Edge computing multi-agent semantic communication method and system

Abstract

The invention discloses an edge computing multi-agent semantic communication method and system, belongs to the technical field of communication information, and aims to solve the technical problem of how to combine semantic communication to realize multi-agent communication in an edge computing scene. On the basis of existing network mobility management, an edge-end agent semantic knowledge model synchronization mechanism and edge computing agent mobility management are added; wherein the side-end agent semantic knowledge model synchronization mechanism specifically comprises the following steps: initial access semantic knowledge model synchronization: adding a semantic knowledge model synchronization message in a process of initial access of a terminal to a core network, sending a semantic knowledge base synchronization request message by a terminal user, and obtaining terminal position information from an existing initial UE message by the core network, the semantic knowledge model is selected and configured in combination with the newly added semantic knowledge model synchronization message; and synchronizing the terminal mobile semantic knowledge model.

Description

Technical Field

[0001] The present invention relates to the field of communication information technology, and specifically to an edge computing multi-agent semantic communication method and system. Background Art

[0002] As an important branch of artificial intelligence, agent technology has been widely used in various scenarios. Agents have the ability to perceive, make decisions and act, which enables them to operate autonomously in complex environments and achieve specific goals through interaction with the environment and other agents. In particular, multi-agent systems (MAS) can effectively enhance intelligent collaboration capabilities through collaborative work, utilize the computing power and data of different agents, and thus improve the efficiency and effectiveness of problem solving. In edge computing scenarios, terminal agents can collaborate with edge computing agents to complete complex tasks such as complex multimodal reasoning, which requires terminal intelligence and edge agents to frequently exchange large amounts of data to maintain efficient collaborative work.

[0003] Existing mobile communication systems mainly rely on information exchange in the form of text, voice and images, which mainly serve the perception of human users. However, with the development of artificial intelligence technology, communication systems need to serve intelligent agents. Semantic communication, as an emerging communication mode, can be used for multi-agent communication. Traditional communication methods mainly focus on the transmission rate and accuracy of data, while semantic communication emphasizes the content and meaning of data, that is, effectively transmitting and understanding the semantic content of information between different intelligent agents.

[0004] Therefore, how to combine semantic communication to realize multi-agent communication in edge computing scenarios is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The technical task of the present invention is to provide an edge computing multi-agent semantic communication method and system to solve the problem of how to combine semantic communication to realize multi-agent communication in edge computing scenarios.

[0006] The technical task of the present invention is achieved in the following way: a semantic communication method for edge computing multi-agents, which is to share knowledge between agents by semantics, add a semantic knowledge model synchronization mechanism for edge agents and mobility management of edge computing agents on the basis of existing network mobility management;

[0007] The synchronization mechanism of the semantic knowledge model of the edge agent is as follows:

[0008] Initial access semantic knowledge model synchronization: During the initial access of the terminal to the core network, a semantic knowledge model synchronization message is added. The terminal user sends a semantic knowledge base synchronization request message. The core network obtains the terminal location information from the existing initial UE messages and selects and configures the semantic knowledge model in combination with the newly added semantic knowledge model synchronization message;

[0009] Terminal mobility semantic knowledge model synchronization: After the terminal initially accesses the core network, as the terminal location changes, the terminal agent collaborates and interacts with the edge agent closest to the location, and the semantic knowledge model is resynchronized; and in the existing handover messages, messages of the edge target area, the terminal semantic knowledge model, and the semantic handover type are added to construct a semantic knowledge model synchronization negotiation mechanism to ensure that after the terminal moves, the semantic knowledge models of the terminal agent and the edge agent are consistent.

[0010] Preferably, the initial access semantic knowledge model synchronization is as follows:

[0011] (1) The terminal powers on and initiates an initial access process to the core network;

[0012] (2) The terminal sends an initial registration message to the core network, and the initial registration message includes the terminal location information;

[0013] (3) The terminal sends a semantic knowledge model synchronization message to the core network, the semantic knowledge model synchronization message; among them, the semantic knowledge model synchronization message includes semantic communication capabilities (such as: supported, not supported), synchronization types (such as: specified by the core network side, specified by the terminal side), and preferred models (such as: semantic model name);

[0014] (4) Determine whether the semantic communication capability is supported or not:

[0015] ① If the semantic communication capability is not supported, the semantic knowledge model process terminates and adjusts to step (9);

[0016] ② If the semantic communication capability is supported, execute step (5);

[0017] (5) The core network uses the terminal location information in the initial registration message as a query condition to select the set Ae of edge computing agents within the terminal's range; and obtains the semantic knowledge models used by each edge computing agent from the set Ae of edge computing agents to construct a semantic knowledge model set Me;

[0018] (6) Determine whether the synchronization type is specified by the core network side or the terminal side:

[0019] ① If the synchronization type value is "specified by the core network side" and the preferred model value is empty, the core network randomly selects a model from the semantic knowledge model set Me and downloads it to the terminal, and executes step (7);

[0020] ② If the synchronization type value is "specified by the core network side" and the preference model value is the name of a specific semantic knowledge model, denoted as m, the situation is as follows:

[0021] If m is in Me, the core network selects model m and sends it to the terminal, and step (7) is executed;

[0022] If the semantic knowledge model m is not in Me, the synchronization fails and step (8) is executed;

[0023] ③ If the synchronization type value is "specified by the terminal side" and the preference model value is empty, step (8) is executed;

[0024] ④ If the synchronization type value is "specified by the terminal side" and the preference model value is the name of a specific semantic knowledge model, denoted as m, the situation is as follows:

[0025] If m is in Me, the core network selects model m and sends it to the terminal, and step (7) is executed;

[0026] If m is not in Me, the core network sends the semantic knowledge model m to the edge computing agent set Ae, and step (7) is executed;

[0027] (7) When the semantic knowledge model synchronization is successful, the core network returns a semantic knowledge model synchronization success response message to the terminal, and step (9) is executed;

[0028] (8) When the semantic knowledge model synchronization fails, the core network returns a semantic knowledge model synchronization failure response message to the terminal;

[0029] (9) The processing ends.

[0030] Preferably, the synchronization of the terminal mobile semantic knowledge model is as follows:

[0031] (1) When the terminal moves, it sends a mobile handover message to the core network; among them, the mobile handover message includes the edge target area and the terminal semantic knowledge model;

[0032] (2) The core network selects the edge computing agent set At within the target area range where the terminal is located based on the edge target area of the terminal handover, obtains the semantic knowledge models used by each edge computing agent from the edge computing agent set At, and constructs a semantic knowledge model set Mt;

[0033] (3) The core network records the terminal semantic knowledge model in the mobile handover message as m;

[0034] (4) Determine whether m is in Mt:

[0035] ① If m is in the set Mt, the synchronization is successful, the terminal semantic knowledge model synchronization is successful, and the core network returns a handover success message;

[0036] ② If m is not in the set Mt and the semantic handover type is "specified by the core network side", the terminal semantic knowledge model synchronization fails, and the core network returns a handover failure message;

[0037] ③ If m is not in the set Mt and the semantic handover type is "specified by the terminal side", the core network distributes the semantic knowledge model m to the edge computing agent set At, the terminal semantic knowledge model synchronization is successful, and the core network returns a handover success message.

[0038] An edge computing multi-agent semantic communication system, which includes a terminal and a core network. The terminal initiates an initial access process to the core network and sends a semantic knowledge model synchronization request message. The synchronization request message includes semantic communication capabilities, synchronization types, and preference models. According to the semantic communication capabilities and synchronization types, the corresponding semantic knowledge model is selected and configured for the terminal. When the terminal moves, the semantic knowledge model is resynchronized through the messages in the existing mobile handover process, including the edge target area, the terminal semantic knowledge model, and the semantic handover type message;

[0039] Among them, when the semantic communication capability is not supported, the semantic knowledge model synchronization process is terminated;

[0040] The selection of the semantic knowledge model is based on the edge computing agent set within the range where the terminal is located and the semantic knowledge models they use.

[0041] Preferably, the initial registration information sent by the terminal to the core network includes the terminal location information.

[0042] More preferably, the semantic knowledge model synchronization request message further includes a semantic knowledge base synchronization request message for establishing semantic communication between the terminal and the core network.

[0043] More preferably, the initial access semantic knowledge model synchronization is as follows:

[0044] (1) The terminal powers on and initiates an initial access process to the core network;

[0045] (2) The terminal sends an initial registration message to the core network, and the initial registration message contains the terminal location information;

[0046] (3) The terminal sends a semantic knowledge model synchronization message to the core network, the semantic knowledge model synchronization message; among them, the semantic knowledge model synchronization message includes semantic communication capabilities (such as: supported, not supported), synchronization types (such as: specified by the core network side, specified by the terminal side), and preference models (such as: semantic model name);

[0047] (4) Determine whether the semantic communication ability is supported or not:

[0048] ① If the semantic communication ability is not supported, the semantic knowledge model process terminates and adjusts to step (9);

[0049] ② If the semantic communication ability is supported, execute step (5);

[0050] (5) The core network uses the terminal location information in the initial registration message as a query condition, selects the edge computing agent set Ae within the range where the terminal is located; and obtains the semantic knowledge models used by each edge computing agent from the edge computing agent set Ae, and constructs the semantic knowledge model set Me;

[0051] (6) Determine whether the synchronization type is specified by the core network side or the terminal side:

[0052] ① If the synchronization type value is "specified by the core network side" and the preference model value is empty, the core network randomly selects a model from the semantic knowledge model set Me and downloads it to the terminal, and executes step (7);

[0053] ② If the synchronization type value is "specified by the core network side" and the preference model value is the name of a specific semantic knowledge model, denoted as m, the situation is as follows:

[0054] If m is in Me, the core network selects model m and downloads it to the terminal, and executes step (7);

[0055] If the semantic knowledge model m is not in Me, the synchronization fails and step (8) is executed;

[0056] ③ If the synchronization type value is "specified by the terminal side" and the preference model value is empty, execute step (8);

[0057] ④ If the synchronization type value is "specified by the terminal side" and the preference model value is the name of a specific semantic knowledge model denoted as m, the situation is as follows:

[0058] If m is in Me, the core network selects model m and downloads it to the terminal, and executes step (7);

[0059] If m is not in Me, the core network downloads the semantic knowledge model m to the edge computing agent set Ae and executes step (7);

[0060] (7) When the semantic knowledge model synchronization is successful, the core network returns a semantic knowledge model synchronization success response message to the terminal and executes step (9);

[0061] (8) When the semantic knowledge model synchronization fails, the core network returns a semantic knowledge model synchronization failure response message to the terminal;

[0062] (9) Processing is completed.

[0063] Preferably, the synchronization of the terminal movement semantic knowledge model is as follows:

[0064] (1) The terminal moves, and sends a handover message to the core network; wherein, the handover message includes the edge target area and the terminal semantic knowledge model;

[0065] (2) The core network selects the set At of edge computing agents within the target area range where the terminal is located based on the edge target area of the terminal handover, obtains the semantic knowledge model used by each edge computing agent from the set At of edge computing agents, and constructs a semantic knowledge model set Mt;

[0066] (3) The core network records the terminal semantic knowledge model in the handover message as m;

[0067] (4) Determine whether m is in Mt:

[0068] ① If m is in the set Mt, the synchronization is successful, the synchronization of the terminal semantic knowledge model is successful, and the core network returns a handover success message;

[0069] ② If m is not in the set Mt, the semantic handover type is "specified by the core network side", the synchronization of the terminal semantic knowledge model fails, and the core network returns a handover failure message;

[0070] ③ If m is not in the set Mt and the semantic handover type is "specified by the terminal side", the core network distributes the semantic knowledge model m to the set At of edge computing agents, the synchronization of the terminal semantic knowledge model is successful, and the core network returns a handover success message.

[0071] An electronic device, comprising: a memory and at least one processor;

[0072] Wherein, a computer program is stored on the memory;

[0073] The at least one processor executes the computer program stored in the memory, so that the at least one processor executes the edge computing multi-agent semantic communication method as described above.

[0074] A computer-readable storage medium, in which a computer program is stored, and the computer program can be executed by a processor to implement the edge computing multi-agent semantic communication method as described above.

[0075] The edge computing multi-agent semantic communication method and system of the present invention have the following advantages:

[0076] (1) The present invention shares knowledge between agents semantically, adding a synchronization mechanism for the semantic knowledge model of edge agents and mobility management for edge computing agents on the basis of existing network mobility management. Among them, the core of semantic communication is knowledge sharing. The same semantic knowledge model is used between agents. The agent inputs the original data into the semantic knowledge model to obtain a semantic vector containing semantic content. The semantic vector is the result of compression and extraction of the original input information, and the space occupied by the semantic vector is smaller than that of the original input information. For semantic communication between agents, the semantic vector is transmitted on the transmission channel. The prerequisite for semantic communication among multiple edge computing agents is that the semantic knowledge models of the terminal agent and the edge agent must be consistent.

[0077] (2) The present invention has a synchronization mechanism for the semantic knowledge model of edge agents, including synchronization of the semantic knowledge model for initial access and synchronization of the semantic knowledge model for terminal mobility, providing a solid foundation for building a more intelligent and flexible multi-agent system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The present invention will be further described below with reference to the accompanying drawings.

[0079] Appendix Figure 1 It is a flowchart of a semantic communication method for multiple edge computing agents. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0080] The semantic communication method and system for multiple edge computing agents of the present invention will be described in detail below with reference to the accompanying drawings of the specification and specific embodiments.

[0081] Embodiment 1:

[0082] As shown in the appendix Figure 1 This embodiment provides a semantic communication method for multiple edge computing agents. This method shares knowledge between agents semantically, adding a synchronization mechanism for the semantic knowledge model of edge agents and mobility management for edge computing agents on the basis of existing network mobility management.

[0083] Among them, the synchronization mechanism for the semantic knowledge model of edge agents is as follows:

[0084] Synchronization of the semantic knowledge model for initial access: During the initial access of the terminal to the core network, a semantic knowledge model synchronization message is added. The terminal user sends a semantic knowledge base synchronization request message. The core network obtains the terminal location information from the existing initial UE message and selects and configures the semantic knowledge model in combination with the newly added semantic knowledge model synchronization message.

[0085] Synchronization of the terminal mobile semantic knowledge model: After the terminal initially accesses the core network, as the terminal's location changes, the terminal agent collaborates and interacts with the edge agent closest to the location, and the semantic knowledge model is resynchronized. In the existing handover messages, messages about the edge target area, the terminal semantic knowledge model, and the semantic handover type are added to build a semantic knowledge model synchronization negotiation mechanism to ensure that after the terminal moves, the semantic knowledge models of the terminal agent and the edge agent are consistent.

[0086] In this embodiment, the synchronization of the initial access semantic knowledge model is as follows:

[0087] (1) The terminal powers on and initiates an initial access process to the core network;

[0088] (2) The terminal sends an initial registration message to the core network, and the initial registration message includes the terminal location information;

[0089] (3) The terminal sends a semantic knowledge model synchronization message to the core network, the semantic knowledge model synchronization message; among them, the semantic knowledge model synchronization message includes semantic communication capabilities (such as: supported, not supported), synchronization types (such as: specified by the core network side, specified by the terminal side), and preference models (such as: semantic model name);

[0090] (4) Determine whether the semantic communication capability is supported or not:

[0091] ① If the semantic communication capability is not supported, the semantic knowledge model process terminates and adjusts to step (9);

[0092] ② If the semantic communication capability is supported, execute step (5);

[0093] (5) The core network uses the terminal location information in the initial registration message as a query condition to select the set Ae of edge computing agents within the terminal's range; and obtains the semantic knowledge models used by each edge computing agent from the set Ae of edge computing agents to construct the semantic knowledge model set Me;

[0094] (6) Determine whether the synchronization type is specified by the core network side or the terminal side:

[0095] ① If the synchronization type value is "specified by the core network side" and the preference model value is empty, the core network randomly selects a model from the semantic knowledge model set Me and downloads it to the terminal, and executes step (7);

[0096] ② If the synchronization type value is "specified by the core network side" and the preference model value is the specific semantic knowledge model name, denoted as m, the situation is as follows:

[0097] If m is in Me, the core network selects the model m and downloads it to the terminal, and executes step (7);

[0098] If the semantic knowledge model m is not in Me, the synchronization fails, and step (8) is executed;

[0099] ③ If the synchronization type value is "specified by the terminal side" and the preference model value is empty, step (8) is executed;

[0100] ④ If the synchronization type value is "specified by the terminal side" and the preference model value is the name of a specific semantic knowledge model denoted as m, the situation is as follows:

[0101] If m is in Me, the core network selects model m and sends it to the terminal, and step (7) is executed;

[0102] If m is not in Me, the core network sends the semantic knowledge model m to the edge computing agent set Ae, and step (7) is executed;

[0103] (7) When the semantic knowledge model synchronization is successful, the core network returns a semantic knowledge model synchronization success response message to the terminal, and step (9) is executed;

[0104] (8) When the semantic knowledge model synchronization fails, the core network returns a semantic knowledge model synchronization failure response message to the terminal;

[0105] (9) The process ends.

[0106] In this embodiment, the synchronization of the terminal mobile semantic knowledge model is as follows:

[0107] (1) The terminal moves in location and sends a mobile handover message to the core network; among them, the mobile handover message includes the edge target area and the terminal semantic knowledge model;

[0108] (2) Based on the edge target area of the terminal handover, the core network selects the edge computing agent set At within the target area range where the terminal is located, obtains the semantic knowledge models used by each edge computing agent from the edge computing agent set At, and constructs a semantic knowledge model set Mt;

[0109] (3) The core network records the terminal semantic knowledge model in the mobile handover message as m;

[0110] (4) Determine whether m is in Mt:

[0111] ① If m is in the set Mt, the synchronization is successful, the terminal semantic knowledge model synchronization is successful, and the core network returns a handover success message;

[0112] ② If m is not in the set Mt, the semantic handover type is "specified by the core network side", the terminal semantic knowledge model synchronization fails, and the core network returns a handover failure message;

[0113] ③ If m is not in the set Mt and the semantic switching type is "specified by the terminal side", the core network will send the semantic knowledge model m to the edge computing agent set At. After the terminal semantic knowledge model is successfully synchronized, the core network returns a switching success message.

[0114] Embodiment 2:

[0115] This embodiment provides an edge computing multi-agent semantic communication system, which includes a terminal and a core network. The terminal initiates an initial access process to the core network and sends a semantic knowledge model synchronization request message. The synchronization request message includes semantic communication capabilities, synchronization types, and preference models. According to the semantic communication capabilities and synchronization types, the corresponding semantic knowledge models are selected and configured for the terminal. When the terminal moves, the semantic knowledge model is resynchronized through the messages in the existing mobile handover process, including the edge target area, the terminal semantic knowledge model, and the semantic switching type message.

[0116] Among them, when the semantic communication capability is not supported, the semantic knowledge model synchronization process is terminated.

[0117] The selection of the semantic knowledge model is based on the edge computing agent set within the range where the terminal is located and the semantic knowledge models they use.

[0118] The initial registration information sent by the terminal to the core network in this embodiment includes the terminal location information.

[0119] The semantic knowledge model synchronization request message in this embodiment also includes a semantic knowledge base synchronization request message, which is used to establish semantic communication between the terminal and the core network.

[0120] In this embodiment, the initial access semantic knowledge model synchronization is as follows:

[0121] (1) The terminal is powered on, and the terminal initiates an initial access process to the core network;

[0122] (2) The terminal sends an initial registration message to the core network, and the initial registration message contains the terminal location information;

[0123] (3) The terminal sends a semantic knowledge model synchronization message to the core network, and the semantic knowledge model synchronization message; among them, the semantic knowledge model synchronization message includes semantic communication capabilities (such as: supported, not supported), synchronization types (such as: specified by the core network side, specified by the terminal side), and preference models (such as: semantic model name);

[0124] (4) Determine whether the semantic communication capability is supported or not:

[0125] ① If the semantic communication capability is not supported, the semantic knowledge model process is terminated, and it is adjusted to step (9);

[0126] ②If the semantic communication ability is supported, execute step (5);

[0127] (5) The core network uses the terminal location information in the initial registration message as a query condition, selects the set Ae of edge computing agents within the range where the terminal is located; and obtains the semantic knowledge models used by each edge computing agent from the set Ae of edge computing agents, and constructs the set Me of semantic knowledge models;

[0128] (6) Determine whether the synchronization type is specified by the core network side or the terminal side:

[0129] ① If the synchronization type value is "specified by the core network side" and the preference model value is empty, the core network randomly selects a model from the set Me of semantic knowledge models and sends it to the terminal, and executes step (7);

[0130] ② If the synchronization type value is "specified by the core network side" and the preference model value is the name of a specific semantic knowledge model, denoted as m, the situation is as follows:

[0131] If m is in Me, the core network selects model m and sends it to the terminal, and executes step (7);

[0132] If the semantic knowledge model m is not in Me, the synchronization fails, and step (8) is executed;

[0133] ③ If the synchronization type value is "specified by the terminal side" and the preference model value is empty, execute step (8);

[0134] ④ If the synchronization type value is "specified by the terminal side" and the preference model value is the name of a specific semantic knowledge model denoted as m, the situation is as follows:

[0135] If m is in Me, the core network selects model m and sends it to the terminal, and executes step (7);

[0136] If m is not in Me, the core network sends the semantic knowledge model m to the set Ae of edge computing agents, and executes step (7);

[0137] (7) When the synchronization of the semantic knowledge model is successful, the core network returns a semantic knowledge model synchronization success response message to the terminal, and executes step (9);

[0138] (8) When the synchronization of the semantic knowledge model fails, the core network returns a semantic knowledge model synchronization failure response message to the terminal;

[0139] (9) The process ends.

[0140] In this embodiment, the synchronization of the terminal mobile semantic knowledge model is as follows:

[0141] (1) The terminal moves its position and sends a handover message to the core network. The handover message includes the edge target area and the terminal semantic knowledge model.

[0142] (2) The core network selects the set of edge computing agents At within the target area range where the terminal is located based on the edge target area of the terminal handover, obtains the semantic knowledge model used by each edge computing agent from the set of edge computing agents At, and constructs the semantic knowledge model set Mt.

[0143] (3) The core network records the terminal semantic knowledge model in the handover message as m.

[0144] (4) Determine whether m is in Mt:

[0145] ① If m is in the set Mt, the synchronization is successful, the terminal semantic knowledge model synchronization is successful, and the core network returns a handover success message.

[0146] ② If m is not in the set Mt, the semantic handover type is "specified by the core network side", the terminal semantic knowledge model synchronization fails, and the core network returns a handover failure message.

[0147] ③ If m is not in the set Mt and the semantic handover type is "specified by the terminal side", the core network distributes the semantic knowledge model m to the set of edge computing agents At, the terminal semantic knowledge model synchronization is successful, and the core network returns a handover success message.

[0148] Embodiment 3:

[0149] This embodiment also provides an electronic device, including: a memory and a processor;

[0150] Among them, the memory stores computer execution instructions;

[0151] The processor executes the computer execution instructions stored in the memory, so that the processor executes the edge computing multi-agent semantic communication method in any embodiment of the present invention.

[0152] The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0153] The memory can be used to store computer programs and / or modules. By running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory, the processor realizes various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function, etc.; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory can also include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash memory card, at least one magnetic disk storage period, flash memory device, or other volatile solid-state storage devices.

[0154] Embodiment 4:

[0155] This embodiment also provides a computer-readable storage medium, in which multiple instructions are stored. The instructions are loaded by the processor to make the processor execute the edge computing multi-agent semantic communication method in any embodiment of the present invention. Specifically, a system or device equipped with a storage medium can be provided. On this storage medium, software program codes for realizing the functions in any one of the above embodiments are stored, and the computer (or CPU or MPU) of the system or device is made to read and execute the program codes stored in the storage medium.

[0156] In this case, the program code read from the storage medium itself can realize the functions in any one of the above embodiments. Therefore, the program code and the storage medium storing the program code constitute a part of the present invention.

[0157] Embodiments of the storage medium for providing program codes include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RYM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Optionally, the program code can be downloaded from a server computer through a communication network.

[0158] In addition, it should be clear that not only can the functions in any one of the above embodiments be realized by executing the program code read by the computer, but also by making the operating system etc. operating on the computer based on the instructions of the program code complete part or all of the actual operations.

[0159] In addition, it can be understood that the program code read from the storage medium is written into the memory provided in the expansion board inserted into the computer or into the memory provided in the expansion unit connected to the computer, and then based on the instructions of the program code, the CPU etc. installed on the expansion board or expansion unit execute part and all of the actual operations, so as to realize the functions in any one of the above embodiments.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An edge computing multi-agent semantic communication method, characterized in that: This method uses semantics to share knowledge between agents, and adds a semantic knowledge model synchronization mechanism for edge agents and mobility management for edge computing agents on the basis of existing network mobility management. The synchronization mechanism of the semantic knowledge model of the edge agent is as follows: Initial access semantic knowledge model synchronization: During the initial access of the terminal to the core network, a semantic knowledge model synchronization message is added. The terminal user sends a semantic knowledge base synchronization request message. The core network obtains the terminal location information from the existing initial UE message and selects and configures the semantic knowledge model in combination with the newly added semantic knowledge model synchronization message. Terminal mobile semantic knowledge model synchronization: After the terminal initially accesses the core network, as the terminal location changes, the terminal agent interacts with the nearby edge agent and the semantic knowledge model is resynchronized. In addition, messages about the edge target area, terminal semantic knowledge model, and semantic switching type are added to the existing switching message to build a semantic knowledge model synchronization negotiation mechanism to ensure that the semantic knowledge models of the terminal agent and the edge agent are consistent after the terminal moves.

2. The edge computing multi-agent semantic communication method according to claim 1, characterized in that: The initial access semantic knowledge model synchronization is as follows: (1) The terminal is powered on and initiates the initial access process to the core network; (2) The terminal sends an initial registration message to the core network, which contains the terminal location information; (3) The terminal sends a semantic knowledge model synchronization message to the core network, wherein the semantic knowledge model synchronization message includes semantic communication capability, synchronization type, and preference model; (4) Determine whether the semantic communication capability is supported or not: ① If the semantic communication capability is not supported, the semantic knowledge model process is terminated and adjusted to step (9); ② If the semantic communication capability is supported, execute step (5); (5) The core network uses the terminal location information in the initial registration message as a query condition to select the edge computing agent set Ae within the range of the terminal; and obtains the semantic knowledge model used by each edge computing agent from the edge computing agent set Ae to construct the semantic knowledge model set Me; (6) Determine whether the synchronization type is set by the core network side or specified by the terminal side: ① If the synchronization type value is "core network side specified" and the preferred model value is empty, the core network randomly selects a model from the semantic knowledge model set Me, sends it to the terminal, and executes step (7); ② If the synchronization type value is "core network side specified", the preference model value is the specific semantic knowledge model name, denoted as m, the situation is as follows: If m is in Me, the core network selects model m and sends it to the terminal, and executes step (7); If the semantic knowledge model m is not in Me, the synchronization fails and step (8) is executed; ③ If the synchronization type value is "terminal side specified" and the preference model value is empty, execute step (8); ④ If the synchronization type value is "terminal side specified", the preference model value is the specific semantic knowledge model name recorded as m, the situation is as follows: If m is in Me, the core network selects model m and sends it to the terminal, and executes step (7); If m is not in Me, the core network sends the semantic knowledge model m to the edge computing agent set Ae and executes step (7); (7) When the semantic knowledge model synchronization is successful, the core network returns a semantic knowledge model synchronization success response message to the terminal and executes step (9); (8) When the semantic knowledge model synchronization fails, the core network returns a semantic knowledge model synchronization failure response message to the terminal; (9)Processing ends.

3. The edge computing multi-agent semantic communication method according to claim 1 or 2, characterized in that: The details of the terminal mobile semantic knowledge model synchronization are as follows: (1) The terminal moves and initiates a mobile handover message to the core network; the mobile handover message includes the edge target area and the terminal semantic knowledge model; (2) The core network selects the edge computing agent set At within the target area where the terminal is located based on the edge target area that the terminal is switching to, obtains the semantic knowledge model used by each edge computing agent from the edge computing agent set At, and constructs the semantic knowledge model set Mt; (3) The core network uses the terminal semantic knowledge model in the mobile handover message as m; (4) Determine whether m is in Mt: ① If m is in the set Mt, the synchronization is successful, the terminal semantic knowledge model is synchronized successfully, and the core network returns a handover success message; ② If m is not in the set Mt, the semantic switching type is "specified on the core network side", the terminal semantic knowledge model synchronization fails, and the core network returns a switching failure message; ③If m is not in the set Mt, the semantic switching type is "terminal side specified", the core network sends the semantic knowledge model m to the edge computing agent set At, the terminal semantic knowledge model is synchronized successfully, and the core network returns a switching success message.

4. An edge computing multi-agent semantic communication system, characterized in that: The system includes a terminal and a core network. The terminal initiates an initial access process to the core network and sends a semantic knowledge model synchronization request message. The synchronization request message includes semantic communication capability, synchronization type and preferred model. According to the semantic communication capability and synchronization type, a corresponding semantic knowledge model is selected and configured to the terminal. When the terminal moves, the edge target area, the terminal semantic knowledge model and the semantic switching type message are added through the message in the existing mobile switching process to resynchronize the semantic knowledge model. Wherein, when the semantic communication capability is not supported, the semantic knowledge model synchronization process is terminated; The selection of the semantic knowledge model is based on the set of edge computing agents within the range of the terminal and the semantic knowledge models they use.

5. The edge computing multi-agent semantic communication system according to claim 4, characterized in that: The initial registration information sent by the terminal to the core network includes the terminal location information.

6. The edge computing multi-agent semantic communication system according to claim 4 or 5, characterized in that: The semantic knowledge model synchronization request message also includes a semantic knowledge base synchronization request message, which is used to establish semantic communication between the terminal and the core network.

7. The edge computing multi-agent semantic communication system according to claim 6, characterized in that: The initial access semantic knowledge model synchronization is as follows: (1) The terminal is powered on and initiates the initial access process to the core network; (2) The terminal sends an initial registration message to the core network, which contains the terminal location information; (3) The terminal sends a semantic knowledge model synchronization message to the core network, wherein the semantic knowledge model synchronization message includes semantic communication capability, synchronization type, and preference model; (4) Determine whether the semantic communication capability is supported or not: ① If the semantic communication capability is not supported, the semantic knowledge model process is terminated and adjusted to step (9); ② If the semantic communication capability is supported, execute step (5); (5) The core network uses the terminal location information in the initial registration message as a query condition to select the edge computing agent set Ae within the range of the terminal; and obtains the semantic knowledge model used by each edge computing agent from the edge computing agent set Ae to construct the semantic knowledge model set Me; (6) Determine whether the synchronization type is set by the core network side or specified by the terminal side: ① If the synchronization type value is "core network side specified" and the preferred model value is empty, the core network randomly selects a model from the semantic knowledge model set Me, sends it to the terminal, and executes step (7); ② If the synchronization type value is "core network side specified", the preference model value is the specific semantic knowledge model name, denoted as m, the situation is as follows: If m is in Me, the core network selects model m and sends it to the terminal, and executes step (7); If the semantic knowledge model m is not in Me, the synchronization fails and step (8) is executed; ③ If the synchronization type value is "terminal side specified" and the preference model value is empty, execute step (8); ④ If the synchronization type value is "terminal side specified", the preference model value is the specific semantic knowledge model name recorded as m, the situation is as follows: If m is in Me, the core network selects model m and sends it to the terminal, and executes step (7); If m is not in Me, the core network sends the semantic knowledge model m to the edge computing agent set Ae and executes step (7); (7) When the semantic knowledge model synchronization is successful, the core network returns a semantic knowledge model synchronization success response message to the terminal and executes step (9); (8) When the semantic knowledge model synchronization fails, the core network returns a semantic knowledge model synchronization failure response message to the terminal; (9)Processing ends.

8. The edge computing multi-agent semantic communication system according to claim 7, characterized in that: The details of the terminal mobile semantic knowledge model synchronization are as follows: (1) The terminal moves and initiates a mobile handover message to the core network; the mobile handover message includes the edge target area and the terminal semantic knowledge model; (2) The core network selects the edge computing agent set At within the target area where the terminal is located based on the edge target area that the terminal is switching to, obtains the semantic knowledge model used by each edge computing agent from the edge computing agent set At, and constructs the semantic knowledge model set Mt; (3) The core network uses the terminal semantic knowledge model in the mobile handover message as m; (4) Determine whether m is in Mt: ① If m is in the set Mt, the synchronization is successful, the terminal semantic knowledge model is synchronized successfully, and the core network returns a handover success message; ② If m is not in the set Mt, the semantic switching type is "specified on the core network side", the terminal semantic knowledge model synchronization fails, and the core network returns a switching failure message; ③If m is not in the set Mt, the semantic switching type is "terminal side specified", the core network sends the semantic knowledge model m to the edge computing agent set At, the terminal semantic knowledge model is synchronized successfully, and the core network returns a switching success message.

9. An electronic device, characterized in that: include: memory and at least one processor; Wherein, the memory stores a computer program; The at least one processor executes the computer program stored in the memory, so that the at least one processor executes the edge computing multi-agent semantic communication method as described in any one of claims 1 to 3.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which can be executed by a processor to implement the edge computing multi-agent semantic communication method as described in any one of claims 1 to 3.

Images