WO2023206049A1 - Ai服务执行方法、装置、网元、存储介质及芯片 - Google Patents

Ai服务执行方法、装置、网元、存储介质及芯片 Download PDF

Info

Publication number
WO2023206049A1
WO2023206049A1 PCT/CN2022/089128 CN2022089128W WO2023206049A1 WO 2023206049 A1 WO2023206049 A1 WO 2023206049A1 CN 2022089128 W CN2022089128 W CN 2022089128W WO 2023206049 A1 WO2023206049 A1 WO 2023206049A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
network element
upf network
request message
service execution
Prior art date
Application number
PCT/CN2022/089128
Other languages
English (en)
French (fr)
Inventor
陈栋
何智斌
Original Assignee
北京小米移动软件有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京小米移动软件有限公司 filed Critical 北京小米移动软件有限公司
Priority to CN202280001396.4A priority Critical patent/CN117296374A/zh
Priority to PCT/CN2022/089128 priority patent/WO2023206049A1/zh
Publication of WO2023206049A1 publication Critical patent/WO2023206049A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements

Definitions

  • the present disclosure relates to the field of communications, and in particular to an AI service execution method, device, network element, storage medium and chip.
  • AI technology Artificial Intelligence, artificial intelligence
  • 5G fifth generation mobile communication system
  • AI technology is highly overlapping.
  • 5G next generation mobile communication system
  • relevant research was carried out after the architecture of 5G networks was determined. Therefore, among related technologies, the application of AI technology based on 5G networks is usually a plug-in application, that is, in traditional Optimizing and transforming the network architecture, this application method lacks a unified AI service work and technical framework. It usually simply superimposes AI services on the existing network process, and cannot realize the tight coupling of AI technology and 5G network.
  • the present disclosure provides an AI service execution method, device, network element, storage medium and chip.
  • an AI service execution method is provided, applied to a session management function SMF network element, and the method includes:
  • the AI service request message is used to indicate the AI service requested by the user equipment
  • an AI service execution method which is applied to the policy control function PCF network element.
  • the method includes:
  • the AI service execution policy is used to instruct the SMF network element to control the target UPF network element to execute the AI service according to the AI service execution policy.
  • the target The UPF network element is one or more of the plurality of UPF network elements.
  • an AI service execution method which is applied to the user plane function UPF network element.
  • the method includes:
  • the AI service execution policy is determined by the SMF network element based on the attribute information of multiple user plane function UPF network elements and the AI service request message.
  • the request message is used to indicate the AI service requested by the user device;
  • an AI service execution method is provided, applied to user equipment, and the method includes:
  • the AI service request message is used to indicate the AI service requested by the user equipment.
  • the AI service request message is used to instruct the SMF network element to perform the following tasks according to multiple user planes.
  • the attribute information of the functional UPF network element and the AI service request message determine the AI service execution strategy, and control the target UPF network element to execute the AI service according to the AI service execution strategy, and the target UPF network element is the plurality of One or more of the UPF network elements;
  • an AI service execution device which is applied to the session management function SMF network element.
  • the device includes:
  • a receiving module configured to receive an AI service request message, where the AI service request message is used to indicate the AI service requested by the user equipment;
  • a determination module configured to determine the AI service execution strategy based on the attribute information of multiple user plane function UPF network elements and the AI service request message;
  • the execution module is configured to control a target UPF network element to execute the AI service according to the AI service execution policy, and the target UPF network element is one or more of the plurality of UPF network elements.
  • an AI service execution device which is applied to the policy control function PCF network element.
  • the device includes:
  • a receiving module configured to receive attribute information of multiple user plane function UPF network elements and an AI service request message of the user equipment sent by the session management function SMF network element, where the AI service request message is used to indicate the AI service requested by the user equipment.
  • a determination module configured to determine an AI service execution strategy based on the attribute information of the plurality of UPF network elements and the AI service request message;
  • a sending module configured to send the AI service execution policy to the SMF network element, where the AI service execution policy is used to instruct the SMF network element to control the target UPF network element to execute the AI service execution policy according to the AI service execution policy.
  • AI service, the target UPF network element is one or more of the multiple UPF network elements.
  • an AI service execution device which is applied to the user plane function UPF network element.
  • the device includes:
  • the receiving module is configured to receive the AI service execution policy sent by the session management function SMF network element.
  • the AI service execution policy is determined by the SMF network element based on the attribute information of multiple user plane function UPF network elements and the AI service request message. , the AI service request message is used to indicate the AI service requested by the user equipment;
  • An execution module configured to execute the AI service according to the AI service execution policy.
  • an AI service execution device is provided, applied to user equipment, and the device includes:
  • a sending module configured to send an AI service request message to the session management function SMF network element, where the AI service request message is used to indicate the AI service requested by the user equipment, and the AI service request message is used to indicate the SMF network element.
  • the element determines the AI service execution strategy based on the attribute information of multiple user plane function UPF network elements and the AI service request message, and controls the target UPF network element to execute the AI service according to the AI service execution strategy.
  • the target UPF network is one or more of the plurality of UPF network elements;
  • the receiving module is configured to receive the service execution result of the AI service sent by the SMF network element.
  • a session management function SMF network element including:
  • Processor memory used to store instructions executable by the processor
  • the processor is configured to implement the steps of any one of the methods in the first aspect when executing the executable instructions.
  • a policy control function PCF network element including:
  • Processor memory used to store instructions executable by the processor
  • the processor is configured to implement the steps of the method in any one of the second aspects when executing the executable instructions.
  • a user plane function UPF network element including:
  • Processor memory used to store instructions executable by the processor
  • the processor is configured to implement the steps of the method in any one of the third aspects when executing the executable instructions.
  • a user equipment including:
  • Processor memory used to store instructions executable by the processor
  • the processor is configured to implement the steps of any one of the methods described in the fourth aspect when executing the executable instructions.
  • a thirteenth aspect provides a computer-readable storage medium on which computer program instructions are stored.
  • the steps of any one of the methods described in the first aspect are implemented, or the program instructions are When executed by a processor, the steps of any one of the methods described in the second aspect are implemented, or when the program instructions are executed by a processor, the steps of any one of the methods described in the third aspect are implemented, or the program instructions are processed When the processor is executed, the steps of any one of the methods described in the fourth aspect are implemented.
  • a fourteenth aspect provides a chip, including a processor and an interface; the processor is used to read instructions to execute the method described in any one of the first aspects, or the processor is used to read instructions to execute the third aspect.
  • the user equipment sends an AI service request message to the SMF network element.
  • the AI service request message is used to indicate the AI service requested by the user equipment.
  • the SMF network element is based on the received AI service request message and multiple UPF networks.
  • the attribute information of the element determines the AI service execution policy for executing the AI service, and determines one or more of the multiple UPF network elements as target UPF network elements, and executes it through the target UPF network element according to the AI service execution policy.
  • the AI service Through the above method, the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • Figure 1 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 2 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 3 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 4 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 5 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 6 is a flow chart of a policy determination method according to an exemplary embodiment
  • Figure 7 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 8 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 9 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 10 is a flow chart of an AI service execution method according to an exemplary embodiment
  • Figure 11 is a block diagram of an AI service execution device according to an exemplary embodiment
  • Figure 12 is a block diagram of an AI service execution device according to an exemplary embodiment
  • Figure 13 is a block diagram of an AI service execution device according to an exemplary embodiment
  • Figure 14 is a block diagram of an AI service execution device according to an exemplary embodiment
  • Figure 15 is a block diagram of an AI service execution device according to an exemplary embodiment.
  • first, second, etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other and do not imply a specific order or importance. In fact, expressions such as “first” and “second” can be used interchangeably.
  • first message frame may also be called a second message frame, and similarly, the second message frame may also be called a first message frame.
  • FIG. 1 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 1, the AI service execution method is used for SMF (Access and Mobility Management Function) In the network element, the method includes the following steps.
  • SMF Access and Mobility Management Function
  • step S101 an AI service request message is received, where the AI service request message is used to indicate the AI service requested by the user equipment.
  • the AI service request message may be initiated by a UE (User Equipment).
  • the UE When the UE needs to request a certain AI service from the core network of the 5G network, the UE generates the AI service request.
  • the AI service request message may include at least AI service parameters indicating the AI service requested by the UE.
  • the AI service parameters may include, for example, AI service identification (AI Service ID), AI service type (AI Service Type), user At least one of the AI data (AI data) used to execute the AI service.
  • the AI service request message may also include other information, such as DNN (Data Network Name), etc.
  • the UE can send the AI service request message to the RAN (Radio Access Network), which can also be called (R)AN.
  • the RAN Radio Access Network
  • the AMF Session Management Function
  • the AMF forwards the AI service request message to the SMF network element. It can be understood that the UE sends the AI service request message. After being sent to the RAN, it is transparently transmitted to the SMF network element through the RAN and AMF network elements in sequence.
  • step S102 the AI service execution policy is determined based on the attribute information of multiple UPF network elements and the AI service request message.
  • the attribute information of the UPF network element can include the AI services that the UPF can support, so that the attribute information of multiple UPF network elements can be determined.
  • the AI services that each UPF network element among multiple UPF network elements can support can determine the AI service requested by the UE according to the AI service request message, thereby determining one or more AI services that can be used to perform the AI service requested by the UE.
  • UPF serves as the following target UPF network element and determines the corresponding AI service execution strategy.
  • the SMF network element can query the NRF (Network Repository Function) network element to obtain the attribute information of the multiple UPF network elements.
  • the NRF network element stores a UPF list.
  • the NRF network element By querying the NRF network element to obtain the UPF list, the multiple UPF network elements can be determined and attribute information of the multiple UPF network elements can be obtained.
  • the plurality of UPF network elements may be part of the UPF network elements or all of the UPF network elements in the UPF list.
  • the SMF network element can determine the AI service execution strategy based on the attribute information of the multiple UPF network elements and the AI service request message.
  • the AI service execution policy may be a PCC (Policy Control and Charging) rule
  • the PCC rule may include at least one or more of the following: the rule identifier of the PCC rule, the AI service The identifier of the assigned target UPF, the priority of the target UPF network element, and other information.
  • a target UPF network element is controlled to execute the AI service according to the AI service execution policy, and the target UPF network element is one or more of the plurality of UPF network elements.
  • the SMF network element After obtaining the AI service execution policy, the SMF network element establishes a connection with the target UPF network element and sends the AI service execution policy to the target UPF network element.
  • the target UPF network element executes what the UE does according to the AI service execution policy.
  • the requested AI service For example, after obtaining the AI service execution policy, the SMF network element establishes a connection with the target UPF network element and sends the AI service execution policy to the target UPF network element.
  • the target UPF network element executes what the UE does according to the AI service execution policy.
  • the requested AI service The requested AI service.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG. 2 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 2, the AI service execution method is used in a PCF network element. The method includes the following steps.
  • step S201 attribute information of multiple UPF network elements and an AI service request message of the UE sent by the SMF network element are received.
  • the AI service request message is used to indicate the AI service requested by the UE.
  • the AI service request message and the method for the UE to send the AI service request message to the SMF network element are the same as in step S101, and reference can be made to the content in step S101, which will not be described again.
  • the SMF network element After acquiring the attribute information of the multiple UPF network elements, the SMF network element sends the attribute information of the multiple UPF network elements and the AI service request message sent by the UE to the PCF network element.
  • step S202 an AI service execution policy is determined based on the attribute information of the multiple UPF network elements and the AI service request message.
  • the PCF network element After receiving the attribute information of the multiple UPF network elements and the AI service request message sent by the SMF network element, the PCF network element analyzes the attribute information of the multiple UPF network elements and the instructions of the AI service request message. AI service, determine one or more UPFs that can be used to execute the AI service requested by the UE as the target UPF network element, and determine the corresponding AI service execution strategy.
  • the AI service execution strategy may be the above-mentioned PCC rule, and The PCC rules described in step S102 are the same and will not be described again.
  • step S203 the AI service execution policy is sent to the SMF network element.
  • the AI service execution policy is used to instruct the SMF network element to control the target UPF network element to execute the AI service according to the AI service execution policy.
  • the target UPF network element The element is one or more of the multiple UPF network elements.
  • the SMF network element After obtaining the AI service execution policy, the SMF network element establishes a connection with the target UPF network element and sends the AI service execution policy to the target UPF network element.
  • the target UPF network element executes what the UE does according to the AI service execution policy.
  • the requested AI service For example, after obtaining the AI service execution policy, the SMF network element establishes a connection with the target UPF network element and sends the AI service execution policy to the target UPF network element.
  • the target UPF network element executes what the UE does according to the AI service execution policy.
  • the requested AI service The requested AI service.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG 3 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 3, the AI service execution method is used in a UPF network element. The method includes the following steps.
  • step S301 receive the AI service execution policy sent by the SMF network element.
  • the AI service execution policy is determined by the SMF network element based on the attribute information of multiple UPF network elements and the AI service request message.
  • the AI service request message is used to Indicates the AI service requested by the UE.
  • the AI service request message and the method for the UE to send the AI service request message to the SMF network element are the same as in step S101, and reference can be made to the content in step S101, which will not be described again.
  • the method for determining the AI service execution policy may refer to steps S102 and S202, which will not be described again.
  • step S302 the AI service is executed according to the AI service execution policy.
  • the AI service execution policy may be the above-mentioned PCC rule, which is the same as the PCC rule described in step S102 and will not be described again.
  • the multiple target UPF network elements include a primary UPF network element and at least one secondary UPF network element.
  • the SMF network element can first send the AI service execution policy to the main UPF network element, and the main UPF network element delivers the AI service execution policy to each secondary UPF network element, so that the main UPF network element and the secondary UPF network element allocate according to their own Enter the AI service execution strategy to execute the AI service.
  • the primary UPF network element may receive an execution result obtained after the secondary UPF among the multiple target UPFs executes the AI service according to the AI service execution policy.
  • the execution result of the primary UPF network element and the execution result of the secondary UPF network element are aggregated to obtain the service execution result of the AI service.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG 4 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 4, the AI service execution method is used in a UE, and the method includes the following steps.
  • an AI service request message is sent to the SMF network element.
  • the AI service request message is used to indicate the AI service requested by the UE.
  • the AI service request message is used to indicate that the SMF network element is based on the attributes of multiple UPF network elements.
  • the information and the AI service request message determine the AI service execution policy, and control the target UPF network element to execute the AI service according to the AI service execution policy.
  • the target UPF network element is one or more of the multiple UPF network elements.
  • the AI service request message and the method for the UE to send the AI service request message to the SMF network element are the same as in step S101, and reference can be made to the content in step S101, which will not be described again.
  • the method for determining the AI service execution policy may refer to steps S102 and S202, which will not be described again.
  • the method for the target UPF network element to execute the AI service according to the AI service execution policy may refer to steps S301 and S302, which will not be described again.
  • step S402 the service execution result of the AI service sent by the SMF network element is received.
  • the target UPF network element executes the AI service according to the AI service execution policy and obtains the service execution result of the AI service. Then the target UPF network element sends the service execution result to the SMF network element, and the SMF network element sends the service execution result.
  • the SMF network element sending the service execution result to the UE may include: the SMF network element first sends the service execution result to the AMF network element, the AMF network element forwards the service execution result to the RAN, and then the RAN sends the service execution result Forwarded to the UE, the UE obtains the service execution result of the AI service.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG. 5 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 5, the AI service execution method includes the following steps.
  • step S501 the SMF network element receives an AI service request message.
  • the AI service request message is used to indicate the AI service requested by the UE.
  • the AI service request message may be initiated by the UE.
  • the UE when the UE needs to request a certain AI service from the core network of the 5G network, the UE generates the AI service request message.
  • AI service request message It may include at least AI service parameters used to indicate the AI service requested by the UE.
  • AI service parameters may include, for example, AI service identification (AI Service ID), AI service type (AI Service Type), and AI data used to execute the AI service. (AI data) at least one.
  • the AI service types may include: video AI services (video), such as image recognition, audio AI services (audio), such as speech recognition, text AI services (text), such as natural language processing, etc.
  • the AI service request message may also include other information, such as one or more of DNN, device information of the UE, user information of the UE, and access type, where
  • the device information of the UE may include, for example, one or more of device identification, device model, and location information of the device.
  • the UE may send the AI service request message to the RAN.
  • the RAN forwards the AI service request message to the AMF network element.
  • the AMF network element Forwarding the AI service request message to the SMF network element can be understood to mean that after the UE sends the AI service request message to the RAN, it is transparently transmitted to the SMF network element through the RAN and AMF network elements in sequence.
  • the AMF network element can store the above-mentioned AI data (AI data) in the AI service request message in the UDSF (Unstructured Data Storage Network Function, unstructured data storage function) network element , used in the subsequent process, the UPF network element calls the AI data from the UDSF network element when executing the AI service.
  • the SMF network element may store the AI data in the AI service request message in the UDSF network element.
  • step S502 the SMF network element obtains attribute information of the multiple UPF network elements.
  • the SMF network element can query the NRF network element to obtain the attribute information of the multiple UPF network elements.
  • the method for the SMF network element to obtain the attribute information of the multiple UPF network elements from the NRF network element can refer to step S102, which will not be described again.
  • the attribute information of the UPF network element may include: the AI service identifier, AI service type and other information of the AI service that the UPF can support, and may also include the location information of the UPF.
  • step S503 the SMF network element determines the AI service execution policy based on the attribute information of multiple UPF network elements and the AI service request message.
  • FIG. 6 is a flowchart of a policy determination method according to an exemplary embodiment. As shown in Figure 6, step S503 may include the following steps.
  • step S5031 the SMF network element sends the attribute information of the multiple UPF network elements and the AI service request message to the PCF network element.
  • step S5032 the PCF network element determines the AI service execution policy based on the attribute information of the multiple UPF network elements and the AI service request message.
  • the PCF network element After receiving the attribute information of the multiple UPF network elements and the AI service request message sent by the SMF network element, the PCF network element analyzes the attribute information of the multiple UPF network elements and the instructions of the AI service request message. AI service, determine one or more UPFs that can be used to execute the AI service requested by the UE as the target UPF network element, and determine the corresponding AI service execution strategy.
  • the AI service execution policy may be the above-mentioned PCC rule, and the PCC rule may include at least one or more of the following: the rule identifier of the PCC rule, the identifier of the target UPF assigned to the AI service, the identifier of the target UPF network element Priority and other information.
  • the AI service request message may include the above-mentioned AI service identifier and AI service type
  • the attribute information of the UPF network element may include information such as the AI service identifier and AI service type of the AI service that the UPF can support
  • the above The steps to determine the execution strategy of the AI service may include the following steps:
  • At least one target UPF network element matching the AI service identifier and AI service type is determined from the multiple UPF network elements.
  • the AI service identifier and AI service type of the AI service that each UPF in the multiple UPF network elements can support can be compared with the AI service identifier and AI service type in the AI service request message of the UE to determine At least one target UPF network element matching the AI service request message.
  • the priority of the at least one target UPF network element can be determined based on the AI services that each UPF target can support and the AI services requested by the AI service request message.
  • the priority of the UPF network element may include whether the UPF is a primary UPF or a secondary UPF, and may also include the priority of the execution order of the at least one target UPF network element when executing the AI service.
  • the location of the UE may also be considered.
  • the AI service request message may also include the location information of the device.
  • the attribute information of the network element may also include the location information of the UPF. Therefore, the AI service identifier, AI service type, and location information of the AI service that each UPF in the multiple UPF network elements can support can be combined with the AI service of the UE. Compare the AI service identifier, AI service type, and device location information in the request message, and determine that one UPF or multiple UPFs that match the above information are the target UPF network elements.
  • the PCF network element can store the AI service execution strategy in UDR (User Data Repository, user data register), and perform step 5033.
  • UDR User Data Repository, user data register
  • step S5033 the PCF network element sends the AI service execution policy to the SMF network element.
  • step S504 the SMF network element controls the target UPF network element to execute the AI service according to the AI service execution policy.
  • FIG. 7 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 7, step S504 may include the following steps.
  • step S5041 the SMF network element sends the AI service execution policy to the target UPF network element.
  • step S5042 the target UPF network element calls the AI data corresponding to the AI service from the UDSF network element.
  • step S5043 the target UPF network element executes the AI service according to the AI service execution policy.
  • target UPFs There may be one or more target UPFs. When there are multiple target UPFs, they may include a primary UPF network element and at least one secondary UPF network element.
  • the SMF network element can first send the AI service execution policy to the main UPF network element, and the main UPF network element delivers the AI service execution policy to each secondary UPF network element, so that the main UPF network element and the secondary UPF network element allocate according to their own Enter the AI service execution strategy to execute the AI service.
  • the AI service can be a single AI service or multiple AI services. If there are multiple AI services, the multiple AI services can jointly complete an AI function, in which each AI service participates.
  • the AI sub-functions are different and may be a single AI sub-function or a combination of multiple AI sub-functions.
  • multiple AI services can also complete multiple AI functions together.
  • UPFs can perform distributed tasks such as federated learning through the 5G N9 interface, and store the output data in UDSF to realize data sharing between UPFs.
  • the AI service requested by the UE is multiple AI services
  • the multiple AI services can be respectively executed by the primary UPF network element and the at least one secondary UPF network element.
  • the primary UPF network element The secondary UPF network element executes one or more AI services among the multiple AI services according to the AI services it supports. For example, different AI services are used to execute different AI algorithms.
  • AI service 1 is used to execute SVM (Support Vector Machine, Support Vector Machine) algorithm
  • AI service 2 is used to execute random forest
  • AI service 3 is used to execute PCA (Principal Component Analysis, principal component analysis) dimensionality reduction algorithm
  • AI service 4 is used to perform K-Means aggregation Class algorithms, etc., thereby enabling multiple UPF network elements to execute multiple different algorithms in a distributed manner through multiple AI services.
  • the priority of the target UPF network element described above can be determined based on the execution order of different UPF network elements when executing different algorithms.
  • the AI service requested by the UE is an AI service
  • the primary UPF network element and the secondary UPF network element can be used to execute part of the AI service, for example, corresponding to the AI service.
  • the AI algorithm is divided into multiple parts (or multiple steps), and the primary UPF network element and the secondary UPF network element respectively execute one or more of the multiple parts (or execute one or more of the multiple steps respectively). steps).
  • the priority of the target UPF network element mentioned above can be the execution order of different UPF network elements when executing different algorithms.
  • the priority of the target UPF network element described above can be determined based on the execution order of multiple parts (or multiple steps) of the AI algorithm.
  • the corresponding AI service execution policy can be generated for the AI service requested by the UE, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a workflow for applying AI services based on the 5G core network. , able to achieve tight coupling of AI technology and 5G networks.
  • FIG 8 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 8, the AI service execution method includes the following steps.
  • step S601 the SMF network element receives an AI service request message.
  • the AI service request message is used to indicate the AI service requested by the UE.
  • step S602 the SMF network element obtains the attribute information of the multiple UPF network elements.
  • step S603 the SMF network element determines the AI service execution policy based on the attribute information of multiple user plane function UPF network elements and the AI service request message.
  • step S604 the SMF network element controls a target UPF network element to execute the AI service according to the AI service execution policy, and the target UPF network element is one or more of the plurality of UPF network elements.
  • steps S601 to S604 are the same as steps S501 to S504. Please refer to the contents of steps S501 to S504, which will not be described again.
  • step S605 the target UPF network element sends the service execution result of the AI service obtained by executing the AI service according to the AI service execution policy to the SMF network element.
  • the primary UPF network element among the target UPF network elements receives the secondary UPF among the multiple target UPFs according to the AI service.
  • the primary UPF network element aggregates the execution results of the primary UPF network element and the execution results of the secondary UPF network element to obtain the service execution result of the AI service. It can be understood that the execution results of the main UPF network element and the execution results of the secondary UPF network element can be part of the results of executing the above-mentioned AI services.
  • the main UPF network element needs to aggregate the execution results to obtain the final service execution result. . For example, since the AI data corresponding to the called AI service is stored in the UDSF network element when the UPF network element executes the AI service, it is unstructured data. Therefore, optionally, each UPF network element can store its own data.
  • the execution results are also stored in the UDSF network element, so the main UPF network element needs to perform structured processing on the execution results in the process of aggregating each execution result to obtain the aggregated service execution results.
  • step S606 the SMF network element sends the service execution result to the UE.
  • the SMF network element may send the service execution result to the AMF network element, and the AMF network element forwards the service execution result to the RAN, and then forwards the RAN to the UE.
  • the corresponding AI service execution policy can be generated for the AI service requested by the UE, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a workflow for applying AI services based on the 5G core network. , able to achieve tight coupling of AI technology and 5G networks.
  • FIG 9 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 9, the AI service execution method includes the following steps.
  • step S701 the SMF network element receives an AI service request message.
  • the AI service request message is used to indicate the AI service requested by the UE.
  • step S702 the SMF network element obtains attribute information of the multiple UPF network elements.
  • step S703 the SMF network element determines the AI service execution policy based on the attribute information of multiple UPF network elements and the AI service request message.
  • step S704 the SMF network element controls a target UPF network element to execute the AI service according to the AI service execution policy, and the target UPF network element is one or more of the plurality of UPF network elements.
  • step S705 the target UPF network element sends the service execution result of the AI service obtained by executing the AI service according to the AI service execution policy to the SMF network element.
  • step S706 the SMF network element sends the service execution result to the UE.
  • steps S701 to S706 are the same as those of steps S601 to S606. Please refer to the contents of steps S601 to S606, which will not be described again.
  • step S707 the SMF network element receives the confirmation message sent by the UE based on the service execution result.
  • the UE replies with the confirmation message for the service execution result.
  • the UE can send the confirmation message to the RAN, and the RAN forwards the confirmation message to the AMF network element, and then the AMF network element forwards the confirmation message to
  • the SMF network element can be understood as transparently transmitting the confirmation message to the SMF network element.
  • step S708 the SMF network element responds to the confirmation message and releases the resources occupied by the AI service.
  • the SMF network element interacts with the PCU to release the resources occupied for executing the above-mentioned AI service. For example, this may include releasing the occupation of the above-mentioned target UPF, and releasing the target UPF after executing the above-mentioned AI service.
  • this may include releasing the occupation of the above-mentioned target UPF, and releasing the target UPF after executing the above-mentioned AI service.
  • it can also be other possible resources.
  • the corresponding AI service execution policy can be generated for the AI service requested by the UE, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a workflow for applying AI services based on the 5G core network. , able to achieve tight coupling of AI technology and 5G networks.
  • FIG 10 is a flow chart of an AI service execution method according to an exemplary embodiment. As shown in Figure 10, the AI service execution method includes the following steps.
  • Step S1 The UE sends an AI service request message to the RAN.
  • the content of the AI service request message may refer to step S501 and will not be described again.
  • Step S2 RAN forwards the AI service request message to the AMF network element.
  • Step S3 The AMF network element forwards the AI service request message to the SMF network element.
  • Step S4 The SMF network element stores the AI data in the AI service request message in the UDSF network element.
  • the AMF network element may also store the AI data in the AI service request message in the UDSF network element in step S3 (not shown in Figure 10).
  • Step S5 The SMF network element obtains attribute information of multiple UPF network elements by querying the NRF network element.
  • attribute information of the UPF network element please refer to the above-mentioned step S502, which will not be described again.
  • Step S6 The SMF network element sends the attribute information of the multiple UPF network elements and the AI service request message to the PCF network element.
  • Step S7 The PCF network element determines the AI service execution policy based on the attribute information of the multiple UPF network elements and the AI service request message.
  • the method for the PCF network element to determine the AI service execution policy may refer to the above step S503, which will not be described again.
  • Step S8 The PCF network element stores the AI service execution policy in the UDR.
  • Step S9 The PCF network element sends the AI service execution policy to the SMF network element.
  • step S8 and step S9 can be executed either first or at the same time.
  • Step S10 The SMF network element establishes a connection with the target UPF network element used to execute the AI service, and sends the AI service execution policy to the target UPF network element used to execute the AI service.
  • Step S11 The target UPF network element retrieves the AI data stored in the UE from the UDSF.
  • Step S12 The target UPF network element executes the AI service based on the AI service execution policy and obtains the service execution result.
  • the method for the UPF network element to perform the AI service can refer to the above step S504, which will not be described again.
  • Step S13 The target UPF sends the service execution result to the SMF network element.
  • Step S14 The SMF network element sends the service execution result to the AMF network element.
  • Step S15 The AMF network element forwards the service execution result to the RAN.
  • Step S16 RAN forwards the service execution result to the UE.
  • Step S17 After receiving the service execution result, the UE sends a reply confirmation message to the RAN.
  • Step S18 RAN forwards the confirmation message to the AMF network element.
  • Step S19 The AMF network element forwards the confirmation message to the SMF network element.
  • Step S20 After receiving the message, the SMF network element releases the resources involved in the AI service.
  • the method of releasing resources may refer to the above-mentioned step S708, which will not be described again.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG. 11 is a block diagram of an AI service execution device according to an exemplary embodiment.
  • the AI service execution device is applied to an SMF network element.
  • the AI service execution device 100 includes a receiving module 111 , a determining module 120 and an execution module 130 .
  • the receiving module 110 is configured to receive an AI service request message, where the AI service request message is used to indicate the AI service requested by the UE;
  • the determination module 120 is configured to determine the AI service execution strategy based on the attribute information of multiple UPF network elements and the AI service request message;
  • the execution module 130 is configured to control a target UPF network element to execute the AI service according to the AI service execution policy, and the target UPF network element is one or more of the plurality of UPF network elements.
  • the determination module 120 may include:
  • the acquisition sub-module is configured to obtain the attribute information of the multiple UPF network elements
  • the first sending submodule is configured to send the attribute information of the multiple UPF network elements and the AI service request message to the PCF network element;
  • the first receiving sub-module is configured to receive the AI service execution policy sent by the PCF network element.
  • the acquisition submodule is configured as:
  • the AI service execution policy includes PCC rules, and the PCC rules include at least one of: the identification of the PCC rule, the identification of the target UPF assigned to the AI service, and the priority of the target UPF network element. .
  • the receiving module 110 is configured as:
  • the AI service request message is sent by the UE to the AMF network element through the RAN.
  • the execution module 130 includes:
  • the second sending sub-module is configured to send the AI service execution policy to the target UPF network element
  • the second receiving submodule is configured to receive the service execution result of the AI service sent by the target UPF network element.
  • the service execution result is generated after the UPF network element executes the AI service based on the AI service execution policy.
  • the target UPF network element includes a primary UPF network element and a secondary UPF network element
  • the second receiving submodule is configured as:
  • the service execution result sent by the primary UPF network element is received.
  • the service execution result is obtained by aggregating the execution result of the primary UPF network element and the execution result of the secondary UPF network element by the primary UPF network element.
  • the AI service execution device 100 may also include:
  • the sending module is configured to send the service execution result to the AMF network element, and the service execution result is used by the AMF network element to forward the service execution result to the UE through the RAN.
  • the AI service execution device 100 may also include: a resource management module;
  • the receiving module 110 is configured to receive a confirmation message sent by the UE based on the service execution result
  • the resource management module is configured to release the resources occupied by executing the AI service in response to the confirmation message.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG. 12 is a block diagram of an AI service execution device according to an exemplary embodiment.
  • the AI service execution device is applied to a PCF network element.
  • the AI service execution device 200 includes a receiving module 210 , a determining module 220 and a sending module 230 .
  • the receiving module 210 is configured to receive the attribute information of multiple UPF network elements sent by the SMF network element and the AI service request message of the UE, where the AI service request message is used to indicate the AI service requested by the UE;
  • the determination module 220 is configured to determine the AI service execution strategy based on the attribute information of the multiple UPF network elements and the AI service request message;
  • the sending module 230 is configured to send the AI service execution policy to the SMF network element.
  • the AI service execution policy is used to instruct the SMF network element to control the target UPF network element to execute the AI service according to the AI service execution policy.
  • the target The UPF network element is one or more of multiple UPF network elements.
  • the AI service execution policy includes PCC rules, and the PCC rules include at least one of: the identification of the PCC rule, the identification of the target UPF assigned to the AI service, and the priority of the target UPF network element.
  • the attribute information of the UPF network element includes at least the AI service supported by the UPF network element, and the AI service request message at least includes the AI service identifier and AI service type;
  • the determination module 220 is configured as:
  • the multiple UPF network elements determine at least one target UPF network element that matches the AI service identifier and the AI service type from the multiple UPF network elements;
  • the priority of the at least one target UPF network element is determined according to the AI service supported by the at least one target UPF network element.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG. 13 is a block diagram of an AI service execution device according to an exemplary embodiment.
  • the AI service execution device is applied to a UPF network element.
  • the AI service execution device 300 includes a receiving module 310 and an execution module 320 .
  • the receiving module 310 is configured to receive the AI service execution policy sent by the SMF network element.
  • the AI service execution policy is determined by the SMF network element based on the attribute information of multiple UPF network elements and the AI service request message.
  • the AI service request message Used to indicate the AI service requested by the UE;
  • the execution module 320 is configured to execute the AI service according to the AI service execution policy.
  • the UPF network element is the main UPF network element among multiple UPF network elements used to execute the AI service execution policy.
  • the execution module 320 is configured as:
  • the execution result of the primary UPF network element and the execution result of the secondary UPF network element are aggregated to obtain the service execution result of the AI service.
  • the AI service request message includes AI data used to execute the AI service
  • the execution module 320 is configured to:
  • the AI data is stored in the UDSF network element by the SMF network element after the AI service request message is sent to the SMF network element, or the AI data is sent by the AI service request message After arriving at the AMF network element, the AMF network element stores it in the UDSF network element;
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • FIG. 14 is a block diagram of an AI service execution device according to an exemplary embodiment.
  • the AI service execution device is applied to a UE.
  • the AI service execution device 200 includes a sending module 410 and an execution module 420 .
  • the sending module 410 is configured to send an AI service request message to the SMF network element.
  • the AI service request message is used to indicate the AI service requested by the UE.
  • the AI service request message is used to indicate that the SMF network element responds to multiple UPF network elements.
  • the attribute information and the AI service request message determine the AI service execution policy, and control the target UPF network element to execute the AI service according to the AI service execution policy.
  • the target UPF network element is one or more of the multiple UPF network elements. ;
  • the receiving module 420 is configured to receive the service execution result of the AI service sent by the SMF network element.
  • the sending module 410 is configured as:
  • the AI service request message is sent to the AMF network element through the RAN, and the AI service request message is used by the AMF network element to forward the AI service request message to the SMF network element.
  • the receiving module 420 is configured as:
  • the service execution result forwarded by the AMF network element is received through the RAN.
  • the service execution result is generated after the target UPF network element executes the AI service and is sent to the SMF network element.
  • the corresponding AI service execution policy can be generated for the AI service requested by the user equipment, and one or more UPF network elements can be scheduled to execute the AI service, thereby providing a work of applying AI services based on the 5G core network.
  • the process can realize the tight coupling of AI technology and 5G network.
  • Embodiments of the present disclosure also provide a computer-readable storage medium on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the AI service execution method provided by the present disclosure are implemented.
  • An embodiment of the present disclosure also provides an SMF network element, including: a processor; a memory used to store executable instructions of the processor; wherein the processor is configured to implement the steps provided by the present disclosure when executing the executable instructions.
  • the steps of the AI service execution method are configured to implement the steps provided by the present disclosure when executing the executable instructions.
  • An embodiment of the present disclosure also provides a PCF network element, including: a processor; a memory used to store executable instructions of the processor; wherein the processor is configured to implement the steps provided by the present disclosure when executing the executable instructions.
  • the steps of the AI service execution method are configured to implement the steps provided by the present disclosure when executing the executable instructions.
  • An embodiment of the present disclosure also provides a UPF network element, including: a processor; a memory used to store executable instructions of the processor; wherein the processor is configured to implement the steps provided by the present disclosure when executing the executable instructions.
  • the steps of the AI service execution method are configured to implement the steps provided by the present disclosure when executing the executable instructions.
  • An embodiment of the present disclosure also provides a UE, including: a processor; a memory for storing executable instructions by the processor; wherein the processor is configured to implement the AI provided by the present disclosure when executing the executable instructions. The steps of the service execution method.
  • FIG. 15 is a block diagram of an AI service execution device according to an exemplary embodiment.
  • the AI service execution device 1500 may be the above-mentioned user equipment, such as a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc. It may also be any one of the above-mentioned SMF network elements, PCF network elements, and UPF network elements.
  • the apparatus 1500 may include one or more of the following components: a processing component 1502, a memory 1504, and a communication component 1516.
  • Processing component 1502 generally controls the overall operations of device 1500, such as operations associated with display, phone calls, data communications, camera operations, and recording operations.
  • the processing component 1502 may include one or more processors 1520 to execute instructions to complete all or part of the steps of the above-mentioned AI service execution method.
  • processing component 1502 may include one or more modules that facilitate interaction between processing component 1502 and other components.
  • processing component 1502 may include a multimedia module to facilitate interaction between multimedia component 1508 and processing component 1502.
  • Memory 1504 is configured to store various types of data to support operations at device 1500 . Examples of such data include instructions for any application or method operating on device 1500, contact data, phonebook data, messages, pictures, videos, etc.
  • Memory 1504 may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EEPROM), Programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.
  • SRAM static random access memory
  • EEPROM electrically erasable programmable read-only memory
  • EEPROM erasable programmable read-only memory
  • EPROM Programmable read-only memory
  • PROM programmable read-only memory
  • ROM read-only memory
  • magnetic memory flash memory, magnetic or optical disk.
  • Communications component 1516 is configured to facilitate wired or wireless communications between device 1500 and other devices.
  • Device 1500 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof.
  • the communication component 1516 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel.
  • the communications component 1516 also includes a near field communications (NFC) module to facilitate short-range communications.
  • NFC near field communications
  • the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.
  • RFID radio frequency identification
  • IrDA infrared data association
  • UWB ultra-wideband
  • Bluetooth Bluetooth
  • apparatus 1500 may be configured by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable Gate array (FPGA), controller, microcontroller, microprocessor or other electronic components are implemented for executing the above AI service execution method.
  • ASICs application specific integrated circuits
  • DSPs digital signal processors
  • DSPDs digital signal processing devices
  • PLDs programmable logic devices
  • FPGA field programmable Gate array
  • controller microcontroller, microprocessor or other electronic components are implemented for executing the above AI service execution method.
  • a non-transitory computer-readable storage medium including instructions such as a memory 1504 including instructions, which can be executed by the processor 1520 of the device 1500 to complete the above AI service execution method is also provided.
  • the non-transitory computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
  • the above device can also be a part of an independent electronic device.
  • the device can be an integrated circuit (Integrated Circuit, IC) or a chip, where the integrated circuit can be an IC. , or it can be a collection of multiple ICs; the chip can include but is not limited to the following types: GPU (Graphics Processing Unit, graphics processor), CPU (Central Processing Unit, central processing unit), FPGA (Field Programmable Gate Array, can Programming logic array), DSP (Digital Signal Processor, digital signal processor), ASIC (Application Specific Integrated Circuit, application specific integrated circuit), SOC (System on Chip, SoC, system on a chip or system-level chip), etc.
  • GPU Graphics Processing Unit, graphics processor
  • CPU Central Processing Unit, central processing unit
  • FPGA Field Programmable Gate Array, can Programming logic array
  • DSP Digital Signal Processor, digital signal processor
  • ASIC Application Specific Integrated Circuit, application specific integrated circuit
  • SOC System on Chip, SoC, system on a chip or system-level chip
  • the above-mentioned integrated circuit or chip can be used to execute executable instructions (or codes) to implement the above-mentioned AI service execution method.
  • the executable instructions can be stored in the integrated circuit or chip, or can be obtained from other devices or devices.
  • the integrated circuit or chip includes a processor, a memory, and an interface for communicating with other devices.
  • the executable instructions can be stored in the processor, and when the executable instructions are executed by the processor, the above-mentioned AI service execution method is implemented; or, the integrated circuit or chip can receive the executable instructions through the interface and transmit them to the processor. Server execution to implement the above AI service execution method.
  • Embodiments of the present disclosure also provide a communication system, which includes: the above-mentioned SMF network element, PCF network element, UPF network element and UE;
  • the SMF network element may include the above-mentioned AI service execution device 100
  • the PCF network element may include the above-mentioned AI service execution device 200
  • the UPF network element may include the above-mentioned AI service execution device 300
  • the UE may include the above-mentioned AI service execution device 400.
  • the communication system may also include the RAN, AMF network element, NRF network element, UDSF network element and UDR mentioned above.
  • the functions performed by the RAN, AMF network element, NRF network element, UDSF network element and UDR can refer to the above embodiments shown in Figures 5 to 10, and will not be described again.
  • the present disclosure also provides another communication system, which includes: the above-mentioned SMF network element, PCF network element, UPF network element and UE;
  • the SMF network element is used to execute the AI service execution method shown in Figure 1
  • the PCF network element can be used to execute the AI service execution method shown in Figure 2
  • the UPF network element can be used to execute the AI service shown in Figure 3
  • Execution method the UE can be used to execute the AI service execution method shown in Figure 4.
  • the communication system may also include the RAN, AMF network element, NRF network element, UDSF network element and UDR mentioned above.
  • the functions performed by the SMF network element, PCF network element, UPF network element, UE, RAN, AMF network element, NRF network element, UDSF network element and UDR can refer to the above embodiments shown in Figures 5 to 10. No longer.
  • a computer program product comprising a computer program executable by a programmable device, the computer program having a function for performing the above when executed by the programmable device.
  • the code part of the AI service execution method is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a function for performing the above when executed by the programmable device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Information Transfer Between Computers (AREA)
  • Computer And Data Communications (AREA)

Abstract

一种AI服务执行方法、装置、网元、存储介质及芯片，涉及通信领域，该方法包括：用户设备向SMF网元发送AI服务请求消息，AI服务请求消息用于指示用户设备请求的AI服务，SMF网元基于接收到的AI服务请求消息以及多个UPF网元的属性信息确定用于执行AI服务的AI服务执行策略，以及在多个UPF网元中确定一个或多个作为目标UPF网元，并通过目标UPF网元根据AI服务执行策略执行AI服务。提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。

Description

AI服务执行方法、装置、网元、存储介质及芯片 技术领域
本公开涉及通信领域，尤其涉及一种AI服务执行方法、装置、网元、存储介质及芯片。
背景技术
目前AI技术(Artificial Intelligence，人工智能)已经成为5G网络的核心技术之一，5G(第五代移动通信系统)网络和AI技术的典型应用场景高度重叠。但是由于AI技术的广泛应用晚于5G网络，是在5G网络的架构确定后才开展的相关研究，因此，在相关技术中，基于5G网络的AI技术的应用通常属于外挂式应用，即在传统网络架构上进行优化和改造，该应用方式缺乏统一的AI服务工作和技术框架，通常是在已有的网络流程中简单的叠加AI服务，无法实现AI技术与5G网络的紧密耦合。
发明内容
为克服相关技术中存在的问题，本公开提供一种AI服务执行方法、装置、网元、存储介质及芯片。
根据本公开实施例的第一方面，提供一种AI服务执行方法，应用于会话管理功能SMF网元，所述方法包括：
接收AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
第二方面，提供一种AI服务执行方法，应用于策略控制功能PCF网元，所述方法包括：
接收会话管理功能SMF网元发送的多个用户面功能UPF网元的属性信息和用户设备的AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
根据所述多个UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
将所述AI服务执行策略发送至所述SMF网元，所述AI服务执行策略用于指示所述SMF网元控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
第三方面，提供一种AI服务执行方法，应用于用户面功能UPF网元，所述方法包括：
接收会话管理功能SMF网元发送的AI服务执行策略，所述AI服务执行策略是所述SMF网元根据多个用户面功能UPF网元的属性信息和AI服务请求消息确定的，所述AI服务请求消息用于指示用户设备请求的AI服务；
根据所述AI服务执行策略执行所述AI服务。
第四方面，提供一种AI服务执行方法，应用于用户设备，所述方法包括：
向会话管理功能SMF网元发送AI服务请求消息，所述AI服务请求消息用于指示所述用户设备请求的AI服务，所述AI服务请求消息用于指示所述SMF网元根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略，并控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个；
接收所述SMF网元发送的所述AI服务的服务执行结果。
第五方面，提供一种AI服务执行装置，应用于会话管理功能SMF网元，所述装置包括：
接收模块，被配置为接收AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
确定模块，被配置为根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
执行模块，被配置为控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
第六方面，提供一种AI服务执行装置，应用于策略控制功能PCF网元，所述装置包括：
接收模块，被配置为接收会话管理功能SMF网元发送的多个用户面功能UPF网元的属性信息和用户设备的AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
确定模块，被配置为根据所述多个UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
发送模块，被配置为将所述AI服务执行策略发送至所述SMF网元，所述AI服务执行策略用于指示所述SMF网元控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
第七方面，提供一种AI服务执行装置，应用于用户面功能UPF网元，所述装置包括：
接收模块，被配置为接收会话管理功能SMF网元发送的AI服务执行策略，所述AI服务执行策略是所述SMF网元根据多个用户面功能UPF网元的属性信息和AI服务请求消息确定的，所述AI服务请求消息用于指示用户设备请求的AI服务；
执行模块，被配置为根据所述AI服务执行策略执行所述AI服务。
第八方面，提供一种AI服务执行装置，应用于用户设备，所述装置包括：
发送模块，被配置为向会话管理功能SMF网元发送AI服务请求消息，所述AI服务请求消息用于指示所述用户设备请求的AI服务，所述AI服务请求消息用于指示所述SMF网元根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略，并控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个；
接收模块，被配置为接收所述SMF网元发送的所述AI服务的服务执行结果。
第九方面，提供一种会话管理功能SMF网元，包括：
处理器；用于存储处理器可执行指令的存储器；
其中，所述处理器被配置为在执行所述可执行指令时，实现第一方面中任一项所述方法的步骤。
第十方面，提供一种策略控制功能PCF网元，包括：
处理器；用于存储处理器可执行指令的存储器；
其中，所述处理器被配置为在执行所述可执行指令时，实现第二方面中任一项所述方法的步骤。
第十一方面，提供一种用户面功能UPF网元，包括：
处理器；用于存储处理器可执行指令的存储器；
其中，所述处理器被配置为在执行所述可执行指令时，实现第三方面中任一项所述方法的步骤。
第十二方面，提供一种用户设备，包括：
处理器；用于存储处理器可执行指令的存储器；
其中，所述处理器被配置为在执行所述可执行指令时，实现第四方面中任一项所述方法的步骤。
第十三方面，提供一种计算机可读存储介质，其上存储有计算机程序指令，所述程序指令被处理器执行时实现第一方面中任一项所述方法的步骤，或所述程序指令被处理器执行时实现第二方面中任一项所述方法的步骤，或所述程序指令被处理器执行时实现第三方面中任一项所述方法的步骤，或所述程序指令被处理器执行时实现第四方面中任一项所述方法的步骤。
第十四方面，提供一种芯片，包括处理器和接口；所述处理器用于读取指令以执行第一方面中任一项所述的方法，或所述处理器用于读取指令以执行第二方面中任一项所述的方法，或所述处理器用于读取指令以执行第三方面中任一项所述的方法，或所述处理器用于读取指令以执行第四方面中任一项所述的方法。
本公开的实施例提供的技术方案可以包括以下有益效果：
在上述技术方案中，用户设备向SMF网元发送AI服务请求消息，该AI服务请求消息用于指示用户设备请求的AI服务，该SMF网元基于接收到的AI服务请求消息以及多个UPF网元的属性信确定用于执行该AI服务的AI服务执行策略，以及在该多个UPF网元中确定一个或多个作为目标UPF网元，并通过目标UPF网元根据该AI服务执行策略执行该AI服务。通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本公开。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并与说明书一起用于解释本公开的原理。
图1是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图2是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图3是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图4是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图5是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图6是根据一示例性实施例示出的一种策略确定方法的流程图；
图7是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图8是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图9是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图10是根据一示例性实施例示出的一种AI服务执行方法的流程图；
图11是根据一示例性实施例示出的一种AI服务执行装置的框图；
图12是根据一示例性实施例示出的一种AI服务执行装置的框图；
图13是根据一示例性实施例示出的一种AI服务执行装置的框图；
图14是根据一示例性实施例示出的一种AI服务执行装置的框图；
图15是根据一示例性实施例示出的一种AI服务执行装置的框图。
具体实施方式
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中 所描述的实施方式并不代表与本公开相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本公开的一些方面相一致的装置和方法的例子。
可以理解的是，本公开中“多个”是指两个或两个以上，其它量词与之类似。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。单数形式的“一种”、“所述”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。
进一步可以理解的是，术语“第一”、“第二”等用于描述各种信息，但这些信息不应限于这些术语。这些术语仅用来将同一类型的信息彼此区分开，并不表示特定的顺序或者重要程度。实际上，“第一”、“第二”等表述完全可以互换使用。例如，在不脱离本公开范围的情况下，第一消息帧也可以被称为第二消息帧，类似地，第二消息帧也可以被称为第一消息帧。
进一步可以理解的是，本公开实施例中尽管在附图中以特定的顺序描述操作，但是不应将其理解为要求按照所示的特定顺序或是串行顺序来执行这些操作，或是要求执行全部所示的操作以得到期望的结果。在特定环境中，多任务和并行处理可能是有利的。
需要说明的是，本申请中所有获取信号、信息或数据的动作都是在遵照所在地国家相应的数据保护法规政策的前提下，并获得由相应装置所有者给予授权的情况下进行的。
图1是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图1所示，该AI服务执行方法用于SMF(Access and Mobility Management Function，接入与移动性管理功能)网元中，该方法包括以下步骤。
在步骤S101中，接收AI服务请求消息，该AI服务请求消息用于指示用户设备请求的AI服务。
示例地，该AI服务请求消息可以由UE(User Equipment，用户设备)发起，在一种实施方式中，在UE需要向5G网络的核心网请求调用某一AI服务时，UE生成该AI服务请求消息，该AI服务请求消息中可以至少包括用于指示UE所请求的AI服务的AI服务参数，AI服务参数例如可以包括AI服务标识(AI Service ID)、AI服务类型(AI Service Type)、用于执行该AI服务的AI数据(AI data)中的至少一种。可选地，该AI服务请求消息还可以包括其他信息，例如DNN(Data Network Name，数据网络名)等。
UE可以将该AI服务请求消息发送至RAN(Radio Access Network，无线接入网)，RAN也可以称为(R)AN，该RAN在接收到该AI服务请求消息后，将该AI服务请求消息转发至AMF(Session Management Function，会话管理功能)网元，该AMF在接收到该AI服务请求消息后，将该AI服务请求消息转发至SMF网元，可以理解为UE将该AI服务请求消息发送给RAN后，依次通过RAN和AMF网元透传至SMF网元。
在步骤S102中，根据多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。
示例地，对于任一UPF(User Plane Function，用户面功能)网元，该UPF网元的属性信息可以包括该UPF能够支持的AI服务，从而根据多个UPF网元的属性信息即可确定该多个UPF网元中的每个UPF网元能够支持的AI服务，根据AI服务请求消息能够确定UE所请求的AI服务，从而可以确定能够用于执行UE所请求的AI服务的一个或多个UPF作为下述的目标UPF网元，以及确定相应地AI服务执行策略。
在一种实施方式中，SMF网元在接收到该AI服务请求消息后，可以向NRF(Network Repository Function，网络存储功能)网元进行查询，以获得该多个UPF网元的属性信息。示例地，该NRF网元存储有UPF列表，通过查询NRF网元获取该UPF列表，可以确定该多个UPF网元，并得到该多个UPF网元的属性信息。其中，该多个UPF网元可以是该 UPF列表中的部分UPF网元或全部UPF网元。
SMF网元在获取到该多个UPF网元的属性信息后，可以根据该多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。示例地，该AI服务执行策略可以是PCC(Policy Control and Charging，策略控制和计费)规则，该PCC规则中可以至少包括以下一种或多种：该PCC规则的规则标识，为该AI服务分配的目标UPF的标识，该目标UPF网元的优先级等信息。
在步骤S103中，控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
示例地，SMF网元在得到AI服务执行策略后，与该目标UPF网元建立连接，并将该AI服务执行策略发送至目标UPF网元，目标UPF网元根据该AI服务执行策略执行UE所请求的AI服务。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图2是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图2所示，该AI服务执行方法用于PCF网元中，该方法包括以下步骤。
在步骤S201中，接收SMF网元发送的多个UPF网元的属性信息和UE的AI服务请求消息，该AI服务请求消息用于指示UE请求的AI服务。
示例地，该AI服务请求消息以及UE向SMF网元发送该AI服务请求消息的方法与步骤S101中相同，可参照步骤S101中的内容，不再赘述。
SMF网元在获取到该多个UPF网元的属性信息后，将该多个UPF网元的属性信息和UE发送的AI服务请求消息，发送给PCF网元。
在步骤S202中，根据该多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。
示例地，PCF网元在接收到SMF网元发送的该多个UPF网元的属性信息和该AI服务请求消息后，通过分析该多个UPF网元的属性信息和该AI服务请求消息所指示的AI服务，确定能够用于执行UE所请求的AI服务的一个或多个UPF作为该目标UPF网元，以及确定相应地AI服务执行策略，该AI服务执行策略可以是上述的PCC规则，与步骤S102中所述的PCC规则相同，不再赘述。
在步骤S203中，将该AI服务执行策略发送至该SMF网元，该AI服务执行策略用于指示该SMF网元控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
示例地，SMF网元在得到AI服务执行策略后，与该目标UPF网元建立连接，并将该AI服务执行策略发送至目标UPF网元，目标UPF网元根据该AI服务执行策略执行UE所请求的AI服务。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图3是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图3所示，该AI服务执行方法用于UPF网元中，该方法包括以下步骤。
在步骤S301中，接收SMF网元发送的AI服务执行策略，该AI服务执行策略是该SMF网元根据多个UPF网元的属性信息和AI服务请求消息确定的，该AI服务请求消息 用于指示UE请求的AI服务。
示例地，该AI服务请求消息以及UE向SMF网元发送该AI服务请求消息的方法与步骤S101中相同，可参照步骤S101中的内容，不再赘述。SMF网元在接收到AI服务请求消息后，确定AI服务执行策略的方法可以参照步骤S102和S202，不再赘述。
在步骤S302中，根据该AI服务执行策略执行该AI服务。
其中，该AI服务执行策略可以是上述的PCC规则，与步骤S102中所述的PCC规则相同，不再赘述。如果用于执行该AI服务执行策略的多个目标UPF网元，则该多个目标UPF网元中包括一个主UPF网元和至少一个次UPF网元。SMF网元可以先将该AI服务执行策略发送至主UPF网元，该主UPF网元将AI服务执行策略下发给各个次UPF网元，从而主UPF网元和次UPF网元根据自身分配到的AI服务执行策略来执行AI服务。
可选地，主UPF网元可以接收该多个目标UPF中的次UPF根据该AI服务执行策略执行该AI服务后得到的执行结果。将该主UPF网元的执行结果和次UPF网元的执行结果进行聚合得到该AI服务的服务执行结果。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图4是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图4所示，该AI服务执行方法用于UE中，该方法包括以下步骤。
在步骤S401中，向SMF网元发送AI服务请求消息，该AI服务请求消息用于指示该UE请求的AI服务，该AI服务请求消息用于指示该SMF网元根据多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略，并控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
示例地，该AI服务请求消息以及UE向SMF网元发送该AI服务请求消息的方法与步骤S101中相同，可参照步骤S101中的内容，不再赘述。SMF网元在接收到AI服务请求消息后，确定AI服务执行策略的方法可以参照步骤S102和S202，不再赘述。目标UPF网元根据该AI服务执行策略执行该AI服务的方法可以参照步骤S301和S302，不再赘述。
在步骤S402中，接收该SMF网元发送的该AI服务的服务执行结果。
示例地，目标UPF网元根据该AI服务执行策略执行该AI服务后得到该AI服务的服务执行结果，而后目标UPF网元将服务执行结果发送给SMF网元，SMF网元将服务执行结果发送给UE，其中SMF网元将服务执行结果发送给UE可以包括：SMF网元先将服务执行结果发送至AMF网元，AMF网元将服务执行结果转发至RAN，再由该RAN将服务执行结果转发至UE，从而UE就得到了该AI服务的服务执行结果。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图5是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图5所示，该AI服务执行方法包括以下步骤。
在步骤S501中，SMF网元接收AI服务请求消息，该AI服务请求消息用于指示UE请求的AI服务。
其中，该AI服务请求消息可以由UE发起，在一种实施方式中，在UE需要向5G网络的核心网请求调用某一AI服务时，UE生成该AI服务请求消息，该AI服务请求消息中可以至少包括用于指示UE所请求的AI服务的AI服务参数，AI服务参数例如可以包括 AI服务标识(AI Service ID)、AI服务类型(AI Service Type)、用于执行该AI服务的AI数据(AI data)中的至少一种。其中AI服务类型可以包括：视频类AI服务(video)，例如图像识别，音频类AI服务(audio)，例如语音识别，文本类AI服务(text)，例如自然语言处理等。
可选地，该AI服务请求消息还可以包括其他信息，例如DNN、该UE的设备信息、该UE的用户信息(User information)、接入类型(Access type)中的一种或多种，其中UE的设备信息例如可以包括设备标识、设备型号、设备的位置信息中的一种或多种。
UE可以将该AI服务请求消息发送至RAN，该RAN在接收到该AI服务请求消息后，将该AI服务请求消息转发至AMF网元，该AMF网元在接收到该AI服务请求消息后，将该AI服务请求消息转发至SMF网元，可以理解为UE将该AI服务请求消息发送给RAN后，依次通过RAN和AMF网元透传至SMF网元。
此外，该AMF网元在接收到AI服务请求消息后，可以将该AI服务请求消息中的上述AI数据(AI data)存储在UDSF(Unstructured Data Storage Network Function，非结构化数据存储功能)网元中，用于后续流程中，UPF网元在执行AI服务时从该UDSF网元中调用该AI数据。或者，也可以在SMF网元接收到该AI服务请求消息后，由SMF网元将该AI服务请求消息中的AI数据存储在UDSF网元中。
在步骤S502中，SMF网元获取该多个UPF网元的属性信息。
示例地，SMF网元在接收到该AI服务请求消息后，可以向NRF网元进行查询，以获得该多个UPF网元的属性信息。SMF网元从该NRF网元获取该多个UPF网元的属性信息的方法可以参照步骤S102，不再赘述。
其中，该对于任一UPF网元，该UPF网元的属性信息可以包括：该UPF能够支持的AI服务的AI服务标识、AI服务类型等信息，还可以包括该UPF的位置信息。
在步骤S503中，SMF网元根据多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。
图6是根据一示例性实施例示出的一种策略确定方法的流程图，如图6所示，该步骤S503可以包括以下步骤。
在步骤S5031中，SMF网元将该多个UPF网元的属性信息和该AI服务请求消息发送至PCF网元。
在步骤S5032中，PCF网元根据该多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。
示例地，PCF网元在接收到SMF网元发送的该多个UPF网元的属性信息和该AI服务请求消息后，通过分析该多个UPF网元的属性信息和该AI服务请求消息所指示的AI服务，确定能够用于执行UE所请求的AI服务的一个或多个UPF作为该目标UPF网元，以及确定相应地AI服务执行策略。该AI服务执行策略可以是上述的PCC规则，该PCC规则中可以至少包括以下一种或多种：该PCC规则的规则标识，为该AI服务分配的目标UPF的标识，该目标UPF网元的优先级等信息。
示例地，由于该AI服务请求消息中可以包括上述的AI服务标识、AI服务类型，UPF网元的属性信息可以包括该UPF能够支持的AI服务的AI服务标识、AI服务类型等信息，因此上述确定该AI服务执行策略的步骤可以包括以下步骤：
首先，根据该多个UPF网元支持的AI服务，从该多个UPF网元中确定与该AI服务标识和AI服务类型匹配的至少一个目标UPF网元。
其次，根据该至少一个目标UPF网元支持的AI服务，确定该至少一个目标UPF网元的优先级。
示例地，可以将该多个UPF网元中的各个UPF能够支持的AI服务的AI服务标识、 AI服务类型与UE的AI服务请求消息中的AI服务标识、AI服务类型进行对比，即可确定与AI服务请求消息匹配的至少一个目标UPF网元。
在确定该至少一个目标UPF网元后，可以根据各个UPF目标能够支持的AI服务，以及AI服务请求消息请求的AI服务，来确定该至少一个目标UPF网元的优先级。其中UPF网元的优先级可以包括该UPF是主UPF还是次UPF，以及还可以包括该至少一个目标UPF网元在执行该AI服务时的执行顺序的优先级。
可选地，在确定与AI服务请求消息匹配的至少一个目标UPF网元的过程中，还可以考虑UE的位置，如上文所述，该AI服务请求消息还可以包括设备的位置信息，在UPF网元的属性信息中也可以包括UPF的位置信息，因此，可以将该多个UPF网元中的各个UPF能够支持的AI服务的AI服务标识、AI服务类型、位置信息，与UE的AI服务请求消息中的AI服务标识、AI服务类型、设备的位置信息进行对比，确定上述各项信息均匹配的一个UPF或多个UPF为目标UPF网元。
PCF网元在确定AI服务执行策略后可以将该AI服务执行策略存储至UDR(User Data Repository，用户数据寄存器)中，以及执行步骤5033。
在步骤S5033中，PCF网元将该AI服务执行策略发送至该SMF网元。
在步骤S504中，SMF网元控制目标UPF网元根据该AI服务执行策略执行该AI服务。
图7是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图7所示，该步骤S504可以包括以下步骤。
在步骤S5041中，SMF网元向该目标UPF网元发送该AI服务执行策略。
在步骤S5042中，目标UPF网元从UDSF网元中调用该AI服务对应的AI数据。
在步骤S5043中，目标UPF网元根据该AI服务执行策略执行该AI服务。
其中，目标UPF可以为一个或多个，在目标UPF在多个的情况下，可以包括一个主UPF网元和至少一个次UPF网元。SMF网元可以先将该AI服务执行策略发送至主UPF网元，该主UPF网元将AI服务执行策略下发给各个次UPF网元，从而主UPF网元和次UPF网元根据自身分配到的AI服务执行策略来执行AI服务。
示例地，可以理解的是，AI服务可以是单个AI服务，也可以是多个AI服务，如果是多个AI服务，则该多个AI服务可以共同完成一个AI功能，其中每个AI服务参与的AI子功能不同，可能是单个AI子功能，也可能是多个AI子功能的组合。或者，多个AI服务也可以共同完成多个AI功能。UPF之间可以通过5G的N9接口进行联邦学习等分布式任务，并将输出的数据存入UDSF实现UPF之间的数据共享。
在一种可能的实现方式中，UE所请求的该AI服务为多个AI服务，则该多个AI服务可以分别由主UPF网元和该至少一个次UPF网元分别执行，主UPF网元和次UPF网元根据其所支持的AI服务来执行该多个AI服务中的一个或多个AI服务，例如，不同的AI服务分别用于执行不同的AI算法，例如AI服务1用于执行SVM(Support Vector Machine，支持向量机)算法、AI服务2用于执行随机森林、AI服务3用于执行PCA(Principal Component Analysis，主成分分析)降维算法、AI服务4用于K-Means聚类算法，等等，从而实现由多个UPF网元通过多个AI服务分布式地执行多种不同的算法。可选地，在此实施方式中，上文所述的目标UPF网元的优先级，可以依据不同的UPF网元在执行不同的算法时的执行顺序确定。
或者，在另一种可能的实现方式中，UE所请求的该AI服务为一个AI服务，主UPF网元和次UPF网元可以分别用于执行该AI服务的一部分，例如将该AI服务对应的AI算法分为多个部分(或者多个步骤)，由主UPF网元和次UPF网元分别执行该多个部分中的一个或多个部分(或者分别执行多个步骤种的一个或多个步骤)。可选地，上文所述的目 标UPF网元的优先级，可以是不同的UPF网元在执行不同的算法时的执行顺序。可选地，在此实施方式中，上文所述的目标UPF网元的优先级，可以依据该AI算法多个部分(或者多个步骤)的执行顺序确定。
通过上述方式，能够针对UE所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图8是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图8所示，该AI服务执行方法包括以下步骤。
在步骤S601中，SMF网元接收AI服务请求消息，该AI服务请求消息用于指示UE请求的AI服务。
在步骤S602中，SMF网元获取该多个UPF网元的属性信息。
在步骤S603中，SMF网元根据多个用户面功能UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。
在步骤S604中，SMF网元控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
其中，步骤S601～步骤S604的方法与步骤S501～步骤S504相同，可参照步骤S501～步骤S504的内容，不再赘述。
在步骤S605中，目标UPF网元将根据该AI服务执行策略执行该AI服务得到的该AI服务的服务执行结果发送至SMF网元。
示例地，如步骤S5041至S5042中所述的步骤，在目标UPF网元为多个的情况下，目标UPF网元中的主UPF网元接收该多个目标UPF中的次UPF根据该AI服务执行策略执行AI服务后得到的执行结果。
主UPF网元将主UPF网元的执行结果和次UPF网元的执行结果进行聚合得到该AI服务的服务执行结果。可以理解的是，主UPF网元的执行结果和次UPF网元的执行结果均可以为执行上述AI服务的一部分结果，需要由主UPF网元将各个执行结果进行聚合后得到最终的服务执行结果。示例地，由于上述在UPF网元在执行AI服务时，调用的该AI服务对应的AI数据存储在UDSF网元中，属于非结构化数据，因此可选地，各个UPF网元可以将各自的执行结果也存储在UDSF网元，因此主UPF网元在将各个执行结果进行聚合的过程中需要对执行结果进行结构化处理，以得到聚合后的服务执行结果。
在步骤S606中，SMF网元将该服务执行结果发送至UE。
示例地，SMF网元可以将该服务执行结果发送至AMF网元，该AMF网元将该服务执行结果转发至RAN，再由转发RAN给该UE。
通过上述方式，能够针对UE所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图9是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图9所示，该AI服务执行方法包括以下步骤。
在步骤S701中，SMF网元接收AI服务请求消息，该AI服务请求消息用于指示UE请求的AI服务。
在步骤S702中，SMF网元获取该多个UPF网元的属性信息。
在步骤S703中，SMF网元根据多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。
在步骤S704中，SMF网元控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
在步骤S705中，目标UPF网元将根据该AI服务执行策略执行该AI服务得到的该AI服务的服务执行结果发送至SMF网元。
在步骤S706中，SMF网元将该服务执行结果发送至UE。
其中，步骤S701～步骤S706的方法与步骤S601～步骤S606相同，可参照步骤S601～步骤S606的内容，不再赘述。
在步骤S707中，SMF网元接收UE基于该服务执行结果发送的确认消息。
示例地，UE在接收到上述服务执行结果后，针对该服务执行结果回复该确认消息，可以将该确认消息发送至RAN，由RAN转发至AMF网元，再由AMF网元将确认消息转发至该SMF网元，可以理解为是将该确认消息透传给SMF网元。
在步骤S708中，SMF网元响应于该确认消息，将执行该AI服务所占用的资源进行释放。
示例地，SMF网元在接收到该确认消息后，与该PCU进行交互，以释放用于执行上述AI服务所占用的资源，例如可以包括释放对于上述目标UPF的占用，释放目标UPF在执行该AI服务时所占用的算力资源，除此之外也可以是其他可能的资源。
通过上述方式，能够针对UE所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图10是根据一示例性实施例示出的一种AI服务执行方法的流程图，如图10所示，该AI服务执行方法包括以下步骤。
步骤S1，UE发送AI服务请求消息至RAN。其中AI服务请求消息的内容可参照步骤S501，不再赘述。
步骤S2，RAN转发该AI服务请求消息至AMF网元。
步骤S3，AMF网元将AI服务请求消息转发至SMF网元。
步骤S4，SMF网元将AI服务请求消息中的AI数据存储在UDSF网元中。
或者，在另一种实现方式中，也可以在步骤S3中由AMF网元将AI服务请求消息中的AI数据存储在UDSF网元中(图10中未示出)。
步骤S5，SMF网元通过查询NRF网元获取多个UPF网元的属性信息。UPF网元的属性信息可以参照上述的步骤S502，不再赘述。
步骤S6，SMF网元将该多个UPF网元的属性信息和该AI服务请求消息发送至PCF网元。
步骤S7，PCF网元根据该多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略。PCF网元确定AI服务执行策的方法可以参照上述步骤S503，不再赘述。
步骤S8，PCF网元将该AI服务执行策略存储至UDR中。
步骤S9，PCF网元将该AI服务执行策略发送至SMF网元。
其中，步骤S8和步骤S9可以先执行其中的任意一步，也可以同时执行。
步骤S10，SMF网元与用于执行该AI服务的目标UPF网元建立连接，并将AI服务执行策略送至用于执行该AI服务的目标UPF网元。
步骤S11，目标UPF网元从UDSF调取UE存储的AI数据。
步骤S12，目标UPF网元基于AI服务执行策略执行该AI服务，得到服务执行结果。UPF网元执行该AI服务的方法可以参照上述步骤S504，不再赘述。
步骤S13，目标UPF将服务执行结果发送至SMF网元。
步骤S14，SMF网元将服务执行结果发送至AMF网元。
步骤S15，AMF网元将服务执行结果转发至RAN。
步骤S16，RAN转发该服务执行结果至UE。
步骤S17，UE接收到服务执行结果后，向RAN发送回复的确认消息。
步骤S18，RAN将确认消息转发至AMF网元。
步骤S19，AMF网元将确认消息转发至SMF网元。
步骤S20，SMF网元收到消息后释放该AI服务涉及的资源。示例地，释放资源的方法可以参照上述步骤S708，不再赘述。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图11是根据一示例性实施例示出的一种AI服务执行装置的框图，该AI服务执行装置应用于SMF网元。参照图11，该AI服务执行装置100包括接收模块111，确定模块120和执行模块130。
接收模块110，被配置为接收AI服务请求消息，该AI服务请求消息用于指示UE请求的AI服务；
确定模块120，被配置为根据多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略；
执行模块130，被配置为控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
可选地，该确定模块120，可以包括：
获取子模块，被配置为获取该多个UPF网元的属性信息；
第一发送子模块，被配置为将该多个UPF网元的属性信息和该AI服务请求消息发送至PCF网元；
第一接收子模块，被配置为接收该PCF网元发送的该AI服务执行策略。
可选地，该获取子模块，被配置为：
从NRF网元获取该多个UPF网元的属性信息。
可选地，该AI服务执行策略包括PCC规则，该PCC规则至少包括：该PCC规则的标识、为该AI服务分配的该目标UPF的标识、该目标UPF网元的优先级中的至少一种。
可选地，接收模块110，被配置为：
接收接入与AMF网元转发的该AI服务请求消息，该AI服务请求消息是该UE通过RAN发送给该AMF网元的。
可选地，该执行模块130，包括：
第二发送子模块，被配置为向该目标UPF网元发送该AI服务执行策略；
第二接收子模块，被配置为接收该目标UPF网元发送的该AI服务的服务执行结果，该服务执行结果由该UPF网元基于该AI服务执行策略执行该AI服务后生成。
可选地，该目标UPF网元包括主UPF网元和次UPF网元，所述第二接收子模块，被配置为：
接收该主UPF网元发送的该服务执行结果，该服务执行结果为该主UPF网元将该主UPF网元的执行结果和该次UPF网元的执行结果进行聚合后得到的。
可选地，该AI服务执行装置100还可以包括：
发送模块，被配置为将该服务执行结果发送至AMF网元，该服务执行结果用于该AMF网元将该服务执行结果通过RAN转发给该UE。
可选地，该AI服务执行装置100还可以包括：资源管理模块；
该接收模块110，被配置为接收该UE基于该服务执行结果发送的确认消息；
该资源管理模块，被配置为响应于该确认消息，将执行该AI服务所占用的资源进行释放。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图12是根据一示例性实施例示出的一种AI服务执行装置的框图，该AI服务执行装置应用于PCF网元。参照图12，该AI服务执行装置200包括接收模块210，确定模块220和发送模块230。
接收模块210，被配置为接收SMF网元发送的多个UPF网元的属性信息和UE的AI服务请求消息，该AI服务请求消息用于指示UE请求的AI服务；
确定模块220，被配置为根据该多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略；
发送模块230，被配置为将该AI服务执行策略发送至该SMF网元，该AI服务执行策略用于指示该SMF网元控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个。
可选地，该AI服务执行策略包括PCC规则，该PCC规则至少包括：该PCC规则的标识、为该AI服务分配的目标UPF的标识、该目标UPF网元的优先级中的至少一种。
可选地，该UPF网元的属性信息至少包括该UPF网元支持的AI服务，该AI服务请求消息至少包括AI服务标识和AI服务类型；
该确定模块220，被配置为：
根据该多个UPF网元支持的AI服务，从该多个UPF网元中确定与该AI服务标识和AI服务类型匹配的至少一个目标UPF网元；
根据该至少一个目标UPF网元支持的AI服务，确定该至少一个目标UPF网元的优先级。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图13是根据一示例性实施例示出的一种AI服务执行装置的框图，该AI服务执行装置应用于UPF网元。参照图13，该AI服务执行装置300包括接收模块310和执行模块320。
接收模块310，被配置为接收SMF网元发送的AI服务执行策略，该AI服务执行策略是该SMF网元根据多个UPF网元的属性信息和AI服务请求消息确定的，该AI服务请求消息用于指示UE请求的AI服务；
执行模块320，被配置为根据该AI服务执行策略执行该AI服务。
可选地，该UPF网元为用于执行该AI服务执行策略的多个UPF网元中的主UPF网元，该执行模块320，被配置为：
为该多个UPF分配该AI服务执行策略，该AI服务执行策略用于指示该多个UPF分别根据分配的该AI服务执行策略执行该AI服务；
接收该多个UPF中的次UPF根据该AI服务执行策略执行该AI服务后得到的执行结果；
将该主UPF网元的执行结果和该次UPF网元的执行结果进行聚合得到该AI服务的服务执行结果。
可选地，该AI服务请求消息中包括用于执行该AI服务的AI数据，该执行模块320，被配置为：
从UDSF网元获取该AI数据；该AI数据是该AI服务请求消息发送至SMF网元后，由该SMF网元存储在该UDSF网元中的，或者该AI数据是该AI服务请求消息发送至AMF网元后，由该AMF网元存储在该UDSF网元中的；
根据该AI服务执行策略和该AI数据执行该AI服务。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
图14是根据一示例性实施例示出的一种AI服务执行装置的框图，该AI服务执行装置应用于UE。参照图14，该AI服务执行装置200包括发送模块410和执行模块420。
发送模块410，被配置为向SMF网元发送AI服务请求消息，该AI服务请求消息用于指示该UE请求的AI服务，该AI服务请求消息用于指示该SMF网元根据多个UPF网元的属性信息和该AI服务请求消息确定AI服务执行策略，并控制目标UPF网元根据该AI服务执行策略执行该AI服务，该目标UPF网元为该多个UPF网元中的一个或多个；
接收模块420，被配置为接收该SMF网元发送的该AI服务的服务执行结果。
可选地，该发送模块410被配置为：
通过RAN向AMF网元发送该AI服务请求消息，该AI服务请求消息用于该AMF网元将该AI服务请求消息转发至该SMF网元。
可选地，该接收模块420被配置为：
通过RAN接收AMF网元转发的该服务执行结果，该服务执行结果是该目标UPF网元执行该AI服务后生成并发送给该SMF网元的。
通过上述方式，能够针对用户设备所请求AI服务生成相应的AI服务执行策略，并调度一个或多个UPF网元来执行该AI服务，从而提供了一种基于5G核心网的应用AI服务的工作流程，能够实现AI技术和5G网络的紧密耦合。
关于上述实施例中的装置，其中各个模块执行操作的具体方式已经在有关该方法的实施例中进行了详细描述，此处将不做详细阐述说明。
本公开实施例还提供一种计算机可读存储介质，其上存储有计算机程序指令，该程序指令被处理器执行时实现本公开提供的AI服务执行方法的步骤。
本公开实施例还提供一种SMF网元，包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为在执行所述可执行指令时，实现本公开提供的AI服务执行方法的步骤。
本公开实施例还提供一种PCF网元，包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为在执行所述可执行指令时，实现本公开提供的AI服务执行方法的步骤。
本公开实施例还提供一种UPF网元，包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为在执行所述可执行指令时，实现本公开提供的AI服务执行方法的步骤。
本公开实施例还提供一种UE，包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为在执行所述可执行指令时，实现本公开提供的AI服务执行方法的步骤。
图15是根据一示例性实施例示出的一种AI服务执行装置的框图。例如，该AI服务执行装置1500可以是为上述的用户设备，例如移动电话，计算机，数字广播终端，消息收发设备，游戏控制台，平板设备，医疗设备，健身设备，个人数字助理等。也可以是是上述的SMF网元、PCF网元、UPF网元中的任一网元。
参照图15，该装置1500可以包括以下一个或多个组件：处理组件1502，存储器1504， 以及通信组件1516。
处理组件1502通常控制装置1500的整体操作，诸如与显示，电话呼叫，数据通信，相机操作和记录操作相关联的操作。处理组件1502可以包括一个或多个处理器1520来执行指令，以完成上述的AI服务执行方法的全部或部分步骤。此外，处理组件1502可以包括一个或多个模块，便于处理组件1502和其他组件之间的交互。例如，处理组件1502可以包括多媒体模块，以方便多媒体组件1508和处理组件1502之间的交互。
存储器1504被配置为存储各种类型的数据以支持在装置1500的操作。这些数据的示例包括用于在装置1500上操作的任何应用程序或方法的指令，联系人数据，电话簿数据，消息，图片，视频等。存储器1504可以由任何类型的易失性或非易失性存储设备或者它们的组合实现，如静态随机存取存储器(SRAM)，电可擦除可编程只读存储器(EEPROM)，可擦除可编程只读存储器(EPROM)，可编程只读存储器(PROM)，只读存储器(ROM)，磁存储器，快闪存储器，磁盘或光盘。
通信组件1516被配置为便于装置1500和其他设备之间有线或无线方式的通信。装置1500可以接入基于通信标准的无线网络，如WiFi，2G或3G，或它们的组合。在一个示例性实施例中，通信组件1516经由广播信道接收来自外部广播管理系统的广播信号或广播相关信息。在一个示例性实施例中，所述通信组件1516还包括近场通信(NFC)模块，以促进短程通信。例如，在NFC模块可基于射频识别(RFID)技术，红外数据协会(IrDA)技术，超宽带(UWB)技术，蓝牙(BT)技术和其他技术来实现。
在示例性实施例中，装置1500可以被一个或多个应用专用集成电路(ASIC)、数字信号处理器(DSP)、数字信号处理设备(DSPD)、可编程逻辑器件(PLD)、现场可编程门阵列(FPGA)、控制器、微控制器、微处理器或其他电子元件实现，用于执行上述AI服务执行方法。
在示例性实施例中，还提供了一种包括指令的非临时性计算机可读存储介质，例如包括指令的存储器1504，上述指令可由装置1500的处理器1520执行以完成上述AI服务执行方法。例如，所述非临时性计算机可读存储介质可以是ROM、随机存取存储器(RAM)、CD-ROM、磁带、软盘和光数据存储设备等。
上述装置除了可以是独立的电子设备外，也可是独立电子设备的一部分，例如在一种实施例中，该装置可以是集成电路(Integrated Circuit，IC)或芯片，其中该集成电路可以是一个IC，也可以是多个IC的集合；该芯片可以包括但不限于以下种类：GPU(Graphics Processing Unit，图形处理器)、CPU(Central Processing Unit，中央处理器)、FPGA(Field Programmable Gate Array，可编程逻辑阵列)、DSP(Digital Signal Processor，数字信号处理器)、ASIC(Application Specific Integrated Circuit，专用集成电路)、SOC(System on Chip，SoC，片上系统或系统级芯片)等。上述的集成电路或芯片中可以用于执行可执行指令(或代码)，以实现上述的AI服务执行方法。其中该可执行指令可以存储在该集成电路或芯片中，也可以从其他的装置或设备获取，例如该集成电路或芯片中包括处理器、存储器，以及用于与其他的装置通信的接口。该可执行指令可以存储于该处理器中，当该可执行指令被处理器执行时实现上述的AI服务执行方法；或者，该集成电路或芯片可以通过该接口接收可执行指令并传输给该处理器执行，以实现上述的AI服务执行方法。
本公开实施例还提供一种通信系统，该通信系统包括：上述的SMF网元、PCF网元、UPF网元和UE；
其中，在一种实施方式中，该SMF网元可以包括上述的AI服务执行装置100，该PCF网元可以包括上述的AI服务执行装置200，该UPF网元可以包括上述的AI服务执行装置300，UE可以包括上述的AI服务执行装置400。
可选地，该通信系统还可以包括上文所述的RAN、AMF网元、NRF网元、UDSF网 元和UDR。该RAN、AMF网元、NRF网元、UDSF网元和UDR所执行的功能可以参照上述图5至图10所示的实施例，不再赘述。
本公开还提供另一种通信系统，该通信系统包括：上述的SMF网元、PCF网元、UPF网元和UE；
其中SMF网元用于执行图1所示的AI服务执行方法，该PCF网元可以用于执行图2所示的AI服务执行方法，该UPF网元可以用于执行图3所示的AI服务执行方法，该UE可以用于执行图4所示的AI服务执行方法。
可选地，该通信系统还可以包括上文所述的RAN、AMF网元、NRF网元、UDSF网元和UDR。
其中，该SMF网元、PCF网元、UPF网元、UE、RAN、AMF网元、NRF网元、UDSF网元和UDR所执行的功能可以参照上述图5至图10所示的实施例，不再赘述。
在另一示例性实施例中，还提供一种计算机程序产品，该计算机程序产品包含能够由可编程的装置执行的计算机程序，该计算机程序具有当由该可编程的装置执行时用于执行上述的AI服务执行方法的代码部分。
本领域技术人员在考虑说明书及实践本公开后，将容易想到本公开的其它实施方案。本申请旨在涵盖本公开的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本公开的真正范围和精神由下面的权利要求指出。
应当理解的是，本公开并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本公开的范围仅由所附的权利要求来限制。

Claims (28)

  1. 一种AI服务执行方法，其特征在于，应用于会话管理功能SMF网元，所述方法包括：
    接收AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
    根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
    控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
  2. 根据权利要求1所述的方法，其特征在于，所述根据多个UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略，包括：
    获取所述多个UPF网元的属性信息；
    将所述多个UPF网元的属性信息和所述AI服务请求消息发送至策略控制功能PCF网元；
    接收所述PCF网元发送的所述AI服务执行策略。
  3. 根据权利要求2所述的方法，其特征在于，所述获取所述多个UPF网元的属性信息，包括：
    从网络存储功能NRF网元获取所述多个UPF网元的属性信息。
  4. 根据权利要求1所述的方法，其特征在于，所述AI服务执行策略包括策略控制和计费PCC规则，所述PCC规则至少包括：所述PCC规则的标识、为所述AI服务分配的所述目标UPF的标识、所述目标UPF网元的优先级中的至少一种。
  5. 根据权利要求1所述的方法，其特征在于，所述接收AI服务请求消息，包括：
    接收接入与移动性管理功能AMF网元转发的所述AI服务请求消息，所述AI服务请求消息是所述用户设备通过无线接入网RAN发送给所述AMF网元的。
  6. 根据权利要求1所述的方法，其特征在于，所述控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，包括：
    向所述目标UPF网元发送所述AI服务执行策略；
    接收所述目标UPF网元发送的所述AI服务的服务执行结果，所述服务执行结果由所述UPF网元基于所述AI服务执行策略执行所述AI服务后生成。
  7. 根据权利要求6所述的方法，其特征在于，所述目标UPF网元包括主UPF网元和次UPF网元，所述接收所述目标UPF网元发送的服务执行结果，包括：
    接收所述主UPF网元发送的所述服务执行结果，所述服务执行结果为所述主UPF网元将所述主UPF网元的执行结果和所述次UPF网元的执行结果进行聚合后得到的。
  8. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    将所述服务执行结果发送至AMF网元，所述服务执行结果用于所述AMF网元将所述服务执行结果通过RAN转发给所述用户设备。
  9. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    接收所述用户设备基于所述服务执行结果发送的确认消息；
    响应于所述确认消息，将执行所述AI服务所占用的资源进行释放。
  10. 一种AI服务执行方法，其特征在于，应用于策略控制功能PCF网元，所述方法包括：
    接收会话管理功能SMF网元发送的多个用户面功能UPF网元的属性信息和用户设备的AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
    根据所述多个UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
    将所述AI服务执行策略发送至所述SMF网元，所述AI服务执行策略用于指示所述SMF网元控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
  11. 根据权利要求10所述的方法，其特征在于，所述AI服务执行策略包括策略控制和计费PCC规则，所述PCC规则至少包括：所述PCC规则的标识、为所述AI服务分配的目标UPF的标识、所述目标UPF网元的优先级中的至少一种。
  12. 根据权利要求10所述的方法，其特征在于，所述UPF网元的属性信息至少包括所述UPF网元支持的AI服务，所述AI服务请求消息至少包括AI服务标识和AI服务类型；
    所述根据所述多个UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略，包括：
    根据所述多个UPF网元支持的AI服务，从所述多个UPF网元中确定与所述AI服务标识和AI服务类型匹配的至少一个目标UPF网元；
    根据所述至少一个目标UPF网元支持的AI服务，确定所述至少一个目标UPF网元的优先级。
  13. 一种AI服务执行方法，其特征在于，应用于用户面功能UPF网元，所述方法包括：
    接收会话管理功能SMF网元发送的AI服务执行策略，所述AI服务执行策略是所述SMF网元根据多个用户面功能UPF网元的属性信息和AI服务请求消息确定的，所述AI服务请求消息用于指示用户设备请求的AI服务；
    根据所述AI服务执行策略执行所述AI服务。
  14. 根据权利要求13所述的方法，其特征在于，所述UPF网元为用于执行所述AI服务执行策略的多个目标UPF网元中的主UPF网元，所述根据所述AI服务执行策略执行所述AI服务，包括：
    为所述多个目标UPF分配所述AI服务执行策略，所述AI服务执行策略用于指示所述多个目标UPF分别根据分配的所述AI服务执行策略执行所述AI服务；
    接收所述多个目标UPF中的次UPF根据所述AI服务执行策略执行所述AI服务后得到的执行结果；
    将所述主UPF网元的执行结果和所述次UPF网元的执行结果进行聚合得到所述AI服务的服务执行结果。
  15. 根据权利要求13或14所述的方法，其特征在于，所述AI服务请求消息中包 括用于执行所述AI服务的AI数据，所述根据所述AI服务执行策略执行所述AI服务，包括：
    从非结构化数据存储功能UDSF网元获取所述AI数据；所述AI数据是所述AI服务请求消息发送至SMF网元后，由所述SMF网元存储在所述UDSF网元中的，或者所述AI数据是所述AI服务请求消息发送至移动性管理功能AMF网元后，由所述AMF网元存储在所述UDSF网元中的；
    根据所述AI服务执行策略和所述AI数据执行所述AI服务。
  16. 一种AI服务执行方法，其特征在于，应用于用户设备，所述方法包括：
    向会话管理功能SMF网元发送AI服务请求消息，所述AI服务请求消息用于指示所述用户设备请求的AI服务，所述AI服务请求消息用于指示所述SMF网元根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略，并控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个；
    接收所述SMF网元发送的所述AI服务的服务执行结果。
  17. 根据权利要求16所述的方法，其特征在于，所述向SMF网元发送AI服务请求消息，包括：
    通过无线接入网RAN向移动性管理功能AMF网元发送所述AI服务请求消息，所述AI服务请求消息用于所述AMF网元将所述AI服务请求消息转发至所述SMF网元。
  18. 根据权利要求16所述的方法，其特征在于，所述接收所述SMF网元发送的所述AI服务的服务执行结果，包括：
    通过RAN接收AMF网元转发的所述服务执行结果，所述服务执行结果是所述目标UPF网元执行所述AI服务后生成并发送给所述SMF网元的。
  19. 一种AI服务执行装置，其特征在于，应用于会话管理功能SMF网元，所述装置包括：
    接收模块，被配置为接收AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
    确定模块，被配置为根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
    执行模块，被配置为控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
  20. 一种AI服务执行装置，其特征在于，应用于策略控制功能PCF网元，所述装置包括：
    接收模块，被配置为接收会话管理功能SMF网元发送的多个用户面功能UPF网元的属性信息和用户设备的AI服务请求消息，所述AI服务请求消息用于指示用户设备请求的AI服务；
    确定模块，被配置为根据所述多个UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略；
    发送模块，被配置为将所述AI服务执行策略发送至所述SMF网元，所述AI服务执行策略用于指示所述SMF网元控制目标UPF网元根据所述AI服务执行策略执行所述 AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个。
  21. 一种AI服务执行装置，其特征在于，应用于用户面功能UPF网元，所述装置包括：
    接收模块，被配置为接收会话管理功能SMF网元发送的AI服务执行策略，所述AI服务执行策略是所述SMF网元根据多个用户面功能UPF网元的属性信息和AI服务请求消息确定的，所述AI服务请求消息用于指示用户设备请求的AI服务；
    执行模块，被配置为根据所述AI服务执行策略执行所述AI服务。
  22. 一种AI服务执行装置，其特征在于，应用于用户设备，所述装置包括：
    发送模块，被配置为向会话管理功能SMF网元发送AI服务请求消息，所述AI服务请求消息用于指示所述用户设备请求的AI服务，所述AI服务请求消息用于指示所述SMF网元根据多个用户面功能UPF网元的属性信息和所述AI服务请求消息确定AI服务执行策略，并控制目标UPF网元根据所述AI服务执行策略执行所述AI服务，所述目标UPF网元为所述多个UPF网元中的一个或多个；
    接收模块，被配置为接收所述SMF网元发送的所述AI服务的服务执行结果。
  23. 一种会话管理功能SMF网元，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为在执行所述可执行指令时，实现权利要求1～9中任一项所述方法的步骤。
  24. 一种策略控制功能PCF网元，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为在执行所述可执行指令时，实现权利要求10～12中任一项所述方法的步骤。
  25. 一种用户面功能UPF网元，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为在执行所述可执行指令时，实现权利要求13～15中任一项所述方法的步骤。
  26. 一种用户设备，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为在执行所述可执行指令时，实现权利要求16～18中任一项所述方法的步骤。
  27. 一种计算机可读存储介质，其上存储有计算机程序指令，其特征在于，所述程序指令被处理器执行时实现权利要求1～9中任一项所述方法的步骤，或所述程序指令被处理器执行时实现权利要求10～12中任一项所述方法的步骤，或所述程序指令被处理器 执行时实现权利要求13～15中任一项所述方法的步骤，或所述程序指令被处理器执行时实现权利要求16～18中任一项所述方法的步骤。
  28. 一种芯片，其特征在于，包括处理器和接口；所述处理器用于读取指令以执行权利要求1～9中任一项所述的方法，或所述处理器用于读取指令以执行权利要求10～12中任一项所述的方法，或所述处理器用于读取指令以执行权利要求13～15中任一项所述的方法，或所述处理器用于读取指令以执行权利要求16～18中任一项所述的方法。
PCT/CN2022/089128 2022-04-25 2022-04-25 Ai服务执行方法、装置、网元、存储介质及芯片 WO2023206049A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202280001396.4A CN117296374A (zh) 2022-04-25 2022-04-25 Ai服务执行方法、装置、网元、存储介质及芯片
PCT/CN2022/089128 WO2023206049A1 (zh) 2022-04-25 2022-04-25 Ai服务执行方法、装置、网元、存储介质及芯片

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/089128 WO2023206049A1 (zh) 2022-04-25 2022-04-25 Ai服务执行方法、装置、网元、存储介质及芯片

Publications (1)

Publication Number Publication Date
WO2023206049A1 true WO2023206049A1 (zh) 2023-11-02

Family

ID=88516425

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/089128 WO2023206049A1 (zh) 2022-04-25 2022-04-25 Ai服务执行方法、装置、网元、存储介质及芯片

Country Status (2)

Country Link
CN (1) CN117296374A (zh)
WO (1) WO2023206049A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112311691A (zh) * 2019-07-26 2021-02-02 华为技术有限公司 策略控制方法、设备及系统
WO2021067913A1 (en) * 2019-10-03 2021-04-08 Ofinno, Llc Configuration of time sensitive bridge during handover
CN113243116A (zh) * 2018-09-30 2021-08-10 华为技术有限公司 本地局域网通信方法、设备及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113243116A (zh) * 2018-09-30 2021-08-10 华为技术有限公司 本地局域网通信方法、设备及系统
CN112311691A (zh) * 2019-07-26 2021-02-02 华为技术有限公司 策略控制方法、设备及系统
WO2021067913A1 (en) * 2019-10-03 2021-04-08 Ofinno, Llc Configuration of time sensitive bridge during handover

Also Published As

Publication number Publication date
CN117296374A (zh) 2023-12-26

Similar Documents

Publication Publication Date Title
US11457082B2 (en) Service processing method and apparatus of service-based architecture
US20230217366A1 (en) Access method, access apparatus, and storage medium
US20220303276A1 (en) Network connection method, hotspot terminal and management terminal
EP3723413B1 (en) Policy control method, apparatus, and system
US20230276430A1 (en) Resource scheduling method and apparatus, communication device and storage medium
US20230055967A1 (en) Communication method and apparatus of multicast and broadcast service, electronic device, and storage medium
CN103036597B (zh) 近距离设备间的资源分享方法和设备
WO2021258370A1 (zh) 一种通信处理方法、通信处理装置及存储介质
US8719332B2 (en) Method and system for content categorization
JP2017504921A (ja) インスタントメッセージング方法、装置、プログラム及び記録媒体
TW202119255A (zh) 推理系統、推理方法、電子設備及電腦儲存媒體
CN112799825A (zh) 一种任务处理方法和网络设备
WO2023109452A1 (zh) 控制方法、电子设备及计算机可读存储介质
WO2021046674A1 (zh) 数据处理方法、装置、电子设备和计算机可读存储介质
WO2017045450A1 (zh) 资源的操作处理方法及装置
US20140051392A1 (en) Method and apparatus for collaborative web resource delivery
WO2018196102A1 (zh) 信息传递方法、装置及计算机可读存储介质
US20220141309A1 (en) Efficient resource representation exchange between service layers
WO2022027358A1 (zh) 一种智能家居系统管理方法、电子设备及存储介质
WO2023206049A1 (zh) Ai服务执行方法、装置、网元、存储介质及芯片
US20180278535A1 (en) Communication device, communication method, and storage medium
WO2023005898A1 (zh) 多终端联合会话管理方法、网络侧设备及终端
WO2023016396A1 (zh) 计算会话更新方法、装置及通信设备
US11019131B2 (en) Method and system of file transfer using device-to-device communication technique in messenger
WO2022032646A1 (zh) 信息传输方法、装置及通信设备

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 202280001396.4

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22938906

Country of ref document: EP

Kind code of ref document: A1