Nothing Special   »   [go: up one dir, main page]

WO2024192584A1 - 算力服务调用方法、装置及存储介质 - Google Patents

算力服务调用方法、装置及存储介质 Download PDF

Info

Publication number
WO2024192584A1
WO2024192584A1 PCT/CN2023/082299 CN2023082299W WO2024192584A1 WO 2024192584 A1 WO2024192584 A1 WO 2024192584A1 CN 2023082299 W CN2023082299 W CN 2023082299W WO 2024192584 A1 WO2024192584 A1 WO 2024192584A1
Authority
WO
WIPO (PCT)
Prior art keywords
computing
computing power
service
request
information
Prior art date
Application number
PCT/CN2023/082299
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 PCT/CN2023/082299 priority Critical patent/WO2024192584A1/zh
Publication of WO2024192584A1 publication Critical patent/WO2024192584A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/50—Service provisioning or reconfiguring

Definitions

  • the present disclosure relates to the field of communication technology, and in particular to a computing power service calling method, device and storage medium.
  • UE user equipment
  • Terminal devices will achieve flexible and direct communication through service-based interfaces. Massive terminal devices will register and store their computing power in the network and collaborate with the network to perform distributed computing, which will reduce the task pressure carried by the network and improve the resource utilization efficiency of idle terminal devices. It is not clear in the relevant technology how user equipment collaborates with the network to provide computing services.
  • the computing power service calling method, device and storage medium proposed in the present invention are used for terminal devices to cooperate with the network to provide computing power services.
  • an embodiment of the present disclosure provides a method for calling a computing power service, which is executed by a network device and includes: receiving a service request sent by a first user equipment UE, wherein the service request is used to request to call a computing power service; and sending indication information for executing the computing power service to at least one third UE, wherein the at least one third UE is a UE that meets the corresponding requirements of the computing power service.
  • an embodiment of the present disclosure provides a method for calling a computing power service, which is executed by a first UE and includes: sending a service request to a computing service module, wherein the service request is used to request to call a computing power service.
  • an embodiment of the present disclosure provides a computing power service calling method, which is executed by a third UE, and the method includes: receiving indication information for executing computing power service sent by a computing service module.
  • an embodiment of the present disclosure provides a computing power service calling method, which is executed by a first network element, and the method includes: sending computing power registration information of at least one second UE to a computing service module, wherein the computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that executes the computing power service.
  • an embodiment of the present disclosure provides a computing service module, which is deployed in a core network and includes:
  • a transceiver unit configured to receive a service request sent by a first user equipment UE, wherein the service request is used to request to call a computing power service;
  • the transceiver unit is also used to: send indication information for executing computing power service to at least one third UE, wherein the at least one third UE is a UE that meets the corresponding requirements of the computing power service.
  • an embodiment of the present disclosure provides a first user equipment UE, including:
  • the transceiver unit is used to send a service request to the computing service module, wherein the service request is used to request to call the computing power service.
  • an embodiment of the present disclosure provides a third user equipment UE, including:
  • the transceiver unit is used to receive the instruction information for executing computing power service sent by the computing service module.
  • an embodiment of the present disclosure provides a first network element, including:
  • a transceiver unit configured to send computing power registration information of at least one second UE to the computing service module
  • the computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that performs computing power services.
  • an embodiment of the present disclosure provides a communication device, comprising a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the computer program stored in the memory so that the device executes the method of any one of the first to fourth aspects above.
  • an embodiment of the present disclosure provides a communication device, comprising: a processor and an interface circuit, wherein the interface circuit is used to receive code instructions and transmit them to the processor; the processor is used to run the code instructions to execute the method of any one of the first to fourth aspects above.
  • an embodiment of the present invention provides a computer-readable storage medium for storing instructions, which, when executed, enables the method of any one of the first to fourth aspects to be implemented.
  • the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method of any one of the first to fourth aspects described above.
  • the present disclosure provides a communication system, the system comprising a computing service module, a first user equipment UE, a third UE, The first network element, wherein
  • the computing service module is used to execute the method of the first aspect
  • the first UE is used to perform the method of the second aspect
  • the third UE is used to perform the method of the third aspect
  • the first network element is used to execute the method of the fourth aspect.
  • the method is executed by a computing service module, and the computing service module is deployed in the core network.
  • the method includes: receiving a service request sent by a first user equipment UE, wherein the service request is used to request to call a computing power service; sending an instruction message for executing the computing power service to at least one third UE, wherein at least one third UE is a UE that meets the corresponding requirements of the computing power service.
  • the scheme of the present disclosure realizes the UE to cooperate with the network to provide computing power services. By registering and storing the computing power capabilities of the UE in the network, the collaborative network performs distributed computing, which reduces the task pressure carried by the network, realizes dynamic and flexible computing power scheduling and deployment, and improves the resource utilization efficiency of idle terminal devices.
  • FIG1 is a schematic diagram of the architecture of a communication system provided by an embodiment of the present disclosure.
  • FIG2 is a flow chart of a method for invoking a computing power service provided by an embodiment of the present disclosure
  • FIG3 is a flow chart of a method for invoking a computing power service provided in another embodiment of the present disclosure
  • FIG4 is a flow chart of a method for invoking a computing power service provided in yet another embodiment of the present disclosure
  • FIG5 is a flow chart of a method for invoking a computing power service provided by another embodiment of the present disclosure.
  • FIG6 is a flow chart of a method for invoking a computing power service provided in another embodiment of the present disclosure.
  • FIG7 is a flow chart of a method for invoking a computing power service provided in yet another embodiment of the present disclosure.
  • FIG8 is a flow chart of a method for invoking a computing power service provided by another embodiment of the present disclosure.
  • FIG9 is a flow chart of a method for invoking a computing power service provided in another embodiment of the present disclosure.
  • FIG10 is a flow chart of a method for invoking a computing power service provided in yet another embodiment of the present disclosure.
  • FIG11 is a flowchart of an interactive method for invoking a computing power service provided by an embodiment of the present disclosure
  • FIG12 is an example diagram of an interactive method for invoking a computing power service provided by an embodiment of the present disclosure
  • FIG13 is a schematic diagram of the structure of a computing service module provided by an embodiment of the present disclosure.
  • FIG14 is a schematic diagram of the structure of a computing service module provided by an embodiment of the present disclosure.
  • FIG15 is a schematic structural diagram of a first UE provided by an embodiment of the present disclosure.
  • FIG16 is a schematic structural diagram of a third UE provided by an embodiment of the present disclosure.
  • FIG17 is a schematic diagram of the structure of a third UE provided by an embodiment of the present disclosure.
  • FIG18 is a schematic diagram of the structure of a first network element provided by an embodiment of the present disclosure.
  • FIG19 is a schematic diagram of the structure of a first network element provided by an embodiment of the present disclosure.
  • FIG20 is a schematic diagram of the structure of a communication device provided by an embodiment of the present disclosure.
  • FIG21 is a schematic structural diagram of a communication device provided by an embodiment of the present disclosure as a chip or a chip system;
  • FIG. 22 is a schematic diagram of the structure of a communication system provided by an embodiment of the present disclosure.
  • first, second, third, etc. may be used to describe various information in the disclosed embodiments, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.
  • first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information.
  • the words "if” and “if” as used herein may be interpreted as “at” or "when” or "in response to determination”.
  • AMF Access and Mobility Management Function
  • Non-Access Stratum NAS
  • NAS Non-Access Stratum
  • Access Stratum security control Core network node-to-node signaling for mobility between 3GPP access networks
  • Idle mode terminal device reachability including control and execution of paging retransmissions
  • Registration area management Support for intra-system and inter-system mobility; Access authentication; Access authorization, including roaming rights check; Mobility management control (subscription and policy); Support for network slicing; Session Management Function (SMF) selection.
  • SMF Session Management Function
  • the SMF network element can perform the following main tasks: session management; terminal device IP address allocation and management; user plane function (UPF) selection and control; configure flow control in UPF to route traffic to the appropriate destination; policy enforcement and quality of service (QoS) control part; downlink data notification.
  • MICO Mobile Initiated Connection Only
  • RRC Inactive Radio Resource Control
  • the SMF network element can perform the following main tasks: session management; terminal device IP address allocation and management; user plane function (UPF) selection and control; configure flow control in UPF to route traffic to the appropriate destination; policy enforcement and quality of service (QoS) control part; downlink data notification.
  • UPF user plane function
  • UDM is used for the management of user identification, contract data, authentication data, and user service network element registration management.
  • the various network elements/functions involved in the embodiments of the present disclosure may be an independent hardware device or a function implemented by computer code in a hardware device, which is not limited in the embodiments of the present disclosure.
  • NEF is a network function that 5GC opens to the outside world and provides a standard interface. Based on 3GPP network functions, NEF exposes functions and events to other systems, providing both openness and system security. The existence of NEF standardizes the display of 5GC functions and facilitates third-party access.
  • AF is similar to an application server, which interacts with other 5G core network control plane NFs and provides business services. AF can exist for different application services and can be owned by operators or trusted third parties.
  • AUSF Authentication Server Function
  • AUSF is a network element that supports unified authentication service functions in 5G, realizes 3GPP and non-3GPP access authentication, and is a network entity in the 5G core network (5GC).
  • 5GC 5G core network
  • PCF Policy Control Function
  • PCF supports a unified policy framework to manage network behavior, provides policy rules to network entities for implementation, and accesses subscription information in the Unified Data Repository (UDR).
  • UDR Unified Data Repository
  • UDR is used by UDM to store subscription data or read subscription data and PCF to store policy data or read policy data.
  • NRF Network Repository Function
  • NRF supports the service discovery function, receives NF discovery requests from NF instances, and provides the information of discovered NF instances (discovered) to the NF instances, while maintaining NF profiles of available NF instances and their supported services.
  • UPF User plane function
  • UPF is a bridge between mobile infrastructure (e.g. RAN) and data network (DN).
  • the UPF is the interconnection point between the two, completing the encapsulation and decapsulation of the GTP-U (GRPS Tunneling Protocol) protocol on the UP.
  • GTP-U GTP-U
  • the UPF is used for the Protocol Data Unit (PDU) Session Anchor Point (Session Anchor Point) within the Radio Access Technologies (RAT) or for mobility between RATs, including sending one or more End Marker Packets (EMP) to the gNB (NG-RAN node).
  • PDU Protocol Data Unit
  • Session Anchor Point Session Anchor Point
  • RAT Radio Access Technologies
  • EMP End Marker Packets
  • the 6G (sixth generation mobile communication system) service-oriented architecture abstracts the functions of network elements into multiple services. Compared with the traditional network architecture, the 6G service-oriented architecture introduces the synaesthesia fusion technology to provide terminal devices with a variety of services (such as perception services, computing power services, artificial intelligence (AI) services, etc.).
  • the 6G service-oriented architecture includes network elements for providing services. Other network elements in the 6G service-oriented architecture achieve deep integration and mutual enhancement of multi-dimensional perception, collaborative communication, and intelligent computing functions through collaboration and sharing with the network elements providing services, thereby enabling the network to have the ability of intelligent interaction and processing of new information flows and wide-area intelligent collaboration.
  • Figure 1 is a schematic diagram of a 6G service-oriented network architecture provided by an embodiment of the present disclosure.
  • Figure 1 includes network elements in the 6G service-oriented network architecture and interfaces for communication between network elements.
  • the dotted box is used to represent the network element that provides services.
  • the network element can realize capabilities such as sensing, AI calculation, and also has storage capabilities.
  • the radio access network (RAN) shown in FIG1 may be an entity on the network side for transmitting or receiving signals.
  • the RAN may be an evolved NodeB (eNB), a transmission reception point (TRP), a next generation NodeB (gNB) in an NR system, a base station in other future mobile communication systems, or an access node in a wireless fidelity (WiFi) system.
  • eNB evolved NodeB
  • TRP transmission reception point
  • gNB next generation NodeB
  • WiFi wireless fidelity
  • the embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the network device.
  • the network device provided in the embodiments of the present disclosure may be composed of a centralized unit (CU) and a distributed unit (DU), wherein the CU may also be referred to as a control unit (control unit).
  • the CU-DU structure may be used to split the protocol layer of the network device, such as the base station, and the functions of some protocol layers are placed in the CU for centralized control, and the functions of the remaining part or all of the protocol layers are distributed in the DU, and the DU is centrally controlled by the CU.
  • the user equipment (UE) (or terminal device) shown in FIG1 can be an entity on the user side for receiving or transmitting signals, such as a mobile phone.
  • the terminal device can also be called a terminal, a mobile station (MS), a mobile terminal (MT), etc.
  • the terminal device can be a car with communication function, a smart car, a mobile phone, a wearable device, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, a wireless terminal device in smart home, etc.
  • the embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the UE.
  • FIG1 also includes network elements such as AMF, SMF, and UDM.
  • the terminal device i.e., the UE in FIG1
  • the terminal device i.e., the UE in FIG1
  • the network element providing the service can provide the terminal device with the service requested by the terminal device.
  • the communication system described in the embodiment of the present disclosure is for the purpose of more clearly illustrating the technical solution of the embodiment of the present disclosure, and does not constitute a limitation on the technical solution provided by the embodiment of the present disclosure.
  • a person skilled in the art can know that with the evolution of the system architecture and the emergence of new business scenarios, the technical solution provided by the embodiment of the present disclosure is also applicable to similar technical problems.
  • the computing service calling method, device, equipment and storage medium provided by the embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.
  • the method can be applied to the fifth generation mobile communication technology (Fifth Generation, 5G) and its subsequent communication technologies, such as the fifth generation mobile communication technology evolution (5G-advanced), the sixth generation mobile communication technology (Sixth Generation, 6G), etc., which are not limited in the present disclosure.
  • 5G fifth generation mobile communication technology
  • 6G sixth generation mobile communication technology
  • Figure 2 is a flow chart of a computing power service calling method provided by an embodiment of the present disclosure, which is executed by a computing service module, and the computing service module is deployed in the core network.
  • the computing service module can be deployed in the core network in the form of a network element device for overall scheduling and deployment of computing power, and its specific presentation form is not limited in the present disclosure.
  • a user plane connection has been established between a UE (referred to as a first UE in this disclosure) and a core network;
  • second UEs that can provide computing power support for the core network have registered their computing power with the computing power storage network element in the core network (referred to as the first network element in this disclosure).
  • third UEs all or part of the second UEs (referred to as third UEs in this disclosure) can receive computing power tasks allocated from the core network.
  • the computing power service calling method may include the following steps:
  • Step 201 Receive a service request sent by a first user equipment UE.
  • the business request is used to request to call a computing power service.
  • the service request includes at least one of the following: an identifier of the first UE; an identifier of the computing power service; a type of the computing power service; computing power requirements; and network bandwidth.
  • the first UE may send a service request when a computing service is needed. For example, if the first UE needs to perform image processing, it may trigger a request to the core network to call a computing service for image processing.
  • the identifier of the UE may be a user permanent identifier (SUPI), or other identifiers that can uniquely identify the UE, which are not limited in the present disclosure.
  • the identifier of the computing service (CalService I D) may be a code that marks the computing service that can be called.
  • the type of the computing service (Cal Service Type) may be a code or field that marks the type of the computing service that can be called.
  • the computing power requirement is the computing power requirement of the UE for the current computing task to be performed.
  • the network bandwidth may be the downlink network bandwidth, so that the network device returns a data packet of the calculation result that meets the bandwidth requirement to the UE.
  • the computing service module can receive the service request sent by the first UE through the user plane connection.
  • the user plane connection between the first UE and the core network can be established in the following manner: when the SMF network element receives the service call request, it means that the terminal device currently requests to call the service.
  • the SMF network element can establish a connection with the first network element (the network element that provides the service requested by the terminal device), so that the SMF network element subsequently interacts with the first network element to establish a user plane connection between the first network element and the terminal device, thereby enabling the first network element to communicate with the terminal device through the user plane to provide the terminal device with the service requested by the terminal device.
  • the computing service module may also receive the service request sent by the first UE through other connections, which is not limited in the present disclosure.
  • Step 202 Send indication information for executing computing power service to at least one third UE.
  • the indication information may be information used to instruct the third UE to perform computing services.
  • the indication information includes data information and/or task requirements required to perform the computing service.
  • the computing service module can send the computing task requested by the first UE as a whole to a third UE to perform computing services.
  • the computing service module may also divide the computing service requested by the UE into one or more sub-services, that is, split the distributed computing service, send indication information to at least one determined third UE based on the divided sub-services, and allocate multiple sub-services to one or more third UEs for calculation, wherein the computing power that the third UE can provide is greater than and similar to the computing power required for the sub-service, and the total computing power capacity of multiple third UEs may be greater than or equal to the computing power requirement of the computing task requested by the first UE, so as to provide the first UE with reasonable computing power service for the service and avoid waste of computing power resources.
  • the method is executed by a computing service module, and the computing service module is deployed in the core network.
  • the method includes: receiving a service request sent by a first user equipment UE, wherein the service request is used to request to call a computing power service; obtaining computing power registration information of at least one second UE from a first network element; based on the computing power registration information of at least one second UE, determining at least one third UE that executes the computing power service from at least one second UE; and sending instruction information for executing the computing power service to at least one third UE.
  • the scheme of the present disclosure realizes that UE cooperates with the network to provide computing power services. By registering and storing the computing power capability of the UE in the network, the collaborative network performs distributed computing, which reduces the task pressure carried by the network, realizes dynamic and flexible computing power scheduling and deployment, and improves the resource utilization efficiency of idle terminal devices.
  • the method further includes: obtaining computing power registration information of at least one second UE from the first network element.
  • the first network element may be a computing power storage network element (see the "storage" network element in FIG. 1), which can store information about computing power nodes that provide computing power services.
  • the UE can collaborate with the core network to perform computing power deployment.
  • the first network element can store information about UEs that can provide computing power services, and can also store The information of other computing nodes, such as the information of computing nodes in the core network or the information of third-party computing nodes outside the core network, is not limited by the present disclosure.
  • the computing service module can coordinate and deploy the computing task requested by the first UE by acquiring the information in the first network element.
  • the first network element may be a network element specifically used to store computing power node information, or it may be a general or shared storage network element in the core network, used to store computing power node information and other information, which is not limited in this disclosure.
  • the first network element may also be deployed outside the core network, and the information stored therein may be exchanged with the core network through the NEF network element, which is not limited in the present disclosure.
  • the second UE may be all or part of the UEs stored in the first network element that can provide collaborative computing services for the core network, wherein computing power information such as the size and type of computing power that the UE can provide is registered in the first network element, and the computing service module may obtain the computing power registration information of all or part of the second UEs stored in the first network element to determine one or more UEs that perform the computing power service.
  • the computing power registration information may be information about the UE computing power capability registered or registered by the UE in the core network (specifically, for example, the first network element).
  • the computing power registration information includes at least the identifier, computing power type, and available computing power size of each second UE.
  • a computing service module receives computing power registration information of one or more second UEs that can provide computing power services sent by a computing power storage network element to determine whether there is at least one third UE among the second UEs that meets the computing power requirements in the service request.
  • the method also includes: determining at least one third UE that performs computing power services from at least one second UE based on the computing power registration information of at least one second UE.
  • this step may be performed between steps 201 and 202.
  • the computing service module may perform service perception on the service request, wherein the service perception is an intelligent assessment of the computing power size and computing power type required for the request according to different types of requests, such as voice processing or image processing.
  • the computing service module determines at least one third UE that performs computing power services from the at least one second UE based on the service perception and the computing power information of each UE registered in the first network element.
  • the computing service module may use a preset algorithm to determine at least one third UE, and the type of the algorithm is not limited in the present disclosure.
  • FIG3 is a flow chart of a method for calling a computing service provided by an embodiment of the present disclosure.
  • the method is executed by a computing service module.
  • the method for calling a computing service may include the following steps:
  • Step 301 Receive a service request sent by a first user equipment UE.
  • the business request is used to request to call the computing power service.
  • step 301 is the same as that of step 201 in the embodiment shown in FIG. 2 , and the relevant description of FIG. 2 may be referred to, and will not be repeated here.
  • Step 302 Send an acquisition request to the first network element.
  • the acquisition request is used to request the computing power registration information of the UE.
  • the acquisition request may include the size of computing power required to perform the computing power service, so as to request the first network element to search for the computing power registration information of the UE that can provide the computing power size; or, the acquisition request may include the computing power type, so as to request the first network element to search for the computing power registration information of the UE that can provide this type of service; or, the acquisition request may include the status of the UE, so as to request the first network element to search for the computing power registration information of the UE in a specific state (for example, a UE in an idle state).
  • the computing power storage network element stores computing power registration information of one or more UEs (i.e., the second UE mentioned above), and the computing power resources of each UE may be the same or different, that is, the size and type of computing power that can be provided may be the same or different.
  • the first network element may also store the status of each UE, such as an idle state or an occupied state.
  • the status of a UE that has been called is stored in the first network element as an occupied state, and it may not be called again before it completes the current computing power service; the status of a UE that has not been called is stored in the computing power storage network element as an idle state, and can provide computing power services for the above services.
  • the first network element may also store the remaining computing power of each UE. For example, if the total computing power of a UE is X, and the UE is called to execute computing service A, and the computing power required to execute computing service A is Y, the first network element may store the remaining computing power XY of the UE, and when computing service B needs to be executed and the computing power required for computing service B is less than XY, the computing service module may To call the UE to execute computing service B.
  • Step 303 Receive computing power registration information of at least one second UE sent by the first network element.
  • the computing power registration information may be information about the UE computing power capability registered or registered by the UE in the core network (specifically, for example, the first network element).
  • the computing power registration information includes at least the identifier, computing power type, and available computing power size of each second UE.
  • a computing service module receives computing power registration information of one or more second UEs that can provide computing power services and sends it from a computing power storage network element, so as to determine whether there is at least one third UE among the second UEs that meets the computing power requirements in the service request based on service perception.
  • Step 304 Determine whether there is at least one third UE from the at least one second UE.
  • the total computing power resources of at least one third UE meet the computing power requirements in the service request.
  • the computing service module can determine whether there is a third UE from the second UE based on the service request and the computing power registration information of at least one second UE, using a resource matching algorithm. For example, the computing service module can extract relevant information in the service request through service perception, such as the identifier of the first UE, the computing power service identifier, the computing power service type, the computing power demand, etc., to determine the computing power size, type, and other information required for the computing task requested by the first UE, and determine whether there is a third UE that can provide computing power services through the computing power registration information of the second UE obtained from the first network element, that is, determine whether there is at least one third UE that meets the computing power demand in the service request.
  • the computing service module measures resource utilization through a specific algorithm to determine one or more third UEs to provide the optimal computing deployment decision.
  • Step 305 Determine that at least one third UE does not exist in the at least one second UE, and send feedback information to the first UE.
  • the feedback information is used to notify the first UE that the call failed.
  • the feedback information may carry a call failure field, which indicates that there is currently no available computing power node to provide computing power services for it, and the call failed, and the first UE may abandon the call of the computing power service.
  • the feedback information may be used to notify the first UE to wait for a UE that meets the computing power requirements of the first UE.
  • the feedback information may carry a waiting field, which indicates that there is currently no available computing power node to provide computing power services for it, and the first UE may wait for other computing power nodes that can provide computing power services for it.
  • the computing service module determines that there is no third UE in the second UE that can provide computing power services for the first UE, that is, there is no third UE that meets the computing power requirements for any of the split sub-services of the second UE, then the computing service module cannot schedule the third UE for the computing task requested by the first UE.
  • the computing service module can send feedback information to the first UE to notify the first UE that the call failed, and the first UE can decide whether to abandon the current computing task, or wait, or change to other computing tasks.
  • the computing service module may also directly notify the first UE to wait through feedback information.
  • the first UE may not give up the current computing task, but wait for a third UE that can provide services to the first UE and meets the computing power requirements of the current computing task.
  • Step 306 Determine that there is at least one third UE in the at least one second UE, and send indication information for executing computing power service to the at least one third UE.
  • the indication information includes data information and/or task requirements required to perform computing services.
  • the computing service module determines that there is at least one third UE in the second UE that can provide computing power services to the first UE, the computing service module will allocate the split sub-services to the third UEs respectively, that is, send indication information to the third UEs respectively to instruct the third UEs to perform computing power services.
  • the indication information carries the data information and/or task requirements required to perform the computing service.
  • the computing service requested by the first UE is to perform image processing on 100 pictures.
  • the computing service module can split the task, for example, into 5 subtasks, and process 20 images under each subtask, and determine the third UE to execute each sub-service in the second UE, for example, 5.
  • the computing service module can send the 20 images and the task requirements for image processing to each third UE, so that each third UE performs the computing and processing tasks for the received images.
  • the computing service module determines that there are multiple UEs in the second UE that meet the computing task requirements, and the multiple UEs are called fourth UEs; then The computing service module may further select one or more third UEs from the plurality of fourth UEs, wherein the selection may be based on one or more dimensions, such as utilization and/or computing efficiency and/or computing capability.
  • an indication information of a computing service can be sent to one or more third UEs that can provide a total computing power greater than and closest to the computing power required by the first UE to request the computing service, so as to avoid wasting terminal device resources.
  • the computing service requested by the first UE is to process 500 images.
  • the indication information of the computing service can be sent to the third UE that can process 550 images to reasonably utilize terminal resources.
  • a UE with high computing efficiency and/or computing power can be selected as the third UE.
  • the computing power service requested by the first UE is to process 500 images.
  • there are two idle third UEs in the second UE one of which has a computing power of 1.5TFlops and the other has a computing power of 2TFlops.
  • the UE with a computing power of 2TFlops is selected as the third UE.
  • the computing power service requested by the first UE is to process 500 images.
  • the second UE there are two idle third UEs in the second UE, one of which has a computing efficiency of 0.1568 (unit computing power/carbon emissions) and the other has a computing efficiency of 0.4325 (unit computing power/carbon emissions).
  • the UE with a computing efficiency of 0.4325 is selected as the third UE.
  • the method is executed by a computing service module, including: receiving a service request sent by a user equipment UE through a user plane connection; sending a computing power registration information acquisition request to a first network element; receiving computing power registration information of at least one second UE sent by the first network element; based on the service request and the computing power registration information of at least one second UE, using a resource matching algorithm, determining whether there is at least one third UE from at least one second UE, wherein the total computing power resources of at least one third UE meet the computing power requirements in the service request; determining that there is no at least one third UE in at least one second UE, and sending feedback information to the first UE; determining that there is at least one third UE in at least one second UE, and sending indication information of executing computing power service to at least one third UE.
  • the scheme of the present disclosure realizes that the computing service module matches the computing power resource information through business perception, allocates the required third UE to the business request, and then enables the first UE to send indication information to the allocated third UE, so that the computing service module efficiently and reasonably calls the third UE, realizes flexible and accurate UE computing power scheduling, and improves network utilization and communication efficiency.
  • FIG4 is a flow chart of a method for calling a computing service provided by an embodiment of the present disclosure.
  • the method is executed by a computing service module. Based on the embodiment shown in FIG3 , the method for calling a computing service may include the following steps:
  • Step 401 Receive a service request sent by a first user equipment UE.
  • the business request is used to request to call the computing power service.
  • Step 402 Obtain computing power registration information of at least one second UE from the first network element.
  • Step 403 Based on the service request and the computing power registration information of at least one second UE, a resource matching algorithm is used to determine whether there is at least one third UE from the at least one second UE, wherein the total computing power resources of the at least one third UE meet the computing power requirements in the service request.
  • steps 401-403 are the same as the relevant steps in the embodiments shown in FIG. 2-FIG . 3 above, and reference may be made to the relevant description of the above embodiments, which will not be repeated here.
  • Step 404 Determine that there is at least one third UE among the at least one second UE, and send indication information for executing computing power service to the at least one third UE.
  • the computing service module determines that there is at least one third UE among the second UEs that can provide computing services for the first UE, the computing service module will allocate the split sub-services to the determined third UEs respectively, that is, send indication information to the determined third UEs respectively to instruct each third UE to perform computing services.
  • step 404 may include: sending indication information to the second network element.
  • the indication information is used to instruct the second network element to send data information and/or task requirements required to perform computing services to at least one third UE.
  • the second network element may be an AMF network element.
  • the computing service may be deployed on multiple computing nodes (for example, when the computing service is executed collaboratively by UE in the present disclosure, the UE may serve as a computing node), and the computing service module may send an indication message to the AMF network element, and the indication message may also include an identifier of the third UE.
  • the AMF receives the indication message and sends the indication message to the AMF network element.
  • the information is sent to the third UE to implement the calling of the computing service module to the third UE.
  • the indication information carries the data information and/or task requirements required to perform the computing service.
  • the computing service requested by the first UE is to process 100 images.
  • the computing service module can split the task, for example, into 5 subtasks, each of which processes 20 images, and determine the third UE to execute each sub-service, for example, 5.
  • the computing service module can send the 20 images and the task requirements for image processing to each third UE, so that each third UE performs the computing and processing tasks for the received images.
  • Step 405 Receive calculation results fed back by at least one third UE.
  • the calculation result is a result obtained by at least one third UE performing a computing service based on the indication information.
  • the third UE performs task processing based on the data information and/or task requirements provided by the received indication information, and the computing service module can receive the calculation results obtained after one or more third UEs determined in the second UE complete the allocated computing power service.
  • Step 406 Aggregate the calculation results to obtain aggregate results.
  • the task division of each third UE may be different, and the computing service module will divide the received computing tasks into multiple steps, such as data cleaning, model training, etc.
  • the computing service module after completing data cleaning, one of the third UEs returns the cleaned data to the computing service module, and the computing service module then sends the data to another third UE for model training. After the training is completed, the data is returned to the computing service module.
  • the computing service module can aggregate the calculation results of each third UE to obtain an aggregated result.
  • each third UE may also be the same, and each third UE may return the calculation result to the calculation service module for aggregation, complete the aggregation processing of the calculation results of each third UE, and obtain the aggregated result.
  • the calculation service module can aggregate and package the calculation results returned by each module, without the need to perform secondary processing on each calculation result.
  • Step 407 Send the aggregation result to the first UE.
  • the computing service module may feed back the aggregation result to the first UE through the user plane connection.
  • the first UE establishes a connection with the core network through the user plane.
  • the computing service module receives the aggregation result, it indicates that the computing service module completes the computing power call service, and sends the above aggregation result to the first UE through the user plane connection established between the computing service module and the terminal device.
  • establishing a connection through the user plane can directly establish a connection with each third UE that provides computing power from the user plane, and can also upload data or parameters, while the control plane cannot send data packets by default.
  • the calculation service module may directly send each calculation result to the first UE without performing aggregation processing, which is not limited in the present disclosure.
  • Step 408 Receive confirmation information sent by the first UE.
  • the first UE sends a confirmation message to the computing service module based on the received aggregation result sent by the computing service module, and the computing service module receives the confirmation message.
  • the first UE may request to call the computing service again or modify it in other ways, which is not limited by the present disclosure.
  • the method further includes: sending a status update request to the first network element.
  • the status update request may be used to request to update the status of the UE stored in the first network element, wherein the status may be an occupied state or an idle state.
  • this step may be an optional step, and the present disclosure does not limit the order of execution.
  • this step may be performed after step 404 and before step 405, that is, when the computing service module sends the computing task to the third UE, the computing service module may send a status update request to the first network element to change the state of the third UE from an idle state to an occupied state.
  • this step may be performed after step 405, that is, when the computing service module receives the computing result fed back by the third UE, the computing service module may send a status update request to the first network element to change the state of the third UE from an occupied state to an idle state. This is not limited in the present disclosure.
  • the status update request can be used to request the remaining computing power of the third UE.
  • the remaining computing power of each UE can also be stored in the first network element.
  • the total computing power of a UE is X
  • the UE is called to execute computing service A.
  • the computing power required to execute computing service A is Y.
  • the first network element can store the remaining computing power XY of the UE, and when computing service B needs to be executed and the computing power required for computing service B is less than XY, the computing service module can call the UE to execute computing service B.
  • the computing service module completes service segmentation and allocation, sends the sub-service to the third UE, and can change the computing power state of the called third UE by sending a status update request to the first network element, thereby changing the called third UE from an idle state to an occupied state.
  • the computing service module receives the computing result fed back by the third UE, indicating that the computing service of the third UE has ended, sends a status update request to the first network element, changes the computing status of the third UE, and changes the third UE that has completed the computing service from an occupied state to an idle state.
  • the method is executed by a computing service module, including: receiving a service request sent by a first user equipment UE, obtaining computing power registration information of at least one second UE from a first network element, and based on the service request and the computing power registration information of at least one second UE, using a resource matching algorithm to determine whether there is at least one third UE from at least one second UE, wherein the total computing power resources of at least one third UE meet the computing power requirements in the service request, determining that there is at least one third UE in at least one second UE, sending indication information for executing computing power service to at least one third UE, receiving the calculation results fed back by at least one third UE, aggregating the calculation results to obtain the aggregated results, sending the aggregated results to the first UE, receiving the confirmation information sent by the first UE, sending a status update request to the first network element, and sending an indication request to the second network element, thereby realizing that the computing power service module reasonably calls
  • FIG5 is a flow chart of a method for invoking a computing power service provided by an embodiment of the present disclosure. The method is executed by a first UE. As shown in FIG5 , the method for invoking a computing power service may include the following steps:
  • Step 501 Send a service request to a computing service module.
  • the business request is used to request to call the computing power service.
  • the service request includes at least one of the following: an identifier of the first UE; an identifier of the computing power service; a type of the computing power service; computing power requirements; and network bandwidth.
  • the identifier of the first UE may be a user permanent identifier (SUbscription Permanent Identifier, SUPI), or other identifiers that can uniquely identify the UE, which are not limited in the present disclosure.
  • the identifier of the computing service (Cal Service ID) may be a code that marks the computing service that can be called.
  • the type of the computing service (Cal Service Type) may be a code or field that marks the type of the computing service that can be called.
  • the computing power requirement is the computing power size requirement of the UE for the current computing task to be performed.
  • the network bandwidth may be the downlink network bandwidth, so that the network device returns a data packet of the calculation result that meets the bandwidth requirement to the UE.
  • the user plane connection between the first UE and the core network may be established in the following manner: when the SMF network element receives a service call request, it indicates that the terminal device currently requests to call a service. At this time, the SMF network element may establish a connection with the first network element (the network element that provides the service requested by the terminal device), so that the SMF network element subsequently interacts with the first network element to establish a user plane connection between the first network element and the terminal device, thereby enabling the first network element to communicate with the terminal device through the user plane to provide the terminal device with the service requested by the terminal device. It should be understood that the present disclosure does not limit the method for establishing a user plane connection, and user plane connections established in other possible ways all fall within the scope of the present disclosure.
  • the method is executed by the first UE, including: sending a service request to the computing service module, wherein the service request is used to request to call the computing service.
  • the user equipment first UE is connected to the computing service module, providing a basis for the first UE to successfully call the core network computing service.
  • FIG6 is a flow chart of a method for invoking a computing power service provided by an embodiment of the present disclosure.
  • the method is executed by a first UE.
  • the method for invoking a computing power service may include the following steps:
  • Step 601 Receive the aggregation result sent by the computing service module.
  • the aggregation result is a result obtained by the computing service module performing aggregation processing on the computing results of at least one third UE.
  • the first UE may receive the aggregated result sent by the computing service module to obtain the result of this computing service call.
  • the first UE may receive various calculation results sent by the computing service module without aggregation processing to obtain the result of this computing service call.
  • Step 602 Send confirmation information to the computing service module.
  • the first UE receives the calculation result of the computing service module and sends a confirmation message to confirm that it has received the computing power calling service.
  • the method is executed by the first UE, including: receiving the aggregation result sent by the computing service module, wherein the aggregation result is the result obtained by the computing service module aggregating the computing results of at least one third UE, and sending confirmation information to the computing service module, so that the UE can receive the computing power service call result in a timely and accurate manner, and feedback the received computing power service, so as to timely change the computing power status of the third UE, thereby improving the resource utilization efficiency of idle terminal devices.
  • the first UE may receive feedback information sent by the computing service module, and the feedback information is used to notify the first UE that the call failed and/or wait for a UE that meets the computing power requirements.
  • the first UE determines whether to abandon the current computing task, or wait, or change to other computing tasks, or in response to the feedback information, waits for a third UE that can provide services to the first UE and meets the computing power requirements of the current computing task.
  • FIG7 is a flow chart of a method for invoking a computing service provided by an embodiment of the present disclosure, the method being executed by a third UE.
  • the method for invoking a computing service may include the following steps:
  • Step 701 Receive instruction information for executing computing power services sent by the computing service module.
  • the indication information includes data information and/or task requirements required to perform the computing service.
  • the third UE receives the indication information of executing computing power service sent by the computing service module. Through the deployment of the computing service module, each third UE can execute the corresponding computing task to provide computing power service.
  • the method is executed by the third UE, including: receiving the instruction information of executing the computing service sent by the computing service module, so that the third UE executes the allocated computing task according to the instruction information.
  • FIG8 is a flow chart of a method for invoking a computing power service provided in an embodiment of the present disclosure.
  • the method is executed by a third UE.
  • the method for invoking a computing power service may include the following steps:
  • Step 801 Based on the indication information, execute the computing service to obtain the computing result.
  • the third UE performs computing services such as data cleaning, model training, etc. based on the indication information to obtain computing results.
  • Different third UEs can perform different computing tasks or the same computing tasks, which depends on the splitting and deployment of the computing tasks requested by the first UE by the computing service module.
  • Step 802 Send the calculation result to the calculation service module.
  • the third UE sends the above calculation result to the calculation service module, and the calculation service module sends the result to the first UE.
  • the computing service module can directly feed back the computing results of each third UE to the first UE, or it can package and summarize the computing results of each third UE and then feed back to the first UE, or it can perform secondary aggregation processing on the computing results of each third UE and then feed back to the first UE, which is not limited by the present disclosure.
  • the method is executed by a third UE, including: executing the computing power service based on the indication information to obtain the calculation result, and sending the calculation result to the computing service module, thereby realizing the sending of the calculation result of the third UE to the computing service module.
  • FIG9 is a flow chart of a method for invoking a computing power service provided in an embodiment of the present disclosure. The method is executed by a first network element. As shown in FIG9 , the method for invoking a computing power service may include the following steps:
  • Step 901 Send computing power registration information of at least one second UE to a computing service module.
  • the computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that performs computing power services.
  • the first network element may be a computing power storage network element (see the "storage" network element in FIG. 1), which can store information about computing power nodes that provide computing power services.
  • the UE can collaborate with the core network to perform computing power deployment.
  • the first network element can store information about UEs that can provide computing power services, and can also store information about other computing power nodes, such as information about computing power nodes in the core network or information about third-party computing power nodes outside the core network.
  • the computing service module can obtain the first The information in the network element coordinates and deploys the computing tasks requested by the first UE.
  • the first network element may be a network element specifically used to store computing power node information, or it may be a general or shared storage network element in the core network, used to store computing power node information and other information, which is not limited in this disclosure.
  • the first network element may also be deployed outside the core network, and the information stored therein may be exchanged with the core network through the NEF network element, which is not limited in the present disclosure.
  • the first network element sends the stored computing power registration information of at least one second UE to the computing service module to help the computing service module select a suitable third UE for computing power service deployment for the service request based on the service perception results and the received computing power resource information of the second UE.
  • the method is executed by the first network element, including: sending computing power registration information of at least one second UE to the computing service module, so that the first network element sends the computing power information of the second UE to the computing power service module to help it complete service allocation.
  • FIG10 is a flow chart of a method for invoking a computing power service provided in an embodiment of the present disclosure. The method is executed by a first network element. As shown in FIG10 , the method for invoking a computing power service may include the following steps:
  • Step 1001 Receive a computing power registration information acquisition request sent by a computing service module.
  • the first network element receives a computing power registration information acquisition request sent by a computing service module to determine a third UE that can provide computing power services.
  • Step 1002 In response to a computing power registration information acquisition request, the computing power registration information of the UE in the idle state is determined as the computing power registration information of at least one second UE.
  • the computing power registration information includes at least the identifier, computing power type, and available computing power of each second UE.
  • the first network element determines the computing power registration information of a UE in an idle state as the computing power registration information of at least one second UE, and the total computing power that can be provided by the second UE meets the computing power required for the service request of the first UE, and can provide a second UE with a computing power close to the required computing power requirement.
  • Step 1003 Receive a status update request sent by the computing service module.
  • the first network element receives a status update request sent by the computing service module after issuing a computing task to the third UE, requesting that the called third UE be updated from an idle state to an occupied state, indicating that the third UE is providing computing services and may not be called to serve other computing services during this period.
  • the first network element receives a status update request sent by the computing service module after receiving the confirmation information sent by the first UE, requesting to update the called third UE from an occupied state to an idle state, indicating that the third UE ends the computing power service and can provide computing power support for other computing power services.
  • Step 1004 In response to the status update request, update the computing power status of at least one third UE, where the computing power status includes an idle state and an occupied state.
  • the first network element receives a status update request sent by the computing power service module, changes the called third UE from an idle state to an occupied state, or changes the third UE that has completed the computing power service from an occupied state to an idle state, and changes the third UE state in time to improve the resource utilization efficiency of the idle terminal device.
  • the method is executed by the first network element, including: receiving a computing power registration information acquisition request sent by the computing service module, and in response to the computing power registration information acquisition request, determining the computing power registration information of the UE in the idle state as the computing power registration information of at least one second UE, receiving a status update request sent by the computing service module, and in response to the status update request, updating the computing power status of at least one third UE, the computing power status including the idle state and the occupied state, and by timely changing the third UE status, the resource utilization efficiency of the idle terminal equipment is improved and the task pressure carried by the network is reduced.
  • FIG11 is a flow chart of an interactive method for invoking a computing service provided by an embodiment of the present disclosure, which can be executed by a communication system, and the communication system includes a first UE, a third UE, a computing service module, and a first network element.
  • the interactive method may include the following steps:
  • the first UE sends a service request to a computing service module.
  • the business request is used to request to call a computing power service.
  • the first network element sends computing power registration information of at least one second UE to the computing service module.
  • computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that performs computing power services.
  • the computing service module determines at least one third UE that performs computing service from at least one second UE.
  • the computing service module sends indication information for executing computing service to at least one third UE.
  • the first UE sends a service request to the computing service module, wherein the service request is used to request to call the computing power service;
  • the first network element sends the computing power registration information of at least one second UE to the computing service module, and the computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that executes the computing power service;
  • the computing service module determines at least one third UE that executes the computing power service from at least one second UE;
  • the computing service module sends indication information for executing the computing power service to at least one third UE, thereby realizing a complete computing power service workflow in the core network, so that the terminal equipment cooperates with the network to provide computing power services, reduces the task pressure carried by the network, and improves the resource utilization efficiency of idle terminal equipment.
  • the computing power service call workflow is shown in Figure 12.
  • the first UE sends a service request to the computing service module in the core network.
  • the request information includes: UE identifier, computing service identifier (CalService ID), computing service type (Cal Service Type), computing power requirement, network bandwidth, etc., as shown in step 1 in the figure.
  • the computing service module receives the business request information and analyzes and translates it to complete the business perception, as shown in step 2 of the figure.
  • the computing service module based on the service perception results, if the computing service module needs to call the UE-side computing power to coordinate the network to complete the service, the computing service module sends a request to obtain the second UE computing power registration information to the first network element Storage network element, and the Storage network element returns the result, as shown in step 3 of the figure.
  • the computing service module performs computing power service deployment according to the computing power registration information received from the second UE, as shown in step 4 of the figure.
  • the computing service module allocates computing tasks and sends data, task requirements and information of the allocated third UE to the access and mobility management function AMF network element, as shown in step 5 in the figure.
  • the computing power service may be deployed on multiple computing power nodes (for example, when the computing power service is executed collaboratively by UE in the present disclosure, the UE can serve as a computing power node), and the AMF sends the corresponding data and task requirements to different third UEs, as shown in step 6 in the figure.
  • the computing service module sends updated third UE status information to the Storage network element, and adjusts the third UE assigned with the task to an occupied state, as shown in step 7 in the figure.
  • each third UE uploads the calculation result to the calculation service module, as shown in step 8 of the figure.
  • the computing service module sends updated third UE status information to the Storage network element, and adjusts the third UE that completes the task to an idle state, as shown in step 9 in the figure.
  • the calculation service module After receiving the calculation results of all subtasks, the calculation service module performs aggregate calculation processing on the received calculation results, as shown in step 10 of the figure.
  • the computing service module returns the final computing result to the first UE requesting the service through the user plane, as shown in step 11 of the figure.
  • the methods provided by the embodiments of the present application are introduced from the perspectives of network equipment and user equipment, respectively.
  • the network equipment and the user equipment may include hardware structures and software modules, and the functions are implemented in the form of hardware structures, software modules, or hardware structures plus software modules.
  • a function of the functions may be executed in the form of hardware structures, software modules, or hardware structures plus software modules.
  • the present disclosure also provides a computing power service calling device. Since the computing power service calling device provided in the embodiment of the present disclosure corresponds to the computing power service calling method provided in the above-mentioned embodiments, the implementation method of the computing power service calling method is also applicable to the computing power service calling device provided in this embodiment and will not be described in detail in this embodiment.
  • FIG13 is a schematic diagram of the structure of a computing service module 1300 provided in an embodiment of the present disclosure.
  • the computing service module is deployed in a core network.
  • the device 1300 may include a transceiver unit 1301 for receiving a service request sent by a first user equipment UE, wherein the service request is used to request to call a computing power service; a determination unit 1302 for determining at least one third UE that executes the computing power service from at least one second UE based on the computing power registration information of at least one second UE.
  • the transceiver unit 1301 is further used to obtain computing power registration information of at least one second UE from the first network element.
  • the transceiver unit 1301 is also used to send indication information for executing computing power services to at least one third UE.
  • the determination unit 1302 can also be used to divide the computing service requested by the first UE into one or more sub-services according to different types of service requests, that is, to split the distributed computing service, send indication information to at least one determined third UE based on the divided sub-services, and allocate multiple sub-services to one or more third UEs for calculation, wherein the computing power that the third UE can provide is greater than and similar to the computing power required for the sub-service, and the total computing power capacity of multiple third UEs can be greater than or equal to the computing power requirement of the computing task requested by the first UE, so as to provide the first UE with reasonable computing power services for the service and avoid waste of computing power resources.
  • the computing service module is deployed in the core network; a service request sent by a first user equipment UE is received; computing power registration information of at least one second UE is obtained from the first network element; at least one third UE that performs computing power service is determined from at least one second UE; indication information for performing computing power service is sent to at least one third UE; a connection can be established between the third UE and the computing service module, and then the computing service module provides the computing power service requested by the first UE to the first UE by communicating with the third UE, thereby realizing the UE collaborative network providing computing power service.
  • the service request includes at least one of the following: an identifier of the first UE; an identifier of the computing power service; a type of computing power service; computing power requirements; and network bandwidth.
  • the transceiver unit 1301 is further used to send a computing power registration information acquisition request to the first network element, wherein the computing power registration information acquisition request is used to request computing power registration information of a UE in an idle state.
  • the transceiver unit 1301 is also used to receive computing power registration information of at least one second UE sent by the first network element, and the computing power registration information includes at least an identifier, computing power type, and available computing power size of each second UE.
  • the determination unit 1302 is also used to determine whether there is at least one third UE from at least one second UE based on the service request and the computing power registration information of at least one second UE, using a resource matching algorithm, wherein the total computing power resources of the at least one third UE meet the computing power requirements in the service request.
  • the determination unit 1302 is also used to determine that there is no at least one third UE in at least one second UE.
  • the transceiver unit 1301 is further used to send feedback information to the first UE, wherein the feedback information is used to notify the first UE of a call failure and/or a UE waiting to meet computing power requirements.
  • the determination unit 1302 is further used to determine that there is at least one third UE in at least one second UE, wherein the indication information includes data information and/or task requirements required to perform the computing service.
  • the transceiver unit 1301 is further used to receive calculation results fed back by at least one third UE, wherein the calculation results are results obtained by at least one third UE performing a computing service based on the indication information.
  • the apparatus further includes an aggregation unit 1303, which is used to: perform aggregation processing on the calculation results to obtain an aggregated result.
  • the transceiver unit 1301 is further configured to send the aggregation result to the first UE.
  • the transceiver unit 1301 is further configured to receive confirmation information sent by the first UE.
  • the transceiver unit 1301 is also used to send a status update request to the first network element, wherein the status update request is used to request the first network element to update the computing power status of at least one third UE, and the computing power status includes an idle state and an occupied state.
  • the transceiver unit 1301 is also used to send indication information to the second network element, wherein the indication information is used to instruct the second network element to send data information and/or task requirements required to perform computing services to at least one third UE.
  • FIG. 15 is a schematic diagram of the structure of a first user equipment UE1400 provided in an embodiment of the present disclosure.
  • the device 1400 may include a transceiver unit 1401 for sending a service request to a computing service module, wherein the service request is used to request to call a computing service.
  • the service request includes at least one of the following: an identifier of the first UE; an identifier of the computing power service; a type of the computing power service; computing power requirements; and network bandwidth.
  • the device is configured on the first UE: the first UE is used to send a business request to the computing service module, wherein the business request is used to request to call the computing power service, thereby connecting the first UE to the computing service module to call the computing power service.
  • the transceiver unit 1401 is also used to receive feedback information sent by the computing service module, wherein the feedback information is used to notify the UE of a call failure and/or the UE waiting to meet computing power requirements.
  • the transceiver unit 1401 is further used to receive an aggregated result sent by a computing service module, wherein the aggregated result is a result obtained by the computing service module performing an aggregation process on a computing result of at least one third UE.
  • the transceiver unit 1401 is further configured to send confirmation information to the computing service module.
  • FIG. 16 is a schematic structural diagram of a third user equipment UE1500 provided in an embodiment of the present disclosure.
  • the device 1500 may include a transceiver unit 1501 for receiving instruction information for executing computing power services sent by a computing service module.
  • the indication information includes data information and/or task requirements required to perform the computing service.
  • the device is configured on the third UE: used to receive the indication information for executing the computing power service sent by the computing service module, wherein the indication information includes the data information and/or task requirements required to execute the computing power service, so that the third UE executes the allocated computing power task according to the indication information.
  • the device also includes a processing unit 1502, and the processing unit 1502 is used to execute computing services and obtain computing results.
  • the transceiver unit 1501 is further configured to send calculation results to the calculation service module.
  • FIG18 is a schematic diagram of the structure of a first network element 1600 provided in an embodiment of the present disclosure.
  • the device 1600 may include a transceiver unit 1601, which is used to send computing power registration information of at least one second UE to the computing service module, wherein the computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that performs computing power services.
  • the device is configured in the first network element: used to send computing power registration information of at least one second UE to the computing service module, wherein the computing power registration information of at least one second UE is used to assist the computing service module in determining at least one third UE that performs the computing power service, thereby determining the third UE that provides the computing power service, and realizing the third UE cooperating with the network to provide computing power services.
  • the transceiver unit 1601 is also used to receive a computing power registration information acquisition request sent by the computing service module.
  • the device also includes a processing unit 1602, which is used to determine the computing power registration information of the UE in the idle state as the computing power registration information of at least one second UE, wherein the computing power registration information includes at least an identifier, computing power type, and available computing power size of each second UE.
  • the transceiver unit 1601 is further configured to receive a status update request sent by the computing service module.
  • the processing unit 1602 is further used to update the computing power status of at least one third UE, and the computing power status includes an idle state and an occupied state.
  • FIG 20 is a schematic diagram of the structure of a communication device 1700 provided in an embodiment of the present application.
  • the communication device 1700 can be a network device, or a terminal device, or a chip, a chip system, or a processor that supports the network device to implement the above method, or a chip, a chip system, or a processor that supports the terminal device to implement the above method.
  • the device can be used to implement the method described in the above method embodiment, and the details can be referred to the description in the above method embodiment.
  • the communication device 1700 may include one or more processors 1701.
  • the processor 1701 may be a general-purpose processor or a dedicated processor, etc. For example, it may be a baseband processor or a central processing unit.
  • the baseband processor may be used to process the communication protocol and the communication data
  • the central processing unit may be used to process the communication device (such as a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.). Control, execute computer programs, and process data of computer programs.
  • the communication device 1700 may further include one or more memories 1702, on which a computer program 1704 may be stored, and the processor 1701 executes the computer program 1704 so that the communication device 1700 performs the method described in the above method embodiment.
  • data may also be stored in the memory 1702.
  • the communication device 1700 and the memory 1702 may be provided separately or integrated together.
  • the communication device 1700 may further include a transceiver 1705 and an antenna 1706.
  • the transceiver 1705 may be referred to as a transceiver unit, a transceiver, or a transceiver circuit, etc., and is used to implement a transceiver function.
  • the transceiver 1705 may include a receiver and a transmitter, the receiver may be referred to as a receiver or a receiving circuit, etc., and is used to implement a receiving function; the transmitter may be referred to as a transmitter or a transmitting circuit, etc., and is used to implement a transmitting function.
  • the communication device 1700 may further include one or more interface circuits 1707.
  • the interface circuit 1707 is used to receive code instructions and transmit them to the processor 1701.
  • the processor 1701 executes the code instructions to enable the communication device 1700 to execute the method described in the above method embodiment.
  • the processor 1701 may include a transceiver for implementing the receiving and sending functions.
  • the transceiver may be a transceiver circuit, an interface, or an interface circuit.
  • the transceiver circuit, interface, or interface circuit for implementing the receiving and sending functions may be separate or integrated.
  • the above-mentioned transceiver circuit, interface, or interface circuit may be used for reading and writing code/data, or the above-mentioned transceiver circuit, interface, or interface circuit may be used for transmitting or delivering signals.
  • the processor 1701 may store a computer program 1703, which runs on the processor 1701 and enables the communication device 1700 to perform the method described in the above method embodiment.
  • the computer program 1703 may be fixed in the processor 1701, in which case the processor 1701 may be implemented by hardware.
  • the communication device 1700 may include a circuit that can implement the functions of sending or receiving or communicating in the aforementioned method embodiments.
  • the processor and transceiver described in the present application can be implemented in an integrated circuit (IC), an analog IC, a radio frequency integrated circuit RFIC, a mixed signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc.
  • the processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (nMetal-oxide-semiconductor, NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (bipolar junction transistor, BJT), bipolar CMOS (Bi CMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.
  • CMOS complementary metal oxide semiconductor
  • N-type metal oxide semiconductor nMetal-oxide-semiconductor
  • PMOS bipolar junction transistor
  • BJT bipolar CMOS
  • SiGe silicon germanium
  • GaAs gallium arsenide
  • the communication device described in the above embodiments may be a network device or a terminal device, but the scope of the communication device described in the present application is not limited thereto, and the structure of the communication device may not be limited by FIG. 20.
  • the communication device may be an independent device or may be part of a larger device.
  • the communication device may be:
  • the IC set may also include a storage component for storing data and computer programs;
  • ASIC such as modem
  • the communication device can be a chip or a chip system
  • the communication device can be a chip or a chip system
  • the schematic diagram of the chip structure shown in Figure 21 includes a processor 1801 and an interface 1802.
  • the number of processors 1801 can be one or more, and the number of interfaces 1802 can be multiple.
  • the chip further includes a memory 1803, and the memory 1803 is used to store necessary computer programs and data.
  • the present application also provides a readable storage medium on which instructions are stored, and when the instructions are executed by a computer, any of the above methods is implemented.
  • Example function
  • the present application also provides a computer program product, which implements the functions of any of the above method embodiments when executed by a computer.
  • Figure 22 is a structural diagram of a communication system provided by an embodiment of the present disclosure.
  • the communication system includes: a computing service module for executing the methods shown in Figures 2 to 4 above; a first UE for executing the methods shown in Figures 5 and 6 above; a third UE for executing the methods shown in Figures 7 and 8 above; and a first network element for executing the methods shown in Figures 9 and 10 above.
  • the computer program product includes one or more computer programs.
  • the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
  • the computer program can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the computer program can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.
  • the computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media integrated. Available media can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., high-density digital video discs (DVD)), or semiconductor media (e.g., solid state disks (SSD)), etc.
  • At least one in the present application can also be described as one or more, and a plurality can be two, three, four or more, which is not limited in the present application.
  • the technical features in the technical feature are distinguished by “first”, “second”, “third”, “A”, “B”, “C” and “D”, etc., and there is no order of precedence or size between the technical features described by the "first”, “second”, “third”, “A”, “B”, “C” and “D”.
  • the corresponding relationships shown in each table in the present application can be configured or predefined.
  • the values of the information in each table are only examples and can be configured as other values, which are not limited by the present application.
  • the corresponding relationships shown in some rows may not be configured.
  • appropriate deformation adjustments can be made based on the above table, such as splitting, merging, etc.
  • the names of the parameters shown in the titles in the above tables can also use other names that can be understood by the communication device, and the values or representations of the parameters can also be other values or representations that can be understood by the communication device.
  • other data structures can also be used, such as arrays, queues, containers, stacks, linear lists, pointers, linked lists, trees, graphs, structures, classes, heaps, hash tables or hash tables.
  • the predefined in the present application may be understood as defined, predefined, stored, pre-stored, pre-negotiated, pre-configured, solidified, or pre-burned.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本公开提出一种算力服务调用方法、装置、及存储介质，属于通信技术领域。该方法包括：接收第一用户设备UE发送的业务请求，其中，业务请求用于请求调用算力服务；向至少一个第三UE发送执行算力服务的指示信息，其中，至少一个第三UE为满足算力服务对应需求的UE；第一UE向计算服务模块发送业务请求；第三UE接收计算服务模块发送的执行算力服务的指示信息；第一网元向计算服务模块发送至少一个第二UE的算力登记信息，从而实现UE协同网络进行分布式计算，提供算力服务。

Description

算力服务调用方法、装置及存储介质 技术领域
本公开涉及通信技术领域，尤其涉及算力服务调用方法、装置及存储介质。
背景技术
在未来网络系统中，用户设备UE有望能够向网络提供计算能力，终端设备将通过基于业务的接口实现灵活、直接的通信，海量终端设备将自身的算力能力登记存储于网络，协同网络进行分布式计算，减轻了网络承载的任务压力，提高了空闲终端设备的资源利用效率。相关技术中对于用户设备如何协同网络提供算力服务尚不明确。
发明内容
本公开提出的算力服务调用方法、装置及存储介质，用于终端设备协同网络提供算力服务。
第一方面，本公开实施例提供一种算力服务调用方法，该方法由网络设备执行，包括：接收第一用户设备UE发送的业务请求，其中，业务请求用于请求调用算力服务；向至少一个第三UE发送执行算力服务的指示信息，其中，所述至少一个第三UE为满足所述算力服务对应需求的UE。
第二方面，本公开实施例提供一种算力服务调用方法，该方法由第一UE执行，方法包括：向计算服务模块发送业务请求，其中，业务请求用于请求调用算力服务。
第三方面，本公开实施例提供一种算力服务调用方法，该方法由第三UE执行，方法包括：接收计算服务模块发送的执行算力服务的指示信息。
第四方面，本公开实施例提供一种算力服务调用方法，方法由第一网元执行，方法包括：向计算服务模块发送至少一个第二UE的算力登记信息，其中，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
第五方面，本公开实施例提供一种计算服务模块，计算服务模块部署于核心网中，包括：
收发单元，用于接收第一用户设备UE发送的业务请求，其中，业务请求用于请求调用算力服务；
收发单元还用于：向至少一个第三UE发送执行算力服务的指示信息，其中，至少一个第三UE为满足算力服务对应需求的UE。
第六方面，本公开实施例提供一种第一用户设备UE，包括：
收发单元，用于向计算服务模块发送业务请求，其中，业务请求用于请求调用算力服务。
第七方面，本公开实施例提供一种第三用户设备UE，包括：
收发单元，用于接收计算服务模块发送的执行算力服务的指示信息。
第八方面，本公开实施例提供一种第一网元，包括：
收发单元，用于向计算服务模块发送至少一个第二UE的算力登记信息，
其中，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
第九方面，本公开实施例提供一种通信装置，装置包括处理器和存储器，其中，存储器中存储有计算机程序，处理器执行存储器中存储的计算机程序，以使装置执行上述第一方面至第四方面任一方面的方法。
第十方面，本公开实施例提供一种通信装置，包括：处理器和接口电路，其中接口电路，用于接收代码指令并传输至处理器；处理器，用于运行代码指令以执行上述第一方面至第四方面任一方面的方法。
第十一方面，本发明实施例提供一种计算机可读存储介质，用于存储有指令，当指令被执行时，使上述第一方面至第四方面的任一方面的方法被实现。
第十二方面，本公开还提供一种包括计算机程序的计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面至第四方面的任一方面的方法。
第十三方面，本公开提供一种通信系统，系统包括计算服务模块、第一用户设备UE、第三UE、 第一网元，其中，
计算服务模块用于执行第一方面的方法；
第一UE用于执行第二方面的方法；
第三UE用于执行第三方面的方法；
第一网元用于执行第四方面的方法。
综上所述，根据本公开提供的算力服务调用方法，该方法由计算服务模块执行，计算服务模块部署于核心网中，方法包括：接收第一用户设备UE发送的业务请求，其中，业务请求用于请求调用算力服务；向至少一个第三UE发送执行算力服务的指示信息，其中，至少一个第三UE为满足算力服务对应需求的UE。本公开的方案实现了UE协同网络提供算力服务，通过将UE的算力能力登记存储于网络，协同网络进行分布式计算，减轻了网络承载的任务压力，实现了动态、灵活的算力调度和部署，提高了空闲终端设备的资源利用效率。
附图说明
本公开上述的和/或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和容易理解，其中：
图1为本公开实施例提供的一种通信系统的架构示意图；
图2为本公开的一个实施例所提供的一种算力服务调用方法的流程示意图；
图3为本公开另一个实施例所提供的一种算力服务调用方法的流程示意图；
图4为本公开再一个实施例所提供的一种算力服务调用方法的流程示意图；
图5为本公开又一个实施例所提供的一种算力服务调用方法的流程示意图；
图6为本公开另一个实施例所提供的一种算力服务调用方法的流程示意图；
图7为本公开再一个实施例所提供的一种算力服务调用方法的流程示意图；
图8为本公开又一个实施例所提供的一种算力服务调用方法的流程示意图；
图9为本公开另一个实施例所提供的一种算力服务调用方法的流程示意图；
图10为本公开再一个实施例所提供的一种算力服务调用方法的流程示意图；
图11为本公开一个实施例提供的一种算力服务调用的交互方法的流程图；
图12为本公开一个实施例所提供的一种算力服务调用的交互方法的示例图；
图13为本公开一个实施例所提供的一种计算服务模块的结构示意图；
图14为本公开一个实施例所提供的一种计算服务模块的结构示意图；
图15为本公开一个实施例所提供的一种第一UE的结构示意图；
图16为本公开一个实施例所提供的一种第三UE的结构示意图；
图17为本公开一个实施例所提供的一种第三UE的结构示意图；
图18为本公开一个实施例所提供的一种第一网元的结构示意图；
图19为本公开一个实施例所提供的一种第一网元的结构示意图；
图20为本公开一个实施例所提供的一种通信装置的结构示意图；
图21为本公开一个实施例所提供的一种通信装置为芯片或芯片系统的结构示意图；
图22为本公开一个实施例所提供的一种通信系统的结构示意图。
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实施例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、本公开实施例的一些方面相一致的装置和方法的例子。
在本公开实施例使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本公开实施例。在本公开实施例和所附权利要求书中所使用的单数形式的“一种”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。还应当理解，本文中使用的术语“和/或”是指并包含一个或多个相关联的列出项目的任何或所有可能组合。
应当理解，尽管在本公开实施例可能采用术语第一、第二、第三等来描述各种信息，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。例如，在不脱离本公开实施例范围的情况下，第一信息也可以被称为第二信息，类似地，第二信息也可以被称为第一信息。取决于语境，如在此所使用的词语“如果”及“若”可以被解释成为“在……时”或“当……时”或“响应于确定”。
下面详细描述本公开的实施例，实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的要素。下面通过参考附图描述的实施例是示例性的，旨在用于解释本公开，而不能理解为对本公开的限制。
为了便于理解，首先介绍本申请涉及的术语。
1、接入和移动性管理功能(Access and Mobility management Function，AMF)网元
支持具有不同移动性管理需求的终端设备。可以执行以下主要任务：非接入层(NAS)信令终端；NAS信令安全；接入层安全控制；用于3GPP接入网之间移动性的核心网节点间信令；空闲模式终端设备可达性(包括寻呼重传的控制和执行)；注册区管理；支持系统内和系统间的移动性；接入认证；访问授权，包括漫游权限检查；移动性管理控制(订阅和策略)；支持网络切片；会话管理功能(Ses sion Management Function，SMF)选择。
2、会话管理功能(Session Management Function，SMF)网元
可以与AMF一起支持定制的移动性管理方案，如“仅限移动启动的连接”(Mobile Initiated Con nection Only，MICO)或RAN增强功能，如“RRC Inactive”状态。SMF网元可以执行以下主要任务：会话管理；终端设备IP地址分配与管理；用户面功能(User Plane Function，UPF)的选择与控制；在UPF配置流量控制，将流量路由到适当的目的地；策略执行和服务质量(Quality of Service，QoS)控制部分；下行数据通知。
3、统一数据管理功能(Unified Data Management，UDM)网元
UDM，用于用户标识、签约数据、鉴权数据的管理、以及用户的服务网元注册管理。
本公开实施例中涉及到的各种网元/功能，其既可以是一个独立的硬件设备，也可以是在硬件设备内的通过计算机代码实现的功能，本公开实施例中并不对此做出限定。
4、网络开放功能(Network Exposure Function，NEF)网元
NEF是5GC对外开放和提供标准接口的网络功能。基于3GPP网络功能通过NEF向其他系统公开功能和事件，既提供开放性，又能保障系统的安全。NEF的存在，将5GC的功能实现标准化的展现，为第三方接入提供便利。
5、应用功能(Application Function，AF)网元
AF类似于应用服务器，其与其他5G核心网控制面NF交互，并提供业务服务。AF可以针对不同的应用服务而存在，可以由运营商或可信的第三方拥有。
6、鉴权服务功能(Authentication Server Function，AUSF)网元
AUSF是5G中支持统一鉴权服务功能的网元，实现3GPP和非3GPP的接入认证，是5G核心网络(5GC)中的网络实体。
7、策略控制功能(Policy Control Fuction，PCF)网元
PCF支持统一的策略框架去管理网络行为，提供策略规则给网络实体去实施执行，访问统一数据仓库(UDR)的订阅信息。
8、统一数据存储(Unified Data Repository，UDR)网元
UDR用于UDM存储订阅数据或读取订阅数据以及PCF存储策略数据或者读取策略数据。
9、网络存储功能(Network Repository Function，NRF)网元
NRF支持服务发现功能，从NF实例接收NF发现请求，并将发现的NF实例(被发现)的信息提供给NF实例，同时可以维护可用NF实例及其支持的服务的NF配置文件。
10、用户面功能(User Plane Function，UPF)网元。
UPF作为移动基础设施(Mobile Infrastructure，例如：RAN)和数据网络DN(Data Network)之 间的互连点，完成UP上GTP-U(GRPS隧道协议)协议的封装和解封装。通常，UPF用于无线接入技术(Radio Access Technologies，RAT)内部、或RAT之间移动性的协议数据单元(Protocol Data Unit，PDU)会话锚点(Session Anchor Point)，包括向gNB(NG-RAN节点)发送一个或多个结束标记数据包(End Marker Packets，EMP)。
为了更好的理解本公开实施例公开的方案，下面首先对本公开实施例适用的通信系统进行描述。
6G(第六代移动通信系统)服务化架构是将网元的功能抽象成多个服务。其中，相对于传统网络架构而言，6G服务化架构中引入了通感融合技术，以为终端设备提供多种服务(如感知服务、算力服务、人工智能(Artificial Intelligence，AI)服务等)。其中，6G服务化架构中包括用于提供服务的网元，6G服务化架构中的其他网元通过与该提供服务的网元之间的协同与共享，实现多维感知、协作通信、智能计算功能的深度融合、互惠增强，进而使网络具备新型信息流智能交互与处理及广域智能协作的能力。
请参见图1，图1为本公开实施例提供的一种6G服务化网络架构的示意图。图1中包括6G服务化网络架构中的网元以及各个网元之间用于通信的接口。参考图1，虚线框用于表示提供服务的网元，如图1所示，该网元可以实现感知(Sensing)、AI计算(Calculation)等能力，并且还具备存储(Stora ge)能力。
以及，图1所示的无线接入网(Radio Access Network，RAN)可以是网络侧的一种用于发射或接收信号的实体。例如，RAN可以为演进型基站(evolved NodeB，eNB)、发送接收点(transmission re ception point，TRP)、NR系统中的下一代基站(next generation NodeB，gNB)、其他未来移动通信系统中的基站或无线保真(wireless fidelity，WiFi)系统中的接入节点等。本公开的实施例对网络设备所采用的具体技术和具体设备形态不做限定。本公开实施例提供的网络设备可以是由集中单元(central u nit，CU)与分布式单元(distributed unit，DU)组成的，其中，CU也可以称为控制单元(control uni t)，采用CU-DU的结构可以将网络设备，例如基站的协议层拆分开，部分协议层的功能放在CU集中控制，剩下部分或全部协议层的功能分布在DU中，由CU集中控制DU。
图1所示的用户设备(User Equipment，UE)(或也可称为终端设备)可以是用户侧的一种用于接收或发射信号的实体，如手机。终端设备也可以称为终端(terminal)、移动台(mobile station，MS)、移动终端设备(mobile terminal，MT)等。终端设备可以是具备通信功能的汽车、智能汽车、手机(m obile phone)、穿戴式设备、平板电脑(Pad)、带无线收发功能的电脑、虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、工业控制(industrial control)中的无线终端设备、无人驾驶(self-driving)中的无线终端设备、远程手术(remote medical surgery)中的无线终端设备、智能电网(smart grid)中的无线终端设备、运输安全(transportation safety)中的无线终端设备、智慧城市(smart city)中的无线终端设备、智慧家庭(smart home)中的无线终端设备等等。本公开的实施例对UE所采用的具体技术和具体设备形态不做限定。
以及，图1中还包括有AMF、SMF、UDM等网元。其中，基于AMF网元和SMF网元的协作调度，可以使得终端设备(即图1中的UE)通过RAN与提供服务的网元建立用户面连接，以使得该提供服务的网元能够向终端设备提供该终端设备所请求调用的服务。关于该部分的具体流程可以参见后续实施例介绍。
可以理解的是，本公开实施例描述的通信系统是为了更加清楚的说明本公开实施例的技术方案，并不构成对于本公开实施例提供的技术方案的限定，本领域普通技术人员可知，随着系统架构的演变和新业务场景的出现，本公开实施例提供的技术方案对于类似的技术问题，同样适用。
下面参考附图对本公开实施例所提供的算力服务调用方法、装置、设备及存储介质进行详细描述。该方法可以应用于第五代移动通信技术(Fifth Generation，5G)及其后续通信技术，诸如第五代移动通信技术演进(5G-advanced)、第六代移动通信技术(Sixth Generation，6G)等，在本公开中不予限制。
图2为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由计算服务模块执行，计算服务模块部署于核心网中。在本公开中，计算服务模块可以以网元设备的形式部署于核心网中，以用于统筹调度和部署算力，其具体呈现形式在本公开中不予限制。
应理解，本公开所提供的算力服务调用方法可以建立在下述前提下：
1、UE(在本公开中称为第一UE)与核心网之间已建立用户面连接；
2、可以为核心网提供算力支持的多个UE(在本公开中称为第二UE)已经向核心网中的算力存储网元(在本公开中称为第一网元)进行算力登记，经核心网中计算服务模块的部署，全部或部分第二U E(在本公开中称为第三UE)可以接收来自核心网分配的算力任务。
如图2所示，该算力服务调用方法可以包括以下步骤：
步骤201、接收第一用户设备UE发送的业务请求。
在本公开的一些实施例中，业务请求用于请求调用算力服务。
在本公开的一些实施例中，业务请求包括以下至少一项：第一UE的标识；算力服务的标识；算力服务的类型；算力需求；网络带宽。
在本公开的一些实施例中，第一UE可以在需要算力服务时发送业务请求。例如，第一UE需要进行图像处理，则可以触发向核心网请求调用图像处理的算力服务。
在本公开的一些可选实施例中，UE的标识可以是用户永久标识符(SUbscription Permanent Identi fier，SUPI)，或其他能够唯一标识UE的标识，在本公开中不予限制。算力服务的标识(CalService I D)可以是标记可被调用的算力服务的代号。算力服务的类型(Cal Service Type)可以是标记可被调用的算力服务的类型的代号或字段。算力需求为UE对当前待进行的计算任务的算力大小需求。网络带宽可以是下行网络带宽，以便于网络设备向UE返回符合带宽要求的计算结果的数据包。
在本公开中，计算服务模块可以通过用户面连接接收第一UE发送的业务请求。其中，第一UE与核心网之间的用户面连接可以是通过如下方式建立的：当SMF网元接收到服务调用请求时，说明终端设备当前请求调用服务，此时，SMF网元可以与第一网元(提供终端设备请求调用的服务的网元)建立连接，以便SMF网元后续通过与第一网元交互来使得第一网元和终端设备建立用户面连接，进而使得第一网元能够通过用户面与终端设备进行通信，来为终端设备提供该终端设备所请求调用的服务。应理解，本公开对用户面连接的建立方式不予限制，以其他可能方式建立的用户面连接均落入本公开范围之内。通过用户面连接发送业务请求以进行算力服务调用，无需再通过控制面连接的信令交互流程，能够提升通信效率。
在本公开中，计算服务模块也可以通过其他连接接收第一UE发送的业务请求，对此本公开不予限制。
步骤202、向至少一个第三UE发送执行算力服务的指示信息。
在本公开的一些实施例中，指示信息可以是用于指示第三UE执行算力服务的信息。
在本公开的一些实施例中，指示信息包括执行算力服务所需的数据信息和/或任务需求。
计算服务模块可以将第一UE请求的计算任务整体下发给一个第三UE执行计算服务。
在本公开的一些实施例中，根据业务请求的类型不同，计算服务模块也可将UE请求的该计算业务分割成一个或多个子业务，即，进行分布式计算业务的拆分，基于分割后的子业务向确定的至少一个第三UE发送指示信息，将多个子业务分别分配给一个或多个第三UE进行计算，其中第三UE所能提供的算力大于且相近于子业务所需算力，多个第三UE的总算力能力可以大于或等于第一UE所请求的计算任务的算力需求，以为第一UE提供对该业务的合理算力服务，避免算力资源浪费。
综上所述，根据本公开提供的算力服务调用方法，该方法由计算服务模块执行，计算服务模块部署于核心网中，方法包括：接收第一用户设备UE发送的业务请求，其中，业务请求用于请求调用算力服务；从第一网元获取至少一个第二UE的算力登记信息；基于至少一个第二UE的算力登记信息，从至少一个第二UE中确定执行算力服务的至少一个第三UE；向至少一个第三UE发送执行算力服务的指示信息。本公开的方案实现了UE协同网络提供算力服务，通过将UE的算力能力登记存储于网络，协同网络进行分布式计算，减轻了网络承载的任务压力，实现了动态、灵活的算力调度和部署，提高了空闲终端设备的资源利用效率。
在一些可选实施例中，该方法还包括：从第一网元获取至少一个第二UE的算力登记信息。
在本公开的一些实施例中，该步骤可以在步骤201和步骤202之间执行。第一网元可以为算力存储网元(参见图1中的“storage”网元)，其中能够存储提供算力服务的算力节点的信息，在本公开中，UE可以协同核心网进行算力部署，第一网元中能够存储可以提供算力服务的UE的信息，还可以存储 其他算力节点的信息，例如核心网中的算力节点的信息或核心网外的第三方算力节点的信息，对此本公开不予限制。计算服务模块可以通过获取第一网元中的信息，统筹部署第一UE所请求执行的计算任务。
在一些可能的实施方式中，第一网元可以是专门用于存储算力节点信息的网元，也可以是核心网中的通用或共享存储网元，用于存储算力节点信息以及其他信息，对此本公开不予限制。
在一些可能的实施方式中，第一网元还可以部署于核心网之外，其存储的信息可以通过NEF网元与核心网进行数据交换，对此本公开不予限制。
在一些可选实施例中，第二UE可以为第一网元存储的、能够为核心网提供协同算力服务的全部或部分UE，其中，UE所能提供的算力大小、算力类型等算力信息登记在第一网元中，计算服务模块可以获取第一网元中存储的全部或部分第二UE的算力登记信息，以从中确定执行算力服务的一个或多个UE。
在本公开的一些实施例中，算力登记信息可以是UE在核心网(具体地，例如第一网元)中注册或登记的关于UE算力能力的信息。
在本公开的一些实施例中，算力登记信息至少包括每个第二UE的标识、算力类型、可用算力大小。
在本公开的一些实施例中，计算服务模块接收算力存储网元发送的可提供算力服务的一个或多个第二UE的算力登记信息，以确定第二UE中是否存在至少一个第三UE满足业务请求中的算力需求。
在一些可选实施例中，该方法还包括：基于至少一个第二UE的算力登记信息，从至少一个第二U E中确定执行算力服务的至少一个第三UE。
在本公开的一些实施例中，该步骤可以在步骤201和202之间执行。计算服务模块响应于接收到业务请求，可以对该业务请求进行业务感知，其中，业务感知为根据请求的类型不同，例如，语音处理或图像处理，对该请求所需算力大小、算力类型等需求进行的智能评估。
在本公开的一些实施例中，计算服务模块基于此业务感知以及登记在第一网元中的各UE的算力信息，从上述至少一个第二UE中确定执行算力服务的至少一个第三UE。
在本公开中计算服务模块可以使用预设算法来确定至少一个第三UE，该算法的类型在本公开中不予限制。
图3为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由计算服务模块执行，如图3所示，该算力服务调用方法可以包括以下步骤：
步骤301、接收第一用户设备UE发送的业务请求。
其中，业务请求用于请求调用算力服务。
在本公开中，步骤301的原理与图2所示实施例中的步骤201相同，可以参照图2相关描述，在此不再赘述。
步骤302、向第一网元发送获取请求。
其中，获取请求用于请求UE的算力登记信息。
在本公开的一些实施例中，获取请求中可以包括执行算力服务所需的算力大小，以请求第一网元查找能够提供该算力大小的UE的算力登记信息；或者，获取请求中可以包括算力类型，以请求第一网元查找能够提供该类型服务的UE的算力登记信息；或者，获取请求中可以包括UE的状态，以请求第一网元查找处于特定状态的UE(例如，处于空闲态的UE)的算力登记信息。
在本公开的一些实施例中，算力存储网元中存储有一个或多个UE(即上述第二UE)的算力登记信息，各UE的算力资源可以相同或不同，即可提供的算力大小、算力类型等相同或不同。
在本公开的一些实施例中，第一网元中还可以存储各个UE的状态，例如空闲态或占用态。已经被调用的UE在第一网元中存储的状态为占用态，在其完成当前算力服务前可以不再被调用；尚未被调用的UE在算力存储网元中存储的状态为空闲态，可为上述业务提供算力服务。
在一些可能的实施方式中，第一网元中还可以存储各个UE的剩余算力。例如，某一UE总算力能力为X，该UE被调用执行算力服务A，执行算力服务A所需算力大小为Y，第一网元可以存储该UE的剩余算力X-Y，并当需要执行算力服务B且算力服务B所需算力大小小于X-Y时，计算服务模块可 以调用该UE执行算力服务B。
步骤303、接收第一网元发送的至少一个第二UE的算力登记信息。
在本公开的一些实施例中，在本公开的一些实施例中，算力登记信息可以是UE在核心网中(具体地，例如第一网元)注册或登记的关于UE算力能力的信息。
在本公开的一些实施例中，算力登记信息至少包括每个第二UE的标识、算力类型、可用算力大小。
在本公开的一些实施例中，计算服务模块接收算力存储网元发送的可提供算力服务的一个或多个第二UE的算力登记信息，以根据业务感知，确定第二UE中是否存在至少一个第三UE满足业务请求中的算力需求。
步骤304、从至少一个第二UE中确定是否存在至少一个第三UE。
在本公开的实施例中，至少一个第三UE的总算力资源满足业务请求中的算力需求。
在本公开的实施例中，计算服务模块可以基于业务请求和至少一个第二UE的算力登记信息，利用资源匹配算法，从第二UE中确定是否存在第三UE。例如，计算服务模块可以通过业务感知提取出业务请求中的相关信息，例如第一UE的标识、算力服务标识、算力服务类型、算力需求等等，确定第一UE请求执行的计算任务需要的算力大小、类型等信息，并通过从第一网元中获取的第二UE的算力登记信息，确定是否存在能够提供算力服务的第三UE，即，确定是否存在满足业务请求中算力需求的至少一个第三UE。
例如，若第一UE请求执行的计算任务的计算量较小，则其需要的算力也较小，此时单独的一个第三UE可以单独完成该算力服务。若第一UE请求执行的计算任务的计算量较大，则其需要的算力也较大，此时单独的一个第三UE可能不能完全完成该算力服务，则需要多个第三UE分布式的计算支持。计算服务模块通过特定算法衡量资源的利用率，从而确定一个或多个第三UE，以提供最优的算力部署决策。
步骤305、确定至少一个第二UE中不存在至少一个第三UE，向第一UE发送反馈信息。
其中，反馈信息用于通知第一UE调用失败。例如，反馈信息中可以携带调用失败字段，该字段表示当前没有可用的算力节点为其提供算力服务，此次调用失败，第一UE可以放弃该次算力服务的调用。
或者，反馈信息可以用于通知第一UE等待满足第一UE所需算力需求的UE。例如，反馈信息中可以携带等待字段，该字段表示当前没有可用的算力节点为其提供算力服务，第一UE可以等待可以为其提供算力服务的其他算力节点。
在本公开的实施例中，若计算服务模块确定第二UE中没有能够为第一UE提供算力服务的第三U E，即，第二UE对于经拆分的任一个子业务均不存在满足算力需求的第三UE，则计算服务模块无法为第一UE请求的计算任务调度第三UE，计算服务模块可以向第一UE发送反馈信息，以通知第一UE调用失败，则第一UE可以决定是否放弃当前计算任务，或等待，或更换其他计算任务。
在一些可选实施例中，计算服务模块也可以通过反馈信息直接通知第一UE等待，第一UE可以不放弃当前计算任务，而是等待能够为第一UE提供服务的、符合当前计算任务的算力需求的第三UE。
步骤306、确定至少一个第二UE中存在至少一个第三UE，向至少一个第三UE发送执行算力服务的指示信息。
其中，指示信息包括执行算力服务所需的数据信息和/或任务需求。
在本公开的实施例中，若计算服务模块确定第二UE中存在至少一个能够为第一UE提供算力服务的第三UE，则计算服务模块将经拆分的子业务分别分配给第三UE，即，分别向第三UE发送指示信息，以指示第三UE执行算力服务。
指示信息中携带有执行该算力服务所需的数据信息和/或任务需求。例如，第一UE请求的算力服务是对100张图片进行图像处理，计算服务模块可以将该任务进行拆分，例如拆分为5个子任务，每个子任务下对20张图像进行处理，并在第二UE中确定执行每个子业务的第三UE，例如5个，计算服务模块可以将20张图像以及进行图像处理的任务需求下发给每个第三UE，以使各个第三UE执行对所接收到的图像的计算处理任务。
计算服务模块确定第二UE中存在多个UE的满足计算任务的需求，称该多个UE为第四UE；则 计算服务模块可以进一步从多个第四UE中选择出一个或多个第三UE。其中，选择可以基于一个或多个维度，如利用率和/或计算效率和/或计算能力等。
在本公开的实施例中，可以向能够提供总体算力大小大于并最接近于第一UE请求算力服务所需算力的大小的一个或多个第三UE发送算力服务的指示信息，以避免终端设备资源的浪费。例如，第一U E请求的算力服务是对500张图像进行图像处理，此时第二UE中空闲态的第三UE有两个，一个可以对550张图像进行处理(利用率为5/5.5)，一个可以对600张图像进行处理(利用率为5/6)，此时可以向可以处理550张图像的第三UE发送算力服务的指示信息，以合理利用终端资源。
类似地，在本公开的实施例中，考虑到不同UE的计算效率和/或计算能力等，可以选择计算效率和/或计算能力高的UE作为第三UE。例如，第一UE请求的算力服务是对500张图像进行图像处理，此时第二UE中空闲态的第三UE有两个，一个UE的计算能力为1.5TFlops，另一个UE的计算能力为2TFlops，则选择计算能力为2TFlops的UE作为第三UE。例如，第一UE请求的算力服务是对500张图像进行图像处理，此时第二UE中空闲态的第三UE有两个，一个UE的计算效率为0.1568(单位计算能力/碳排放)，另一个UE的计算效率为0.4325(单位计算能力/碳排放)，则选择计算效率为0.4325的UE作为第三UE。
综上所述，根据本公开提供的算力服务调用方法，该方法由计算服务模块执行，包括：接收用户设备UE通过用户面连接发送的业务请求；向第一网元发送算力登记信息获取请求；接收第一网元发送的至少一个第二UE的算力登记信息；基于业务请求和至少一个第二UE的算力登记信息，利用资源匹配算法，从至少一个第二UE中确定是否存在至少一个第三UE，其中，至少一个第三UE的总算力资源满足业务请求中的算力需求；确定至少一个第二UE中不存在至少一个第三UE,向第一UE发送反馈信息；确定至少一个第二UE中存在至少一个第三UE,向至少一个第三UE发送执行算力服务的指示信息。本公开的方案实现了计算服务模块通过业务感知与算力资源信息相匹配，为业务请求分配所需第三UE，进而使第一UE向所分配的第三UE发送指示信息，以使计算服务模块高效、合理地调用第三UE，实现了灵活、准确的UE算力调度，提高了网络利用率和通信效率。
图4为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由计算服务模块执行，基于图3所示的实施例，该算力服务调用方法可以包括以下步骤：
步骤401、接收第一用户设备UE发送的业务请求。
其中，业务请求用于请求调用算力服务。
步骤402、从第一网元获取至少一个第二UE的算力登记信息。
步骤403、基于业务请求和至少一个第二UE的算力登记信息，利用资源匹配算法，从至少一个第二UE中确定是否存在至少一个第三UE，其中，至少一个第三UE的总算力资源满足业务请求中的算力需求。
在本公开中，步骤401-403的原理与上述图2-图3所示实施例中的相关步骤相同，可以参照上述实施例的相关描述，在此不再赘述。
步骤404、确定至少一个第二UE中存在至少一个第三UE，向至少一个第三UE发送执行算力服务的指示信息。
在本公开的实施例中，若计算服务模块确定第二UE中存在至少一个能够为第一UE提供算力服务的第三UE，则计算服务模块将经拆分的子业务分别分配给确定的第三UE，即，分别向确定的第三UE发送指示信息，以指示各个第三UE执行算力服务。
在一种可选实施例中，步骤404可以包括：向第二网元发送指示信息。
其中，指示信息用于指示第二网元向至少一个第三UE发送执行算力服务所需的数据信息和/或任务需求。
在本公开的一些实施例中，第二网元可以是AMF网元。具体地，算力服务可能会部署在多个算力节点(例如，在本公开中通过UE协同执行算力服务时，UE可以作为算力节点)，计算服务模块可以向AMF网元发送指示信息，指示信息中还可以包括第三UE的标识，AMF接收到指示信息并将该指示信 息发送至第三UE，以实现计算服务模块对第三UE的调用。
指示信息中携带有执行该算力服务所需的数据信息和/或任务需求。例如，第一UE请求的算力服务是对100张图片进行图像处理，计算服务模块可以将该任务进行拆分，例如拆分为5个子任务，每个子任务下对20张图像进行处理，并确定执行每个子业务的第三UE，例如5个，计算服务模块可以将20张图像以及进行图像处理的任务需求下发给每个第三UE，以使各个第三UE执行对所接收到的图像的计算处理任务。
步骤405、接收至少一个第三UE反馈的计算结果。
其中，计算结果为至少一个第三UE基于指示信息执行算力服务得到的结果。
在本公开的实施例中，第三UE依据接收到指示信息提供的数据信息和/或任务需求进行任务处理，计算服务模块可以接收第二UE中确定的一个或多个第三UE完成所分配的算力服务后得到的计算结果。
步骤406、对计算结果进行聚合处理，以得到聚合结果。
在本公开的一些实施例中，每个第三UE的任务分工可以不同，计算服务模块会将接收到的算力任务分成多个步骤，例如：数据清洗、模型训练等，在一种示例中，其中一个第三UE在完成数据清洗后，将该数据清洗后的数据返回至计算服务模块，计算服务模块再将该数据发送至另一个第三UE进行模型训练，待训练完成后将该数据返回至计算服务模块，计算服务模块可以对各第三UE的计算结果进行聚合处理，得到聚合结果。
在本公开的一些实施例中，各第三UE的任务分工也可以相同，各第三UE可以将计算结果返回计算服务模块进行汇总，完成对各第三UE计算结果的聚合处理，得到聚合结果。在该情形下，相较于各第三UE分工不同的情形，计算服务模块可以对各模块返回的计算结果进行汇总和打包，无需再对各计算结果进行二次处理。
步骤407、将聚合结果发送至第一UE。
在本公开的一些实施例中，计算服务模块可以将聚合结果通过用户面连接反馈给第一UE。第一U E通过用户面与核心网建立连接，当计算服务模块接收到聚合结果，说明计算服务模块完成算力调用服务，通过计算服务模块和终端设备建立的用户面连接，将上述聚合结果发送至第一UE。与通过控制面建立连接不同，通过用户面建立连接，可以实现直接从用户面与各提供算力的第三UE直接建立连接，同时可以实现数据或参数的上传，而控制面默认不能进行数据包发送。
应理解，上述步骤406-407可以是可选步骤。换言之，计算服务模块可以直接将各计算结果发送至第一UE，不进行聚合处理，对此本公开不予限制。
步骤408、接收第一UE发送的确认信息。
在本公开的一些实施例中，第一UE依据接收到的计算服务模块发送的聚合结果，向计算服务模块发送确认信息，计算服务模块接收该确认信息。在一些实施例中，若聚合结果未达到预设要求，第一U E可能再次请求调用算力服务或进行其他方式修改，本公开对此不予限制。
在一些实施例中，该方法还包括：向第一网元发送状态更新请求。
在本公开的实施例中，状态更新请求可以用于请求更新第一网元中所存储的UE的状态。其中，该状态可以是占用态或空闲态。
应理解，该步骤可以是可选步骤，并且本公开不限制其执行顺序。在一种可能的实施方式中，该步骤可以在步骤404之后、步骤405之前执行，即，当计算服务模块将计算任务下发至第三UE时，计算服务模块可以向第一网元发送状态更新请求，将该第三UE的状态由空闲态变更为占用态。在另一种可能的实施方式中，该步骤可以在步骤405之后执行，即，当计算服务模块接收到第三UE反馈的计算结果时，计算服务模块可以向第一网元发送状态更新请求，将该第三UE的状态由占用态变更为空闲态。在本公开中不予限制。
在本公开的一些实施例中，状态更新请求可以用于请求第三UE的剩余算力。其中，第一网元中还可以存储各个UE的剩余算力。例如，某一UE总算力能力为X，该UE被调用执行算力服务A，执行算力服务A所需算力大小为Y，第一网元可以存储该UE的剩余算力X-Y，并当需要执行算力服务B且算力服务B所需算力大小小于X-Y时，计算服务模块可以调用该UE执行算力服务B。
在本公开的一些实施例中，计算服务模块完成业务分割及分配，将子业务发送至第三UE，可以通过向第一网元发送状态更新请求来变更被调用的第三UE的算力状态，将被调用的第三UE由空闲态变更为占用态。
在本公开的一些实施例中，计算服务模块接收到第三UE反馈的计算结果，表明第三UE的算力服务结束，向第一网元发送状态更新请求，变更第三UE的算力状态，将已完成算力服务的第三UE由占用态变更为空闲态。
综上所述，根据本公开提供的算力服务调用方法，该方法由计算服务模块执行，包括：接收第一用户设备UE发送的业务请求，从第一网元获取至少一个第二UE的算力登记信息，基于业务请求和至少一个第二UE的算力登记信息，利用资源匹配算法，从至少一个第二UE中确定是否存在至少一个第三UE，其中，至少一个第三UE的总算力资源满足业务请求中的算力需求，确定至少一个第二UE中存在至少一个第三UE，向至少一个第三UE发送执行算力服务的指示信息，接收至少一个第三UE反馈的计算结果，对计算结果进行聚合处理，以得到聚合结果，将聚合结果发送至第一UE，接收第一UE发送的确认信息，向第一网元发送状态更新请求，向第二网元发送指示请求，从而实现了算力服务模块合理调用UE侧进行算力服务，有效减轻了网络承载的任务压力，提高了对算力任务的处理速度。
图5为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由第一UE执行，如图5所示，该算力服务调用方法可以包括以下步骤：
步骤501、向计算服务模块发送业务请求。
其中，业务请求用于请求调用算力服务。
在本公开的一些实施例中，业务请求包括以下至少一项：第一UE的标识；算力服务的标识；算力服务的类型；算力需求；网络带宽。
在本公开的一些可选实施例中，第一UE的标识可以是用户永久标识符(SUbscription Permanent I dentifier，SUPI)，或其他能够唯一标识UE的标识，在本公开中不予限制。算力服务的标识(Cal Serv ice ID)可以是标记可被调用的算力服务的代号。算力服务的类型(Cal Service Type)可以是标记可被调用的算力服务的类型的代号或字段。算力需求为UE对当前待进行的计算任务的算力大小需求。网络带宽可以是下行网络带宽，以便于网络设备向UE返回符合带宽要求的计算结果的数据包。
在本公开的一些实施例中，第一UE与核心网之间的用户面连接可以是通过如下方式建立的：当S MF网元接收到服务调用请求时，说明终端设备当前请求调用服务，此时，SMF网元可以与第一网元(提供终端设备请求调用的服务的网元)建立连接，以便SMF网元后续通过与第一网元交互来使得第一网元和终端设备建立用户面连接，进而使得第一网元能够通过用户面与终端设备进行通信，来为终端设备提供该终端设备所请求调用的服务。应理解，本公开对用户面连接的建立方式不予限制，以其他可能方式建立的用户面连接均落入本公开范围之内。
综上所述，根据本公开提供的算力服务调用方法，该方法由第一UE执行，包括：向计算服务模块发送业务请求，其中，业务请求用于请求调用算力服务。从而将用户设备第一UE同计算服务模块连接起来，为第一UE能够成功调用核心网算力服务提供了基础。
图6为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由第一UE执行，如图7所示，在执行上述步骤601之后，该算力服务调用方法可以包括以下步骤：
步骤601、接收计算服务模块发送的聚合结果。
其中，聚合结果为计算服务模块对至少一个第三UE的计算结果进行聚合处理得到的结果。
在本公开的一些实施例中，第一UE可以接收计算服务模块发送的聚合结果，以获取本次算力服务调用的结果。
在本公开的一些实施例中，第一UE可以接收计算服务模发送的不进行聚合处理的各计算结果，以获取本次算力服务调用的结果。
步骤602、向计算服务模块发送确认信息。
在本公开的一些实施例中，第一UE接收到计算服务模块的计算结果，发送确认信息，确认其收到该算力调用服务。
综上所述，根据本公开提供的算力服务调用方法，该方法由第一UE执行，包括：接收计算服务模块发送的聚合结果，其中，聚合结果为计算服务模块对至少一个第三UE的计算结果进行聚合处理得到的结果，向计算服务模块发送确认信息，使UE能及时、准确收到算力服务调用结果，并反馈收到的算力服务，以及时变更第三UE算力状态，提高了空闲终端设备的资源利用效率。
在本公开的一些实施例中，第一UE可以接收计算服务模块发送的反馈信息，该反馈信息用于通知第一UE调用失败和/或等待满足算力需求的UE。第一UE响应于该反馈信息，确定是否放弃当前计算任务，或等待，或更换其他计算任务，或者响应于该反馈信息，等待能够为第一UE提供服务的、符合当前计算任务的算力需求的第三UE。
图7为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由第三UE执行。如图8所示，该算力服务调用方法可以包括以下步骤：
步骤701、接收计算服务模块发送的执行算力服务的指示信息。
在本公开的一些实施例中，指示信息包括执行算力服务所需的数据信息和/或任务需求。
在本公开的一些实施例中，第三UE接收计算服务模块发送的执行算力服务的指示信息，通过计算服务模块的部署，各第三UE可以执行相应的计算任务，以提供算力服务。
综上所述，根据本公开提供的算力服务调用方法，该方法由第三UE执行，包括：接收计算服务模块发送的执行算力服务的指示信息。从而使第三UE依据指示信息执行所分配的算力任务。
图8为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由第三UE执行，如图8所示，在执行上述步骤701之后，该算力服务调用方法可以包括以下步骤：
步骤801、基于指示信息，执行算力服务，以得到计算结果。
在本公开的一些实施例中，第三UE基于指示信息，执行算力服务，例如数据清洗、模型训练等，得到计算结果。不同的第三UE可以执行不同的算力任务，也可以执行相同的算力任务，其取决于计算服务模块对第一UE所请求的计算任务的拆分和部署。
步骤802、向计算服务模块发送计算结果。
在本公开的一些实施例中，第三UE将上述计算结果发送至计算服务模块，通过计算服务模块将该结果发送至第一UE。
此外，计算服务模块可以将各第三UE的计算结果直接反馈至第一UE，也可以将各第三UE的计算结果进行打包汇总后再反馈至第一UE，还可以将各第三UE的计算结果进行二次聚合处理后在反馈至第一UE，对此本公开不予限制。
综上所述，根据本公开提供的算力服务调用方法，该方法由第三UE执行，包括：基于指示信息，执行算力服务，以得到计算结果，向计算服务模块发送计算结果，从而实现将第三UE的计算结果发送至计算服务模块。
图9为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由第一网元执行，如图9所示，该算力服务调用方法可以包括以下步骤：
步骤901、向计算服务模块发送至少一个第二UE的算力登记信息。
其中，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
在本公开的一些实施例中，第一网元可以为算力存储网元(参见图1中的“storage”网元)，其中能够存储提供算力服务的算力节点的信息，在本公开中，UE可以协同核心网进行算力部署，第一网元中能够存储可以提供算力服务的UE的信息，还可以存储其他算力节点的信息，例如核心网中的算力节点的信息或核心网外的第三方算力节点的信息，对此本公开不予限制。计算服务模块可以通过获取第一 网元中的信息，统筹部署第一UE所请求执行的计算任务。
在一些可能的实施方式中，第一网元可以是专门用于存储算力节点信息的网元，也可以是核心网中的通用或共享存储网元，用于存储算力节点信息以及其他信息，对此本公开不予限制。
在一些可能的实施方式中，第一网元还可以部署于核心网之外，其存储的信息可以通过NEF网元与核心网进行数据交换，对此本公开不予限制。
在本公开的一些实施例中，第一网元将存储的至少一个第二UE的算力登记信息发送至计算服务模块，帮助计算服务模块根据业务感知结果和接收到的第二UE的算力资源信息，为该服务请求选择合适的第三UE进行算力服务部署。
综上所述，根据本公开提供的算力服务调用方法，该方法由第一网元执行，包括：向计算服务模块发送至少一个第二UE的算力登记信息，从而实现第一网元向算力服务模块发送第二UE的算力信息，帮助其完成业务分配。
图10为本公开实施例所提供的一种算力服务调用方法的流程示意图，该方法由第一网元执行，如图10所示，该算力服务调用方法可以包括以下步骤：
步骤1001、接收计算服务模块发送的算力登记信息获取请求。
在本公开的一些实施例中，第一网元接收计算服模块发送的算力登记信息获取请求，以确定能提供算力服务的第三UE。
步骤1002、响应于算力登记信息获取请求，将处于空闲态的UE的算力登记信息确定为至少一个第二UE的算力登记信息。
其中，算力登记信息至少包括每个第二UE的标识、算力类型、可用算力大小。
在本公开的一些实例中，第一网元将处于空闲态的UE的算力登记信息确定为至少一个第二UE的算力登记信息，该第二UE所能提供的总算力满足第一UE业务请求所需算力，且可以提供算力大小接近于所需算力需求的第二UE。
步骤1003、接收计算服务模块发送的状态更新请求。
在本公开的一些实施例中，第一网元接收计算服务模块向第三UE下发算力任务后发送的状态更新请求，请求将调用的第三UE由空闲态更新为占用态，表明该第三UE在提供算力服务，此期间可以不被调用服务其他算力业务。
在本公开的一些实施例中，第一网元接收计算服务模块接收第一UE发送的确认信息后发送的状态更新请求，请求将调用的第三UE由占用态更新为空闲态，表明该第三UE结束算力服务，可以为其他算力业务提供算力支持。
步骤1004、响应于状态更新请求，更新至少一个第三UE的算力状态，算力状态包括空闲态、占用态。
在本公开的一些实施例中，第一网元接收算力服务模块发送的状态更新请求，将被调用的第三UE由空闲态变更为占用态或将完成算力服务的第三UE由占用态变更为空闲态，及时变更第三UE状态以提高空闲终端设备的资源利用效率。
综上所述，根据本公开提供的算力服务调用方法，该方法由第一网元执行，包括：接收计算服务模块发送的算力登记信息获取请求，响应于算力登记信息获取请求，将处于空闲态的UE的算力登记信息确定为至少一个第二UE的算力登记信息，接收计算服务模块发送的状态更新请求，响应于状态更新请求，更新至少一个第三UE的算力状态，算力状态包括空闲态、占用态，通过及时变更第三UE状态，提高空闲终端设备的资源利用效率，减轻网络承载的任务压力。
图11为本公开一个实施例提供的一种算力服务调用的交互方法的流程图，该方法可以由通信系统执行，通信系统中包括第一UE、第三UE、计算服务模块、第一网元。如图11所示，该交互方法可以包括以下步骤：
1101、第一UE向计算服务模块发送业务请求。
在本公开的一些实施例中，业务请求用于请求调用算力服务。
1102、第一网元向计算服务模块发送至少一个第二UE的算力登记信息。
在本公开的一些实施例中，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
1103、计算服务模块从至少一个第二UE中确定执行算力服务的至少一个第三UE。
1104、计算服务模块向至少一个第三UE发送执行算力服务的指示信息。
上述步骤1101-1104的原理与上述图2-10所示实施例各个步骤的原理相同，可以参照图2-10相关描述，在此不再赘述。
综上所述，根据本公开提供的算力服务调用方法，通过第一UE、第三UE、计算服务模块以及第一网元之间的交互，第一UE向计算服务模块发送业务请求，其中，业务请求用于请求调用算力服务；第一网元向计算服务模块发送至少一个第二UE的算力登记信息，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE；计算服务模块从至少一个第二UE中确定执行算力服务的至少一个第三UE；计算服务模块向至少一个第三UE发送执行算力服务的指示信息，从而实现了一种完整的核心网中的算力服务工作流程，以使得终端设备协同网络提供算力服务，减轻了网络承载的任务压力，提高了空闲终端设备的资源利用效率。
作为一种示例，算力服务调用工作流程具体如图12所示。
其中，第一UE向核心网内的计算服务模块发送业务请求，该请求信息包含：UE标识符、算力服务的标识(CalService ID)、算力服务的类型(Cal Service Type)、算力大小需求、网络带宽等，如图示步骤1所示。
其中，计算服务模块收到业务请求信息并进行分析转译，完成业务感知，如图示步骤2所示。
其中，计算服务模块根据业务感知结果，如果需要调用UE侧算力协同网络完成业务，计算服务模块向第一网元Storage网元发送第二UE算力登记信息获取请求，Storage网元返回结果，如图示步骤3所示。
其中，计算服务模块根据收到的第二UE的算力登记信息，进行算力服务部署，如图示步骤4所示。
其中，计算服务模块将算力任务进行分配，将数据、任务需求和分配的第三UE的信息发送至接入与移动性管理功能AMF网元，如图示步骤5所示。
其中，算力服务可能会部署在多个算力节点(例如，在本公开中通过UE协同执行算力服务时，UE可以作为算力节点)，AMF将对应的数据、任务需求下发至不同的第三UE，如图示步骤6所示。
其中，计算服务模块向Storage网元发送更新第三UE状态信息，将分配任务的第三UE调整为占用状态，如图示步骤7所示。
其中，各第三UE将计算结果上传至计算服务模块，如图示步骤8所示。
其中，计算服务模块向Storage网元发送更新第三UE状态信息，将完成任务的第三UE调整为空闲状态，如图示步骤9所示。
其中，计算服务模块收到所有子任务计算结果后，对收到的计算结果进行聚合计算处理，如图示步骤10所示。
其中，计算服务模块通过用户面将最终计算结果返回请求业务的第一UE，如图示步骤11所示。
上述本申请提供的实施例中，分别从网络设备、用户设备的角度对本申请实施例提供的方法进行了介绍。为了实现上述本申请实施例提供的方法中的各功能，网络设备和用户设备可以包括硬件结构、软件模块，以硬件结构、软件模块、或硬件结构加软件模块的形式来实现上述各功能。上述各功能中的某个功能可以以硬件结构、软件模块、或者硬件结构加软件模块的方式来执行。
与上述几种实施例提供的网络设备管理方法相对应，本公开还提供一种算力服务调用装置，由于本公开实施例提供的算力服务调用装置与上述几种实施例提供的算力服务调用方法相对应，因此算力服务调用方法的实施方式也适用于本实施例提供的算力服务调用装置，在本实施例中不再详细描述。
图13为本公开实施例提供的一种计算服务模块1300的结构示意图，计算服务模块部署于核心网中。
如图13所示，该装置1300可以包括收发单元1301，接收第一用户设备UE发送的业务请求，其中，业务请求用于请求调用算力服务；确定单元1302，用于基于至少一个第二UE的算力登记信息，从至少一个第二UE中确定执行算力服务的至少一个第三UE。
在一些实施例中，收发单元1301还用于从第一网元获取至少一个第二UE的算力登记信息。
在一些实施例中，收发单元1301还用于向至少一个第三UE发送执行算力服务的指示信息。
在一些实施例中，确定单元1302还可以用于根据业务请求的类型不同，将第一UE请求的该计算业务分割成一个或多个子业务，即，进行分布式计算业务的拆分，基于分割后的子业务向确定的至少一个第三UE发送指示信息，将多个子业务分别分配给一个或多个第三UE进行计算，其中第三UE所能提供的算力大于且相近于子业务所需算力，多个第三UE的总算力能力可以大于或等于第一UE所请求的计算任务的算力需求，以为第一UE提供对该业务的合理算力服务，避免算力资源浪费。
综上所述，根据本公开实施例提供的计算服务模块，计算服务模块部署于核心网中；接收第一用户设备UE发送的业务请求；从第一网元获取至少一个第二UE的算力登记信息；从至少一个第二UE中确定执行算力服务的至少一个第三UE；向至少一个第三UE发送执行算力服务的指示信息；可以使得第三UE与计算服务模块建立连接，进而使得计算服务模块通过与第三UE通信来为第一UE提供该第一UE所请求调用的算力服务，从而实现UE协同网络提供算力服务。
在本公开的一些实施例中，业务请求包括以下至少一项第一UE的标识；算力服务的标识；算力服务的类型；算力需求；网络带宽。
在本公开的一些实施例中，收发单元1301还用于向第一网元发送算力登记信息获取请求，其中，算力登记信息获取请求用于请求处于空闲态的UE的算力登记信息。
在本公开的一些实施例中，收发单元1301还用于接收第一网元发送的至少一个第二UE的算力登记信息，算力登记信息至少包括每个第二UE的标识、算力类型、可用算力大小。
在本公开的一些实施例中，确定单元1302还用于基于业务请求和至少一个第二UE的算力登记信息，利用资源匹配算法，从至少一个第二UE中确定是否存在至少一个第三UE，其中，至少一个第三UE的总算力资源满足业务请求中的算力需求。
在本公开的一些实施例中，确定单元1302还用于确定至少一个第二UE中不存在至少一个第三U E。
在本公开的一些实施例中，收发单元1301还用于向第一UE发送反馈信息，其中，反馈信息用于通知第一UE调用失败和/或等待满足算力需求的UE。
在本公开的一些实施例中，确定单元1302还用于确定至少一个第二UE中存在至少一个第三UE，其中，指示信息包括执行算力服务所需的数据信息和/或任务需求。
在本公开的一些实施例中，收发单元1301还用于接收至少一个第三UE反馈的计算结果，其中，计算结果为至少一个第三UE基于指示信息执行算力服务得到的结果。
在本公开的一些实施例中，装置还包括聚合单元1303，用于：对计算结果进行聚合处理，得到聚合结果。
在本公开的一些实施例中，收发单元1301还用于将聚合结果发送至第一UE。
在本公开的一些实施例中，收发单元1301还用于接收第一UE发送的确认信息。
在本公开的一些实施例中，收发单元1301还用于向第一网元发送状态更新请求，其中，状态更新请求用于请求第一网元更新至少一个第三UE的算力状态，算力状态包括空闲态、占用态。
在本公开的一些实施例中，收发单元1301还用于向第二网元发送指示信息，其中，指示信息用于指示第二网元向至少一个第三UE发送执行算力服务所需的数据信息和/或任务需求。
图15为本公开实施例提供的一种第一用户设备UE1400的结构示意图。
如图15所示，该装置1400可以包括收发单元1401，用于向计算服务模块发送业务请求，其中，业务请求用于请求调用算力服务。
在本公开的一些实施例中，业务请求包括以下至少一项：第一UE的标识；算力服务的标识；算力服务的类型；算力需求；网络带宽。
综上，根据本公开实施例提供的算力服务调用装置，装置配置于第一UE：第一UE用于向计算服务模块发送业务请求，其中，业务请求用于请求调用算力服务，从而使第一UE与计算服务模块连接，以调用算力服务。
在本公开的一些实施例中，收发单元1401还用于接收计算服务模块发送的反馈信息，其中，反馈信息用于通知UE调用失败和/或等待满足算力需求的UE。
在本公开的一些实施例中，收发单元1401还用于接收计算服务模块发送的聚合结果，其中，聚合结果为计算服务模块对至少一个第三UE的计算结果进行聚合处理得到的结果。
在本公开的一些实施例中，收发单元1401还用于向计算服务模块发送确认信息。
图16为本公开实施例提供的一种第三用户设备UE1500的结构示意图。
如图16所示，该装置1500可以包括收发单元1501，用于接收计算服务模块发送的执行算力服务的指示信息。
在本公开的一些实施例中，指示信息包括执行算力服务所需的数据信息和/或任务需求。
综上，根据本公开实施例提供的算力服务调用装置，装置配置于第三UE：用于接收计算服务模块发送的执行算力服务的指示信息，其中，指示信息包括执行算力服务所需的数据信息和/或任务需求，从而使第三UE依据指示信息执行所分配的算力任务。
在本公开的一些实施例中，装置还包括处理单元1502，处理单元1502用于执行算力服务，得到计算结果。
在本公开的一些实施例中，收发单元1501还用于向计算服务模块发送计算结果。
图18为本公开实施例提供的一种第一网元1600的结构示意图。
如图18所示，该装置1600可以包括收发单元1601，用于向计算服务模块发送至少一个第二UE的算力登记信息，其中，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
综上，根据本公开实施例提供的算力服务调用装置，装置配置于第一网元：用于向计算服务模块发送至少一个第二UE的算力登记信息，其中，至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE，从而确定提供算力服务的第三UE，实现第三UE协同网络提供算力服务。
在本公开的一些实施例中，收发单元1601还用于接收计算服务模块发送的算力登记信息获取请求。
在本公开的一些实施例中，装置还包括处理单元1602，处理单元1602用于将处于空闲态的UE的算力登记信息确定为至少一个第二UE的算力登记信息，其中，算力登记信息至少包括每个第二UE的标识、算力类型、可用算力大小。
在本公开的一些实施例中，收发单元1601还用于接收计算服务模块发送的状态更新请求。
在本公开的一些实施例中，处理单元1602还用于更新至少一个第三UE的算力状态，算力状态包括空闲态、占用态。
请参见图20，图20是本申请实施例提供的一种通信装置1700的结构示意图。通信装置1700可以是网络设备，也可以是终端设备，也可以是支持网络设备实现上述方法的芯片、芯片系统、或处理器等，还可以是支持终端设备实现上述方法的芯片、芯片系统、或处理器等。该装置可用于实现上述方法实施例中描述的方法，具体可以参见上述方法实施例中的说明。
通信装置1700可以包括一个或多个处理器1701。处理器1701可以是通用处理器或者专用处理器等。例如可以是基带处理器或中央处理器。基带处理器可以用于对通信协议以及通信数据进行处理，中央处理器可以用于对通信装置(如，基站、基带芯片，终端设备、终端设备芯片，DU或CU等)进行 控制，执行计算机程序，处理计算机程序的数据。
可选的，通信装置1700中还可以包括一个或多个存储器1702，其上可以存有计算机程序1704，处理器1701执行计算机程序1704，以使得通信装置1700执行上述方法实施例中描述的方法。可选的，存储器1702中还可以存储有数据。通信装置1700和存储器1702可以单独设置，也可以集成在一起。
可选的，通信装置1700还可以包括收发器1705、天线1706。收发器1705可以称为收发单元、收发机、或收发电路等，用于实现收发功能。收发器1705可以包括接收器和发送器，接收器可以称为接收机或接收电路等，用于实现接收功能；发送器可以称为发送机或发送电路等，用于实现发送功能。
可选的，通信装置1700中还可以包括一个或多个接口电路1707。接口电路1707用于接收代码指令并传输至处理器1701。处理器1701运行代码指令以使通信装置1700执行上述方法实施例中描述的方法。
在一种实现方式中，处理器1701中可以包括用于实现接收和发送功能的收发器。例如该收发器可以是收发电路，或者是接口，或者是接口电路。用于实现接收和发送功能的收发电路、接口或接口电路可以是分开的，也可以集成在一起。上述收发电路、接口或接口电路可以用于代码/数据的读写，或者，上述收发电路、接口或接口电路可以用于信号的传输或传递。
在一种实现方式中，处理器1701可以存有计算机程序1703，计算机程序1703在处理器1701上运行，可使得通信装置1700执行上述方法实施例中描述的方法。计算机程序1703可能固化在处理器1701中，该种情况下，处理器1701可能由硬件实现。
在一种实现方式中，通信装置1700可以包括电路，电路可以实现前述方法实施例中发送或接收或者通信的功能。本申请中描述的处理器和收发器可实现在集成电路(integrated circuit，IC)、模拟IC、射频集成电路RFIC、混合信号IC、专用集成电路(application specific integrated circuit，ASIC)、印刷电路板(printed circuit board，PCB)、电子设备等上。该处理器和收发器也可以用各种IC工艺技术来制造，例如互补金属氧化物半导体(complementary metal oxide semiconductor，CMOS)、N型金属氧化物半导体(nMetal-oxide-semiconductor，NMOS)、P型金属氧化物半导体(positive channel metal oxide semiconductor，PMOS)、双极结型晶体管(bipolar junction transistor，BJT)、双极CMOS(Bi CMOS)、硅锗(SiGe)、砷化镓(GaAs)等。
以上实施例描述中的通信装置可以是网络设备或者终端设备，但本申请中描述的通信装置的范围并不限于此，而且通信装置的结构可以不受图20的限制。通信装置可以是独立的设备或者可以是较大设备的一部分。例如通信装置可以是：
(1)独立的集成电路IC，或芯片，或，芯片系统或子系统；
(2)具有一个或多个IC的集合，可选的，该IC集合也可以包括用于存储数据，计算机程序的存储部件；
(3)ASIC，例如调制解调器(Modem)；
(4)可嵌入在其他设备内的模块；
(5)接收机、终端设备、智能终端设备、蜂窝电话、无线设备、手持机、移动单元、车载设备、网络设备、云设备、人工智能设备等等；
(6)其他等等。
对于通信装置可以是芯片或芯片系统的情况，可参见图21所示的芯片的结构示意图。图21所示的芯片包括处理器1801和接口1802。其中，处理器1801的数量可以是一个或多个，接口1802的数量可以是多个。
可选的，芯片还包括存储器1803，存储器1803用于存储必要的计算机程序和数据。
本领域技术人员还可以了解到本申请实施例列出的各种说明性逻辑块(illustrative logical block)和步骤(step)可以通过电子硬件、电脑软件，或两者的结合进行实现。这样的功能是通过硬件还是软件来实现取决于特定的应用和整个系统的设计要求。本领域技术人员可以对于每种特定的应用，可以使用各种方法实现的功能，但这种实现不应被理解为超出本申请实施例保护的范围。
本申请还提供一种可读存储介质，其上存储有指令，该指令被计算机执行时实现上述任一方法实施 例的功能。
本申请还提供一种计算机程序产品，该计算机程序产品被计算机执行时实现上述任一方法实施例的功能。
图22为本公开实施例所提供的一种通信系统的结构示意图，如图22所示，该通信系统包括：计算服务模块用于执行上述图2—图4所示方法；第一UE用于执行上述图5和6所示方法；第三UE用于执行上述图7和8所示方法；第一网元用于执行上述图9和10所示方法。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。计算机程序产品包括一个或多个计算机程序。在计算机上加载和执行计算机程序时，全部或部分地产生按照本申请实施例的流程或功能。计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。计算机程序可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，计算机程序可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital s ubscriber line，DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，高密度数字视频光盘(digital video disc，DVD))、或者半导体介质(例如，固态硬盘(solid state disk，SSD))等。
本领域普通技术人员可以理解：本申请中涉及的第一、第二等各种数字编号仅为描述方便进行的区分，并不用来限制本申请实施例的范围，也表示先后顺序。
本申请中的至少一个还可以描述为一个或多个，多个可以是两个、三个、四个或者更多个，本申请不做限制。在本申请实施例中，对于一种技术特征，通过“第一”、“第二”、“第三”、“A”、“B”、“C”和“D”等区分该种技术特征中的技术特征，该“第一”、“第二”、“第三”、“A”、“B”、“C”和“D”描述的技术特征间无先后顺序或者大小顺序。
本申请中各表所示的对应关系可以被配置，也可以是预定义的。各表中的信息的取值仅仅是举例，可以配置为其他值，本申请并不限定。在配置信息与各参数的对应关系时，并不一定要求必须配置各表中示意出的所有对应关系。例如，本申请中的表格中，某些行示出的对应关系也可以不配置。又例如，可以基于上述表格做适当的变形调整，例如，拆分，合并等等。上述各表中标题示出参数的名称也可以采用通信装置可理解的其他名称，其参数的取值或表示方式也可以通信装置可理解的其他取值或表示方式。上述各表在实现时，也可以采用其他的数据结构，例如可以采用数组、队列、容器、栈、线性表、指针、链表、树、图、结构体、类、堆、散列表或哈希表等。
本申请中的预定义可以理解为定义、预先定义、存储、预存储、预协商、预配置、固化、或预烧制。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (36)

  1. 一种算力服务调用方法，其特征在于，所述方法由网络设备执行，所述方法包括：
    接收第一用户设备UE发送的业务请求，其中，所述业务请求用于请求调用算力服务；
    向至少一个第三UE发送执行所述算力服务的指示信息，其中，所述至少一个第三UE为满足所述算力服务对应需求的UE。
  2. 根据权利要求1所述的方法，其特征在于，所述业务请求包括以下至少一项：
    所述第一UE的标识；
    所述算力服务的标识；
    所述算力服务的类型；
    算力需求；
    网络带宽。
  3. 根据权利要求1或2所述的方法，其特征在于，所述方法还包括：
    从第一网元获取至少一个第二UE的算力登记信息；
    基于所述至少一个第二UE的算力登记信息，从所述至少一个第二UE中确定执行所述至少一个第三UE。
  4. 根据权利要求3所述的方法，其特征在于，所述从第一网元获取至少一个第二UE的算力登记信息包括：
    向所述第一网元发送获取请求，其中，所述获取请求用于请求UE的算力登记信息；
    接收所述第一网元发送的所述至少一个第二UE的算力登记信息，所述算力登记信息至少包括每个第二UE的标识、算力类型或可用算力大小。
  5. 根据权利要求3或4所述的方法，其特征在于，所述至少一个第二UE为空闲态的UE。
  6. 根据权利要求3至5中任一项所述的方法，其特征在于，所述基于所述至少一个第二UE的算力登记信息，从所述至少一个第二UE中确定所述至少一个第三UE包括：
    基于所述业务请求和所述至少一个第二UE的算力登记信息，从所述至少一个第二UE中确定所述至少一个第三UE，
    其中，所述至少一个第三UE的总算力资源满足所述业务请求中的算力需求。
  7. 根据权利要求1至6中任一项所述的方法，其特征在于，所述方法包括：
    其中，所述指示信息包括执行所述算力服务所需的数据信息和/或任务需求。
  8. 根据权利要求1至7中任一项所述的方法，其特征在于，所述方法还包括：
    接收所述至少一个第三UE反馈的计算结果，
    其中，所述计算结果为所述至少一个第三UE基于所述指示信息执行所述算力服务得到的结果。
  9. 根据权利要求1至8中任一项所述的方法，其特征在于，所述方法还包括：
    对所述计算结果进行聚合处理，得到聚合结果；
    将所述聚合结果发送至所述第一UE；
    接收所述第一UE发送的确认信息。
  10. 根据权利要求1至9中任一项所述的方法，其特征在于，所述方法还包括：
    向所述第一网元发送状态更新请求，
    其中，所述状态更新请求用于请求所述第一网元更新所述至少一个第三UE的算力状态，所述算力状态包括空闲态、占用态。
  11. 根据权利要求1至9中任一项所述的方法，其特征在于，所述方法还包括：
    向所述第一网元发送状态更新请求，
    其中，所述状态更新请求用于请求所述至少一个第三UE的剩余算力。
  12. 根据权利要求1至11中任一项所述的方法，其特征在于，所述向所述至少一个第三UE发送 执行所述算力服务的指示信息包括：
    向第二网元发送所述指示信息，
    其中，所述指示信息用于指示向所述至少一个第三UE发送执行所述算力服务所需的数据信息和/或任务需求。
  13. 根据权利要求1至6中任一项所述的方法，其特征在于，所述方法还包括：
    基于所述业务请求和所述至少一个第二UE的算力登记信息确定所述至少一个第二UE中不存在所述至少一个第三UE；
    向所述第一UE发送反馈信息；
    其中，所述反馈信息用于通知所述第一UE调用失败和/或等待满足所述算力需求的UE。
  14. 一种算力服务调用方法，其特征在于，所述方法由网络设备执行，所述方法包括：
    接收第一用户设备UE发送的业务请求，其中，所述业务请求用于请求调用算力服务；
    基于所述业务请求和所述至少一个第二UE的算力登记信息，从所述至少一个第二UE中确定所述至少一个第三UE，
    确定是否存在至少一个第三UE，其中，所述至少一个第三UE为满足所述算力服务对应需求的UE。
  15. 根据权利要求14所述的方法，其特征在于，所述方法还包括：
    在确定不存在所述至少一个第三UE的情况下，向所述第一UE发送反馈信息；
    其中，所述反馈信息用于通知所述第一UE调用失败和/或等待满足所述算力需求的UE。
  16. 根据权利要求14所述的方法，其特征在于，所述方法还包括：
    在确定不存在所述至少一个第三UE的情况下，向所述至少一个第三UE发送执行所述算力服务的指示信息，
    其中，所述指示信息包括执行所述算力服务所需的数据信息和/或任务需求。
  17. 根据权利要求14至16中任一项所述的方法，其特征在于，所述确定是否存在至少一个第三UE包括：
    获取至少一个第二UE的算力登记信息；
    基于所述业务请求和所述至少一个第二UE的算力登记信息，从所述至少一个第二UE中确定是否存在所述至少一个第三UE。
  18. 一种算力服务调用方法，其特征在于，所述方法由第一UE执行，所述方法包括：
    向计算服务模块发送业务请求，其中，所述业务请求用于请求调用算力服务。
  19. 根据权利要求18所述的方法，其特征在于，所述业务请求包括以下至少一项：
    所述第一UE的标识；
    所述算力服务的标识；
    所述算力服务的类型；
    算力需求；
    网络带宽。
  20. 根据权利要求19或20所述的方法，其特征在于，所述方法还包括：
    接收所述计算服务模块发送的反馈信息，
    其中，所述反馈信息用于通知所述第一UE调用失败和/或等待满足所述算力需求的UE。
  21. 根据权利要求19或20所述的方法，其特征在于，所述方法还包括：
    接收所述计算服务模块发送的聚合结果，其中，所述聚合结果为所述计算服务模块对至少一个第三UE的计算结果进行聚合处理得到的结果；
    向所述计算服务模块发送确认信息。
  22. 一种算力服务调用方法，其特征在于，所述方法由第三UE执行，所述方法包括：
    接收计算服务模块发送的执行算力服务的指示信息。
  23. 根据权利要求22所述的方法，其特征在于，
    所述指示信息包括执行所述算力服务所需的数据信息和/或任务需求。
  24. 根据权利要求23所述的方法，其特征在于，所述方法还包括：
    基于所述指示信息，执行所述算力服务，以得到计算结果；
    向所述计算服务模块发送所述计算结果。
  25. 一种算力服务调用方法，其特征在于，所述方法由第一网元执行，所述方法包括：
    向计算服务模块发送至少一个第二UE的算力登记信息，
    其中，所述至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
  26. 根据权利要求25所述的方法，其特征在于，所述方法还包括：
    接收所述计算服务模块发送的获取请求；
    响应于所述获取请求，将处于空闲态的UE的算力登记信息确定为所述至少一个第二UE的算力登记信息，
    其中，所述算力登记信息至少包括每个第二UE的标识、算力类型、可用算力大小。
  27. 根据权利要求25或26所述的方法，其特征在于，所述方法包括：
    接收所述计算服务模块发送的状态更新请求；
    响应于所述状态更新请求，更新所述至少一个第三UE的算力状态，所述算力状态包括空闲态、占用态。
  28. 根据权利要求25或26所述的方法，其特征在于，所述方法包括：
    接收所述计算服务模块发送的状态更新请求；
    响应于所述状态更新请求，更新所述至少一个第三UE的剩余算力。
  29. 一种计算服务模块，其特征在于，所述计算服务模块部署于核心网中，包括：
    收发单元，用于接收第一用户设备UE发送的业务请求，其中，所述业务请求用于请求调用算力服务；
    向至少一个第三UE发送执行所述算力服务的指示信息，其中，所述至少一个第三UE为满足所述算力服务对应需求的UE。
  30. 一种第一用户设备UE，其特征在于，包括：
    收发单元，用于向计算服务模块发送业务请求，其中，所述业务请求用于请求调用算力服务。
  31. 一种第三用户设备UE，其特征在于，包括：
    收发单元，用于接收计算服务模块发送的执行算力服务的指示信息。
  32. 一种第一网元，其特征在于，包括：
    收发单元，用于向计算服务模块发送至少一个第二UE的算力登记信息，
    其中，所述至少一个第二UE的算力登记信息用于辅助计算服务模块确定执行算力服务的至少一个第三UE。
  33. 一种通信装置，其特征在于，所述装置包括处理器和存储器，其中，所述存储器中存储有计算机程序，所述处理器执行所述存储器中存储的计算机程序，以使所述装置执行：
    如权利要求1至13中任一项所述的方法；或者
    如权利要求14至17中任一项所述的方法；或者
    如权利要求18至21中任一项所述的方法；或者
    如权利要求22至24中任一项所述的方法；或者
    如权利要求25至28中任一项所述的方法。
  34. 一种通信装置，其特征在于，包括：处理器和接口电路，其中
    所述接口电路，用于接收代码指令并传输至所述处理器；
    所述处理器，用于运行所述代码指令以执行：
    如权利要求1至13中任一项所述的方法；或者
    如权利要求14至17中任一项所述的方法；或者
    如权利要求18至21中任一项所述的方法；或者
    如权利要求22至24中任一项所述的方法；或者
    如权利要求25至28中任一项所述的方法。
  35. 一种计算机可读存储介质，用于存储有指令，当所述指令被执行时，使以下方法被实现：
    如权利要求1至13中任一项所述的方法；或者
    如权利要求14至17中任一项所述的方法；或者
    如权利要求18至21中任一项所述的方法；或者
    如权利要求22至24中任一项所述的方法；或者
    如权利要求25至28中任一项所述的方法。
  36. 一种通信系统，其特征在于，所述系统包括计算服务模块、第一用户设备UE、第三UE、第一网元，其中，
    所述计算服务模块用于执行如权利要求1至13中任一项所述的方法或14至17中任一项所述的方法；
    所述第一UE用于执行如权利要求18至21中任一项所述的方法；
    所述第三UE用于执行如权利要求22至24中任一项所述的方法；
    所述第一网元用于执行如权利要求25至28中任一项所述的方法。
PCT/CN2023/082299 2023-03-17 2023-03-17 算力服务调用方法、装置及存储介质 WO2024192584A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2023/082299 WO2024192584A1 (zh) 2023-03-17 2023-03-17 算力服务调用方法、装置及存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2023/082299 WO2024192584A1 (zh) 2023-03-17 2023-03-17 算力服务调用方法、装置及存储介质

Publications (1)

Publication Number Publication Date
WO2024192584A1 true WO2024192584A1 (zh) 2024-09-26

Family

ID=92840735

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/082299 WO2024192584A1 (zh) 2023-03-17 2023-03-17 算力服务调用方法、装置及存储介质

Country Status (1)

Country Link
WO (1) WO2024192584A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112988377A (zh) * 2021-01-05 2021-06-18 腾讯科技（深圳）有限公司 用于云服务的资源分配方法、系统和介质
WO2021208914A1 (zh) * 2020-04-15 2021-10-21 展讯半导体(南京)有限公司 基于网络调度的算力共享方法及相关产品
WO2021208915A1 (zh) * 2020-04-13 2021-10-21 展讯半导体(南京)有限公司 算力共享方法及相关设备
WO2022143744A1 (zh) * 2020-12-31 2022-07-07 维沃移动通信有限公司 信息处理方法、装置、设备及存储介质
CN114756340A (zh) * 2022-03-17 2022-07-15 中国联合网络通信集团有限公司 算力调度系统、方法、装置和存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021208915A1 (zh) * 2020-04-13 2021-10-21 展讯半导体(南京)有限公司 算力共享方法及相关设备
WO2021208914A1 (zh) * 2020-04-15 2021-10-21 展讯半导体(南京)有限公司 基于网络调度的算力共享方法及相关产品
WO2022143744A1 (zh) * 2020-12-31 2022-07-07 维沃移动通信有限公司 信息处理方法、装置、设备及存储介质
CN112988377A (zh) * 2021-01-05 2021-06-18 腾讯科技（深圳）有限公司 用于云服务的资源分配方法、系统和介质
CN114756340A (zh) * 2022-03-17 2022-07-15 中国联合网络通信集团有限公司 算力调度系统、方法、装置和存储介质

Similar Documents

Publication Publication Date Title
KR20230008870A (ko) 통신 방법, 장치, 및 시스템
CN114079995A (zh) 一种中继管理方法及通信装置
WO2021134701A1 (zh) D2d通信方法、装置及系统
WO2021163901A1 (zh) 一种会话处理方法及其装置
WO2024092523A1 (zh) 侧行链路定位消息的发送方法、接收方法及其装置
WO2024092772A1 (zh) 侧行链路sl通信方法及装置
WO2024092577A1 (zh) 一种组定位方法、装置、设备及存储介质
WO2024065127A1 (zh) 控制中继设备信息发送的方法及其装置
WO2024192584A1 (zh) 算力服务调用方法、装置及存储介质
WO2024182951A1 (zh) 一种算力服务调用方法、装置及存储介质
WO2024197472A1 (zh) 一种算力登记与注销方法、装置及存储介质
EP4030689A1 (en) Data transmission method and apparatus, system, and storage medium
WO2024138338A1 (zh) 一种服务调用方法/装置/设备及存储介质
WO2023185496A1 (zh) 冗余传输请求方法及装置
WO2023212960A1 (zh) 扩展现实业务策略实现方法及装置
WO2024221358A1 (zh) 通信控制方法、装置及存储介质
WO2024012376A1 (zh) 一种通信方法、通信装置及通信系统
WO2023184191A1 (zh) 一种扩展现实多媒体xrm业务的处理方法及其装置
WO2024138564A1 (zh) 一种服务质量分配方法及其装置
WO2024197939A1 (zh) 定位方法、装置、设备及存储介质
WO2024148626A1 (zh) 一种终端标识确定方法/装置/设备及存储介质
WO2024207520A1 (zh) 网络选择方法和装置
WO2024148489A1 (zh) 一种个人物联网网络pin元素注册的方法及通信装置
WO2024065136A1 (zh) 一种控制方法及其装置
WO2024145902A1 (zh) 密钥获取方法、装置、设备及芯片系统