Communication method, communication apparatus and communication system

Abstract

A communication method, a communication apparatus and a communication system. On the basis of the method, the same computing node executes a first computing task and a second computing task, and if the difference between the times at which the two computing tasks occur is less than a preset time threshold value, different uplink transmission delays are configured for the two computing tasks, such that data of one computing task first reaches the computing node and the computing task is executed by the computing node, and after the execution of the computing task is completed or nearly completed, data of the other computing task then reaches the computing node, and the other computing task is executed by the computing node. Since the data of the two computing tasks does not reach the computing node at the same time or at similar times, the two computing tasks can be executed by the computing node, without it being necessary to wait for a long time after same reach the computing node, or, there is no need for the computing node to simultaneously process a plurality of computing tasks, thereby reducing the generation time of a first token of each computing task, and thus helping to ensure an end-to-end delay.

Description

A communication method, communication device and communication system

[0001] Cross-reference to related applications

[0002] This application claims priority to Chinese Patent Application No. 202411759226.X, filed on November 29, 2024, entitled "A Communication Method, Communication Device and Communication System", the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of communication technology, and in particular to a communication method, communication device and communication system. Background Technology

[0004] With the widespread deployment of networks, mobile internet has achieved tremendous success, gradually becoming one of the main ways for users to access the internet, leading to an explosive growth in mobile traffic. Traditional centralized anchor point deployment methods are increasingly unable to support this rapidly growing mobile traffic model. On the one hand, in networks with centralized anchor point gateway deployment, the increased traffic ultimately concentrates at the gateway and core equipment room, placing increasingly higher demands on backhaul network bandwidth, equipment room throughput, and gateway specifications. On the other hand, the long-distance backhaul network from the access network to the anchor point gateway and the complex transmission environment also result in significant latency and jitter in user packet transmission. The backhaul network mainly includes intermediate links from the core network (or backbone network) to the edge subnet.

[0005] Against this backdrop, the industry has proposed the concept of edge computing. Edge computing enables distributed local processing of business traffic by moving business processing capabilities to the network edge, avoiding excessive traffic concentration and thus significantly reducing the specification requirements for core data centers and centralized gateways. At the same time, edge computing also shortens the distance of the backhaul network, reduces end-to-end latency and jitter of user packets, and makes the deployment of ultra-low latency services possible.

[0006] On the other hand, with the rapid development of artificial intelligence (AI) services, generative AI services are becoming increasingly diverse and evolving into AI agents. Since AI services require significant computing power, power consumption, and storage resources, they pose new challenges to terminal devices. Due to resource limitations on the edge side, cloud-edge-device computing collaboration is a new trend that can alleviate the computing pressure on the edge side. Therefore, in one implementation, computing nodes with AI computing-related business processing capabilities can be deployed within the local data network. In this case, the computing node can be an application server, a functional unit deployed within the network, or it can be deployed within or near access network equipment. The computing node can provide computing power, offering computing resources to terminal devices or third parties.

[0007] A compute node can receive one or more compute tasks (or inference tasks) from the same or different terminal devices. These compute tasks may be related to the models deployed on the compute node. Once a compute node receives all or most of the data for a compute task, it can begin performing computations (or inferences) based on the received data and output the corresponding computation results (or inference results). However, when a compute node executes multiple compute tasks concurrently, it may increase the computation latency of each compute task, making it impossible to guarantee the end-to-end latency of each compute task. Summary of the Invention

[0008] This application provides a communication method, communication device, and communication system to ensure end-to-end latency between terminal devices and computing nodes.

[0009] In a first aspect, embodiments of this application provide a communication method that can be applied to the network side, such as a computing management function network element on the network side, a module (e.g., a circuit, chip, or chip system) in the computing management function network element, or a logical node, logical module, or software that can implement all or part of the functions of the computing management function network element. The method includes: determining that the difference between the occurrence time of the first computing task and the occurrence time of the second computing task is less than a preset time threshold, based on the occurrence time of the first computing task and the occurrence time of the second computing task; wherein the first computing task and the second computing task are executed by the same computing node; determining a first uplink transmission delay between the first computing task and the computing node based on a first end-to-end delay corresponding to the first computing task and a first computing delay of the first computing task on the computing node; and determining a second uplink transmission delay between the second computing task and the computing node based on a second end-to-end delay corresponding to the second computing task and a second computing delay of the second computing task on the computing node; wherein the first uplink transmission delay and the second uplink transmission delay are different; determining a QoS profile in the QoS file of the first quality of service flow (QoS flow) corresponding to the first computing task based on the first uplink transmission delay; and determining a QoS profile in the QoS file of the second quality of service flow corresponding to the second computing task based on the second uplink transmission delay; and sending the QoS profiles of the first QoS flow and the second QoS flow to the access network device.

[0010] Based on the above scheme, the same computing node executes the first and second computing tasks. If the occurrence times of the two computing tasks are relatively close, i.e., the difference between their occurrence times is less than a preset time threshold, then by configuring different uplink transmission delays for the two computing tasks, the data for one computing task arrives at the computing node first and is executed by the computing node. After the execution of the first computing task is completed or nearly completed, the other computing task arrives at the computing node and is executed by the computing node. Since the data for the two computing tasks do not arrive at the computing node at the same or similar times, neither computing task needs to wait for a long time before it can be executed by the computing node, or the computing node does not need to process multiple computing tasks simultaneously. This reduces the generation time of the first token for each computing task, improves the computing efficiency of the computing node, and thus helps to ensure end-to-end latency.

[0011] In one possible implementation, the quality of service (QoS) file for the first QoS stream sent to the access network device includes the first uplink transmission delay and the first downlink transmission delay; the QoS file for the second QoS stream sent to the access network device includes the second uplink transmission delay and the second downlink transmission delay. The calculation function network element determines the QoS file based on the determined uplink and downlink transmission delays.

[0012] In one possible implementation, the first quality of service (QoS) flow corresponding to the first computing task refers to the data packets of the first computing task being transmitted on the first QoS flow. The second QoS flow corresponding to the second computing task refers to the data packets of the second computing task being transmitted on the second QoS flow.

[0013] In one possible implementation, the first computational latency is the maximum or average computational latency of the first computational task on the computational node, and the second computational latency is the maximum or average computational latency of the second computational task on the computational node.

[0014] In one possible implementation, the first computational delay indicates the delay between the computing node receiving data and issuing a response.

[0015] In one possible implementation, the service quality file of the first service quality stream and the service quality file of the second service quality stream are sent to the access network device. The method further includes: sending the service quality file of the first service quality stream to the first access network device and sending the service quality file of the second service quality stream to the second access network device.

[0016] In one possible implementation, the method further includes: determining the occurrence time of the first computing task based on the period of the first computing task; and determining the occurrence time of the second computing task based on the period of the second computing task.

[0017] Based on the above scheme, determining the occurrence time of a computing task based on its cycle can accurately determine the occurrence time of the computing task, which helps to determine the uplink transmission delay of the computing task.

[0018] In one possible implementation, the method further includes: receiving a first policy and charging control (PCC) rule and a second policy and charging control rule from a policy control network element, wherein the first policy and charging control rule indicates the period of the first computing task and the second policy and charging control rule indicates the period of the second computing task; or, receiving a third policy and charging control rule from a policy control network element, wherein the third policy and charging control rule indicates the period of the first computing task and the period of the second computing task.

[0019] In one possible implementation, when the first calculation delay is less than the second calculation delay, the first uplink transmission delay is less than the second uplink transmission delay.

[0020] Based on the above scheme, an appropriate uplink transmission delay can be determined so that the data of the two computing tasks do not arrive at the computing node at the same or similar time. Therefore, after the two computing tasks arrive at the computing node, they can be executed by the computing node without waiting for a long time, which reduces the computing delay of each computing task and helps to ensure end-to-end latency.

[0021] In one possible implementation, the method further includes: determining a first downlink transmission delay of the first computing task between the first terminal device and the computing node based on the first end-to-end delay, the first computing delay, and the first uplink transmission delay; and determining a second downlink transmission delay of the second computing task between the second terminal device and the computing node based on the second end-to-end delay, the second computing delay, and the second uplink transmission delay.

[0022] In one possible implementation, the method further includes: determining that both the first computing task and the second computing task are processed by the computing node.

[0023] Secondly, embodiments of this application provide a communication method that can be applied to the network side, such as a network-side computing management function network element, a module (e.g., circuit, chip, or chip system) within the computing management function network element, or a logical node, logical module, or software capable of implementing all or part of the computing management function network element's functions. The method includes: determining multiple different uplink transmission delays between a terminal device and the computing node based on a first end-to-end delay corresponding to a first computing task and a first computing delay of the first computing task on the computing node; determining multiple quality of service (QoS) files for a first QoS stream corresponding to the first computing task; wherein each QoS file includes one of the multiple different uplink transmission delays, and the different QoS files include different uplink transmission delays; and sending first indication information and the multiple QoS files to an access network device, wherein the first indication information is used to instruct the access network device to select one of the multiple QoS files based on the waiting time for sending data from the first computing task.

[0024] Based on the above scheme, the computing management function network element determines multiple quality of service (QoS) files for the first QoS flow corresponding to the first computing task. Each QoS file includes one of multiple different uplink transmission delays, and the uplink transmission delays included in different QoS files are different. Then, the computing management function network element instructs the access network device to select one of the multiple QoS files based on the waiting time for sending the data of the first computing task, and uses the selected QoS file to control the uplink transmission delay when the first QoS flow sends the data of the first computing task. This ensures that the data of the first computing task can be processed immediately or as quickly as possible after arriving at the computing node, reducing queuing time and reducing the generation time of the first token of the first computing task, thereby ensuring the end-to-end latency of the first computing task.

[0025] In one possible implementation, the multiple quality of service files sent to the access network device for the first quality of service stream include the multiple first uplink transmission delays and the multiple first downlink transmission delays.

[0026] In one possible implementation, the first quality of service flow corresponding to the first computing task refers to the data packets of the first computing task being transmitted on the first quality of service flow.

[0027] In one possible implementation, the first computational latency is the maximum or average computational latency of the first computational task on the computational node.

[0028] In one possible implementation, the first computational delay indicates the delay between the computing node receiving data and issuing a response.

[0029] It is worth noting that the first quality of service flow corresponding to the first computing task refers to the data packets of the first computing task being transmitted on the first quality of service flow.

[0030] In one possible implementation, determining multiple different uplink transmission delays between the terminal device and the computing node based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node includes: determining that the first computing task is a non-periodic computing task, and then determining the multiple different uplink transmission delays based on the first end-to-end delay and the first computing delay.

[0031] Based on the above scheme, determining multiple different uplink transmission delays for non-periodic computing tasks, rather than determining multiple different uplink transmission delays for all computing tasks, can reduce resource overhead.

[0032] In one possible implementation, the method further includes: receiving a policy charging control rule from a policy control network element, the policy charging control rule being used to indicate that the first computing task is a non-periodic computing task.

[0033] In one possible implementation, the method further includes: determining multiple different downlink transmission delays of the first computing task between the terminal device and the computing node based on the first end-to-end delay, the first computing delay, and the multiple different uplink transmission delays.

[0034] Thirdly, embodiments of this application provide a communication method that can be applied to the network side, such as an access network device, a module (e.g., a circuit, chip, or chip system) within the access network device, or a logical node, logical module, or software capable of implementing all or part of the functions of the access network device. The method includes: receiving first indication information and multiple quality of service (QoS) files for a first computing task corresponding to a first QoS stream; wherein each QoS file includes one of multiple different uplink transmission delays, and the uplink transmission delays included in different QoS files are different; the first indication information is used to instruct the access network device to select one QoS file from the multiple QoS files based on the waiting time for sending data for the first computing task; selecting one QoS file from the multiple QoS files of the first QoS stream based on the waiting time for sending data for the first computing task; and scheduling the data for the first computing task based on the selected QoS file.

[0035] Based on the above scheme, the computing management function network element determines multiple quality of service (QoS) files for the first QoS stream corresponding to the first computing task. Each QoS file includes one of multiple different uplink transmission delays, and the uplink transmission delays included in different QoS files are different. Then, the computing management function network element instructs the access network device to select one of the multiple QoS files based on the waiting time for sending the data of the first computing task, and uses the selected QoS file to send the data of the first computing task. This ensures that the data of the first computing task can be processed immediately or as quickly as possible after arriving at the computing node, reducing queuing waiting time and reducing the generation time of the first token of the first computing task, thereby ensuring the end-to-end latency of the first computing task.

[0036] In one possible implementation, the method further includes: receiving second indication information from the computing node, the second indication information being used to indicate the waiting time.

[0037] Based on the above scheme, the computing node can indicate the waiting time to the access network device, which can accurately indicate the waiting time for sending the data of the first computing task.

[0038] In one possible implementation, the first computational latency is the maximum or average computational latency of the first computational task on the computational node.

[0039] In one possible implementation, the first computational delay indicates the delay between the computing node receiving data and issuing a response.

[0040] In one possible implementation, the waiting time indicates the waiting time by which the computing node can meet the first computational latency requirement. One possible explanation is that when a computing node processes multiple tasks, it needs to use resources such as memory and computing power simultaneously, which can lead to the computational latency failing to meet requirements. When the number of tasks processed by the computing node decreases, the load on the computing node decreases, and for a single task, the generation time of the first token is shortened, thereby reducing the real-time computational latency.

[0041] In one possible implementation, the waiting time indicates the computational latency required by one or more computational tasks being processed by the computing node. The computational latency can indicate one or more of the following: maximum computational latency, average computational latency, remaining computational latency, or real-time computational latency of the currently processed computing tasks. The computational latency can be one or more computational latencies required by one or more computing tasks, or it can be the computational latency required by the computing node to process one or more computing tasks.

[0042] In one possible implementation, the waiting time is used by the access network device to determine the waiting time of the first computing task at the computing node.

[0043] In one possible implementation, the second indication information includes a first parameter and a second parameter corresponding to the second computing task, the second computing task being the computing task being processed by the computing node; the first parameter is used to indicate the time interval between two adjacent tokens that generate the second computing task, and the second parameter is used to indicate the number of remaining tokens for the generated second computing task; the method further includes: determining the waiting time based on the first parameter and the second parameter.

[0044] Based on the above scheme, by using the first parameter and the second parameter to indicate the waiting time for sending the data of the first computing task, the waiting time for sending the data of the first computing task can be accurately indicated.

[0045] In one possible implementation, the method further includes: determining the waiting time based on the computation delay of the third computation task; wherein the third computation task is the computation task that most recently performed data transmission.

[0046] Based on the above scheme, the access network device determines the waiting time based on the computation delay of the third computing task, which can accurately indicate the waiting time for sending data of the first computing task.

[0047] In one possible implementation, the method further includes: receiving third indication information from a terminal device, the third indication information being used to indicate the data volume of the first computing task; and scheduling the data of the first computing task according to the third indication information.

[0048] In one possible implementation, the method further includes: receiving fourth indication information from a terminal device, the fourth indication information indicating either the buffer time for the data of the first computing task to arrive at the access layer of the terminal device or the remaining time for the data of the first computing task to arrive at the terminal device. The remaining time indicates that the data of the first computing task needs to arrive at the access network device within the remaining time.

[0049] Based on the above scheme, the access network device can determine the resources used to send the data of the first computing task according to the third indication information, and prepare the corresponding resources in advance to ensure that the data of the first computing task can be scheduled and transmitted as soon as possible, thereby ensuring the end-to-end latency of the first computing task.

[0050] Fourthly, this application provides a communication device that has the functions of the first aspect described above. For example, the communication device includes modules, units, or means for performing the operations involved in the first aspect. These modules, units, or means can be implemented by software, hardware, or a combination of software and hardware.

[0051] Fifthly, this application provides a communication device that has the functions of the second aspect above. For example, the communication device includes modules, units or means for performing the operations involved in the second aspect above. These modules, units or means can be implemented by software, hardware or a combination of software and hardware.

[0052] Sixthly, this application provides a communication device that has the functions of the third aspect above. For example, the communication device includes modules, units or means for performing the operations involved in the third aspect above. These modules, units or means can be implemented by software, hardware or a combination of software and hardware.

[0053] In a seventh aspect, this application provides a communication device including an interface circuit and one or more processors. The one or more processors are coupled to a memory. The memory stores part or all of the necessary computer programs or instructions for implementing the functions described in the first or second aspect. The one or more processors are executable to carry out the computer programs or instructions, which, when executed, cause the communication device to implement the methods in any possible design or implementation of the first or second aspect. The interface circuit is used to implement communication functions within the communication device and / or communication functions between the communication device and other devices or components.

[0054] The aforementioned communication device may be a computing management function network element, a module (e.g., a circuit, chip, or chip system) within a computing management function network element, or a logic node, logic module, or software capable of implementing all or part of the functions of a computing management function network element.

[0055] Eighthly, this application provides a communication device including an interface circuit and one or more processors. The one or more processors are coupled to a memory. The memory stores part or all of the necessary computer program or instructions for implementing the functions described in the third aspect above. The one or more processors are executable to carry out the computer program or instructions, causing the communication device to implement the methods in any possible design or implementation of the third aspect above. The interface circuit is used to implement the communication functions within the communication device and / or the communication functions between the communication device and other devices or components.

[0056] The aforementioned communication device may be an access network device, a module (e.g., a circuit, chip, or chip system) within the access network device, or a logical node, logical module, or software capable of implementing all or part of the functions of the access network device.

[0057] Ninthly, this application provides a computer-readable storage medium storing a computer program or instructions that, when executed, implement the method in any of the possible designs of the first to third aspects described above.

[0058] In a tenth aspect, this application provides a computer program product comprising a computer program or instructions that, when executed, implement the method in any of the possible designs of the first to third aspects described above.

[0059] Eleventhly, this application provides a communication system, including a computing management function network element and an access network device. The computing management function network element is used to execute the first aspect and any possible implementation method of the first aspect. The access network device is used to receive the quality of service (QoS) files of a first QoS stream and a second QoS stream.

[0060] In a twelfth aspect, this application provides a communication system including a computing management function network element and an access network device. The computing management function network element is used to execute the second aspect and any possible implementation of the second aspect. The access network device is used to execute the third aspect and any possible implementation of the third aspect. Attached Figure Description

[0061] Figure 1 is a schematic diagram of a 5G network architecture based on a service-oriented architecture;

[0062] Figure 2(a) is a flowchart illustrating a communication method provided in an embodiment of this application;

[0063] Figure 2(b) is a flowchart illustrating a communication method provided in an embodiment of this application;

[0064] Figure 3 is a flowchart illustrating a communication method provided in an embodiment of this application;

[0065] Figure 4 is a flowchart illustrating a communication method provided in an embodiment of this application;

[0066] Figure 5 is a flowchart illustrating a communication method provided in an embodiment of this application;

[0067] Figure 6 is a schematic diagram of a communication device provided in an embodiment of this application;

[0068] Figure 7 is a schematic diagram of a communication device provided in an embodiment of this application. Detailed Implementation

[0069] To address the challenges of wireless broadband technology and maintain the leading edge of the 3rd Generation Partnership Project (3GPP) network, the 3GPP standards group has developed a 5G network architecture. This architecture not only supports access to the 5G core network (CN) using radio access technologies defined by the 3GPP standards group (such as Long Term Evolution (LTE) and 5G Radio Access Network (RAN) technologies), but also supports access to the core network using non-3GPP access technologies through non-3GPP interworking functions (N3IWF) or next-generation packet data gateways (ngPDG).

[0070] Figure 1 is a schematic diagram of a service-oriented 5G network architecture. The 5G network architecture shown in Figure 1 may include access network equipment and core network equipment. Terminal devices access the DN through access network equipment and core network equipment. The core network equipment includes, but is not limited to, some or all of the following network elements: authentication server function (AUSF) network element, unified data management (UDM) network element, unified data repository (UDR) network element, network repository function (NRF) network element, network exposure function (NEF) network element, application function (AF) network element, policy control function (PCF) network element, access and mobility management function (AMF) network element, session management function (SMF) network element, user plane function (UPF) network element, and compute management function (CMF) network element.

[0071] Access network equipment, sometimes also called RAN nodes, RAN entities, or access nodes, is used to help terminal equipment achieve wireless access.

[0072] In one possible scenario, the access network device can be a base station, an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a base station in a future mobile communication system, or an access node in a wireless fidelity (WiFi) system. The access network device can be a macro base station, a micro base station, an indoor station, a relay node, or a donor node. Optionally, the access network device can also be a server, a wearable device, a vehicle, or an in-vehicle device. For example, the access network device in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of the access network device in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The access network device can also be equipped with communication modules, circuits, or chips that perform corresponding communication functions. The access network device can also be configured with program instructions for performing corresponding communication functions and corresponding program instructions. The access network device in this application may also be a logical node, logical module, or software that can implement all or part of the functions of the access network device.

[0073] In another possible scenario, multiple access network devices collaborate to assist terminal devices in achieving wireless access, with each access network device performing a portion of the base station's functions. For example, the access network devices can be a central unit (CU), a distributed unit (DU), a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU), etc. The CU and DU can be configured separately or included in the same network element, such as a baseband unit (BBU). The RU can be included in radio frequency equipment or radio frequency units, such as a remote radio unit (RRU), an active antenna unit (AAU), or a remote radio head (RRH).

[0074] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an open radio access network (ORAN) system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through a software module, a hardware module, or a combination of software and hardware modules.

[0075] Terminal devices can also be called terminals, user equipment (UE), mobile stations, mobile terminals, etc. Terminal devices can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), the Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, etc. Terminal devices can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, transportation vehicles with wireless communication capabilities, communication modules, etc. The embodiments of this application do not limit the device form of the terminal device. Terminal devices typically contain communication modules, circuits, or chips that perform corresponding communication functions. Terminal devices can also be configured with program instructions for performing corresponding communication functions.

[0076] Access network equipment and terminal equipment can be fixed or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can be deployed in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the application scenarios of the access network equipment and terminal equipment.

[0077] AMF network elements include functions such as performing mobility management or access authentication / authorization. They are also responsible for transmitting user policies between terminal devices and PCF network elements.

[0078] SMF network elements include functions such as performing session management, executing control policies issued by PCF network elements, selecting UPF network elements, or allocating Internet Protocol (IP) addresses to terminal devices.

[0079] UPF network elements include functions such as user plane data forwarding, session / flow-based billing statistics, and bandwidth limiting.

[0080] UDM network elements include functions such as managing contracted data or authorizing user access.

[0081] UDR includes functions for accessing data of various types, such as contract data, policy data, or application data.

[0082] NEF network elements are used to support the opening of capabilities and events, enabling third parties to indirectly interact with certain network elements within the 3GPP network.

[0083] AF (Application Provider) network elements convey application-side requests to the network side, such as QoS requirements or user state event subscriptions. AFs can be third-party functional entities or application services deployed by operators, such as IP Multimedia Subsystem (IMS) voice call services. AF network elements include AFs within the core network (i.e., operator-owned AFs) and third-party AFs (such as an enterprise's application server).

[0084] PCF network elements include policy control functions such as billing at the session and service flow levels, QoS bandwidth guarantee and mobility management, or terminal device policy decisions.

[0085] NRF network elements can be used to provide network element discovery functionality, providing network element information corresponding to the network element type based on requests from other network elements. NRF network elements also provide network element management services, such as network element registration, updates, deregistration, or network element status subscription and push.

[0086] The AUSF network element is responsible for authenticating users to determine whether to allow users or devices to access the network.

[0087] CMF network elements are responsible for establishing computing connections, managing computing plane sessions, and selecting computing nodes with service processing capabilities.

[0088] A Domain Provider (DN) is a network located outside of the carrier's network. A carrier's network can connect to multiple DNs, and various services can be deployed on a DN, providing data and / or voice services to terminal devices. For example, a DN might be the private network of a smart factory. Sensors installed in the workshop can act as terminal devices, and a control server for these sensors is deployed within the DN. The control server provides services to the sensors. Sensors can communicate with the control server, receive instructions from it, and transmit the collected sensor data back to the control server accordingly. Another example is a DN serving as an internal office network for a company. Employees' mobile phones or computers can act as terminal devices, accessing information and data resources on the company's internal office network.

[0089] In Figure 1, Nausf, Npcf, Nudr, Nudm, Naf, Namf, Nsmf, Nnef, and Nnrf are the service-based interfaces (SBIs) provided by AUSF, PCF, UDR, UDM, AF, AMF, SMF, NEF, and NRF, respectively, used to invoke the corresponding service-based operations. N1, N2, N3, N4, and N6 are interface sequence numbers, and their meanings are as follows:

[0090] 1) N1: The interface between the AMF network element and the terminal device, which can be used to transmit non-access stratum (NAS) signaling (such as QoS rules from the AMF network element) to the terminal device.

[0091] 2) N2: The interface between the AMF network element and the access network equipment, which can be used to transmit radio bearer control information from the core network side to the access network equipment.

[0092] 3) N3: The interface between the access network equipment and the UPF network element, mainly used to transmit uplink and downlink user plane data between the access network equipment and the UPF network element.

[0093] 4) N4: The interface between SMF network elements and UPF network elements. It can be used to transmit information between the control plane and the user plane, including the distribution of forwarding rules, QoS rules, traffic statistics rules, etc. from the control plane to the user plane, as well as the reporting of information from the user plane.

[0094] 5) N6: The interface between the UPF network element and the DN, used to transmit uplink and downlink user data streams between the UPF network element and the DN.

[0095] In the architecture shown in Figure 1, the various network functional elements are connected via a service-oriented bus and interact through service-oriented interfaces. The advantages of a service-oriented bus include improved network flexibility, openness, scalability, and intelligence, enabling support for diverse service scenarios and requirements. The service-oriented bus can be used to transmit various types of data and signaling. For example, it can be used to transmit latency-sensitive real-time signaling (e.g., service-oriented interface call signaling between network functional elements), latency-sensitive real-time data (e.g., real-time AI inference data), and non-real-time data (e.g., offline AI training data). Furthermore, when transmitting this data or signaling, the service-oriented bus couples the data or signaling together; that is, the service-oriented bus can simultaneously transmit real-time signaling, real-time data, and non-real-time data.

[0096] It should be noted that the term "network element" can be omitted when describing the above network elements (such as SMF network elements, UPF network elements, etc.). For example, an SMF network element can be abbreviated as SMF, a UPF network element as UPF, and so on. This abbreviated description is also used in Figure 1.

[0097] It is understood that the aforementioned network element or function can be a network component in a hardware device, a software function running on dedicated hardware, or a virtualized function instantiated on a platform (e.g., a cloud platform). Optionally, the aforementioned network element or function can be implemented by one device, multiple devices working together, or a functional module within a single device; this application embodiment does not specifically limit this.

[0098] The computing management function network element and user plane network element in this application can be the CMF network element and UPF network element in Figure 1, respectively, or they can be network elements in future communication networks that have the functions of the aforementioned CMF network element and UPF network element. This application does not limit them in this regard.

[0099] With the widespread deployment of networks, mobile internet has achieved tremendous success, gradually becoming one of the main ways for users to access the internet, leading to an explosive growth in mobile traffic. Traditional centralized anchor point deployment methods are increasingly unable to support this rapidly growing mobile traffic model. On the one hand, in networks with centralized anchor point gateway deployment, the increased traffic ultimately concentrates at the gateway and core equipment room, placing increasingly higher demands on backhaul network bandwidth, equipment room throughput, and gateway specifications. On the other hand, the long-distance backhaul network from the access network to the anchor point gateway and the complex transmission environment also result in significant latency and jitter in user packet transmission. The backhaul network mainly includes intermediate links from the core network (or backbone network) to the edge subnet.

[0100] Against this backdrop, the industry has proposed the concept of edge computing. Edge computing enables distributed local processing of business traffic by moving business processing capabilities to the network edge, avoiding excessive traffic concentration and thus significantly reducing the specification requirements for core data centers and centralized gateways. At the same time, edge computing also shortens the distance of the backhaul network, reduces end-to-end latency and jitter of user packets, and makes the deployment of ultra-low latency services possible.

[0101] On the other hand, with the rapid development of AI services, generative AI services are gradually becoming more diverse and evolving into AI agents. Since AI services require significant computing power, power consumption, and storage resources, this presents new challenges for terminal devices. Due to resource limitations on the edge side, cloud-edge-device computing collaboration is a new trend that can alleviate the computing pressure on the edge side. Therefore, in one implementation, computing nodes with AI computing-related business processing capabilities can be deployed locally. In this case, the computing node can be an application server, a functional unit deployed within the network, or it can be deployed within or near the access network equipment.

[0102] A computing node can receive multiple computing tasks (or inference tasks) from the same or different terminal devices. Once a computing node receives all or most of the data for a computing task, it can begin performing computations (or inferences) based on the received data and output the corresponding computation results (or inference results). However, when a computing node executes multiple computing tasks concurrently, it may increase the computation latency of each computing task, making it impossible to guarantee the end-to-end latency of each computing task.

[0103] For example, both terminal device #1 and terminal device #2 have computing tasks to execute. Terminal device #1 sends multiple data for computing task #1 to the computing node, and terminal device #2 sends multiple data for computing task #2 to the same computing node. For the sake of fairness, the data from terminal device #1 and terminal device #2 are scheduled in turn, for example, the sending order is: data #1 from terminal device #1, data #1 from terminal device #2, data #2 from terminal device #1, and so on. Based on this scheduling method, after receiving data #1 from terminal device #1, the computing node needs to wait to receive other data from computing task #1 of terminal device #1. Only after receiving all or most of the data from computing task #1 will it begin computation or inference. However, because the computing node receives data from computing task #1 from terminal device #1 and data from computing task #2 from terminal device #2 simultaneously, the time when the computing node starts executing computing task #1 is delayed, thus increasing the real-time computation latency of computing task #1. Here, real-time computation latency refers to the real-time duration between the computing node receiving data from computing task #1 (e.g., the first data) and the computing node outputting the computation result of computing task #1. Because the computation latency of computing task #1 increases, the end-to-end latency of computing task #1 cannot be guaranteed either. Here, end-to-end latency refers to the duration between terminal device #1 sending data from computing task #1 (e.g., the first data) and receiving the computation result of computing task #1. Similarly, the computation latency of computing task #2 also increases, and therefore, the end-to-end latency of computing task #2 cannot be guaranteed either.

[0104] To address the aforementioned issues, this application provides corresponding solutions.

[0105] The communication method and communication device will be further described below with reference to the accompanying drawings. It is understood that this application uses computing management function network elements, access network equipment, and computing nodes as examples of the execution entities in this interactive illustration, but this application does not limit the execution entities in the interactive illustration. For example, the method executed by the computing management function network element in this application can also be implemented by modules (e.g., circuits, chips, or chip systems) within the computing management function network element, or by logical nodes, logical modules, or software capable of implementing all or part of the functions of the computing management function network element. Similarly, the method executed by the computing node in this application can also be implemented by modules (e.g., circuits, chips, or chip systems) within the computing node, or by logical nodes, logical modules, or software capable of implementing all or part of the functions of the computing node. Likewise, the method executed by the access network equipment in this application can also be implemented by modules (e.g., circuits, chips, or chip systems) within the access network equipment, or by logical nodes, logical modules, or software capable of implementing all or part of the functions of the access network equipment.

[0106] In this application, a computing node can be configured with one or more computing power applications and corresponding computing power resources for each application. Each computing power application supports processing corresponding computing tasks or large model inference requests. A computing node is also referred to as a computing unit, a far edge intelligent node (FEIN), or a computation execution function (CEF) network element. Optionally, a computing task can be uniquely identified by any one or more of the following: task ID, address, computing model identifier, or port number. Optionally, each computing task corresponds to one or more computing models. In one possible implementation, a computing task refers to a computing model on a computing node that can be distinguished based on the type of service it supports, such as supporting artificial intelligence models, large language models, large model inference, image rendering services, image recognition, human-computer interaction, or image-text question answering models; it can also be distinguished based on different computing services, such as deployment by operators, Wenxin Yiyan, Kimi, or other manufacturers; or it can be distinguished by the service parameters of different supported computing services. Computing tasks can also be distinguished by one or more of the three distinction methods mentioned above.

[0107] In this application, the computational model can be deployed on computational nodes. As one implementation method, the terms "computational model" and "computational node" can be considered equivalent; that is, the node where the computational model is deployed is the computational node. In this case, all computational nodes in this application can be considered to be replaced by computational models. The computational model can also be a generative pre-trained transformer (GPT) model, a mixture of experts (MoE) model, a Llama (Large Language Model Meta AI) model, or other large language models, etc., and this invention does not limit this.

[0108] Figure 2(a) is a flowchart illustrating a communication method provided in an embodiment of this application. The method includes the following steps:

[0109] Step 201a: The computing management function network element determines, based on the occurrence time of the first computing task and the occurrence time of the second computing task, that the difference between the occurrence time of the first computing task and the occurrence time of the second computing task is less than a preset time threshold.

[0110] The first and second computing tasks are executed by the same computing node.

[0111] The computational task here (e.g., the first computational task or the second computational task) refers to a task that requires relevant reasoning or computation. For example, a computational task might be rendering an image, rendering a video, or a question-and-answer format, etc. This application does not limit the specific type of computational task.

[0112] The first computing task is initiated by a first terminal device, and the second computing task is initiated by a second terminal device. The first terminal device and the second terminal device can be the same terminal device or different terminal devices. The first computing task and the second computing task can correspond to the same computing model (or inference model), that is, the first computing task and the second computing task are processed by the same computing model.

[0113] The occurrence time of the first computing task refers to the time when the first terminal device initiates the first computing task, or the time when the first terminal device sends the data for the first computing task to the computing node. The occurrence time of the second computing task refers to the time when the second terminal device initiates the second computing task, or the time when the second terminal device sends the data for the second computing task to the computing node.

[0114] For example, both the first and second computing tasks are periodic computing tasks. For instance, in industrial applications, multiple industrial robots (i.e., the first or second terminal device, etc.) periodically collect images and send the collected image information (i.e., data from the first or second computing task, etc.) to a cloud server (i.e., a computing node) at a rate of 10 images per second. The server then analyzes the image information to obtain the calculation results (or inference results, processing results) and sends the calculation results back to the industrial robots.

[0115] Specifically, determining that the difference between the occurrence time of the first computing task and the occurrence time of the second computing task is less than a preset time threshold can be understood as determining that the first and second computing tasks are executed concurrently. That is, it means that the first terminal device and the second terminal device send the data of the first computing task and the data of the second computing task to the same computing node at the same or similar times. The size of this preset time threshold can be pre-configured by network administrators in the computing management function network element, or it can be sent from the policy control network element to the computing management function network element.

[0116] As one implementation method, the computing management function network element determines the occurrence time of the first computing task based on its cycle. For example, the computing management function network element can determine multiple occurrence times of the first computing task based on its cycle and the time when the first computing task is first initiated. As an example, if the time when the first computing task is first initiated is t1, and the cycle of the first computing task is T1, then the multiple occurrence times of the first computing task are t1, t1+T1, t1+2T1, t1+3T1, and so on.

[0117] As one implementation method, the computing management function network element determines the occurrence time of the second computing task based on its cycle. For example, the computing management function network element can determine multiple occurrence times of the second computing task based on its cycle and the time when the second computing task is first initiated. As an example, if the time when the second computing task is first initiated is t2, and the cycle of the second computing task is T1, then the multiple occurrence times of the second computing task are t2, t2+T2, t2+2T2, t2+3T2, and so on.

[0118] Based on the above example, the difference between the occurrence time of the first computation task and the occurrence time of the second computation task being less than a preset time threshold may include, but is not limited to, one or more of the following: the difference between t1 and t2 is less than a preset time threshold, the difference between t1 and t2+T2 is less than a preset time threshold, the difference between t1 and t2+2T2 is less than a preset time threshold, the difference between t1 and t2+3T2 is less than a preset time threshold, the difference between t1+T1 and t2 is less than a preset time threshold, the difference between t1+T1 and t2+T2 is less than a preset time threshold, the difference between t1+T1 and t2+2T2 is less than a preset time threshold, and the difference between t1+T1 and t2+3T2 is less than a preset time threshold.

[0119] As one implementation method, the computing management function network element can receive a first policy charging control rule and a second policy charging control rule from the policy control network element. The first policy charging control rule indicates the period of the first computing task, and the second policy charging control rule indicates the period of the second computing task.

[0120] As another implementation method, the computing management function network element can receive a third policy billing control rule from the policy control network element. The third policy billing control rule indicates the period of the first computing task and the period of the second computing task.

[0121] Step 202a: The computing management function network element determines the first uplink transmission delay of the first computing task between the first terminal device and the computing node based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node; and determines the second uplink transmission delay of the second computing task between the second terminal device and the computing node based on the second end-to-end delay corresponding to the second computing task and the second computing delay of the second computing task on the computing node.

[0122] The first uplink transmission delay is different from the second uplink transmission delay.

[0123] As one implementation method, the computing management function network element also determines the first downlink transmission delay of the first computing task between the first terminal device and the computing node based on the first end-to-end delay, the first computing delay, and the first uplink transmission delay. And it determines the second downlink transmission delay of the second computing task between the second terminal device and the computing node based on the second end-to-end delay, the second computing delay, and the second uplink transmission delay.

[0124] In one possible implementation, the first end-to-end latency corresponding to the first computing task refers to the latency between the first terminal device sending data for the first computing task and the first terminal device receiving the computation result (or inference result, processing result) of the first computing task from the computing node. For example, a first end-to-end latency of 500 milliseconds (ms) means that the latency between the first terminal device sending data for the first computing task to the computing node and the first terminal device receiving the computation result of the first computing task is 500ms. The first end-to-end latency corresponding to the first computing task includes three parts: the first part is the first uplink transmission latency of the first computing task, that is, the latency between the first terminal device sending data for the first computing task and the computing node receiving the data for the first computing task, where the data can be part or all of the data for the first computing task; the second part is the first computing latency of the first computing task, that is, the latency between the computing node receiving the data for the first computing task (e.g., the first data) and the computing node sending the computation result of the first computing task; the third part is the downlink transmission latency of the first computing task, that is, the latency between the computing node sending the computation result of the first computing task and the first terminal device receiving the computation result of the first computing task. Wherein, the first uplink transmission delay + the first computation delay + the first downlink transmission delay ≤ the first end-to-end delay. In one application scenario, assuming the computing node is deployed near the access network device, the first uplink transmission delay consists of the following delays: uplink transmission delay 1 for the transmission of data of the first computation task between the first terminal device and the access network device, and uplink transmission delay 2 for the transmission of data of the first computation task between the access network device and the computing node; the first downlink transmission delay consists of the following delays: downlink transmission delay 1 for the transmission of data of the first computation task between the computing node and the access network device, and downlink transmission delay 2 for the transmission of data of the first computation task between the access network device and the first terminal device. In another application scenario, assuming the computing node is deployed within the DN, the first uplink transmission delay consists of the following delays: uplink transmission delay 1 for the transmission of the first computing task between the first terminal device and the access network device, uplink transmission delay 2 for the transmission of the first computing task between the access network device and the UPF network element, and uplink transmission delay 3 for the transmission of the first computing task between the UPF network element and the computing node; the first downlink transmission delay consists of the following delays: downlink transmission delay 1 for the transmission of the first computing task data between the computing node and the UPF network element, downlink transmission delay 2 for the transmission of the first computing task data between the UPF network element and the access network device, and downlink transmission delay 3 for the transmission of the first computing task data between the access network device and the first terminal device.In another application scenario, assuming the computing node is deployed after the UPF, the first uplink transmission delay consists of the following delays: uplink transmission delay 1 for the first computing task transmitted between the first terminal device and the access network device, uplink transmission delay 2 for the first computing task transmitted between the access network device and the UPF network element, and uplink transmission delay 3 for the first computing task transmitted between the UPF network element and the computing node; the first downlink transmission delay consists of the following delays: downlink transmission delay 1 for the data of the first computing task transmitted between the computing node and the UPF network element, downlink transmission delay 2 for the data of the first computing task transmitted between the UPF network element and the access network device, and downlink transmission delay 3 for the data of the first computing task transmitted between the access network device and the first terminal device. For example, the first end-to-end delay corresponding to the first computing task may be 500ms, the first computing delay of the first computing task on the computing node may be 200ms, the first uplink transmission delay may be 100ms, and the first downlink transmission delay may be 200ms.

[0125] In one possible implementation, the second end-to-end latency corresponding to the second computing task refers to the time delay between the second terminal device sending data for the second computing task and the second terminal device receiving the computing result (or inference result, processing result) of the second computing task from the computing node. For example, a second end-to-end latency of 500ms means that the time delay between the second terminal device sending data for the second computing task to the computing node and the second terminal device receiving the computing result of the second computing task is 500ms. The second end-to-end latency corresponding to the second computing task includes three parts: the first part is the second uplink transmission latency of the second computing task, that is, the time delay between the second terminal device sending data for the second computing task and the computing node receiving the data for the second computing task, where the data can be part or all of the data for the second computing task; the second part is the second computing latency of the second computing task, that is, the time delay between the computing node receiving the data for the second computing task (for example, it could be the first data) and the computing node sending the computing result of the second computing task; the third part is the downlink transmission latency of the second computing task, that is, the time delay between the computing node sending the computing result of the second computing task and the second terminal device receiving the computing result of the second computing task. Wherein, the second uplink transmission delay + the second computation delay + the second downlink transmission delay ≤ the second end-to-end delay. In one application scenario, assuming the computing node is deployed near the access network device, the second uplink transmission delay consists of the following delays: uplink transmission delay 1 for the transmission of data of the second computation task between the second terminal device and the access network device, and uplink transmission delay 2 for the transmission of data of the second computation task between the access network device and the computing node; the second downlink transmission delay consists of the following delays: downlink transmission delay 1 for the transmission of data of the second computation task between the computing node and the access network device, and downlink transmission delay 2 for the transmission of data of the second computation task between the access network device and the second terminal device. In another application scenario, assuming the computing node is deployed within the DN, the second uplink transmission delay consists of the following delays: uplink transmission delay 1 for the transmission of the second computing task between the second terminal device and the access network device, uplink transmission delay 2 for the transmission of the second computing task between the access network device and the UPF network element, and uplink transmission delay 3 for the transmission of the second computing task between the UPF network element and the computing node; the second downlink transmission delay consists of the following delays: downlink transmission delay 1 for the transmission of the second computing task data between the computing node and the UPF network element, downlink transmission delay 2 for the transmission of the second computing task data between the UPF network element and the access network device, and downlink transmission delay 3 for the transmission of the second computing task data between the access network device and the second terminal device.In another application scenario, assuming the computing node is deployed after the UPF, the second uplink transmission delay consists of the following delays: uplink transmission delay 1 for the second computing task transmitted between the second terminal device and the access network device, uplink transmission delay 2 for the second computing task transmitted between the access network device and the UPF network element, and uplink transmission delay 3 for the second computing task transmitted between the UPF network element and the computing node; the second downlink transmission delay consists of the following delays: downlink transmission delay 1 for the data of the second computing task transmitted between the computing node and the UPF network element, downlink transmission delay 2 for the data of the second computing task transmitted between the UPF network element and the access network device, and downlink transmission delay 3 for the data of the second computing task transmitted between the access network device and the second terminal device. For example, the second end-to-end delay corresponding to the second computing task may be 500ms, the second computing delay of the second computing task on the computing node may be 200ms, the second uplink transmission delay may be 200ms, and the second downlink transmission delay may be 100ms.

[0126] As one implementation method, when the first calculation delay is less than the second calculation delay, the first uplink transmission delay is less than the second uplink transmission delay; when the first calculation delay is greater than the second calculation delay, the first uplink transmission delay is greater than the second uplink transmission delay.

[0127] As another implementation method, before step 202a, the computing management function network element also determines that the first computing task and the second computing task are processed by the same computing node. That is, when it is determined that the first computing task and the second computing task are processed by the same computing node, and it is determined that the difference between the occurrence time of the first computing task and the occurrence time of the second computing task is less than a preset time threshold, the above step 202a is triggered.

[0128] Step 203a: The computing management function network element determines the quality of service parameters in the quality of service file of the first quality of service flow corresponding to the first computing task based on the first uplink transmission delay, and determines the quality of service parameters in the quality of service file of the second quality of service flow corresponding to the second computing task based on the second uplink transmission delay.

[0129] The quality parameters in the quality of service file of the first quality of service flow include at least the latency parameter of the first quality of service flow. This latency parameter indicates the uplink and / or downlink latency of the first computing task's transmission over the network. It can be understood that the uplink latency is a portion or all of the first uplink transmission latency. In one possible implementation, the latency parameter is a portion of the first transmission latency. This latency can be any one or more of the following: the transmission latency of the first computing task on the air interface (i.e., the access network), the transmission latency of the first computing task from the terminal device to the UPF network element, and the transmission latency of the first computing task from the terminal device to the computing node. Here, the transmission latency refers to the upper limit of the latency requirement for data packets to be transmitted between the terminal device and the access network device, UPF network element, or computing node, or the upper limit of the latency requirement for the completion of data packet transmission of the first computing task between the terminal device and the access network device, UPF network element, or computing node. In one possible implementation, the latency parameter is the total latency in the first transmission latency. This latency can be any one or more of the following: the transmission latency of the first computing task from the terminal device to the UPF network element, or the transmission latency of the first computing task from the terminal device to the computing node. Here, the transmission latency refers to the upper limit of the latency requirement for data packets to be transmitted between the terminal device and the UPF network element or the computing node, or the upper limit of the latency requirement for the completion of data packet transmission of the first computing task between the terminal device and the access network device, UPF network element, or computing node. In another possible implementation, the access network device determines the transmission latency in the wireless network based on this latency parameter, such as determining the transmission latency on the air interface side (i.e., the access network) or the transmission latency between the terminal device and the user plane functional network element. In addition to the quality parameters, the QoS file of the first QoS flow may also contain other information, such as the identifier of the first QoS flow, i.e., the QoS flow identity (QFI).

[0130] The quality parameters in the quality of service file of the second quality of service flow include at least the latency parameters of the second quality of service flow. These latency parameters indicate the uplink and / or downlink latency of the second computing task transmitted on the wireless network. This uplink latency can be understood as a portion of the second uplink transmission latency. In addition to the quality parameters, the quality of service file of the second quality of service flow may also contain other information, such as the identifier of the second quality of service flow, i.e., QFI.

[0131] Step 204a: The network element performing the calculation and management functions sends the quality of service (QoS) files of the first QoS stream and the second QoS stream to the access network equipment.

[0132] The access network device can schedule the data of the first computing task sent through the first quality of service stream according to the quality of service file of the first quality of service stream, so that the data of the first computing task sent by the first terminal device reaches the access network device after a first uplink transmission delay. And according to the quality of service file of the second quality of service stream, the access network device can schedule the data of the second computing task sent through the second quality of service stream, so that the data of the second computing task sent by the second terminal device reaches the access network device after a second uplink transmission delay.

[0133] Based on the above scheme, the same computing node executes the first and second computing tasks. If the occurrence times of the two computing tasks are relatively close, i.e., the difference between their occurrence times is less than a preset time threshold, then by configuring different uplink transmission delays for the two computing tasks, the data for one computing task arrives at the computing node first and is executed by the computing node. After the execution of the first computing task is completed or nearly completed, the other computing task arrives at the computing node and is executed by the computing node. Since the data for the two computing tasks do not arrive at the computing node at the same or similar times, neither computing task needs to wait for a long time before it can be executed by the computing node, or the computing node does not need to process multiple computing tasks simultaneously. This reduces the generation time of the first token for each computing task, improves the computing efficiency of the computing node, and thus helps to ensure end-to-end latency.

[0134] The above explanation uses two computing tasks as an example. The method embodiments described above still apply when three or more computing tasks are involved. For three or more computing tasks, at least two of the three or more computing tasks have different uplink transmission delays. For example, the end-to-end delays of computing tasks #1, #2, and #3 are all 500ms. Specifically, the computing delay of computing task #1 is 100ms, the uplink transmission delay is 200ms, and the downlink transmission delay is 200ms. The computing delay of computing task #2 is 100ms, the uplink transmission delay is 100ms, and the downlink transmission delay is 300ms. The computing delay of computing task #3 is 200ms, the uplink transmission delay is 200ms, and the downlink transmission delay is 100ms.

[0135] This application does not limit the implementation method of the computing management function network element to obtain the end-to-end latency corresponding to the computing task. For example, two different methods are given below.

[0136] Method 1: The computing management function network element obtains the end-to-end latency corresponding to the computing task from the policy charging control rules.

[0137] For example, the aforementioned first strategy billing control rule includes the identifier of the first computing task and the first end-to-end latency corresponding to the first computing task. Similarly, the aforementioned second strategy billing control rule includes the identifier of the second computing task and the second end-to-end latency corresponding to the second computing task.

[0138] For example, the aforementioned third strategy billing control rule includes the identifier of the first computing task, the first end-to-end delay corresponding to the first computing task, the identifier of the second computing task, and the second end-to-end delay corresponding to the second computing task.

[0139] In method 2, the computing management function network element determines the end-to-end latency corresponding to the computing task based on the local policy.

[0140] The local policy can be pre-configured on the computing management function network element, or it can be received by the computing management network element from other network elements, such as being reported in advance by the terminal device to the computing management function network element.

[0141] For example, the computing management function network element determines the first end-to-end latency corresponding to the first computing task and the second end-to-end latency corresponding to the second computing task based on local policies.

[0142] This application does not limit the implementation method of the computing management function network element obtaining the computing latency corresponding to the computing task. For example, two different methods are given below.

[0143] In method A, the computing management function network element obtains the computing latency corresponding to the computing task from the local network, network management network, NRF network element, or UDR network element.

[0144] For example, the computing management function network element obtains the first computing latency corresponding to the first computing task and the second computing latency corresponding to the second computing task from the local network, network management network, NRF network element or UDR network element.

[0145] In method B, the computing management function network element obtains the computing latency corresponding to the computing task from the policy-based charging control rules.

[0146] For example, the aforementioned first strategy billing control rule includes the identifier of the first computing task and the first computing delay corresponding to the first computing task. Similarly, the aforementioned second strategy billing control rule includes the identifier of the second computing task and the second computing delay corresponding to the second computing task.

[0147] For example, the aforementioned third strategy billing control rule includes the identifier of the first computing task, the first computing delay corresponding to the first computing task, the identifier of the second computing task, and the second computing delay corresponding to the second computing task.

[0148] As one implementation method, the computing management function network element can also send first information to the first terminal device and / or computing node. This first information is used to map the data of the first computing task to a first quality of service (QoS) stream for transmission. The first terminal device can transmit the data of the first computing task via the first QoS stream based on the first information. The computing node can transmit the computation result of the first computing task via the first QoS stream or another QoS stream (e.g., a third QoS stream) based on the first information, for example, by adding an identifier of the third QoS stream to the downlink data packet carrying the computation result.

[0149] As one implementation method, the computing management function network element can also send second information to the second terminal device and / or computing node. This second information is used to map the data of the second computing task to a second Quality of Service (QoS) stream for transmission. The second terminal device can transmit the data of the second computing task via the second QoS stream based on the second information, for example, by adding an identifier of the second QoS stream to the uplink data packet carrying the data. The computing node can transmit the computation result of the second computing task via the second QoS stream or other QoS streams (such as a fourth QoS stream) based on the second information, for example, by adding an identifier of the fourth QoS stream to the downlink data packet carrying the computation result, to ensure the latency parameters of the second computing task.

[0150] As one implementation method, the first computational delay in the above embodiment can specifically be the delay between the computing node receiving the data of the first computational task (e.g., the first data) and the computing node sending the computation result of the first computational task under the first segmentation method. The first task segmentation method is used to indicate the task segmentation method or model segmentation method when the first terminal device and the computing node jointly execute the first computational task.

[0151] As one implementation method, the second computational delay in the above embodiment can specifically be the delay between the computing node receiving the data of the second computational task (e.g., the first data) and the computing node sending the computation result of the second computational task under the second segmentation method. The second task segmentation method is used to indicate the task segmentation method or model segmentation method when the second terminal device and the computing node jointly execute the second computational task.

[0152] In this embodiment, when a computing task (e.g., the first computing task or the second computing task) requires a task partitioning method, it means that a terminal device (e.g., the first terminal device or the second terminal device) and a computing node need to jointly execute a computing task. For example, if computing task #1 is not partitioned, and the computing node executes computing task #1 entirely, the computation latency of computing task #1 on the computing node is 400ms. Under task partitioning method #1, the computation latency of computing task #1 on the computing node is 100ms. Under task partitioning method #2, the computation latency of computing task #1 on the computing node is 200ms. Under task partitioning method #3, the computation latency of computing task #1 on the computing node is 300ms. It can be seen that when the computing task is partitioned, since the terminal device also executes part of the computing task, the latency of the computing node executing the computing task is reduced.

[0153] In addition, there is another task partitioning method where the terminal device (e.g., a first terminal device or a second terminal device) tokenizes the computing task (e.g., a first computing task or a second computing task). That is, the terminal device converts the data packet of the computing task into multiple tokens, and then sends these tokens to the computing node. The computing node then processes these tokens to obtain the computing result. Based on this method, since the computing node does not need to perform the tokenization operation, the computation latency of the computing node executing the computing task can be reduced.

[0154] Figure 2(b) is a flowchart illustrating a communication method provided in an embodiment of this application. The method includes the following steps:

[0155] Step 201b: The computing management function network element determines multiple different uplink transmission delays between the terminal device and the computing node for the first computing task based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node.

[0156] The first computational task here refers to a task that requires relevant reasoning or computation. For example, the first computational task could be rendering an image, rendering a video, or a question-and-answer format. This application does not limit the specific type of the first computational task.

[0157] The first computing task is initiated by the terminal device. This first computing task can correspond to a computing model (or inference model), that is, the first computing task is processed by the computing model.

[0158] For example, the first computing task is a non-periodic computing task.

[0159] As one implementation method, if the computing management function network element determines that the first computing task is an aperiodic computing task, then it determines multiple different uplink transmission delays between the terminal device and the computing node based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node. That is, when the computing management function network element determines that the first computing task is an aperiodic computing task, it triggers the computing management function network element to execute step 201b.

[0160] As one implementation method, the computing management function network element can receive policy charging control rules from the policy control network element. These policy charging control rules are used to indicate that the first computing task is a non-periodic computing task. For example, the policy charging control rules also include a first end-to-end delay corresponding to the first computing task and / or a first computing delay of the first computing task on the computing node.

[0161] In one possible implementation, the computing management function network element also determines multiple different downlink transmission delays of the first computing task between the terminal device and the computing node based on the first end-to-end delay, the first computing delay, and multiple different uplink transmission delays.

[0162] In one possible implementation, the first end-to-end latency corresponding to the first computing task refers to the latency between the terminal device sending data for the first computing task and the terminal device receiving the computing result (or inference result, processing result) of the first computing task from the computing node. For example, a first end-to-end latency of 500ms means that the latency between the terminal device sending data for the first computing task to the computing node and the terminal device receiving the computing result of the first computing task is 500ms. The first end-to-end latency corresponding to the first computing task includes three parts: the first part is the uplink transmission latency of the first computing task (also called the first uplink transmission latency), which is the latency between the terminal device sending data for the first computing task and the computing node receiving the data for the first computing task, where the data can be part or all of the data for the first computing task; the second part is the first computing latency of the first computing task, which is the latency between the computing node receiving the data for the first computing task (e.g., the first data) and the computing node sending the computing result of the first computing task; the third part is the downlink transmission latency of the first computing task (also called the first downlink transmission latency), which is the latency between the computing node sending the computing result of the first computing task and the terminal device receiving the computing result of the first computing task. Wherein, uplink transmission delay + first computation delay + downlink transmission delay ≤ first end-to-end delay. In one application scenario, assuming the computing node is deployed near the access network device, the uplink transmission delay consists of the following delays: uplink transmission delay 1 for transmitting data of the first computation task between the terminal device and the access network device, and uplink transmission delay 2 for transmitting data of the first computation task between the access network device and the computing node; the downlink transmission delay consists of the following delays: downlink transmission delay 1 for transmitting data of the first computation task between the computing node and the access network device, and downlink transmission delay 2 for transmitting data of the first computation task between the access network device and the terminal device. In another application scenario, assuming the computing node is deployed within the DN, the uplink transmission latency consists of the following latency: uplink transmission latency 1 for the transmission of the first computing task between the terminal device and the access network device, uplink transmission latency 2 for the transmission of the first computing task between the access network device and the UPF network element, and uplink transmission latency 3 for the transmission of the first computing task between the UPF network element and the computing node; the downlink transmission latency consists of the following latency: downlink transmission latency 1 for the transmission of the first computing task's data between the computing node and the UPF network element, downlink transmission latency 2 for the transmission of the first computing task's data between the UPF network element and the access network device, and downlink transmission latency 3 for the transmission of the first computing task's data between the access network device and the terminal device.In another application scenario, assuming the computing node is deployed after the UPF network element, the second uplink transmission delay consists of the following delays: uplink transmission delay 1 for the second computing task transmitted between the second terminal device and the access network device, uplink transmission delay 2 for the second computing task transmitted between the access network device and the UPF network element, and uplink transmission delay 3 for the second computing task transmitted between the UPF network element and the computing node; the second downlink transmission delay consists of the following delays: downlink transmission delay 1 for the data of the second computing task transmitted between the computing node and the UPF network element, downlink transmission delay 2 for the data of the second computing task transmitted between the UPF network element and the access network device, and downlink transmission delay 3 for the data of the second computing task transmitted between the access network device and the second terminal device. For example, the first end-to-end delay corresponding to the first computing task is 500ms, the first computing delay of the first computing task on the computing node is 200ms, the uplink transmission delay is 100ms, and the downlink transmission delay is 200ms.

[0163] In this step, the computing management function network element determines several different uplink transmission delays for the first computing task between the terminal device and the computing node. For example, the first end-to-end delay is 500ms, the first computing delay is 100ms, and the following combination of uplink and downlink transmission delays is determined:

[0164] (1) Uplink transmission delay: 100ms, downlink transmission delay: 300ms;

[0165] (2) Uplink transmission delay: 200ms, downlink transmission delay: 200ms;

[0166] (3) Uplink transmission delay: 300ms, downlink transmission delay: 100ms.

[0167] Step 202b: The computing management function network element determines multiple quality of service files for the first quality of service flow corresponding to the first computing task.

[0168] Each of the multiple quality of service (QoS) files includes one of several different uplink transmission delays, and the uplink transmission delays included in the different QoS files are different.

[0169] The quality parameters in the quality of service file of the first quality of service flow include at least the latency parameter of the first quality of service flow. This latency parameter indicates the uplink and / or downlink latency of the first computing task's transmission over the network. It can be understood that the uplink latency is a portion or all of the first uplink transmission latency. In one possible implementation, the latency parameter is a portion of the first transmission latency. This latency can be any one or more of the following: the transmission latency of the first computing task on the air interface (i.e., the access network), the transmission latency of the first computing task from the terminal device to the UPF network element, and the transmission latency of the first computing task from the terminal device to the computing node. Here, the transmission latency refers to the upper limit of the latency requirement for data packets to be transmitted between the terminal device and the access network device, UPF network element, or computing node, or the upper limit of the latency requirement for the completion of data packet transmission of the first computing task between the terminal device and the access network device, UPF network element, or computing node. In one possible implementation, the latency parameter is the total latency in the first transmission latency. This latency can be any one or more of the following: the transmission latency of the first computing task from the terminal device to the UPF network element, or the transmission latency of the first computing task from the terminal device to the computing node. Here, the transmission latency refers to the upper limit of the latency requirement for data packets to be transmitted between the terminal device and the UPF network element or the computing node, or the upper limit of the latency requirement for the completion of data packet transmission of the first computing task between the terminal device and the access network device, UPF network element, or computing node. In another possible implementation, the access network device determines the transmission latency in the wireless network based on this latency parameter, such as determining the transmission latency on the air interface side (i.e., the access network) or the transmission latency between the terminal device and the user plane functional network element. In addition to the quality parameters, the quality of service file of the first quality of service flow may also contain other information, such as the identifier of the first quality of service flow, i.e., QFI.

[0170] Continuing with the example, the computing management function network element determines three Quality of Service (QoS) files for the QoS flow corresponding to the first computing task. QoS file 1 includes a latency parameter 1, which indicates the uplink latency 1 for the first computing task's transmission over the wireless network, and this uplink latency 1 ≤ 100ms. QoS file 2 includes a latency parameter 2, which indicates the uplink latency 2 for the first computing task's transmission over the wireless network, and this uplink latency 2 ≤ 200ms. QoS file 3 includes a latency parameter 3, which indicates the uplink latency 3 for the first computing task's transmission over the wireless network, and this uplink latency 3 ≤ 300ms.

[0171] In step 203b, the computing management function network element sends a first indication message and multiple quality of service (QoS) files of the first QoS flow to the access network device. Correspondingly, the access network device receives the first indication message and the multiple QoS files of the first QoS flow.

[0172] The first instruction information is used to instruct the access network device to select one of the multiple quality of service files of the first quality of service stream based on the waiting time for sending the data of the first computing task.

[0173] Step 204b: The access network device selects a quality of service file from multiple quality of service files in the first quality of service stream based on the waiting time for sending the data of the first computing task.

[0174] Step 205b: The access network device schedules the data for the first computing task according to the selected quality of service file.

[0175] For example, the access network device may determine the waiting time for sending data for the first computing task according to any of the following methods.

[0176] Method 1: The access network device receives a second indication information from the computing node, which is used to indicate the waiting time.

[0177] The second indication information may include a waiting time for sending data for the first computing task. This waiting time may be, for example, the remaining computing time compared to other computing tasks (such as the second computing task) being processed by the computing node. By sending the waiting time to the access network device, the computing node enables the access network device to control the arrival time of the data for the first computing task at the computing node based on the waiting time, ensuring that the data for the first computing task can be processed immediately or as quickly as possible after arriving at the computing node, thus ensuring the end-to-end latency of the first computing task. For example, assuming the waiting time indicated by the second indication information is 100ms, the access network device can select the aforementioned Quality of Service (QoS) file 1 of the first QoS stream and use QoS file 1 to control the uplink transmission latency of the data for the first computing task sent by the first QoS stream. Since the uplink transmission latency of the data for the first computing task is 100ms, the data for the first computing task can be processed immediately or as quickly as possible after arriving at the computing node.

[0178] Optionally, the second indication information may not include the waiting time for sending data for the first computing task, but instead include parameters for determining the waiting time. The access network device determines the waiting time for sending data for the first computing task based on these parameters. For example, the second indication information includes a first parameter and a second parameter corresponding to the second computing task, where the second computing task is the computing task being processed by the computing node. The first parameter indicates the time interval between two adjacent tokens that generate the second computing task, and the second parameter indicates the number of remaining tokens for the generated second computing task. The access network device can then determine the waiting time for sending data for the first computing task based on the first and second parameters. For example, the waiting time for sending data for the first computing task is equal to t*N, where t is the time interval between two adjacent tokens that generate the second computing task, and N is the number of remaining tokens for the generated second computing task.

[0179] Method 2: The access network device determines the waiting time for sending data for the first computing task based on the computing delay of the third computing task.

[0180] The third computing task is the computing task in which the access network device recently transmitted data. The computing delay of the third computing task is used to indicate the delay between when the computing node receives the data of the third computing task and when it sends the computing result of the second computing task.

[0181] For example, if the terminal device starts transmitting data for the third computing task at time t1, and the uplink transmission delay of the third computing task is 100ms and the computing delay of the third computing task is 200ms, and if the terminal device needs to send data for the first computing task at time t1+100ms, the access network device can determine that the waiting time for sending the data for the first computing task is 200ms, i.e., (t1+100+200)-(t1+100)=200ms. Therefore, the service quality file 2 of the aforementioned first service quality flow can be selected, and the service quality file 2 can be used to control the uplink transmission delay of the first service quality flow sending the data for the first computing task. The uplink transmission delay of the data for the first computing task is 200ms, so that the data for the first computing task can be processed immediately or as soon as possible after arriving at the computing node, thereby ensuring the end-to-end latency of the first computing task.

[0182] As one implementation method, the access network device can also receive third indication information from the terminal device. This third indication information is used to indicate the amount of data for the first computing task. Thus, the access network device can schedule the data of the first computing task according to the third indication information, that is, determine the resources used to send the data of the first computing task and prepare the corresponding resources in advance to ensure that the data of the first computing task can be scheduled and transmitted as soon as possible, thereby ensuring the end-to-end latency of the first computing task.

[0183] Based on the above scheme, the computing management function network element determines multiple quality of service (QoS) files for the first QoS stream corresponding to the first computing task. Each QoS file includes one of multiple different uplink transmission delays, and the uplink transmission delays included in different QoS files are different. Then, the computing management function network element instructs the access network device to select one of the multiple QoS files based on the waiting time for sending the data of the first computing task, and uses the selected QoS file to send the data of the first computing task. This ensures that the data of the first computing task can be processed immediately or as quickly as possible after arriving at the computing node, reducing queuing waiting time and reducing the generation time of the first token of the first computing task, thereby ensuring the end-to-end latency of the first computing task.

[0184] The embodiments of Figures 2(a) and 2(b) described above will be explained below with reference to the examples in Figures 3, 4 and 5.

[0185] Figure 3 is a flowchart illustrating a communication method provided in an embodiment of this application. This method is applicable to application scenarios where a single terminal device has multiple computing tasks that need to be executed concurrently, and each computing task can be a periodic computing task.

[0186] The method includes the following steps:

[0187] Step 300: The terminal device completes the attachment process.

[0188] During the attach process, the terminal device receives an attach response message from the AMF, which may include an identifier of the allowed computing tasks for the terminal device.

[0189] Step 301: The terminal device sends a compute plane connection request to the AMF. Correspondingly, the AMF receives the compute plane connection request.

[0190] The compute plane connection request includes an identifier of the terminal device. Optionally, the compute plane connection request also includes an identifier of the requested compute task. Optionally, the compute plane connection request also includes the requested end-to-end latency corresponding to the requested compute task. The requested compute task is included among the allowed compute tasks. The requested end-to-end latency refers to the time delay between the terminal device sending the data for the requested compute task and the terminal device receiving the compute result of the compute task.

[0191] For example, the requested computing tasks include computing task 1 and computing task 2, and the end-to-end latency corresponding to the requested computing tasks includes end-to-end latency #1 and end-to-end latency #2. Here, end-to-end latency #1 refers to the latency between the terminal device sending data for computing task 1 and the terminal device receiving the computing result of computing task 1, and end-to-end latency #2 refers to the latency between the terminal device sending data for computing task 2 and the terminal device receiving the computing result of computing task 2.

[0192] The compute plane connection request can be a new message name used to create a connection between the terminal device and the compute node, or it can be an enhancement of an existing PDU session creation request. This invention does not limit the scope of the request.

[0193] Step 302: The AMF sends a computational plane connection request to the CMF. Correspondingly, the CMF receives the computational plane connection request.

[0194] This compute plane connection request is the same as the compute plane connection request in step 301, meaning the AMF forwards the compute plane connection request received from the terminal device to the CMF. Optionally, the AMF can select a CMF for forwarding based on the location information of the terminal device.

[0195] Step 303, optionally, the CMF sends a compute management association establishment request to the PCF. Correspondingly, the PCF receives the compute management association establishment request.

[0196] The optional parameters for establishing a computation plane management association include the identifier of the requested computation task and the end-to-end latency of the request corresponding to the requested computation task.

[0197] Step 304: PCF determines the PCC rule.

[0198] In one possible implementation, the PCF determines the PCC rules based on the local configuration.

[0199] In one possible implementation, the PCF obtains the terminal device's subscription data from the UDM. This subscription data includes the identifier of the allowed computing tasks of the terminal device and the allowed end-to-end latency corresponding to the allowed computing tasks. The allowed end-to-end latency refers to the time delay between the terminal device sending the data corresponding to the allowed computing task and the terminal device receiving the computing result of that task.

[0200] PCF determines PCC rules based on the identifier of the allowed computing task of the terminal device and the allowed end-to-end latency corresponding to the allowed computing task, as well as the identifier of the requested computing task and the requested end-to-end latency corresponding to the requested computing task. The PCC rules include the identifier of the requested computing task, the end-to-end latency corresponding to the requested computing task, and the periodic information of the requested computing task.

[0201] For example, for any computing task in the requested computing task (taking the first computing task as an example), if the end-to-end latency of the request corresponding to the first computing task is greater than or equal to the allowed end-to-end latency corresponding to the first computing task, then the PCC rule includes the identifier of the first computing task and the latency requirement of the request corresponding to the identifier of the first computing task.

[0202] For example, for any computing task in the requested computing task (taking the first computing task as an example), if the end-to-end latency of the request corresponding to the first computing task is less than the allowed end-to-end latency corresponding to the first computing task, then the PCC rule includes the identifier of the first computing task and the allowed latency requirement corresponding to the identifier of the first computing task.

[0203] Step 305, optionally, the PCF sends a compute management association establishment response to the CMF. Correspondingly, the PCF receives the compute management association establishment response.

[0204] The computational plane management association establishes responses including PCC rules.

[0205] Alternatively, steps 301 to 303 can be omitted. Instead, the AF sends a compute plane connection request to the PCF via the NEF. This request includes one or more of the following: the identifier of the terminal device, the identifier of the requested compute task, and the requested end-to-end latency corresponding to the compute task. This triggers the PCF to execute steps 304 and 305.

[0206] As another implementation method, steps 301 to 302 above can be omitted. Instead, the AF sends a computing plane connection request to the CMF via the NEF. The computing plane connection request includes one or more of the following: the identifier of the terminal device, the identifier of the requested computing task, and the end-to-end delay corresponding to the requested computing task. This triggers the CMF to execute step 303 and the PCF to execute steps 304 and 305.

[0207] Step 306: CMF determines the uplink and downlink transmission delays of the computation task requested in the PCC rule between the computing node and the terminal device.

[0208] Based on the cycle of the requested computation tasks indicated by the PCC rule, the CMF determines that the occurrence times of multiple computation tasks are relatively close, i.e., the difference in the occurrence times of the multiple computation tasks is less than a preset time threshold. Then, it determines the uplink and downlink transmission delays of the computation tasks requested in the PCC rule between the computing node and the terminal device. Specifically, for the multiple computation tasks indicated by the PCC rule, at least two of the computation tasks have different uplink transmission delays. For example, the uplink transmission delay of the first computation task is different from that of the second computation task.

[0209] For example, for a computation task in a PCC rule, the computation latency of the computation task on the computing node can be determined first. Then, based on the end-to-end latency of the computation task and the computation latency of the computation task on the computing node, the uplink transmission latency of the computation task between the computing node and the terminal device can be determined. After determining the uplink transmission latency, the downlink transmission latency of the computation task between the computing node and the terminal device can be determined based on the end-to-end latency of the computation task, the computation latency of the computation task on the computing node, and the uplink transmission latency of the computation task between the computing node and the terminal device.

[0210] The computation latency of the requested computing task on the computing node can be either the maximum computation latency of the task on the computing node or the computation latency of the task on the computing node under a certain task partitioning method. The computation latency of the same computing task on the computing node varies under different partitioning methods. Furthermore, the computation latency of the computing task on the computing node when the computing task is partitioned is less than the computation latency of the computing task on the computing node when the task is not partitioned (i.e., the maximum computation latency). This is because after the computing task is partitioned, it is jointly executed by the terminal device and the computing node, thus reducing the computation latency of the computing node.

[0211] Step 307: The CMF sends a compute plane connection configuration message to the compute node. Correspondingly, the compute node receives the compute plane connection configuration message.

[0212] The compute plane connection configuration message includes one or more of the following: first rule, tunnel information of the access network device, address of the terminal device (e.g., IP address), task partitioning method corresponding to the requested compute task, or compute latency corresponding to the requested compute task.

[0213] The first rule is used to indicate the correspondence between at least one QFI and at least one quintuple of information, and / or to indicate the correspondence between at least one QFI and the identifier of the requested computation task.

[0214] Optionally, the tunnel information of the access network device is at the terminal device level.

[0215] In step 308, the CMF sends a computing plane connection configuration message to the access network device via the AMF. Correspondingly, the access network device receives the computing plane connection configuration message.

[0216] The compute plane connection configuration message includes one or more of the following: tunnel information of the compute node, a QoS profile, or the compute latency of the requested compute task on the compute node. A QFI is used to uniquely identify a QoS flow. The QoS profile includes the QFI and QoS parameters, such as latency parameters. These latency parameters indicate the uplink or downlink transmission latency of data packets for transmitting compute tasks via QoS flows between the terminal device and the user plane network element, or they indicate the uplink or downlink transmission latency of data packets for transmitting compute tasks via QoS flows between the terminal device and the compute node (e.g., deployed within the access network equipment).

[0217] Optionally, the tunnel information of the computing node is at the terminal device granularity.

[0218] In step 309, the CMF sends a compute plane connection configuration message to the terminal device via the AMF. Correspondingly, the terminal device receives the compute plane connection configuration message.

[0219] The compute plane connection configuration message includes one or more of the following: QoS rules, indication information, task splitting method corresponding to the requested compute task, or address of the compute node (e.g., IP address).

[0220] The QoS rule is used to indicate the correspondence between at least one QFI and at least one 5-tuple information, and / or to indicate the correspondence between at least one QFI and the identifier of the requested computing task.

[0221] This instruction information is used to instruct the terminal device to carry the identifier of the computing task in the header of the data packet of the computing task.

[0222] Step 310: The access network device sends an RRC configuration message to the terminal device. Correspondingly, the terminal device receives the RRC configuration message.

[0223] This RRC configuration message is used to configure air interface resources for the terminal device.

[0224] After completing the above configurations, the terminal device, access network device, and computing node can transmit uplink data packets and / or downlink data packets.

[0225] In the uplink direction, when a terminal device needs to send uplink data packets for a computing task, it determines the corresponding QFI based on QoS rules and the five-tuple information (or the identifier of the computing task corresponding to the uplink data packet), and adds the QFI to the header of the uplink data packet, thus mapping the uplink data packet to the corresponding QoS stream for transmission. The terminal device sends the uplink data packet to the access network device, and the access network device performs corresponding QoS processing on the uplink data packet according to the QoS parameters of the QoS file corresponding to the QoS stream. In one possible implementation, when the access network device and the computing node are directly connected, the access network device sends the uplink data packet to the computing node through the tunnel corresponding to the uplink data packet between the access network device and the computing node.

[0226] In the downlink direction, the compute node receives the uplink data packet, processes it accordingly to obtain the computation result, and carries the computation result in the downlink data packet. In one possible implementation, the compute node further determines the corresponding QFI based on the first rule and the five-tuple information corresponding to the downlink data packet (or the identifier of the computation task corresponding to the downlink data), and adds the QFI to the header of the downlink data packet, thus mapping the downlink data packet to the corresponding QoS stream for transmission. In another possible implementation, when the access network device and the compute node are directly connected, the compute node sends the downlink data packet to the access network device through the tunnel corresponding to the downlink data packet between the compute node and the access network device. In this scenario, the header of the downlink data packet carries the QFI, and the body of the downlink data packet carries the computation result. The access network device determines the QoS file corresponding to the QFI based on the downlink data packet and performs corresponding QoS processing on the downlink data packet according to the QoS file. The access network device then sends the downlink data packet to the terminal device, thereby enabling the terminal device to receive the downlink data packet.

[0227] Based on the above scheme, multiple computing tasks from a single terminal device are executed by the same computing node. If the occurrence times of these multiple computing tasks are relatively close, i.e., the difference in their occurrence times is less than a preset time threshold, then by configuring different uplink transmission delays for these multiple computing tasks, the data of these multiple computing tasks arrive at the computing node in sequence and are executed by the computing node in order. After one computing task is completed or nearly completed, the next computing task arrives at the computing node and is executed by the computing node. Since the data of these multiple computing tasks do not arrive at the computing node at the same or similar times, each computing task does not need to wait for a long time after arriving at the computing node before it can be executed, or the computing node does not need to process multiple computing tasks simultaneously. This reduces the generation time of the first token for each computing task, improves the computing efficiency of the computing node, and thus helps to ensure end-to-end latency.

[0228] Figure 4 is a flowchart illustrating a communication method provided in an embodiment of this application. This method is applicable to application scenarios where multiple terminal devices have computational tasks that need to be executed concurrently, and each computational task can be a periodic task. Each terminal device hosts one or more computational tasks. For example, terminal device 1 hosts computational task 1 to be processed, while terminal device 2 hosts computational tasks 2 and 3 to be processed. These computational tasks are concurrent, meaning they occur within the same or similar timeframes.

[0229] The method includes the following steps:

[0230] Step 400: The terminal device completes the attachment process.

[0231] During the attach process, the terminal device receives an attach response message from the AMF, which may include an identifier of the allowed computing tasks for the terminal device.

[0232] Step 401: The terminal device sends a compute plane connection request to the AMF. Correspondingly, the AMF receives the compute plane connection request.

[0233] The compute plane connection request includes an identifier of the terminal device. Optionally, the compute plane connection request also includes an identifier of the requested compute task. Optionally, the compute plane connection request also includes the requested end-to-end latency corresponding to the requested compute task. The requested compute task is included among the allowed compute tasks. The requested end-to-end latency refers to the time delay between the terminal device sending the data for the requested compute task and the terminal device receiving the compute result of the compute task.

[0234] For example, the requested computing tasks include computing task 1 and computing task 2, and the end-to-end latency corresponding to the requested computing tasks includes end-to-end latency #1 and end-to-end latency #2. Here, end-to-end latency #1 refers to the latency between the terminal device sending data for computing task 1 and the terminal device receiving the computing result of computing task 1, and end-to-end latency #2 refers to the latency between the terminal device sending data for computing task 2 and the terminal device receiving the computing result of computing task 2.

[0235] Step 402: AMF sends a computational plane connection request to CMF. Correspondingly, CMF receives the computational plane connection request.

[0236] This compute plane connection request is the same as the compute plane connection request in step 401, which means that the AMF will forward the compute plane connection request received from the terminal device to the CMF.

[0237] Step 403, optionally, the CMF sends a computational plane management association establishment request to the PCF. Correspondingly, the PCF receives the computational plane management association establishment request.

[0238] The computation plane management association establishment request includes the identifier of the requested computation task and the end-to-end latency of the request corresponding to the requested computation task.

[0239] Step 404, optionally, PCF determines the PCC rule.

[0240] In one possible implementation, the PCF determines the PCC rules based on the local configuration.

[0241] In one possible implementation, the PCF obtains the terminal device's subscription data from the UDM. This subscription data includes the identifier of the allowed computing tasks of the terminal device and the allowed end-to-end latency corresponding to the allowed computing tasks. The allowed end-to-end latency refers to the time delay between the terminal device sending the data corresponding to the allowed computing task and the terminal device receiving the computing result of that task.

[0242] PCF determines PCC rules based on the identifier of the allowed computing task of the terminal device and the allowed end-to-end latency corresponding to the allowed computing task, as well as the identifier of the requested computing task and the requested end-to-end latency corresponding to the requested computing task. The PCC rules include the identifier of the requested computing task, the end-to-end latency corresponding to the requested computing task, and the period information of the requested computing task. The period information is used to indicate the period.

[0243] For example, for any computing task in the requested computing task (taking the first computing task as an example), if the end-to-end latency of the request corresponding to the first computing task is greater than or equal to the allowed end-to-end latency corresponding to the first computing task, then the PCC rule includes the identifier of the first computing task and the latency requirement of the request corresponding to the identifier of the first computing task.

[0244] For example, for any computing task in the requested computing task (taking the first computing task as an example), if the end-to-end latency of the request corresponding to the first computing task is less than the allowed end-to-end latency corresponding to the first computing task, then the PCC rule includes the identifier of the first computing task and the allowed latency requirement corresponding to the identifier of the first computing task.

[0245] Step 405, optionally, the PCF sends a compute plane management association establishment response to the CMF. Correspondingly, the PCF receives the compute plane management association establishment response.

[0246] The computational plane management association establishes responses including PCC rules.

[0247] Alternatively, steps 401 to 403 can be omitted. Instead, the AF sends a compute plane connection request to the PCF via the NEF. This request includes one or more of the following: the identifier of the terminal device, the identifier of the requested compute task, and the requested end-to-end latency corresponding to the compute task. This triggers the PCF to execute steps 404 and 405.

[0248] As another implementation method, steps 401 to 402 above can be omitted. Instead, the AF sends a computing plane connection request to the CMF via the NEF. The computing plane connection request includes one or more of the following: the identifier of the terminal device, the identifier of the requested computing task, and the end-to-end delay of the requested computing task. This triggers the CMF to execute step 403, and the PCF to execute steps 404 and 405.

[0249] Steps 401 to 405 described above are the process for one terminal device. When there are multiple terminal devices, steps 401 to 405 are executed for each terminal device.

[0250] The above steps are for a single terminal device. When multiple terminal devices are involved, the above steps are repeated and will not be elaborated further.

[0251] Step 406a: The CMF determines the computing nodes corresponding to the computing tasks requested by each terminal device. In other words, the CMF determines the computing tasks from one or more terminal devices that are executed on the computing nodes.

[0252] When CMF processes computation tasks requested from multiple terminal devices, that is, for each terminal request, it determines the computation node to execute the computation task. Different computation tasks from the same terminal device can be executed by the same computation node or by different computation nodes.

[0253] For example, terminal device 1 requests computing tasks including computing task 1, computing task 2, and computing task 3, while terminal device 2 requests computing tasks including computing task 4 and computing task 5. CMF determines that computing node 1 executes computing task 1, computing task 2, and computing task 4, and computing node 1 executes computing task 3 and computing task 5.

[0254] Step 406b: CMF determines the uplink and downlink transmission delays between the computing node and the terminal device for the computation tasks requested in the PCC rules corresponding to each terminal device.

[0255] Based on the period of the computation tasks requested by each terminal device according to the PCC rules, the CMF determines that the occurrence times of the computation tasks requested by different terminal devices are relatively close, that is, the difference in the occurrence times of the computation tasks requested by different terminal devices is less than a preset time threshold. Therefore, it determines the uplink and downlink transmission delays between the computing node and the terminal devices for the computation tasks requested by different terminal devices. These computation tasks requested by different terminal devices are executed by the same computing node. For example, the computation tasks requested by different terminal devices here include computation task 3 and computation task 5 in the example above.

[0256] For computing tasks requested by different terminal devices, there are at least two computing tasks with different uplink transmission delays. For example, the uplink transmission delay of the first computing task corresponding to terminal device 1 is different from the uplink transmission delay of the second computing task corresponding to terminal device 2.

[0257] For example, for a computation task in a PCC rule, the computation latency of the computation task on the computing node can be determined first. Then, based on the end-to-end latency of the computation task and the computation latency of the computation task on the computing node, the uplink transmission latency of the computation task between the computing node and the terminal device can be determined. After determining the uplink transmission latency, the downlink transmission latency of the computation task between the computing node and the terminal device can be determined based on the end-to-end latency of the computation task, the computation latency of the computation task on the computing node, and the uplink transmission latency of the computation task between the computing node and the terminal device.

[0258] The computation latency of the requested computing task on the computing node can be either the maximum computation latency of the task on the computing node or the computation latency of the task on the computing node under a certain task partitioning method. The computation latency of the same computing task on the computing node varies under different partitioning methods. Furthermore, the computation latency of the computing task on the computing node when the computing task is partitioned is less than the computation latency of the computing task on the computing node when the task is not partitioned (i.e., the maximum computation latency). This is because after the computing task is partitioned, it is jointly executed by the terminal device and the computing node, thus reducing the computation latency of the computing node.

[0259] In step 407, the CMF sends a compute plane connection configuration message to the compute node. Correspondingly, the compute node receives the compute plane connection configuration message.

[0260] The compute plane connection configuration message includes one or more of the following: first rule, tunnel information of the access network device, address of the terminal device (e.g., IP address), task partitioning method corresponding to the requested compute task, or compute latency corresponding to the requested compute task.

[0261] The first rule is used to indicate the correspondence between at least one QFI and at least one quintuple of information, and / or to indicate the correspondence between at least one QFI and the identifier of the requested computation task.

[0262] Optionally, the tunnel information of the access network device is at the terminal device level.

[0263] In step 408, the CMF sends a computing plane connection configuration message to the access network device via the AMF. Correspondingly, the access network device receives the computing plane connection configuration message.

[0264] The compute plane connection configuration message includes one or more of the following: tunnel information of the compute node, a QoS profile, or the compute latency of the requested compute task on the compute node. A QFI is used to uniquely identify a QoS flow. The QoS profile includes the QFI and QoS parameters, such as latency parameters. These latency parameters indicate the uplink or downlink transmission latency of data packets for transmitting compute tasks via QoS flows between the terminal device and user plane network elements, or they indicate the uplink or downlink transmission latency of data packets for transmitting compute tasks via QoS flows between the terminal device and a compute node (e.g., deployed within the network).

[0265] Optionally, the tunnel information of the computing node is at the terminal device granularity.

[0266] In step 409, the CMF sends a compute plane connection configuration message to the terminal device via the AMF. Correspondingly, the terminal device receives the compute plane connection configuration message.

[0267] The compute plane connection configuration message includes one or more of the following: QoS rules, indication information, task splitting method corresponding to the requested compute task, or address of the compute node (e.g., IP address).

[0268] The QoS rule is used to indicate the correspondence between at least one QFI and at least one 5-tuple information, and / or to indicate the correspondence between at least one QFI and the identifier of the requested computing task.

[0269] This instruction information is used to instruct the terminal device to carry the identifier of the computing task in the header of the data packet of the computing task.

[0270] Step 410: The access network device sends an RRC configuration message to the terminal device. Correspondingly, the terminal device receives the RRC configuration message.

[0271] This RRC configuration message is used to configure air interface resources for the terminal device.

[0272] After completing the above configurations, the terminal device, access network device, and computing node can transmit uplink data packets and / or downlink data packets.

[0273] In the uplink direction, when a terminal device needs to send uplink data packets for a computing task, it determines the corresponding QFI based on QoS rules and the five-tuple information (or the identifier of the computing task corresponding to the uplink data packet), and adds the QFI to the header of the uplink data packet, thus mapping the uplink data packet to the corresponding QoS stream for transmission. The terminal device sends the uplink data packet to the access network device, and the access network device performs corresponding QoS processing on the uplink data packet according to the QoS parameters of the QoS file corresponding to the QoS stream. In one possible implementation, when the access network device and the computing node are directly connected, the access network device sends the uplink data packet to the computing node through the tunnel corresponding to the uplink data packet between the access network device and the computing node.

[0274] In the downlink direction, the compute node receives the uplink data packet, processes it accordingly to obtain the computation result, and carries the computation result in the downlink data packet. In one possible implementation, the compute node further determines the corresponding QFI based on the first rule and the five-tuple information corresponding to the downlink data packet (or the identifier of the computation task corresponding to the downlink data), and adds the QFI to the header of the downlink data packet, thus mapping the downlink data packet to the corresponding QoS stream for transmission. In another possible implementation, when the access network device and the compute node are directly connected, the compute node sends the downlink data packet to the access network device through the tunnel corresponding to the downlink data packet between the compute node and the access network device. In this scenario, the header of the downlink data packet carries the QFI, and the body of the downlink data packet carries the computation result. The access network device determines the QoS file corresponding to the QFI based on the downlink data packet and performs corresponding QoS processing on the downlink data packet according to the QoS file. The access network device then sends the downlink data packet to the terminal device, thereby enabling the terminal device to receive the downlink data packet.

[0275] Steps 407 to 410 described above are the process for one terminal device. When there are multiple terminal devices, steps 407 to 410 are executed for each terminal device.

[0276] Based on the above scheme, multiple computing tasks from multiple terminal devices are executed by a single computing node. If the occurrence times of these multiple computing tasks are relatively close, i.e., the difference in their occurrence times is less than a preset time threshold, then by configuring different uplink transmission delays for these multiple computing tasks, the data of these multiple computing tasks arrive at the computing node in sequence and are executed by the computing node in order. Once one computing task is completed or nearly completed, the next computing task arrives at the computing node and is executed by the computing node. Because the data of these multiple computing tasks do not arrive at the computing node at the same or similar times, each computing task can be executed by the computing node without waiting for a long time, or the computing node does not need to process multiple computing tasks simultaneously. This reduces the generation time of the first token for each computing task, improves the computing efficiency of the computing node, and thus helps to ensure end-to-end latency.

[0277] Figure 5 is a flowchart illustrating a communication method provided in an embodiment of this application. This method is applicable to application scenarios where a single terminal device has multiple non-periodic computing tasks, and also applicable to application scenarios where multiple terminal devices have non-periodic concurrent computing tasks.

[0278] The method includes the following steps:

[0279] Step 500: The terminal device completes the attachment process.

[0280] During the attach process, the terminal device receives an attach response message from the AMF, which may include an identifier of the allowed computing tasks for the terminal device.

[0281] Step 501: The terminal device sends a compute plane connection request to the AMF. Correspondingly, the AMF receives the compute plane connection request.

[0282] The compute plane connection request includes an identifier of the terminal device. Optionally, the compute plane connection request also includes an identifier of the requested compute task. Optionally, the compute plane connection request also includes the requested end-to-end latency corresponding to the requested compute task. The requested compute task is included among the allowed compute tasks. The requested end-to-end latency refers to the time delay between the terminal device sending the data for the requested compute task and the terminal device receiving the compute result of the compute task.

[0283] For example, the requested computing tasks include computing task 1 and computing task 2, and the end-to-end latency corresponding to the requested computing tasks includes end-to-end latency #1 and end-to-end latency #2. Here, end-to-end latency #1 refers to the latency between the terminal device sending data for computing task 1 and the terminal device receiving the computing result of computing task 1, and end-to-end latency #2 refers to the latency between the terminal device sending data for computing task 2 and the terminal device receiving the computing result of computing task 2.

[0284] Step 502: The AMF sends a computational plane connection request to the CMF. Correspondingly, the CMF receives the computational plane connection request.

[0285] This compute plane connection request is the same as the compute plane connection request in step 501, which means that the AMF will forward the compute plane connection request received from the terminal device to the CMF.

[0286] Step 503, optionally, the CMF sends a computational plane management association establishment request to the PCF. Correspondingly, the PCF receives the computational plane management association establishment request.

[0287] The computation plane management association establishment request includes the identifier of the requested computation task and the end-to-end latency of the request corresponding to the requested computation task.

[0288] Step 504, optionally, PCF determines the PCC rule.

[0289] In one possible implementation, the PCF determines the PCC rules based on the local configuration.

[0290] In one possible implementation, the PCF obtains the terminal device's subscription data from the UDM. This subscription data includes the identifier of the allowed computing tasks of the terminal device and the allowed end-to-end latency corresponding to the allowed computing tasks. The allowed end-to-end latency refers to the time delay between the terminal device sending the data corresponding to the allowed computing task and the terminal device receiving the computing result of that task.

[0291] PCF determines PCC rules based on the identifier of the allowed computing task of the terminal device and the allowed end-to-end latency corresponding to the allowed computing task, as well as the identifier of the requested computing task and the requested end-to-end latency corresponding to the requested computing task. The PCC rules include the identifier of the requested computing task, the end-to-end latency corresponding to the requested computing task, and the periodicity information of the requested computing task. The periodicity information is used to indicate that the requested computing task is a non-periodic computing task.

[0292] For example, for any computing task in the requested computing task (taking the first computing task as an example), if the end-to-end latency of the request corresponding to the first computing task is greater than or equal to the allowed end-to-end latency corresponding to the first computing task, then the PCC rule includes the identifier of the first computing task and the latency requirement of the request corresponding to the identifier of the first computing task.

[0293] For example, for any computing task in the requested computing task (taking the first computing task as an example), if the end-to-end latency of the request corresponding to the first computing task is less than the allowed end-to-end latency corresponding to the first computing task, then the PCC rule includes the identifier of the first computing task and the allowed latency requirement corresponding to the identifier of the first computing task.

[0294] Step 505, optionally, the PCF sends a compute plane management association establishment response to the CMF. Correspondingly, the PCF receives the compute plane management association establishment response.

[0295] The computational plane management association establishes responses including PCC rules.

[0296] Alternatively, steps 501 to 503 can be omitted. Instead, the AF sends a compute plane connection request to the PCF via the NEF. This request includes one or more of the following: the identifier of the terminal device, the identifier of the requested compute task, and the end-to-end latency corresponding to the requested compute task. This triggers the PCF to execute steps 504 and 505.

[0297] As another implementation method, steps 501 to 502 above can be omitted. Instead, the AF sends a computing plane connection request to the CMF via the NEF. The computing plane connection request includes one or more of the following: the identifier of the terminal device, the identifier of the requested computing task, and the end-to-end delay corresponding to the requested computing task. This triggers the CMF to execute step 503 and the PCF to execute steps 504 and 505.

[0298] Step 506: CMF determines the multiple uplink transmission delays and multiple downlink transmission delays for each requested computation task in the PCC rule between the compute node and the terminal device.

[0299] If the computation task requested by CMF according to the PCC rule is an aperiodic computation task, then CMF determines multiple uplink transmission delays and corresponding multiple downlink transmission delays between the computing node and the terminal device for the computation task requested in the PCC rule. These multiple uplink transmission delays are all distinct from each other.

[0300] For example, for computation task 1, CMF determines several different uplink transmission delays between the terminal device and the computation node. For instance, the end-to-end delay for computation task 1 is 500ms, the computation delay for computation task 1 is 100ms, and the following combinations of uplink and downlink transmission delays for computation task 1 are determined:

[0301] (1) Uplink transmission delay: 100ms, downlink transmission delay: 300ms;

[0302] (2) Uplink transmission delay: 200ms, downlink transmission delay: 200ms;

[0303] (3) Uplink transmission delay: 300ms, downlink transmission delay: 100ms.

[0304] For example, for a computation task in a PCC rule, the computation latency of the computation task on the computing node can be determined first. Then, based on the end-to-end latency of the computation task and the computation latency of the computation task on the computing node, multiple uplink transmission latencies between the computing node and the terminal device can be determined. After determining the multiple uplink transmission latencies, multiple downlink transmission latencies between the computing node and the terminal device can be determined based on the end-to-end latency of the computation task, the computation latency of the computation task on the computing node, and the multiple uplink transmission latencies of the computation task between the computing node and the terminal device.

[0305] The computation latency of the requested computing task on the computing node can be either the maximum computation latency of the task on the computing node or the computation latency of the task on the computing node under a certain task partitioning method. The computation latency of the same computing task on the computing node varies under different partitioning methods. Furthermore, the computation latency of the computing task on the computing node when the computing task is partitioned is less than the computation latency of the computing task on the computing node when the task is not partitioned (i.e., the maximum computation latency). This is because after the computing task is partitioned, it is jointly executed by the terminal device and the computing node, thus reducing the computation latency of the computing node.

[0306] Step 507: The CMF sends a compute plane connection configuration message to the compute node. Correspondingly, the compute node receives the compute plane connection configuration message.

[0307] The compute plane connection configuration message includes one or more of the following: first rule, tunnel information of the access network device, address of the terminal device (e.g., IP address), task partitioning method corresponding to the requested compute task, or compute latency corresponding to the requested compute task.

[0308] The first rule is used to indicate the correspondence between at least one QFI and at least one quintuple of information, and / or to indicate the correspondence between at least one QFI and the identifier of the requested computation task.

[0309] Optionally, the tunnel information of the access network device is at the terminal device level.

[0310] In step 508, the CMF sends a computing plane connection configuration message to the access network device via the AMF. Correspondingly, the access network device receives the computing plane connection configuration message.

[0311] The compute plane connection configuration message includes one or more of the following: tunnel information of the compute node, multiple QoS profiles, indication information, or the compute latency of the requested compute task on the compute node. A QFI is used to uniquely identify a QoS flow. Each QoS profile includes a QFI and QoS parameters, such as latency parameters. These latency parameters indicate the uplink or downlink transmission latency of data packets for transmitting compute tasks via a QoS flow between the terminal device and user plane network elements, or the uplink or downlink transmission latency of data packets for transmitting compute tasks via a QoS flow between the terminal device and a compute node (e.g., deployed within the network). Latency parameters in different QoS profiles for the same QoS flow correspond to different uplink transmission latencies. Furthermore, different QoS profiles for the same QoS flow contain the same QFI, which indicates the QoS flow.

[0312] For example, CMF determines three QoS files for the QoS stream corresponding to computation task 1. QoS file 1 has an uplink transmission latency of 100ms. The QoS parameters in QoS file 1 include latency parameter 1, which indicates the uplink latency 1 for computation task 1's transmission over the wireless network, and this uplink latency 1 ≤ 100ms. QoS file 2 has an uplink transmission latency of 200ms. The QoS parameters in QoS file 2 include latency parameter 2, which indicates the uplink latency 2 for computation task 1's transmission over the wireless network, and this uplink latency 2 ≤ 200ms. QoS file 3 has an uplink transmission latency of 300ms. The QoS parameters in QoS file 3 include latency parameter 3, which indicates the uplink latency 3 for computation task 1's transmission over the wireless network, and this uplink latency 3 ≤ 300ms.

[0313] The above instruction information is used to instruct the access network device to select one of the multiple QoS files of the QoS stream based on the waiting time for sending the data of the computing task.

[0314] Optionally, the tunnel information of the computing node is at the terminal device granularity.

[0315] In step 509, the CMF sends a compute plane connection configuration message to the terminal device via the AMF. Correspondingly, the terminal device receives the compute plane connection configuration message.

[0316] The compute plane connection configuration message includes one or more of the following: QoS rules, indication information, task splitting method corresponding to the requested compute task, or address of the compute node (e.g., IP address).

[0317] The QoS rule is used to indicate the correspondence between at least one QFI and at least one 5-tuple information, and / or to indicate the correspondence between at least one QFI and the identifier of the requested computing task.

[0318] This instruction information is used to instruct the terminal device to carry the identifier of the computing task in the header of the data packet of the computing task.

[0319] Step 510: The access network device sends an RRC configuration message to the terminal device. Correspondingly, the terminal device receives the RRC configuration message.

[0320] This RRC configuration message is used to configure air interface resources for the terminal device.

[0321] After completing the above configurations, the terminal device, access network device, and computing node can transmit uplink data packets and / or downlink data packets.

[0322] In the uplink direction, when the terminal device needs to send uplink data packets for computing tasks, the terminal device determines the corresponding QFI based on the QoS rules and the five-tuple information corresponding to the uplink data packet (or the identifier of the computing task corresponding to the uplink data), and adds the QFI to the header of the uplink data packet to map the uplink data packet to the corresponding QoS stream for transmission. The terminal device sends the uplink data packet to the access network device. Based on the QFI in the uplink data packet, the access network device selects a QoS file from multiple QoS files corresponding to the QFI, and performs corresponding QoS processing on the uplink data packet based on the QoS parameters of the selected QoS file, so that the uplink data packet of the computing task is sent to the computing node according to the corresponding uplink transmission delay. For example, the specific implementation method of the access network device selecting a QoS file from multiple QoS files can be referred to the relevant description in step 205b of the embodiment of Figure 2(b) above, which will not be repeated here. In one possible implementation, when the access network device and the computing node are directly connected, the access network device sends the uplink data packet to the computing node through the tunnel between the access network device and the computing node corresponding to the uplink data packet.

[0323] In the downlink direction, the compute node receives the uplink data packet, processes it accordingly to obtain the computation result, and carries the computation result in the downlink data packet. In one possible implementation, the compute node further determines the corresponding QFI based on the first rule and the five-tuple information corresponding to the downlink data packet (or the identifier of the computation task corresponding to the downlink data), and adds the QFI to the header of the downlink data packet, thus mapping the downlink data packet to the corresponding QoS stream for transmission. In another possible implementation, when the access network device and the compute node are directly connected, the compute node sends the downlink data packet to the access network device through the tunnel corresponding to the downlink data packet between the compute node and the access network device. In this scenario, the header of the downlink data packet carries the QFI, and the body of the downlink data packet carries the computation result. The access network device determines the QoS file corresponding to the QFI based on the downlink data packet, and performs corresponding QoS processing on the downlink data packet according to the QoS file, thus ensuring that the downlink data packet of the computation task is sent to the terminal device according to the corresponding downlink transmission delay.

[0324] Based on the above scheme, CMF determines multiple QoS files for the QoS stream corresponding to the computing task. Each QoS file includes one of multiple uplink transmission delays, and the uplink transmission delays included in different QoS files are different. Then, CMF instructs the access network device to select one of the multiple QoS files based on the waiting time for sending the computing task data, and uses the selected QoS file to send the computing task data. This ensures that the computing task data can be processed immediately or as quickly as possible after arriving at the computing node, reducing queuing time and reducing the generation time of the first token of the computing task, thereby ensuring the end-to-end latency of the computing task.

[0325] Figure 6 illustrates a possible exemplary block diagram of the communication device involved in the embodiments of this application. As shown in Figure 6, the communication device 600 may include modules or units for implementing the methods described above. In one possible design, the communication device 600 includes a processing unit 602 and a communication unit 603. Optionally, the communication device 600 may further include a storage unit 601 for storing device program code and / or data.

[0326] The communication device 600 can also be a network-side device in the above embodiments, such as a network-side computing management function network element, a module (e.g., circuit, chip or chip system) in the computing management function network element, or a logic node, logic module or software that can implement all or part of the computing management function network element functions.

[0327] For example, in one embodiment, the processing unit 602 is configured to determine, based on the occurrence time of the first computing task and the occurrence time of the second computing task, that the difference between the occurrence time of the first computing task and the occurrence time of the second computing task is less than a preset time threshold; wherein, the first computing task and the second computing task are executed by the same computing node; determine, based on the first end-to-end latency corresponding to the first computing task and the first computing latency of the first computing task on the computing node, a first uplink transmission latency of the first computing task between the first terminal device and the computing node; and determine, based on the second end-to-end latency corresponding to the second computing task and the second computing task on the computing node, a first uplink transmission latency of the first computing task between the first terminal device and the computing node; and determine, based on the second end-to-end latency corresponding to the second computing task and the second computing task on the computing node, a second uplink transmission latency of the first computing task between the first terminal device and the computing node; and determine, based on the second end-to-end latency corresponding to the second computing task and the second computing task on the computing node, a second uplink transmission latency of the first computing task between the first terminal device and the computing node; and determine, based on the second end-to-end latency of ... The second computation delay on the computing node determines the second uplink transmission delay of the second computation task between the second terminal device and the computing node; wherein the first uplink transmission delay is different from the second uplink transmission delay; based on the first uplink transmission delay, the quality of service parameters in the quality of service file of the first quality of service stream corresponding to the first computation task are determined; and based on the second uplink transmission delay, the quality of service parameters in the quality of service file of the second quality of service stream corresponding to the second computation task are determined; the communication unit 603 is used to send the quality of service file of the first quality of service stream and the quality of service file of the second quality of service stream to the access network device.

[0328] In one possible implementation, the processing unit 602 is further configured to determine the occurrence time of the first computing task based on the cycle of the first computing task; and to determine the occurrence time of the second computing task based on the cycle of the second computing task.

[0329] In one possible implementation, the communication unit 603 is further configured to receive a first policy charging control rule and a second policy charging control rule from the policy control network element, wherein the first policy charging control rule indicates the period of the first computing task and the second policy charging control rule indicates the period of the second computing task; or, to receive a third policy charging control rule from the policy control network element, wherein the third policy charging control rule indicates the period of the first computing task and the period of the second computing task.

[0330] In one possible implementation, when the first calculation delay is less than the second calculation delay, the first uplink transmission delay is less than the second uplink transmission delay.

[0331] In one possible implementation, the processing unit 602 is further configured to determine a first downlink transmission delay of the first computing task between the first terminal device and the computing node based on the first end-to-end delay, the first computing delay, and the first uplink transmission delay; and to determine a second downlink transmission delay of the second computing task between the second terminal device and the computing node based on the second end-to-end delay, the second computing delay, and the second uplink transmission delay.

[0332] In one possible implementation, the processing unit 602 is further configured to determine that both the first computing task and the second computing task are processed by the computing node.

[0333] For example, in another embodiment, the processing unit 602 is configured to determine multiple different uplink transmission delays between the terminal device and the computing node based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node; and to determine multiple quality of service (QoS) files for the first QoS stream corresponding to the first computing task; wherein each QoS file includes one of the multiple different uplink transmission delays, and the different QoS files include different uplink transmission delays; the communication unit 603 is configured to send first indication information and the multiple QoS files to the access network device, wherein the first indication information is used to instruct the access network device to select one of the multiple QoS files based on the waiting time for sending the data of the first computing task.

[0334] In one possible implementation, the processing unit 602 is configured to determine multiple different uplink transmission delays between the terminal device and the computing node based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node, including: determining the multiple different uplink transmission delays based on the first end-to-end delay and the first computing delay when the first computing task is a non-periodic computing task.

[0335] In one possible implementation, the communication unit 603 is further configured to receive policy charging control rules from the policy control network element, the policy charging control rules being used to indicate that the first computing task is a non-periodic computing task.

[0336] In one possible implementation, the processing unit 602 is further configured to determine, based on the first end-to-end delay, the first computation delay, and the plurality of different uplink transmission delays, a plurality of different downlink transmission delays between the terminal device and the computing node.

[0337] The communication device 600 can also be a network-side device in the above embodiments, such as a network-side access network device, a module (e.g., circuit, chip or chip system) in the access network device, or a logic node, logic module or software that can implement all or part of the functions of the access network device.

[0338] For example, in one embodiment, the communication unit 603 is configured to receive first indication information and multiple quality of service (QoS) files of a first QoS stream corresponding to a first computing task; wherein each QoS file includes one of the multiple different uplink transmission delays, and the uplink transmission delays included in different QoS files are different; the first indication information is used to instruct the access network device to select one of the multiple QoS files based on the waiting time for sending the data of the first computing task; the processing unit 602 is configured to select one QoS file from the multiple QoS files of the first QoS stream based on the waiting time for sending the data of the first computing task; the communication unit 603 is further configured to schedule the data of the first computing task based on the selected QoS file.

[0339] In one possible implementation, the communication unit 603 is further configured to receive second indication information from the computing node, the second indication information being used to indicate the waiting time.

[0340] In one possible implementation, the second indication information includes a first parameter and a second parameter corresponding to the second computing task, the second computing task being the computing task being processed by the computing node; the first parameter is used to indicate the time interval between two adjacent tokens that generate the second computing task, and the second parameter is used to indicate the number of remaining tokens for the generated second computing task; the processing unit 602 is further configured to determine the waiting time based on the first parameter and the second parameter.

[0341] In one possible implementation, the processing unit 602 is further configured to determine the waiting time based on the computation delay of the third computation task; wherein the third computation task is the computation task that most recently performed data transmission.

[0342] In one possible implementation, the communication unit 603 is further configured to receive third indication information from the terminal device, the third indication information being used to indicate the amount of data for the first computing task; the processing unit 602 is further configured to schedule the data of the first computing task according to the third indication information.

[0343] It is understood that the division of units in the above-described device is merely a logical functional division. One function can correspond to one functional unit, or two or more functions can be integrated into one functional unit. In actual implementation, all or some units can be integrated onto a single physical entity, or distributed across different physical entities. Furthermore, the aforementioned functional units can be implemented in hardware, software, or a combination of both. Whether a function is executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for specific applications, but such implementations should not be considered beyond the scope of this application.

[0344] In one example, the functional unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as: one or more application-specific integrated circuits (ASICs), or one or more central processing units (CPUs), one or more microcontroller units (MCUs), one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0345] In one example, storage unit 601 may include random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory and / or registers, etc.

[0346] Figure 7 illustrates a possible exemplary block diagram of a communication device according to an embodiment of this application. The communication device 700 shown in Figure 7 includes a processor 710 and an interface circuit 720. The processor 710 and the interface circuit 720 are coupled to each other. It is understood that the interface circuit 720 can be a transceiver or an input / output interface. Optionally, the communication device 700 may further include a memory 730 for storing instructions executed by the processor 710, or storing input data required for the processor 710 to execute instructions, or storing data generated after the processor 710 executes instructions.

[0347] When the communication device 700 is used to implement the above method embodiment, the processor 710 is used to implement the function of the processing unit 602, and the interface circuit 720 is used to implement the function of the communication unit 603.

[0348] It is understood that the processor in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0349] This application provides a chip (or chip system) including a processor for implementing any of the above-described method embodiments.

[0350] This application provides a computer-readable storage medium storing a computer program or instructions that, when executed, implement any of the above-described method embodiments.

[0351] This application provides a computer program product, which includes a computer program or instructions that, when executed, implement any of the above-described method embodiments.

[0352] This application provides a communication system, including the computing management function network element and access network device in the above method embodiments.

[0353] The method steps in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, compact disc read-only memory (CD-ROM), or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Furthermore, the ASIC can reside in a first network element or a store-and-forward terrestrial function network element. Alternatively, the processor and storage medium can exist as discrete components in access network equipment or terminal equipment.

[0354] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. A computer program is a set of instructions that directs each step of an action of an electronic computer or other device with message processing capabilities. It is typically written in a programming language and runs on a target architecture. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. 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 integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video optical disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium can be volatile or non-volatile, or it can include both types of storage media.

[0355] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0356] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates an "or" relationship between the preceding and following related objects; in the formulas of this application, the character " / " indicates a "division" relationship between the preceding and following related objects.

[0357] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers described above does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

[0358] The terms "system" and "network" in this application embodiment are used interchangeably. "At least one" refers to one or more, and "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, "at least one of A, B, or C" includes A, B, C, AB, AC, BC, or ABC; "at least one of A, B, and C" can also be understood as including A, B, C, AB, AC, BC, or ABC. Furthermore, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in this application embodiment are used to distinguish multiple objects and are not used to limit the order, sequence, priority, or importance of multiple objects.

[0359] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0360] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0361] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0362] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0363] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A communication method, characterized in that, The method includes: Based on the occurrence time of the first computing task and the occurrence time of the second computing task, it is determined that the difference between the occurrence time of the first computing task and the occurrence time of the second computing task is less than a preset time threshold; wherein, the first computing task and the second computing task are executed by the same computing node; Based on the first end-to-end latency corresponding to the first computing task and the first computing latency of the first computing task on the computing node, a first uplink transmission latency of the first computing task between the first terminal device and the computing node is determined; and based on the second end-to-end latency corresponding to the second computing task and the second computing latency of the second computing task on the computing node, a second uplink transmission latency of the second computing task between the second terminal device and the computing node is determined; wherein, the first uplink transmission latency and the second uplink transmission latency are different; Based on the first uplink transmission delay, determine the quality of service parameters in the quality of service file of the first quality of service stream corresponding to the first computing task; and based on the second uplink transmission delay, determine the quality of service parameters in the quality of service file of the second quality of service stream corresponding to the second computing task. Send the quality of service (QoS) files of the first QoS stream and the second QoS stream to the access network equipment.

2. The method as described in claim 1, characterized in that, The method further includes: The occurrence time of the first computing task is determined based on the cycle of the first computing task. The occurrence time of the second computing task is determined based on the cycle of the second computing task.

3. The method as described in claim 2, characterized in that, The method further includes: Receive a first policy charging control rule and a second policy charging control rule from the policy control network element, wherein the first policy charging control rule indicates the period of the first computing task, and the second policy charging control rule indicates the period of the second computing task; or, Receive a third policy billing control rule from the policy control network element, the third policy billing control rule indicating the period of the first computing task and the period of the second computing task.

4. The method according to any one of claims 1 to 3, characterized in that, If the first calculation delay is less than the second calculation delay, then the first uplink transmission delay is less than the second uplink transmission delay.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Based on the first end-to-end delay, the first computation delay, and the first uplink transmission delay, the first downlink transmission delay of the first computation task between the first terminal device and the computation node is determined; Based on the second end-to-end delay, the second computation delay, and the second uplink transmission delay, the second downlink transmission delay of the second computation task between the second terminal device and the computation node is determined.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: It is determined that both the first computing task and the second computing task are processed by the computing node.

7. A communication method, characterized in that, The method includes: Based on the first end-to-end latency corresponding to the first computing task and the first computing latency of the first computing task on the computing node, determine multiple different uplink transmission latencies of the first computing task between the terminal device and the computing node. Determine multiple quality of service (QoS) files for the first QoS stream corresponding to the first computing task; wherein each QoS file includes one of the multiple different uplink transmission delays, and the uplink transmission delays included in different QoS files are different; Send a first indication message and the plurality of quality of service (QoS) files to the access network device. The first indication message is used to instruct the access network device to select one of the plurality of QoS files based on the waiting time for sending the data of the first computing task.

8. The method as described in claim 7, characterized in that, The step of determining multiple different uplink transmission delays between the terminal device and the computing node based on the first end-to-end delay corresponding to the first computing task and the first computing delay of the first computing task on the computing node includes: If the first computing task is determined to be an aperiodic computing task, then the multiple different uplink transmission delays are determined based on the first end-to-end delay and the first computing delay.

9. The method as described in claim 8, characterized in that, The method further includes: The system receives policy charging control rules from the policy control network element, wherein the policy charging control rules indicate that the first computing task is a non-periodic computing task.

10. The method according to any one of claims 7 to 9, characterized in that, The method further includes: Based on the first end-to-end delay, the first computation delay, and the multiple different uplink transmission delays, the multiple different downlink transmission delays of the first computation task between the terminal device and the computation node are determined.

11. A communication method, characterized in that, The method includes: The system receives a first instruction information and multiple service quality files corresponding to a first service quality stream for a first computing task; wherein each service quality file in the multiple service quality files includes one of the multiple different uplink transmission delays, and the uplink transmission delays included in the different service quality files are different; the first instruction information is used to instruct the access network device to select one of the multiple service quality files based on the waiting time for sending the data of the first computing task. Based on the waiting time for sending the data of the first computing task, select one quality of service file from multiple quality of service files of the first quality of service stream; The data for the first computing task is scheduled according to the selected quality of service file.

12. The method as described in claim 11, characterized in that, The method further includes: Receive a second indication message from the computing node, the second indication message being used to indicate the waiting time.

13. The method as described in claim 12, characterized in that, The second indication information includes a first parameter and a second parameter corresponding to the second computing task, whereby the second computing task is the computing task being processed by the computing node. The first parameter is used to indicate the time interval between two adjacent tokens that generate the second computing task, and the second parameter is used to indicate the number of remaining tokens in the generated second computing task; The method further includes: The waiting time is determined based on the first parameter and the second parameter.

14. The method as described in claim 11, characterized in that, The method further includes: The waiting time is determined based on the computation delay of the third computation task; The third computing task is the computing task that most recently transmitted data.

15. The method according to any one of claims 11 to 14, characterized in that, The method further includes: Receive third indication information from the terminal device, the third indication information being used to indicate the amount of data for the first computing task; The data for the first computing task is scheduled according to the third instruction information.

16. A communication device, characterized in that, Includes a module for performing the method of any one of claims 1 to 6, or the method of any one of claims 7 to 10, or the method of any one of claims 11 to 15.

17. A communication device, characterized in that, The device includes a processor and an interface circuit, the processor being configured to communicate with other devices via the interface circuit to implement the method of any one of claims 1 to 6, or the method of any one of claims 7 to 10, or the method of any one of claims 11 to 15.

18. A computer program product, characterized in that, The computer program product includes instructions that, when executed, implement the method of any one of claims 1 to 6, or the method of any one of claims 7 to 10, or the method of any one of claims 11 to 15.

19. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions, which, when executed, implement the method of any one of claims 1 to 6, or the method of any one of claims 7 to 10, or the method of any one of claims 11 to 15.

20. A communication system, characterized in that, This includes network elements with computing and management functions, and access network equipment; The computing management function network element is used to implement the method described in any one of claims 1 to 6; The access network device is used to receive the quality of service (QoS) file of the first QoS stream and the QoS file of the second QoS stream.

21. A communication system, characterized in that, This includes network elements with computing and management functions, and access network equipment; The computing management function network element is used to implement the method described in any one of claims 7 to 10; The access network device is used to implement the method according to any one of claims 11 to 15.

Images