Communication method and apparatus

Abstract

The present application relates to the field of wireless communications. Provided are a communication method and apparatus. In the method, a network node can send a first scheduling result and a first quality-of-service requirement to at least one computing node, receive first feedback information from the at least one computing node, and determine a second scheduling result on the basis of the first feedback information, wherein the first scheduling result can indicate a first mode for executing a target task by means of the at least one computing node; the first feedback information can indicate that the quality of service of executing the target task on the basis of the first mode cannot meet the first quality-of-service requirement; and the second scheduling result indicates a second mode for executing the target task by means of the at least one computing node, and the second mode is different from the first mode. In the above process, the network node can update a scheduling result of the target task in a timely manner on the basis of feedback from the at least one computing node, such that the updated scheduling result is more suitable for the computing node, thereby ensuring the quality of service of the target task, and improving the user experience.

Description

Communication method and device

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on December 29, 2023, with application number 202311864243.5 and invention name “Communication Method and Device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of wireless communications, and in particular to communication methods and devices. Background Art

[0003] In a communication system, a radio access network (RAN) node can obtain a communication quality of service (QoS) profile and indicate the communication QoS profile to a terminal and an application server so that the terminal and the application server can determine the scheduling result of a target task corresponding to the communication QoS profile. Afterwards, the terminal and the application server can execute the target task according to the scheduling result of the target task to ensure the quality of data transmission. In the above method, since the scheduling result of the target task is associated with the communication QoS profile, the communication QoS can be guaranteed. Therefore, the above method can be used in scenarios where the scheduling result of the target task requires communication QoS.

[0004] However, with the development of communication technology, communication scenarios are becoming more and more complex. In these communication scenarios, if only the scheduling results of the target task have requirements for communication QoS, the scheduling results of the target task may not be suitable for the terminal or application server, thereby affecting the service quality of the entire task. Summary of the Invention

[0005] The present application provides a communication method and apparatus that can determine appropriate scheduling results for target tasks and ensure the service quality of the target tasks.

[0006] To achieve the above objectives, this application adopts the following technical solutions:

[0007] In a first aspect, a communication method is provided, which can be executed by a network node. The network node here can refer to the network node itself or a processor, module, logic node, chip, or chip system in the network node that implements the method.

[0008] The method includes: sending a first scheduling result and a first service quality requirement to at least one computing node, wherein the first scheduling result indicates a first way for the at least one computing node to perform a target task; receiving first feedback information from the at least one computing node, wherein the first feedback information indicates that the service quality of the at least one computing node performing the target task according to the first way cannot meet the first service quality requirement; and determining a second scheduling result based on the first feedback information, wherein the second scheduling result indicates a second way for the at least one computing node to perform the target task, and the second way is different from the first way.

[0009] Based on the method provided in the first aspect above, the network node can send a first scheduling result and a first service quality requirement to at least one computing node, and adjust the scheduling strategy of the target task in combination with the first feedback information sent by at least one computing node, so that the adjusted scheduling result (such as the second scheduling result) is more suitable for the computing node, so as to ensure the service quality of the target task and improve the user experience. Optionally, the computing power of the computing node (such as the available computing resources of the computing node) is usually dynamically changing, so the first service quality requirement can be designed as a service quality related to computing. In this way, the network node can adjust the scheduling strategy of the target task in a timely manner according to the changes in the computing power of at least one computing node, so that the adjusted scheduling result is more in line with the actual situation of the computing node, thereby ensuring the computing service quality of the target task.

[0010] In a possible implementation manner, the method further includes: sending first condition information, where the first condition information is used to indicate a condition for triggering the first feedback information.

[0011] Based on the possible implementation manner described above, a node (such as at least one computing node) that receives the first condition information may determine whether to send the first feedback information, or determine when to send the first feedback information.

[0012] In one possible implementation, the condition for triggering the first feedback information includes at least one of the following: the service quality of the at least one computing node performing the target task according to the first method does not meet the first service quality requirement; or, the service quality of the at least one computing node performing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to the first threshold value; or, the feedback time indicated by the first condition information arrives.

[0013] Based on the above possible implementations, if the above conditions include that the service quality of at least one computing node performing the target task according to the first method does not meet the first service quality requirement, then the at least one computing node may send first feedback information to the network node if the service quality of the target task performed according to the first method does not meet the first service quality requirement. If the above conditions include that the service quality of at least one computing node performing the target task according to the first method does not meet the first service quality requirement, and the deviation therebetween is greater than or equal to a first threshold value, then the at least one computing node may send first feedback information to the network node if the service quality of the target task performed according to the first method does not meet the first service quality requirement, and the deviation therebetween is greater than or equal to the first threshold value. If the above conditions include that the feedback time indicated by the first condition information arrives, then the at least one computing node may send the first feedback information to the network node if the feedback time indicated by the first condition information arrives.

[0014] In a possible implementation, the method further includes: determining the first quality of service requirement based on the computing quality of service requirement of the target task and the computing power of the at least one computing node.

[0015] Based on the above possible implementation methods, the network node can determine the first service quality requirement based on the computing service quality requirement of the target task and the computing power of at least one computing node, so that at least one computing node can determine the computing service quality it needs to provide when executing the target task according to the first method.

[0016] In one possible implementation, the computing service quality requirement of the target task or the first service quality requirement indicates at least one of the following: computing type requirement, computing amount requirement, computing delay requirement, computing energy consumption requirement or computing accuracy requirement.

[0017] Based on the above possible implementation methods, at least one computing node can determine one or more of the computing type, computing amount, computing delay, computing energy consumption or computing accuracy required to perform the target task according to the first method.

[0018] In one possible implementation, the method also includes: sending a second quality of service requirement to the communication node; receiving second feedback information from the communication node, the second feedback information indicating that the quality of service provided by the communication node cannot meet the second quality of service requirement; determining a second scheduling result based on the first feedback information, including: determining the second scheduling result based on the first feedback information and the second feedback information.

[0019] Based on the above possible implementation methods, the network node can send a second quality of service requirement to the communication node, and adjust the scheduling strategy of the target task in combination with the first feedback information sent by at least one computing node and the second feedback information sent by the communication node, so that the adjusted scheduling result (such as the second scheduling result) is more suitable for the computing node and the communication node to ensure the service quality of the target task and improve the user experience. Optionally, the state of the channel is usually dynamically changing, so the quality of service provided by the communication node is also dynamically changing, so the second quality of service requirement can be designed as a quality of service related to communication. In this way, the network node can also adjust the scheduling strategy of the target task in a timely manner in combination with the changes in the quality of service provided by the communication node, so that the adjusted scheduling result is more in line with the actual situation of the communication node, thereby ensuring the communication service quality of the target task.

[0020] In a possible implementation manner, the method further includes: sending second condition information, where the second condition information is used to indicate a condition for triggering the second feedback information.

[0021] Based on the above possible implementation manner, a node that receives the second condition information, such as a communication node, may determine whether to send the second feedback information, or determine when to send the second feedback information.

[0022] In one possible implementation, the condition for triggering the second feedback information includes at least one of the following: the quality of service provided by the communication node cannot meet the second quality of service requirement; or, the quality of service provided by the communication node cannot meet the second quality of service requirement, and the deviation between the two is greater than or equal to the second threshold value; or, the feedback time indicated by the second condition information arrives.

[0023] Based on the above possible implementations, if the above conditions include the quality of service provided by the communication node failing to meet the second quality of service requirement, the communication node may send second feedback information to the network node if the provided quality of service fails to meet the second quality of service requirement. If the above conditions include the quality of service provided by the communication node failing to meet the second quality of service requirement, and the deviation therebetween is greater than or equal to a second threshold value, the communication node may send second feedback information to the network node if the provided quality of service fails to meet the second quality of service requirement, and the deviation therebetween is greater than or equal to the second threshold value. If the above conditions include the arrival of the feedback time indicated by the second condition information, the communication node may send the second feedback information to the network node if the feedback time indicated by the second condition information arrives.

[0024] In a possible implementation, the method further includes: determining the second quality of service requirement according to a network state and a communication quality of service requirement of the target task.

[0025] Based on the above possible implementation manner, the network node may determine the second QoS requirement according to the network state and the QoS requirement of the target task, so that the communication node may determine the QoS of the communication service it needs to provide.

[0026] In one possible implementation, the communication service quality requirement of the target task or the second service quality requirement indicates at least one of the following: packet delay budget, guaranteed bit rate, guaranteed stream bit rate, maximum burst data volume, packet error rate or guaranteed transmission data size.

[0027] Based on the above possible implementation methods, the communication node can determine one or more of the packet delay budget, guaranteed bit rate, guaranteed flow bit rate, maximum burst data volume, packet error rate or guaranteed transmission data size that it needs to provide.

[0028] In one possible implementation, the method also includes: obtaining the network status, the service quality requirements of the target task, and the computing power of the at least one computing node, the service quality requirements of the target task including the communication service quality requirements of the target task and the computing service quality requirements of the target task; determining the first scheduling result based on the network status, the service quality requirements of the target task, and the computing power of the at least one computing node.

[0029] Based on the above possible implementation methods, the network node can determine the first scheduling result according to the network status, the service quality requirements of the target task and the computing power of at least one computing node, so that the communication node can provide corresponding communication service quality for the target task, at least one computing node can provide corresponding computing service quality for the target task, and the above communication service quality and computing service quality meet the service quality requirements of the target task.

[0030] In a possible implementation, obtaining the quality of service requirement of the target task includes: receiving a computing service request, where the computing service request includes quality of service requirement information of the target task.

[0031] Based on the above possible implementations, the network node may obtain the quality of service requirements of the target task from the computing service request.

[0032] In a possible implementation, obtaining the service quality requirement of the target task includes: receiving a computing service request, the computing service request including a service identifier corresponding to the target task; and obtaining the service quality requirement of the target task according to the service identifier.

[0033] Based on the above possible implementation, the network node can obtain the service identifier from the computing service request and obtain the service quality requirement of the target task based on the service identifier. For example, the network node can obtain the service quality requirement of the target task from the core network based on the service identifier.

[0034] In a possible implementation, the method further includes: sending the second scheduling result.

[0035] Based on the above possible implementation manner, the computing node that receives the second scheduling result can be enabled to execute the target task according to the second scheduling result.

[0036] In a possible implementation, the method is applied to a radio access network node, a centralized unit, a distributed unit, an intelligent controller of a radio access network, or a core network element.

[0037] Based on the above possible implementation methods, wireless access network nodes, centralized units, distributed units, intelligent controllers of wireless access networks or core network elements can execute the method provided by the first aspect, thereby improving the flexibility and diversity of the equipment deployed by the method provided by the first aspect.

[0038] In a possible implementation manner, the first manner instructs the at least one computing node to execute a first subtask in the target task; the second manner instructs the at least one computing node to execute a second subtask in the target task.

[0039] Based on the above possible implementations, before and after the scheduling policy of the target task is adjusted, at least one computing node may execute different subtasks to provide corresponding quality of service.

[0040] In a second aspect, a communication method is provided, which can be executed by a computing node. The computing node here can refer to the computing node itself, or a processor, module, logic node, chip, or chip system in the computing node that implements the method.

[0041] The method includes: receiving a first scheduling result and a first service quality requirement, wherein the first scheduling result indicates a first way for at least one computing node to execute a target task; executing the target task according to the first way; and sending first feedback information, wherein the first feedback information is used to indicate that the service quality of executing the target task according to the first way cannot meet the first service quality requirement.

[0042] Based on the method provided in the second aspect above, the computing node can send first feedback information, indicating that the service quality of executing the target task according to the first method cannot meet the first service quality requirement, so that the node that receives the first feedback information, such as the network node, adjusts the scheduling strategy of the target task according to the first feedback information.

[0043] In a possible implementation manner, the method further includes: receiving first condition information, where the first condition information is used to indicate a condition for triggering the first feedback information.

[0044] Based on the foregoing possible implementation manner, the computing node may determine whether to send the first feedback information, or determine when to send the first feedback information according to the first condition information.

[0045] In one possible implementation, the condition for triggering the first feedback information includes at least one of the following: the service quality of executing the target task according to the first method does not meet the first service quality requirement; or, the service quality of executing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to the first threshold value; or, the feedback time indicated by the first condition information arrives.

[0046] Based on the above possible implementations, if the above conditions include that the quality of service of the target task executed according to the first method does not meet the first quality of service requirement, the computing node may send first feedback information to the network node if the quality of service of the target task executed according to the first method does not meet the first quality of service requirement. If the above conditions include that the quality of service of the target task executed according to the first method does not meet the first quality of service requirement, and the deviation therebetween is greater than or equal to a first threshold value, the computing node may send first feedback information to the network node if the quality of service of the target task executed according to the first method does not meet the first quality of service requirement, and the deviation therebetween is greater than or equal to the first threshold value. If the above conditions include that the feedback time indicated by the first condition information has arrived, the computing node may send first feedback information to the network node if the feedback time indicated by the first condition information has arrived.

[0047] In a possible implementation, the first quality of service requirement indicates at least one of the following: a calculation type requirement, a calculation amount requirement, a calculation delay requirement, a calculation energy consumption requirement, or a calculation accuracy requirement.

[0048] Based on the above possible implementation methods, the computing node determines one or more of the computing type, computing amount, computing latency, computing energy consumption or computing accuracy required to perform the target task according to the first method.

[0049] In a possible implementation, the method further includes: sending a computing service request, where the computing service request includes at least one of service quality requirement information of the target task or a service identifier corresponding to the target task.

[0050] Based on the above possible implementation methods, a node that receives a computing service request, such as a network node, can obtain the service quality requirements of the target task, or obtain the service identifier corresponding to the target task, and then obtain the service quality requirements of the target task based on the service identifier.

[0051] In one possible implementation, the method further includes: receiving a second scheduling result, the second scheduling result indicating a second way for the at least one computing node to execute the target task, the second way being different from the first way; and executing the target task according to the second scheduling result.

[0052] Based on the above possible implementation methods, the computing node may receive the second scheduling result and execute the target task according to the second scheduling result.

[0053] In a possible implementation manner, the first manner instructs the at least one computing node to execute a first subtask in the target task; the second manner instructs the at least one computing node to execute a second subtask in the target task.

[0054] Based on the above possible implementation methods, before and after the scheduling policy of the target task is adjusted, the computing node can execute different subtasks to provide corresponding service quality.

[0055] In a third aspect, a communication device is provided for implementing the above-mentioned method. The communication device may be the network node described in the first aspect; alternatively, the communication device may be the computing node described in the second aspect. The communication device includes modules, units, or means corresponding to the above-mentioned method. The modules, units, or means may be implemented in hardware, software, or by hardware executing corresponding software implementations. The hardware or software includes one or more modules or units corresponding to the above-mentioned functions.

[0056] In conjunction with the third aspect above, in one possible implementation, the communication device may include an interface module and a processing module. The interface module, which may also be referred to as an interface unit, is configured to implement the sending and / or receiving functions described in any of the above aspects and any possible implementations thereof. The interface module may be comprised of an interface circuit, a transceiver, a transceiver, or a communication interface. The processing module may be configured to implement the processing functions described in any of the above aspects and any possible implementations thereof. The processing module may be, for example, a processor.

[0057] In combination with the third aspect above, in a possible implementation, the interface module includes a sending module and a receiving module, which are respectively used to implement the sending and receiving functions in any of the above aspects and any possible implementations thereof.

[0058] In a fourth aspect, a communication device is provided, comprising: a processor; the processor is coupled to a memory, and after reading instructions from the memory, executes the method described in any of the above aspects according to the instructions. The communication device may be the network node described in the first aspect; or the communication device may be the computing node described in the second aspect.

[0059] In conjunction with the fourth aspect above, in one possible implementation, the communication device further includes a memory for storing program instructions and data. Optionally, the memory is integrated with the processor; or the memory is independent of the processor.

[0060] In conjunction with the fourth aspect above, in one possible implementation, the processor and / or memory further includes an artificial intelligence (AI) module for implementing AI-related functions. The AI ​​module can implement AI functions through software, hardware, or a combination of software and hardware. For example, the AI ​​module includes a radio access network (RAN) intelligent controller (RIC) module.

[0061] In conjunction with the fourth aspect above, in one possible implementation, the communication device is a chip or a chip system. Optionally, when the communication device is a chip system, it can be composed of a chip or include a chip and other discrete devices.

[0062] In a fifth aspect, a communication device is provided, comprising: a processor and an interface circuit; the interface circuit being configured to receive a computer program or instruction and transmit it to the processor; and the processor being configured to execute the computer program or instruction, thereby causing the communication device to perform the method described in any of the above aspects. The communication device may be the network node described in the first aspect; alternatively, the communication device may be the computing node described in the second aspect.

[0063] In conjunction with the fifth aspect above, in one possible implementation, the processor further includes an AI module for implementing AI-related functions. The AI ​​module can implement AI functions through software, hardware, or a combination of software and hardware. For example, the AI ​​module includes a RIC module.

[0064] In conjunction with the fifth aspect above, in one possible implementation, the communication device is a chip or a chip system. Optionally, when the communication device is a chip system, it can be composed of a chip or include a chip and other discrete devices.

[0065] In a sixth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium. When the computer-readable storage medium is run on a computer, the computer can execute the method described in any one of the above aspects.

[0066] In a seventh aspect, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to execute the method described in any one of the above aspects.

[0067] In an eighth aspect, a communication system is provided, which includes a network node for executing the method described in the first aspect and a computing node for executing the method described in the second aspect.

[0068] In conjunction with the eighth aspect, in one possible implementation, the communication system further includes a communication node. The network node is further configured to send a second quality of service requirement to the communication node and receive second feedback information from the communication node. The second feedback information may indicate that the quality of service provided by the communication node fails to meet the second quality of service requirement, and the second feedback information may be used to determine a second scheduling result. The communication node is configured to receive the second quality of service requirement from the network node and send the second feedback information.

[0069] Among them, the technical effects brought about by any possible implementation method in the third to eighth aspects can be referred to the technical effects brought about by any aspect in the first to second aspects or different possible implementation methods in any aspect, and will not be repeated here.

[0070] It is understandable that, provided that the solutions are not contradictory, the solutions in each aspect can be combined. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] FIG1A is a schematic diagram of downlink data transmission between layers provided by the present application;

[0072] FIG1B is a schematic diagram of a centralized unit (CU) and a distributed unit (DU) provided by the present application;

[0073] FIG1C is a first schematic diagram of a RAN node provided by this application;

[0074] FIG1D is a second schematic diagram of a RAN node provided in this application;

[0075] FIG1E is a third schematic diagram of a RAN node provided in this application;

[0076] FIG1F is a schematic diagram of a task provided in this application;

[0077] FIG1G is a schematic diagram of end-cloud collaboration for a virtual reality (VR) task provided by this application;

[0078] FIG1H is a schematic diagram of a deep neural network (DNN) model provided in this application;

[0079] FIG2 is a schematic diagram of the communication system architecture provided by this application;

[0080] FIG3A is a schematic diagram of a first deployment method of network nodes provided in this application;

[0081] FIG3B is a second schematic diagram of a deployment method of network nodes provided in this application;

[0082] FIG3C is a third schematic diagram of the deployment method of the network nodes provided in this application;

[0083] FIG3D is a fourth schematic diagram of a deployment method of network nodes provided in this application;

[0084] FIG3E is a fifth schematic diagram of a deployment method of network nodes provided in this application;

[0085] FIG4 is a schematic diagram of the hardware structure of the communication device provided in this application;

[0086] FIG5 is a flow chart of the communication method provided in this application;

[0087] FIG6 is a second flow chart of the communication method provided by this application;

[0088] FIG7 is a schematic diagram of the structure of the communication device provided in this application. DETAILED DESCRIPTION

[0089] Before introducing the technical solution of this application, the relevant technical terms involved in this application are explained. It is understood that these explanations are intended to make this application easier to understand and should not be regarded as limiting the scope of protection claimed in this application.

[0090] 1. Terminal

[0091] The terminal in this application is a device with wireless transceiver and computing functions. The terminal can be deployed on land, including indoors, outdoors, handheld or vehicle-mounted; it can also be deployed on the water (such as ships, etc.); it can also be deployed in the air (for example, on airplanes, balloons and satellites, etc.). The terminal can also be called a terminal device, and the terminal device can be a user equipment (UE), where the UE includes a handheld device, a vehicle-mounted device, a wearable device or a computing device with wireless communication functions. Exemplarily, the UE can be a mobile phone, a tablet computer or a computer with wireless transceiver functions. The terminal device can also be a VR terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in unmanned driving, a wireless terminal in telemedicine, a wireless terminal in a smart grid, a wireless terminal in a smart city, or a wireless terminal in a smart home, etc.

[0092] As an example and not a limitation, in this application, the terminal may be a wearable device. Wearable devices may also be referred to as wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. For example, a wearable device is not only a hardware device, but also a device that achieves powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include devices that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as devices that focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0093] In the present application, the terminal may also be a terminal in an Internet of Things (IoT) system. IoT is an important component of the future development of information technology. Its main technical feature is to connect objects to the network through communication technology, thereby realizing an intelligent network of human-machine interconnection and object-to-object interconnection. The terminal in the present application may be a terminal in machine type communication (MTC). The terminal of the present application may be an on-board module, on-board module, on-board component, on-board chip or on-board unit built into a vehicle as one or more components or units. The vehicle may implement the method of the present application through the built-in on-board module, on-board module, on-board component, on-board chip or on-board unit.

[0094] 2. RAN Node

[0095] The RAN node in this application can be any device with wireless transceiver functions, which can provide wireless access services for terminals. Optionally, the RAN node can also have computing capabilities. RAN nodes may include but are not limited to: evolved base stations (NodeB or eNB or e-NodeB, evolutionary Node B) in long term evolution (LTE), evolved base stations (next generation eNB, ng-eNB) in next generation LTE, base stations (gNodeB or gNB) or transmission receiving points (TRP) in new radio (NR), base stations of subsequent evolution of the third generation partnership project (3GPP), access nodes in wireless fidelity (WiFi) systems, wireless relay nodes, wireless backhaul nodes, etc. The base station can be: a macro base station, a micro base station, a pico base station, a small station, a relay station, or a balloon station, etc. Multiple base stations can support networks of the same technology mentioned above, or they can support networks of different technologies mentioned above. The base station may include one or more co-sited or non-co-sited TRPs. The RAN node may also be at least one of a radio controller, a CU, a DU, a centralized unit control plane (CU-CP) node, a centralized unit user plane (CU-UP) node, a baseband processing unit (BBU) or a remote radio unit (RRU) in a cloud radio access network (CRAN) scenario. The following description takes the RAN node as a base station as an example. The multiple RAN nodes may be base stations of the same type or different types. The base station may communicate with the terminal or with the terminal through a relay station. The terminal may communicate with multiple base stations of different technologies. For example, the terminal may communicate with a base station supporting an LTE network or a base station supporting a 5G network, and may also support dual connections with base stations of an LTE network and a base station of a 5G network.

[0096] In this application, the CU and DU can be set separately, or can also be included in the same network element, such as the BBU. The RU can be included in a radio frequency device or radio frequency unit, such as an RRU, an active antenna unit (AAU), or a remote radio head (RRH). It can be understood that the CU can be divided into a network device in the access network, or the CU can be divided into a network device in the core network, without limitation here.

[0097] In different systems, CU (or CU-CP and CU-UP), DU or RU may also have different names, but those skilled in the art can understand their meanings. For example, in an open radio access network (ORAN) system, CU may also be called O-CU (open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, CU-UP may also be called O-CU-UP, and RU may also be called O-RU. For the convenience of description, this application takes CU, CU-CP, CU-UP, DU and RU as examples for description. Any unit of CU (or CU-CP, CU-UP), DU and RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0098] 3. Protocol layer

[0099] In the present application, the terminal and the RAN node can communicate through the protocol layer. The protocol layer may include a control plane (CP) protocol layer and a user plane (UP) protocol layer. Among them, the control plane protocol layer may include a radio resource control (RRC) layer, a packet data convergence protocol (PDCP) layer, a radio link control (RLC) layer, a media access control (MAC) layer and a physical (PHY) layer and other protocol layers. The user plane protocol layer may include a PDCP layer, an RLC layer, a MAC layer and a physical layer and other protocol layers. Optionally, for the user plane protocol layer, a service data adaptation protocol (SDAP) layer may also be included above the PDCP layer.

[0100] The following description uses the user plane protocol layer as an example. User plane data between RAN nodes and terminals can pass through user plane protocol layers, such as the SDAP layer, PDCP layer, RLC layer, MAC layer, and physical layer. The SDAP layer, PDCP layer, RLC layer, MAC layer, and physical layer are collectively referred to as the access layer. Depending on the direction of user plane data transmission, each of these layers is further divided into a transmitting portion and a receiving portion. The following description uses downlink data transmission as an example.

[0101] As shown in Figure 1A, it is a schematic diagram of the transmission of downlink data between layers. The arrows in Figure 1A indicate the transmission direction of downlink data, that is, after the SDAP layer of the RAN node obtains the downlink data from the upper layer, it transmits the downlink data to the PDCP layer, RLC layer and MAC layer in sequence, and then the MAC layer generates a transport block (TB), which is then wirelessly transmitted through the physical layer to transmit the transport block to the physical layer of the terminal. Subsequently, the physical layer of the terminal delivers the downlink data to the upper layer. It can be understood that data can be encapsulated accordingly in each layer. For example, the data received by a layer from the upper layer of the layer is regarded as the service data unit (SDU) of the layer, which becomes the protocol data unit (PDU) after being encapsulated by the layer and then passed to the next layer.

[0102] Optionally, in addition to the access layer, the protocol layer of the terminal also includes an application layer (app layer) and a non-access layer (NAS). The non-access layer can be used to forward user data, such as forwarding uplink data received from the application layer to the SDAP layer or forwarding downlink data received from the SDAP layer to the application layer. The application layer can be used to provide services to applications installed in the terminal. For example, downlink data received by the terminal can be submitted by the physical layer to the upper layer to reach the application layer, and then provided to the application by the application layer. For another example, the application layer can obtain data generated by the application and submit the data to the lower layer to reach the physical layer, and then send it to other communication devices via the physical layer.

[0103] 4. CU and DU

[0104] As mentioned above, a RAN node can include a CU and a DU. In other words, the RAN node's functions can be split, with some functions deployed on the CU and the remaining functions on the DU. Signaling generated by the CU can be sent to terminals via the DU, and vice versa.

[0105] This application does not limit the number of CUs or DUs included in a RAN node. For example, a RAN node may include one CU and multiple DUs. These multiple DUs may be centrally controlled by one CU. The interface between the CU and the DU may be referred to as the F1 interface. The interface between the CU and the DU can be further divided into a control plane interface (F1-C) and a user plane interface (F1-U).

[0106] In some embodiments, the CU and DU can be divided based on the protocol layers of the wireless network. For example, the PDCP layer and protocol layers above it are located in the CU, and protocol layers below the PDCP layer are located in the DU. Specifically, this can be shown in Figure 1B. It should be understood that Figure 1B is only an example of the division of the CU and DU based on the protocol layers. In specific applications, other division methods are possible and are not limited.

[0107] In some embodiments, the CU may be further separated into a CU-CP node and a CU-UP node. The CU-CP node and the CU-UP node may be coupled with the DU to jointly perform the functions of a RAN node.

[0108] 5. RAN intelligent controller (RIC)

[0109] In the ORAN architecture, RAN nodes include a RIC (such as a near-real-time RIC (nrt-RIC)), an O-CU, and an O-DU. This means that the functions of the RAN node can be split and deployed on the RIC, O-CU, and O-DU, respectively. For example, in the ORAN architecture, the RAN node may be shown in Figure 1C or Figure 1D.

[0110] Optionally, the RIC may also include a non-real-time RIC (Non-RT-RIC). For example, the RAN node may be as shown in FIG1E .

[0111] In a possible design, in Figure 1C, Figure 1D, or Figure 1E, the O-CU may include the RRC layer and the PDCP layer, the O-DU may include the RLC layer, the MAC layer, and the physical layer, and the nrt-RIC may implement application layer functions, as well as be used to perform delay-sensitive functions such as load balancing, switching, and interference detection. Optionally, the nrt-RIC may also implement AI-related functions, making full use of AI and machine learning to control and optimize wireless access network components and resources based on applications running on the network. The non-RT-RIC may be used to perform one or more of the following non-real-time operations: service orchestration, policy management, or analysis, etc.

[0112] 6. Extended reality (XR)

[0113] XR refers to various virtual and real-world environments, as well as human-machine interactions, generated by computing technologies and wearable devices. These environments offer multiple perspectives and strong interactivity, providing users with a completely new experience and possessing enormous application value and commercial potential. XR primarily encompasses virtual-reality interaction technologies such as VR, AR, and mixed reality (MR). These technologies are widely applicable in entertainment, gaming, healthcare, advertising, industry, online education, and engineering, among other fields. The following sections introduce VR, AR, and MR technologies, respectively.

[0114] VR technology combines computer graphics, multimedia, and other technologies to simulate the functions of human senses such as vision, hearing, and touch, making people feel as if they are actually there, immersed in a computer-generated virtual world. They can also communicate in real time through language, gestures, etc., enhancing the sense of immersion. Through VR technology, people can not only experience the real world realistically, but also break through the limitations of time and space and experience the wonderful experience of entering a virtual world. VR technology usually requires users to wear an XR terminal (such as a head-mounted device) to simulate vision and / or hearing and / or touch for the user. VR technology can also track the user's movements to timely update the simulated visual and / or auditory and / or tactile content. For example, VR technology can process the user's status information (such as the user's location information and posture information) to display scene content corresponding to the user's status information on the XR terminal.

[0115] AR technology can use computer technology to overlay virtual information onto the real world, displaying it through devices such as mobile phones, tablets, and glasses for people to perceive, thereby achieving a grand fusion of the real and the virtual and enriching the real world. In short, it is to give physical objects more information, enhance the sense of three-dimensionality, and strengthen the visual effects and interactive experience. For example, AR technology can process perceived visual information (usually including depth information) to merge virtual information with the real world and be perceived by users, thereby achieving an "enhancement" of the real world.

[0116] MR technology can mix the real world and the virtual world to produce a new visual environment that contains both physical entities and virtual information, and the content seen in the visual environment is "real-time".

[0117] 7. Multi-node collaboration

[0118] In recent years, computing services such as video rendering services like cloud gaming and VR, and AI services like terminal visual cognition, AR, and MR, have placed increasing demands on network transmission bandwidth and terminal computing power. However, network transmission bandwidth and terminal computing power are limited and can no longer meet the needs of these computing services.

[0119] Taking the video rendering business as an example, if video rendering is performed on the terminal, the terminal's computing power is insufficient to achieve purely local high-definition video rendering. If video rendering is performed on the cloud, it cannot meet the low latency requirements and is limited by the network's transmission capacity, the picture will have black edge effects, distortion, and other phenomena.

[0120] Taking AI services as an example, if AI services are executed on the terminal, the terminal's computing power and power are insufficient to support local AI reasoning. If AI services are executed in the cloud, while meeting low latency requirements, AI reasoning in the cloud requires a large uplink bandwidth, which will limit uplink coverage and the number of users.

[0121] Therefore, in order to take into account the computing power of the terminal and the transmission capacity of the network, a solution for multi-node collaboration to jointly process tasks is proposed. Among them, a task may refer to the work of processing data through multiple steps. The multiple steps may be parallel steps. For example, Task 1 shown in Figure 1F includes two parallel steps, such as S1a and S1b. Alternatively, the multiple steps may be serial steps. For example, Task 2 shown in Figure 1F includes three serial steps, such as S1 to S3. Alternatively, the multiple steps may be parallel and serial steps. For example, Task 3 shown in Figure 1F includes two parallel steps, such as S1a and S1b, and a step serial with these two steps, such as S2.

[0122] In order to facilitate the description of the multiple steps included in a task, one step or multiple related steps in the multiple steps can be regarded as a subtask, that is, a task can include multiple subtasks. For example, Task 1 shown in Figure 1F includes two subtasks, one subtask includes S1a, and the other subtask includes S1b. For another example, Task 2 shown in Figure 1F includes two subtasks, one subtask includes S1, and the other subtask includes S2 to S3; or, one subtask includes S1 to S2, and the other subtask includes S3. Alternatively, Task 2 shown in Figure 1F includes three subtasks, the first subtask includes S1, the second subtask includes S2, and the third subtask includes S3. For another example, Task 3 shown in Figure 1F includes two subtasks, one subtask includes S1a and S1b, and the other subtask includes S2; or, one subtask includes S1a, and the other subtask includes S1b and S2.

[0123] In this application, a computing service may include one or more tasks. If a computing service includes a task, the terminal and the cloud can each perform a part of the subtasks to reduce the task's demand on the terminal and computing power and the demand for network transmission bandwidth. If a computing service includes multiple tasks, the terminal and the cloud can collaborate to jointly process all or part of the multiple tasks. For example, for each of the all or part of the tasks, the terminal and the cloud can each perform a part of the subtasks to reduce the multiple tasks' demand on the terminal and computing power and the demand for network transmission bandwidth. In addition, when a computing service includes multiple tasks, the logic of the terminal and the cloud to execute each task is similar, so for ease of description, this application takes the terminal and the cloud executing a task (such as the target task in the following embodiment) as an example for description. These descriptions apply to the terminal and the cloud executing any task, and a unified explanation is made here, which will not be repeated later.

[0124] It can be understood that the terminal or cloud in this application can be collectively referred to as a computing node. The computing node can be any device with computing and communication capabilities. In addition to the terminal and the cloud, the computing node can also be a RAN node, a functional module of a RAN node (such as CU, DU or RIC, etc.), a core network element, a server, an application server, a cloud server, a cloud platform (cloud platform), a mobile edge computing (MEC) platform or a computing execution entity (CEF), etc., without limitation. Among them, the introduction of the terminal and the RAN node can refer to the explanation of the technical terms involved in this application in the above text. The core network network element is, for example, one or more of the following network elements: a user plane function (UPF) network element, an access management function (AMF) network element, a session management function (SMF) network element, a policy control function (PCF) network element or an application function (AF) network element. The cloud platform can be located in the data network behind the UPF network element, and can interact with the fifth-generation system (5GS) through the N6 interface for user-plane application layer data to provide computing services.

[0125] In summary, a task can be divided into multiple subtasks, and these multiple subtasks can be executed by multiple computing nodes. It should be understood that this application does not limit the number of computing nodes that execute a task. Taking Task 2 shown in Figure 1F as an example, Task 2 can be divided into 3 subtasks, each of which is executed by 3 computing nodes, or Task 2 can be divided into 2 subtasks, each of which is executed by 2 computing nodes.

[0126] The following uses the terminal and cloud as examples to introduce the specific process of multiple computing nodes executing tasks.

[0127] Referring to Figure 1G , the VR task can be divided into subtasks 101 and 102. The terminal can obtain initial data (e.g., image information), input the initial data into subtask 101, obtain intermediate data, and send the intermediate data to the cloud. After receiving the intermediate data, the cloud inputs it into subtask 102 to obtain the target data.

[0128] It is understandable that when dividing a task, there may be multiple ways of dividing the task, and different ways of dividing the task may correspond to different communication requirements.

[0129] For example, taking the video rendering task as an example, the video rendering task can be divided into foreground and background separation. For example, the foreground part of the picture is generally difficult to predict and has a small rendering volume, so it can be rendered locally by the terminal. The background part of the picture is predictable and has a large rendering volume, so it can be rendered by the cloud. Alternatively, the video rendering task can be divided according to the user's gaze point. For example, the picture at the center of the gaze point can be rendered locally by the terminal, and the picture around the gaze point can be rendered by the cloud. Alternatively, the video rendering task can be divided by object, such as selecting different nodes for rendering based on the rendering calculation and data volume of different objects.

[0130] For video rendering tasks, different splitting methods correspond to different amounts of video data to be transmitted, so different splitting methods correspond to different communication requirements. For example, in Table 1, the average bit rate of data transmission required by pure cloud rendering is 4Mbps, the peak is 40Mbps, and the allowed transmission delay is 25ms. The average bit rate of data transmission required by end-cloud collaborative rendering is 0.49Mbps~1.8Mbps, the peak is 3.8Mbps, and the allowed transmission delay is 50ms. It can be seen that end-cloud collaborative processing of video rendering tasks can reduce the downlink transmission rate requirements. Therefore, choosing an appropriate splitting method for video rendering tasks can increase the number of downlink transmission users accessing the network.

[0131] Table 1

[0132] Taking AI computing tasks as an example, the computational model for an AI computing task can be divided into multiple sub-models, with one or more sub-models considered a sub-task. For example, the AI ​​computing model is a neural network (NN) model or a DNN model. NN / DNN models can include multiple layers, such as input layers, convolution layers, pooling layers, and fully connected layers. These layers can be divided, for example, with the input and convolution layers deployed on the terminal and the pooling and fully connected layers deployed on the cloud. In this way, the terminal can use perception modules such as radar, cameras, controllers, or microphones to perceive user behavior data (such as user location, posture, or voice information) or surrounding environment data (such as videos or images of the surrounding environment), sequentially input this data into the input and convolution layers, generate intermediate data, and send this intermediate data to the cloud. Subsequently, after receiving the intermediate data, the cloud sequentially inputs the intermediate data into the pooling and fully connected layers to generate the target data. Optionally, the cloud can also send the target data to the terminal.

[0133] Further research revealed that when tasks are split, different locations of task splitting (hereinafter referred to as task splitting points) may result in different amounts of intermediate data and different computing loads on the terminal.

[0134] Exemplarily, taking the DNN model shown in Figure 1H as an example, the DNN model includes an input layer, a convolutional layer 1, a pooling layer 1, a convolutional layer 2, a pooling layer 2, a convolutional layer 3, a pooling layer 3, a convolutional layer 4, a pooling layer 4, a convolutional layer 5, an activation function (such as a rectified linear unit (ReLU)), a pooling layer 5, a fully connected layer 1, a fully connected layer 2 and an output layer. The initial data is input into the above layers in sequence to obtain the target data. Figure 1H also shows five candidate split points, namely candidate split point 0 to candidate split point 4, and these five candidate split points have different positions in the DNN model. It can be understood that the layers before the candidate split point can be deployed on the terminal, and the layers after the candidate split point can be deployed on the cloud. For example, candidate split point 0 is located before the input layer, so the terminal does not perform any tasks and sends the initial data to the cloud, which then performs all tasks. Candidate split point 1 is located after pooling layer 1, so the terminal performs tasks before pooling layer 1, and the cloud performs tasks after pooling layer 1. Candidate split point 2 is located after pooling layer 2, so the terminal performs tasks before pooling layer 2, and the cloud performs tasks after pooling layer 2. Candidate split point 3 is located after pooling layer 5, so the terminal performs tasks before pooling layer 5, and the cloud performs tasks after pooling layer 5. Candidate split point 4 is located after the output layer, so the terminal performs all tasks, the cloud does not perform any tasks, and the terminal does not send any intermediate data to the cloud. The approximate output uplink data size (i.e., the amount of intermediate data) and the computational load in the terminal (required UL data rate) corresponding to each candidate split point are different. Taking candidate split points 1 and 2 as examples, in Table 2, the approximate output uplink data size corresponding to candidate split point 1 is 120 Mbit / s, and the corresponding terminal computational load is low (low); the approximate output uplink data size corresponding to candidate split point 2 is 24 Mbit / s, and the corresponding terminal computational load is high (high). A larger approximate output uplink data size indicates more data the terminal needs to transmit, i.e., a larger amount of intermediate data. Therefore, the smaller the terminal computational load, i.e., the lower the terminal computational load. Conversely, a smaller approximate output uplink data size indicates less data the terminal needs to transmit, and the higher the terminal computational load. Therefore, while meeting the terminal computational load requirements, selecting an appropriate task split point can reduce the data rate to be transmitted, thereby increasing the number of users with network access.

[0135] Table 2

[0136] In summary, the task scheduling results can affect the size of intermediate data, communication requirements, and the computing load of the terminal, so determining the task scheduling results is crucial. The task scheduling results can indicate how at least one computing node should execute the task—in other words, which subtasks within the task the computing node should execute.

[0137] Currently, RAN nodes can obtain a communication QoS profile and indicate the communication QoS profile to the computing node, so that the computing node can determine the task scheduling result corresponding to the communication QoS profile. The computing node can then execute the task based on the task scheduling result to ensure data transmission quality. The above method can be used in scenarios where the task scheduling result has communication QoS requirements. However, with the development of communication technology, communication scenarios are becoming increasingly complex. In these communication scenarios, if only considering the case where the task scheduling result has communication QoS requirements, the task scheduling result will not be suitable for the computing node, thereby affecting the service quality of the entire task.

[0138] To address the above-mentioned issues, the present application provides a communication method. In this method, a network node may send a first scheduling result and a first quality of service requirement to at least one computing node. The first scheduling result may indicate a first method for the at least one computing node to perform a target task. After receiving the above information, the at least one computing node may perform the target task according to the first method and send first feedback information to the network node. The first feedback information may indicate that the quality of service of performing the target task according to the first method cannot meet the first quality of service requirement. After receiving the first feedback information, the network node may determine a second scheduling result based on the first feedback information. The second scheduling result indicates a second method for the at least one computing node to perform the target task, which is different from the first method. In the above process, the network node may promptly update the scheduling result of the target task based on the feedback from the at least one computing node, so that the updated scheduling result is more suitable for the computing node, thereby ensuring the quality of service of the target task and improving the user experience. The specific process of this method will be described in detail in the method shown in Figures 5 or 6 below and will not be repeated here.

[0139] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0140] The communication method provided in this application can be used in various communication systems. For example, the communication system can be a universal mobile telecommunications system (UMTS) system, a long term evolution (LTE) system, a fifth generation (5G) communication system, a wireless fidelity (WiFi) system, a communication system related to the third generation partnership project (3GPP), a future evolution communication system (such as: a sixth generation (6G) communication system, etc.), or a system that integrates multiple systems, etc., without limitation. Among them, 5G can also be called new radio (NR). The method provided in this application is described below using the communication system 20 shown in Figure 2 as an example. Figure 2 is only a schematic diagram and does not constitute a limitation on the applicable scenarios of the technical solution provided in this application.

[0141] Figure 2 shows a schematic diagram of the architecture of a communication system 20 provided in this application. In Figure 2, the communication system 20 may include a network node 201 and at least one computing node 202 that can communicate with the network node 201. For example, the at least one computing node may include one or more of a mobile phone 2021, a pad 2022, AR glasses 2023, or an application server 2024. For other descriptions of the computing nodes, please refer to the corresponding description above and will not be repeated here.

[0142] In the present application, the network node 201 may have a task management function (TMF) or a computing management function (CMF). For example, the network node 201 may adjust the scheduling result of the target task based on the feedback from the computing node to ensure the service quality of the target task. Optionally, the network node 201 may also perceive the network status, and / or the computing power status of the computing node, and / or the task topology of the target task (such as which subtasks the target task includes, etc.). Optionally, the network node 201 may also determine a first scheduling result so that the computing node executes the target task according to the first scheduling result. Optionally, the network node 201 may also determine a first quality of service requirement so that when the quality of service provided by the computing node when executing the target task according to the first scheduling result does not meet the first quality of service requirement, the network node 201 sends feedback information to the network node 201.

[0143] It is understood that network node 201 can be deployed on the RAN side or the core network side, without limitation. For example, the functions of network node 201, such as TMF or CMF, can be deployed on a RAN node. As another example, the functions of network node 201 can be deployed on one or more network elements in the core network, such as an AMF network element, an SMF network element, or a newly added network element in the core network. The following describes the deployment of network node 201 using the deployment of TMF on a RAN node as an example.

[0144] Referring to Figure 3A , the TMF can be deployed independently on a RAN node. For example, the RAN node can have both TMF and gNB functionality, and the TMF can communicate with the gNB over the E1 interface. Alternatively, the E1 interface can be the interface between the TMF and the CU in the gNB.

[0145] As shown in Figure 3B, the TMF can be deployed on the CU in a RAN node. For example, the RAN node has gNB functionality, the gNB includes the CU and DU, and the CU has the TMF.

[0146] As shown in Figure 3C, the TMF can be deployed on the DU in a RAN node. For example, the RAN node has gNB functionality, the gNB includes a CU and DU, and the DU has the TMF.

[0147] As shown in Figure 3D, the TMF can be deployed on the RIC, such as the nrt-RIC. For example, a RAN node may have gNB functionality, with the gNB comprising the nrt-RIC, O-CU, and O-DU. The nrt-RIC may also have the TMF. The TMF can communicate with the O-CU or O-DU over the E2 interface. Third-party applications can also be deployed on the nrt-RIC.

[0148] Referring to Figure 3E , the TMF can be deployed on a RIC, such as a non-RT-RIC. For example, a RAN node may have gNB functionality. The gNB includes a non-RT-RIC, an nrt-RIC, an O-CU, and an O-DU. The non-RT-RIC may also have a TMF. The TMF can communicate with the nrt-RIC over the A1 interface.

[0149] It should be understood that the above examples are only examples of deployment methods of the network node 201. In specific applications, the network node 201 can also have other deployment methods, such as the functions of the network node 201 can also be deployed on the O-CU or O-DU, without limitation.

[0150] It is understood that the communication system 20 shown in FIG2 is for illustrative purposes only and is not intended to limit the technical solutions of the present application. Those skilled in the art will appreciate that, in a specific implementation, the communication system 20 may further include other devices, and the number of network nodes and computing nodes may be determined based on specific needs without limitation.

[0151] Optionally, each node in Figure 2 of the present application (such as network node 201 or any computing node, etc.) can also be referred to as a communication device, which can be a general device or a dedicated device. This application does not make specific limitations on this.

[0152] Optionally, the relevant functions of each node in FIG. 2 of the present application (e.g., network node 201 or any computing node, etc.) can be implemented by a single device, or by multiple devices, or by one or more functional modules within a single device, and this application does not impose any specific restrictions on this. It is understood that the above functions can be network elements in hardware devices, software functions running on dedicated hardware, or a combination of hardware and software, or virtualized functions instantiated on a platform (e.g., a cloud platform).

[0153] In a specific implementation, each node in FIG. 2 of the present application (e.g., network node 201 or any computing node, etc.) may adopt the structure shown in FIG. 4 , or include the components shown in FIG. 4 . FIG. 4 is a schematic diagram of the hardware structure of a communication device applicable to the present application. The communication device 40 includes at least one processor 401 and at least one communication interface 404 for implementing the method provided in the present application. The communication device 40 may also include a communication line 402 and a memory 403.

[0154] The processor 401 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present application.

[0155] The communication link 402 may include a path for transmitting information between the above components, such as a bus.

[0156] Communication interface 404 is used to communicate with other devices or communication networks. Communication interface 404 can be any transceiver-like device, such as an Ethernet interface, a radio access network (RAN) interface, a wireless local area network (WLAN) interface, a transceiver, a pin, a bus, an interface circuit, or a transceiver circuit.

[0157] The memory 403 can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this. The memory can be independent and coupled to the processor 401 via the communication line 402. The memory 403 can also be integrated with the processor 401. The memory provided in this application can generally be non-volatile.

[0158] Among them, the memory 403 is used to store computer-executable instructions involved in executing the solution provided by this application, and is controlled by the processor 401. The processor 401 is used to execute the computer-executable instructions stored in the memory 403, thereby implementing the method provided by this application. Alternatively, optionally, in this application, the processor 401 may also perform the processing-related functions of the method provided below in this application, and the communication interface 404 is responsible for communicating with other devices or communication networks, which is not specifically limited in this application.

[0159] Optionally, the computer-executable instructions in this application may also be referred to as application code, which is not specifically limited in this application.

[0160] The coupling in this application is an indirect coupling or communication connection between devices, units or modules, which can be electrical, mechanical or other forms, and is used for information exchange between devices, units or modules.

[0161] As an embodiment, the processor 401 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 4 .

[0162] As an embodiment, the communication device 40 may include multiple processors, such as processor 401 and processor 407 in FIG4 . Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. The processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0163] As an embodiment, the communication device 40 may further include an output device 405 and / or an input device 406. The output device 405 is coupled to the processor 401 and can display information in a variety of ways. For example, the output device 405 can be a liquid crystal display (LCD), a light emitting diode (LED) display device, a cathode ray tube (CRT) display device, or a projector. The input device 406 is coupled to the processor 401 and can receive user input in a variety of ways. For example, the input device 406 can be a mouse, a keyboard, a touch screen device, or a sensor device.

[0164] It is understandable that the composition structure shown in Figure 4 does not constitute a limitation on the communication device. In addition to the components shown in Figure 4, the communication device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0165] The method provided by the present application will be described below with reference to the accompanying drawings. Each network element in the following embodiment may include the components shown in FIG4 , which will not be described in detail.

[0166] It is understandable that the message names between the nodes or the names of the parameters in the messages in the following embodiments of the present application are merely examples, and other names may be used in specific implementations, and the present application does not impose any specific limitations on this.

[0167] It can be understood that in this application, "sending information to... (such as a computing node)" can be understood as the destination end of the information being the computing node. It can include sending information to the computing node directly or indirectly. "Receiving information from... (such as a computing node)" can be understood as the source end of the information being the computing node, which can include receiving information from the computing node directly or indirectly. The information may be processed as necessary between the source end and the destination end of the information transmission, such as format changes, etc., but the destination end can understand the valid information from the source end. Similar expressions in this application can be understood similarly and will not be repeated here.

[0168] It is understood that in this application, " / " can indicate that the objects associated with each other are in an "or" relationship, for example, A / B can mean A or B; "and / or" can be used to describe that there are three relationships between the associated objects, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. In addition, expressions similar to "at least one of A, B and C" or "at least one of A, B or C" are usually used to indicate any of the following: A exists alone; B exists alone; C exists alone; A and B exist at the same time; A and C exist at the same time; B and C exist at the same time; A, B and C exist at the same time. The above uses A, B and C as an example to illustrate the optional items of the item. When there are more elements in the expression, the meaning of the expression can be obtained according to the above rules.

[0169] In order to facilitate the description of the technical solutions of the present application, in the present application, words such as "first" and "second" may be used to distinguish between technical features with the same or similar functions. The words such as "first" and "second" do not limit the quantity and execution order, and the words such as "first" and "second" do not necessarily limit them to be different. In the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" should not be interpreted as being more preferred or more advantageous than other embodiments or design. The use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0170] It is understood that the "embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, the various embodiments in the entire specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It is understood that in the various embodiments of the present application, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the present application.

[0171] It can be understood that in the present application, "used to indicate" can include direct indication and indirect indication, and can also include explicit indication and implicit indication. When describing that a certain information is used to indicate A, it can include that the information directly indicates A or indirectly indicates A, and it does not mean that the information must carry A. The information indicated by a certain information is called information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated, such as but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated. The information to be indicated can also be indirectly indicated by indicating other information, where there is an association between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while the other parts of the information to be indicated are known or agreed in advance. For example, the indication of specific information can also be achieved by means of the arrangement order of each information agreed in advance (for example, stipulated by the protocol), thereby reducing the indication overhead to a certain extent.

[0172] It can be understood that in this application, "when...", "in the case of...", "if" and "if" all mean that corresponding processing will be taken under certain objective circumstances, and do not limit the time, nor do they require judgment actions when implementing them, nor do they mean that there are other limitations.

[0173] The term "simultaneously" in this application may be understood as at the same time point, within a period of time, or within the same cycle.

[0174] In this application, "greater than or equal to" can be replaced by "greater than" or "equal to"; "less than or equal to" can be replaced by "less than" or "equal to". For example, "A is greater than or equal to B" can be replaced by "A is greater than B" or "A is equal to B"; "A is less than or equal to B" can be replaced by "A is less than B" or "A is equal to B".

[0175] It can be understood that some optional features in the present application can be implemented independently in certain scenarios without relying on other features, such as the solution on which it is currently based, to solve corresponding technical problems and achieve corresponding effects. They can also be combined with other features according to needs in certain scenarios. For example, S500a to S500b in the following embodiments of the present application can be implemented independently without relying on the method shown in Figure 5. For another example, S501a and S503a in the following embodiments of the present application can be implemented independently without relying on the method shown in Figure 5, so that the network node determines the second scheduling result based on the second feedback information. Accordingly, the device provided in the present application can also implement these features or functions accordingly, which will not be described in detail here.

[0176] It is understandable that the same step or steps or technical features with the same function in different embodiments of the present application can be referenced to each other.

[0177] It is understood that in the present application, the network nodes and / or computing nodes may perform some or all of the steps in the present application. These steps are merely examples, and the present application may also perform other steps or variations of various steps. In addition, the steps may be performed in a different order than presented in the present application, and it is possible that not all of the steps in the present application need to be performed.

[0178] It is understandable that the method provided below in this application uses a network node and a computing node as an example of the execution subject of the interaction diagram to illustrate the method, but this application does not limit the execution subject of the interaction diagram. For example, the network node in the method provided in the following embodiment of this application may also be a chip, chip system, or processor that supports the network node to implement the method, or a logical node, logic module, or software that can implement all or part of the network node function; the computing node in the method provided below in this application may also be a chip, chip system, or processor that supports the computing node to implement the method, or a logical node, logic module, or software that can implement all or part of the computing node function.

[0179] As shown in FIG5 , a communication method provided by the present application may include the following steps:

[0180] S501: A network node sends a first scheduling result and a first quality of service requirement to at least one computing node. Correspondingly, the at least one computing node receives the first scheduling result and the first quality of service requirement from the network node.

[0181] In the present application, the network node may be the network node 201 in the communication system 20 shown in FIG. 2 , and the at least one computing node may be the at least one computing node 202 in the communication system 20 .

[0182] In this application, the first scheduling result may indicate a first method for at least one computing node to execute a target task. The first method may indicate that at least one computing node executes a first subtask within the target task. For example, the first scheduling result includes an identifier for the first subtask. It should be understood that the first scheduling result may also indicate that at least one computing node executes the entire target task, without limitation. For an introduction to the target task and the first subtask, reference may be made to the previous description of tasks and subtasks, and will not be repeated here.

[0183] In this application, the first quality of service requirement may indicate the quality of service that at least one computing node must meet to execute the target task according to the first method. Optionally, the first quality of service requirement may indicate a quality of service requirement related to computing. For example, the first quality of service requirement may indicate at least one of the following: a computing type requirement, a computing amount requirement, a computing latency requirement, a computing energy consumption requirement, or a computing accuracy requirement. Each of the above requirements is described in detail below.

[0184] (1) Calculation type requirements

[0185] The computing type requirement indicated by the first quality of service requirement can enable at least one computing node to determine the computing type required to perform the target task according to the first method. The computing type can also be replaced by a computing power type.

[0186] One possible design is to divide the computing types according to the computing power of physical hardware, including CPU, graphics processing unit (GPU), neural network processing unit (NPU) or tensor processing unit (TPU), etc.

[0187] Another possible design is to classify computing types according to business types, including image recognition, video recognition, compression, graphics rendering, AI training, AI reasoning, perception processing, high-performance computing, or big data offline analysis.

[0188] Optionally, the first quality of service requirement may further indicate parameters corresponding to the computing type. For example, the first quality of service requirement may further indicate parameters such as the model of the physical hardware, memory, or main frequency, or the first quality of service requirement may further indicate the algorithm used by the service type, such as the recognition algorithm used for image recognition.

[0189] (2) Computational requirements

[0190] The computational requirement indicated by the first quality of service requirement can enable at least one computing node to determine the computational requirement required to execute the target task in the first manner. The computational requirement here can also be replaced by input / output data throughput, computing power or operation power, etc.

[0191] For example, the computational workload requirement may indicate how many operations or floating-point operations a processor of at least one computing node needs to perform per second.

[0192] Optionally, the first quality of service requirement may further indicate parameters corresponding to the computational effort. For example, the first quality of service requirement may further indicate the hardware model, memory (e.g., required bytes of content capacity), input / output (I / O) speed (e.g., required bandwidth for read / write operations per second (IOPS)), or main frequency (e.g., required Hz of the main frequency), etc., required to perform the target task.

[0193] (3) Calculation delay requirements

[0194] The computational latency requirement indicated by the first quality of service requirement may enable the at least one computing node to determine the latency requirement that needs to be met for executing the target task in the first manner. For example, the computational latency requirement may indicate how long the at least one computing node needs to complete the first subtask.

[0195] (4) Calculate energy consumption requirements

[0196] The computing energy consumption requirement indicated by the first quality of service requirement may enable at least one computing node to determine the computing energy consumption required to execute the target task according to the first method. The computing energy consumption may also be replaced by an indicator that can be converted to or from computing energy consumption, such as computing power consumption or computing power. The unit of computing energy consumption may be watts (W), joules (W), etc.

[0197] Exemplarily, the computing energy consumption requirement indicates how much energy or power is required for at least one computing node to consume when executing the target task according to the first method.

[0198] It is understood that the computing energy consumption may be the total energy consumption including transmission energy consumption and computing energy consumption, or the average computing energy consumption over a period of time, or the computing energy consumption for completing one process or within a short period of time. For example, the computing energy consumption may be the computing energy consumption for processing each video frame, the computing energy consumption for executing the first subtask, or the average computing energy consumption over a predefined period of time.

[0199] Optionally, the calculated energy consumption may also include thermal design power consumption (TDP). TDP is an indicator that reflects the heat release of a processor (such as a CPU or GPU), and may refer to the amount of heat released when the processor reaches maximum load, and may be measured in watts.

[0200] (5) Calculation accuracy requirements

[0201] The computational accuracy requirement indicated by the first quality of service requirement may indicate the error requirement between the computational result of the target task (such as the inference result of an AI model, the computational result of an algorithm, or the training result of an AI model) and the true value, such as requiring the error between the two to be less than or equal to 5%. Therefore, this computational accuracy requirement enables at least one computing node to determine the allowable error for executing the target task according to the first method.

[0202] Optionally, the calculation accuracy requirement may also indicate the probability of successful calculation or the probability of error in calculation, etc.

[0203] It should be understood that the above requirements are merely examples of computing-related QoS requirements that may be indicated by the first QoS requirement. In specific applications, the first QoS requirement may indicate more or fewer QoS requirements than the above, without limitation.

[0204] In order to better understand the method provided in the present application, S501 is introduced below by taking the case where the number of at least one computing node is 1 and the case where the number of at least one computing node is greater than 1 as examples.

[0205] Scenario 1: The number of at least one computing node is 1. The following description takes the computing node as the first computing node as an example. The first computing node is, for example, the mobile phone 2021, pad 2022, AR glasses 2023, or application server 2024 in Figure 2.

[0206] In one possible implementation, the network node sends a first scheduling result and a first service quality requirement to the first computing node, so that the first computing node executes the target task according to the first scheduling result and determines the service quality that needs to be met to execute the target task according to the first scheduling result.

[0207] For example, taking the target task as Task 2 shown in FIG1F , the first scheduling result may instruct the first computing node to execute S1 to S2 (i.e., the first subtask includes S1 to S2), and the content of the first quality of service requirement indication may be as shown in Table 3. After receiving the first scheduling result, the first computing node may determine that it is to execute S1 to S2 of Task 2. After receiving the first quality of service requirement, the first computing node may determine the following information: (1) The computing type corresponding to S1 to S2 is CPU, the model of the CPU is XX, and the main frequency is 3.2 GHz; (2) Execution of S1 to S2 requires the CPU to execute 100 million floating-point operations per second, and the required I / O bandwidth is 5 Gbits / S; (3) Execution of S1 to S2 cannot exceed 20 ms; (4) The energy consumption required to execute S1 to S2 is 2 watts; (5) The error between the result of executing S1 to S2 and the true value must be less than or equal to 3%.

[0208] Table 3

[0209] It should be understood that Table 3 is merely an example of the information indicated by the first quality of service requirement. In specific applications, the first quality of service requirement may indicate more or less information than shown in Table 3, without limitation.

[0210] Optionally, in scenario 1, the first scheduling result may instruct the first computing node to execute the entire target task.

[0211] Scenario 2: The number of at least one computing node is greater than 1. The following is explained using two computing nodes (such as a second computing node and a third computing node) as an example. For example, the second computing node is the mobile phone 2021 in Figure 2, and the third computing node is the pad 2022 in Figure 2; or, the second computing node is the AR glasses 2023 in Figure 2, and the third computing node is the application server 2024 in Figure 2. It is understandable that when the number of at least one computing node is greater than 2, the operation of each computing node is similar to the operation of the second computing node or the operation of the third node. Please refer to the corresponding description below and no further details will be given.

[0212] In one possible implementation, the network node sends the first scheduling result and the first quality of service requirement to the second computing node, and also sends the first scheduling result and the first quality of service requirement to the third computing node. This allows the second computing node to execute the target task based on the first scheduling result and determine the quality of service it needs to meet when executing the target task based on the first scheduling result. Furthermore, the third computing node can execute the target task based on the first scheduling result and determine the quality of service it needs to meet when executing the target task based on the first scheduling result.

[0213] It can be understood that the first scheduling result sent by the network node to the second computing node (for the convenience of description, hereinafter referred to as the scheduling result of the second computing node) and the first scheduling result sent by the network node to the third computing node (for the convenience of description, hereinafter referred to as the scheduling result of the third computing node) may be the same or different.

[0214] In one case, the first scheduling result can indicate the subtask that the corresponding computing node needs to execute. Taking the target task as Task 2 shown in Figure 1F as an example, the scheduling result of the second computing node can instruct the second computing node to execute S1-S2, and the scheduling result of the third computing node can instruct the third computing node to execute S3.

[0215] In another case, the first scheduling result indicates the subtasks that each computing node needs to execute. Still taking Task 2 as the target task shown in Figure 1F as an example, the scheduling results of the second computing node and the scheduling results of the third computing node both indicate that the second computing node executes S1-S2 and the third computing node executes S3.

[0216] It is understandable that the first quality of service requirement (for ease of description, hereinafter referred to as the quality of service requirement of the second computing node) sent by the network node to the second computing node and the first quality of service requirement (for ease of description, hereinafter referred to as the quality of service requirement of the third computing node) sent by the network node to the third computing node may be the same or different. For example, if the quality of service that the second computing node needs to meet in order to perform the target task is the same as the quality of service that the third computing node needs to meet in order to perform the target task, then the quality of service requirement of the second computing node and the quality of service requirement of the third computing node are the same; if the quality of service that the second computing node needs to meet in order to perform the target task is different from the quality of service that the third computing node needs to meet in order to perform the target task, then the quality of service requirement of the second computing node and the quality of service requirement of the third computing node are different.

[0217] It can be understood that the “quality of service requirements of the second computing node and the quality of service requirements of the third computing node are different” here may mean that the computing quality of service requirements indicated by the quality of service requirements of the second computing node are different from the quality of service requirements of the third computing node. For example, the quality of service requirements of the second computing node indicate the computing type requirements and the computing delay requirements, and the quality of service requirements of the third computing node indicate the computing type requirements, the computing amount requirements and the computing delay requirements. Alternatively, the “quality of service requirements of the second computing node and the quality of service requirements of the third computing node are different” here may mean that the computing quality of service requirements indicated by the quality of service requirements of the second computing node are the same as the quality of service requirements of the third computing node, but the specific requirements indicated by any one of the computing quality of service requirements are different. For example, the quality of service requirements of the second computing node and the quality of service requirements of the third computing node both indicate computing amount requirements, but the computing amount requirement indicated by the former indicates an I / O bandwidth of 5 Gbits / S, and the computing amount requirement indicated by the latter indicates an I / O bandwidth of 7 Gbits / S.

[0218] It can be understood that the content of the service quality requirement of the second computing node or the service quality requirement indication of the third computing node is similar to that in Table 3. For details, please refer to the corresponding description in Scenario 1 and no further details will be given.

[0219] It is understandable that the manner in which the network node sends the first scheduling result and the first quality of service requirement to at least one computing node is related to the location of the network node in the network and the location of the computing node in the network. The following describes the manner in which the network node 201 sends the first scheduling result to the mobile phone 2021, the pad 2022, the AR glasses 2023, and the application server 2024 in conjunction with Figure 2 and Figures 3A to 3E. The manner in which the first quality of service requirement is sent is similar to that of the first scheduling result and will not be described in detail.

[0220] Exemplarily, if the network node 201 is independently deployed on a RAN node (as shown in FIG3A ), the RAN node may send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via the air interface. For example, the first scheduling result is carried in downlink control information (DCI), a medium access control control element (MAC CE), a PDCP control PDU message, or an RRC message. The RAN node may send the first scheduling result to the application server 2024 via the core network. For example, the RAN node may carry the first scheduling result in a user plane part (GTP-U) header of at least one uplink data transmitted via the NG3 interface. After the RAN node sends the GTP-U header to the UPF network element, the UPF network element detects the GTP-U header information of the uplink data, and notifies the application server 2024 of the first scheduling result through the NEF network element or the local NEF (local-NEF) network element in the form of an application programming interface (API).

[0221] For example, if the network node 201 is deployed on a CU in a RAN node (as shown in FIG3B ), the CU may send the first scheduling result to the DU via the F1 interface, and the DU may send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via the DCI or MAC CE; alternatively, the CU may send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via a PDCP Control PDU message or an RRC message. The CU may send the first scheduling result to the application server 2024 via the core network (such as a UPF network element, a NEF network element, etc. in the core network).

[0222] For example, if the network node 201 is deployed on a DU in a RAN node (as shown in FIG3C ), the DU may send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via DCI or MAC CE; alternatively, the DU may send the first scheduling result to the CU via the F1 interface, and the CU may send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via a PDCP Control PDU message or an RRC message. Furthermore, if the CU is separated into a CU-CP and a CU-UP, the DU may send the first scheduling result to the CU-UP via the CU-CP, or the DU may send the first scheduling result to the CU-UP via the F1-u interface (e.g., the GTP-U header of the uplink data transmitted via the F1-u interface carries the first scheduling result), so that the CU-UP sends the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023. It can be understood that after receiving the first scheduling result, the CU or CU-UP can also send the first scheduling result to the application server 2024 through the core network (such as the UPF network element, NEF network element, etc. in the core network).

[0223] For example, if the network node 201 is deployed on the nrt-RIC in the RAN node (as shown in FIG3D ), the nrt-RIC can send the first scheduling result to the O-CU via the E2 interface. After receiving the first scheduling result, the O-CU can send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via a PDCP Control PDU message or an RRC message. Alternatively, the nrt-RIC can send the first scheduling result to the O-DU via the E2 interface. After receiving the first scheduling result, the O-DU can send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023 via a DCI or MAC CE. Furthermore, if the O-CU is separated into an O-CU-CP and an O-CU-UP, the nrt-RIC can send the first scheduling result to the O-CU-UP via the O-CU-CP, or the nrt-RIC can send the first scheduling result to the O-CU-UP so that the O-CU-UP can send the first scheduling result to the mobile phone 2021, the pad 2022, or the AR glasses 2023. It can be understood that after receiving the first scheduling result, O-CU or O-CU-UP can also send the first scheduling result to the application server 2024 through the core network (such as the UPF network element, NEF network element, etc. in the core network).

[0224] For example, if the network node 201 is deployed on the Non-RT-RIC in the RAN node (as shown in Figure 3E), the Non-RT-RIC can send the first scheduling result to the nrt-RIC through the A1 interface, so that the nrt-RIC can send the first scheduling result to the mobile phone 2021, pad 2022, AR glasses 2023 and application server 2024 in the above manner.

[0225] Optionally, the network node may also send parameter information required to execute the target task according to the first scheduling result to at least one computing node, such as one or more of the AI ​​model, algorithm or program information used to execute the target task.

[0226] S502: At least one computing node executes the target task according to the first scheduling result.

[0227] It is understood that at least one computing node can execute the target task according to the first method. For example, for scenario 1 above, the first computing node executes the target task according to the first method indicated by the first scheduling result. For scenario 2 above, the second computing node executes the target task according to the first method corresponding to the second computing node, and the third computing node executes the target task according to the first method corresponding to the third computing node.

[0228] Optionally, if the computing node executing the target task includes an application server, the application server may further send a computing result obtained by executing the target task to the terminal.

[0229] Optionally, if the at least one computing node cannot meet the first quality of service requirement, the at least one computing node may skip S502 and directly execute S503 to indicate to the network node that the first quality of service requirement cannot be met.

[0230] S503: At least one computing node sends first feedback information to the network node. Correspondingly, the network node receives the first feedback information from the at least one computing node.

[0231] In one possible implementation, if the service quality of at least one computing node performing the target task according to the first method cannot meet the first service quality requirement, the at least one computing node sends a first feedback message to the network node. The first feedback message can be used to indicate that the service quality of at least one computing node performing the target task according to the first method cannot meet the first service quality requirement. For example, the first feedback message includes 1 bit, and when the value of the 1 bit is "0" or "1", it indicates that the service quality of at least one computing node performing the target task according to the first method cannot meet the first service quality requirement. The "service quality of at least one computing node performing the target task according to the first method" here can be the service quality of a single detection of the computing node, or the service quality counted within a certain time period or cycle, such as the average value of the service quality within the time period or cycle, the maximum value of the service quality, or the minimum value of the service quality, etc.

[0232] Optionally, the first feedback information may further indicate the quality of service of at least one computing node when executing the target task according to the first method, so that the network node can adjust the scheduling strategy of the target task based on the quality of service and the first quality of service requirement. For example, the first feedback information may indicate that when at least one computing node executes the target task according to the first method, the CPU performs 100 million floating-point operations per second; and / or, indicate that the latency of at least one computing node when executing the target task according to the first method is 25ms; and / or indicate that the computing energy consumption of at least one computing node when executing the target task according to the first method is 10 watts; and / or indicate that the accuracy of at least one computing node when executing the target task according to the first method is 5%. Alternatively, the first feedback information may further indicate the deviation between the quality of service of at least one computing node when executing the target task according to the first method and the first quality of service requirement, so that the network node can adjust the scheduling strategy of the target task based on the deviation.

[0233] Exemplarily, taking scenario 1 as an example, if the calculation type of the first computing node does not meet the calculation type indicated by the first quality of service requirement, the first feedback information indicates that the calculation type of the first computing node does not meet the calculation type indicated by the first quality of service requirement, and optionally, the first feedback information also indicates the calculation type supported by the first computing node. And / or, if the I / O bandwidth of the first computing node when performing the target task according to the first method is less than the I / O bandwidth indicated by the first quality of service requirement, the first feedback information indicates that the calculation amount of the first computing node when performing the target task according to the first method does not meet the calculation amount requirement indicated by the first quality of service requirement, and optionally, the first feedback information also indicates the I / O bandwidth of the first computing node when performing the target task according to the first method or indicates the deviation between the I / O bandwidth of the first computing node when performing the target task according to the first method and the I / O bandwidth indicated by the first quality of service requirement. And / or, if the computational delay when the first computing node performs the target task according to the first method is greater than the delay indicated by the first quality of service requirement, the first feedback information indicates that the computational delay when the first computing node performs the target task according to the first method does not meet the computational delay requirement indicated by the first quality of service requirement. Optionally, the first feedback information further indicates the computational delay when the first computing node performs the target task according to the first method, or indicates the deviation between the computational delay when the first computing node performs the target task according to the first method and the computational delay indicated by the first quality of service requirement. And / or, if the computational energy consumption when the first computing node performs the target task according to the first method is greater than the computational energy consumption indicated by the first quality of service requirement, the first feedback information indicates that the computational energy consumption when the first computing node performs the target task according to the first method does not meet the computational energy consumption requirement indicated by the first quality of service requirement. Optionally, the first feedback information further indicates the computational energy consumption when the first computing node performs the target task according to the first method, or indicates the deviation between the computational energy consumption when the first computing node performs the target task according to the first method and the computational energy consumption indicated by the first quality of service requirement. And / or, if the computing accuracy of the first computing node when performing the target task according to the first method is lower than the computing accuracy indicated by the first service quality requirement, the first feedback information indicates that the computing accuracy of the first computing node when performing the target task according to the first method does not meet the computing accuracy requirement indicated by the first service quality requirement. Optionally, the first feedback information also indicates the computing accuracy of the first computing node when performing the target task according to the first method or indicates the deviation between the computing accuracy of the first computing node when performing the target task according to the first method and the computing accuracy indicated by the first service quality requirement.

[0234] It can be understood that the logic of the second computing node sending the first feedback information and the logic of the third computing node sending the first feedback information are similar to the logic of the first computing node sending the first feedback information, and are not repeated here.

[0235] It is understandable that the first feedback information sent by the second computing node and the first feedback information sent by the third computing node may be different. In addition, the second computing node and the third computing node do not necessarily both send the first feedback information. For example, if the service quality of the second computing node performing the target task according to the first method cannot meet the service quality requirements of the second computing node, and the service quality of the third computing node performing the target task according to the first method meets the service quality requirements of the third computing node, then the second computing node sends the first feedback information, and the third computing node does not send the first feedback information, and vice versa.

[0236] Optionally, the network node sends first condition information to at least one computing node. The first condition information may indicate a condition for triggering the first feedback information. Thus, after receiving the first condition information, the at least one computing node may determine whether to trigger the first feedback information, or when to trigger the first feedback information, based on the first condition information. The network node may send the first condition information to the at least one computing node before S503. For example, the network node may send the first condition information simultaneously with the first scheduling information and the first quality of service requirement, or the network node may send the first condition information after S501 and before S502, without limitation.

[0237] In one possible design, the condition for triggering the first feedback information includes at least one of the following: the service quality of at least one computing node performing the target task according to the first method does not meet the first service quality requirement; or the service quality of at least one computing node performing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to a first threshold; or the feedback time indicated by the first condition information arrives. The following uses the example of the first computing node sending the first feedback information to illustrate.

[0238] Exemplarily, if the conditions that trigger the first computing node to send the first feedback information include: the service quality of the first computing node when executing the target task according to the first method does not meet the first service quality requirement, then the first computing node can send the first feedback information to the network node when the service quality of the first computing node when executing the target task according to the first method does not meet the first service quality requirement.

[0239] Exemplarily, if the condition triggering the first computing node to send the first feedback information includes: the service quality of the first computing node when executing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to the first threshold value, then the first computing node may send the first feedback information to the network node if the service quality of the first computing node when executing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to the first threshold value. For example, if the actual computing energy consumption of the first computing node when executing the target task according to the first method is greater than the computing energy consumption indicated by the first service quality requirement, and the deviation between the two is greater than the first threshold value corresponding to the computing energy consumption, and / or the actual computing amount of the first computing node when executing the target task according to the first method is greater than the computing amount indicated by the first service quality requirement, and the deviation between the two is greater than the first threshold value corresponding to the computing amount, and / or the actual computing delay of the first computing node when executing the target task according to the first method is greater than the computing delay indicated by the first service quality requirement, and the deviation between the two is greater than the first threshold value corresponding to the computing delay, then the first computing node sends the first feedback information to the network node, and the first feedback information may indicate at least one of the above deviations.

[0240] Exemplarily, if the condition that triggers the first computing node to send the first feedback information includes: the feedback time indicated by the first condition information arrives, then the first computing node can indicate to the network node the service quality of the first computing node executing the first subtask when the feedback time indicated by the first condition information arrives. Taking the first condition information indicating a duration T1 as an example, after receiving the first condition information, the first computing node can start a timer with a duration of T1. When the timer times out, the first computing node sends the first feedback information to the network node. The first feedback information indicates the service quality of the first computing node executing the target task according to the first method, and / or the first feedback information indicates whether the service quality of the first computing node executing the target task according to the first method meets the first service quality.

[0241] Exemplarily, if the conditions that trigger the first computing node to send the first feedback information include: the service quality of the first computing node performing the target task according to the first method does not meet the first service quality requirement and the feedback time indicated by the first condition information arrives, then the first computing node may send the first feedback information if the service quality of the target task performed according to the first method does not meet the first service quality requirement, and / or the first computing node sends the first feedback information when the feedback time indicated by the first condition information arrives.

[0242] It is understood that the logic for the second and third computing nodes to send the first feedback information is similar to that of the first computing node. For details, see the description of the first computing node sending the first feedback information. Furthermore, the first threshold value corresponding to the second computing node, the threshold value corresponding to the third computing node, and the threshold value corresponding to the first computing node may be different, and the feedback time corresponding to the second computing node, the feedback time corresponding to the third computing node, and the feedback time corresponding to the first computing node may be different.

[0243] S504: The network node determines a second scheduling result according to the first feedback information.

[0244] In the present application, the second scheduling result may indicate a second method for at least one computing node to execute the target task. The second method differs from the first method in that it may indicate at least one computing node to execute a second subtask within the target task. For example, the second scheduling result may include an identifier for the second subtask. It should be understood that the second scheduling result may also indicate at least one computing node to execute the entire target task, without limitation.

[0245] Optionally, the network node may further determine a computing service quality requirement corresponding to the second scheduling result, i.e., a third service quality requirement. The third service quality requirement may indicate the service quality that must be met by at least one computing node in executing the target task according to the second method. The content of the third service quality requirement is similar to that of the first service quality requirement, and reference may be made to the description of the first service quality requirement above. Furthermore, the process by which the network node determines the third service quality requirement is similar to the process by which the network node determines the first service quality requirement in the following embodiments, and reference may be made to the corresponding description in S500b below.

[0246] As will be appreciated, for scenario 1, the network node determines the second scheduling result based on the first feedback information sent by the first computing node. For scenario 2, the network node determines the second scheduling result based on the first feedback information sent by the second computing node and / or the first feedback information sent by the third computing node. The following describes scenario 1 in detail.

[0247] Exemplarily, if the first feedback information indicates that the computing type of the first computing node does not meet the computing type indicated by the first quality of service requirement, the network node can adjust the scheduling strategy of the target task so that the computing type of the subtask executed by the first computing node is the computing type it supports. In other words, the computing type corresponding to the second subtask or the computing type indicated by the third quality of service is the computing type supported by the first computing node. Taking the target task as Task 2 shown in Figure 1F as an example, if the first scheduling result instructs the first computing node to execute S1 to S3, where the computing type corresponding to S1 to S2 is CPU and the computing type corresponding to S3 is GPU, and the first computing node supports CPU but not GPU, then the second scheduling result instructs the first computing node to execute S1 to S2, and the computing type indicated by the third quality of service requirement is required to be CPU.

[0248] Exemplarily, if the first feedback information indicates that the computational load of the first computing node executing the target task according to the first method does not meet the computational load requirement indicated by the first quality of service requirement, or the first feedback information indicates that the computational load of the first computing node executing the target task according to the first method does not meet the computational load requirement indicated by the first quality of service requirement, and the deviation between the two is greater than or equal to the first threshold value, then the network node can adjust the scheduling strategy of the target task to a task splitting mode with a smaller computational load on the first computing node, such as reducing the amount of tasks executed by the first computing node to avoid excessive computational load on the first computing node. In other words, the computational load of the first computing node executing the target task according to the second method is less than the computational load of the first computing node executing the target task according to the first method. Optionally, the computational load requirement indicated by the third quality of service requirement is lower than or equal to the computational load requirement indicated by the first quality of service requirement. Taking Task 2 shown in Figure 1F as an example, if the first scheduling result instructs the first computing node to execute S1 to S3, and the I / O bandwidth indicated by the first quality of service requirement is 5 Gbits / S, then the second scheduling result can instruct the first computing node to execute S1 or S1 to S2, and the I / O bandwidth indicated by the third quality of service requirement is 3 Gbits / S.

[0249] For example, if the first feedback information indicates that the computing delay when the first computing node performs the target task according to the first method does not meet the computing delay requirement indicated by the first quality of service requirement, or the first feedback information indicates that the computing delay when the first computing node performs the target task according to the first method does not meet the computing delay requirement indicated by the first quality of service requirement, and the deviation between the two is greater than or equal to the first threshold value, then the network node can adjust the scheduling strategy of the target task to a task splitting mode with a smaller computing delay for the first computing node, such as shutting down certain functions of the target task, or reducing the computing accuracy, or replacing complex algorithms with simple algorithms, etc., to meet the computing delay requirement and ensure end-to-end delay. Taking the first computing node as a cloud server as an example, when the first computing node feedbacks that its computing delay deviates from the delay indicated by the first quality of service requirement by more than 10ms, the network node can instruct the first computing node to shut down some rendering effect functions, such as shutting down reflection rendering and / or dynamic diffuse global illumination (DDGI) rendering. It can be understood that the computing delay of the first computing node performing the target task according to the second method is less than the computing delay of the first computing node performing the target task according to the first method. Optionally, the calculation delay requirement indicated by the third quality of service requirement is lower than or equal to the calculation delay requirement indicated by the first quality of service requirement.

[0250] Exemplarily, if the first feedback information indicates that the computing energy consumption of the first computing node when executing the target task according to the first method does not meet the computing energy consumption requirement indicated by the first quality of service requirement, or the first feedback information indicates that the computing energy consumption of the first computing node when executing the target task according to the first method does not meet the computing energy consumption requirement indicated by the first quality of service requirement, and the deviation between the two is greater than or equal to the first threshold value, then the network node can adjust the scheduling strategy of the target task to a task splitting mode with lower computing energy consumption of the first computing node, such as reducing the amount of tasks executed by the first computing node to avoid excessive computing energy consumption of the first computing node. In other words, the computing energy consumption of the first computing node executing the target task according to the second method is less than the computing energy consumption of the first computing node executing the target task according to the first method. Optionally, the computing energy consumption requirement indicated by the third quality of service requirement is lower than or equal to the computing energy consumption requirement indicated by the first quality of service requirement. Taking Task 2 shown in Figure 1F as an example, where the target task is Task 2, if the first scheduling result instructs the first computing node to execute S1 to S3, and the computing energy consumption indicated by the first quality of service requirement is 5 watts, then the second scheduling result may instruct the first computing node to execute S1 or S1 to S2, and the computing energy consumption indicated by the third quality of service requirement is 5 watts.

[0251] For example, if the first feedback information indicates that the computational accuracy of the first computing node when executing the target task according to the first method does not meet the computational accuracy requirement indicated by the first quality of service requirement, or if the first feedback information indicates that the computational accuracy of the first computing node when executing the target task according to the first method does not meet the computational accuracy requirement indicated by the first quality of service requirement, and the deviation between the two is greater than or equal to a first threshold, the network node may adjust the scheduling policy for the target task so that the first computing node executes a subtask with lower computational accuracy. In other words, the computational accuracy of the first computing node when executing the target task according to the second method is lower than the computational accuracy of the first computing node when executing the target task according to the first method. Optionally, the computational accuracy requirement indicated by the third quality of service requirement is lower than the computational accuracy requirement indicated by the first quality of service requirement. Taking the target task as Task 2 shown in Figure 1F as an example, if the first scheduling result instructs the first computing node to execute S1-S3, where the computational accuracy requirement for S1 is 10%, the computational accuracy requirements for S2 and S3 are 3%, and the computational accuracy indicated by the first quality of service requirement is 3%, then the second scheduling result may instruct the first computing node to execute S1, with the computational accuracy indicated by the third quality of service requirement being 10%.

[0252] For example, if the first feedback information indicates that the computational amount of the target task executed by the first computing node according to the first method does not meet the computational amount requirement indicated by the first quality of service requirement, and the computational delay when the first computing node executes the target task according to the first method does not meet the computational delay requirement indicated by the first quality of service requirement, then the network node can adjust the scheduling strategy of the target task to a task splitting mode in which the computational load of the first computing node is smaller and the computational delay is smaller.

[0253] It can be understood that for scenario 2, the logic of the network node determining the second scheduling result is similar to that in scenario 1. The difference is that when the second computing node and the third computing node both send the first feedback information, the network node must consider not only the situation of the second computing node but also the situation of the third computing node.

[0254] For example, if the first feedback information sent by the second computing node indicates that the computing delay when the second computing node performs the target task according to the first method does not meet the computing delay requirement indicated by the service quality requirement of the second computing node, and the first feedback information sent by the third computing node indicates that the computing energy consumption when the third computing node performs the target task according to the first method does not meet the computing energy consumption requirement indicated by the service quality requirement of the third computing node, then the network node can adjust the scheduling strategy of the target task to a task splitting mode in which the computing delay of the second computing node is smaller and the computing energy consumption of the third computing node is smaller.

[0255] Optionally, the network node may send the second scheduling result and the third quality of service requirement to the at least one computing node. Accordingly, the at least one computing node receives the second scheduling result and the third quality of service requirement. Thereafter, the at least one computing node may execute the target task based on the second scheduling result.

[0256] It is understandable that the computing node to which the network node sends the second scheduling result and the computing node to which the network node sends the first scheduling result may be the same or different. For example, if the first scheduling result instructs the terminal to perform all subtasks of the target task, the network node sends the first scheduling result to the terminal, and the second scheduling result instructs the terminal and the cloud to respectively perform a portion of the subtasks of the target task, then the network node sends the second scheduling result to the terminal and the cloud. For another example, if the first scheduling result instructs the terminal and the cloud to perform the target task in a first manner, then the network node sends the first scheduling result to the terminal and the cloud, and the second scheduling result instructs the terminal and the cloud to perform the target task in a second manner, then the network node sends the second scheduling result to the terminal and the cloud.

[0257] Optionally, if the service quality of at least one computing node executing the target task according to the second scheduling result does not meet the third service quality requirement, at least one computing node can send feedback information to the network node so that the network device can continue to adjust the scheduling strategy of the target task according to the feedback information to ensure the computing service quality of the target task.

[0258] Optionally, the network node may also send parameter information required to execute the target task according to the second scheduling result to at least one computing node, such as one or more of the AI ​​model, algorithm or program information used to execute the target task.

[0259] Based on the method shown in FIG5 , the network node can send a first scheduling result and a first service quality requirement to at least one computing node. After receiving the first scheduling result and the first service quality requirement, at least one computing node can execute the target task according to the first scheduling result and send a first feedback message to the network node to indicate that the service quality of executing the target task according to the first scheduling result cannot meet the first service quality requirement. After receiving the first feedback message, the network node can adjust the scheduling strategy of the target task in combination with the first feedback message, so that the adjusted scheduling result (such as the second scheduling result) is more suitable for the computing node, so as to ensure the computing service quality of the target task and improve the user experience. It can be understood that the computing power of the computing node (such as the available computing resources of the computing node) is usually dynamically changing. Through the method shown in FIG5 , this change can be fed back to the network node, so that the second scheduling result determined by the network node is more in line with the actual situation of the computing node, thereby ensuring the computing service quality of the target task.

[0260] Optionally, in a possible implementation of the method shown in FIG5 , the network node may further send a second quality of service requirement to the communication node so that the communication node can determine the quality of service it needs to provide and provide corresponding transmission capabilities based on the quality of service. Specifically, as shown in FIG6 , the method shown in FIG5 may further include the following steps:

[0261] S501a: The network node sends a second quality of service requirement to the communication node. Correspondingly, the communication node receives the second quality of service requirement from the network node.

[0262] In the present application, the communication node may be a RAN node. Optionally, the functions of the network node may be deployed on the RAN node.

[0263] In the present application, the second quality of service requirement may indicate the quality of service that the communication node needs to provide. Optionally, the second quality of service requirement may indicate a quality of service requirement related to communication. For example, the second quality of service requirement may indicate at least one of the following: packet delay budget (PDB), guaranteed bit rate (GBR), guaranteed flow bit rate (GFBR), maximum data burst volume (MDBV), packet error rate (PER) or guaranteed transmission data size. It can be understood that the above-mentioned quality of service information may be the quality of service requirement corresponding to uplink transmission, or the quality of service requirement corresponding to downlink transmission, or the quality of service requirement for uplink transmission and downlink transmission together. Taking PDB as an example, the PDB may be an uplink PDB, indicating the PDB that needs to be met for uplink transmission; or, the PDB may be a downlink PDB, indicating the PDB that needs to be met for downlink transmission; or, the PDB may be the sum of uplink and downlink PDBs, indicating the PDB that needs to be met for uplink transmission and downlink transmission in total.

[0264] It should be understood that the above requirements are merely examples of communication-related QoS requirements that may be indicated by the second QoS requirement. In specific applications, the second QoS requirement may indicate more or fewer QoS requirements than the above, without limitation.

[0265] It is understandable that S501a can be executed before S501, or after S501, or simultaneously with S501, without limitation.

[0266] S503a: The communication node sends second feedback information to the network node. Correspondingly, the network node receives the second feedback information from the communication node.

[0267] In one possible implementation, if the quality of service provided by the communication node cannot meet the second quality of service requirement, the communication node sends second feedback information to the network node. The second feedback information can be used to indicate that the quality of service provided by the communication node cannot meet the second quality of service requirement. For example, the second feedback information includes 1 bit, and when the value of the 1 bit is "0" or "1", it indicates that the quality of service provided by the communication node cannot meet the second quality of service requirement. The "quality of service provided by the communication node" here can be the service quality of a single detection of the communication node, or the service quality counted within a certain time period or cycle, such as the average value of the service quality counted within the time period or cycle, the maximum value of the service quality, or the minimum value of the service quality, etc.

[0268] Optionally, the second feedback information may indicate the quality of service provided by the communication node, so that the network node can adjust the scheduling policy for the target task based on the quality of service and the second quality of service requirement. For example, the second feedback information may indicate the packet transmission delay provided by the communication node, and / or the bit rate provided by the communication node, and / or the stream bit rate provided by the communication node, and / or the MDBV provided by the communication node, and / or the PER provided by the communication node, and / or the guaranteed transmission data size provided by the communication node. Alternatively, the second feedback information may also indicate the deviation between the quality of service provided by the communication node and the second quality of service requirement, so that the network node can adjust the scheduling policy for the target task based on the deviation.

[0269] Exemplarily, if the packet transmission delay provided by the communication node is greater than the second quality of service requirement (PDB), the second feedback information indicates that the packet transmission delay provided by the communication node does not meet the PDB indicated by the second quality of service requirement. Optionally, the second feedback information also indicates the packet transmission delay provided by the communication node or the deviation between the packet transmission delay provided by the communication node and the PDB indicated by the second quality of service requirement. And / or, if the bit rate provided by the communication node is less than the second quality of service requirement (GBR), the second feedback information indicates that the bit rate provided by the communication node does not meet the GBR indicated by the second quality of service requirement. Optionally, the second feedback information also indicates the bit rate provided by the communication node or the deviation between the bit rate provided by the communication node and the GBR indicated by the second quality of service requirement. And / or, if the stream bit rate provided by the communication node is less than the second quality of service requirement (GFBR), the second feedback information indicates that the stream bit rate provided by the communication node does not meet the GFBR indicated by the second quality of service requirement. Optionally, the second feedback information also indicates the stream bit rate provided by the communication node or the deviation between the stream bit rate provided by the communication node and the GFBR indicated by the second quality of service requirement. And / or, if the MDBV provided by the communication node is less than the second quality of service requirement MDBV, the second feedback information indicates that the MDBV provided by the communication node does not meet the MDBV indicated by the second quality of service requirement. Optionally, the second feedback information further indicates the MDBV provided by the communication node or the deviation between the MDBV provided by the communication node and the MDBV indicated by the second quality of service requirement. And / or, if the PER provided by the communication node is greater than the second quality of service requirement PER, the second feedback information indicates that the PER provided by the communication node does not meet the PER indicated by the second quality of service requirement. Optionally, the second feedback information further indicates the PER provided by the communication node or the deviation between the PER provided by the communication node and the PER indicated by the second quality of service requirement. And / or, if the transmission data size provided by the communication node is less than the guaranteed transmission data size indicated by the second quality of service requirement, the second feedback information indicates that the transmission data size provided by the communication node does not meet the guaranteed transmission data size indicated by the second quality of service requirement. Optionally, the second feedback information further indicates the transmission data size provided by the communication node or the deviation between the transmission data size provided by the communication node and the guaranteed transmission data size indicated by the second quality of service requirement.

[0270] Optionally, the network node sends second condition information to the communication node. The second condition information may indicate conditions for triggering the second feedback information. Thus, after receiving the second condition information, the communication node may determine whether to trigger the second feedback information, or when to trigger the second feedback information, based on the second condition information. The network node may send the second condition information to the communication node before S503a. For example, the network node may send the second condition information simultaneously with the second quality of service requirement, or the network node may send the first condition information after S501a and before S502, without limitation.

[0271] In one possible design, the conditions for triggering the second feedback information include at least one of the following: the quality of service provided by the communication node does not meet the second quality of service requirement; or, the quality of service provided by the communication node does not meet the second quality of service requirement, and the deviation between the two is greater than or equal to the second threshold value; or, the feedback time indicated by the second condition information arrives.

[0272] Exemplarily, if the condition triggering the communication node to send the second feedback information includes: the service quality provided by the communication node does not meet the second service quality requirement, then the communication node can send the second feedback information to the network node if the service quality it provides does not meet the second service quality requirement.

[0273] Exemplarily, if the conditions that trigger the communication node to send the second feedback information include: the service quality provided by the communication node does not meet the second service quality requirement, and the deviation between the two is greater than or equal to the second threshold value, then the communication node may send the second feedback information to the network node when the service quality it provides does not meet the second service quality requirement and the deviation between the two is greater than or equal to the second threshold value. For example, if the packet transmission delay actually provided by the communication node is greater than the second service quality requirement indication PDB, and the deviation between the two is greater than the second threshold value corresponding to the packet transmission delay, and / or, the bit rate actually provided by the communication node is greater than the second service quality requirement indication GBR, and the deviation between the two is greater than the second threshold value corresponding to the bit rate, and / or, the stream bit rate actually provided by the communication node is greater than the second service quality requirement indication GFBR, and the deviation between the two is greater than the second threshold value corresponding to the stream bit rate, and / or, the MDBV actually provided by the communication node is less than the second service quality requirement indication MDBV, and the deviation between the two is greater than the second threshold value corresponding to the MDBV, and / or, if the PER actually provided by the communication node is greater than the second service quality requirement indication PER, and the deviation between the two is greater than the second threshold value corresponding to the PER, then the communication node sends second feedback information to the network node, and the second feedback information may indicate at least one of the above deviations.

[0274] Exemplarily, if the condition triggering the communication node to transmit the second feedback information includes the arrival of a feedback time indicated by the second condition information, the communication node may indicate the quality of service provided by the communication node to the network node upon arrival of the feedback time indicated by the second condition information. For example, if the second condition information indicates a duration T2, upon receiving the second condition information, the communication node may start a timer with a duration of T2. When the timer expires, the communication node transmits the second feedback information to the network node. The second feedback information indicates the quality of service provided by the communication node and / or indicates whether the quality of service provided by the communication node meets the second quality of service.

[0275] Exemplarily, if the conditions that trigger the communication node to send the second feedback information include: the service quality provided by the communication node does not meet the second service quality requirement, and the deviation between the two is greater than or equal to the second threshold value, and the feedback time indicated by the second condition information arrives, then the communication node may send the second feedback information when the provided service quality does not meet the second service quality requirement and the deviation between the two is greater than or equal to the second threshold value, and / or the communication node sends the second feedback information when the feedback time indicated by the second condition information arrives.

[0276] It is understandable that S503a can be executed before S503, or after S503, or simultaneously with S503, without limitation.

[0277] It is understandable that if the communication node sends the second feedback information to the network node, and at least one computing node does not send the first feedback information to the network node, the network node can determine the second scheduling result based on the second feedback information. In this way, the adjusted scheduling result (such as the second scheduling result) can be better adapted to the current communication capabilities of the network, thereby ensuring the communication service quality of the target task and improving the user experience.

[0278] Optionally, the network node may also determine a communication service quality requirement corresponding to the second scheduling result, namely, a fourth service quality requirement. The fourth service quality requirement may indicate the service quality that the communication node must meet when at least one computing node executes the target task according to the second scheduling result. The content of the fourth service quality requirement is similar to that of the second service quality requirement, and reference may be made to the description of the second service quality requirement above. Furthermore, the process by which the network node determines the fourth service quality requirement is similar to the process by which the network node determines the second service quality requirement in the following embodiments, and reference may be made to the corresponding description in S500b below.

[0279] Exemplarily, if the second feedback information indicates that the packet transmission delay provided by the communication node does not meet the PDB indicated by the second quality of service requirement, the network node can adjust the scheduling strategy of the target task to a task splitting mode with looser network PDB requirements, such as increasing the number of at least one computing node to meet the delay requirements. In other words, the PDB indicated by the fourth quality of service requirement is greater than the PDB indicated by the second quality of service requirement. Taking the target task as a video rendering task as an example, if the first scheduling result indicates that the first computing node performs all tasks of the video rendering task, and the second quality of service requirement indicates that the PDB is 30ms, then the second scheduling result can instruct the first computing node and one or more nodes other than the first computing node to collaboratively perform the video rendering task, and the PDB indicated by the fourth quality of service requirement is 50ms.

[0280] For example, if the second feedback information indicates that the bit rate provided by the communication node does not meet the GBR indicated by the second quality of service requirement, the network node may adjust the scheduling strategy of the target task to a task splitting mode with more relaxed network GBR requirements, such as adjusting the scheduling strategy of the target task from the pure cloud rendering mode in Table 1 to the end-cloud collaborative rendering mode to meet the transmission rate requirement. In other words, the GBR indicated by the fourth quality of service requirement is less than the GBR indicated by the second quality of service requirement.

[0281] For example, if the second feedback information indicates that the stream bit rate provided by the communication node does not meet the GFBR indicated by the second quality of service requirement, the network node may adjust the scheduling strategy of the target task to a task splitting mode with more relaxed network GFBR requirements, such as adjusting the scheduling strategy of the target task from candidate split point 1 in Table 2 to candidate split point 2 to meet the transmission rate requirement. In other words, the GFBR indicated by the fourth quality of service requirement is less than the GFBR indicated by the second quality of service requirement.

[0282] For example, if the second feedback information indicates that the MDBV provided by the communication node does not meet the MDBV indicated by the second quality of service requirement, the network node may adjust the scheduling strategy of the target task to a task splitting mode with more relaxed network MDBV requirements, such as adjusting the scheduling strategy of the target task from 4K screen rendering to 2K screen rendering to meet the data volume requirement. In other words, the MDBV indicated by the fourth quality of service requirement is smaller than the MDBV indicated by the second quality of service requirement.

[0283] For example, if the second feedback information indicates that the PER provided by the communication node does not meet the PER indicated by the second quality of service requirement, the network node may adjust the scheduling strategy of the target task to a task splitting mode that has more relaxed requirements on the network PER, such as adjusting the scheduling strategy of the target task from rendering at a frame rate of 90 fps to rendering at a frame rate of 60 fps, to meet the PER requirement. In other words, the PER indicated by the fourth quality of service requirement is greater than the PER indicated by the second quality of service requirement.

[0284] For example, if the second feedback information indicates that the transmission data size provided by the communication node does not meet the guaranteed transmission data size indicated by the second quality of service requirement, the network node may adjust the scheduling policy for the target task to a task splitting mode with more relaxed network transmission data size requirements, such as adjusting the scheduling policy for the target task from enabling DDGI rendering in the cloud to disabling DDGI rendering in the cloud, in order to meet the transmission data size requirement. In other words, the guaranteed transmission data size indicated by the fourth quality of service requirement is smaller than the guaranteed transmission data size indicated by the second quality of service requirement.

[0285] It is understandable that if the communication node sends the second feedback information to the network node and at least one computing node sends the first feedback information to the network node, the network node may determine the second scheduling result by combining the first feedback information and the second feedback information. For example, S504 may be replaced by the following steps:

[0286] S504a: The network node determines a second scheduling result according to the first feedback information and the second feedback information.

[0287] It can be understood that the network node determines the second scheduling result based on the first feedback information and the second feedback information, which is similar to the logic of the network node determining the second scheduling result based on the first feedback information and the second feedback information respectively. The difference is that in S504a, the network node not only considers the first feedback information but also the second feedback information.

[0288] For example, if the first feedback information indicates that the computing energy consumption of the first computing node when executing the target task according to the first method does not meet the computing energy consumption requirement indicated by the first quality of service requirement, and the second feedback information indicates that the packet transmission delay provided by the communication node does not meet the PDB indicated by the second quality of service requirement, then the network node can adjust the scheduling strategy of the target task to a task splitting mode in which the computing energy consumption of the first computing node is lower and the network PDB requirements are more relaxed, so as to avoid excessive computing energy consumption of the first computing node and meet the delay requirements.

[0289] It can be understood that the state of the channel usually changes dynamically, so the service quality provided by the communication node also changes dynamically. The communication node can feed back this change to the network node by sending second feedback information to the network node, so that the second scheduling result determined by the network node is more in line with the actual situation of the communication node, thereby ensuring the communication service quality of the target task.

[0290] Optionally, the network node may send a fourth quality of service requirement to the communication node, and accordingly, the communication node receives the fourth quality of service requirement.

[0291] Optionally, if the service quality provided by the communication node does not meet the fourth service quality requirement, the communication node can send feedback information to the network node so that the network device can continue to adjust the scheduling strategy of the target task according to the feedback information to ensure the communication service quality of the target task.

[0292] Optionally, in a possible implementation of the method shown in FIG5 , the network node may obtain the network status, the quality of service requirements of the target task, and the computing power of at least one computing node to determine the first scheduling result based on this information. Specifically, as shown in FIG6 , the method shown in FIG5 further includes the following steps:

[0293] S500a: The network node obtains the network status, the service quality requirement of the target task, and the computing power of at least one computing node.

[0294] In this application, the network status may indicate one or more of the current network's PDB, GBR, GFBR, MDBV, PER, or guaranteed transmission data size. The above information may be the network status of uplink transmission, the network status of downlink transmission, or the network status of both uplink and downlink transmission. In addition, the above information is only an example of the network status. In specific applications, the network status may include more or less information than the above information. For example, the network status also indicates information such as uplink / downlink channel status, uplink / downlink available bandwidth resources, or uplink / downlink network congestion status. The congestion status information may be information related to the queue length of data packets in the layer 1 or layer 2 protocol stack on the RAN side, such as buffer status report (BSR) related information.

[0295] In this application, the QoS requirement of the target task may include the communication QoS requirement of the target task and the computing QoS requirement of the target task. The communication QoS requirement of the target task may indicate at least one of the following: the PDB of the target task, the GBR of the target task, the GFBR of the target task, the MDBV of the target task, the PER of the target task, or the guaranteed transmission data size of the target task. Alternatively, the communication QoS requirement of the target task may indicate at least one of the following: the PDB of each subtask in the target task, the GBR of each subtask in the target task, the GFBR of each subtask in the target task, the MDBV of each subtask in the target task, the PER of each subtask in the target task, or the guaranteed transmission data size of each subtask in the target task.

[0296] The computing service quality requirement of the target task may indicate at least one of the following: the computing type requirement of the target task, the computing amount requirement of the target task, the computing delay requirement of the target task, the computing energy consumption requirement of the target task, or the computing accuracy requirement of the target task. Alternatively, the computing service quality requirement of the target task may indicate at least one of the following: the computing type requirement of each subtask in the target task, the computing amount requirement of each subtask in the target task, the computing delay requirement of each subtask in the target task, the computing energy consumption requirement of each subtask in the target task, or the computing accuracy requirement of each subtask in the target task. The above-mentioned computing type requirement, computing amount requirement, computing delay requirement, computing energy consumption requirement, or computing accuracy requirement are described above.

[0297] In this application, the computing power of at least one computing node may indicate at least one of the following: the computing type supported by at least one computing node, the computing amount supported by at least one computing node, the computing delay supported by at least one computing node, the computing energy consumption supported by at least one computing node, or the computing accuracy supported by at least one computing node.

[0298] In one possible implementation, the network node may periodically or irregularly monitor the network status to obtain the latest network status. Alternatively, the RAN node periodically reports the latest network status to the network node. Alternatively, the network node triggers the RAN node to report the latest network status to the network node.

[0299] In one possible implementation, at least one computing node periodically reports its computing power to the network node. Alternatively, the network node triggers at least one computing node to report its computing power to the network node.

[0300] In one possible implementation, at least one computing node, such as a terminal or application server, sends the quality of service requirements of a target task to a network node. For example, the first computing node, the second computing node, or the third computing node sends a computing service request to the network node. The computing service request includes information about the quality of service requirements of the target task. After receiving the computing service request, the network node can obtain the quality of service requirements of the target task. Alternatively, the network node obtains the quality of service requirements of the target task from a core network element. For example, the first computing node, the second computing node, or the third computing node sends a computing service request to the network node. The computing service request includes a service identifier corresponding to the target task. After receiving the computing service request, the network node obtains the quality of service requirements of the target task based on the service identifier. For example, the network node sends the service identifier to a PCF network element, and the PCF network element can locally obtain the quality of service requirements of the target task based on the service identifier, or the PCF network element can obtain the quality of service requirements of the target task from an application server based on the service identifier. Thereafter, the PCF network element can send the quality of service requirements of the target task to the network node.

[0301] The service identifier can be used to indicate the service corresponding to the target task, such as a rendering service or an image recognition service. A service can correspond to one or more tasks, such as a rendering service can correspond to multiple rendering tasks using different rendering methods.

[0302] S500b: The network node determines a first scheduling result according to the network status, the quality of service requirement of the target task, and the computing power of at least one computing node.

[0303] In one possible design, the manner in which at least one computing node executes the target task, i.e., the manner in which the target task is divided, corresponds to different network states and different computing powers of the computing nodes. The network node may determine a first scheduling result based on the information obtained in S500b and the above-mentioned correspondence.

[0304] As an example, the network node may determine, based on the corresponding relationship, at least one splitting method for the target task that satisfies the network state and the computing power of at least one computing node, determine a splitting method from these splitting methods that satisfies the quality of service requirements of the target task, and determine this splitting method as the first scheduling result. It will be understood that if there are multiple splitting methods that satisfy the quality of service requirements of the target task, the network node may select one of them as the first scheduling result, or the network node may indicate the multiple splitting methods to at least one computing node, and the at least one computing node may determine the first scheduling result.

[0305] For example, if the target task can be executed by both compute node 1 and compute node 2, the network status includes the uplink PDB, and the computing power of the compute nodes includes computing energy consumption, the above correspondence can be shown in Table 4. If the network status indicates that the current network uplink PDB is 20ms, the computing energy consumption supported by compute node 1 is 3 watts, and the computing energy consumption supported by compute node 2 is 5 watts, then according to Table 4, it can be determined that the target task splitting methods that meet the network status, computing power of compute node 1, and computing power of compute node 2 include splitting methods 1 to 3. Subsequently, the network node can determine the splitting method that meets the quality of service requirements of the target task from splitting methods 1 to 3. For example, if the communication service quality indicator PDB of the target task is 45ms, and the computing service quality indicator of the target task is 8 watts, then according to Table 4, it can be determined that the splitting methods that meet the service quality requirements of the target task are splitting method 2 and splitting method 3. The network node can determine splitting method 2 or splitting method 3 as the first scheduling result, or the network node can indicate these two splitting methods to computing node 1 or computing node 2, and computing node 1 or computing node 2 determines the first scheduling result, or computing node 1 or computing node 2 negotiates the first scheduling result. Exemplarily, computing node 1 can be a terminal, computing node 2 can be a cloud, splitting method 1 can be that neither the terminal nor the cloud performs DDGI rendering, splitting method 2 can be that the terminal performs DDGI rendering and the cloud does not perform DDGI rendering, and splitting method 3 can be that the cloud performs DDGI rendering and the terminal does not perform DDGI rendering.

[0306] Table 4

[0307] It can be understood that Table 4 is only an example of the division method of the target task and the correspondence between different network states and different computing powers of computing nodes. In specific applications, the correspondence can also be in other forms, or can include more or fewer rows / columns than Table 4, without limitation.

[0308] As another example, the network node may determine, based on the above-described correspondence, at least one splitting method that meets the quality of service requirements of the target task, determine among these splitting methods a splitting method that meets the network status and the computing power of at least one computing node, and determine this splitting method as the first scheduling result. It will be understood that if there are multiple splitting methods that meet the network status and the computing power of at least one computing node, the network node may select one of them as the first scheduling result, or the network node may indicate the multiple splitting methods to at least one computing node, and the at least one computing node may determine the first scheduling result.

[0309] Still take the corresponding relationship shown in Table 4 as an example. If the communication service quality indication PDB of the target task is 45ms, and the computing service quality indication of the target task is 8 watts, then according to Table 4, it can be determined that the splitting methods that meet the service quality requirements of the target task are splitting method 2 and splitting method 3. Subsequently, the network node can determine the splitting method that meets the network status, the computing power of computing node 1, and the computing power of computing node 2 in splitting method 1 and splitting method 2. For example, if the network status indicates that the uplink PDB of the current network is 35ms, the computing energy consumption supported by computing node 1 is 3 watts, and the computing energy consumption supported by computing node 2 is 5 watts, then according to Table 4, it can be determined that the splitting method that meets the network status, the computing power of computing node 1, and the computing power of computing node 2 is splitting method 3, and the network node can determine splitting method 3 as the first scheduling result.

[0310] In one possible implementation, the network node may determine the first QoS requirement based on the target task's computing QoS requirement and the computing power of at least one computing node. For example, after determining the first scheduling result, the network node may determine the first QoS requirement based on the target task's computing QoS requirement, the computing power of at least one computing node, and the first scheduling result. It is understood that at least one computing node has the capability to provide the first QoS requirement.

[0311] It is understandable that the first quality of service requirement may indicate a set of quality of service requirements (e.g., a set of quality of service requirements indicating at least one of the following: computing type requirement, computing amount requirement, computing delay requirement, computing energy consumption requirement, or computing accuracy requirement), and at least one computing node needs to meet this set of quality of service requirements when performing the target task according to the first method. Alternatively, the first quality of service requirement may indicate multiple sets of quality of service requirements (e.g., each set of quality of service requirements indicates at least one of the following: computing type requirement, computing amount requirement, computing delay requirement, computing energy consumption requirement, or computing accuracy requirement), and at least one computing node needs to meet at least one set of quality of service requirements among these multiple sets of quality of service requirements when performing the target task according to the first method.

[0312] Using Table 4 as an example, if the target task's communication QoS indicator (PDB) is 45ms, the target task's computing QoS indicator (TSQS) indicates the target task's total computing energy consumption is 8 watts, the first scheduling result indicates segmentation method 3, and the first QoS requirement indicates a set of QoS requirements, then the first QoS requirement sent to compute node 1 may indicate that the computing energy consumption of compute node 1 is less than 3 watts, and the first QoS requirement sent to compute node 2 may indicate that the computing energy consumption of compute node 2 is less than 5 watts. Therefore, the computing energy consumption of compute node 1 executing the target task according to the first method must be less than 3 watts, and the computing energy consumption of compute node 2 executing the target task according to the first method must be less than 5 watts. Alternatively, the first QoS requirement sent to compute node 1 may indicate that the computing energy consumption of compute node 1 is less than or equal to 2 watts, and the first QoS requirement sent to compute node 2 may indicate that the computing energy consumption of compute node 2 is less than 6 watts. Therefore, the computing energy consumption of compute node 1 executing the target task according to the first method must be less than or equal to 2 watts, and the computing energy consumption of compute node 2 executing the target task according to the first method must be less than 6 watts.

[0313] Using Table 4 as an example, if the target task's communication QoS indicator (PDB) is 45ms, the target task's computing QoS indicator (TSQS) indicates the target task's total computing energy consumption is 8 watts, the first scheduling result indicates segmentation mode 3, and the first QoS requirement indicates two sets of QoS requirements, then the first QoS requirement sent to computing node 1 may indicate that computing node 1's computing energy consumption is less than or equal to 2 watts, or less than or equal to 1.5, and the first QoS requirement sent to computing node 2 may indicate that computing node 2's computing energy consumption is less than 6 watts, or less than or equal to 5.5 watts. Therefore, the computing energy consumption of computing node 1 executing the target task according to the first mode must be less than or equal to 2 watts, or less than or equal to 1.5 watts, and the computing energy consumption of computing node 2 executing the target task according to the first mode must be less than or equal to 6 watts, or less than or equal to 5.5 watts.

[0314] In one possible implementation, the network node may determine the second QoS requirement based on the network status and the communication QoS requirement of the target task. For example, after determining the first scheduling result, the network node may determine the second QoS requirement based on the network status, the computing QoS requirement of the target task, and the first scheduling result. It is understood that the communication node has the ability to provide the second QoS requirement.

[0315] It is understandable that the second quality of service requirement may indicate a set of quality of service requirements (e.g., a set of quality of service requirements indicating at least one of the following: PDB, GBR, GFBR, MDBV, PER, or guaranteed data size), and the communication node needs to provide the quality of service indicated by this set of quality of service requirements. Alternatively, the second quality of service requirement may indicate multiple sets of quality of service requirements (e.g., each set of quality of service requirements indicating at least one of the following: PDB, GBR, GFBR, MDBV, PER, or guaranteed data size), and the communication node needs to provide the quality of service indicated by at least one set of quality of service requirements among the multiple sets of quality of service requirements.

[0316] Still taking Table 4 as an example, if the communication service quality indicator PDB of the target task is 45ms, the computing service quality indicator of the target task is 8 watts, the first scheduling result indicates segmentation mode 3, and the second service quality requirement indicates a set of service quality requirements, then the second service quality requirement may indicate that the uplink PDB is less than 45ms, so the communication node needs to provide an uplink PDB less than 45ms. Alternatively, the second service quality requirement may indicate that the uplink PDB is less than or equal to 30ms, so the communication node needs to provide an uplink PDB less than or equal to 30ms. Alternatively, the second service quality requirement may indicate that the uplink PDB is less than or equal to 35ms, so the communication node needs to provide an uplink PDB less than or equal to 35ms. If the first scheduling result indicates segmentation mode 2, the second service quality requirement may indicate that the uplink PDB is less than 45ms, so the communication node needs to provide an uplink PDB less than 45ms. Alternatively, the second service quality requirement may indicate that the uplink PDB is less than or equal to 40ms, so the communication node needs to provide an uplink PDB less than or equal to 40ms. Alternatively, the second quality of service requirement may indicate that the uplink PDB is less than or equal to 42 ms, so the communication node needs to provide an uplink PDB that is less than or equal to 42 ms.

[0317] Still taking Table 4 as an example, if the communication service quality indication PDB of the target task is 45ms, the computing service quality indication of the target task is the total computing energy consumption of 8 watts, the first scheduling result indicates the segmentation method 3, and the second service quality requirement indicates two sets of service quality requirements, then the second service quality requirement can indicate that the uplink PDB is less than 45ms or less than or equal to 30ms, so the communication node needs to provide an uplink PDB less than 45ms, or less than or equal to 30ms.

[0318] It is understandable that the target tasks in this application can also be replaced by business traffic patterns, such as business bit rate, resolution, or frame rate. Scheduling target tasks can be understood as adjusting business traffic patterns, such as adjusting business bit rate, resolution, or frame rate. Different business traffic patterns can also have different communication service quality requirements and / or computing service quality requirements. In other words, the business process pattern can be determined using the method provided in this application.

[0319] It can be understood that the actions of the network node or computing node or communication node in the above steps can be executed by the processor 401 in the communication device 40 shown in Figure 4 calling the application code stored in the memory 403, and this application does not impose any restrictions on this.

[0320] The various embodiments mentioned above in this application can be combined without limitation if there is no contradiction between the solutions.

[0321] The above description primarily describes the solution provided by this application from the perspective of interaction between various nodes. Accordingly, this application also provides a communication device, which may be a network node in the above-described method embodiments, or a device comprising such a network node, or a component usable in a network node; alternatively, the communication device may be a computing node in the above-described method embodiments, or a device comprising such a computing node, or a component usable in a computing node. It will be understood that, in order to implement the aforementioned functions, the aforementioned network node or computing node, etc., includes hardware structures and / or software modules corresponding to the respective functions. Those skilled in the art will readily appreciate that, in conjunction with the various exemplary units and algorithmic operations described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is implemented in hardware or in a hardware-driven manner by computer software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0322] The present application can divide the network nodes or computing nodes into functional modules according to the above method examples. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or software functional modules. It is understood that the division of modules in this application is schematic and is only a logical functional division. In actual implementation, other division methods may be used.

[0323] For example, FIG7 shows a schematic diagram of the structure of a communication device 70, where the functional modules are divided in an integrated manner. Communication device 70 includes an interface module 701 and a processing module 702. Interface module 701, also known as an interface unit, is used to perform transceiver operations and may be, for example, an interface circuit, a transceiver, a transceiver, or a communication interface. Processing module 702, also known as a processing unit, is used to perform operations other than transceiver operations and may be, for example, a processing circuit or a processor.

[0324] In some embodiments, the communication device 70 may further include a storage module (not shown in FIG. 7 ) for storing program instructions and data.

[0325] In some embodiments, the communication device 70 may further include an AI module (not shown in FIG. 7 ) for implementing AI-related functions. The AI ​​module may implement AI functions through software, hardware, or a combination of software and hardware. For example, the AI ​​module includes an RIC module. Optionally, the AI ​​module and the storage module are integrated into a single module, or the AI ​​module and the processing module 702 are integrated into a single module.

[0326] Exemplarily, the communication device 70 is used to implement the function of a network node. The communication device 70 is, for example, the network node described in the embodiment shown in FIG5 or the embodiment shown in FIG6.

[0327] The interface module 701 is configured to send a first scheduling result and a first quality of service requirement to at least one computing node. The first scheduling result may indicate a first method for the at least one computing node to execute a target task. For example, the interface module 701 may be configured to execute S501.

[0328] The interface module 701 is further configured to receive first feedback information from at least one computing node. The first feedback information may indicate that the quality of service of the at least one computing node performing the target task according to the first method fails to meet a first quality of service requirement. For example, the interface module 701 may also be configured to execute S503.

[0329] Processing module 702 is configured to determine a second scheduling result based on the first feedback information. The second scheduling result may indicate a second method for at least one computing node to execute the target task, and the second method may be different from the first method. For example, processing module 702 may be configured to execute S504.

[0330] In a possible implementation, the interface module 701 is further configured to send first condition information, where the first condition information is used to indicate a condition for triggering the first feedback information.

[0331] In one possible implementation, the conditions for triggering the first feedback information include at least one of the following: the service quality of at least one computing node performing the target task according to the first method does not meet the first service quality requirement; or, the service quality of at least one computing node performing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to the first threshold value; or, the feedback time indicated by the first condition information arrives.

[0332] In a possible implementation, the processing module 702 is further configured to determine a first quality of service requirement based on a computing quality of service requirement of the target task and a computing power of at least one computing node.

[0333] In one possible implementation, the computing service quality requirement or the first service quality requirement of the target task indicates at least one of the following: computing type requirement, computing amount requirement, computing delay requirement, computing energy consumption requirement, or computing accuracy requirement.

[0334] In one possible implementation, the interface module 701 is further used to send a second quality of service requirement to the communication node; the interface module 701 is further used to receive second feedback information from the communication node, the second feedback information indicating that the quality of service provided by the communication node cannot meet the second quality of service requirement; the processing module 702 is specifically used to determine a second scheduling result based on the first feedback information and the second feedback information.

[0335] In a possible implementation, the interface module 701 is further configured to send second condition information, where the second condition information is used to indicate a condition for triggering the second feedback information.

[0336] In one possible implementation, the conditions for triggering the second feedback information include at least one of the following: the quality of service provided by the communication node cannot meet the second quality of service requirement; or, the quality of service provided by the communication node cannot meet the second quality of service requirement, and the deviation between the two is greater than or equal to the second threshold value; or, the feedback time indicated by the second condition information arrives.

[0337] In a possible implementation, the processing module 702 is further configured to determine a second quality of service requirement according to the network state and the communication quality of service requirement of the target task.

[0338] In one possible implementation, the communication service quality requirement or the second service quality requirement of the target task indicates at least one of the following: packet delay budget, guaranteed bit rate, guaranteed flow bit rate, maximum burst data volume, packet error rate or guaranteed transmission data size.

[0339] In one possible implementation, the processing module 702 is also used to obtain the network status, the service quality requirements of the target task, and the computing power of at least one computing node, where the service quality requirements of the target task include the communication service quality requirements of the target task and the computing service quality requirements of the target task; the processing module 702 is also used to determine the first scheduling result based on the network status, the service quality requirements of the target task, and the computing power of at least one computing node.

[0340] In a possible implementation, the processing module 702 is specifically configured to receive a computing service request through the interface module 701 , where the computing service request includes service quality requirement information of a target task.

[0341] In one possible implementation, the processing module 702 is specifically configured to receive a computing service request through the interface module 701, where the computing service request includes a service identifier corresponding to the target task; and the processing module 702 is further specifically configured to obtain a service quality requirement of the target task based on the service identifier.

[0342] In a possible implementation, the interface module 701 is further configured to send the second scheduling result.

[0343] In a possible implementation, the communication device 70 is a radio access network node, a centralized unit, a distributed unit, an intelligent controller of a radio access network, or a core network element.

[0344] In a possible implementation, a first method instructs at least one computing node to execute a first subtask in a target task; a second method instructs at least one computing node to execute a second subtask in the target task.

[0345] When used to implement the function of a network node, for other functions that the communication device 70 can implement, please refer to the relevant introduction of the embodiment shown in Figure 5 or the embodiment shown in Figure 6, and no further details will be given.

[0346] Alternatively, illustratively, the communication device 70 is used to implement the function of a computing node. The communication device 70 is, for example, a computing node in the at least one computing node described in the embodiment shown in FIG5 or the embodiment shown in FIG6.

[0347] The interface module 701 is configured to receive a first scheduling result and a first quality of service requirement. The first scheduling result indicates a first manner in which at least one computing node executes a target task. For example, the interface module 701 may be configured to execute S501.

[0348] The processing module 702 is configured to execute the target task according to the first method. For example, the processing module 702 may be configured to execute S502.

[0349] The interface module 701 is further configured to send first feedback information, wherein the first feedback information is used to indicate that the service quality of executing the target task in the first manner cannot meet the first service quality requirement. For example, the interface module 701 can also be configured to execute S503.

[0350] In a possible implementation, the interface module 701 is further configured to receive first condition information, where the first condition information is used to indicate a condition for triggering the first feedback information.

[0351] In one possible implementation, the condition for triggering the first feedback information includes at least one of the following: the service quality of executing the target task according to the first method does not meet the first service quality requirement; or, the service quality of executing the target task according to the first method does not meet the first service quality requirement, and the deviation between the two is greater than or equal to the first threshold value; or, the feedback time indicated by the first condition information arrives.

[0352] In a possible implementation, the first quality of service requirement indicates at least one of the following: a calculation type requirement, a calculation amount requirement, a calculation delay requirement, a calculation energy consumption requirement, or a calculation accuracy requirement.

[0353] In a possible implementation, the interface module 701 is further configured to send a computing service request, where the computing service request includes at least one of the service quality requirement information of the target task or the service identifier corresponding to the target task.

[0354] In one possible implementation, the interface module 701 is further used to receive a second scheduling result, which indicates a second way for at least one computing node to execute the target task, where the second way is different from the first way; the processing module 702 is further used to execute the target task according to the second scheduling result.

[0355] In a possible implementation, a first method instructs at least one computing node to execute a first subtask in a target task; a second method instructs at least one computing node to execute a second subtask in the target task.

[0356] When used to implement the function of a computing node, for other functions that the communication device 70 can implement, reference can be made to the relevant introduction of the embodiment shown in FIG5 or the embodiment shown in FIG6 , and no further details will be given.

[0357] In a simple embodiment, those skilled in the art may conceive that the communication device 70 may be in the form shown in Figure 4. For example, the processor 401 in Figure 4 may call computer-executable instructions stored in the memory 403 to enable the communication device 70 to execute the method described in the above embodiment.

[0358] Exemplarily, the functions / implementation processes of the interface module 701 and the processing module 702 in FIG7 can be implemented by the processor 401 in FIG4 calling computer-executable instructions stored in the memory 403. Alternatively, the functions / implementation processes of the processing module 702 in FIG7 can be implemented by the processor 401 in FIG4 calling computer-executable instructions stored in the memory 403, and the functions / implementation processes of the interface module 701 in FIG7 can be implemented by the communication interface 404 in FIG4.

[0359] It is understandable that one or more of the above modules or units can be implemented by software, hardware or a combination of the two. When any of the above modules or units is implemented by software, the software exists in the form of computer program instructions and is stored in a memory, and a processor can be used to execute the program instructions and implement the above method flow. The processor can be built into an SoC (system on chip) or an ASIC, or it can be an independent semiconductor chip. In addition to the core used to execute software instructions to perform calculations or processing within the processor, it can further include necessary hardware accelerators, such as field programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits that implement dedicated logic operations.

[0360] When the above modules or units are implemented in hardware, the hardware can be any one or any combination of a CPU, a microprocessor, a digital signal processing (DSP) chip, a microcontroller unit (MCU), an artificial intelligence processor, an ASIC, a SoC, an FPGA, a PLD, a dedicated digital circuit, a hardware accelerator or a non-integrated discrete device, which can run the necessary software or not rely on the software to execute the above method flow.

[0361] Optionally, the present application also provides a chip system, comprising: at least one processor and an interface, wherein the at least one processor is coupled to a memory via the interface, and when the at least one processor executes a computer program or instruction in the memory, the method in any of the above method embodiments is executed. In one possible implementation, the chip system also includes a memory. Optionally, the chip system can be composed of a chip, or can include a chip and other discrete devices, which is not specifically limited in this application.

[0362] Optionally, the present application also provides a computer-readable storage medium. All or part of the processes in the above-mentioned method embodiments can be completed by a computer program to instruct the relevant hardware. The program can be stored in the above-mentioned computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. The computer-readable storage medium can be an internal storage unit of the communication device of any of the above-mentioned embodiments, such as a hard disk or memory of the communication device. The above-mentioned computer-readable storage medium can also be an external storage device of the above-mentioned communication device, such as a plug-in hard disk, a smart memory card (smart media card, SMC), a secure digital (secure digital, SD) card, a flash card (flash card), etc. equipped on the above-mentioned communication device. Furthermore, the above-mentioned computer-readable storage medium can also include both the internal storage unit of the above-mentioned communication device and an external storage device. The above-mentioned computer-readable storage medium is used to store the above-mentioned computer program and other programs and data required by the above-mentioned communication device. The above-mentioned computer-readable storage medium can also be used to temporarily store data that has been output or is to be output.

[0363] Optionally, the present application also provides a computer program product. All or part of the processes in the above method embodiments may be completed by a computer program instructing related hardware. The program may be stored in the above computer program product, and when executed, the program may include the processes in the above method embodiments.

[0364] Optionally, the present application also provides a computer instruction. All or part of the process in the above method embodiment can be completed by the computer instruction to instruct the relevant hardware (such as a computer, processor, network node or computing node, etc.). The program can be stored in the above computer-readable storage medium or in the above computer program product.

[0365] Optionally, the present application also provides a communication system, comprising at least two of the following nodes: a network node in the above embodiment, and at least one computing node or communication node.

[0366] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0367] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0368] The units described as separate components may or may not be physically separate, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple places. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0369] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0370] The above is only a specific embodiment of the present application, but the scope of protection of this application is not limited to this. Any changes or substitutions within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A communication method, characterized in that, The method includes: Sending a first scheduling result and a first quality of service (QoS) requirement to at least one computing node, where the first scheduling result indicates a first way for the at least one computing node to execute a target task; Receiving first feedback information from the at least one computing node, where the first feedback information indicates that the QoS of executing the target task by the at least one computing node according to the first way cannot meet the first QoS requirement; Determining a second scheduling result according to the first feedback information, where the second scheduling result indicates a second way for the at least one computing node to execute the target task, and the second way is different from the first way.

2. The method according to claim 1, characterized in that, The method further includes: Sending first condition information, where the first condition information is used to indicate the condition for triggering the first feedback information.

3. The method according to claim 2, wherein The condition for triggering the first feedback information includes at least one of the following: The QoS of executing the target task by the at least one computing node according to the first way does not meet the first QoS requirement; or, The QoS of executing the target task by the at least one computing node according to the first way does not meet the first QoS requirement, and the deviation between them is greater than or equal to a first threshold; Or, The feedback time indicated by the first condition information arrives.

4. The method according to any one of claims 1 to 3, characterized in that The method further includes: Determining the first QoS requirement according to the computing QoS requirement of the target task and the computing power of the at least one computing node.

5. The method according to claim 4, characterized in that The computing QoS requirement of the target task or the first QoS requirement indicates at least one of the following: computing type requirement, computing volume requirement, computing latency requirement, computing energy consumption requirement, or computing accuracy requirement.

6. The method according to any one of claims 1-5, characterized in that, The method further includes: Sending a second QoS requirement to a communication node; Receiving second feedback information from the communication node, where the second feedback information indicates that the QoS provided by the communication node cannot meet the second QoS requirement; The determining the second scheduling result according to the first feedback information includes: Determining the second scheduling result according to the first feedback information and the second feedback information.

7. The method according to claim 6, characterized in that, The method further includes: Sending second condition information, where the second condition information is used to indicate the condition for triggering the second feedback information.

8. The method according to claim 7, wherein The condition for triggering the second feedback information includes at least one of the following: The QoS provided by the communication node cannot meet the second QoS requirement; or, The QoS provided by the communication node cannot meet the second QoS requirement, and the deviation between them is greater than or equal to a second threshold; or, The feedback time indicated by the second condition information arrives.

9. The method according to any one of claims 6-8, characterized in that, The method further includes: Determining the second QoS requirement according to the network state and the communication QoS requirement of the target task.

10. The method according to claim 7, wherein The communication QoS requirement of the target task or the second QoS requirement indicates at least one of the following: packet delay budget, guaranteed bit rate, guaranteed flow bit rate, maximum burst data volume, packet error rate, or guaranteed transmission data size.

11. The method according to any one of claims 1 to 10, characterized in that, The method further includes: Obtain the network status, the quality of service requirements of the target task, and the computing power of the at least one computing node, where the quality of service requirements of the target task include the communication quality of service requirements and the computing quality of service requirements of the target task; Determine the first scheduling result according to the network status, the quality of service requirements of the target task, and the computing power of the at least one computing node.

12. The method according to claim 11, wherein The obtaining the quality of service requirements of the target task includes: Receive a computing service request, where the computing service request includes the quality of service requirement information of the target task.

13. The method according to claim 11, wherein The obtaining the quality of service requirements of the target task includes: Receive a computing service request, where the computing service request includes the service identifier corresponding to the target task; Obtain the quality of service requirements of the target task according to the service identifier.

14. The method according to any one of claims 1-13, characterized in that, The method further includes: Send the second scheduling result.

15. The method according to any one of claims 1 to 14, characterized in that, The method is applied to a radio access network node, a central unit, a distributed unit, an intelligent controller of a radio access network, or a core network element.

16. The method according to any one of claims 1 to 15, characterized in that, The first mode instructs the at least one computing node to execute the first subtask in the target task; the second mode instructs the at least one computing node to execute the second subtask in the target task.

17. A communication method, characterized in that, The method includes: Receive a first scheduling result and a first quality of service requirement, where the first scheduling result instructs the at least one computing node to execute the target task in a first mode; Execute the target task according to the first mode; Send first feedback information, where the first feedback information is used to indicate that the quality of service for executing the target task according to the first mode cannot meet the first quality of service requirement.

18. The method according to claim 17, wherein The method further includes: Receive first condition information, where the first condition information is used to indicate the condition for triggering the first feedback information.

19. The method according to claim 18, wherein The condition for triggering the first feedback information includes at least one of the following: The quality of service for executing the target task according to the first mode does not meet the first quality of service requirement; or, The quality of service for executing the target task according to the first mode does not meet the first quality of service requirement, and the deviation between the two is greater than or equal to a first threshold; or, The feedback moment indicated by the first condition information arrives.

20. The method according to any one of claims 17-19, characterized in that, The first quality of service requirement indicates at least one of the following: computing type requirement, computing volume requirement, computing delay requirement, computing energy consumption requirement, or computing accuracy requirement.

21. The method according to any one of claims 17 - 20, characterized in that, The method further includes: Send a computing service request, where the computing service request includes at least one of the quality of service requirement information of the target task or the service identifier corresponding to the target task.

22. The method according to any one of claims 17-21, characterized in that, The method further includes: Receive a second scheduling result, where the second scheduling result instructs the at least one computing node to execute the target task in a second mode, and the second mode is different from the first mode; Execute the target task according to the second scheduling result.

23. The method according to claim 22, wherein The first mode instructs the at least one computing node to execute the first subtask in the target task; the second mode instructs the at least one computing node to execute the second subtask in the target task.

24. A communication device, characterized in that, Comprising units or modules for performing the method according to any one of claims 1 to 16, or comprising units or modules for performing the method according to any one of claims 17 to 23.

25. A communication device, characterized in that, Comprising: A processor, the processor being coupled to a memory for storing programs or instructions, which when executed by the processor cause the device to perform the method according to any one of claims 1 to 16, or to perform the method according to any one of claims 17 to 23.

26. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed, the computer is caused to perform the method according to any one of claims 1 to 16, or the method according to any one of claims 17 to 23.

27. A computer program product, comprising computer program code, characterized in that, When the computer program code runs on a computer, the computer is caused to implement the method according to any one of claims 1 to 16, or to implement the method according to any one of claims 17 to 23.

Images