ​​management and orchestration function includes:

[0028] The service request is sent to the radio access network management function, so that the radio access network management function forwards the service request to the AI ​​management and orchestration function.

[0029] In some embodiments, the cloud node is a single cloud node;

[0030] The AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notifies the data management unit to send the model training data and the service rendering data to the cloud node for model training and service rendering, and generates a service processing result and sends it to the wireless access network, including:

[0031] The AI ​​management and orchestration function uses the AI ​​service orchestration unit to send the AI ​​computing power request to the cloud management function, so that the cloud management function matches a single cloud node that meets the AI ​​computing power request from the covered nodes according to the computing power usage of the nodes covered by the cloud management function and sends the node information of the single cloud node to the data management unit; the node information of the single cloud node is used to instruct the data management unit to send the model training data and service rendering data corresponding to the data request to the single cloud node; the single cloud node is used to perform model training and service rendering according to the model training data and service rendering data corresponding to the data request, generate a service processing result and send it to the wireless access network.

[0032] In some embodiments, the single cloud node is used to perform model training and service rendering according to the model training data and service rendering data corresponding to the data request, generate a service processing result and send it to the wireless access network, including:

[0033] The single cloud node performs AI model training according to the model training data, determines the trained AI model, and uses the trained AI model to complete model reasoning to determine the model reasoning result;

[0034] The single cloud node performs service rendering according to the service rendering data and determines a service rendering function;

[0035] The single cloud node determines the service processing result according to the model reasoning result and the service rendering function and sends it to the wireless access network.

[0036] In some embodiments, before receiving the service request sent by the terminal, the method further includes:

[0037] Receiving the registration application sent by the terminal;

[0038] Sending the registration application to a core network, so that the core network completes device authentication and authorization of the terminal according to the registration application, generates a registration response and sends it to the wireless access network;

[0039] The registration response is received, and the registration response is sent to the terminal to complete the terminal registration.

[0040] According to another aspect of the present disclosure, there is also provided a service providing device, which is applied to a wireless access network, including:

[0041] A service request receiving module is used to receive a service request sent by a terminal; the service request includes: a data request and an artificial intelligence AI computing power request;

[0042] A cloud node computing power scheduling module is used to send the service request to the AI ​​management and orchestration function if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

[0043] According to another aspect of the present disclosure, an electronic device is also provided, which includes: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned service providing methods by executing the executable instructions.

[0044] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the service providing method described in any one of the above is implemented.

[0045] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, any one of the above service providing methods is implemented.

[0046] The service providing method and device, storage medium and electronic device provided in the embodiments of the present disclosure, when applied to a wireless access network, include: receiving a service request sent by a terminal; the service request includes: a data request and an artificial intelligence (AI) computing power request; if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set and AI computing power request corresponding to the data request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

[0047] When the terminal for business processing accesses the wireless network, it first determines whether the computing power resources on the wireless access network side are sufficient in the business request sent by the terminal. If the computing power on the wireless access network side is insufficient, the business request is forwarded to the AI ​​management and orchestration function driven by AI, seeking a cloud node that can meet the AI ​​computing power request, and intelligently scheduling the data set corresponding to the data request to the cloud node for model training and business rendering, generating business processing results and sending them to the wireless access network, and the wireless network access network provides AI-based business services to the terminal according to the business processing results. The mechanism of this embodiment can dynamically adjust the computing power allocation to ensure that the computing needs are always met, thereby improving the continuity of the business and user experience. Especially in the face of scenarios with high computing needs, this flexible scheduling scheme avoids the performance degradation caused by insufficient local computing power, configures the business processing results on the wireless access network side, and can coordinate the wireless network to provide low-latency access services, realizing the deep integration of cloud computing resources and wireless access networks, optimizing the transmission path of business data flows, effectively reducing network delays, and enhancing the overall service quality. It can significantly reduce business processing delays, improve the utilization of computing resources, and achieve dual guarantees of high business reliability and performance. Compared with the traditional static or single computing power allocation method, the present invention provides a smarter and more dynamic computing power scheduling solution, reduces resource waste, and improves the service capability and response speed of the network under high load conditions. Through end-to-end AI management and orchestration functions, it is possible to connect the intelligent capabilities and data of cloud nodes distributed in the network, make full use of and coordinate computing resources in different domains, improve the network's intelligent service capabilities and the flexibility of computing power calls; uniformly schedule data resources in the network to achieve efficient use of data. By adopting AI collaborative technology, resource coordination and optimization between terminal domains, wireless network domains, and core network domains are achieved. Through real-time analysis and adjustment of cross-domain data, resources in different domains can be intelligently allocated to optimize overall network performance according to the current network status and business needs. It significantly improves resource utilization and network efficiency in a multi-domain environment, solves the problem of multi-domain collaboration in the prior art, and supports more complex and diverse business scenarios. In addition, this method is based on the evolution of the current network architecture and has good compatibility with the existing technology architecture.

[0048] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0050] Figure 1 A schematic diagram showing the system structure of a service providing method in an embodiment of the present disclosure.

[0051] Figure 2 A schematic diagram of a service providing method in an embodiment of the present disclosure is shown.

[0052] Figure 3 A schematic diagram of an end-to-end AI management and orchestration framework of a service provision method in an embodiment of the present disclosure is shown.

[0053] Figure 4 A terminal registration interaction diagram of a service providing method in an embodiment of the present disclosure is shown.

[0054] Figure 5 An interaction diagram of a wireless access network locally providing a business service in a service providing method according to an embodiment of the present disclosure is shown.

[0055] Figure 6 An interaction diagram showing a wireless access network scheduling cloud node to provide business services in a service providing method in an embodiment of the present disclosure.

[0056] Figure 7 An interaction diagram showing a wireless access network scheduling a single cloud node to provide business services in a service providing method according to an embodiment of the present disclosure.

[0057] Figure 8 An interaction diagram showing a service providing method in an embodiment of the present disclosure for providing business services to a Metaverse terminal.

[0058] Fig. 9 A schematic diagram of a service providing device in an embodiment of the present disclosure is shown.

[0059] Fig.10 A structural block diagram of a computer device showing a service providing method in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0061] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0062] For ease of understanding, before introducing the embodiments of the present disclosure, several terms involved in the embodiments of the present disclosure are first explained as follows:

[0063] End-to-end AI management and orchestration function, E2EAMO: Deployed on the network management system, it logically connects the AI ​​functions of RAN, CN, and the cloud, and is responsible for the unified orchestration and management of network intelligence.

[0064] Radio Access Network, RAN.

[0065] Core Network,CN core network.

[0066] Network Data Analytics Function, NWDAF Network data analysis function: The network element in 5GC is responsible for collecting, analyzing and processing the data generated in the network.

[0067] 3GPP Third Generation Partnership Project.

[0068] Open Radio Access Network Alliance, O-RAN Open Wireless Network Alliance.

[0069] The specific implementation of the embodiment of the present disclosure is described in detail below with reference to the accompanying drawings.

[0070] Figure 1 FIG. 1 shows an exemplary application system architecture diagram to which the service providing method in the embodiment of the present disclosure can be applied. Figure 1 As shown, the system architecture includes a terminal device 101 , a network 102 and a network side device 103 .

[0071] The network 102 is a medium for providing a communication link between the terminal device 101 and the network-side device 103 , and may be a wired network or a wireless network.

[0072] Optionally, the wireless network or wired network described above uses standard communication technology and / or protocol. The network is usually the Internet, but it can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or any combination of a virtual private network). In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Mark-up Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPSec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0073] Optionally, the terminal device in the embodiment of the present disclosure may also be referred to as UE (User Equipment). In a specific implementation, the terminal device may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), a personal digital assistant (Personal Digital Assistant, PDA), a mobile Internet device (Mobile Internet Device, MID), a wearable device (Wearable Device) or a vehicle-mounted device, etc. It should be noted that the specific type of the terminal device is not limited in the embodiment of the present invention.

[0074] The network side device may be a base station, a relay or an access point, etc. The base station may be a base station of 5G or later versions (e.g., 5G NR NB), or a base station in other communication systems (e.g., an eNB base station). It should be noted that the specific type of the network side device is not limited in the embodiments of the present disclosure.

[0075] Those skilled in the art will know that Figure 1The number of terminals, networks, and network-side devices in the example is only illustrative, and any number of terminals, networks, and network-side devices may be provided according to actual needs. The embodiments of the present disclosure are not limited to this.

[0076] Under the above system architecture, a service providing method is provided in an embodiment of the present disclosure, and the method can be executed by any electronic device with computing and processing capabilities.

[0077] In some embodiments, the service provision method provided in the embodiments of the present disclosure may be executed by a terminal device of the above-mentioned system architecture; in other embodiments, the service provision method provided in the embodiments of the present disclosure may be executed by a server in the above-mentioned system architecture; in other embodiments, the service provision method provided in the embodiments of the present disclosure may be implemented by a terminal device and a server in the above-mentioned system architecture through interaction.

[0078] Currently, 3GPP's AI is one-to-one for application scenarios, and each working group independently designs AI. This approach leads to the superposition of AI functions in the network, and the problem that AI capabilities such as computing power and models cannot be coordinated across domains. Although the 3GPP SA (Service and Architecture) plenary level has begun to discuss the research alignment of AI / ML, there is still no end-to-end AI framework and solution.

[0079] The integration of network and intelligence is one of the important directions of 6G network evolution. As network computing power deployment becomes more extensive and decentralized, how to coordinate computing resources in different locations and domains in the network has become an important challenge in 6G network design. There are significant gaps between RAN, CN, and network management systems in terms of deployment location, functional architecture, and hardware standards. Realizing cross-domain AI collaboration from the perspective of AI management and orchestration will be an inevitable requirement for 6G system architecture design, and it is also the key to improving the overall intelligence level of the network, optimizing resource utilization, and improving user experience.

[0080] In view of this, based on the cross-domain intelligent architecture, this patent studies the support process of the cross-domain intelligent architecture for actual business from a practical perspective, and provides support for new business of 6G network endogenous intelligent service. This innovative solution solves the problem of the existing network's business needs such as terminal-side AI, wireless access network-side AI, and core network AI coordination difficulties, and at the same time provides a feasible path for network intelligence to endogenously better serve business.

[0081] Figure 2 A schematic diagram of a service providing method in an embodiment of the present disclosure is shown. Figure 2 As shown, the service providing method provided in the embodiment of the present disclosure, applied to a wireless access network, includes the following steps:

[0082] Step S202: receiving a service request sent by a terminal; the service request includes: a data request and an artificial intelligence AI computing power request;

[0083] Step S204: If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

[0084] When the terminal for business processing accesses the wireless network, it first determines whether the computing power resources on the wireless access network side are sufficient in the business request sent by the terminal. If the computing power on the wireless access network side is insufficient, the business request is forwarded to the AI ​​management and orchestration function driven by AI, seeking a cloud node that can meet the AI ​​computing power request, and intelligently scheduling the data set corresponding to the data request to the cloud node for model training and business rendering, generating business processing results and sending them to the wireless access network, and the wireless network access network provides AI-based business services to the terminal according to the business processing results. The mechanism of this embodiment can dynamically adjust the computing power allocation to ensure that the computing needs are always met, thereby improving the continuity of the business and user experience. Especially in the face of scenarios with high computing needs, this flexible scheduling scheme avoids the performance degradation caused by insufficient local computing power, configures the business processing results on the wireless access network side, and can coordinate the wireless network to provide low-latency access services, realizing the deep integration of cloud computing resources and wireless access networks, optimizing the transmission path of business data flows, effectively reducing network delays, and enhancing the overall service quality. It can significantly reduce business processing delays, improve the utilization of computing resources, and achieve dual guarantees of high business reliability and performance. Compared with the traditional static or single computing power allocation method, the present invention provides a smarter and more dynamic computing power scheduling solution, reduces resource waste, and improves the service capability and response speed of the network under high load conditions. Through end-to-end AI management and orchestration functions, it is possible to connect the intelligent capabilities and data of cloud nodes distributed in the network, make full use of and coordinate computing resources in different domains, improve the network's intelligent service capabilities and the flexibility of computing power calls; uniformly schedule data resources in the network to achieve efficient use of data. By adopting AI collaborative technology, resource coordination and optimization between terminal domains, wireless network domains, and core network domains are achieved. Through real-time analysis and adjustment of cross-domain data, resources in different domains can be intelligently allocated to optimize overall network performance according to the current network status and business needs. It significantly improves resource utilization and network efficiency in a multi-domain environment, solves the problem of multi-domain collaboration in the prior art, and supports more complex and diverse business scenarios. In addition, this method is based on the evolution of the current network architecture and has good compatibility with the existing technology architecture.

[0085] Figure 3 The end-to-end AI management and orchestration framework is shown in Figure 1. Figure 3 As shown, on the basis of designing end-to-end intelligent management and orchestration functions, the global resource scheduling module in the cloud network intelligent layer is used to realize the integrated scheduling and coordination of cloud and network resources in the cross-domain network. This embodiment realizes intelligent resource scheduling and computing power allocation: the computing power allocation strategy is flexibly adjusted according to the needs of the service terminals. When the computing power resources on the wireless access network side are insufficient, the computing power request is intelligently forwarded to the global resource scheduling module to realize the efficient utilization and on-demand allocation of computing power resources. In addition, this embodiment also realizes the integrated orchestration capability of cloud network fusion: through the end-to-end AI management and orchestration module, on the premise of meeting the computing needs of the business and the network access latency requirements, the intelligent fusion scheduling of cloud nodes and wireless network resources is realized to ensure business stability and performance.

[0086] This embodiment involves a global resource scheduling module across RAN, CN, cloud, and cloud network intelligent layer. The main network element functions involved in this technical solution include but are not limited to traditional 3GPP base stations, O-RAN base stations, OAM, CN, and also include RAN intelligent network elements, cloud network intelligent layer and public cloud for 6G networks. Among them, the end-to-end intelligent management and orchestration function is connected with the intelligent functions of different network domains through service interfaces or dedicated interfaces, providing unified network intelligent management and orchestration functions for 6G networks. The global resource scheduling communicates with the cloud intelligent management function, RAN intelligent management function and CN management function across domains through service interfaces, thereby realizing the key enabling technology for 6G end-to-end network intelligent services.

[0087] In the embodiment, before receiving the service request sent by the terminal, the method further includes:

[0088] Receiving the registration application sent by the terminal;

[0089] Sending the registration application to the core network, so that the core network completes device authentication and authorization of the terminal according to the registration application, generates a registration response and sends it to the wireless access network;

[0090] Receive the registration response, send the registration response to the terminal, and complete the terminal registration.

[0091] Figure 4 The terminal registration interaction diagram is shown as Figure 4 As shown in the figure, the terminal registration interaction process includes:

[0092] The terminal sends a registration application to the wireless access network;

[0093] The wireless access network forwards the registration application to the core network;

[0094] The core network performs device registration authentication and authorization on the terminal according to the registration application. After completing device authentication and authorization, the core network generates a registration response and sends it to the wireless access network;

[0095] The wireless access network receives the registration response and sends the registration response to the terminal to complete the terminal registration. The terminal is registered in the wireless access network and the core network. In one embodiment, the terminal may be a Metaverse terminal. The AI ​​computing power request may be the number of GPU computing units.

[0096] In the embodiment, it also includes:

[0097] If the available computing resources of the wireless access network can satisfy the AI ​​computing power request, the data request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request to the wireless access network;

[0098] Receive the data set corresponding to the data request, schedule the corresponding computing power according to the AI ​​computing power request to perform model training and business rendering, and generate business processing results;

[0099] Provide AI-based business services to terminals based on business processing results.

[0100] When judging the service request sent by the terminal, if the computing resources on the wireless access network side are sufficient and the available computing resources of the wireless access network can meet the AI ​​computing request, then the data request is sent to the AI ​​management and orchestration function;

[0101] The AI ​​management and orchestration function receives the data request, retrieves the data set corresponding to the data request, and sends the data set to the wireless access network;

[0102] The wireless access network receives the data set corresponding to the data request, schedules the corresponding computing power according to the AI ​​computing power request to perform model training and service rendering, and generates service processing results; the wireless access network provides AI-based service services to the terminal based on the service processing results.

[0103] When the computing power on the wireless access network side can meet the AI ​​computing power request, there is no need to schedule the computing power of the distributed cloud nodes. Model training and business rendering are directly performed locally on the wireless access network side to generate business processing results, and provide AI-based business services for the terminal. This embodiment designs an intelligent scheduling strategy based on computing power requirements. This strategy uses an AI algorithm to monitor the computing power status of the wireless access network in real time, and makes intelligent decisions on whether to use local computing power or forward computing tasks to the cloud network orchestration module. If the local computing power is sufficient, business processing will be performed directly locally to improve the response speed of business services; if the local computing power is insufficient, the task will be scheduled to the most suitable cloud node for processing to ensure efficient use of resources and timely response to tasks.

[0104] In an embodiment, the AI ​​management and orchestration function includes: a data management unit; the data management unit stores a data set; the data set includes: model training data and business rendering data;

[0105] If the available computing resources of the wireless access network can satisfy the AI ​​computing power request, the data request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request to the wireless access network, including:

[0106] Compare the currently available computing resources of the wireless access network with the AI ​​computing power request;

[0107] If the available computing resources of the wireless access network can meet the AI ​​computing power request, the data request will be sent to the data management unit of the AI ​​management and orchestration function, so that the data management unit will send the model training data and business rendering data corresponding to the data request to the wireless access network.

[0108] Receive the data set corresponding to the data request, schedule the corresponding computing power according to the AI ​​computing power request to perform model training and business rendering, and generate business processing results, including:

[0109] Receive model training data and business rendering data;

[0110] Perform AI model training based on model training data to determine the trained AI model;

[0111] Use the trained AI model to complete model reasoning and determine the model reasoning results;

[0112] Perform business rendering according to business rendering data and determine business rendering functions;

[0113] Determine the business processing results based on the model reasoning results and business rendering functions.

[0114] Figure 5 An interaction diagram of a wireless access network providing local business services is shown, such as Figure 5 As shown in the figure, the main interaction process includes:

[0115] The terminal sends a service request to the wireless access network;

[0116] The wireless access network receives the service request and parses the AI ​​computing power request in the service request; compares the currently available computing resources of the wireless access network with the AI ​​computing power request, and if the available computing resources of the wireless access network can meet the AI ​​computing power request, sends the data request to the data management unit of the AI ​​management and orchestration function through the wireless access network management function;

[0117] The data management unit receives the data request, obtains corresponding model training data and service rendering data according to the data request, and sends the model training data and service rendering data to the wireless access network;

[0118] The wireless access network receives the model training data and service rendering data; the wireless access network performs AI model training based on the model training data and determines the trained AI model; the wireless access network uses the trained AI model to complete model reasoning and determines the model reasoning result; the wireless access network performs service rendering based on the service rendering data and determines the service rendering function; the wireless access network determines the service processing result based on the model reasoning result and the service rendering function. The wireless access network provides AI-based service services to the terminal based on the service processing result.

[0119] In the embodiment, the AI ​​management and orchestration function includes: a data management unit and an AI business orchestration unit; the data management unit stores a data set; the data set includes: model training data and business rendering data;

[0120] If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates service processing results and sends them to the wireless access network, including:

[0121] If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function uses the data management unit to match the model training data and service rendering data corresponding to the data request, and the AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notify the data management unit to send the model training data and service rendering data to the cloud node for model training and service rendering, and generate a service processing result and send it to the wireless access network. Among them, sending the service request to the AI ​​management and orchestration function includes: sending the service request to the wireless access network management function, so that the wireless access network management function forwards the service request to the AI ​​management and orchestration function.

[0122] Figure 6 An interaction diagram of a wireless access network scheduling cloud node to provide business services is shown, such as Figure 6 As shown in the figure, the main interaction process includes:

[0123] The terminal sends a service request to the wireless access network;

[0124] The wireless access network receives the service request and parses the AI ​​computing power request in the service request; compares the currently available computing resources of the wireless access network with the AI ​​computing power request, and if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, sends the service request to the wireless access network management function;

[0125] The wireless access network management function receives the service request and forwards the service request to the AI ​​management and orchestration function;

[0126] The AI ​​management and orchestration function receives the service request and classifies the service request into a data request and an AI computing power request. The AI ​​management and orchestration function includes a data management unit and an AI service orchestration unit. The AI ​​management and orchestration function allocates the data request to the data management unit, and the AI ​​management and orchestration function allocates the AI ​​computing power request to the AI ​​service orchestration unit.

[0127] The data management unit receives the data request, matches the corresponding model training data and business rendering data according to the data request, and waits for the call;

[0128] The AI ​​service orchestration unit receives the AI ​​computing power request, matches the corresponding cloud node from the multiple cloud nodes covered according to the AI ​​computing power request, and notifies the data management unit;

[0129] The data management unit receives the notification and sends the model training data and business rendering data to the cloud node;

[0130] The cloud node receives the model training data and the service rendering data, performs model training and service rendering according to the model training data and the service rendering data, generates a service processing result and sends it to the wireless access network;

[0131] The wireless access network receives the service processing results and provides AI-based service to the terminal based on the service processing results.

[0132] In the embodiment, the cloud node is a single cloud node;

[0133] The AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notify the data management unit to send the model training data and service rendering data to the cloud node for model training and service rendering, and generate service processing results and send them to the wireless access network, including:

[0134] The AI ​​management and orchestration function uses the AI ​​business orchestration unit to send the AI ​​computing power request to the cloud management function, so that the cloud management function matches a single cloud node that meets the AI ​​computing power request from the covered nodes according to the computing power usage of the nodes covered by the cloud management function and sends the node information of the single cloud node to the data management unit; the node information of the single cloud node is used to instruct the data management unit to send the model training data and business rendering data corresponding to the data request to the single cloud node; the single cloud node is used to perform model training and business rendering according to the model training data and business rendering data corresponding to the data request, generate business processing results and send them to the wireless access network.

[0135] Among them, a single cloud node is used to perform model training and service rendering according to the model training data and service rendering data corresponding to the data request, generate service processing results and send them to the wireless access network, including:

[0136] A single cloud node performs AI model training based on model training data, determines the trained AI model, and uses the trained AI model to complete model inference to determine the model inference result;

[0137] A single cloud node performs business rendering based on business rendering data and determines the business rendering function;

[0138] A single cloud node determines the service processing results based on the model reasoning results and service rendering functions and sends them to the wireless access network.

[0139] Figure 7 An interaction diagram of a wireless access network scheduling a single cloud node to provide business services is shown, such as Figure 7 As shown in the figure, the main interaction process includes:

[0140] The terminal sends a service request to the wireless access network;

[0141] The wireless access network receives the service request and parses the AI ​​computing power request in the service request; compares the currently available computing resources of the wireless access network with the AI ​​computing power request, and if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, sends the service request to the wireless access network management function;

[0142] The wireless access network management function receives the service request and forwards the service request to the AI ​​management and orchestration function;

[0143] The AI ​​management and orchestration function receives the service request and classifies the service request into a data request and an AI computing power request. The AI ​​management and orchestration function includes a data management unit and an AI service orchestration unit. The AI ​​management and orchestration function allocates the data request to the data management unit, and the AI ​​management and orchestration function allocates the AI ​​computing power request to the AI ​​service orchestration unit.

[0144] The data management unit receives the data request, matches the corresponding model training data and business rendering data according to the data request, and is ready to wait for the call;

[0145] The AI ​​service orchestration unit receives the AI ​​computing power request. When scheduling a single cloud node, the AI ​​service orchestration unit directly sends the AI ​​computing power request to the cloud management function;

[0146] The cloud management function receives AI computing power requests; the cloud management function covers multiple cloud nodes, and the cloud management function can obtain the computing power usage of each cloud node; the cloud management function matches a single cloud node that meets the AI ​​computing power request from the covered nodes based on the computing power usage of the covered nodes; the cloud management function sends the node information of the matched single cloud node to the data management unit to notify the data management unit; the cloud management function sends the AI ​​computing power request to the single cloud node.

[0147] The data management unit receives the node information of the matched single cloud node, matches the prepared data request with the corresponding model training data and business rendering data, and sends them to the matched single cloud node;

[0148] A single cloud node receives model training data and business rendering data, performs AI model training in the single cloud node according to the model training data, determines the trained AI model, and uses the trained AI model to complete model reasoning to determine the model reasoning result; further, in the single cloud node, performs business rendering according to the business rendering data, determines the business rendering function; further, in the single cloud node, determines the business processing result according to the model reasoning result and the business rendering function and sends it to the wireless access network;

[0149] The wireless access network receives the service processing result, and provides AI-based service to the terminal according to the service processing result.

[0150] In the embodiment, when the cloud node is a cloud computing power chain composed of multiple cloud nodes;

[0151] The AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notify the data management unit to send the model training data and service rendering data to the cloud node for model training and service rendering, and generate service processing results and send them to the wireless access network, including:

[0152] The AI ​​management and orchestration function uses the AI ​​business orchestration unit to send the AI ​​computing power request to the cloud management function, so that the cloud management function matches the cloud computing power chain that meets the AI ​​computing power request from the covered nodes according to the computing power usage of the nodes covered by the cloud management function and sends the node information of multiple cloud nodes in the cloud computing power chain to the data management unit; the node information of multiple cloud nodes in the cloud computing power chain is used to instruct the data management unit to send the model training data and business rendering data corresponding to the data request to the cloud computing power chain; the cloud computing power chain is used to perform model training and business rendering according to the model training data and business rendering data corresponding to the data request, generate business processing results and send them to the wireless access network.

[0153] The cloud computing chain is used to perform model training and service rendering according to the model training data and service rendering data corresponding to the data request, generate service processing results and send them to the wireless access network, including:

[0154] The cloud computing power chain receives the model training data and business rendering data corresponding to the data request, and divides the model training data and business rendering data into multiple data subsets, each of which includes part of the model training data and part of the business rendering data;

[0155] Distribute multiple data subsets to multiple cloud nodes in the cloud computing chain;

[0156] Each of the multiple cloud nodes performs AI model training according to part of the model training data, determines the trained part of the AI ​​model, and uses the trained part of the AI ​​model to complete model reasoning to determine the part of the model reasoning results;

[0157] Perform business rendering according to partial business rendering data and determine partial business rendering functions;

[0158] The cloud computing power chain organizes partial model inference results from multiple cloud nodes into complete model inference results;

[0159] The cloud computing power chain organizes partial business rendering functions in multiple cloud nodes into complete business rendering functions;

[0160] The cloud computing chain will determine the business processing results and send them to the wireless access network based on the complete model reasoning results and complete business rendering functions.

[0161] The embodiment also provides an interaction diagram for a wireless access network to schedule multiple cloud nodes to provide business services, and the main interaction process includes:

[0162] The terminal sends a service request to the wireless access network;

[0163] The wireless access network receives the service request and parses the AI ​​computing power request in the service request; compares the currently available computing resources of the wireless access network with the AI ​​computing power request, and if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, sends the service request to the wireless access network management function;

[0164] The wireless access network management function receives the service request and forwards the service request to the AI ​​management and orchestration function;

[0165] The AI ​​management and orchestration function receives the service request and classifies the service request into a data request and an AI computing power request. The AI ​​management and orchestration function includes a data management unit and an AI service orchestration unit. The AI ​​management and orchestration function allocates the data request to the data management unit, and the AI ​​management and orchestration function allocates the AI ​​computing power request to the AI ​​service orchestration unit.

[0166] The data management unit receives the data request, matches the corresponding model training data and business rendering data according to the data request, and is ready to wait for the call;

[0167] The AI ​​service orchestration unit receives the AI ​​computing power request. When scheduling the cloud computing power chain composed of multiple cloud nodes, the AI ​​service orchestration unit decomposes the AI ​​computing power request into multiple AI computing power sub-requests to provide AI computing power services through cross-node collaboration. The AI ​​service orchestration unit sends the multiple AI computing power sub-requests to the cloud management function.

[0168] The cloud management function receives multiple AI computing power sub-requests. The cloud management function covers multiple cloud nodes, and the cloud management function can obtain the computing power usage of each cloud node; the cloud management function matches multiple cloud nodes that meet multiple AI computing power sub-requests from the covered nodes according to the computing power usage of the covered nodes to form a cloud computing power chain; the cloud management function sends the node information of the cloud computing power chain to the data management unit to notify the data management unit; the cloud management function sends the AI ​​computing power request to the cloud computing power chain.

[0169] The data management unit receives the node information of the cloud computing chain, matches the prepared data request with the corresponding model training data and business rendering data, and sends them to the matched cloud computing chain;

[0170] The cloud computing chain receives the model training data and business rendering data corresponding to the data request, divides the model training data and business rendering data into multiple data subsets, each of which includes part of the model training data and part of the business rendering data; distributes the multiple data subsets to multiple cloud nodes in the cloud computing chain; each of the multiple cloud nodes performs AI model training according to part of the model training data, determines the trained part of the AI ​​model, and uses the trained part of the AI ​​model to complete model reasoning and determine part of the model reasoning results; performs business rendering according to part of the business rendering data, and determines part of the business rendering function; the cloud computing chain organizes part of the model reasoning results in multiple cloud nodes into complete model reasoning results; the cloud computing chain organizes part of the business rendering functions in multiple cloud nodes into complete business rendering functions; the cloud computing chain determines the business processing results based on the complete model reasoning results and the complete business rendering function and sends them to the wireless access network.

[0171] The wireless access network receives the service processing result, and provides AI-based service to the terminal according to the service processing result.

[0172] The disclosed embodiment provides a cross-domain AI collaborative optimization solution. An AI collaborative mechanism is introduced between the terminal domain, wireless network domain, and core network domain to dynamically analyze and intelligently schedule multi-domain resources. While the AI ​​algorithm monitors the network status in real time, it optimizes the resource allocation strategy to ensure coordination and efficient operation among domains. The solution realizes intelligent and refined management of resources, which can significantly improve the overall resource utilization of the network and reduce service response time. This collaborative optimization not only improves network performance, but also reduces performance bottlenecks caused by improper resource allocation, ensuring the stability and efficiency of the service.

[0173] The disclosed embodiments provide an intelligent computing power allocation and scheduling mechanism. When a future business terminal (such as a metaverse terminal) device accesses a wireless network, the system first determines whether the computing power resources on the wireless access network side are sufficient. If sufficient, local computing services and network access are directly provided; otherwise, the computing power request is forwarded to the AI-driven cloud network orchestration module to seek cloud computing nodes that meet the needs and intelligently schedule resources. This mechanism can dynamically adjust the computing power allocation to ensure that computing needs are always met, thereby improving business continuity and user experience. Especially in the face of metaverse scenarios with high computing needs, this flexible scheduling solution avoids performance degradation caused by insufficient local computing power, while effectively reducing network latency and enhancing overall service quality.

[0174] The disclosed embodiment provides a cloud-network integrated orchestration function. Through the end-to-end AI management and orchestration module, the network and computing resource status are monitored in real time, the best cloud node is intelligently selected to handle computing power requirements, and the wireless network is coordinated to provide low-latency access services. This function realizes the deep integration of cloud computing resources and wireless access networks, and optimizes the transmission path of business data flows. This cloud-network fusion solution can significantly reduce the overall latency, improve the efficiency of network transmission and computing, and meet the latency-sensitive business needs of the metaverse. Intelligent orchestration also improves the flexibility and scalability of the system, providing support for more high-performance applications in the future. In addition, by intelligently optimizing the data transmission path, network congestion and latency are further reduced, providing users with a smooth and efficient user experience.

[0175] Compared with the prior art, the cloud-network integrated orchestration function of this embodiment can significantly reduce the service processing delay, improve the utilization of computing resources, and achieve dual guarantees of high service reliability and performance. Compared with the traditional static or single computing power allocation method, the intelligent computing power allocation and scheduling mechanism provides a smarter and more dynamic computing power scheduling solution, reduces resource waste, and improves the service capability and response speed of the network under high load conditions. Compared with traditional domain management and inefficient resource coordination, the cross-domain AI collaborative optimization solution significantly improves the resource utilization and network efficiency in a multi-domain environment, solves the problem of multi-domain collaboration in the prior art, and supports more complex and diverse business scenarios. Compared with the traditional AI model training method, data collaborative management and AI model optimization significantly shortens the data transmission delay and improves the training and reasoning efficiency of the AI ​​model through intelligent data push and computing power allocation. At the same time, the collaborative optimization of computing power and data resources is achieved, providing more efficient and intelligent AI service capabilities for complex application scenarios.

[0176] The disclosed embodiments provide an integrated orchestration function for cloud-network fusion. A cloud-network orchestration module driven by AI is proposed to achieve deep integration and intelligent scheduling of cloud computing nodes and wireless access network resources. The module dynamically selects and allocates resources that meet business needs by analyzing the computing power and delay status of each node in the network in real time, ensuring efficient processing and rapid response of the business. The core innovation of this cloud-network fusion orchestration mechanism in resource scheduling is the intelligent selection of dynamic allocation methods for cloud nodes, as well as low-latency and high-efficiency business support capabilities. The integrated orchestration function can significantly reduce business processing delays, improve the utilization rate of computing resources, and achieve dual guarantees of high business reliability and performance.

[0177] The disclosed embodiments provide an intelligent computing power allocation and scheduling mechanism. An intelligent scheduling strategy based on computing power demand is designed. The strategy uses an AI algorithm to monitor the computing power status of the wireless access network in real time, and makes intelligent decisions on whether to use local computing power or forward computing tasks to the cloud network orchestration module. If the local computing power is insufficient, the task will be quickly scheduled to the most appropriate cloud node for processing to ensure efficient use of resources and timely response to tasks. The flexibility and intelligence of the computing power scheduling mechanism, especially the automatic switching of local and cloud resources, and the key technology of AI control computing power allocation. Compared with traditional static or single computing power allocation methods, it provides a smarter and more dynamic computing power scheduling solution, reduces resource waste, and improves the network's service capabilities and response speed under high load conditions.

[0178] The disclosed embodiments provide a cross-domain AI collaborative optimization solution. AI collaborative technology is used to achieve resource coordination and optimization between the terminal domain, wireless network domain, and core network domain. Through real-time analysis and adjustment of cross-domain data, the solution can intelligently allocate resources in different domains according to the current network status and business needs, and optimize the overall network performance. The AI ​​collaborative optimization technology, especially the cross-domain resource intelligent coordination mechanism and its dynamic analysis and optimization function in a multi-domain environment. Compared with traditional domain management and inefficient resource coordination, it significantly improves resource utilization and network efficiency in a multi-domain environment, solves the problem of multi-domain collaboration in the prior art, and supports more complex and diverse business scenarios.

[0179] The disclosed embodiments provide data collaborative management and AI model optimization. Integrate data management into the AI ​​computing power scheduling mechanism. When allocating computing power, not only the computing power of each node is considered, but also the data management module is used to coordinate the push of data to efficiently transmit the data to the cloud node for AI model training. This module supports the intelligent flow and distribution of data, ensuring that the AI ​​model can complete efficient training and reasoning at appropriate nodes, thereby improving the overall performance of the AI ​​algorithm. A mechanism combining computing power scheduling with data collaborative management, especially a technical solution for combining data push with AI model optimization, and its efficient operation mode in a cloud network environment. Compared with traditional AI model training methods, intelligent data push and computing power allocation significantly shorten the data transmission delay and improve the training and reasoning efficiency of the AI ​​model. At the same time, the collaborative optimization of computing power and data resources is achieved, providing more efficient and intelligent AI service capabilities for complex application scenarios.

[0180] Figure 8 An interaction diagram for providing business services to Metaverse terminals is shown, such as Figure 8As shown, this embodiment takes the Metaverse terminal's demand for intelligence as the starting point, and describes the Metaverse terminal request content including AI computing requirements such as the number of GPUs, and data set requirements for Metaverse business intelligent rendering, where the data includes data for AI model training and business rendering data. At the same time, due to the limitations of its own computing resources, the Metaverse terminal in this solution needs to push the business model reasoning and rendering capabilities to the RAN node after completing the business intelligence model training, so as to provide stable services for the Metaverse terminal through this node. The implementation process of the method is described in detail as follows.

[0181] In the first step, the Metaverse terminal first sends a registration request to the RAN access point. The RAN completes device authentication and authorization at the core network (CN) through a series of standardized processes. Then, the Metaverse terminal sends a service request to the RAN node. As mentioned above, the service request includes AI computing requirements such as the number of GPUs and data set requirements for intelligent rendering of Metaverse services, where the data includes data for AI model training and service rendering data. The RAN node then checks whether its available computing resources meet the requirements. If so, it forwards the data requirements to the data module in end-to-end AI management and orchestration. The data module provides data to the RAN node side for AI model training and rendering requirements related to Metaverse services, and provides AI-based Metaverse services for Metaverse terminals. If both computing requirements and data requirements are not met, the request is directly forwarded to end-to-end AI management and orchestration.

[0182] Step 2: End-to-end AI management and orchestration. After receiving the forwarded business request, data requests and AI-related computing power requests are first classified, and the required data is matched in the data management module. The business request is sent to the cloud intelligent management function according to the decomposition of the AI ​​computing power demand by the AI ​​business orchestration side. In this embodiment, only a single cloud node is considered to provide AI computing power services. If cross-node collaborative provision of AI computing power services is required, the AI ​​business orchestration module will establish a "computing power chain" to plan how the AI ​​business is split, which nodes are specifically trained, and the order of input and output after node training. As shown above, this node considers a single cloud node to provide computing power services, so the AI ​​business orchestration module forwards the AI ​​computing power demand to the cloud intelligent management function. This function module selects a node that can complete the business demand to provide computing power services based on the computing power usage of the existing coverage nodes, and completes the training of the AI ​​model at this node; at the same time, after the cloud intelligent management function selects a node, the selected node information will be synchronized to the data management module. After receiving the information, the data management module pushes the business demand data to the computing power processing node for AI model training and business rendering.

[0183] The third step is to provide computing cloud nodes to complete AI model training, then complete AI model reasoning and required business rendering functions locally according to business needs, and then forward the processed results to the RAN node connected to the Metaverse terminal, and provide AI-based Metaverse services at the RAN node.

[0184] The disclosed embodiments can realize intelligent and dynamic resource allocation and management in a multi-domain network environment, meeting the requirements of high computing demand and low latency application scenarios such as the metaverse. Through AI-driven cloud-network fusion scheduling and data collaborative management, the utilization efficiency of network and computing resources is significantly improved, ensuring the efficient operation and stability of the business, providing strong support for emerging technology applications, and having broad commercialization and application prospects. It can be applied to future 6G networks to help realize cross-domain intelligent collaboration and network endogenous intelligence. It can be used in metaverse applications, smart cities, industrial Internet of Things, Internet of Vehicles and other scenarios. Through intelligent cloud-network fusion scheduling and resource optimization technology, network resource utilization and service performance are improved. It can support emerging high-demand application scenarios such as the metaverse and smart medical care, and provide users with a high-quality experience with low latency and high reliability. At the same time, this technology has the potential to promote the intelligent development of communication networks and create significant economic and innovation value for enterprises and industries.

[0185] It should be noted that the acquisition, storage, use, and processing of data in the technical solution disclosed in this disclosure are in compliance with the relevant provisions of national laws and regulations, and various types of data such as personal identity data, operation data, behavioral data, etc. related to individuals, customers, and groups obtained in the embodiments of this disclosure have all been authorized.

[0186] Based on the same inventive concept, the present disclosure also provides a service providing device in the following embodiments. Since the principle of solving the problem in the device embodiment is similar to that in the above method embodiment, the implementation of the device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0187] Fig. 9 A schematic diagram of a service providing device in an embodiment of the present disclosure is shown. Fig. 9 As shown, the device comprises:

[0188] The service request receiving module 901 is used to receive a service request sent by a terminal; the service request includes: a data request and an artificial intelligence AI computing power request;

[0189] The cloud node computing power scheduling module 902 is used to send the service request to the AI ​​management and orchestration function if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

[0190] It should be noted that the above-mentioned service request receiving module 901 and cloud node computing power scheduling module 902 correspond to S202 to S204 in the method embodiment, and the examples and application scenarios implemented by the above-mentioned modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned method embodiment. It should be noted that the above-mentioned modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0191] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0192] Refer to the following Fig.10 The electronic device 1000 according to this embodiment of the present disclosure is described. Fig.10 The electronic device 1000 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0193] like Fig.10 As shown, the electronic device 1000 is in the form of a general computing device. The components of the electronic device 1000 may include but are not limited to: the at least one processing unit 1010, the at least one storage unit 1020, and a bus 1030 connecting different system components (including the storage unit 1020 and the processing unit 1010).

[0194] Among them, the storage unit stores a program code, and the program code can be executed by the processing unit 1010, so that the processing unit 1010 performs the steps described in the above "Exemplary Method" section of this specification according to various exemplary embodiments of the present disclosure. For example, the processing unit 1010 can perform the following steps of the above method embodiment: applied to a wireless access network, including: receiving a service request sent by a terminal; the service request includes: a data request and an artificial intelligence AI computing power request; if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

[0195] The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 10201 and / or a cache storage unit 10202 , and may further include a read-only storage unit (ROM) 10203 .

[0196] The storage unit 1020 may also include a program / utility 10204 having a set (at least one) of program modules 10205, such program modules 10205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0197] Bus 1030 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0198] The electronic device 1000 may also communicate with one or more external devices 1040 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1000, and / or communicate with any device that enables the electronic device 1000 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 1050. Furthermore, the electronic device 1000 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 1060. As shown, the network adapter 1060 communicates with other modules of the electronic device 1000 via a bus 1030. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0199] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0200] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer program product, which includes: a computer program, which implements the above service providing method when executed by a processor.

[0201] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. A program product capable of implementing the above method of the present disclosure is stored thereon. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above “Exemplary Method” section of this specification.

[0202] More specific examples of computer-readable storage media in the present disclosure may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0203] In the present disclosure, a computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein a readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0204] Alternatively, the program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0205] In a specific implementation, the program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0206] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0207] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0208] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0209] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A service providing method, characterized in that: Applied to wireless access networks, including: Receiving a service request sent by a terminal; the service request includes: a data request and an artificial intelligence (AI) computing power request; If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

2. The service providing method according to claim 1, characterized in that: Also includes: If the available computing resources of the wireless access network can satisfy the AI ​​computing power request, sending the data request to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request to the wireless access network; Receive a data set corresponding to the data request, schedule corresponding computing power according to the AI ​​computing power request to perform model training and business rendering, and generate business processing results; According to the business processing result, provide AI-based business services for the terminal.

3. The service providing method according to claim 2, characterized in that: The AI ​​management and orchestration function includes: a data management unit; the data management unit stores a data set; the data set includes: model training data and business rendering data; If the available computing resources of the wireless access network can satisfy the AI ​​computing power request, sending the data request to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request to the wireless access network, including: Comparing currently available computing resources of the wireless access network with the AI ​​computing power request; If the available computing resources of the wireless access network can meet the AI ​​computing power request, the data request is sent to the data management unit of the AI ​​management and orchestration function, so that the data management unit sends the model training data and business rendering data corresponding to the data request to the wireless access network.

4. The service providing method according to claim 3, characterized in that: Receiving a data set corresponding to the data request, scheduling corresponding computing power according to the AI ​​computing power request to perform model training and business rendering, and generating business processing results, including: Receiving the model training data and business rendering data; Performing AI model training according to the model training data to determine a trained AI model; Use the trained AI model to complete model reasoning and determine the model reasoning results; Performing service rendering according to the service rendering data to determine the service rendering function; A business processing result is determined according to the model reasoning result and the business rendering function.

5. The service providing method according to claim 1, characterized in that: The AI ​​management and orchestration function includes: a data management unit and an AI business orchestration unit; the data management unit stores a data set; the data set includes: model training data and business rendering data; If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering and generates a service processing result and sends it to the wireless access network, including: If the available computing resources of the wireless access network cannot meet the AI ​​computing power request, the service request is sent to the AI ​​management and orchestration function, so that the AI ​​management and orchestration function uses the data management unit to match the model training data and service rendering data corresponding to the data request, and the AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notify the data management unit to send the model training data and service rendering data to the cloud node for model training and service rendering, and generate a service processing result and send it to the wireless access network.

6. The service providing method according to claim 5, characterized in that: Send the business request to the AI ​​management and orchestration functions, including: The service request is sent to the radio access network management function, so that the radio access network management function forwards the service request to the AI ​​management and orchestration function.

7. The service providing method according to claim 5, characterized in that: The cloud node is a single cloud node; The AI ​​management and orchestration function uses the AI ​​service orchestration unit to match the corresponding cloud node according to the AI ​​computing power request and notifies the data management unit to send the model training data and the service rendering data to the cloud node for model training and service rendering, and generates a service processing result and sends it to the wireless access network, including: The AI ​​management and orchestration function uses the AI ​​service orchestration unit to send the AI ​​computing power request to the cloud management function, so that the cloud management function matches a single cloud node that meets the AI ​​computing power request from the covered nodes according to the computing power usage of the nodes covered by the cloud management function and sends the node information of the single cloud node to the data management unit; the node information of the single cloud node is used to instruct the data management unit to send the model training data and service rendering data corresponding to the data request to the single cloud node; the single cloud node is used to perform model training and service rendering according to the model training data and service rendering data corresponding to the data request, generate a service processing result and send it to the wireless access network.

8. The service providing method according to claim 7, characterized in that: The single cloud node is used to perform model training and service rendering according to the model training data and service rendering data corresponding to the data request, generate a service processing result and send it to the wireless access network, including: The single cloud node performs AI model training according to the model training data, determines the trained AI model, and uses the trained AI model to complete model reasoning to determine the model reasoning result; The single cloud node performs service rendering according to the service rendering data and determines a service rendering function; The single cloud node determines the service processing result according to the model reasoning result and the service rendering function and sends it to the wireless access network.

9. The service providing method according to claim 1, characterized in that: Before receiving the service request sent by the terminal, it also includes: Receiving the registration application sent by the terminal; Sending the registration application to a core network, so that the core network completes device authentication and authorization of the terminal according to the registration application, generates a registration response and sends it to the wireless access network; The registration response is received, and the registration response is sent to the terminal to complete the terminal registration.

10. A service providing device, characterized in that: Applied to wireless access networks, including: A service request receiving module is used to receive a service request sent by a terminal; the service request includes: a data request and an artificial intelligence AI computing power request; A cloud node computing power scheduling module is used to send the service request to the AI ​​management and orchestration function if the available computing resources of the wireless access network cannot meet the AI ​​computing power request, so that the AI ​​management and orchestration function sends the data set corresponding to the data request and the AI ​​computing power request to the cloud node matching the AI ​​computing power request for model training and service rendering, and generates a service processing result and sends it to the wireless access network; the service processing result is used to provide AI-based service services for the terminal in the wireless access network.

11. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the service providing method according to any one of claims 1 to 7 by executing the executable instructions.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the service providing method according to any one of claims 1 to 7 is implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the service providing method according to any one of claims 1 to 7 is implemented.

Images