Communication method, device and equipment

Abstract

The embodiment of the invention provides a communication method, device and equipment, and relates to the technical field of AIs, and the method comprises the steps: obtaining the respective AI calculation capability of a plurality of second network function equipment; dividing the first AI task into a plurality of first AI sub-tasks based on the respective AI computing capabilities of the plurality of second network function devices, and selecting a plurality of third network function devices for executing the plurality of first AI sub-tasks from the plurality of second network function devices; and sending respective corresponding first AI subtasks to the plurality of third network function devices, and receiving first AI subtask execution results sent by the plurality of third network function devices, thereby achieving the purpose of executing the AI task in the mobile communication network.

Description

Technical Field

[0001] The embodiments of the present application relate to artificial intelligence (AI) technology, and in particular to a communication method, apparatus and device. Background Art

[0002] AI is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0003] Currently, research on mobile communication networks supporting AI tasks is gradually being carried out. How to perform AI tasks in mobile communication networks is a technical problem that needs to be solved urgently in this application. Summary of the invention

[0004] The embodiments of the present application provide a communication method, apparatus and device, which is a method for jointly executing AI tasks in a mobile communication network, and solves the technical problem of how to execute AI tasks in a mobile communication network.

[0005] In a first aspect, an embodiment of the present application provides a communication method, which is applied to a first network function device, and the method includes: obtaining the AI ​​computing capability of each of multiple second network function devices; dividing the first AI task into multiple first AI subtasks based on the AI ​​computing capability of each of the multiple second network function devices, and selecting multiple third network function devices from the multiple second network function devices for executing the multiple first AI subtasks; sending the first AI subtasks corresponding to each of the multiple third network function devices, and receiving the first AI subtask execution results sent by each of the multiple third network function devices.

[0006] In a second aspect, an embodiment of the present application provides a communication method, which is applied to a third network function device, and the method includes: sending the AI ​​computing capability of the third network function device to the first network function device; receiving a first AI subtask corresponding to the third network function device; executing the first AI subtask to obtain an execution result of the first AI subtask; and sending the execution result of the first AI subtask to the first network function device.

[0007] In a third aspect, an embodiment of the present application provides a communication device, comprising: a transceiver module and a processing module; the transceiver module is used to obtain the AI ​​computing capabilities of each of the multiple second network function devices; the processing module is used to divide the first AI task into multiple first AI sub-tasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple third network function devices from the multiple second network function devices to execute the multiple first AI sub-tasks; the transceiver module is also used to send the first AI sub-tasks corresponding to each of the multiple third network function devices, and receive the first AI sub-task execution results sent by each of the multiple third network function devices.

[0008] In a fourth aspect, an embodiment of the present application provides a communication device, comprising: a transceiver module and a processing module; the transceiver module is used to send the AI ​​computing capability of the communication device to a first network function device; the transceiver module is also used to receive a first AI subtask corresponding to the communication device; the processing module is used to execute the first AI subtask and obtain the execution result of the first AI subtask; the transceiver module is also used to send the execution result of the first AI subtask to the first network function device.

[0009] In a fifth aspect, an embodiment of the present application provides a network function device, comprising: a processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, and executing the method in the first aspect, the second aspect or their respective implementation methods.

[0010] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a computer program, wherein the computer program enables a computer to execute a method as in the first aspect, the second aspect, or each of their implementations.

[0011] In a seventh aspect, an embodiment of the present application provides a computer program product, comprising computer program instructions, which enable a computer to execute a method as in the first aspect, the second aspect, or each of their implementations.

[0012] In an eighth aspect, an embodiment of the present application provides a computer program, which enables a computer to execute a method as described in the first aspect, the second aspect, or any of their implementations.

[0013] The technical solution provided by the embodiment of the present application solves the technical problem of how to perform AI tasks in a mobile communication network. Furthermore, this joint execution method of AI tasks can improve the execution efficiency of AI tasks. And the communication method provided by the embodiment of the present application can reduce the amount of data transmission between communication devices compared to the centralized AI task processing method, thereby reducing the network transmission load. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1A schematic diagram of an application scenario provided for an embodiment of the present application;

[0015] Figure 2 A schematic diagram of another application scenario provided for an embodiment of the present application;

[0016] Figure 3 An interactive flow chart of a communication method provided in an embodiment of the present application;

[0017] Figure 4A A schematic diagram of the connection between network function devices provided in one embodiment of the present application;

[0018] Figure 4B A schematic diagram of connections between network function devices provided in another embodiment of the present application;

[0019] Figure 5 A schematic diagram of a communication device 500 provided in an embodiment of the present application;

[0020] Figure 6 A schematic diagram of a communication device 600 provided in an embodiment of the present application;

[0021] Figure 7 It is a schematic block diagram of a network function device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0024] The embodiments of the present application may involve AI technology.

[0025] Among them, AI is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0026] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. Basic artificial intelligence technologies generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, mechatronics and other technologies. Artificial intelligence software technologies mainly include computer vision technology, speech processing technology, natural language processing technology, as well as machine learning / deep learning, autonomous driving, smart transportation and other major directions.

[0027] Currently, research on mobile communication networks supporting AI tasks is gradually being carried out. How to perform AI tasks in mobile communication networks is a technical problem that needs to be solved urgently in this application.

[0028] In order to solve this technical problem, the embodiment of the present application proposes to send the AI ​​task to multiple communication devices, and jointly execute the AI ​​task through these communication devices.

[0029] For example, Figure 1 A schematic diagram of an application scenario provided in an embodiment of the present application, such as Figure 1 As shown, the application scenario may include: a first network function device 110 and multiple second network function devices 120, wherein the first network function device 110 and each second network function device 120 may be directly or indirectly connected via wired or wireless communication, and this application does not impose any limitation thereto.

[0030] In some implementations, the first network function device 110 and each second network function device 120 may be core network devices in a mobile communication network, but is not limited thereto.

[0031] For example, the first network function device 110 may be any of the following, but is not limited thereto: an access and mobility management function (AMF) entity, a session management function (SMF) entity, a user plane function (UPF) entity, a policy control function (PCF) entity, a unified data management (UDM) entity, a data network (DN), an application function (AF) entity, an authentication server function (AUSF) entity, and a network slice selection function (NSSF) entity. Similarly, each second network function device 120 may also be any of the above devices.

[0032] In some implementation methods, the mobile communication network involved in the embodiments of the present application may be as follows, but not limited to: Global System of Mobile communication (GSM) system, Code Division Multiple Access (CDMA) system, Wideband Code Division Multiple Access (WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD) system, advanced long term evolution (LTE-A) system, new radio (NR) system, NR system evolution system, LTE on unlicensed spectrum (LTE-based access to unlicensed spectrum, LTE-U) system, NR on unlicensed spectrum (NR-based access to unlicensed spectrum, NR-U) system, Universal Mobile Telecommunication System (UMTS), etc.

[0033] It should be understood that Figure 1 The application scenario shown includes a first network function device and two second network function devices. In fact, it may include other numbers of first network function devices and second network function devices, and this application does not impose any limitation on this.

[0034] For example, Figure 2 Another application scenario diagram provided in the embodiment of the present application is as follows: Figure 2 As shown, the application scenario may include: a first network function device 210, multiple second network function devices 220 and a fifth network function device 230, wherein the first network function device 210 and the fifth network function device 230 may be directly or indirectly connected via a wired communication mode, which is not limited in this application. The fifth network function device 230 and each second network function device 220 may be directly or indirectly connected via a wired communication mode, which is not limited in this application.

[0035] It should be understood that the explanation of the first network function device, the second network function device and the mobile communication network can be referred to above, and the embodiments of the present application will not be repeated here.

[0036] In some implementations, the fifth network function device 230 is used to implement AI data forwarding between the first network function device 210 and each second network function device 220. The fifth network function device 230 may be an independent physical device, or a device cluster or distributed system composed of multiple physical devices.

[0037] In some implementations, the AI ​​data may include at least one of the following, but is not limited thereto: AI computing capability, AI subtasks, AI subtask execution results, and instruction messages.

[0038] In some implementations, the fifth network function device 230 may also be a bus, but is not limited thereto.

[0039] In some possible implementations, the fifth network function device 230 is also referred to as an AI task communication agent, an AI task communication agent device, an AI task communication agent node, etc., which is not limited in the embodiments of the present application.

[0040] It should be understood that Figure 2 The application scenario shown includes a first network function device, two second network function devices and one fifth network function device. In fact, it may include other numbers of first network function devices, second network function devices and fifth network function devices, and this application does not impose any restrictions on this.

[0041] The technical solution of this application will be described in detail below:

[0042] Figure 3 An interactive flow chart of a communication method provided in an embodiment of the present application, the method can be executed by a first network function device, multiple second network function devices and multiple third network function devices, such as Figure 3 As shown, the method may include:

[0043] S310: Multiple second network function devices send their respective AI computing capabilities to the first network function device;

[0044] In some implementations, for each second network function device among multiple second network function devices, the AI ​​computing capability of the second network function device includes, but is not limited to: the AI ​​model supported by the second network function device and the amount of computing resources of the second network function device.

[0045] In some implementations, for each second network function device among the plurality of second network function devices, the second network function device may support one or more AI models.

[0046] In some implementations, the AI ​​model may be a neural network model, but is not limited thereto.

[0047] In some implementations, for each second network function device among the plurality of second network function devices, the computing resource amount of the second network function device includes at least one of the following, but is not limited to: the total computing resource amount, the currently available computing resource amount.

[0048] In some implementations, for each of the plurality of second network function devices, the second network function device may actively send its AI computing capability. For example, the second network function device may periodically report its AI computing capability.

[0049] In some implementations, for each second network function device among the plurality of second network function devices, the second network function device may send the AI ​​computing capability of the second network function device to the first network function device after receiving the request message sent by the first network function device, wherein the request message is used to request the AI ​​computing capability of the second network function device.

[0050] S320: The first network function device divides the first AI task into a plurality of first AI subtasks based on the AI ​​computing capabilities of the plurality of second network function devices, and selects a plurality of third network function devices from the plurality of second network function devices for executing the plurality of first AI subtasks;

[0051] It should be understood that the plurality of third network function devices are part or all of the plurality of second network function devices. The third network function device refers to a second network function device selected from the plurality of second network function devices.

[0052] In some implementations, the first network function device may divide the first AI task into multiple first AI subtasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple third network function devices from the multiple second network function devices to execute the multiple first AI subtasks.

[0053] For example, assuming that the AI ​​task is to predict the load of the network in a period of time in the future, the second network function device A can predict the user plane load of the network in the future, and the second network function device B can predict the control plane load of the network in the future. Based on this, the first network function device can divide the AI ​​task into two subtasks, including: predicting the user plane load of the network in the future and predicting the control plane load of the network in the future. At the same time, the first network function device can select the second network function device A and the second network function device B as the third network function device.

[0054] For example, assuming that the AI ​​task is to predict the voice traffic and mobile Internet service data volume of a certain object in a certain period of time in the future, the second network function device A can predict the voice traffic volume of the object in the future, and the second network function device B can predict the mobile Internet service data volume of the object in the future. Based on this, the first network function device can divide the AI ​​task into two subtasks, including: predicting the voice traffic volume of the object in the future and predicting the mobile Internet service data volume of the object in the future. At the same time, the first network function device can select the second network function device A and the second network function device B as the third network function device.

[0055] S330: The first network function device sends the first AI subtask corresponding to each of the plurality of third network function devices;

[0056] For example, assuming that the AI ​​task is to predict the network load in the future, the first network function device can send the following AI subtask to the third network function device A: predict the user plane load of the current network in the future, and send the following AI subtask to the third network function device B: predict the control plane load of the current network in the future.

[0057] For example, assuming that the AI ​​task is to predict the voice traffic and mobile Internet service data volume of a certain object in a certain period of time in the future, the first network function device can send the following AI subtask to the third network function device A: predict the voice traffic of the object in the future period of time, and send the following AI subtask to the third network function device B: predict the mobile Internet service data volume of the object in the future period of time.

[0058] S340: The plurality of third network function devices execute their respective corresponding first AI subtasks, and obtain respective first AI subtask execution results of the plurality of third network function devices;

[0059] In some possible implementations, for each third network function device among multiple third network function devices, the third network function device can input the first AI subtask corresponding to the third network function device into the AI ​​model corresponding to the first AI subtask to obtain the execution result of the first AI subtask.

[0060] For example, assuming that the AI ​​subtask corresponding to the third network function device A is: predicting the user plane load of the current network within a period of time in the future, then the third network function device A can input the AI ​​subtask into the AI ​​model used to predict the user plane load of the current network within a period of time in the future, and the result of the AI ​​subtask execution is: the user plane load of the current network within a period of time in the future (such as within 10 minutes) is 60%. Similarly, assuming that the AI ​​subtask corresponding to the third network function device B is: predicting the control plane load of the current network within a period of time in the future, then the third network function device B can input the AI ​​subtask into the AI ​​model used to predict the control plane load of the current network within a period of time in the future, and the result of the AI ​​subtask execution is: the control plane load of the network within a period of time in the future (such as within 10 minutes) is 30%.

[0061] For example, assuming that the AI ​​subtask corresponding to the third network function device A is to predict the voice service volume of the object in the future, the third network function device A can input the AI ​​subtask into the AI ​​model used to predict the voice service volume of the object in the future, and the result of the AI ​​subtask execution is: the call duration of the voice service of the object in the future (such as within 10 minutes) is 3 minutes. Similarly, assuming that the AI ​​subtask corresponding to the third network function device B is to predict the mobile Internet service data volume of the object in the future, the third network function device B can input the AI ​​subtask into the AI ​​model used to predict the mobile Internet service data volume of the object in the future, and the result of the AI ​​subtask execution is: the mobile Internet service data volume of the object in the future (such as within 10 minutes) is 1GB.

[0062] S350: Multiple third network function devices send their respective first AI subtask execution results to the first network function device.

[0063] In some implementations, for any third network function device among multiple third network function devices, after the third network function device starts executing the corresponding first AI subtask, it can send the first AI subtask execution result to the first network function device immediately after generating the execution result.

[0064] In some other possible implementations, for any third network function device among the plurality of third network function devices, after the third network function device starts to execute the corresponding first AI subtask, the first AI subtask execution result may be periodically sent to the first network function device. The sending period of the first AI subtask execution result may be configured by the first network function device to the third network function device, but is not limited thereto.

[0065] In some implementations, the first network function device ends after obtaining the respective first AI subtask execution results sent by multiple third network function devices.

[0066] In some possible implementations, after the first network function device determines that the first AI task has been executed, the first network function device may send a first indication message to multiple third network function devices; wherein the first indication message is used to indicate that the first AI task has been executed.

[0067] In some possible implementations, after the first network function device obtains the respective first AI subtask execution results sent by multiple third network function devices, it can further process these first AI subtask execution results.

[0068] For example, suppose that the first network function device obtains the AI ​​subtask execution result of the third network function device A: the user plane load of the network in the future period (such as 10 minutes) is 90%, and the obtained AI subtask execution result of the third network function device A is: the control plane load of the network in the future period (such as 10 minutes) is 30%. Further, the first network function device conducts a comprehensive analysis of the two AI subtask execution results and determines that the network user plane resources need to be adjusted to adapt to the challenges brought about by the increase in the amount of network user plane data in the future.

[0069] For example, suppose that the first network function device obtains the AI ​​subtask execution result of the third network function device A: the call duration of the voice service of the object in the future (such as 10 minutes) is 3 minutes, and the obtained AI subtask execution result of the third network function device B is: the data volume of the mobile Internet service of the object in the future (such as 10 minutes) is 1GB. Further, the first network function device conducts a comprehensive analysis of the two AI subtask execution results, and determines that the Internet service consumption of the object is large, and it is necessary to optimize the protocol data unit (PDU) session resource guarantee for the user's Internet service. .

[0070] In some implementations, if the first AI task needs to be updated, the first network function device can update the first AI task to obtain the second AI task. Further, the first network function device divides the second AI task into multiple second AI subtasks based on the AI ​​computing capabilities of the multiple second network function devices, and selects multiple fourth network function devices from the multiple second network function devices for executing the multiple second AI subtasks; the first network function device sends the corresponding second AI subtasks to the multiple fourth network function devices; the multiple fourth network function devices execute the corresponding second AI subtasks to obtain the execution results of the second AI subtasks of the multiple fourth network function devices; the multiple fourth network function devices send the execution results of the second AI subtasks to the first network function device.

[0071] For example, assuming that the AI ​​task is to predict the voice traffic and mobile Internet service data volume of a certain object in the future, and the update task corresponding to the AI ​​task is to predict the voice traffic, 4G mobile Internet service data volume and 5G mobile Internet service data volume of a certain object in the future, assuming that the second network function device A can predict the voice traffic of a certain object in the future, the second network function device B can predict the 4G mobile Internet service data volume of a certain object in the future, and the second network function device C can predict the 5G mobile Internet service data volume of a certain object in the future, based on this, the first network function device can divide the AI ​​task into three subtasks, including: predicting the voice traffic, 4G mobile Internet service data volume and 5G mobile Internet service data volume of a certain object in the future, and the first network function device can select the second network function device A, the second network function device B and the second network function device C as the fourth network function device. Further, the first network function device can send the corresponding AI subtasks to the second network function device A, the second network function device B and the second network function device C. Among them, the second network function device A can input the corresponding AI subtask into the AI ​​model used to predict the voice service volume of the object in the future, and the obtained AI subtask execution result is: the call duration of the voice service of the object in the future (such as within 10 minutes) is 3 minutes. The second network function device B can input the corresponding AI subtask into the AI ​​model used to predict the 4G mobile Internet service data volume of the object in the future, and the obtained AI subtask execution result is: the 4G mobile Internet service data volume of the object in the future (such as within 10 minutes) is 1GB. The second network function device C can input the corresponding AI subtask into the AI ​​model used to predict the 5G mobile Internet service data volume of the object in the future, and the obtained AI subtask execution result is: the 5G mobile Internet service data volume of the object in the future (such as within 10 minutes) is 2GB. Furthermore, the second network function device A, the second network function device B and the second network function device C can send their respective AI subtask execution results to the first network function device.

[0072] It should be understood that the processing method for the second AI task can refer to the processing method for the first AI task, and this embodiment of the present application will not be described in detail.

[0073] The communication method provided in the embodiment of the present application is a method for jointly executing AI tasks in a mobile communication network, which solves the technical problem of how to execute AI tasks in a mobile communication network. Furthermore, this method of jointly executing AI tasks can improve the execution efficiency of AI tasks. And if all AI subtasks are executed by one communication device, then the communication device needs to obtain a large amount of data from other communication devices, and then execute all AI subtasks based on these data, thereby realizing the execution of AI tasks. The communication method provided in the embodiment of the present application is relative to this centralized AI task processing method. It can reduce the amount of data transmission between communication devices, thereby reducing the network transmission load.

[0074] It should be understood that mobile communication networks usually involve basic services, for example, the SMF entity needs to be responsible for the session management function in the core network, the AMF entity needs to be responsible for the mobility management function in the core network, etc. In order to avoid the impact of AI services on these basic services, in an embodiment of the present application, two channels may be included between the first network function device and each second network function device, namely: an AI data transmission channel and a basic service data transmission channel.

[0075] In some implementations, the AI ​​data transmission channel and the basic business data transmission channel may be configured with different configuration information, and the configuration information may include at least one of the following: an Internet Protocol (IP) address and a port number.

[0076] For example, Figure 4A A schematic diagram of the connection between network function devices provided in an embodiment of the present application, such as Figure 4A As shown, there are AI data transmission channels and basic business data transmission channels between network function device 1 and network function device 2, there are AI data transmission channels and basic business data transmission channels between network function device 1 and network function device 3, and there are AI data transmission channels and basic business data transmission channels between network function device 2 and network function device 3.

[0077] In some implementations, any network function device involved in the embodiments of the present application may include: an AI function module and a basic business function module, wherein the AI ​​function is used to execute methods related to AI data, for example, the communication method provided in the embodiments of the present application may be executed. The basic business function module is also called a business logic function module, which can be used to execute methods related to the basic business data of the network function device other than the AI ​​function, such as session management, mobility management, etc.

[0078] Based on this, the AI ​​data transmission channel can connect the AI ​​function modules of two network function devices, and the basic business data transmission channel can connect the basic business function modules of two network function devices.

[0079] For example, Figure 4B A schematic diagram of connections between network function devices provided in another embodiment of the present application is shown in FIG. Figure 4B As shown, there is an AI data transmission channel between the AI ​​function module in network function device 1 and the AI ​​function module in network function device 2, and there is a basic business data transmission channel between the basic business function module in network function device 1 and the basic business function module in network function device 2. There is an AI data transmission channel between the AI ​​function module in network function device 1 and the AI ​​function module in network function device 3, and there is a basic business data transmission channel between the basic business function module in network function device 1 and the basic business function module in network function device 3. There is an AI data transmission channel between the AI ​​function module in network function device 2 and the AI ​​function module in network function device 4, and there is a basic business data transmission channel between the basic business function module in network function device 2 and the basic business function module in network function device 3.

[0080] It should be understood that the basic business data transmission channel is also referred to as a non-AI data transmission channel or a business logic channel, etc., and the embodiments of the present application do not limit this. Among them, the basic business data transmission channel is used to transmit the basic business data (ie, non-AI data) of the network function device.

[0081] In some implementations, non-AI data refers to data other than AI data. For example, messages related to session management functions and messages related to mobility management functions can be referred to as non-AI data.

[0082] It should be understood that the AI ​​data transmission channel is a transmission channel for transmitting AI data, wherein the AI ​​data may include at least one of the following, but is not limited thereto: AI computing capability, AI subtask, AI subtask execution result, and indication message. The AI ​​data may be an IP data packet or an Ethernet data frame.

[0083] Based on this, the embodiments of the present application provide the following implementation methods:

[0084] In some possible implementations, the first network function device obtains the AI ​​computing capabilities of each of the multiple second network function devices, including: the first network function device receives the AI ​​computing capabilities sent by the multiple second network function devices through their respective corresponding AI data transmission channels.

[0085] For example, the first network function device can receive the AI ​​computing capability of the second network function device A sent by the second network function device A through the AI ​​data transmission channel 1, and can receive the AI ​​computing capability of the second network function device B sent by the second network function device B through the AI ​​data transmission channel 2.

[0086] In some possible implementations, the first network function device sends each corresponding first AI subtask to multiple third network function devices, including: the first network function device sends each corresponding first AI subtask to multiple third network function devices through each corresponding AI data transmission channel of the multiple third network function devices.

[0087] For example, the first network function device can send the corresponding first AI subtask through the AI ​​data transmission channel 1 between it and the third network function device A, and can send the corresponding first AI subtask through the AI ​​data transmission channel 2 between it and the third network function device B.

[0088] In some possible implementations, the first network function device receives the first AI subtask execution results sent by multiple third network function devices respectively, including: the first network function device receives the first AI subtask execution results sent by multiple third network function devices through their respective corresponding AI data transmission channels.

[0089] For example, the first network function device can receive the first AI subtask execution result sent by the third network function device A through the AI ​​data transmission channel 1, and can receive the first AI subtask execution result sent by the third network function device B through the AI ​​data transmission channel 2.

[0090] In some possible implementations, the first network function device may update the first AI task to obtain a second AI task; divide the second AI task into multiple second AI subtasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple fourth network function devices from the multiple second network function devices to execute the multiple second AI subtasks; send the corresponding second AI subtasks to the multiple fourth network function devices through the AI ​​data transmission channels corresponding to the multiple fourth network function devices; and receive the second AI subtask execution results sent by the multiple fourth network function devices through the corresponding AI data transmission channels.

[0091] For example, the first network function device may send the corresponding second AI subtask through the AI ​​data transmission channel 1 between it and the fourth network function device A, and receive the second AI subtask execution result sent by the fourth network function device A through the AI ​​data transmission channel 1. And the first network function device may send the corresponding second AI subtask through the AI ​​data transmission channel 2 between it and the fourth network function device B, and receive the second AI subtask execution result sent by the fourth network function device B through the AI ​​data transmission channel 2.

[0092] In some possible implementations, if the first AI task has been executed and completed, the first network function device sends a first indication message to multiple third network function devices through the AI ​​data transmission channels corresponding to each of the multiple third network function devices; wherein the first indication message is used to indicate that the first AI task has been executed and completed.

[0093] For example, the first network function device can send a first indication message to the third network function device A through the AI ​​data transmission channel 1 between it and the third network function device A, and can send a first indication message to the third network function device B through the AI ​​data transmission channel 2 between it and the third network function device B.

[0094] In an embodiment of the present application, two types of channels may be included between the first network function device and each second network function device, namely: an AI data transmission channel and a basic service data transmission channel, thereby avoiding the impact of AI services on basic services and further improving the reliability of the mobile communication network.

[0095] In some implementations, for each second network function device among the plurality of second network function devices, the AI ​​data transmission channel between the first network function device and the second network function device includes: a default transmission channel and a dedicated transmission channel. The default transmission channel is used to transmit the AI ​​data before the first AI subtask corresponding to the second network function device is executed. The dedicated transmission channel is used to transmit the AI ​​data after the first AI subtask corresponding to the second network function device is executed.

[0096] In some implementations, for each second network function device among the multiple second network function devices, the default transmission channel between the first network function device and the second network function device may always exist to maintain a normal heartbeat connection.

[0097] In some implementations, for each second network function device among the multiple second network function devices, a dedicated transmission channel between the first network function device and the second network function device can be established on demand after receiving the first AI subtask.

[0098] Based on this, the embodiments of the present application provide the following implementation methods, but are not limited thereto:

[0099] In some possible implementations, the first network function device receives AI computing capabilities sent by multiple second network function devices through their respective corresponding AI data transmission channels, including: the first network function device receives AI computing capabilities sent by multiple second network function devices through their respective corresponding default transmission channels.

[0100] For example, the first network function device can receive the AI ​​computing capability of the second network function device A sent by the second network function device A through the default transmission channel 11, and can receive the AI ​​computing capability of the second network function device B sent by the second network function device B through the default transmission channel 21.

[0101] In some possible implementations, the first network function device sends the first AI subtasks corresponding to each of the multiple third network function devices through the AI ​​data transmission channels corresponding to each of the multiple third network function devices, including: the first network function device sends the first AI subtasks corresponding to each of the multiple third network function devices through the default transmission channels corresponding to each of the multiple third network function devices.

[0102] For example, the first network function device can send the corresponding first AI subtask through the default transmission channel 11 between it and the third network function device A, and can send the corresponding first AI subtask through the default transmission channel 21 between it and the third network function device B.

[0103] In some possible implementations, the first network function device receives the first AI subtask execution results sent by multiple third network function devices through their respective corresponding AI data transmission channels, including: the first network function device receives the first AI subtask execution results sent by multiple third network function devices through their respective corresponding dedicated transmission channels.

[0104] For example, the first network function device can receive the first AI subtask execution result sent by the third network function device A through the dedicated transmission channel 12, and can receive the first AI subtask execution result sent by the third network function device B through the dedicated transmission channel 22.

[0105] In some possible implementations, the first network function device may update the first AI task to obtain a second AI task; divide the second AI task into multiple second AI subtasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple fourth network function devices from the multiple second network function devices to execute the multiple second AI subtasks; send the corresponding second AI subtasks to the multiple fourth network function devices through the corresponding default transmission channels of the multiple fourth network function devices; and receive the second AI subtask execution results sent by the multiple fourth network function devices through the corresponding dedicated transmission channels.

[0106] For example, the first network function device may send the corresponding second AI subtask through the default transmission channel 12 between it and the fourth network function device A, and receive the second AI subtask execution result sent by the fourth network function device A through the dedicated transmission channel 12. And the first network function device may send the corresponding second AI subtask through the default transmission channel 22 between it and the fourth network function device B, and receive the second AI subtask execution result sent by the fourth network function device B through the dedicated transmission channel 22.

[0107] In some possible implementations, if the first AI task has been executed and completed, the first network function device sends a first indication message to multiple third network function devices through the dedicated transmission channels corresponding to each of the multiple third network function devices; wherein the first indication message is used to indicate that the first AI task has been executed and completed.

[0108] For example, the first network function device can send a first indication message to the third network function device A through the dedicated transmission channel 12 between it and the third network function device A, and can send a first indication message to the third network function device B through the dedicated transmission channel 22 between it and the third network function device B.

[0109] In some implementations, the dedicated transmission channel uniquely corresponds to the first AI task, or the dedicated transmission channel is a dedicated transmission channel shared by the first AI task and other AI tasks. Typically, the first network function device determines the characteristics of the dedicated transmission channel corresponding to each AI task, and AI tasks with the same dedicated transmission channel characteristics can share the same dedicated transmission channel. For example, the first AI task and the second AI task share a dedicated transmission channel.

[0110] In some implementations, for any second network function device among the plurality of second network function devices, the dedicated transmission channel between the first network function device and the second network function device may be deleted after the AI ​​task corresponding to the dedicated transmission channel is executed. For example, assuming that the AI ​​task corresponding to the dedicated transmission channel is the first AI task, after the first AI task is executed, the dedicated transmission channel is deleted. For example, assuming that the AI ​​tasks corresponding to the dedicated transmission channel are the first AI task and the second AI task, after the first AI task and the second AI task are executed, the dedicated transmission channel is deleted.

[0111] In an embodiment of the present application, the AI ​​data transmission channel between the first network function device and each second network function device may include: a default transmission channel and a dedicated transmission channel. After the first network function device sends the corresponding AI subtask to the second network function device, the dedicated transmission channel can be used to transmit relevant data of the AI ​​subtask, thereby improving the execution efficiency of the AI ​​subtask, and further improving the execution efficiency of the AI ​​task.

[0112] In some implementations, for each second network function device, the first network function device and the second network function device may be connected via a fifth network function device, for example, Figure 2 As shown, the fifth network function device is used to forward data between the first network function device and the second network function device. The fifth network function device has its own IP address and port number.

[0113] Based on this, the embodiments of the present application propose the following possible implementation methods, but are not limited thereto:

[0114] In some possible implementations, the first network function device obtains the respective AI computing capabilities of the plurality of second network function devices, including: the first network function device receives the respective AI computing capabilities of the plurality of second network function devices forwarded via the fifth network function device.

[0115] For example, the second network function device A can send the AI ​​computing capability of the second network function device A to the fifth network function device, and the fifth network function device then forwards the AI ​​computing capability of the second network function device A to the first network function device. Similarly, the second network function device B can send the AI ​​computing capability of the second network function device B to the fifth network function device, and the fifth network function device then forwards the AI ​​computing capability of the second network function device B to the first network function device.

[0116] In some possible implementations, the first network function device sends each corresponding first AI subtask to multiple third network function devices, including: the first network function device sends each corresponding first AI subtask to multiple third network function devices through the fifth network function device.

[0117] For example, the first network function device sends the first AI subtask corresponding to the second network function device A to the fifth network function device, and the fifth network function device then forwards the first AI subtask corresponding to the second network function device A to the second network function device A. Similarly, the first network function device sends the first AI subtask corresponding to the second network function device B to the fifth network function device, and the fifth network function device then forwards the first AI subtask corresponding to the second network function device B to the second network function device B.

[0118] In some possible implementations, the first network function device receives the first AI subtask execution results respectively sent by multiple third network function devices, including: the first network function device receives the first AI subtask execution results respectively forwarded by multiple third network function devices through the fifth network function device.

[0119] For example, the second network function device A can send the first AI subtask execution result corresponding to the second network function device A to the fifth network function device, and the fifth network function device then forwards the first AI subtask execution result corresponding to the second network function device A to the first network function device. Similarly, the second network function device B can send the first AI subtask execution result corresponding to the second network function device B to the fifth network function device, and the fifth network function device then forwards the first AI subtask execution result corresponding to the second network function device B to the first network function device.

[0120] In some possible implementations, the first network function device may update the first AI task to obtain a second AI task; divide the second AI task into multiple second AI subtasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple fourth network function devices from the multiple second network function devices to execute the multiple second AI subtasks; send the corresponding second AI subtasks to the multiple fourth network function devices through the fifth network function device; and receive the second AI subtask execution results sent by the multiple fourth network function devices through their corresponding dedicated transmission channels through the fifth network function device.

[0121] For example, the first network function device may send the second AI subtask corresponding to the fourth network function device A to the fifth network function device, and the fifth network function device may forward the second AI subtask corresponding to the fourth network function device A to the fourth network function device A. After the fourth network function device A completes the execution of the second AI subtask, it may send the corresponding second AI subtask execution result to the fifth network function device, and the fifth network function device may forward the second AI subtask execution result to the first network function device. Similarly, the first network function device may send the second AI subtask corresponding to the fourth network function device B to the fifth network function device, and the fifth network function device may forward the second AI subtask corresponding to the fourth network function device B to the fourth network function device B. After the fourth network function device B completes the execution of the second AI subtask, it may send the corresponding second AI subtask execution result to the fifth network function device, and the fifth network function device may forward the second AI subtask execution result to the first network function device.

[0122] In some possible implementations, if the first AI task has been executed and completed, the first network function device sends a first indication message to multiple third network function devices through the fifth network function device; wherein the first indication message is used to indicate that the first AI task has been executed and completed.

[0123] For example, the first network function device may send a first indication message to the fifth network function device, and the fifth network function device may then forward the first indication message to the third network function device A and the third network function device B.

[0124] It should be understood that when the first network function device and the second network function device forward data through the fifth network function device, the above-mentioned AI data transmission channel and the basic business data transmission channel are also applicable to this scenario. Among them, in this scenario, the AI ​​data transmission channel may include two parts: the AI ​​data transmission channel between the first network function device and the fifth network function device, and the AI ​​data transmission channel between the fifth network function device and the second network function device. The default transmission channel may also include two parts: the default transmission channel between the first network function device and the fifth network function device, and the default transmission channel between the fifth network function device and the second network function device. The dedicated transmission channel may also include two parts: a dedicated transmission channel between the first network function device and the fifth network function device, and a dedicated transmission channel between the fifth network function device and the second network function device.

[0125] In an embodiment of the present application, data can be forwarded between the first network function device and each second network function device through a fifth network function device. The fifth network function device can be understood as a bus, that is, the first network function device and multiple second network function devices can all be connected to the bus, and data transmission is performed based on the bus. Compared with the scheme of connecting the first network function device to each second network function device, the method provided in the embodiment of the present application can reduce the number of connections, thereby saving network transmission resources.

[0126] In the embodiment of the present application, the first network function device may select the third network function device in any of the following achievable ways, but is not limited thereto:

[0127] In some possible implementations, the first network function device selects a plurality of third network function devices corresponding to the plurality of first AI subtasks from the plurality of second network function devices, including:

[0128] S320-1: The first network function device selects a plurality of sixth network function devices corresponding to the plurality of first AI subtasks from the plurality of second network function devices based on the respective AI computing capabilities of the plurality of second network function devices;

[0129] It should be understood that the sixth network function device refers to the network function device preliminarily selected by the first network function device from among the plurality of second network function devices.

[0130] In some implementations, for any second network function device among multiple second network function devices, the AI ​​computing capability of the second network function device includes: the AI ​​model supported by the second network function device and the computing resource amount of the second network function device, but is not limited thereto.

[0131] Correspondingly, S320-1A may include: for any second network function device among multiple second network function devices, if the AI ​​model supported by the second network function device matches at least one first AI subtask among multiple first AI subtasks, and the computing resource usage of the second network function device meets the computing resource usage of at least one first AI subtask, the first network function device uses the second network function device as the sixth network function device.

[0132] In some implementations, the computing resource usage includes at least one of the following, but is not limited thereto: total computing resources, and currently available computing resources.

[0133] Accordingly, the first network function device may detect whether the computing resource amount of the second network function device meets the computing resource usage amount of at least one first AI subtask in the following achievable manner, but is not limited thereto:

[0134] In some possible implementations, if the amount of computing resources includes: the total amount of computing resources; the first network function device can calculate the product of the total amount of computing resources and a preset coefficient to obtain an estimated amount of computing resources; if the computing resource usage of the above-mentioned at least one first AI subtask is less than or equal to the estimated amount of computing resources, it is determined that the amount of computing resources of the second network function device meets the computing resource usage of the at least one first AI subtask; if the computing resource usage of the above-mentioned at least one first AI subtask is greater than the estimated amount of computing resources, it is determined that the amount of computing resources of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

[0135] In some implementations, the preset coefficient may be any real number greater than 0 and less than or equal to 1. For example, the value of the preset coefficient may be 0.2, 0.3, 0.5, etc., which is not limited in the embodiment of the present application.

[0136] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the total computing resource amount of the second network function device A is 100, and the preset coefficient is 0.2, then 100*0.2=20 is the estimated computing resource amount. Since the computing resource usage 10 corresponding to the first AI subtask 1 and the first AI subtask 2 is less than the estimated computing resource amount, it can be determined that the computing resource amount of the second network function device A meets the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0137] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 30, and the total computing resource amount of the second network function device A is 100, and the preset coefficient is 0.2, then 100*0.2=20 is the estimated computing resource amount. Since the computing resource usage 30 corresponding to the first AI subtask 1 and the first AI subtask 2 is greater than the estimated computing resource amount, it can be determined that the computing resource amount of the second network function device A does not meet the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0138] In some implementations, if the amount of computing resources includes: the currently available amount of computing resources, then if the computing resource usage of at least one first AI subtask is less than or equal to the currently available amount of computing resources, it is determined that the amount of computing resources of the second network function device meets the computing resource usage of the at least one first AI subtask; if the computing resource usage of at least one first AI subtask is greater than the currently available amount of computing resources, it is determined that the amount of computing resources of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

[0139] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the currently available computing resources of the second network function device A is 20. Since the computing resource usage 10 corresponding to the first AI subtask 1 and the first AI subtask 2 is less than the currently available computing resources, it can be determined that the computing resource usage of the second network function device A meets the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0140] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 30, and the currently available computing resources of the second network function device A is 20. Since the computing resource usage 30 corresponding to the first AI subtask 1 and the first AI subtask 2 is greater than the currently available computing resources, it can be determined that the computing resource usage of the second network function device A does not meet the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0141] In some possible implementations, the amount of computing resources includes: the total amount of computing resources and the currently available amount of computing resources. If the total amount of computing resources is greater than or equal to a preset value, and the computing resource usage of at least one first AI subtask is less than or equal to the currently available amount of computing resources, the first network function device determines that the amount of computing resources of the second network function device meets the computing resource usage of the at least one first AI subtask; if the total amount of computing resources is less than the preset value, or the computing resource usage of at least one first AI subtask is greater than the currently available amount of computing resources, the first network function device determines that the amount of computing resources of the second network function device does not meet the computing resource usage of at least one first AI subtask.

[0142] In some possible implementations, the preset value may be 10, 20, 100, etc., which is not limited in the embodiments of the present application.

[0143] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the total computing resource amount of the second network function device A is 100, assuming that the preset value is 50, and the currently available computing resources are 20. Since the total computing resources 100 are greater than the preset value 50, the computing resource usage 10 corresponding to the first AI subtask 1 and the first AI subtask 2 is less than the currently available computing resources 20, then it can be determined that the computing resource amount of the second network function device A meets the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0144] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the total computing resource amount of the second network function device A is 100, assuming that the preset value is 120, and the current available computing resources are 20. Since the total computing resources 100 are less than the preset value 120, it can be determined that the computing resources of the second network function device A do not meet the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0145] In some possible implementations, if the amount of computing resources includes: the total amount of computing resources and the currently available amount of computing resources, the first network function device calculates the product of the total amount of computing resources and a preset coefficient to obtain an estimated amount of computing resources; based on the estimated amount of computing resources and the currently available amount of computing resources, it is detected whether the amount of computing resources of the second network function device meets the computing resource usage of at least one first AI subtask.

[0146] It should be understood that the explanation of the preset coefficient can be found above, and the embodiments of the present application will not be repeated here.

[0147] In some possible implementations, the first network function device determines the smaller of the estimated amount of computing resources and the currently available amount of computing resources; if the computing resource usage of the at least one first AI subtask is less than or equal to the smaller one, it is determined that the computing resource usage of the second network function device meets the computing resource usage of the at least one first AI subtask; if the computing resource usage of the at least one first AI subtask is greater than the smaller one, it is determined that the computing resource usage of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

[0148] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the total computing resource amount of the second network function device A is 100, and the preset coefficient is 0.3, then 100*0.3=30 is the estimated computing resource amount, assuming that the current available computing resources are 20, the smaller of the estimated computing resources and the current available computing resources is the current available computing resources. Since the computing resource usage 10 corresponding to the first AI subtask 1 and the first AI subtask 2 is less than the currently available computing resources, it can be determined that the computing resource amount of the second network function device A meets the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0149] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the total computing resource amount of the second network function device A is 100, and the preset coefficient is 0.05, then 100*0.05=5 is the estimated computing resource amount. Assuming that the current available computing resources are 20, the smaller of the estimated computing resources and the current available computing resources is the estimated computing resource amount. Since the computing resource usage 10 corresponding to the first AI subtask 1 and the first AI subtask 2 is greater than the estimated computing resource amount, it can be determined that the computing resource amount of the second network function device A does not meet the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0150] In some possible implementations, the first network function device determines the larger of the estimated amount of computing resources and the currently available amount of computing resources; if the computing resource usage of the at least one first AI subtask is less than or equal to the larger one, it is determined that the computing resource usage of the second network function device meets the computing resource usage of the at least one first AI subtask; if the computing resource usage of the at least one first AI subtask is greater than the larger one, it is determined that the computing resource usage of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

[0151] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 10, and the total computing resource amount of the second network function device A is 100, and the preset coefficient is 0.3, then 100*0.3=30 is the estimated computing resource amount, assuming that the current available computing resources are 20, the larger of the estimated computing resources and the current available computing resources is the estimated computing resource amount. Since the computing resource usage 10 corresponding to the first AI subtask 1 and the first AI subtask 2 is less than the estimated computing resource amount, it can be determined that the computing resource amount of the second network function device A meets the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0152] For example, assuming that the second network function device A has an AI model that supports the first AI subtask 1 and an AI model that supports the first AI subtask 2, and the computing resource usage corresponding to the first AI subtask 1 and the first AI subtask 2 is 20, and the total computing resource amount of the second network function device A is 100, and the preset coefficient is 0.05, then 100*0.05=5 is the estimated computing resource amount, assuming that the current available computing resources are 10, the larger of the estimated computing resources and the current available computing resources is the current available computing resources. Since the computing resource usage 20 corresponding to the first AI subtask 1 and the first AI subtask 2 is greater than the current available computing resources, it can be determined that the computing resource amount of the second network function device A does not meet the computing resource usage of the first AI subtask 1 and the first AI subtask 2.

[0153] S320-2: The first network function device sends each corresponding first AI subtask to multiple sixth network function devices;

[0154] S320-3: The plurality of sixth network function devices respectively determine whether they agree to the first AI subtask assigned to themselves;

[0155] S320-4: The first network function device receives a second indication message respectively sent by a plurality of sixth network function devices;

[0156] The following is an explanation in combination with S320-2 to S320-4:

[0157] It should be understood that, for any sixth network function device among the multiple sixth network function devices, the second indication message sent by the sixth network function device is used to indicate whether the sixth network function device agrees to the first AI subtask assigned to itself.

[0158] Among them, for any sixth network function device among the multiple sixth network function devices, the sixth network function device may determine whether it agrees to assign the first AI subtask to the sixth network function device in any of the following achievable ways, but is not limited thereto:

[0159] In some possible implementations, the sixth network function device can determine whether the computing resource usage of the first AI subtask allocated to it meets the computing resource usage; if the computing resource usage of the sixth network function device meets the computing resource usage of the first AI subtask allocated to it, the sixth network function device agrees to the first AI subtask allocated to it; if the computing resource usage of the sixth network function device does not meet the computing resource usage of the first AI subtask allocated to it, the sixth network function device does not agree to the first AI subtask allocated to it.

[0160] In some implementations, the sixth network function device and the first network function device may use the same method to determine whether the computing resource usage of at least one first AI subtask meets the computing resource usage.

[0161] For example, assume that the first network function device adopts the following method: the first network function device calculates the product of the total computing resource amount and a preset coefficient to obtain an estimated computing resource amount; if the computing resource usage of at least one first AI subtask allocated to communication device A is less than or equal to the estimated computing resource amount, it is determined that the computing resource amount of communication device A meets the computing resource usage of at least one first AI subtask; if the computing resource usage of the above-mentioned at least one first AI subtask is greater than the estimated computing resource amount, it is determined that the computing resource amount of communication device A does not meet the computing resource usage of at least one first AI subtask. Similarly, after receiving the above-mentioned at least one first AI subtask, the communication device A may also calculate the product of the total computing resource amount of the communication device A and the preset coefficient to obtain the estimated computing resource amount; if the computing resource usage of at least one first AI subtask allocated to the communication device A is less than or equal to the estimated computing resource amount, it is determined that the computing resource amount of the communication device A meets the computing resource usage of at least one first AI subtask; if the computing resource usage of the above-mentioned at least one first AI subtask is greater than the estimated computing resource amount, it is determined that the computing resource amount of the communication device A does not meet the computing resource usage of at least one first AI subtask.

[0162] It should be understood that, in this implementable manner, the reason why the sixth network function device is asked to reconfirm whether it agrees to the first AI subtask assigned to itself is because the amount of computing resources of the sixth network function device may change.

[0163] In some possible implementations, the sixth network function device and the first network function device may use different methods to determine whether the computing resource usage of at least one first AI subtask meets the computing resource usage.

[0164] For example, it is assumed that the first network function device adopts the following method: the first network function device calculates the product of the total computing resource amount and the preset coefficient to obtain the estimated computing resource amount; if the computing resource usage of at least one first AI subtask allocated to the communication device A is less than or equal to the estimated computing resource amount, it is determined that the computing resource amount of the communication device A meets the computing resource usage of at least one first AI subtask; if the computing resource usage of the at least one first AI subtask is greater than the estimated computing resource amount, it is determined that the computing resource amount of the communication device A does not meet the computing resource usage of the at least one first AI subtask. After receiving the at least one first AI subtask, the communication device A may also adopt the following method: if the computing resource usage of at least one first AI subtask allocated to the communication device A is less than or equal to the currently available computing resource amount, it is determined that the computing resource amount of the communication device A meets the computing resource usage of at least one first AI subtask; if the computing resource usage of the at least one first AI subtask is greater than the currently available computing resource amount, it is determined that the computing resource amount of the communication device A does not meet the computing resource usage of at least one first AI subtask.

[0165] S320-5: The first network function device determines multiple third network function devices based on the second indication messages respectively sent by the multiple sixth network function devices.

[0166] In some implementations, if the second indication messages sent by multiple sixth network function devices respectively indicate that these sixth network function devices agree to be assigned to the first AI subtask, the first network function device can use these sixth network function devices as third network function devices.

[0167] In some possible implementations, if there is at least one sixth network function device among multiple sixth network function devices, and the second indication message sent by it indicates that it does not agree to the first AI subtask assigned to itself, then the first network function device can reselect a sixth network function device from multiple second network function devices for the first AI subtask, and use the reselected sixth network function device as the third network function device, or send the first AI subtask to the reselected sixth network function device so that the sixth network function device determines whether it agrees to the first AI subtask assigned to itself. If it agrees, the first network function device uses the sixth network function device as the third network function device. If it disagrees, the first network function device continues to reselect the sixth network function device for the first AI subtask until the reselection is successful.

[0168] It should be understood that regarding the method for the reselected sixth network function device to determine whether it agrees to the first AI subtask assigned to itself, reference can be made to the above, and the embodiments of the present application will not be repeated here.

[0169] In some other possible implementations, the first network function device selects multiple third network function devices corresponding to multiple first AI subtasks from multiple second network function devices, including: the first network function device directly selects multiple third network function devices corresponding to multiple first AI subtasks from multiple second network function devices based on the respective AI computing capabilities of the multiple second network function devices.

[0170] In other words, in this implementable manner, the first network function device can select the third network function device by itself.

[0171] It should be understood that, regarding how the first network function device selects the third network function device, reference may be made to the above selection method for the sixth network function device, and this embodiment of the present application will not be repeated herein.

[0172] In the embodiment of the present application, the first network function device can directly select multiple third network function devices corresponding to multiple first AI subtasks from multiple second network function devices based on the AI ​​computing capabilities of each of the multiple second network function devices, or can select multiple sixth network function devices corresponding to multiple first AI subtasks from multiple second network function devices based on the AI ​​computing capabilities of each of the multiple second network function devices; the third network function device is determined in combination with whether the sixth network function device agrees to the first AI subtask assigned to it. In this way, the first network function device will assign the first AI subtask to the third network function device only when the third network function device is capable of processing the corresponding first AI subtask, thereby improving the reliability and effectiveness of AI task execution.

[0173] Figure 5 A schematic diagram of a communication device 500 provided in an embodiment of the present application is shown as follows: Figure 5 As shown, the device 500 includes: a transceiver module 510 and a processing module 520; wherein the transceiver module 510 is used to obtain the AI ​​computing capabilities of each of the multiple second network function devices; the processing module 520 is used to divide the first AI task into multiple first AI subtasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple third network function devices from the multiple second network function devices to execute the multiple first AI subtasks; the transceiver module 510 is also used to send the first AI subtasks corresponding to each of the multiple third network function devices, and receive the first AI subtask execution results sent by each of the multiple third network function devices.

[0174] In some possible implementations, the communication device 500 and multiple second network function devices respectively include: AI data transmission channels; accordingly, the transceiver module 510 is specifically used to: receive AI computing capabilities sent by multiple second network function devices through their respective corresponding AI data transmission channels; send their respective first AI subtasks to multiple third network function devices through their respective corresponding AI data transmission channels; receive the first AI subtask execution results sent by multiple third network function devices through their respective corresponding AI data transmission channels.

[0175] In some possible implementations, the AI ​​data transmission channel includes: a default transmission channel and a dedicated transmission channel; wherein, for any second network function device among a plurality of second network function devices, the default transmission channel between the communication device 500 and the second network function device is used to transmit the AI ​​data before the first AI subtask corresponding to the second network function device is executed, and the dedicated transmission channel between the communication device 500 and the second network function device is used to transmit the AI ​​data after the first AI subtask corresponding to the second network function device is executed; accordingly, the transceiver module 510 is specifically used to: receive the AI ​​computing power sent by a plurality of second network function devices through their respective corresponding default transmission channels; send the respective first AI subtasks to a plurality of third network function devices through their respective corresponding default transmission channels; receive the execution results of the first AI subtasks sent by a plurality of third network function devices through their respective corresponding dedicated transmission channels.

[0176] In some possible implementations, the processing module 520 is also used to: update the first AI task to obtain a second AI task; divide the second AI task into multiple second AI subtasks based on the AI ​​computing capabilities of each of the multiple second network function devices, and select multiple fourth network function devices from the multiple second network function devices to execute the multiple second AI subtasks; the transceiver module 510 is also used to: send the corresponding second AI subtasks to the multiple fourth network function devices through the default transmission channels corresponding to the multiple fourth network function devices; receive the second AI subtask execution results sent by the multiple fourth network function devices through the corresponding dedicated transmission channels.

[0177] In some possible implementations, the transceiver module 510 is also used to: if the first AI task has been executed and completed, send a first indication message to multiple third network function devices through the dedicated transmission channels corresponding to each of the multiple third network function devices; wherein the first indication message is used to indicate that the first AI task has been executed and completed.

[0178] In some implementations, the dedicated transmission channel uniquely corresponds to the first AI task, or the dedicated transmission channel is a dedicated transmission channel shared by the first AI task and other AI tasks.

[0179] In some possible implementations, the communication device 500 is connected to multiple second network function devices through a fifth network function device; accordingly, the transceiver module 510 is specifically used to: receive the respective AI computing capabilities forwarded by the multiple second network function devices through the fifth network function device; send the respective corresponding first AI subtasks to multiple third network function devices through the fifth network function device; receive the respective first AI subtask execution results forwarded by the multiple third network function devices through the fifth network function device.

[0180] In some possible implementations, the processing module 520 is specifically used to: select multiple sixth network function devices corresponding to multiple first AI subtasks from multiple second network function devices based on the AI ​​computing capabilities of each of the multiple second network function devices; send the first AI subtasks corresponding to each of the multiple sixth network function devices; receive the second indication message sent by each of the multiple sixth network function devices; wherein, for any sixth network function device among the multiple sixth network function devices, the second indication message sent by the sixth network function device is used to indicate whether the sixth network function device agrees to the first AI subtask assigned to the sixth network function device; determine multiple third network function devices based on the second indication message sent by each of the multiple sixth network function devices.

[0181] In some possible implementations, for any second network function device among multiple second network function devices, the AI ​​computing capability of the second network function device includes: the AI ​​model supported by the second network function device and the computing resource usage of the second network function device; accordingly, the processing module 520 is specifically used to: for any second network function device among the multiple second network function devices, if the AI ​​model supported by the second network function device matches at least one first AI subtask among multiple first AI subtasks, and the computing resource usage of the second network function device meets the computing resource usage of at least one first AI subtask, then the second network function device is used as the sixth network function device.

[0182] In some possible implementations, the amount of computing resources includes: the total amount of computing resources; the processing module 520 is also used to: calculate the product of the total amount of computing resources and a preset coefficient to obtain an estimated amount of computing resources; if the computing resource usage of at least one first AI subtask is less than or equal to the estimated amount of computing resources, then determine that the amount of computing resources of the second network function device meets the computing resource usage of at least one first AI subtask; if the computing resource usage of at least one first AI subtask is greater than the estimated amount of computing resources, then determine that the amount of computing resources of the second network function device does not meet the computing resource usage of at least one first AI subtask.

[0183] In some possible implementations, the amount of computing resources includes: the currently available amount of computing resources; the processing module 520 is also used to: if the computing resource usage of at least one first AI subtask is less than or equal to the currently available amount of computing resources, then determine that the computing resource usage of the second network function device meets the computing resource usage of at least one first AI subtask; if the computing resource usage of at least one first AI subtask is greater than the currently available amount of computing resources, then determine that the computing resource usage of the second network function device does not meet the computing resource usage of at least one first AI subtask.

[0184] In some implementations, the amount of computing resources includes: the total amount of computing resources and the currently available amount of computing resources; the processing module 520 is also used to: if the total amount of computing resources is greater than or equal to a preset value, and the computing resource usage of at least one first AI subtask is less than or equal to the currently available amount of computing resources, then determine that the amount of computing resources of the second network function device meets the computing resource usage of at least one first AI subtask; if the total amount of computing resources is less than the preset value, or the computing resource usage of at least one first AI subtask is greater than the currently available amount of computing resources, then determine that the amount of computing resources of the second network function device does not meet the computing resource usage of at least one first AI subtask.

[0185] In some possible implementations, the amount of computing resources includes: the total amount of computing resources and the currently available amount of computing resources; the processing module 520 is also used to: calculate the product of the total amount of computing resources and a preset coefficient to obtain an estimated amount of computing resources; based on the estimated amount of computing resources and the currently available amount of computing resources, detect whether the amount of computing resources of the second network function device meets the computing resource usage of at least one first AI subtask.

[0186] In some possible implementations, the processing module 520 is specifically used to: determine the smaller of the estimated amount of computing resources and the currently available amount of computing resources; if the computing resource usage of at least one first AI subtask is less than or equal to the smaller one, determine that the computing resource usage of the second network function device meets the computing resource usage of at least one first AI subtask; if the computing resource usage of at least one first AI subtask is greater than the smaller one, determine that the computing resource usage of the second network function device does not meet the computing resource usage of at least one first AI subtask.

[0187] It should be understood that the device embodiment and the method embodiment may correspond to each other, and similar descriptions may refer to the method embodiment. To avoid repetition, they will not be described here. Specifically, Figure 5 The apparatus 500 shown may perform Figure 3 The method embodiment corresponding to the first network function device in the apparatus 500, and the aforementioned and other operations and / or functions of each module in the apparatus 500 are respectively to implement Figure 3 For the sake of brevity, the corresponding processes in each method corresponding to the first network function device are not repeated here.

[0188] The above describes the device 500 of the embodiment of the present application from the perspective of the functional module in conjunction with the accompanying drawings. It should be understood that the functional module can be implemented in hardware form, can be implemented by instructions in software form, and can also be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software form instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or a combination of hardware and software modules in the decoding processor to perform. Optionally, the software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory, and completes the steps in the above method embodiment in conjunction with its hardware.

[0189] Figure 6 A schematic diagram of a communication device 600 provided in an embodiment of the present application is shown as follows: Figure 6 As shown, the device 600 includes: a transceiver module 610 and a processing module 620; wherein the transceiver module 610 is used to send the AI ​​computing capability of the communication device 600 to the first network function device; the transceiver module 610 is also used to receive the first AI subtask corresponding to the communication device 600; the processing module 620 is used to execute the first AI subtask and obtain the execution result of the first AI subtask; the transceiver module 610 is also used to send the execution result of the first AI subtask to the first network function device.

[0190] In some possible implementations, the first network function device and the communication apparatus 600 respectively include: an AI data transmission channel; accordingly, the transceiver module 610 is specifically used to: send the AI ​​computing capability of the third network function device to the first network function device through the AI ​​data transmission channel; receive the first AI subtask corresponding to the third network function device through the AI ​​data transmission channel; and send the execution result of the first AI subtask to the first network function device through the AI ​​data transmission channel.

[0191] In some possible implementations, the AI ​​data transmission channel includes: a default transmission channel and a dedicated transmission channel; the default transmission channel is used to transmit AI data before the first AI subtask corresponding to the communication device 600 is executed, and the dedicated transmission channel is used to transmit AI data after the first AI subtask corresponding to the communication device 600 is executed; accordingly, the transceiver module 610 is specifically used to: send the AI ​​computing capability of the communication device 600 to the first network function device through the default transmission channel; receive the first AI subtask corresponding to the communication device 600 through the default transmission channel; and send the execution result of the first AI subtask to the first network function device through the dedicated transmission channel.

[0192] In some possible implementations, if the first AI task has been completed, the transceiver module 610 is further used to: receive a first indication message sent by the first network function device; wherein the first indication message is used to indicate that the first AI task has been completed.

[0193] In some implementations, the dedicated transmission channel uniquely corresponds to the first AI task, or the dedicated transmission channel is a dedicated transmission channel shared by the first AI task and other AI tasks.

[0194] In some possible implementations, the first network function device is connected to the communication device 600 through the fifth network function device; accordingly, the transceiver module 610 is specifically used to: forward the AI ​​computing power of the communication device 600 to the first network function device through the fifth network function device; receive the first AI subtask sent by the first network function device through the fifth network function device; and forward the execution result of the first AI subtask through the fifth network function device.

[0195] It should be understood that the device embodiment and the method embodiment may correspond to each other, and similar descriptions may refer to the method embodiment. To avoid repetition, they will not be described here. Specifically, Figure 6 The device 600 shown may perform Figure 3 The method embodiment corresponding to the third network function device in the apparatus 600, and the aforementioned and other operations and / or functions of each module in the apparatus 600 are respectively to implement Figure 3For the sake of brevity, the corresponding processes in each method corresponding to the third network function device are not repeated here.

[0196] The above describes the device 600 of the embodiment of the present application from the perspective of the functional module in conjunction with the accompanying drawings. It should be understood that the functional module can be implemented in hardware form, can be implemented by instructions in software form, and can also be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software form instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or a combination of hardware and software modules in the decoding processor to perform. Optionally, the software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory, and completes the steps in the above method embodiment in conjunction with its hardware.

[0197] Figure 7 It is a schematic block diagram of a network function device provided in an embodiment of the present application.

[0198] like Figure 7 As shown, the network function device may include:

[0199] The memory 710 and the processor 720, the memory 710 is used to store the computer program and transmit the program code to the processor 720. In other words, the processor 720 can call and run the computer program from the memory 710 to implement the method in the embodiment of the present application.

[0200] For example, the processor 720 may be configured to execute the above method embodiments according to instructions in the computer program.

[0201] In some embodiments of the present application, the processor 720 may include but is not limited to:

[0202] General-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.

[0203] In some embodiments of the present application, the memory 710 includes but is not limited to:

[0204] Volatile memory and / or non-volatile memory. Among them, the non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0205] In some embodiments of the present application, the computer program may be divided into one or more modules, which are stored in the memory 710 and executed by the processor 720 to complete the method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the network function device.

[0206] like Figure 7 As shown, the network function device may also include:

[0207] The transceiver 730 may be connected to the processor 720 or the memory 710 .

[0208] The processor 720 may control the transceiver 730 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices. The transceiver 730 may include a transmitter and a receiver. The transceiver 730 may further include an antenna, and the number of antennas may be one or more.

[0209] It should be understood that the various components in the network functional device are connected via a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus and a status signal bus.

[0210] The present application also provides a computer storage medium on which a computer program is stored, and when the computer program is executed by a computer, the computer can perform the method of the above method embodiment. In other words, the present application embodiment also provides a computer program product containing instructions, and when the instructions are executed by a computer, the computer can perform the method of the above method embodiment.

[0211] When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integration. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a solid state drive (solid state disk, SSD)), etc.

[0212] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0213] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the module is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, 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 modules, which can be electrical, mechanical or other forms.

[0214] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. For example, each functional module in each embodiment of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0215] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A communication method, characterized in that: The method is applied to a first network function device, and the method includes: Obtaining artificial intelligence AI computing capabilities of each of the plurality of second network function devices; Divide the first AI task into a plurality of first AI subtasks based on the AI ​​computing capabilities of the plurality of second network function devices, and select a plurality of third network function devices from the plurality of second network function devices for executing the plurality of first AI subtasks; Send the first AI subtasks corresponding to each of the plurality of third network function devices, and receive the first AI subtask execution results sent by each of the plurality of third network function devices.

2. The method according to claim 1, characterized in that: The first network function device and the plurality of second network function devices respectively include: an AI data transmission channel; The obtaining of the artificial intelligence AI computing capabilities of each of the plurality of second network function devices includes: Receiving AI computing capabilities sent by the plurality of second network function devices through their respective corresponding AI data transmission channels; The sending the first AI subtasks corresponding to each of the plurality of third network function devices includes: Sending the first AI subtasks corresponding to each of the plurality of third network function devices through the AI ​​data transmission channels corresponding to each of the plurality of third network function devices; The receiving the first AI subtask execution results respectively sent by the plurality of third network function devices includes: Receive the first AI subtask execution results sent by the multiple third network function devices through their respective corresponding AI data transmission channels.

3. The method according to claim 2, characterized in that The AI ​​data transmission channel includes: a default transmission channel and a dedicated transmission channel; wherein, for any second network function device among the plurality of second network function devices, the default transmission channel between the first network function device and the second network function device is used to transmit AI data before the first AI subtask corresponding to the second network function device is executed, and the dedicated transmission channel between the first network function device and the second network function device is used to transmit AI data after the first AI subtask corresponding to the second network function device is executed; The receiving the AI ​​computing capabilities sent by the plurality of second network function devices through their respective corresponding AI data transmission channels includes: Receiving the AI ​​computing capabilities sent by the plurality of second network function devices through their respective corresponding default transmission channels; The sending the first AI subtasks corresponding to each of the plurality of third network function devices through the AI ​​data transmission channels corresponding to each of the plurality of third network function devices comprises: Sending the first AI subtasks corresponding to each of the plurality of third network function devices through the default transmission channels corresponding to each of the plurality of third network function devices; The receiving the first AI subtask execution results sent by the plurality of third network function devices through their respective corresponding AI data transmission channels includes: Receive the first AI subtask execution results sent by the multiple third network function devices through their respective corresponding dedicated transmission channels.

4. The method according to claim 3, characterized in that Also includes: Updating the first AI task to obtain a second AI task; Dividing the second AI task into a plurality of second AI subtasks based on the AI ​​computing capabilities of the plurality of second network function devices, and selecting a plurality of fourth network function devices from the plurality of second network function devices for executing the plurality of second AI subtasks; Sending the second AI subtasks corresponding to each of the plurality of fourth network function devices through the default transmission channels corresponding to each of the plurality of fourth network function devices; Receive the second AI subtask execution results sent by the multiple fourth network function devices through their respective corresponding dedicated transmission channels.

5. The method according to claim 3, characterized in that: Also includes: If the first AI task has been executed and completed, sending a first indication message to the multiple third network function devices through the dedicated transmission channels corresponding to each of the multiple third network function devices; The first indication message is used to indicate that the first AI task has been completed.

6. The method according to claim 3, characterized in that: The dedicated transmission channel uniquely corresponds to the first AI task, or the dedicated transmission channel is a dedicated transmission channel shared by the first AI task and other AI tasks.

7. The method according to any one of claims 1 to 6, characterized in that: The first network function device is connected to the plurality of second network function devices via a fifth network function device; The obtaining of the artificial intelligence AI computing capabilities of each of the plurality of second network function devices includes: receiving respective AI computing capabilities forwarded by the plurality of second network function devices through the fifth network function device; The sending the first AI subtasks corresponding to each of the plurality of third network function devices includes: Sending the first AI subtasks corresponding to each of the plurality of third network function devices through the fifth network function device; The receiving the first AI subtask execution results respectively sent by the plurality of third network function devices includes: Receive respective first AI subtask execution results forwarded by the plurality of third network function devices through the fifth network function device.

8. The method according to any one of claims 1 to 6, characterized in that: The selecting, from the plurality of second network function devices, a plurality of third network function devices corresponding to the plurality of first AI subtasks comprises: Selecting a plurality of sixth network function devices corresponding to the plurality of first AI subtasks from the plurality of second network function devices based on the respective AI computing capabilities of the plurality of second network function devices; Sending the first AI subtasks corresponding to each of the plurality of sixth network function devices; Receiving a second indication message sent by each of the plurality of sixth network function devices; wherein, for any sixth network function device among the plurality of sixth network function devices, the second indication message sent by the sixth network function device is used to indicate whether the sixth network function device agrees to the first AI subtask assigned to the sixth network function device; The plurality of third network function devices are determined based on the second indication messages respectively sent by the plurality of sixth network function devices.

9. The method according to claim 8, characterized in that For any second network function device among the plurality of second network function devices, the AI ​​computing capability of the second network function device includes: an AI model supported by the second network function device and an amount of computing resources of the second network function device; The selecting, from among the plurality of second network function devices, a plurality of sixth network function devices corresponding to the plurality of first AI subtasks based on the respective AI computing capabilities of the plurality of second network function devices comprises: For any second network function device among the multiple second network function devices, if the AI ​​model supported by the second network function device matches at least one first AI subtask among the multiple first AI subtasks, and the computing resource usage of the second network function device meets the computing resource usage of the at least one first AI subtask, then the second network function device is used as the sixth network function device.

10. The method according to claim 9, characterized in that The amount of computing resources includes: the amount of currently available computing resources; the method further includes: If the computing resource usage of the at least one first AI subtask is less than or equal to the currently available computing resource amount, determining that the computing resource amount of the second network function device meets the computing resource usage of the at least one first AI subtask; If the computing resource usage of the at least one first AI subtask is greater than the currently available computing resource amount, it is determined that the computing resource amount of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

11. The method according to claim 9, characterized in that The computing resource amount includes: the total computing resource amount and the currently available computing resource amount; the method further includes: If the total computing resource amount is greater than or equal to the preset value, and the computing resource usage of the at least one first AI subtask is less than or equal to the currently available computing resource amount, determining that the computing resource amount of the second network function device meets the computing resource usage of the at least one first AI subtask; If the total computing resource amount is less than the preset value, or the computing resource usage of the at least one first AI subtask is greater than the currently available computing resource amount, it is determined that the computing resource amount of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

12. The method according to claim 9, characterized in that The computing resource amount includes: the total computing resource amount and the currently available computing resource amount; the method further includes: Calculating the product of the total computing resource amount and a preset coefficient to obtain an estimated computing resource amount; Based on the estimated amount of computing resources and the currently available amount of computing resources, detect whether the amount of computing resources of the second network function device meets the computing resource usage of the at least one first AI subtask.

13. The method according to claim 12, characterized in that The detecting, based on the estimated amount of computing resources and the currently available amount of computing resources, whether the amount of computing resources of the second network function device meets the amount of computing resources used by the at least one first AI subtask includes: Determining the smaller of the estimated amount of computing resources and the currently available amount of computing resources; If the computing resource usage of the at least one first AI subtask is less than or equal to the smaller one, determining that the computing resource usage of the second network function device meets the computing resource usage of the at least one first AI subtask; If the computing resource usage of the at least one first AI subtask is greater than the smaller one, it is determined that the computing resource usage of the second network function device does not meet the computing resource usage of the at least one first AI subtask.

14. A communication method, characterized in that: The method is applied to a third network function device, and the method includes: Sending the AI ​​computing capability of the third network function device to the first network function device; Receiving a first AI subtask corresponding to the third network function device; Execute the first AI subtask to obtain the first AI subtask execution result; Send the first AI subtask execution result to the first network function device.

15. The method according to claim 14, characterized in that The first network function device and the third network function device include: an AI data transmission channel; The sending the AI ​​computing capability of the third network function device to the first network function device includes: sending the AI ​​computing capability of the third network function device to the first network function device through the AI ​​data transmission channel; The receiving a first AI subtask corresponding to the third network function device includes: Receiving a first AI subtask corresponding to the third network function device through the AI ​​data transmission channel; The sending the first AI subtask execution result to the first network function device includes: The first AI subtask execution result is sent to the first network function device through the AI ​​data transmission channel.

16. The method according to claim 15, characterized in that The AI ​​data transmission channel includes: a default transmission channel and a dedicated transmission channel; the default transmission channel is used to transmit the AI ​​data before the first AI subtask corresponding to the third network function device is executed, and the dedicated transmission channel is used to transmit the AI ​​data after the first AI subtask corresponding to the third network function device is executed; The sending the AI ​​computing capability of the third network function device to the first network function device through the AI ​​data transmission channel includes: Sending the AI ​​computing capability of the third network function device to the first network function device through the default transmission channel; The receiving, through the AI ​​data transmission channel, the first AI subtask corresponding to the third network function device includes: Receiving a first AI subtask corresponding to the third network function device through the default transmission channel; The sending the first AI subtask execution result to the first network function device through the AI ​​data transmission channel includes: The first AI subtask execution result is sent to the first network function device through the dedicated transmission channel.

17. The method according to claim 16, characterized in that Also includes: If the first AI task has been executed and completed, receiving a first indication message through the dedicated transmission channel; The first indication message is used to indicate that the first AI task has been completed.

18. A communication device, characterized in that: include: Transceiver module and processing module; The transceiver module is used to obtain the AI ​​computing capability of each of the plurality of second network function devices; The processing module is used to divide the first AI task into a plurality of first AI subtasks based on the AI ​​computing capabilities of the plurality of second network function devices, and select a plurality of third network function devices from the plurality of second network function devices for executing the plurality of first AI subtasks; The transceiver module is also used to send the first AI subtasks corresponding to each of the multiple third network function devices, and receive the first AI subtask execution results sent by each of the multiple third network function devices.

19. A communication device, characterized in that: include: Transceiver module and processing module; The transceiver module is used to send the AI ​​computing capability of the communication device to the first network function device; The transceiver module is also used to receive a first AI subtask corresponding to the communication device; The processing module is used to execute the first AI subtask and obtain the first AI subtask execution result; The transceiver module is also used to send the execution result of the first AI subtask to the first network function device.

20. A network function device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, and the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 16.

Images