Communication device, method, apparatus, storage medium, chip, product, and program

Abstract

The embodiment of the invention provides communication equipment, a communication method, a communication device, a storage medium, a chip, a product and a program. When the communication equipment is management equipment, the management equipment comprises at least one target AI / ML entity; each target AI / ML entity is used for executing AI / ML related tasks corresponding to the management equipment.

Description

Communication equipment, methods, devices, storage media, chips, products and programs Technical Field

[0001] The embodiments of the present application relate to the field of mobile communication technologies, and specifically to a communication device, method, apparatus, storage medium, chip, product, and program. Background Art

[0002] Given the tremendous success of artificial intelligence (AI) or machine learning (ML) technology in computer vision, natural language processing, and other fields, the communications field has begun to try to use AI technology to seek new technical ideas to solve technical problems that are limited by traditional methods.

[0003] However, there is no technology in the related art that combines AI technology or ML technology with a communication system.

[0004] Summary of the Invention

[0005] Embodiments of the present application provide a communication device, method, apparatus, storage medium, chip, product, and program.

[0006] In a first aspect, an embodiment of the present application provides a management device, comprising at least one target AI / ML entity; each of the target AI / ML entities is configured to execute an AI / ML-related task corresponding to the management device.

[0007] In a second aspect, an embodiment of the present application provides a core network device, wherein the core network device includes at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the core network device.

[0008] In a third aspect, an embodiment of the present application provides an access network device, wherein the access network device includes at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the access network device.

[0009] In a fourth aspect, an embodiment of the present application provides a terminal device, comprising at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the terminal device.

[0010] In a fifth aspect, an embodiment of the present application provides a communication method, the method comprising:

[0011] The communication device performs an AI / ML-related task corresponding to the communication device; wherein the AI / ML-related task corresponding to the communication device includes at least one of the following:

[0012] AI / ML related tasks generated by the communication device;

[0013] AI / ML-related tasks requested by a target AI / ML entity deployed on a device other than the communication device;

[0014] AI / ML-related tasks sent to a target AI / ML entity deployed on a device other than the communication device;

[0015] Training, reasoning, or monitoring the AI / ML model in the communication device.

[0016] In a sixth aspect, an embodiment of the present application provides a communication device, the communication device comprising:

[0017] an execution unit, configured to execute an AI / ML-related task corresponding to the communication device; wherein the AI / ML-related task corresponding to the communication device includes at least one of the following:

[0018] AI / ML related tasks generated by the communication device;

[0019] AI / ML-related tasks requested by a target AI / ML entity deployed on a device other than the communication device;

[0020] AI / ML-related tasks sent to a target AI / ML entity deployed on a device other than the communication device;

[0021] Training, reasoning, or monitoring the AI / ML model in the communication device.

[0022] In a seventh aspect, an embodiment of the present application provides a communication device, including: a processor and a memory,

[0023] The memory stores a computer program executable on the processor.

[0024] The above method is implemented when the processor executes the program.

[0025] In an eighth aspect, an embodiment of the present application provides a computer storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above method.

[0026] In a ninth aspect, an embodiment of the present application provides a chip, comprising: a processor, configured to call and run a computer program from a memory to implement the above method.

[0027] In the tenth aspect, an embodiment of the present application provides a computer program product, which includes a computer storage medium, the computer storage medium storing a computer program, and the computer program including instructions that can be executed by at least one processor, and when the instructions are executed by the at least one processor, the above method is implemented.

[0028] In an eleventh aspect, an embodiment of the present application provides a computer program, which enables a computer to execute the above method.

[0029] In an embodiment of the present application, the management device includes at least one target AI / ML entity; each target AI / ML entity is configured to execute an AI / ML-related task corresponding to the management device. Thus, by including at least one target AI / ML entity in the management device, the target AI / ML entity is enabled to execute the AI / ML-related task corresponding to the management device, thereby achieving a solution that combines AI or ML technology with a communication system and improving information processing efficiency through AI or ML technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0031] FIG1 is a schematic diagram of an application scenario of an embodiment of the present application;

[0032] FIG2 is a schematic diagram of the structure of a management device provided in an embodiment of the present application;

[0033] FIG3 is a schematic diagram of the structure of a core network device provided in an embodiment of the present application;

[0034] FIG4 is a schematic structural diagram of an access network device provided in an embodiment of the present application;

[0035] FIG5 is a schematic structural diagram of a terminal device provided in an embodiment of the present application;

[0036] FIG6 is a schematic diagram of the architecture of a communication system provided in an embodiment of the present application;

[0037] FIG7 is a schematic diagram of communication between a management device and other communication nodes provided in an embodiment of the present application;

[0038] FIG8 is a schematic diagram of another management device communicating with other communication nodes provided in an embodiment of the present application;

[0039] FIG9 is a schematic diagram of a core network device communicating with other communication nodes according to an embodiment of the present application;

[0040] FIG10 is a schematic diagram of another core network device communicating with other communication nodes provided in an embodiment of the present application;

[0041] FIG11 is a schematic diagram of communication between an access network device and other communication nodes provided in an embodiment of the present application;

[0042] FIG12 is a schematic diagram of another access network device communicating with other communication nodes according to an embodiment of the present application;

[0043] FIG13 is a schematic structural diagram of a terminal device provided in an embodiment of the present application;

[0044] FIG14 is a flow chart of a communication method provided in an embodiment of the present application;

[0045] FIG15 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0046] FIG16 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0047] FIG17 is a schematic structural diagram of a chip according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0049] The technical solutions described in the embodiments of the present application can be combined arbitrarily without conflict. In the description of the present application, "multiple" means two or more, unless otherwise clearly defined.

[0050] FIG1 is a schematic diagram of an application scenario of an embodiment of the present application. As shown in FIG1 , a communication system 100 may include a terminal device 110 , an access network device 120 , a core network (CN) device 130 , and a management device 140 .

[0051] The access network equipment may provide communication coverage for a specific geographical area and may communicate with terminal devices 110 (eg, UE) located within the coverage area.

[0052] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: 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, Advanced Long Term Evolution (LTE-A) system, New Radio (NR) system, NR system evolution system, LTE on unlicensed spectrum (LTE-U) system, NR on unlicensed spectrum (NR-U) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Wireless Fidelity (WiFi), LTE Time Division Duplex (TDD), ... Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidelity (WiFi), Wireless Fidel System, UMTS), Internet of Things (IoT) system, Narrow Band Internet of Things (NB-IoT) system, enhanced Machine-Type Communications (eMTC) system, or future communication systems (such as 6G and 7G communication systems).

[0053] The terminal device in this application can be a device with wireless communication capabilities, which can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; can also be deployed on the water (such as ships, etc.); can also be deployed in the air (such as airplanes, balloons and satellites, etc.). The terminal device in this application can be called user equipment (UE), mobile station (MS), mobile terminal (MT), subscriber unit, subscriber station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device. The terminal device may include one of the following or a combination of at least two of the following: Internet of Things (IoT) devices, satellite terminals, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, servers, mobile phones, tablet computers, computers with wireless transceiver capabilities, handheld computers, desktop computers, personal digital assistants, portable media players, smart speakers, navigation devices, smart watches, smart glasses, smart necklaces and other wearable devices, pedometers, digital TVs, Virtual Reality (VR) terminal devices, Augmented Reality (AR) terminal devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, and wireless terminals in smart homes. The terminal device 110 includes wireless terminals in a home, as well as vehicles, onboard devices, onboard modules, wireless modems, handheld devices, customer premises equipment (CPE), smart home appliances, etc. in the Internet of Vehicles system. Optionally, the terminal device 110 can be any terminal device, including but not limited to terminal devices that use wired or wireless connections with the access network device 120 and other terminal devices. Optionally, the terminal device 110 can be used for device-to-device (D2D) communication.

[0054] The access network device 120 may include one of the following or a combination of at least two: an evolved base station (eNB or eNodeB) in a Long Term Evolution (LTE) system, a next generation radio access network (NG RAN) device, a base station (gNB) in an NR system, a small station, a micro station, a wireless controller in a cloud radio access network (CRAN), a wireless fidelity (Wi-Fi) access point, a transmission reception point (TRP), a relay station, an access point, a vehicle-mounted device, a wearable device, a hub, a switch, a bridge, a router, a network device in a future evolved public land mobile network (PLMN), etc.

[0055] The core network device 130 may be a 5G core network (5G Core, 5GC) device, and the core network device 130 may include one of the following or a combination of at least two: an access and mobility management function (AMF) network element, an authentication server function (AUSF) network element, a user plane function (UPF) network element, a session management function (SMF) network element, a location management function (LMF) network element, a policy control function (PCF) network element, and a unified data management (UDM) network element. In other embodiments, the core network device may also be an evolved packet core (EPC) device of an LTE network, for example, a session management function + core network data gateway (SMF+PGW-C) device. It should be understood that SMF+PGW-C can simultaneously implement the functions that SMF and PGW-C can implement. During the network evolution process, the core network device 130 may also be called other names, or a new network entity may be formed by dividing the functions of the core network, which is not limited in the embodiments of the present application.

[0056] The functional units in the communication system 100 may also establish connections and implement communication via next generation (NG) network interfaces.

[0057] For example, the terminal device establishes an air interface connection with the access network device through the NR interface for transmitting user plane data and control plane signaling; the terminal device can establish a control plane signaling connection with the AMF through the NG interface 1 (referred to as N1); the access network device, such as the next generation wireless access base station (gNB), can establish a user plane data connection with the UPF through the NG interface 3 (referred to as N3); the access network device can establish a control plane signaling connection with the AMF through the NG interface 2 (referred to as N2); the UPF can establish a control plane signaling connection with the SMF through the NG interface 4 (referred to as N4); the UPF can exchange user plane data with the data network through the NG interface 6 (referred to as N6); the AMF can establish a control plane signaling connection with the SMF through the NG interface 11 (referred to as N11); the SMF can establish a control plane signaling connection with the PCF through the NG interface 7 (referred to as N7).

[0058] Figure 1 exemplarily shows a base station, a core network device and two terminal devices. Optionally, the wireless communication system 100 may include multiple base station devices and each base station may include other numbers of terminal devices within its coverage area, which is not limited in this embodiment of the present application.

[0059] It should be noted that Figure 1 is merely an example of a system applicable to this application. Of course, the methods described in the embodiments of this application can also be applied to other systems. Furthermore, the terms "system" and "network" are often used interchangeably herein. The term "and / or" herein simply describes an association relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " generally indicates that the associated objects are in an "or" relationship. It should also be understood that the "indication" mentioned in the embodiments of this application can be a direct indication, an indirect indication, or an indication of an association relationship. For example, "A indicates B" can mean that A directly indicates B, for example, B can obtain information through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can obtain information through C; or it can mean that A and B have an association relationship. It should also be understood that the "correspondence" mentioned in the embodiments of this application can mean that there is a direct or indirect correspondence between two objects, or that there is an association relationship between the two objects, or a relationship between an indicator and the indicated, a configuration and the configured, and so on. It should also be understood that the “predefined”, “protocol agreement”, “predetermined” or “predefined rules” mentioned in the embodiments of the present application can be implemented by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in devices (for example, including terminal devices and network devices). This application does not limit its specific implementation method. For example, predefined can refer to what is defined in the protocol. It should also be understood that in the embodiments of the present application, the “protocol” may refer to a standard protocol in the field of communications, for example, it may include LTE protocols, NR protocols, and related protocols used in future communication systems, and this application does not limit this.

[0060] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following relevant technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

[0061] 5G introduces several features:

[0062] The Self-Organizing Networks (SON) and Minimization of Drive Tests (MDT) mechanisms are primarily used to collect data related to function optimization, such as handover optimization, coverage optimization, and random access optimization. The data collection is conducted by Operation Administration and Maintenance (OAM). How the collected data is used and how optimization actions are defined are all implemented.

[0063] The Network Data Analytics Function (NWDAF) is deployed in the core network and is primarily used for function optimization analysis. For each optimization objective, the NWDAF entity defines the type of data to be collected and the possible data sources. It also provides possible optimization analysis behavior outputs. Of course, how to use the collected data to obtain the optimization analysis behavior output is also handled by the implementation method.

[0064] The Management Data Analytics (MDA) function is deployed in the Operation Management and Maintenance (OAM) and is primarily used for functional optimization analysis. For each optimization objective, the MDA entity defines the type of data to be collected and the possible data sources. It also provides possible optimization analysis behavior outputs. Of course, how to use the collected data to obtain the optimization analysis behavior outputs is also handled through implementation. Unlike the NWDAF function, the MDA functional entity can even use the data output by the NWDAF functional entity.

[0065] 5G's existing data collection and data analysis functions basically adopt a centralized processing model. This processing model has at least one of the following problems: data collection is too centralized, which is not conducive to accurate, timely and secure data transmission; data analysis and processing are too centralized, which places too high demands on the resources and functions of the data processing center and is not conducive to timely problem solving nearby; the data analysis and processing supervision mechanism is imperfect, and it is impossible to effectively evaluate the data analysis and processing results, so as to timely adjust the data analysis and processing strategy.

[0066] Therefore, the above shortcomings determine that the existing function optimization of 5G basically adopts a post-processing mode, that is, optimization measures are attempted only after the problem occurs, which makes it difficult to solve the problem in real time.

[0067] However, with the continuous development of AI / ML technology, the combination of communication technology and AI / ML technology is one of the trends in future communications. However, the various functions of the communication system often have very high real-time requirements. In most cases, problems are expected to be solved in real time or near real time. The existing centralized processing mode of 5G is basically unable to achieve the goal of using AI / ML tools to optimize communication problems in real time or near real time.

[0068] Unless otherwise specified, the communication device in the embodiments of the present application may represent at least one of the following: a management device, a core network device, an access network device, and a terminal device.

[0069] To facilitate understanding of the technical solutions of the embodiments of the present application, the technical solutions of the present application are described in detail below through specific embodiments. The above related technologies can be combined arbitrarily with the technical solutions of the embodiments of the present application as optional solutions, and all of them fall within the scope of protection of the embodiments of the present application. The embodiments of the present application include at least part of the following contents.

[0070] Figure 2 is a structural diagram of a management device provided in an embodiment of the present application. As shown in Figure 2, the management device includes at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the management device.

[0071] Optionally, the AI / ML entity in any embodiment of the present application may be an AI entity or a ML entity.

[0072] Optionally, the management device in the embodiment of the present application may be an OAM device. Optionally, the management device may communicate with one or more communication nodes in a core network device. Optionally, the management device may communicate with one or more communication nodes in an access network device.

[0073] Optionally, the functions of the management device may include: link performance detection: monitoring various link performances, including measurement of packet loss, delay and jitter, as well as statistics of various types of traffic; fault detection and alarm: detecting the connectivity of the link by sending detection messages, and notifying the network administrator in a timely manner when a link fails; loop test: detecting link failures by looping back non-Ethernet OAM protocol messages.

[0074] Optionally, the core network device may include one or more communication nodes, or it can be said that one or more communication nodes may be deployed in the core network device. Optionally, in any embodiment of the present application, the communication node included in the core network device or the communication node deployed in the core network device may include at least one of the following: access and mobility management function AMF network element, session management function SMF network element, user plane function UPF network element, unified data management UDM network element, policy control function PCF network element, location management function LMF network element, authentication server function AUSF network element.

[0075] Optionally, the one or more communication nodes in the access network device may include at least one of the following: one or more base stations, one or more centralized units (CU), one or more distributed units (DU), one or more centralized unit control planes (CU-CP), one or more centralized unit user planes (CU-UP), and one or more access network protocol layer entities. Among them, the base station may include a next generation Node B (gNB), a base station in 6G, or a base station in 7G.

[0076] Optionally, the protocol layer entity in the access network or the protocol layer entity in the access network device includes at least one of the following: a non-access stratum (NAS) entity, a service data adaptation protocol (SDAP) entity, a radio resource control (RRC) entity, a packet data convergence protocol (PDCP) entity, a radio link control (RLC) entity, a backhaul adaptation protocol (BAP) entity, a media access control (MAC) entity, and a physical layer (PHY) entity.

[0077] For example, the management device and / or the target AI / ML entity in the management device can communicate with the AMF network element in the core network device and / or the target AI / ML entity in the AMF network element. For another example, the management device and / or the target AI / ML entity in the management device can communicate with a base station in the access network device and / or the target AI / ML entity of the base station, or can communicate with multiple base stations in the access network device and / or the target AI / ML entities of the multiple base stations respectively. For another example, the management device and / or the target AI / ML entity in the management device can communicate with the distributed unit in the access network device and / or the target AI / ML entity in the distributed unit, and / or the management device and / or the target AI / ML entity in the management device can communicate with the centralized unit user plane in the access network device and / or the target AI / ML entity in the centralized unit user plane, and / or the management device and / or the target AI / ML entity in the management device can communicate with the non-access layer entity in one or more base stations and / or the target AI / ML entity in the non-access layer entity, and / or the management device and / or the target AI / ML entity in the management device can communicate with the MAC entity in one or more base stations and / or the target AI / ML entity in the MAC entity.

[0078] Optionally, in any embodiment of the present application, communication between A and B may be direct communication between A and B via an interface protocol, or communication between A and B may be indirect communication between A and B via other devices or other entities, and the present application does not limit this. A may represent any communication node or any protocol layer entity, and B may represent any communication node or any protocol layer entity.

[0079] Optionally, in any embodiment of the present application, at least one target AI / ML entity may be an AI / ML entity of the same type, or may be an AI / ML entity of a different type.

[0080] Optionally, in any embodiment of the present application, at least one target AI / ML entity may include at least one of the following: at least one intra-node AI / ML entity (referred to as AI entity in the accompanying drawings), at least one cross-node AI / ML entity (referred to as cross-node AI coordination entity in the accompanying drawings), at least one central entity (referred to as cross-node AI coordination entity central entity in the accompanying drawings), at least one sub-entity (referred to as cross-node AI coordination entity sub-entity in the accompanying drawings), and at least one specific entity.

[0081] Optionally, in any embodiment of the present application, the AI / ML-related tasks corresponding to the management device may include at least one of the following: tasks generated by the management device, tasks requested by a target AI / ML entity deployed by a device other than the management device, and tasks interacting with a target AI / ML entity deployed by a device other than the management device.

[0082] For example, each of the target AI / ML entities in the management device can be used for at least one of the following: sending data and / or AI / ML models for AI / ML to a target AI / ML entity deployed on a device other than the management device, receiving data and / or AI / ML models for AI / ML sent by a target AI / ML entity deployed on a device other than the management device, generating data and / or AI / ML models for AI / ML, and training, reasoning, or monitoring an AI / ML model based on the data for AI / ML.

[0083] In an embodiment of the present application, the management device includes at least one target AI / ML entity; each target AI / ML entity is configured to execute an AI / ML-related task corresponding to the management device. Thus, by including at least one target AI / ML entity in the management device, the target AI / ML entity is enabled to execute the AI / ML-related task corresponding to the management device, thereby achieving a solution that combines AI or ML technology with a communication system and improving information processing efficiency through AI or ML technology.

[0084] In some embodiments, the at least one target AI / ML entity includes at least one intra-node AI / ML entity and / or at least one cross-node AI / ML entity; each cross-node AI / ML entity or each intra-node AI / ML entity is configured to process at least one of the following:

[0085] tasks requested by non-target AI / ML entities in the management device;

[0086] The tasks generated by the cross-node AI / ML entity or the intra-node AI / ML entity;

[0087] Tasks requested by other cross-node AI / ML entities deployed in the management device;

[0088] Tasks requested by AI / ML entities in other nodes deployed in the management device;

[0089] Tasks requested by the target AI / ML entity deployed on devices other than the management device.

[0090] Optionally, different intra-node AI / ML entities can process the same, different, or partially identical tasks. Optionally, different cross-node AI / ML entities can process the same, different, or partially identical tasks. Optionally, any intra-node AI / ML entity can process the same, different, or partially identical tasks as any cross-node AI / ML entity.

[0091] Optionally, each cross-node AI / ML entity or each intra-node AI / ML entity may be pre-configured for processing tasks, or configured by other AI / ML entities or devices other than itself.

[0092] Optionally, the intra-node AI / ML entity in any embodiment of the present application may be an intra-node AI / ML functional entity. Optionally, the cross-node AI / ML entity in any embodiment of the present application may be a cross-node AI / ML coordination entity.

[0093] Optionally, the management device may include a target AI / ML entity and non-target AI / ML entities other than the target AI / ML entity.

[0094] Optionally, the target AI / ML entity deployed on a device other than the management device may include at least one of the following: an intra-node AI / ML entity, a cross-node AI / ML entity, a central entity, a child entity, or a specific entity deployed on a device other than the management device. Optionally, the device other than the management device may include at least one of the following: a core network device, an access network device, or a terminal device.

[0095] In some embodiments, any two cross-node AI / ML entities deployed in the management device communicate directly through an interface protocol.

[0096] In some embodiments, the management device includes at least one central entity, and any two cross-node AI / ML entities deployed in the management device communicate indirectly through one or more of the central entities.

[0097] In some embodiments, any intra-node AI / ML entity deployed in the management device communicates directly with the cross-node AI / ML entity via an interface protocol.

[0098] In some embodiments, the management device includes at least one central entity, and any intra-node AI / ML entity deployed in the management device communicates indirectly with the cross-node AI / ML entity through one or more of the central entities.

[0099] In some embodiments, the AI / ML entities in any two nodes deployed in the management device communicate directly with each other through an interface protocol.

[0100] In some embodiments, the management device includes at least one central entity, and the AI / ML entities in any two nodes deployed in the management device communicate indirectly through one or more of the central entities.

[0101] Optionally, the central device in any embodiment of the present application may be a cross-node AI / ML coordination entity central entity.

[0102] Optionally, the connection between any intra-node AI / ML entity and the cross-node AI / ML entity may include the connection between any intra-node AI / ML entity and one or more cross-node AI / ML entities.

[0103] Optionally, in any embodiment of the present application, direct communication between any two target AI / ML entities via an interface protocol may correspond to the existence of a communication connection between the any two target AI / ML entities. Optionally, in any embodiment of the present application, indirect communication between any two target AI / ML entities may correspond to the absence of a direct communication connection between the any two target AI / ML entities.

[0104] In some embodiments, the cross-node AI / ML entity deployed in the management device communicates directly with the target AI / ML entity deployed in a device outside the management device through an interface protocol.

[0105] In some embodiments, the cross-node AI / ML entity deployed in the management device indirectly communicates with the target AI / ML entity deployed on a device other than the management device through the cross-node AI / ML entity deployed on a device other than the management device.

[0106] In some embodiments, the AI / ML entity in the node deployed in the management device directly communicates with the target AI / ML entity deployed in a device outside the management device through an interface protocol.

[0107] In some embodiments, the intra-node AI / ML entity deployed in the management device communicates indirectly with the target AI / ML entity deployed on a device outside the management device through a cross-node AI / ML entity deployed on a device outside the management device.

[0108] In some embodiments, the at least one target AI / ML entity includes: at least one central entity and at least one sub-entity, each of the central entities being associated with one or more sub-entities;

[0109] Wherein, any of the central entities is used for at least one of the following:

[0110] Processing tasks requested by non-target AI / ML entities in the management device;

[0111] Processing tasks requested by other central entities deployed in the management device;

[0112] Processing tasks requested by sub-entities deployed in the management device;

[0113] Forwarding tasks triggered or responded to by a target AI / ML entity deployed on a device other than the management device to one or more corresponding sub-entities;

[0114] Forward tasks triggered or responded to by one or more sub-entities to the corresponding target AI / ML entity deployed on a device other than the management device.

[0115] Optionally, the sub-entity in any embodiment of the present application can also be called at least one of the following: cross-node AI / ML coordination entity sub-entity, coordination sub-entity, AI / ML sub-entity, AI / ML coordination sub-entity, cross-node AI / ML coordination sub-entity, cross-node sub-entity, cross-node AI / ML sub-entity, cross-node coordination sub-entity, etc.

[0116] In some embodiments, any of the central entities is used for at least one of the following:

[0117] Forwarding a task sent by an AI / ML entity in a node deployed by a device other than the management device to one or more sub-entities in the management device associated with the AI / ML entity in the node that sent the task;

[0118] Forwarding tasks sent by a cross-node AI / ML entity deployed on a device other than the management device to one or more sub-entities in the management device associated with the cross-node AI / ML entity that sent the task;

[0119] Forwarding tasks sent by AI / ML entities in nodes deployed by devices other than the management device to one or more sub-entities associated with the task type in the management device;

[0120] Forward tasks sent by cross-node AI / ML entities deployed on devices other than the management device to one or more sub-entities associated with the task type in the management device.

[0121] Optionally, in any embodiment of the present application, different central entities may be used to process the same or different tasks. Optionally, in any embodiment of the present application, different sub-entities may be used to process the same or different tasks.

[0122] Optionally, the tasks that each central entity or each sub-entity is used to process may be pre-configured or configured by other AI / ML entities or devices other than itself.

[0123] Optionally, the management device may pre-store a correspondence between the AI / ML entity and the sub-entities within the node that sends the task, and the management device may determine one or more sub-entities associated with the AI / ML entity within the node that sends the task based on the correspondence. Optionally, the management device may determine one or more sub-entities associated with the AI / ML entity within the node that sends the task based on the identifier of the AI / ML entity within the node that sends the task. For example, the management device may use a load balancing algorithm based on the identifier of the AI / ML entity within the node that sends the task to determine one or more sub-entities associated with the AI / ML entity within the node that sends the task.

[0124] Optionally, in any embodiment of the present application, one or more sub-entities associated with a task type may include: one or more sub-entities associated with the task type of the sending task. Optionally, a sub-entity can handle tasks of one task type, and / or a sub-entity can handle tasks of multiple task types. Optionally, in any embodiment of the present application, different task types may include at least one of the following differences: different AI / ML models associated with the tasks, different data management tasks, different storage management tasks, different computing power management tasks, different model management tasks, etc.

[0125] In some embodiments, the central entity deployed in the management device directly communicates with the target AI / ML entity deployed in a device outside the management device through an interface protocol, and / or indirectly communicates with the target AI / ML entity deployed in a device outside the management device through a cross-node AI / ML entity deployed in a device outside the management device.

[0126] In some embodiments, any two central entities deployed in the management device communicate directly with each other through an interface protocol.

[0127] In some embodiments, any two central entities deployed in the management device communicate indirectly through a specific entity.

[0128] In some embodiments, a part of the central entities deployed in the management device communicate directly with each other through an interface protocol, and another part of the central entities communicate indirectly with each other through a specific entity. For example, for a central entity deployed in the management device, it can communicate directly with other central devices deployed in the management device and / or communicate indirectly through a specific entity. For example, the management device may include central entity 1, central entity 2 and central entity 3, and central device 1 and central entity 2 can communicate directly through an interface protocol, and central entity 1 and central entity 3 can communicate indirectly through a specific entity. For another example, the management device may include central entity 1, central entity 2, central entity 3 and central device 4, and central device 1 and central entity 2 can communicate directly through an interface protocol, and central entity 3 and central entity 4 can communicate indirectly through a specific entity.

[0129] Optionally, the specific entity in any embodiment of the present application may also be referred to as a task coordination center entity, a task coordination entity, a coordination entity, or a coordination center entity, etc.

[0130] Optionally, the indirect communication between any two central entities through a specific entity in any embodiment of the present application may include: indirect communication between any two central entities through one or more specific entities.

[0131] In some embodiments, any intra-node AI / ML entity deployed in the management device may or may not have a communication connection with one or more cross-node AI / ML entities deployed in the management device. For example, any intra-node AI / ML entity deployed in the management device may or may not have a communication connection with each cross-node AI / ML entity deployed in the management device. For another example, any intra-node AI / ML entity deployed in the management device may or may not have a communication connection with a part of the cross-node AI / ML entities deployed in the management device, and may not have a communication connection with another part of the cross-node AI / ML entities deployed in the management device.

[0132] In some embodiments, the AI / ML entity in any node deployed in the management device may or may not be in communication connection with one or more central entities deployed in the management device. For example, the AI / ML entity in any node deployed in the management device may or may not be in communication connection with each of the central entities deployed in the management device. For another example, the AI / ML entity in any node deployed in the management device may or may not be in communication connection with a portion of the central entities deployed in the management device, but may or may not be in communication connection with another portion of the central entities deployed in the management device.

[0133] In some embodiments, the AI / ML entity in any node deployed in the management device may or may not be in communication connection with one or more sub-entities deployed in the management device. For example, the AI / ML entity in any node deployed in the management device may or may not be in communication connection with each sub-entity deployed in the management device. For another example, the AI / ML entity in any node deployed in the management device may or may not be in communication connection with a portion of the sub-entities deployed in the management device, and may or may not be in communication connection with another portion of the sub-entities deployed in the management device.

[0134] In some embodiments, any cross-node AI / ML entity deployed in the management device may or may not be in communication with one or more central entities deployed in the management device. For example, any cross-node AI / ML entity deployed in the management device may or may not be in communication with each of the central entities deployed in the management device. For another example, any cross-node AI / ML entity deployed in the management device may or may not be in communication with a portion of the central entities deployed in the management device, and may or may not be in communication with another portion of the central entities deployed in the management device.

[0135] In some embodiments, any cross-node AI / ML entity deployed in the management device may or may not have a communication connection with one or more sub-entities deployed in the management device. For example, any cross-node AI / ML entity deployed in the management device may or may not have a communication connection with each sub-entity deployed in the management device. For another example, any cross-node AI / ML entity deployed in the management device may or may not have a communication connection with a portion of the sub-entities deployed in the management device, and may or may not have a communication connection with another portion of the sub-entities deployed in the management device.

[0136] Optionally, any two AI / ML entities that are communicatively connected in any embodiment of the present application may communicate directly through an interface protocol. Any two AI / ML entities that are not communicatively connected in any embodiment of the present application will not be able to communicate, or may communicate indirectly through other AI / ML entities.

[0137] In any embodiment of the present application, any two AI / ML entities may include any two of the following entities: one or more intra-node AI / ML entities, one or more cross-node AI / ML entities, one or more central entities, one or more sub-entities, or one or more specific entities. Any two AI / ML entities may be of the same type or different types.

[0138] In some embodiments, the intra-node AI / ML entity, cross-node AI / ML entity, central entity, sub-entity, or specific entity deployed in the management device includes at least one of the following:

[0139] A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0140] A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0141] A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling;

[0142] A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching;

[0143] The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

[0144] Optionally, the model management unit in any embodiment of the present application may also be referred to as a model lifecycle management unit.

[0145] In some embodiments, the intra-node AI / ML entity deployed in the management device is used for at least one of the following:

[0146] Sending, assigning, or forwarding a task to a first designated AI / ML entity;

[0147] Receive and process tasks sent, assigned, or forwarded by the first designated AI / ML entity;

[0148] Forwarding tasks sent, assigned, or forwarded by the first designated AI / ML entity;

[0149] Among them, the first designated AI / ML entity includes at least one of the following: other node-intra-AI / ML entities deployed in the management device, cross-node AI / ML entities deployed in the management device, central entities deployed in the management device, sub-entities deployed in the management device, node-intra-AI / ML entities deployed in devices other than the management device, cross-node AI / ML entities deployed in devices other than the management device, central entities deployed in devices other than the management device, and sub-entities deployed in devices other than the management device.

[0150] In some embodiments, the cross-node AI / ML entity deployed in the management device is used for at least one of the following:

[0151] Sending, assigning, or forwarding a task to a second designated AI / ML entity;

[0152] Receive and process tasks sent, assigned, or forwarded by the second designated AI / ML entity;

[0153] forwarding tasks sent, assigned, or forwarded by a second designated AI / ML entity;

[0154] The second designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the management device, other cross-node AI / ML entities deployed in the management device, a central entity deployed in the management device, a sub-entity deployed in the management device, an intra-node AI / ML entity deployed on a device other than the management device, a cross-node AI / ML entity deployed on a device other than the management device, a central entity deployed on a device other than the management device, and a sub-entity deployed on a device other than the management device.

[0155] In some embodiments, the central entity deployed in the management device is used for at least one of the following:

[0156] Sending, assigning, or forwarding tasks to a third designated AI / ML entity;

[0157] Receive and process tasks sent, assigned, or forwarded by a third designated AI / ML entity;

[0158] Forwarding tasks sent, assigned, or forwarded by a third designated AI / ML entity;

[0159] Among them, the third designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the management device, a cross-node AI / ML entity deployed in the management device, other central entities deployed in the management device, a sub-entity deployed in the management device, an intra-node AI / ML entity deployed on a device other than the management device, a cross-node AI / ML entity deployed on a device other than the management device, a central entity deployed on a device other than the management device, and a sub-entity deployed on a device other than the management device.

[0160] In some embodiments, the sub-entity deployed in the management device is used for at least one of the following:

[0161] Sending, assigning, or forwarding tasks to a fourth designated AI / ML entity;

[0162] Receive and process tasks sent, assigned, or forwarded by the fourth designated AI / ML entity;

[0163] forwarding tasks sent, assigned, or forwarded by a fourth designated AI / ML entity;

[0164] Among them, the fourth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the management device, a cross-node AI / ML entity deployed in the management device, a central entity deployed in the management device, other sub-entities deployed in the management device, an intra-node AI / ML entity deployed on a device other than the management device, a cross-node AI / ML entity deployed on a device other than the management device, a central entity deployed on a device other than the management device, and a sub-entity deployed on a device other than the management device.

[0165] In some embodiments, the specific entity deployed in the management device is used for at least one of the following:

[0166] Sending, assigning or forwarding tasks to a central entity deployed in the management device;

[0167] Receive and process tasks sent, assigned or forwarded by a central entity deployed in the management device;

[0168] Forwarding tasks sent, assigned or forwarded by a central entity deployed in the management device.

[0169] In any embodiment of the present application, the task includes at least one of the following:

[0170] Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0171] Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0172] Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery;

[0173] Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

[0174] In some embodiments, a newly added or newly activated target AI / ML entity in the management device sends an initialization message to at least one deployed AI / ML entity;

[0175] The initialization message includes at least one of the following of the newly joined or newly activated target AI / ML entity: communication address information, identification information, supported function scope information, and deployment location information;

[0176] The at least one deployed AI / ML entity includes: one or more target AI / ML entities deployed in the management device, and / or one or more target AI / ML entities deployed in a device other than the management device.

[0177] Optionally, in any embodiment of the present application, a newly joined or newly activated target AI / ML entity may send an initialization message to at least one deployed AI / ML entity through direct communication, or may send an initialization message to at least one deployed AI / ML entity through indirect communication.

[0178] Optionally, the initialization message in any embodiment of the present application may also be referred to as an initialization protocol message.

[0179] Optionally, the newly joined or newly activated target AI / ML entity in the management device is pre-configured with the communication address information and / or identification information of at least one deployed AI / ML entity, so that the newly joined or newly activated target AI / ML entity in the management device sends an initialization message to the at least one deployed AI / ML entity.

[0180] Optionally, the functional scope information supported in any embodiment of the present application may indicate at least one of the following: one or more tasks that can be performed by the newly added or newly activated target AI / ML entity.

[0181] Optionally, the deployment location information in any embodiment of the present application may include at least one of the following: physical location information, a location identifier, an identifier of the node where the location is located, and an identifier of the AI / ML entity connected thereto.

[0182] In some embodiments, any of the deployed AI / ML entities sends a first response message or a second response message to the newly joined or newly activated target AI / ML entity;

[0183] The first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity; and the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

[0184] Optionally, in any embodiment of the present application, for a deployed AI / ML entity, a first response message or a second response message can be sent to a newly joined or newly activated target AI / ML entity, thereby indicating to the newly joined or newly activated target AI / ML entity whether to accept the newly joined or newly activated target AI / ML entity.

[0185] Optionally, in any embodiment of the present application, the first response message may include one or more bits, and the values ​​of the one or more bits may be used to indicate the acceptance of the newly joined or newly activated target AI / ML entity; the second response message may include one or more bits, and the values ​​of the one or more bits may be used to indicate the rejection of the newly joined or newly activated target AI / ML entity.

[0186] In some embodiments, the first response message includes at least one of the following for any of the deployed AI / ML entities: communication address information, identification information, supported functional scope information, and deployment location information. The second response message includes: reason information for rejection and / or duration information for rejection.

[0187] Optionally, in any embodiment of the present application, by including at least one of the following in the first response message: communication address information, identification information, supported function scope information, and deployment location information, it can be characterized that the first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity, and by including rejection reason information and / or rejection duration information in the second response message, it can be characterized that the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

[0188] Optionally, the rejection reason information may include at least one of the following: failure to meet the functional requirements of the target AI / ML entity, and the connection capability of the target AI / ML entity reaching an upper limit.

[0189] Optionally, the rejection duration information may indicate the length of time during which further retransmission of the initialization message is prohibited. Optionally, if a newly joined or newly activated target AI / ML entity receives the rejection duration information, it may retransmit the initialization message to the deployed AI / ML entity that sent the second response message at the end of the rejection duration information.

[0190] In some embodiments, any target AI / ML entity deployed in the management device sends a first notification message to at least one fifth designated AI / ML entity; the first notification message is used to instruct the any target AI / ML entity to perform a deactivation operation or a deletion operation on itself;

[0191] When any one of the target AI / ML entities receives a confirmation message sent by each of the fifth designated AI / ML entities, the any one of the target AI / ML entities performs a deactivation operation or a deletion operation on itself;

[0192] The fifth designated AI / ML entity includes: other target AI / ML entities deployed in the management device that are communicatively connected to any target AI / ML entity, and / or target AI / ML entities deployed in a device other than the management device that are communicatively connected to any target AI / ML entity.

[0193] In this case, any target AI / ML entity deployed in the management device proactively initiates a deactivation operation or deletion operation on its own, thereby achieving its own deactivation or deletion.

[0194] Optionally, when any one or more of the at least one fifth designated AI / ML entities sends at least one of a rejection message, a denial message, a non-confirmation message, and a rejection reason to any target AI / ML entity, any target AI / ML entity will not perform its own deactivation operation or deletion operation.

[0195] In some embodiments, any target AI / ML entity deployed in the management device receives a second notification message sent by a fifth designated AI / ML entity; the second notification message is used to instruct the any target AI / ML entity to be deactivated or deleted;

[0196] Any target AI / ML entity performs a deactivation operation or a deletion operation on itself based on the second notification message;

[0197] The fifth designated AI / ML entity includes: other target AI / ML entities deployed in the management device that are in communication connection with any target AI / ML entity, and / or target AI / ML entities deployed in a device other than the management device that are in communication connection with any target AI / ML entity.

[0198] In this case, any target AI / ML entity deployed in the management device passively deactivates or deletes itself.

[0199] Optionally, any target AI / ML entity deployed in the management device can determine whether it is executing a task and / or whether there is a task to be executed upon receiving the second notification message. If both are negative, the any target AI / ML entity performs its own deactivation operation or deletion operation based on the second notification message.

[0200] In some embodiments, the management device may send capability information to a communication node deployed by a device other than the management device, or the management device may receive capability information sent by a communication node deployed by a device other than the management device;

[0201] The capability information indicates at least one of the following:

[0202] Whether to support deployment of AI / ML entities within the node;

[0203] Whether to support deployment of cross-node AI / ML entities;

[0204] The maximum number of AI / ML entities supported within a node;

[0205] The maximum number of supported cross-node AI / ML entities;

[0206] The maximum number of sub-entities supported;

[0207] Target units and / or protocol layer entities in communication nodes where AI / ML entities within the nodes can be deployed;

[0208] Target units and / or protocol layer entities in communication nodes where AI / ML entities within the node cannot be deployed;

[0209] Communication nodes and / or management devices where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously;

[0210] Communication nodes and / or management devices where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously;

[0211] Among them, the communication nodes deployed in devices other than the management device include at least one of the following: communication nodes in core network devices, communication nodes in access network devices, target units in access network devices, protocol layer entities in access network devices, terminal devices, and protocol layer entities in terminal devices.

[0212] Optionally, in any embodiment of the present application, unless otherwise specified, the communication node may be a node in a core network device or a node in an access network device, and the specific one may be determined by referring to the semantics of the context.

[0213] The following describes the structure of the core network equipment:

[0214] Figure 3 is a structural diagram of a core network device provided in an embodiment of the present application. As shown in Figure 3, the core network device includes at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the core network device.

[0215] Optionally, the core network device or one or more communication nodes in the core network device can communicate with the management device. Optionally, the core network device or one or more communication nodes in the core network device can communicate with the access network device. Optionally, the core network device or one or more communication nodes in the core network device can communicate with the terminal device through the access network device.

[0216] For example, the communication node in the core network device and / or the target AI / ML entity in the core network device can communicate with the management device or the target AI / ML entity in the management device. Optionally, the communication node in the core network device and / or the target AI / ML entity in the core network device can communicate with the communication node in the access network device and / or the target AI / ML entity in the access network device. Optionally, the communication node in the core network device and / or the target AI / ML entity in the core network device can communicate with the terminal device and / or the target AI / ML entity in the terminal device through the communication node in the access network device and / or the target AI / ML entity in the access network device.

[0217] Optionally, in any embodiment of the present application, the AI / ML-related tasks corresponding to the core network device may include at least one of the following: tasks generated by a communication node in the core network device, tasks requested by a target AI / ML entity deployed on a device other than the core network device, and tasks interacting with a target AI / ML entity deployed on a device other than the core network device.

[0218] For example, each of the target AI / ML entities in the core network device can be used for at least one of the following: sending data and / or AI / ML models for AI / ML to a target AI / ML entity deployed on a device outside the core network device, receiving data and / or AI / ML models for AI / ML sent by a target AI / ML entity deployed on a device outside the core network device, generating data and / or AI / ML models for AI / ML, and training, reasoning, or monitoring the AI / ML model based on the data for AI / ML.

[0219] In some embodiments, the target AI / ML entity deployed in the core network device includes an intra-node AI / ML entity; the core network device further includes at least one communication node;

[0220] One or more intra-node AI / ML entities are deployed in each of the communication nodes;

[0221] The AI / ML entity in each node is used to perform AI / ML related tasks of the corresponding communication node.

[0222] Optionally, in any embodiment of the present application, for a communication node, a non-target AI / ML entity of the communication node may generate a task and send it to a target AI / ML entity of the communication node, so that the target AI / ML entity of the communication node processes the task.

[0223] Optionally, the tasks processed by the AI / ML entities in different nodes may be the same, different, or partially the same. Optionally, the tasks processed by the AI / ML entities in different nodes in the same communication node may be the same, different, or partially the same.

[0224] Optionally, the intra-node AI / ML entity in communication node A is at least used to execute AI / ML related tasks of communication node A, and the intra-node AI / ML entity in communication node B is at least used to execute AI / ML related tasks of communication node B.

[0225] In some embodiments, the target AI / ML entity deployed in the core network device includes a cross-node AI / ML entity; the core network device further includes at least one communication node;

[0226] Each of the cross-node AI / ML entities is communicatively connected to one or more of the communication nodes;

[0227] Each of the cross-node AI / ML entities is configured to execute AI / ML-related tasks corresponding to one or more of the communication nodes.

[0228] Optionally, in any embodiment of the present application, for a communication node, the communication node may generate a task and send the task to a corresponding cross-node AI / ML entity so that the corresponding cross-node AI / ML entity processes the task.

[0229] Optionally, in any embodiment of the present application, the tasks used for processing by different cross-node AI / ML entities may be the same or different or partially the same. Optionally, the communication nodes to which different cross-node AI / ML entities are communicated may be the same or different or partially the same. Exemplarily, cross-node AI / ML entity A is communicated with communication node A and communication node B respectively, and cross-node AI / ML entity B is communicated with communication node A and communication node B respectively. Again exemplarily, cross-node AI / ML entity A is communicated with communication node A and communication node B respectively, and cross-node AI / ML entity B is communicated with communication node A and communication node C respectively. Again exemplarily, cross-node AI / ML entity A is communicated with communication node A and communication node B respectively, and cross-node AI / ML entity B is communicated with communication node C and communication node D respectively.

[0230] Optionally, each cross-node AI / ML entity is configured to execute an AI / ML-related task corresponding to a communication node with which each cross-node AI / ML entity has a communication connection.

[0231] In some embodiments, the core network device further comprises at least one communication node; the at least one target AI / ML entity comprises: at least one intra-node AI / ML entity and at least one cross-node AI / ML entity;

[0232] Any of the communication nodes includes one or more intra-node AI / ML entities; each intra-node AI / ML entity is used to perform AI / ML related tasks of the corresponding communication node;

[0233] Each of the cross-node AI / ML entities is in communication with one or more of the intra-node AI / ML entities; each of the cross-node AI / ML entities is configured to perform AI / ML-related tasks corresponding to one or more of the communication nodes.

[0234] Optionally, the AI / ML entity within each node is used to execute the AI / ML related tasks of the corresponding communication node, and / or, the AI / ML entity within each node is used to execute tasks sent by the cross-node AI / ML entity.

[0235] In some embodiments, intra-node AI / ML entities deployed in any two communication nodes in the core network device communicate directly via an interface protocol.

[0236] In some embodiments, intra-node AI / ML entities deployed in any two communication nodes in the core network device communicate indirectly through one or more cross-node AI / ML entities.

[0237] In some embodiments, the intra-node AI / ML entities deployed in some of the communication nodes in the core network device communicate directly with each other via an interface protocol, and the intra-node AI / ML entities deployed in another part of the communication nodes communicate indirectly with each other via one or more cross-node AI / ML entities. For example, the intra-node AI / ML entity in communication node A and the intra-node AI / ML entity in communication node B can communicate directly via an interface protocol, and the intra-node AI / ML entity in communication node A and the intra-node AI / ML entity in communication node C can communicate indirectly via one or more cross-node AI / ML entities.

[0238] Optionally, for a core network device, when multiple intra-node AI / ML entities are deployed in a communication node, any two of the multiple intra-node AI / ML entities can communicate directly through an interface protocol. Optionally, when intra-node AI / ML entities are deployed in different communication nodes, the intra-node AI / ML entities belonging to different communication nodes can communicate directly through an interface protocol, or can communicate indirectly through one or more cross-node AI / ML entities.

[0239] In some embodiments, any two cross-node AI / ML entities in the core network device communicate directly with each other via an interface protocol.

[0240] In some embodiments, the core network device includes at least one central entity, and any two cross-node AI / ML entities deployed in the core network device communicate indirectly through one or more of the central entities.

[0241] In some embodiments, the core network device includes at least one central entity, wherein a portion of the cross-node AI / ML entities in the core network device communicate directly with each other via an interface protocol, while another portion of the cross-node AI / ML entities communicate indirectly with each other via one or more of the central entities. For example, cross-node AI / ML entity A and cross-node AI / ML entity B can communicate directly via an interface protocol, while cross-node AI / ML entity A and cross-node AI / ML entity C can communicate indirectly via one or more of the central entities.

[0242] In some embodiments, the cross-node AI / ML entity deployed in the core network device directly communicates with the target AI / ML entity deployed in a device outside the core network device through an interface protocol.

[0243] In some embodiments, the cross-node AI / ML entity deployed in the core network device indirectly communicates with the target AI / ML entity deployed in the device outside the core network device through the cross-node AI / ML entity deployed in the device outside the core network device.

[0244] In some embodiments, the intra-node AI / ML entity deployed in the core network device directly communicates with the target AI / ML entity deployed in a device outside the core network device through an interface protocol.

[0245] In some embodiments, the intra-node AI / ML entity deployed in the core network device indirectly communicates with the target AI / ML entity deployed on a device outside the core network device through the cross-node AI / ML entity deployed in the core network device and / or the cross-node AI / ML entity deployed on a device outside the core network device.

[0246] Optionally, devices other than the core network device in any embodiment of the present application may include at least one of the following: a management device, an access network device, and a terminal device.

[0247] Optionally, the target AI entity deployed in a device other than the core network device in any embodiment of the present application includes at least one of the following: an intra-node AI entity, a cross-node AI entity, a central entity, a sub-entity, and a specific entity.

[0248] In some embodiments, each of the cross-node AI / ML entities or each of the intra-node AI / ML entities is configured to process at least one of the following:

[0249] a task requested by at least one communication node included in the core network device;

[0250] The tasks generated by the cross-node AI / ML entity or the intra-node AI / ML entity;

[0251] Tasks requested by other cross-node AI / ML entities deployed in the core network devices;

[0252] Tasks requested by AI / ML entities in other nodes deployed in the core network device;

[0253] Tasks requested by the target AI / ML entity deployed on devices other than the core network device.

[0254] In some embodiments, the at least one target AI / ML entity includes: at least one central entity and at least one sub-entity, each of the central entities being associated with one or more sub-entities;

[0255] Wherein, any of the central entities is used for at least one of the following:

[0256] Processing a task requested by at least one communication node included in the core network device;

[0257] Processing tasks requested by other central entities deployed in the core network equipment;

[0258] Processing tasks requested by sub-entities deployed in the core network device;

[0259] Forwarding tasks triggered or responded to by an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the core network device to one or more sub-entities deployed in the core network device, or to a target AI / ML entity deployed in a device outside the core network device;

[0260] Forward tasks triggered or responded to by one or more sub-entities deployed in the core network device, or by a target AI / ML entity deployed on a device outside the core network device, to an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the core network device.

[0261] In some embodiments, any of the central entities is used for at least one of the following:

[0262] Forwarding a task triggered or responded to by an intra-node AI / ML entity deployed in the core network device to one or more sub-entities in the core network device associated with the intra-node AI / ML entity that triggered or responded to the task;

[0263] Forwarding a task triggered or responded to by a cross-node AI / ML entity deployed in the core network device to one or more sub-entities in the core network device associated with the cross-node AI / ML entity that triggered or responded to the task;

[0264] Forwarding tasks triggered or responded to by AI / ML entities in nodes deployed in the core network device to one or more sub-entities associated with the task type in the core network device;

[0265] Forwarding tasks triggered or responded to by the cross-node AI / ML entity deployed in the core network device to one or more sub-entities associated with the task type in the core network device;

[0266] Forwarding a task sent by an AI / ML entity in a node deployed on a device other than the core network device to one or more sub-entities in the core network device associated with the AI / ML entity in the node that sent the task;

[0267] Forwarding a task sent by a cross-node AI / ML entity deployed on a device other than the core network device to one or more sub-entities in the core network device associated with the cross-node AI / ML entity that sent the task;

[0268] Forwarding tasks sent by the AI / ML entity in the node deployed by the device outside the core network device to one or more sub-entities associated with the task type in the core network device;

[0269] Forwarding tasks sent by a cross-node AI / ML entity deployed on a device other than the core network device to one or more sub-entities associated with the task type in the core network device.

[0270] Optionally, the core network device may pre-store a correspondence between the AI / ML entity and the sub-entities within the node of the triggering or responding task, and the core network device may determine one or more sub-entities associated with the AI / ML entity within the node of the triggering or responding task based on the correspondence. Optionally, the core network device may determine one or more sub-entities associated with the AI / ML entity within the node of the triggering or responding task based on the identifier of the AI / ML entity within the node of the triggering or responding task. For example, the core network device may use a load balancing algorithm based on the identifier of the AI / ML entity within the node of the triggering or responding task to determine one or more sub-entities associated with the AI / ML entity within the node of the triggering or responding task.

[0271] Optionally, in any embodiment of the present application, one or more sub-entities associated with a task type may include: one or more sub-entities associated with a task type that triggers or responds to a task. Optionally, a sub-entity may process tasks of one task type, and / or a sub-entity may process tasks of multiple task types. Optionally, different task types may include at least one of the following differences: different AI / ML models associated with the tasks, different data management tasks, different storage management tasks, different computing power management tasks, different model management tasks, etc.

[0272] In some embodiments, the central entity deployed in the core network device directly communicates with the target AI / ML entity deployed on a device outside the core network device through an interface protocol, and / or indirectly communicates with the target AI / ML entity deployed on a device outside the core network device through a cross-node AI / ML entity deployed on a device outside the core network device.

[0273] In some embodiments, any two central entities deployed in the core network device communicate directly with each other through an interface protocol.

[0274] In some embodiments, any two central entities deployed in the core network device communicate indirectly through a specific entity.

[0275] In some embodiments, a part of the central entities deployed in the core network device communicate directly with each other through an interface protocol, and another part of the central entities communicate indirectly with each other through a specific entity. Exemplarily, for a central entity deployed in the core network device, it can communicate directly with other central devices deployed in the core network device and / or communicate indirectly through a specific entity. For example, the core network device may include central entity 1, central entity 2 and central entity 3, and central device 1 and central entity 2 can communicate directly through an interface protocol, and central entity 1 and central entity 3 can communicate indirectly through a specific entity. For another example, the core network device may include central entity 1, central entity 2, central entity 3 and central device 4, and central device 1 and central entity 2 can communicate directly through an interface protocol, and central entity 3 and central entity 4 can communicate indirectly through a specific entity.

[0276] Optionally, any intra-node AI / ML entity deployed in the core network device has or does not have a communication connection with one or more cross-node AI / ML entities deployed in the core network device. Optionally, any intra-node AI / ML entity deployed in the core network device has or does not have a communication connection with one or more central entities deployed in the core network device. Optionally, any intra-node AI / ML entity deployed in the core network device has or does not have a communication connection with one or more sub-entities deployed in the core network device.

[0277] Optionally, any cross-node AI / ML entity deployed in the core network device has a communication connection or no communication connection with one or more central entities deployed in the core network device. Optionally, any cross-node AI / ML entity deployed in the core network device has a communication connection or no communication connection with one or more sub-entities deployed in the core network device.

[0278] In some embodiments, the intra-node AI / ML entity, cross-node AI / ML entity, central entity, sub-entity, or specific entity deployed in the core network device includes at least one of the following:

[0279] A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0280] A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0281] A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling;

[0282] A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching;

[0283] The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

[0284] In some embodiments, the intra-node AI / ML entity deployed in the core network device is used for at least one of the following:

[0285] Sending, assigning, or forwarding tasks to a sixth designated AI / ML entity;

[0286] Receive and process tasks sent, assigned, or forwarded by the sixth designated AI / ML entity;

[0287] forwarding tasks sent, assigned, or forwarded by the sixth designated AI / ML entity;

[0288] Among them, the sixth designated AI / ML entity includes at least one of the following: other node AI / ML entities deployed in the core network device, cross-node AI / ML entities deployed in the core network device, central entities deployed in the core network device, sub-entities deployed in the core network device, node AI / ML entities deployed by devices other than the core network device, cross-node AI / ML entities deployed by devices other than the core network device, central entities deployed by devices other than the core network device, and sub-entities deployed by devices other than the core network device.

[0289] In some embodiments, the cross-node AI / ML entity deployed in the core network device is used for at least one of the following:

[0290] Sending, assigning, or forwarding tasks to a seventh designated AI / ML entity;

[0291] Receive and process tasks sent, assigned, or forwarded by the seventh designated AI / ML entity;

[0292] forwarding tasks sent, assigned, or forwarded by the seventh designated AI / ML entity;

[0293] Among them, the seventh designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the core network device, other cross-node AI / ML entities deployed in the core network device, a central entity deployed in the core network device, a sub-entity deployed in the core network device, an intra-node AI / ML entity deployed by a device other than the core network device, a cross-node AI / ML entity deployed by a device other than the core network device, a central entity deployed by a device other than the core network device, and a sub-entity deployed by a device other than the core network device.

[0294] In some embodiments, the central entity deployed in the core network device is used for at least one of the following:

[0295] Sending, assigning, or forwarding tasks to an eighth designated AI / ML entity;

[0296] Receive and process tasks sent, assigned, or forwarded by the eighth designated AI / ML entity;

[0297] forwarding tasks sent, assigned, or forwarded by the eighth designated AI / ML entity;

[0298] Among them, the eighth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the core network device, a cross-node AI / ML entity deployed in the core network device, other central entities deployed in the core network device, a sub-entity deployed in the core network device, an intra-node AI / ML entity deployed on a device other than the core network device, a cross-node AI / ML entity deployed on a device other than the core network device, a central entity deployed on a device other than the core network device, and a sub-entity deployed on a device other than the core network device.

[0299] In some embodiments, the sub-entity deployed in the core network device is used for at least one of the following:

[0300] Sending, assigning, or forwarding tasks to a ninth designated AI / ML entity;

[0301] Receive and process tasks sent, assigned, or forwarded by the ninth designated AI / ML entity;

[0302] Forwarding tasks sent, assigned, or forwarded by the ninth designated AI / ML entity;

[0303] Among them, the ninth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the core network device, a cross-node AI / ML entity deployed in the core network device, a central entity deployed in the core network device, other sub-entities deployed in the core network device, an intra-node AI / ML entity deployed by a device other than the core network device, a cross-node AI / ML entity deployed by a device other than the core network device, a central entity deployed by a device other than the core network device, and a sub-entity deployed by a device other than the core network device.

[0304] In some embodiments, the specific entity deployed in the core network device is used for at least one of the following:

[0305] Sending tasks, assigning tasks, or forwarding tasks to a central entity deployed in the core network device;

[0306] Receive and process tasks sent, assigned, or forwarded by a central entity deployed in the core network device;

[0307] Forwarding tasks sent, assigned or forwarded by a central entity deployed in the core network device.

[0308] In some embodiments, the task includes at least one of the following:

[0309] Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0310] Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0311] Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery;

[0312] Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

[0313] In some embodiments, a newly joined or newly activated target AI / ML entity in the core network device sends an initialization message to at least one deployed AI / ML entity;

[0314] The initialization message includes at least one of the following of the newly joined or newly activated target AI / ML entity: communication address information, identification information, supported function scope information, and deployment location information;

[0315] The at least one deployed AI / ML entity includes: one or more target AI / ML entities deployed in the core network device, and / or one or more target AI / ML entities deployed in a device other than the core network device.

[0316] Optionally, the newly added or newly activated target AI / ML entity in the core network device is pre-configured with the communication address information and / or identification information of at least one deployed AI / ML entity, so that the newly added or newly activated target AI / ML entity in the core network device sends an initialization message to at least one deployed AI / ML entity.

[0317] In some embodiments, any of the deployed AI / ML entities sends a first response message or a second response message to the newly joined or newly activated target AI / ML entity;

[0318] The first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity; and the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

[0319] In some embodiments, the first response message includes at least one of the following of any of the deployed AI / ML entities: communication address information, identification information, supported function scope information, and deployment location information;

[0320] The second response message includes: rejection reason information and / or rejection duration information.

[0321] In some embodiments, any target AI / ML entity deployed in the core network device sends a first notification message to at least one tenth designated AI / ML entity; the first notification message is used to instruct the any target AI / ML entity to perform a deactivation operation or a deletion operation on itself;

[0322] When any one of the target AI / ML entities receives a confirmation message sent by each of the tenth designated AI / ML entities, the any one of the target AI / ML entities performs a deactivation operation or a deletion operation on itself;

[0323] Among them, the tenth designated AI / ML entity includes: other target AI / ML entities deployed in the core network device that have a communication connection with any target AI / ML entity, and / or, target AI / ML entities deployed in devices other than the core network device that have a communication connection with any target AI / ML entity.

[0324] In this case, any target AI / ML entity deployed in the core network device actively initiates its own deactivation operation or deletion operation, thereby achieving its own deactivation or deletion.

[0325] Optionally, when any one or more of the at least one tenth designated AI / ML entity sends at least one of a rejection message, a denial message, a non-confirmation message, and a rejection reason to any target AI / ML entity, any target AI / ML entity will not perform its own deactivation operation or deletion operation.

[0326] In some embodiments, any target AI / ML entity deployed in the core network device receives a second notification message sent by the tenth designated AI / ML entity; the second notification message is used to instruct the any target AI / ML entity to be deactivated or deleted;

[0327] Any target AI / ML entity performs a deactivation operation or a deletion operation on itself based on the second notification message;

[0328] Among them, the tenth designated AI / ML entity includes: other target AI / ML entities deployed in the core network device that have a communication connection with any target AI / ML entity, and / or, target AI / ML entities deployed in devices other than the core network device that have a communication connection with any target AI / ML entity.

[0329] In this case, any target AI / ML entity deployed in the core network device passively deactivates or deletes itself.

[0330] Optionally, any target AI / ML entity deployed in the core network device can determine whether it is executing a task and / or whether there is a task to be executed upon receiving the second notification message. If both are negative, the any target AI / ML entity performs its own deactivation operation or deletion operation based on the second notification message.

[0331] Optionally, the core network device sends first indication information to the terminal device, where the first indication information is used to indicate at least one of the following:

[0332] Instructing the terminal device to newly deploy at least one intra-node AI / ML entity; instructing the terminal device to delete at least one intra-node AI / ML entity; instructing the terminal device to activate at least one intra-node AI / ML entity; instructing the terminal device to deactivate at least one intra-node AI / ML entity; instructing the terminal device to modify at least one intra-node AI / ML entity.

[0333] Optionally, the core network device receives second indication information sent by the terminal device, where the second indication information is used for at least one of the following:

[0334] Indicates at least one intra-node AI / ML entity that the terminal device expects to add; indicates at least one intra-node AI / ML entity that the terminal device expects to delete; indicates at least one intra-node AI / ML entity that the terminal device expects to activate; indicates at least one intra-node AI / ML entity that the terminal device expects to deactivate; indicates at least one intra-node AI / ML entity that the terminal device has added; indicates at least one intra-node AI / ML entity that the terminal device has deleted; indicates at least one intra-node AI / ML entity that the terminal device has activated; indicates at least one intra-node AI / ML entity that the terminal device has deactivated; indicates at least one intra-node AI / ML entity that the terminal device has modified.

[0335] In some embodiments, the communication node deployed in the core network device may send capability information to other communication nodes deployed in the core network device, or the communication node deployed in the core network device may send capability information to a communication node deployed in a device other than the core network device, or the communication node deployed in the core network device may receive capability information sent by a communication node deployed in a device other than the core network device;

[0336] The capability information indicates at least one of the following:

[0337] Whether to support deployment of AI / ML entities within the node;

[0338] Whether to support deployment of cross-node AI / ML entities;

[0339] The maximum number of AI / ML entities supported within a node;

[0340] The maximum number of supported cross-node AI / ML entities;

[0341] The maximum number of sub-entities supported;

[0342] Target units and / or protocol layer entities in communication nodes where AI / ML entities within the nodes can be deployed;

[0343] Target units and / or protocol layer entities in communication nodes where AI / ML entities within the node cannot be deployed;

[0344] Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously;

[0345] Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously;

[0346] The communication nodes deployed in devices other than the core network device include at least one of the following: a management device, a communication node in an access network device, a target unit in an access network device, a protocol layer entity in an access network device, a terminal device, and a protocol layer entity in a terminal device;

[0347] The communication nodes deployed in the core network device include at least one of the following: access and mobility management function AMF network element, session management function SMF network element, user plane function UPF network element, unified data management UDM network element, policy control function PCF network element, location management function LMF network element, and authentication server function AUSF network element.

[0348] The following describes the structure of the access network equipment:

[0349] Figure 4 is a structural diagram of an access network device provided in an embodiment of the present application. As shown in Figure 4, the access network device includes at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the access network device.

[0350] Optionally, the access network device or one or more communication nodes in the access network device can communicate with the management device. Optionally, the access network device or one or more communication nodes in the access network device can communicate with the core network device. Optionally, the access network device or one or more communication nodes in the access network device can communicate with the terminal device.

[0351] Optionally, the communication node in the access network device and / or the target AI / ML entity in the access network device can communicate with the management device or the target AI / ML entity in the management device. Optionally, the communication node in the access network device and / or the target AI / ML entity in the access network device can communicate with the communication node in the core network device and / or the target AI / ML entity in the core network device. Optionally, the communication node in the access network device and / or the target AI / ML entity in the access network device can communicate with the terminal device and / or the target AI / ML entity in the terminal device.

[0352] Optionally, in any embodiment of the present application, the AI / ML-related tasks corresponding to the access network device may include at least one of the following: tasks generated by a communication node in the access network device, tasks requested by a target AI / ML entity deployed on a device other than the access network device, and tasks interacting with a target AI / ML entity deployed on a device other than the access network device.

[0353] For example, each of the target AI / ML entities in the access network device can be used for at least one of the following: sending data and / or AI / ML models for AI / ML to a target AI / ML entity deployed on a device other than the access network device, receiving data and / or AI / ML models for AI / ML sent by a target AI / ML entity deployed on a device other than the access network device, generating data and / or AI / ML models for AI / ML, and training, reasoning, or monitoring the AI / ML model based on the data for AI / ML.

[0354] In some embodiments, the target AI / ML entity includes the intra-node AI / ML entity; the access network device further includes at least one communication node;

[0355] The intra-node AI / ML entity includes at least one of the following: an access network AI / ML entity, a target unit AI / ML entity, and a protocol layer AI / ML entity; the target unit includes at least one of the following: a centralized unit CU, a distributed unit DU, a centralized unit control plane CU-CP, and a centralized unit user plane CU-UP;

[0356] One or more access network AI / ML entities are deployed in the access network device, and / or one or more target unit AI / ML entities are deployed in the target unit in the communication node, and / or one or more protocol layer AI / ML entities are deployed in the protocol layer entity in the communication node;

[0357] Among them, the access network AI / ML entity is used to perform AI / ML related tasks corresponding to the access network device, the target unit AI / ML entity is used to perform AI / ML related tasks corresponding to the target unit, and the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer.

[0358] Optionally, taking the communication node in the access network device as a base station as an example, the base station may include a target unit and a non-target unit, the access network AI / ML entity may be deployed in the non-target unit, the target unit may include a protocol layer entity and a non-protocol layer entity, the target unit AI / ML entity may be deployed in the non-protocol layer entity, and the protocol layer AI / ML entity may be deployed in the protocol layer entity.

[0359] Optionally, for any base station (also called base station equipment), a base station can be deployed with at least one of the following: one or more access network AI / ML entities, one or more target unit AI / ML entities, and one or more protocol layer AI / ML entities.

[0360] Optionally, the access network AI / ML entity is used to perform AI / ML-related tasks corresponding to the base station to which it belongs. Optionally, the target unit AI / ML entity is used to perform AI / ML-related tasks corresponding to the target unit to which it belongs. Optionally, the protocol layer AI / ML entity is used to perform AI / ML-related tasks corresponding to the protocol layer entity to which it belongs.

[0361] Optionally, the access network AI / ML entities in different base stations may be used to process the same, different, or partially identical tasks. Optionally, the target unit AI / ML entities in different base stations may be used to process the same, different, or partially identical tasks. Optionally, the protocol layer AI / ML entities in different base stations may be used to process the same, different, or partially identical tasks.

[0362] Optionally, the tasks processed by any two of the one or more access network AI / ML entities, the one or more target unit AI / ML entities, and the one or more protocol layer AI / ML entities may be the same, different, or partially the same.

[0363] In some embodiments, the target AI / ML entity includes a cross-node AI / ML entity; the access network device further includes at least one communication node;

[0364] Each of the cross-node AI / ML entities is communicatively connected to one or more of the communication nodes;

[0365] Each of the cross-node AI / ML entities is configured to execute AI / ML-related tasks corresponding to one or more of the communication nodes.

[0366] In some embodiments, the access network device further comprises at least one communication node; the at least one target AI / ML entity comprises: at least one intra-node AI / ML entity and at least one cross-node AI / ML entity;

[0367] The intra-node AI / ML entity includes at least one of the following: an access network AI / ML entity, a target unit AI / ML entity, and a protocol layer AI / ML entity; the target unit includes at least one of the following: a centralized unit CU, a distributed unit DU, a centralized unit control plane CU-CP, and a centralized unit user plane CU-UP;

[0368] One or more of the access network AI / ML entities are deployed in the access network device, and / or one or more of the target unit AI / ML entities are deployed in the target unit in the access network device, and / or one or more of the protocol layer AI / ML entities are deployed in the protocol layer entity in the access network device;

[0369] The access network AI / ML entity is used to perform AI / ML related tasks corresponding to the access network device, the target unit AI / ML entity is used to perform AI / ML related tasks corresponding to the target unit, and the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer;

[0370] Each of the cross-node AI / ML entities is in communication with one or more of the intra-node AI / ML entities, and each of the cross-node AI / ML entities is configured to perform AI / ML-related tasks corresponding to one or more of the communication nodes.

[0371] Optionally, the access network AI / ML entity and / or the target unit AI / ML entity and / or the protocol layer AI / ML entity may also be used to execute tasks sent by cross-node AI / ML entities.

[0372] In some embodiments, the intra-node AI / ML entities deployed in any two of the communication nodes in the access network device communicate directly via an interface protocol.

[0373] In some embodiments, intra-node AI / ML entities deployed in any two communication nodes in the access network device communicate indirectly through one or more cross-node AI / ML entities.

[0374] In some embodiments, the intra-node AI / ML entities deployed in some of the communication nodes in the access network device communicate directly with each other, while the intra-node AI / ML entities deployed in another portion of the communication nodes communicate indirectly through one or more cross-node AI / ML entities. For example, the intra-node AI / ML entity in communication node A and the intra-node AI / ML entity in communication node B can communicate directly through an interface protocol, and the intra-node AI / ML entity in communication node A and the intra-node AI / ML entity in communication node C can communicate indirectly through one or more cross-node AI / ML entities.

[0375] Optionally, for an access network device, when multiple intra-node AI / ML entities are deployed in a communication node, any two of the multiple intra-node AI / ML entities may communicate directly via an interface protocol, or may communicate indirectly via other intra-node AI / ML entities. Optionally, when the intra-node AI / ML entities are deployed in different communication nodes, the intra-node AI / ML entities belonging to different communication nodes may communicate directly via an interface protocol, or may communicate indirectly via one or more cross-node AI / ML entities.

[0376] In some embodiments, any two target unit AI / ML entities deployed in the access network device communicate with each other via one or more of the access network AI / ML entities. For example, any two target unit AI / ML entities in a base station communicate with each other via one or more of the access network AI / ML entities. Alternatively, in other embodiments, any two target unit AI / ML entities in a base station communicate directly with each other via an interface protocol.

[0377] In some embodiments, any two protocol layer AI / ML entities communicate with each other through one or more access network AI / ML entities. Alternatively, in other embodiments, any two protocol layer AI / ML entities in a base station communicate directly with each other through an interface protocol.

[0378] In some embodiments, any two protocol layer AI / ML entities communicate with each other through one or more target unit AI / ML entities. Alternatively, in other embodiments, any two protocol layer AI / ML entities in a base station communicate directly with each other through an interface protocol.

[0379] Exemplarily, any two protocol layer AI / ML entities in a base station may communicate through one or more of the access network AI / ML entities, and / or may communicate through one or more of the target unit AI / ML entities.

[0380] In some embodiments, any two cross-node AI / ML entities deployed in the access network device communicate directly with each other via an interface protocol.

[0381] In some embodiments, the access network device includes at least one central entity, and any two cross-node AI / ML entities deployed in the access network device communicate indirectly through one or more of the central entities.

[0382] In some embodiments, the access network device includes at least one central entity, wherein a portion of the cross-node AI / ML entities deployed in the access network device communicate directly with each other via an interface protocol, and another portion of the cross-node AI / ML entities deployed communicate indirectly with each other via one or more of the central entities. For example, cross-node AI / ML entity A and cross-node AI / ML entity B can communicate directly via an interface protocol, while cross-node AI / ML entity A and cross-node AI / ML entity C can communicate indirectly via one or more of the central entities.

[0383] In some embodiments, the cross-node AI / ML entity in the access network device directly communicates with the target AI / ML entity deployed in a device outside the access network device through an interface protocol.

[0384] In some embodiments, the cross-node AI / ML entity deployed in the access network device indirectly communicates with the target AI / ML entity deployed in a device outside the access network device through the cross-node AI / ML entity deployed in a device outside the access network device.

[0385] In some embodiments, the intra-node AI / ML entity deployed in the access network device directly communicates with the target AI / ML entity deployed in a device outside the access network device via an interface protocol.

[0386] In some embodiments, the intra-node AI / ML entity deployed in the access network device indirectly communicates with the target AI / ML entity deployed in a device outside the access network device through the cross-node AI / ML entity deployed in the access network device and / or the cross-node AI / ML entity deployed in a device outside the access network device.

[0387] Optionally, devices other than the access network device in any embodiment of the present application may include at least one of the following: a management device, a core network device, and a terminal device.

[0388] Optionally, the target AI entity deployed in a device other than the access network device in any embodiment of the present application includes at least one of the following: an intra-node AI entity, a cross-node AI entity, a central entity, a sub-entity, and a specific entity.

[0389] In some embodiments, each of the cross-node AI / ML entities or each of the intra-node AI / ML entities is configured to process at least one of the following:

[0390] a task requested by at least one communication node included in the access network device;

[0391] The tasks generated by the cross-node AI / ML entity or the intra-node AI / ML entity;

[0392] Tasks requested by other cross-node AI / ML entities deployed in the access network device;

[0393] Tasks requested by AI / ML entities in other nodes deployed in the access network device;

[0394] Tasks requested by the target AI / ML entity deployed on devices other than the access network device.

[0395] In some embodiments, the at least one target AI / ML entity includes: at least one central entity and at least one sub-entity, each of the central entities being associated with one or more sub-entities;

[0396] Wherein, any of the central entities is used for at least one of the following:

[0397] Processing a task requested by at least one communication node included in the access network device;

[0398] Processing tasks requested by other central entities deployed in the access network equipment;

[0399] Processing tasks requested by sub-entities deployed in the access network device;

[0400] Forwarding tasks triggered or responded to by an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the access network device to one or more sub-entities deployed in the access network device, or to a target AI / ML entity deployed in a device outside the access network device;

[0401] Forwarding tasks triggered or responded to by one or more sub-entities deployed in the access network device, or a target AI / ML entity deployed on a device outside the access network device, to an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the access network device.

[0402] In some embodiments, any of the central entities is used for at least one of the following:

[0403] Forwarding a task triggered or responded to by an intra-node AI / ML entity deployed in the access network device to one or more sub-entities in the access network device associated with the intra-node AI / ML entity that triggered or responded to the task;

[0404] Forwarding a task triggered or responded to by a cross-node AI / ML entity deployed in the access network device to one or more sub-entities in the access network device associated with the cross-node AI / ML entity that triggered or responded to the task;

[0405] Forwarding tasks triggered or responded to by an AI / ML entity in a node deployed in the access network device to one or more sub-entities associated with the task type in the access network device;

[0406] Forwarding tasks triggered or responded to by the cross-node AI / ML entity deployed in the access network device to one or more sub-entities associated with the task type in the access network device;

[0407] Forwarding a task sent by an AI / ML entity in a node deployed on a device other than the access network device to one or more sub-entities in the access network device associated with the AI / ML entity in the node that sent the task;

[0408] Forwarding a task sent by a cross-node AI / ML entity deployed on a device other than the access network device to one or more sub-entities in the access network device associated with the cross-node AI / ML entity that sent the task;

[0409] Forwarding tasks sent by the AI / ML entity in the node deployed by the device outside the access network device to one or more sub-entities associated with the task type in the access network device;

[0410] Forwarding tasks sent by a cross-node AI / ML entity deployed on a device other than the access network device to one or more sub-entities associated with the task type in the access network device.

[0411] Optionally, the access network device may pre-store a correspondence between the AI / ML entity and the sub-entities within the node of the triggering or responding task, and the access network device may determine one or more sub-entities associated with the AI / ML entity within the node of the triggering or responding task based on the correspondence. Optionally, the access network device may determine one or more sub-entities associated with the AI / ML entity within the node of the triggering or responding task based on the identifier of the AI / ML entity within the node of the triggering or responding task. For example, the access network device may use a load balancing algorithm to determine one or more sub-entities associated with the AI / ML entity within the node of the triggering or responding task based on the identifier of the AI / ML entity within the node of the triggering or responding task.

[0412] In some embodiments, the central entity deployed in the access network device directly communicates with the target AI / ML entity deployed in a device outside the access network device through an interface protocol, and / or communicates with the target AI / ML entity deployed in a device outside the access network device through a cross-node AI / ML entity deployed in a device outside the access network device.

[0413] In some embodiments, any two central entities deployed in the access network device communicate directly with each other through an interface protocol.

[0414] In some embodiments, any two central entities deployed in the access network device communicate indirectly through a specific entity.

[0415] In some embodiments, a part of the central entities deployed in the access network device communicate directly with each other through an interface protocol, and another part of the central entities communicate indirectly with each other through a specific entity. For example, for a central entity deployed in the access network device, it can communicate directly with other central devices deployed in the access network device and / or communicate indirectly through a specific entity. For example, the access network device may include central entity 1, central entity 2 and central entity 3, and central device 1 and central entity 2 can communicate directly through an interface protocol, and central entity 1 and central entity 3 can communicate indirectly through a specific entity. For another example, the access network device may include central entity 1, central entity 2, central entity 3 and central device 4, and central device 1 and central entity 2 can communicate directly through an interface protocol, and central entity 3 and central entity 4 can communicate indirectly through a specific entity.

[0416] Optionally, any intra-node AI / ML entity deployed in the access network device has or does not have a communication connection with one or more cross-node AI / ML entities deployed in the access network device. Optionally, any intra-node AI / ML entity deployed in the access network device has or does not have a communication connection with one or more central entities deployed in the access network device. Optionally, any intra-node AI / ML entity deployed in the access network device has or does not have a communication connection with one or more sub-entities deployed in the access network device.

[0417] Optionally, any cross-node AI / ML entity deployed in the access network device has or does not have a communication connection with one or more central entities deployed in the access network device. Optionally, any cross-node AI / ML entity deployed in the access network device has or does not have a communication connection with one or more sub-entities deployed in the access network device.

[0418] In some embodiments, the intra-node AI / ML entity, cross-node AI / ML entity, central entity, sub-entity, or specific entity deployed in the access network device includes at least one of the following:

[0419] A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0420] A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0421] A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling;

[0422] A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching;

[0423] The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

[0424] In some embodiments, the intra-node AI / ML entity deployed in the access network device is used for at least one of the following:

[0425] Sending, assigning, or forwarding tasks to the eleventh designated AI / ML entity;

[0426] Receive and process tasks sent, assigned, or forwarded by the eleventh designated AI / ML entity;

[0427] Forwarding tasks sent, assigned, or forwarded by the eleventh designated AI / ML entity;

[0428] Among them, the eleventh designated AI / ML entity includes at least one of the following: other node AI / ML entities deployed in the access network device, cross-node AI / ML entities deployed in the access network device, central entities deployed in the access network device, sub-entities deployed in the access network device, node AI / ML entities deployed by devices other than the access network device, cross-node AI / ML entities deployed by devices other than the access network device, central entities deployed by devices other than the access network device, and sub-entities deployed by devices other than the access network device.

[0429] In some embodiments, the cross-node AI / ML entity deployed in the access network device is used for at least one of the following:

[0430] Sending, assigning, or forwarding tasks to a twelfth designated AI / ML entity;

[0431] Receive and process tasks sent, assigned, or forwarded by the twelfth designated AI / ML entity;

[0432] forwarding tasks sent, assigned, or forwarded by the twelfth designated AI / ML entity;

[0433] Among them, the twelfth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the access network device, other cross-node AI / ML entities deployed in the access network device, a central entity deployed in the access network device, a sub-entity deployed in the access network device, an intra-node AI / ML entity deployed by a device other than the access network device, a cross-node AI / ML entity deployed by a device other than the access network device, a central entity deployed by a device other than the access network device, and a sub-entity deployed by a device other than the access network device.

[0434] In some embodiments, the central entity deployed in the access network device is used for at least one of the following:

[0435] Send, assign, or forward tasks to the 13th designated AI / ML entity;

[0436] Receive and process tasks sent, assigned, or forwarded by the 13th designated AI / ML entity;

[0437] Forwarding tasks sent, assigned, or forwarded by the thirteenth designated AI / ML entity;

[0438] Among them, the thirteenth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the access network device, a cross-node AI / ML entity deployed in the access network device, other central entities deployed in the access network device, a sub-entity deployed in the access network device, an intra-node AI / ML entity deployed by a device other than the access network device, a cross-node AI / ML entity deployed by a device other than the access network device, a central entity deployed by a device other than the access network device, and a sub-entity deployed by a device other than the access network device.

[0439] In some embodiments, the sub-entity deployed in the access network device is used for at least one of the following:

[0440] Sending, assigning, or forwarding tasks to the fourteenth designated AI / ML entity;

[0441] Receive and process tasks sent, assigned, or forwarded by the fourteenth designated AI / ML entity;

[0442] Forwarding tasks sent, assigned, or forwarded by the fourteenth designated AI / ML entity;

[0443] Among them, the fourteenth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the access network device, a cross-node AI / ML entity deployed in the access network device, a central entity deployed in the access network device, other sub-entities deployed in the access network device, an intra-node AI / ML entity deployed by a device other than the access network device, a cross-node AI / ML entity deployed by a device other than the access network device, a central entity deployed by a device other than the access network device, and a sub-entity deployed by a device other than the access network device.

[0444] In some embodiments, the specific entity deployed in the access network device is used for at least one of the following:

[0445] Sending tasks, assigning tasks, or forwarding tasks to a central entity deployed in the access network device;

[0446] Receive and process tasks sent, assigned or forwarded by a central entity deployed in the access network device;

[0447] Forwarding tasks sent, assigned or forwarded by a central entity deployed in the access network device.

[0448] In some embodiments, the task includes at least one of the following:

[0449] Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0450] Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0451] Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery;

[0452] Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

[0453] In some embodiments, a newly joined or newly activated target AI / ML entity in the access network device sends an initialization message to at least one deployed AI / ML entity;

[0454] The initialization message includes at least one of the following of the newly joined or newly activated target AI / ML entity: communication address information, identification information, supported function scope information, and deployment location information;

[0455] The at least one deployed AI / ML entity includes: one or more target AI / ML entities deployed in the access network device, and / or one or more target AI / ML entities deployed in a device outside the access network device.

[0456] Optionally, the newly joined or newly activated target AI / ML entity in the access network device is pre-configured with the communication address information and / or identification information of at least one deployed AI / ML entity, so that the newly joined or newly activated target AI / ML entity in the access network device sends an initialization message to at least one deployed AI / ML entity.

[0457] In some embodiments, any of the deployed AI / ML entities sends a first response message or a second response message to the newly joined or newly activated target AI / ML entity;

[0458] The first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity; and the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

[0459] In some embodiments, the first response message includes at least one of the following of any of the deployed AI / ML entities: communication address information, identification information, supported function scope information, and deployment location information;

[0460] The second response message includes: rejection reason information and / or rejection duration information.

[0461] In some embodiments, any target AI / ML entity deployed in the access network device sends a first notification message to at least one fifteenth designated AI / ML entity; the first notification message is used to instruct the any target AI / ML entity to perform a deactivation operation or a deletion operation on itself;

[0462] When any one of the target AI / ML entities receives a confirmation message sent by each of the fifteenth designated AI / ML entities, the any one of the target AI / ML entities performs a deactivation operation or a deletion operation on itself;

[0463] Among them, the fifteenth designated AI / ML entity includes: other target AI / ML entities deployed in the access network device that are communicatively connected to any of the target AI / ML entities, and / or target AI / ML entities deployed on devices other than the access network device that are communicatively connected to any of the target AI / ML entities.

[0464] In this case, any target AI / ML entity deployed in the access network device proactively initiates its own deactivation or deletion operation, thereby achieving its own deactivation or deletion.

[0465] Optionally, when any one or more of the at least one fifteenth designated AI / ML entity sends at least one of a rejection message, a denial message, a non-confirmation message, and a rejection reason to any target AI / ML entity, any target AI / ML entity will not perform its own deactivation operation or deletion operation.

[0466] In some embodiments, any target AI / ML entity deployed in the access network device receives a second notification message sent by the fifteenth designated AI / ML entity; the second notification message is used to instruct the any target AI / ML entity to be deactivated or deleted;

[0467] Any target AI / ML entity performs a deactivation operation or a deletion operation on itself based on the second notification message;

[0468] Among them, the fifteenth designated AI / ML entity includes: other target AI / ML entities deployed in the access network device that have a communication connection with any target AI / ML entity, and / or, target AI / ML entities deployed in a device other than the access network device that have a communication connection with any target AI / ML entity.

[0469] In this case, any target AI / ML entity deployed in the access network device passively deactivates or deletes itself.

[0470] Optionally, any target AI / ML entity deployed in the access network device can determine whether it is executing a task and / or whether there is a task to be executed upon receiving the second notification message. If both are negative, the any target AI / ML entity performs its own deactivation operation or deletion operation based on the second notification message.

[0471] Optionally, the access network device sends first indication information to the terminal device, where the first indication information is used to indicate at least one of the following:

[0472] Instructing the terminal device to newly deploy at least one intra-node AI / ML entity; instructing the terminal device to delete at least one intra-node AI / ML entity; instructing the terminal device to activate at least one intra-node AI / ML entity; instructing the terminal device to deactivate at least one intra-node AI / ML entity; instructing the terminal device to modify at least one intra-node AI / ML entity.

[0473] Optionally, the access network device receives second indication information sent by the terminal device, where the second indication information is used for at least one of the following:

[0474] Indicates at least one intra-node AI / ML entity that the terminal device expects to add; indicates at least one intra-node AI / ML entity that the terminal device expects to delete; indicates at least one intra-node AI / ML entity that the terminal device expects to activate; indicates at least one intra-node AI / ML entity that the terminal device expects to deactivate; indicates at least one intra-node AI / ML entity that the terminal device has added; indicates at least one intra-node AI / ML entity that the terminal device has deleted; indicates at least one intra-node AI / ML entity that the terminal device has activated; indicates at least one intra-node AI / ML entity that the terminal device has deactivated; indicates at least one intra-node AI / ML entity that the terminal device has modified.

[0475] In some embodiments, the communication node deployed in the access network device may send capability information to other communication nodes deployed in the access network device, or the communication node deployed in the access network device may send capability information to a communication node deployed in a device other than the access network device, or the communication node deployed in the access network device may receive capability information sent by a communication node deployed in a device other than the access network device;

[0476] The capability information indicates at least one of the following: whether deployment of intra-node AI / ML entities is supported; whether deployment of cross-node AI / ML entities is supported; the maximum number of supported intra-node AI / ML entities; the maximum number of supported cross-node AI / ML entities; the maximum number of supported sub-entities; target units and / or protocol layer entities in communication nodes where intra-node AI / ML entities can be deployed; target units and / or protocol layer entities in communication nodes where intra-node AI / ML entities cannot be deployed; communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously; communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously;

[0477] The communication node deployed in the device other than the access network device includes at least one of the following: a management device, a communication node in a core network device, a terminal device, and a protocol layer entity in a terminal device;

[0478] The communication nodes deployed in the access network device include at least one of the following: a base station, a centralized unit CU, a distributed unit DU, a centralized unit control plane CU-CP, a centralized unit user plane CU-UP, and a protocol layer entity within the access network, wherein the protocol layer entity within the access network includes at least one of the following: a non-access layer NAS entity, a service data adaptation protocol SDAP entity, a radio resource control RRC entity, a packet data convergence protocol PDCP entity, a radio link control RLC entity, a backhaul adaptation protocol BAP entity, a media access control MAC entity, and a physical layer PHY entity.

[0479] The following describes the structure of the terminal device:

[0480] Figure 5 is a structural diagram of a terminal device provided in an embodiment of the present application. As shown in Figure 5, the terminal device includes at least one target AI / ML entity; each of the target AI / ML entities is used to perform AI / ML-related tasks corresponding to the terminal device.

[0481] Optionally, the terminal device can communicate with the access network device. Optionally, the terminal device can communicate with the core network device through the access network device. Optionally, the terminal device and / or the target AI / ML entity in the terminal device can communicate with the access network device and / or the target AI / ML entity in the access network device. Optionally, the terminal device and / or the target AI / ML entity in the terminal device can communicate with the core network device and / or the target AI / ML entity in the core network device through the access network device and / or the target AI / ML entity in the access network device.

[0482] Optionally, in any embodiment of the present application, the AI / ML-related tasks corresponding to the terminal device may include at least one of the following: tasks generated by the terminal device, tasks requested by a target AI / ML entity deployed on a device other than the terminal device, and tasks interacting with a target AI / ML entity deployed on a device other than the terminal device. Optionally, the terminal device may include a target AI / ML entity and a non-target AI / ML entity, and the target AI / ML entity and the non-target AI / ML entity may communicate with each other. The tasks generated by the terminal device may be tasks generated by the target AI / ML entity and / or tasks generated by the non-target AI / ML entity.

[0483] For example, each of the target AI / ML entities in the terminal device can be used for at least one of the following: sending data and / or AI / ML models for AI / ML to a target AI / ML entity deployed on a device other than the terminal device, receiving data and / or AI / ML models for AI / ML sent by a target AI / ML entity deployed on a device other than the terminal device, generating data and / or AI / ML models for AI / ML, and training, reasoning, or monitoring the AI / ML model based on the data for AI / ML.

[0484] In some embodiments, the target AI / ML entity includes the intra-node AI / ML entity; the intra-node AI / ML entity includes at least one of the following: a terminal device AI / ML entity, a protocol layer AI / ML entity;

[0485] One or more terminal device AI / ML entities are deployed in the terminal device, and / or one or more protocol layer AI / ML entities are deployed in the protocol layer entity included in the terminal device;

[0486] The terminal device AI / ML entity is used to perform AI / ML related tasks corresponding to the terminal device, and the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer.

[0487] Optionally, the terminal device may include a protocol layer entity and a non-protocol layer entity, the terminal device AI / ML entity may be deployed in the non-protocol layer entity, and the protocol layer AI / ML entity may be deployed in the protocol layer entity.

[0488] Optionally, a terminal device may be deployed with at least one of the following: one or more terminal device AI / ML entities, one or more protocol layer AI / ML entities.

[0489] Optionally, the terminal device AI / ML entity is used to perform AI / ML related tasks corresponding to the terminal device. Optionally, the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer entity to which it belongs.

[0490] Optionally, different terminal device AI / ML entities in a terminal device may be used to process the same, different, or partially identical tasks. Optionally, different protocol layer AI / ML entities in a terminal device may be used to process the same, different, or partially identical tasks. Optionally, any terminal device AI / ML entity and any protocol layer AI / ML entity in a terminal device may be used to process the same, different, or partially identical tasks.

[0491] Optionally, in any embodiment of the present application, the protocol layer entities in the terminal device include at least one of the following: non-access layer NAS entity, service data adaptation protocol SDAP entity, radio resource control RRC entity, packet data convergence protocol PDCP entity, radio link control RLC entity, backhaul adaptation protocol BAP entity, media access control MAC entity, and physical layer PHY entity.

[0492] Optionally, the terminal device AI / ML entity and / or the protocol layer AI / ML entity may also be used to execute a task sent by a target AI / ML entity of a device external to the terminal device.

[0493] In some embodiments, any two protocol layer AI / ML entities deployed in the terminal device communicate directly through an interface protocol, or communicate indirectly through one or more terminal device AI / ML entities.

[0494] In some embodiments, the terminal device AI / ML entity deployed in the terminal device directly communicates with the access network AI / ML entity deployed in the access network device through an air interface protocol, or directly communicates with the target unit AI / ML entity deployed in the access network device through an air interface protocol.

[0495] In some embodiments, the protocol layer AI / ML entity deployed in the terminal device communicates directly with the protocol layer AI / ML entity deployed in the access network device through an air interface protocol, or communicates indirectly with the protocol layer AI / ML entity deployed in the access network device through at least one of the terminal device AI / ML entity deployed in the terminal device, the access network AI / ML entity deployed in the access network device, and the target unit AI / ML entity deployed in the access network device.

[0496] In some embodiments, the terminal device AI / ML entity deployed in the terminal device and / or the protocol layer AI / ML entity deployed in the terminal device communicates indirectly with the intra-node AI / ML entity and / or cross-node AI / ML entity deployed in the core network device through the access network device.

[0497] In some embodiments, the intra-node AI / ML entity deployed in the terminal device includes at least one of the following:

[0498] A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0499] A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0500] A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling;

[0501] A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching;

[0502] The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

[0503] In some embodiments, the intra-node AI / ML entity deployed in the terminal device is used for at least one of the following:

[0504] Sending, assigning, or forwarding tasks to the sixteenth designated AI / ML entity;

[0505] Receive and process tasks sent, assigned, or forwarded by the sixteenth designated AI / ML entity;

[0506] forwarding tasks sent, assigned, or forwarded by the sixteenth designated AI / ML entity;

[0507] Among them, the sixteenth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in an access network device, a cross-node AI / ML entity deployed in an access network device, a central entity deployed in an access network device, a sub-entity deployed in an access network device, an intra-node AI / ML entity deployed in a core network device, a cross-node AI / ML entity deployed in a core network device, an intra-node AI / ML entity deployed in a management device, and a cross-node AI / ML entity deployed in a management device.

[0508] In some embodiments, the task includes at least one of the following:

[0509] Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0510] Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0511] Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery;

[0512] Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

[0513] In some embodiments, the terminal device receives first indication information sent by an access network device or a core network device, and the first indication information is used to indicate at least one of the following: instructing the terminal device to newly deploy at least one intra-node AI / ML entity; instructing the terminal device to delete at least one intra-node AI / ML entity; instructing the terminal device to activate at least one intra-node AI / ML entity; instructing the terminal device to deactivate at least one intra-node AI / ML entity; instructing the terminal device to modify at least one intra-node AI / ML entity.

[0514] Optionally, the first indication information may also be used to indicate an identifier of an AI / ML entity within the node.

[0515] Optionally, in other embodiments, the terminal device may proactively perform at least one of the following operations: newly deploy at least one intra-node AI / ML entity, delete at least one intra-node AI / ML entity, activate at least one intra-node AI / ML entity, deactivate at least one intra-node AI / ML entity, or modify at least one intra-node AI / ML entity. Optionally, before proactively performing one of the aforementioned operations, the terminal device may send a first indication to the access network device or the core network device.

[0516] In some embodiments, the terminal device is further used to send second indication information to the access network device or the core network device, where the second indication information is used for at least one of the following:

[0517] Indicates at least one intra-node AI / ML entity that the terminal device expects to add; indicates at least one intra-node AI / ML entity that the terminal device expects to delete; indicates at least one intra-node AI / ML entity that the terminal device expects to activate; indicates at least one intra-node AI / ML entity that the terminal device expects to deactivate; indicates at least one intra-node AI / ML entity that the terminal device has added; indicates at least one intra-node AI / ML entity that the terminal device has deleted; indicates at least one intra-node AI / ML entity that the terminal device has activated; indicates at least one intra-node AI / ML entity that the terminal device has deactivated; indicates at least one intra-node AI / ML entity that the terminal device has modified.

[0518] In some embodiments, a target AI / ML entity deployed in the terminal device may send capability information to other target AI / ML entities deployed in the terminal device, or the terminal device may send capability information to a communication node deployed in a device other than the terminal device, or the terminal device may receive capability information sent by a communication node deployed in a device other than the terminal device;

[0519] The capability information indicates at least one of the following: whether deployment of intra-node AI / ML entities is supported; whether deployment of cross-node AI / ML entities is supported; the maximum number of supported intra-node AI / ML entities; the maximum number of supported cross-node AI / ML entities; the maximum number of supported sub-entities; protocol layer entities of intra-node AI / ML entities deployable in the terminal device; protocol layer entities of intra-node AI / ML entities not deployable in the terminal device; target units and / or protocol layer entities of intra-node AI / ML entities deployable in the communication node; target units and / or protocol layer entities of intra-node AI / ML entities not deployable in the communication node; communication nodes on which intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously; communication nodes on which intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously;

[0520] The communication node deployed in the device other than the terminal device includes at least one of the following: a management device, a communication node in a core network device, a communication node in an access network device, a target unit in an access network device, and a protocol layer entity in an access network device;

[0521] The communication node deployed in the terminal device includes at least one of the following: a terminal device, a protocol layer entity within the terminal device, wherein the protocol layer entity within the terminal device includes at least one of the following: a non-access layer NAS entity, a service data adaptation protocol SDAP entity, a radio resource control RRC entity, a packet data convergence protocol PDCP entity, a radio link control RLC entity, a backhaul adaptation protocol BAP entity, a media access control MAC entity, and a physical layer PHY entity.

[0522] This application can achieve the intrinsic integration of AI / ML technology and any communication function by deploying at least one target AI / ML entity on demand in any communication node in the communication system, providing theoretical support for the realization of automated communication networks by AI / ML technology. On the one hand, the distributed deployment of target AI / ML entities (for example, AI / ML entities within a node) is conducive to the timely and local resolution of communication technology problems, and to a certain extent avoids the problem of over-centralized collection of sensitive data; on the other hand, the local centralized deployment of target AI / ML entities (for example, cross-node AI / ML entities) can well solve technical problems involving multiple communication nodes and realize cross-node collaboration of AI / ML tasks; on the other hand, the formulation of a unified AI / ML architecture and protocol interface (for example, AI / ML task interaction between different devices) is conducive to the global sharing of AI / ML data, greatly improving the efficiency of data use and transmission; finally, ubiquitous AI / ML deployment can easily allow any node in the communication system to establish resource and data sharing connections with devices outside the communication system (for example, various Internet cloud servers and cloud platforms), expanding the business scope of operators and breaking the constraints of operators as data pipeline manufacturers.

[0523] The following further explains at least some of the beneficial effects of the above embodiments through multiple examples:

[0524] Example 1: The physical layer of an access network device or a terminal device needs to train a beam management AI / ML model. The input of the beam management AI / ML model may include at least one of the following: the signal measurement result of at least one beam measured by the terminal device, the beam identifier of at least one beam measured by the terminal device, the beam transmission configuration configured by the network device, the antenna type or shape configured by the network device, the location information of the terminal device, etc. The output of the beam management AI / ML model may include at least one of the following: the inferred signal measurement result of at least one beam, and the inferred beam identifier of at least one beam.

[0525] If the target AI / ML entity is deployed at the physical layer of the access network equipment and / or terminal equipment, the training of the beam management AI / ML model can be carried out at the physical layer of the access network equipment and / or terminal equipment. The collection and processing of relevant model training input data will be very timely. At the same time, it also prevents sensitive data such as terminal location information from being collected or perceived by other unrelated communication nodes, reflecting the advantage of distributed deployment of target AI / ML entities.

[0526] Example 2: The core network equipment needs to train an access network load management AI / ML model. The input of the access network load management AI / ML model may include: historical load information and / or time distribution information of each access network connected to the core network. The output of the access network load management AI / ML model may include: predicted historical load information and / or time distribution information of each access network connected to the core network within a certain period of time in the future.

[0527] If a target AI / ML entity is deployed in the core network device and a cross-node AI / ML entity is deployed in the access network device, the cross-node AI / ML entity deployed in the access network device can collect the historical load information and time distribution information of each access network managed by the cross-node AI / ML entity respectively, and summarize the collected information to the core network device through the interface between the target AI / ML entity deployed in the core network device and the cross-node AI / ML entity deployed in the access network device. The core network device uses the summarized data for model training, thereby realizing the solution of cross-node communication problems by using the cross-node AI / ML entity.

[0528] Example 3: The access network device needs to train a mobility management AI / ML model. The input of the mobility management AI / ML model may include at least one of the following: the historical movement trajectory of the terminal device, the current location of the terminal device, the load status of other access network devices, and the signal measurement results of the neighboring cells and / or serving cells measured by the terminal device. The output of the mobility management AI / ML model may include: switching to the target cell.

[0529] If the target AI / ML entity is deployed in the access network device, the access network devices can easily share the trained model or the data required for the shared model input, thereby realizing the rapid sharing of AI / ML data in the communication system.

[0530] Example 4: An internet company wants to train a smart home alarm system (for example, to promptly sound an alarm when a person falls accidentally). The input of the smart home alarm system may include at least one of the following: historical human behavior, home layout information, and human gesture information. The output of the smart home alarm system may include: turning on or off a home appliance and / or sounding an alarm in the event of an accident.

[0531] Internet companies alone would find it difficult to collect much training data. However, if ubiquitous AI / ML entities were defined within the communication system, these AI / ML entities deployed at each communication node could easily collect and share all types of data gathered by these nodes. Because these AI / ML entities are flexibly connected and have a unified architecture, internet companies could easily access AI / ML data collected by these nodes through standard interfaces, thereby training a smart home alarm system. Furthermore, operators could generate additional revenue by sharing collected AI / ML data, thereby expanding their business growth opportunities.

[0532] Of course, there are many other functions that can be optimized by using AI / ML entities on various devices. Here are some typical examples:

[0533] Functions that can be optimized for terminal devices include at least one of the following: layer 1 / layer 3 measurement optimization, paging optimization, cell reselection optimization, beam management optimization, CSI feedback optimization, mobility optimization, positioning optimization, etc.;

[0534] The functions that can be optimized for access network equipment include at least one of the following: resource scheduling optimization, load balancing optimization, coverage optimization, mobility optimization, data retransmission strategy optimization, network energy saving optimization, etc.

[0535] The functions that can be optimized for core network equipment include at least one of the following: paging optimization, mobility optimization, load optimization, etc.;

[0536] The functions that can be optimized for the management device include at least one of the following: energy saving optimization, resource management optimization, etc.

[0537] It should be noted that the above is only an illustrative example of the beneficial effects produced by the target AI / ML entity in the communication device provided in this application. The target AI / ML entity in the communication device provided in this application can also produce other beneficial effects, which can be reflected in various aspects of the communication system. This application will not go into details.

[0538] It should be noted that any solution generated based on the communication device of this application should be within the scope of protection of this application.

[0539] The following describes the communication system according to an embodiment of the present application from another aspect.

[0540] A communication system generally consists of terminal equipment, access network equipment, and core network equipment. Optionally, the communication system may also include operation, management, and maintenance (OAM) equipment (i.e., corresponding to the management equipment mentioned above), which is generally used for communication network management and maintenance. In order to use AI / ML tools to solve problems that arise in the communication system throughout the entire process (including "pre-problem solving", "in-process problem solving", or "post-problem solving"), this application proposes a hybrid AI / ML entity deployment architecture. The so-called hybrid architecture is a special architecture that combines centralized and distributed architectures.

[0541] FIG6 is a schematic diagram of the architecture of a communication system provided in an embodiment of the present application. As shown in FIG6 , the communication system includes: a management device, a core network device, an access network device, and a terminal device.

[0542] The management device may include or deploy a cross-node AI coordination entity (i.e., corresponding to the above-mentioned cross-node AI / ML entity).

[0543] The core network device may include at least one communication node. As shown in Figure 6, the at least one communication node in the core network device includes CN node 1 to CN node 4 and an AMF network element. An AI entity (corresponding to the intra-node AI / ML entity in the above embodiment) may be deployed in each device in CN node 1 to CN node 4 and the AMF network element. The core network device may also include or deploy a cross-node AI coordination entity.

[0544] The access network device may include at least one communication node. As shown in Figure 6, the at least one communication node in the access network device includes gNB1 and gNB2. gNB1 and gNB2 include a centralized unit (CU) and a distributed unit (DU). gNB1 and gNB2 also include an AI entity (corresponding to the access network AI / ML entity in the above embodiment). The access network device may also include or deploy a cross-node AI coordination entity.

[0545] The communication system includes two terminal devices, which may be terminal device 1 (UE1) and terminal device 2 (UE2). Each of terminal device 1 and terminal device 2 may be deployed with an AI entity.

[0546] It should be noted that Figure 6 only shows an exemplary implementation of a communication system architecture. In other embodiments, the communication system may have other architectures. For example, in some embodiments, the management device also includes or is deployed with an AI entity, or the management device includes or is deployed with an AI entity, but does not include or deploy a cross-node AI coordination entity; for another example, at least one device among CN node1 to CN node4 and the AMF network element may not have an AI entity deployed; for another example, the core network device may not include or deploy a cross-node AI coordination entity; for another example, at least one of gNB1 and gNB2 may not have an access network AI / ML entity deployed; for another example, at least one of gNB1 and gNB2 may be deployed with a protocol layer AI / ML entity and / or a target unit AI / ML; for another example, the access network device may not include or deploy a cross-node AI coordination entity; for another example, the communication system may include other numbers of terminal devices, and at least some of the terminal devices in the communication system may include or deploy AI entities.

[0547] Overall, the overall architecture of a communication system that includes AI / ML capabilities has at least one of the following characteristics:

[0548] Feature 1: Any communication node can deploy (or include) at least one intra-node AI / ML functional entity, including at least one of the following:

[0549] Management equipment, such as OAM equipment;

[0550] Core network equipment, such as at least one of AMF, SMF, UPF, UDM, PCF, and LMF;

[0551] Access network equipment, such as at least one of the following: gNB, CU, DU, CU-CP, CU-UP, and protocol layer entities within the access network (including at least one of SDAP, RRC, PDCP, RLC, BAP, MAC, and PHY);

[0552] Terminal equipment, such as the UE device as a whole and / or protocol layer entities within the terminal device (including at least one of NAS, SDAP, RRC, PDCP, RLC, BAP, MAC, and PHY);

[0553] In any embodiment of the present application, it is allowed that a part of the communication nodes deploy intra-node AI / ML functional entities (i.e., corresponding to the above-mentioned intra-node AI / ML entities), while at the same time another part of the communication nodes do not deploy any intra-node AI / ML functional entities. This application does not limit this.

[0554] Feature 2: For communication nodes deployed with intra-node AI / ML functional entities, the intra-node AI / ML functional entities deployed on the communication nodes can also be associated with one or more cross-node AI / ML coordination entities (the cross-node AI / ML coordination entity can be within the device, or can be outside the device, such as in the OAM device in Figure 6, the cross-node AI / ML coordination entity can be within the device, such as in the core network device in Figure 6, and the cross-node AI / ML coordination entity can be outside the communication node).

[0555] In any embodiment of the present application, it is allowed that some communication nodes deploy intra-node AI / ML functional entities but these intra-node AI / ML functional entities are not associated with any cross-node AI / ML coordination entity for communication, and this application does not limit this.

[0556] Feature 3: Task interaction between AI / ML functional entities deployed within any two communicating nodes is achieved through the following methods:

[0557] Method 1: Direct communication between AI / ML functional entities deployed within any two communication nodes;

[0558] Method 2: Intra-node AI / ML functional entities deployed on any two communicating nodes communicate indirectly through the cross-node AI / ML coordination entity.

[0559] Method 3: The AI / ML functional entities deployed in some communication nodes communicate directly with each other, while the AI / ML functional entities deployed in other communication nodes communicate indirectly through the cross-node AI / ML coordination entity.

[0560] In the first implementation, all interaction tasks between the intra-node AI / ML functional entities deployed in any two communicating nodes are implemented through the first implementation. In this case, the intra-node AI / ML functional entity deployed in each communicating node also takes on the functions of the cross-node AI / ML coordination entity. That is, the functions of the cross-node AI / ML coordination entity are part of the functions included in the intra-node AI / ML functional entity deployed in each communicating node.

[0561] In the second implementation mode, all interactive tasks between the intra-node AI / ML functional entities deployed on any two communicating nodes are implemented through mode 2. In this case, the intra-node AI / ML functional entities deployed on any two communicating nodes cannot directly interact with each other, but indirectly interact with each other through the cross-node AI / ML coordination entity.

[0562] In the third implementation method, part of the interaction tasks between the AI / ML functional entities deployed within any two communication nodes are implemented through method one, while the remaining part of the interaction tasks between the AI / ML functional entities deployed within any two communication nodes are implemented through method two. The task interaction method depends on the specific task type.

[0563] It should be noted that the communication system involved in the present application supports at least one of the above three implementation methods. For example, assuming that there are six communication nodes A, B, C, D, E, and F in the communication system, the interaction tasks between the AI / ML functional entities in the nodes deployed by A and B are all implemented through method one, and the interaction tasks between the AI / ML functional entities in the nodes deployed by C and D are all implemented through method two, and the interaction tasks between the AI / ML functional entities in the nodes deployed by E and F are all implemented through method three; of course, it is also allowed that the interaction tasks between the AI / ML functional entities in the nodes deployed by A and B are all implemented through method one, and the interaction tasks between the AI / ML functional entities in the nodes deployed by A and C are all implemented through method two, and the interaction tasks between the AI / ML functional entities in the nodes deployed by A and D are all implemented through method three. This application does not limit this.

[0564] The following describes the communication connection structure of the intra-node AI / ML functional entities and / or associated cross-node AI / ML coordination entities deployed on each communication node. The functions and communication connection structure of the intra-node AI / ML functional entities and / or associated cross-node AI / ML coordination entities deployed above may vary for different communication nodes.

[0565] The following describes how to manage devices:

[0566] FIG7 is a schematic diagram of a management device communicating with other communication nodes according to an embodiment of the present application. As shown in FIG7 , the management device may include a cross-node AI / ML coordination entity. Optionally, the management device may also include an AI entity, and the AI ​​entity may be communicatively connected to the cross-node AI / ML coordination entity. The cross-node AI / ML coordination entity in the management device may communicate indirectly with the AI ​​entity in the other communication node 1 through the cross-node AI / ML coordination entity, and the cross-node AI / ML coordination entity in the management device may communicate directly with the AI ​​entity in the other communication node 2.

[0567] In some embodiments, the management device may deploy at least one cross-node AI / ML coordination entity, each cross-node AI / ML coordination entity being used to handle tasks of interaction with intra-node AI / ML functional entities deployed with other communication nodes and / or tasks of interaction with cross-node AI / ML coordination entities associated with other communication nodes.

[0568] Optionally, the management device may further deploy at least one intra-node AI / ML functional entity, where the at least one intra-node AI / ML functional entity is used to process tasks generated within the management device.

[0569] The cross-node AI / ML coordination entity deployed on the management device can communicate directly with the intra-node AI / ML functional entity deployed on other communication nodes, or communicate indirectly with the intra-node AI / ML functional entity deployed on other communication nodes through another cross-node AI / ML coordination entity. The intra-node AI / ML functional entity deployed in the management device is an optional entity, and there may or may not be a connection between it and the cross-node AI / ML coordination entity deployed in the management device.

[0570] Optionally, if the management device deploys at least two cross-node AI / ML coordination entities, there are two ways for any two cross-node AI / ML coordination entities to communicate:

[0571] Method 1: Any two cross-node AI / ML coordination entities communicate directly through an interface protocol;

[0572] Method 2: Any two cross-node AI / ML coordination entities communicate indirectly through a cross-node AI / ML coordination entity central entity. That is, there is no direct communication interface between any two cross-node AI / ML coordination entities. Instead, indirect communication is achieved through a cross-node AI / ML coordination entity central entity forwarding tasks.

[0573] Figure 8 is a schematic diagram of another management device communicating with other communication nodes provided in an embodiment of the present application. As shown in Figure 8, the management device deploys at least one cross-node AI / ML coordination entity central entity and at least one cross-node AI / ML coordination entity sub-entity (for example, Figure 8 shows cross-node AI / ML coordination entity sub-entities 1-4), and each cross-node AI / ML coordination entity central entity is associated with at least one cross-node AI / ML coordination entity sub-entity, wherein the cross-node AI / ML coordination entity central entity is used to forward tasks sent by intra-node AI / ML functional entities or other cross-node AI / ML coordination entities deployed by other communication nodes to corresponding cross-node AI / ML coordination entity sub-entities for processing; or, forward tasks triggered or responded to by cross-node AI / ML coordination entity sub-entities to corresponding intra-node AI / ML functional entities or other cross-node AI / ML coordination entities deployed by other communication nodes for processing.

[0574] Optionally, the management device may further deploy at least one intra-node AI / ML functional entity, where the at least one intra-node AI / ML functional entity is used to process tasks generated within the management device.

[0575] The cross-node AI / ML coordination entity central entity deployed by the management device can communicate directly with the intra-node AI / ML functional entity (i.e., AI entity) deployed by other communication nodes 2, or communicate indirectly with the intra-node AI / ML functional entity (i.e., AI entity) deployed by other communication nodes 1 through another cross-node AI / ML coordination entity; there is a connection between the cross-node AI / ML coordination entity central entity deployed by the management device and at least one deployed cross-node AI / ML coordination entity sub-entity, wherein the cross-node AI / ML coordination entity central entity is mainly responsible for the routing and allocation of tasks, while the cross-node AI / ML coordination entity sub-entity is used to process specific tasks.

[0576] Optionally, there may be two principles for dividing sub-entities of the cross-node AI / ML coordination entity:

[0577] The first principle: divide according to the granularity of communication nodes.

[0578] All tasks involved in the intra-node AI / ML functional entity deployed on any communication node are processed by the same cross-node AI / ML coordination entity sub-entity. The cross-node AI / ML coordination entity sub-entities associated with the intra-node AI / ML functional entities deployed on different communication nodes may be the same or different. At this time, the intra-node AI / ML functional entity deployed on the communication node and the cross-node AI / ML coordination entity sub-entity are a one-to-one mapping relationship (one communication node is associated with one cross-node AI / ML coordination entity sub-entity) or a many-to-one mapping relationship (multiple communication nodes are associated with one cross-node AI / ML coordination entity sub-entity).

[0579] The second principle: divide according to the granularity of task types.

[0580] The tasks involved in the intra-node AI / ML functional entity deployed on a communication node are one or more, and each task is associated with a cross-node AI / ML coordination entity sub-entity. The cross-node AI / ML coordination entity sub-entities associated between the tasks may be the same or different. At this time, the intra-node AI / ML functional entity deployed on the communication node and the cross-node AI / ML coordination entity sub-entity are in a one-to-one mapping relationship (the intra-node AI / ML functional entity deployed on a communication node contains only one task scenario; or the intra-node AI / ML functional entity deployed on a communication node contains at least two task scenarios, but the cross-node AI / ML coordination entity sub-entities associated with at least two tasks are the same) or a one-to-many mapping relationship (the intra-node AI / ML functional entity deployed on a communication node contains at least two task scenarios, and there are at least two tasks with different cross-node AI / ML coordination entity sub-entities associated).

[0581] It should be noted that from the perspective of the communication node, the relationship between the intra-node AI / ML functional entity deployed on the same communication node and the cross-node AI / ML coordination entity sub-entity may be the first or second type mentioned above, but if there are more than one intra-node AI / ML functional entities deployed on the same communication node (for example, more than one intra-node AI / ML functional entity may be deployed in a UE or CU or DU or gNB), then the relationship between any intra-node AI / ML functional entity deployed on the same communication node and the cross-node AI / ML coordination entity sub-entity may be the first or second type mentioned above; from the perspective of the cross-node AI / ML coordination entity sub-entity, the relationship between the same cross-node AI / ML coordination entity sub-entity and the intra-node AI / ML functional entities deployed on different communication nodes may be the first and / or second type mentioned above (wherein, the 'and' relationship is for the relationship between the intra-node AI / ML functional entity deployed on communication node 1 and the cross-node AI / ML coordination entity sub-entity as the first type mentioned above, and the relationship between the intra-node AI / ML functional entity deployed on communication node 2 and the same cross-node AI / ML coordination entity sub-entity as the second type mentioned above).

[0582] Optionally, at least one intra-node AI / ML functional entity may be deployed in the management device, which may or may not be connected to the cross-node AI / ML coordination entity central entity or the cross-node AI / ML coordination entity sub-entity deployed in the management device.

[0583] Optionally, if the management device deploys at least two cross-node AI / ML coordination entity center entities, there are two ways for any two cross-node AI / ML coordination entity center entities to communicate:

[0584] Method 1: Any two cross-node AI / ML coordination entity central entities communicate directly through the interface protocol;

[0585] Method 2: Any two cross-node AI / ML coordination entity center entities communicate indirectly through a task coordination center entity. That is, there is no direct communication interface between any two cross-node AI / ML coordination entity center entities. Instead, indirect communication is achieved by forwarding tasks through a task coordination center entity.

[0586] The other communication nodes 1 and 2 in FIG. 7 and FIG. 8 are other communication nodes except the management device.

[0587] The following is an explanation of the core network equipment (such as at least one of AMF, SMF, UPF, UDM, PCF, and LMF). First of all, it should be noted that the following core network may refer to the core network equipment unless otherwise specified:

[0588] FIG9 is a schematic diagram of a core network device communicating with other communication nodes according to an embodiment of the present application. As shown in FIG9 , the core network deploys at least one cross-node AI / ML coordination entity, each of which is configured to handle at least one of the following tasks:

[0589] Other tasks requested by cross-node AI / ML coordination entities deployed in the core network;

[0590] The task of interaction between AI / ML functional entities deployed within any two communication nodes in the core network;

[0591] The task of interaction between the AI / ML functional entity deployed in any communication node in the core network and the AI / ML functional entity deployed in other communication nodes other than the core network communication node;

[0592] The task of interaction between the intra-node AI / ML functional entity deployed in any communication node in the core network and the cross-node AI / ML coordination entity associated with other communication nodes except the core network communication node.

[0593] Any communication node (such as AMF, any device from CN node1 to CN node4) can deploy at least one intra-node AI / ML functional entity according to functional requirements.

[0594] The cross-node AI / ML coordination entity deployed in the core network can communicate directly with the intra-node AI / ML functional entity (i.e., AI entity) deployed in other communication nodes 1, or communicate indirectly with the intra-node AI / ML functional entity (i.e., AI entity) deployed in other communication nodes 2 through another cross-node AI / ML coordination entity. The communication mode between the intra-node AI / ML functional entities deployed in any two communication nodes in the core network is any of the following:

[0595] Direct communication between intra-node AI / ML functional entities deployed in any two communication nodes in the core network (without communication through cross-node AI / ML coordination entities), for example: the communication node AMF sends a task request directly to the other core network communication node; or

[0596] The intra-node AI / ML functional entities deployed in any two communication nodes in the core network communicate indirectly through the cross-node AI / ML coordination entity deployed in the core network. For example, the cross-node AI / ML coordination entity forwards the task request sent by the AMF to the intra-node AI / ML functional entity deployed in other core network communication nodes for processing according to the task target.

[0597] Optionally, if the core network deploys at least two cross-node AI / ML coordination entities, there are two ways for any two cross-node AI / ML coordination entities to communicate:

[0598] Method 1: Any two cross-node AI / ML coordination entities communicate directly through an interface protocol;

[0599] Method 2: Any two cross-node AI / ML coordination entities communicate indirectly through a cross-node AI / ML coordination entity central entity. That is, there is no direct communication interface between any two cross-node AI / ML coordination entities. Instead, indirect communication is achieved through a cross-node AI / ML coordination entity central entity forwarding tasks.

[0600] FIG10 is a schematic diagram of another core network device communicating with other communication nodes according to an embodiment of the present application. As shown in FIG10 , the core network device deploys at least one cross-node AI / ML coordination entity central entity and at least one cross-node AI / ML coordination entity sub-entity (i.e., cross-node AI / ML coordination entity sub-entities 1-4 in FIG10 ). Each cross-node AI / ML coordination entity central entity is associated with at least one cross-node AI / ML coordination entity sub-entity. The cross-node AI / ML coordination entity central entity is used to process at least one of the following tasks:

[0601] Other cross-node AI / ML coordination tasks requested by central entities deployed in the core network;

[0602] Forward (or route) a task request triggered (or responded to) by an intra-node AI / ML functional entity deployed on any communication node in the core network to a cross-node AI / ML coordination entity sub-entity deployed on the core network, or an intra-node AI / ML functional entity deployed on a communication node other than the core network communication node, or a cross-node AI / ML coordination entity associated with a communication node other than the core network communication node for processing;

[0603] Forward (or route) a task request triggered (or responded to) by a cross-node AI / ML coordination entity sub-entity deployed in the core network, or an intra-node AI / ML functional entity deployed in other communication nodes other than the core network communication node, or an inter-node AI / ML coordination entity associated with other communication nodes other than the core network communication node, to an intra-node AI / ML functional entity deployed in one or more communication nodes in the core network for processing.

[0604] Any communication node (such as AMF, any device from CN node1 to CN node4) can deploy at least one intra-node AI / ML functional entity according to functional requirements.

[0605] The cross-node AI / ML coordination entity central entity deployed by the core network device can communicate directly with the intra-node AI / ML functional entity (i.e., AI entity) deployed by other communication nodes 1, or communicate indirectly with the intra-node AI / ML functional entity (i.e., AI entity) deployed by other communication nodes 2 through another cross-node AI / ML coordination entity; there is a connection between the cross-node AI / ML coordination entity central entity deployed by the core network device and at least one deployed cross-node AI / ML coordination entity sub-entity, wherein the cross-node AI / ML coordination entity central entity is mainly responsible for the routing and allocation of tasks, while the cross-node AI / ML coordination entity sub-entity is used to handle specific tasks. Among them, the division principle of the cross-node AI / ML coordination entity sub-entities deployed by the core network is similar to the description in the second case when the aforementioned communication node is a management device, and will not be repeated here.

[0606] Optionally, if the core network deploys at least two cross-node AI / ML coordination entity center entities, there are two ways for any two cross-node AI / ML coordination entity center entities to communicate:

[0607] Method 1: Any two cross-node AI / ML coordination entity central entities communicate directly through the interface protocol;

[0608] Method 2: Any two cross-node AI / ML coordination entity center entities communicate indirectly through a task coordination center entity. That is, there is no direct communication interface between any two cross-node AI / ML coordination entity center entities. Instead, indirect communication is achieved by forwarding tasks through a task coordination center entity.

[0609] Optionally, for the communication node being a core network device, and for any of the above two situations, it also includes: any core network communication node (such as AMF, SMF, UPF, UDM, PCF or LMF) may also have a direct connection with a cross-node AI / ML coordination entity that does not belong to the core network or with an intra-node AI / ML functional entity deployed at a communication node that does not belong to the core network.

[0610] Example 1: Direct connection and communication between the AI / ML functional entity deployed in the AMF node of the core network communication node and the AI / ML functional entity deployed in the gNB node;

[0611] Example 2: The intra-node AI / ML functional entity deployed in the core network communication node AMF and the intra-node AI / ML functional entity deployed in the gNB are indirectly connected and communicated through the cross-node AI / ML coordination entity deployed in the access network.

[0612] Among them, the other communication nodes 1 and other communication nodes 2 in Figures 9 and 10 are other communication nodes except the core network equipment.

[0613] The following describes the communication nodes in the access network equipment (e.g., gNB, CU, DU, CU-CP, CU-UP, and at least one of the protocol layer entities in the access network). It should be noted that the following access network, unless otherwise specified, may refer to the access network equipment:

[0614] FIG11 is a schematic diagram of communication between an access network device and other communication nodes according to an embodiment of the present application. As shown in FIG11 , the access network deploys at least one cross-node AI / ML coordination entity, and each cross-node AI / ML coordination entity is configured to handle at least one of the following tasks:

[0615] Tasks requested by other cross-node AI / ML coordination entities deployed in the access network;

[0616] The task of interaction between AI / ML functional entities deployed within any two communication nodes in the access network;

[0617] The task of interaction between the intra-node AI / ML functional entity deployed in any communication node in the access network and the intra-node AI / ML functional entity deployed in other communication nodes except the access network communication node;

[0618] The task of interaction between the intra-node AI / ML functional entity deployed in any communication node in the access network and the cross-node AI / ML coordination entity associated with other communication nodes except the access network communication node.

[0619] Any communication node (e.g., a gNB) may deploy at least one intra-node AI / ML functional entity (corresponding to a first-level intra-node AI / ML functional entity) based on functional requirements. Optionally, a target unit (e.g., a CU, DU, CU-CP, or CU-UP) within a communication node within an access network device may also deploy at least one intra-node AI / ML functional entity (corresponding to a second-level intra-node AI / ML functional entity) based on functional requirements. Optionally, a protocol layer entity (e.g., NAS, SDAP, RRC, PDCP, RLC, BAP, MAC, or PHY) within the access network may also deploy at least one intra-node AI / ML functional entity (corresponding to a third-level intra-node AI / ML functional entity) based on functional requirements. For example, in Figure 11 , an AI entity may be deployed in an entity outside the target unit of gNB1, gNB2, and gNB3. Optionally, the AI ​​entity deployed in an entity outside the target unit of gNB1, gNB2, and gNB3 may communicate with the cross-node AI / ML coordination entity. For example, in Figure 11, AI entities can be deployed in the RRC layer, PDCP layer, MAC layer, and PHY layer in gNB1, gNB2, and gNB3.

[0620] The cross-node AI / ML coordination entity deployed in the access network can communicate directly with the intra-node AI / ML functional entity (i.e., AI entity) deployed in other communication nodes 1, or communicate indirectly with the intra-node AI / ML functional entity (i.e., AI entity) deployed in other communication nodes 2 through another cross-node AI / ML coordination entity. The communication mode between the intra-node AI / ML functional entities deployed in any two communication nodes in the access network is any of the following:

[0621] Direct communication between the intra-node AI / ML functional entities deployed on any two communication nodes in the access network (without going through the inter-node AI / ML coordination entity). For example, the intra-node AI / ML functional entity deployed on the communication node gNB1 CU sends a task request directly to the intra-node AI / ML functional entity deployed on the communication node gNB1 DU.

[0622] The intra-node AI / ML functional entities deployed on any two communication nodes in the access network communicate indirectly through the inter-node AI / ML coordination entity deployed on the access network. For example, the inter-node AI / ML coordination entity forwards the task request sent by gNB1 CU to the intra-node AI / ML functional entity deployed on the communication node gNB2CU for processing based on the task objective.

[0623] Intra-node AI / ML functional entities deployed at any two communication nodes within the access network communicate indirectly through the intra-node AI / ML functional entity deployed within the access network (e.g., within the gNB). Example 1: The intra-node AI / ML functional entity deployed within gNB1 forwards the task request sent by the intra-node AI / ML functional entity deployed at the gNB1 CU to the intra-node AI / ML functional entity deployed at the communication node gNB1 DU for processing based on the task objective. Example 2: The intra-node AI / ML functional entity deployed within gNB1 forwards the task request sent by the intra-node AI / ML functional entity deployed at the gNB1 CU RRC protocol layer to the intra-node AI / ML functional entity deployed at the communication node gNB1 CU PDCP protocol layer for processing based on the task objective. Example 3: The intra-node AI / ML functional entity deployed within gNB1 CU forwards the task request sent by the intra-node AI / ML functional entity deployed at the gNB1 CU RRC protocol layer to the intra-node AI / ML functional entity deployed at the communication node gNB1 CU PDCP protocol layer for processing based on the task objective.

[0624] Optionally, if at least two cross-node AI / ML coordination entities are deployed in the access network, there are two communication modes between any two cross-node AI / ML coordination entities:

[0625] Method 1: Any two cross-node AI / ML coordination entities communicate directly through an interface protocol;

[0626] Method 2: Any two cross-node AI / ML coordination entities communicate indirectly through a cross-node AI / ML coordination entity central entity. That is, there is no direct communication interface between any two cross-node AI / ML coordination entities. Instead, indirect communication is achieved through a cross-node AI / ML coordination entity central entity forwarding tasks.

[0627] FIG12 is a schematic diagram of another access network device communicating with other communication nodes according to an embodiment of the present application. As shown in FIG12 , the access network device deploys at least one cross-node AI / ML coordination entity central entity and at least one cross-node AI / ML coordination entity sub-entity. Each cross-node AI / ML coordination entity central entity is associated with at least one cross-node AI / ML coordination entity sub-entity. The cross-node AI / ML coordination entity central entity is used to process at least one of the following tasks:

[0628] Other cross-node AI / ML coordination tasks requested by central entities deployed in the access network;

[0629] Forward (or route) a task request triggered (or responded to) by an intra-node AI / ML functional entity deployed at any communication node in the access network to a cross-node AI / ML coordination entity sub-entity deployed at the access network, or an intra-node AI / ML functional entity deployed at a communication node other than the access network communication node, or a cross-node AI / ML coordination entity associated with a communication node other than the access network communication node for processing;

[0630] The task request triggered (or responded to) by a cross-node AI / ML coordination entity sub-entity deployed in the access network, or an intra-node AI / ML functional entity deployed in other communication nodes other than the access network communication node, or a cross-node AI / ML coordination entity associated with other communication nodes other than the access network communication node is forwarded (or routed) to an intra-node AI / ML functional entity deployed in one or more communication nodes in the access network for processing.

[0631] A communication node (e.g., a gNB) within any access network device may deploy at least one intra-node AI / ML functional entity (corresponding to a first-level intra-node AI / ML functional entity) based on functional requirements. Optionally, a communication node (e.g., a CU, DU, CU-CP, or CU-UP) within the access network device may also deploy at least one intra-node AI / ML functional entity (corresponding to a second-level intra-node AI / ML functional entity) based on functional requirements. Optionally, an access network protocol layer entity (e.g., NAS, SDAP, RRC, PDCP, RLC, BAP, MAC, or PHY) may also deploy at least one intra-node AI / ML functional entity (corresponding to a third-level intra-node AI / ML functional entity) based on functional requirements. For example, in Figure 12 , an AI entity may be deployed in an entity outside the target cell of gNB1, gNB2, and gNB3. Optionally, the AI ​​entity deployed in an entity outside the target cell of gNB1, gNB2, and gNB3 may communicate with the cross-node AI / ML coordination entity. For example, in Figure 12, AI entities can be deployed in the RRC layer, PDCP layer, MAC layer, and PHY layer in gNB1, gNB2, and gNB3.

[0632] The cross-node AI / ML coordination entity central entity deployed in the access network can communicate directly with the intra-node AI / ML functional entity (i.e., AI entity) deployed in other communication nodes 1, or communicate indirectly with the intra-node AI / ML functional entity (i.e., AI entity) deployed in other communication nodes 2 through another cross-node AI / ML coordination entity; there is a connection between the cross-node AI / ML coordination entity central entity deployed in the access network device and at least one deployed cross-node AI / ML coordination entity sub-entity, wherein the cross-node AI / ML coordination entity central entity is mainly responsible for the routing and allocation of tasks, while the cross-node AI / ML coordination entity sub-entity is used to handle specific tasks. Among them, the division principle of the cross-node AI / ML coordination entity sub-entities deployed in the access network is similar to the description in the second case when the aforementioned communication node is a management device, and will not be repeated here.

[0633] Optionally, if at least two cross-node AI / ML coordination entity center entities are deployed in the access network, there are two ways of communication between any two cross-node AI / ML coordination entity center entities:

[0634] Method 1: Any two cross-node AI / ML coordination entity central entities communicate directly through the interface protocol;

[0635] Method 2: Any two cross-node AI / ML coordination entity center entities communicate indirectly through a task coordination center entity. That is, there is no direct communication interface between any two cross-node AI / ML coordination entity center entities. Instead, indirect communication is achieved by forwarding tasks through a task coordination center entity.

[0636] Optionally, any access network communication node (such as gNB, CU, DU, CU-CP, CU-UP or protocol layer entity within the access network) may also have a direct connection with a cross-node AI / ML coordination entity that does not belong to the access network or an intra-node AI / ML functional entity deployed at a communication node that does not belong to the access network.

[0637] Example 1: Direct connection and communication between the AI / ML functional entity deployed in the node of the access network communication node gNB and the AI / ML functional entity deployed in the node of the AMF;

[0638] Example 2: The AI / ML functional entity deployed in the access network communication node gNB and the AI / ML functional entity deployed in the AMF node communicate indirectly through the cross-node AI / ML coordination entity deployed in the core network;

[0639] Example 3: Direct connection and communication between the AI / ML functional entity deployed in the access network communication node gNB or gNB CU or gNB DU and the AI / ML functional entity deployed in the node of the UE.

[0640] The other communication nodes 1 and 2 described in FIG. 11 and FIG. 12 are other communication nodes except the access network devices.

[0641] The following describes the terminal device:

[0642] Figure 13 is a structural diagram of a terminal device provided in an embodiment of the present application. As shown in Figure 13, any terminal device can deploy at least one intra-node AI / ML functional entity according to functional requirements (that is, deploying the AI ​​entity in an entity outside the protocol layer entity, corresponding to the first-level intra-node AI / ML functional entity); optionally, the protocol layer entity inside the terminal device (such as: NAS, SDAP, RRC, PDCP, RLC, BAP, MAC or PHY) can also deploy at least one intra-node AI / ML functional entity according to functional requirements (that is, deploying the AI ​​entity in the protocol layer entity, corresponding to the second-level intra-node AI / ML functional entity). Exemplarily, Figure 13 shows that AI entities are deployed in RRC, PDCP, RLC, MAC and PHY layer entities.

[0643] Any two intra-node AI / ML functional entities deployed at the protocol layer within a terminal device (i.e., second-level intra-node AI / ML functional entities deployed within the terminal device) can communicate directly or indirectly through first-level intra-node AI / ML functional entities deployed within the terminal device.

[0644] The AI / ML functional entity within the first-level node deployed inside the terminal device can also communicate directly with the AI / ML functional entity within the first-level node or the AI / ML functional entity within the second-level node deployed in the access network device through the air interface protocol;

[0645] The AI / ML functional entity in the second-level node deployed inside the terminal device can also communicate directly with the AI / ML functional entity in the third-level node deployed in the access network device through the air interface protocol;

[0646] The AI / ML functional entity in the second-level node deployed inside the terminal device can also communicate indirectly with the AI / ML functional entity in the third-level node deployed on the access network device through the AI / ML functional entity in the first-level node deployed inside the terminal device and the AI / ML functional entity in the first-level node deployed on the access network device (or through the AI / ML functional entity in the first-level node deployed inside the terminal device and the AI / ML functional entity in the second-level node deployed on the access network device);

[0647] The AI / ML functional entity within the first-level node deployed inside the terminal device or the AI / ML functional entity within the second-level node deployed inside the terminal device can also communicate indirectly with the AI / ML functional entity within the node deployed by the core network device (such as: AMF device) by forwarding messages through the access network.

[0648] The following describes the functions of the intra-node AI / ML functional entities and / or associated cross-node AI / ML coordination entities deployed on each communication node:

[0649] In some scenarios, for a communication node that is a management device (such as an OAM device), a core network device (such as at least one of AMF, SMF, UPF, UDM, PCF, and LMF), an access network device (gNB, CU, DU, CU-CP, CU-UP, or at least one of the protocol layer entities within the access network), or a terminal device (the entire UE device and / or the protocol layer entity within the terminal device), the node AI / ML functional entity deployed in the above communication node has at least one of the following functions:

[0650] General functions (applicable both internally and externally, i.e., these functions can be used by the AI / ML functional entity deployed within the communication node, and can also be used in the external tasks of the AI / ML functional entity deployed within the communication node):

[0651] Data management functions, including at least one of data collection (for example, data includes model input data, model output data, and intermediate model data), data storage, data modification or update, data deletion, data replication, and data forwarding operations;

[0652] Storage management functions, including at least one of remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting operations;

[0653] Computing power management functions, including at least one of remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery operations;

[0654] Model lifecycle management functions, including at least one of model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, deactivation, and model deletion;

[0655] The task management function is used to manage and control any internal or external task, including at least one of task generation, task acceptance or rejection, task splitting, task allocation, task monitoring, task update, and task deletion operations.

[0656] In some embodiments, the gNB wishes to train a resource scheduling AI / ML model. The task management function first generates the task and then splits the task into subtasks (including data collection subtasks, storage scheduling subtasks, computing power scheduling subtasks, and model lifecycle management subtasks). The data collection subtask is assigned to the data management function, the storage scheduling subtask is assigned to the storage management function, the computing power scheduling subtask is assigned to the computing power management function, and the model lifecycle management subtask is assigned to the model lifecycle management function. Task monitoring is responsible for monitoring the completion status of each subtask until it is completed. If necessary, the subtask can be updated. After the task is completed, the corresponding task process is deleted.

[0657] For any of the above communication nodes, the intra-node AI / ML functional entity and / or the associated cross-node AI / ML coordination entity may perform at least one of the following:

[0658] Actively sending, allocating, or forwarding a task to a first AI / ML functional entity (corresponding to the first designated AI / ML entity, the sixth designated AI / ML entity, or the eleventh designated AI / ML entity in the above embodiment);

[0659] Receive and process tasks proactively sent, assigned, or forwarded by the first AI / ML functional entity;

[0660] Forwarding tasks proactively sent, assigned, or forwarded by the first AI / ML functional entity.

[0661] The first AI / ML functional entity is a cross-node AI / ML coordination entity or an intra-node AI / ML functional entity deployed in other communication nodes or a cross-node AI / ML coordination entity central entity.

[0662] Among them, the task types involved in the above external functions include at least one of the following: data management tasks; storage management tasks; computing power management tasks; model lifecycle management tasks.

[0663] The above tasks are transmitted through the interface protocol or the air interface protocol. For the specific functional meaning, please refer to the corresponding functional description in the above general function section and will not be repeated here.

[0664] In other scenarios, for the cross-node AI / ML coordination entity, cross-node AI / ML coordination entity central entity, cross-node AI / ML coordination entity sub-entity, or task coordination center entity deployed on the management device, core network, or access network, the main functions of each entity are as follows:

[0665] A cross-node AI / ML coordination entity includes at least one of the following functions:

[0666] Actively sending, allocating, or forwarding a task to a second AI / ML functional entity (corresponding to the second designated AI / ML entity, the seventh designated AI / ML entity, or the twelfth designated AI / ML entity in the above embodiment);

[0667] Receive and process tasks proactively sent, assigned, or forwarded by the second AI / ML functional entity;

[0668] Forwards tasks proactively sent, assigned, or forwarded by the second AI / ML functional entity.

[0669] Optionally, the second AI / ML functional entity is another cross-node AI / ML coordination entity, an intra-node AI / ML functional entity deployed on another communication node, or a cross-node AI / ML coordination entity central entity. Optionally, each of the above tasks is transmitted via an interface protocol or an air interface protocol. For specific functional meanings, please refer to the corresponding functional description in the general function section of Scenario 1 above and will not be repeated here.

[0670] The cross-node AI / ML coordination entity central entity includes at least one of the following functions:

[0671] Actively sending, allocating, or forwarding tasks to a third AI / ML functional entity (corresponding to the third designated AI / ML entity, the eighth designated AI / ML entity, or the thirteenth designated AI / ML entity in the above embodiment);

[0672] Receive and process tasks proactively sent, assigned, or forwarded by the third AI / ML functional entity;

[0673] Forwards tasks proactively sent, assigned, or forwarded by the third AI / ML functional entity.

[0674] Optionally, the third AI / ML functional entity is a cross-node AI / ML coordination entity, an intra-node AI / ML functional entity deployed at another communication node, or another cross-node AI / ML coordination entity central entity, a cross-node AI / ML coordination entity sub-entity, or a task coordination central entity. Optionally, each of the above tasks is transmitted via an interface protocol or an air interface protocol. For specific functional meanings, please refer to the corresponding functional description in the general functions section of Scenario 1 above and will not be repeated here.

[0675] The cross-node AI / ML coordination entity sub-entity includes at least one of the following functions:

[0676] Actively send or assign tasks to the cross-node AI / ML coordination entity central entity;

[0677] Receive and process tasks actively sent, assigned, or forwarded by the cross-node AI / ML coordination entity center (this may also include task rejection situations).

[0678] Optionally, the above tasks are transmitted through an interface protocol or an air interface protocol. For specific functional meanings, please refer to the corresponding functional description in the general function part of the above scenario 1, which will not be repeated here.

[0679] Optionally, the task coordination center entity (corresponding to the specific entity mentioned above) includes at least one of the following functions:

[0680] Actively send, assign, or forward tasks to the cross-node AI / ML coordination entity central entity;

[0681] Receive and process tasks proactively sent, assigned, or forwarded by the cross-node AI / ML coordination entity center;

[0682] Forwarding cross-node AI / ML coordination entity The central entity actively sends, assigns, or forwards tasks.

[0683] Optionally, the above tasks are transmitted through an interface protocol or an air interface protocol. For specific functional meanings, please refer to the corresponding functional description in the general function part of the above scenario 1, which will not be repeated here.

[0684] Optionally, the intra-node AI / ML functional entity deployed in each communication node and / or the associated cross-node AI / ML coordination entity may also be initialized, activated, deactivated or deleted.

[0685] Introduction to the initialization process: Before a newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity is officially deployed, it is necessary to use an initialization protocol message (corresponding to the initialization message in the above embodiment) to allow the newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity to establish a connection with the already officially deployed intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity (hereinafter referred to as the target intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity). After the connection is established, the newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity is considered to be officially deployed. The above initialization protocol message is sent by the newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity to the already officially deployed intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity. The message includes at least one of the following information:

[0686] Communication address information or identification information of the newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity;

[0687] The scope of functions supported by the newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity;

[0688] The communication node to which the newly added intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity belongs.

[0689] Optionally, after receiving the initialization protocol message sent by the newly joined intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity, the behavior of the target intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity also includes sending a first response message to the newly joined intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity, where the first response message is used to accept the initialization protocol message sent by the newly joined intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity; or, sending a second response message to the newly joined intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity, where the second response message is used to reject the initialization protocol message sent by the newly joined intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity.

[0690] The first response message includes at least one of the following information:

[0691] Communication address information or identification information of the AI / ML functional entity within the target node and / or the cross-node AI / ML coordination entity;

[0692] The scope of functions supported by the AI / ML functional entity within the target node and / or the cross-node AI / ML coordination entity;

[0693] The communication node to which the AI / ML functional entity within the target node and / or the cross-node AI / ML coordination entity belongs.

[0694] The second response message includes the reason for rejection (for example, failure to meet the functional requirements of the AI / ML functional entity within the target node and / or the cross-node AI / ML coordination entity; or, the connection capacity of the AI / ML functional entity within the target node and / or the cross-node AI / ML coordination entity reaches the upper limit) and / or the duration of rejection (how long the initialization protocol message is not allowed to be sent again).

[0695] Introduction to the activation or deactivation process: After the intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities are deployed, if some intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities require temporary maintenance or have no tasks for a long time (hereinafter referred to as source node intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities), the source node intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities can actively or passively perform deactivation operations.

[0696] Active method: After the AI / ML functional entity in the source node and / or the cross-node AI / ML coordination entity actively performs the deactivation operation, it also includes notifying the AI / ML functional entities in other connected nodes and / or the cross-node AI / ML coordination entities; or, before the AI / ML functional entity in the source node and / or the cross-node AI / ML coordination entity actively performs the deactivation operation, it also includes notifying the AI / ML functional entities in other connected nodes and / or the cross-node AI / ML coordination entities. After the AI / ML functional entities in other connected nodes and / or the cross-node AI / ML coordination entities reply to the confirmation message, the AI / ML functional entity in the source node and / or the cross-node AI / ML coordination entity actively performs the deactivation operation.

[0697] Passive mode: The AI / ML functional entity in the source node and / or the cross-node AI / ML coordination entity performs the deactivation operation after receiving the deactivation message sent by the fourth AI / ML functional entity.

[0698] The fourth AI / ML functional entity is another cross-node AI / ML coordination entity or an intra-node AI / ML functional entity deployed at another communication node or a cross-node AI / ML coordination entity central entity or a cross-node AI / ML coordination entity sub-entity or a task coordination center entity.

[0699] Optionally, similar to the above-mentioned deactivation operation, the AI / ML functional entity within the source node and / or the cross-node AI / ML coordination entity can actively or passively perform the activation operation. The specific process can be referred to the description of the deactivation operation behavior and will not be repeated here.

[0700] Deletion operation: The AI / ML functional entity within the source node and / or the cross-node AI / ML coordination entity can be deleted actively or passively. Once the deletion is successful, the next deployment still needs to go through the above initialization process.

[0701] Optionally, in the embodiment of the present application, any two communication nodes (e.g., communication node 1 and communication node 2) may exchange capability information, where the capability information includes at least one of the following capabilities:

[0702] Whether it supports the deployment of at least one intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity;

[0703] The maximum number of supported intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities;

[0704] Which intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities cannot be deployed simultaneously?

[0705] Which intra-node AI / ML functional entities and / or cross-node AI / ML coordination entities can be deployed simultaneously.

[0706] After sending the capability information to the communication node 2, the communication node 1 further includes receiving instruction information sent by the communication node 2, where the instruction information is used to configure or activate at least one intra-node AI / ML functional entity and / or cross-node AI / ML coordination entity.

[0707] The communication node 1 or the communication node 2 is any one of the following communication nodes:

[0708] Management equipment, such as OAM equipment;

[0709] Core network equipment, such as at least one of AMF, SMF, UPF, UDM, PCF, and LMF;

[0710] Access network equipment, such as at least one of the following: gNB, CU, DU, CU-CP, CU-UP, and protocol layer entities within the access network (including at least one of NAS, SDAP, RRC, PDCP, RLC, BAP, MAC, and PHY);

[0711] Terminal equipment, such as the UE device as a whole and / or the protocol layer entities within the terminal device (including at least one of NAS, SDAP, RRC, PDCP, RLC, BAP, MAC, and PHY).

[0712] The embodiments of the present application propose a design method for a communication system architecture, through which the following beneficial effects can be produced: AI / ML functions can be flexibly deployed at any communication node in the communication system, so that the AI / ML functions have good scalability in the communication system, and can serve more potential usage scenarios than the architecture of centralized plug-in AI / ML function modules; the use of AI / ML functions can solve communication problems in real time or near real time, and compared with the existing communication system's "after-the-fact solution" mode, it can further achieve "pre-solution" and "in-process solution" problems, thereby improving the robustness of the communication system and the timeliness of problem solving; a unified AI / ML function architecture is conducive to realizing functions such as global data sharing and task sharing in the communication system, thereby improving the efficiency of using AI / ML functions in the communication system.

[0713] The present application proposes a design method for a communication system architecture that enables the generalized application of AI / ML functions in communication systems, namely, an endogenously intelligent communication system. The method includes flexible deployment of AI / ML functional entities at any communication node in the communication system, flexible connection relationships between AI / ML functional entities, definition of general and / or external functions of AI / ML functional entities, definition of initialization, activation, deactivation, or deletion operations for AI / ML functional entities, and definition of AI / ML function-related capability interaction behaviors between communication nodes.

[0714] The following is an example of the tasks that can be performed by the target AI / ML entity in each device:

[0715] The AI / ML algorithms or AI / ML models corresponding to one or more target AI / ML entities in the management device can be applied to energy-saving scenarios. For example, the AI / ML algorithms or AI / ML models corresponding to the management device can be fed with historical energy consumption information recorded by each access network device over a period of time, and the AI / ML algorithms or AI / ML models corresponding to the management device can output energy-saving strategies for the future period of time.

[0716] The AI / ML algorithms or AI / ML models corresponding to one or more target AI / ML entities in the core network device can be applied to load balancing scenarios. For example, the AI / ML algorithm or AI / ML model corresponding to the core network device can be input with the following information: the current load information of each AMF entity and / or the historical load information recorded by each AMF entity over a period of time, and output to the AI / ML algorithm or AI / ML model corresponding to the core network device: the load information and / or load distribution strategy of each AMF entity over a period of time in the future.

[0717] The AI / ML algorithms or AI / ML models corresponding to one or more target AI / ML entities in the access network device can be applied to random access scenarios. For example:

[0718] At least one of the following can be input to the AI / ML algorithm or AI / ML model corresponding to the access network device: terminal device geographic location information; serving cell measurement results; measurement results of at least one neighboring cell; random access channel busyness assessment results; channel interference assessment results; and historical random access reports. Optionally, the random access channel busyness assessment results can be used to assess the random access channel load level, thereby assisting the terminal device in deciding whether to initiate a random access attempt. Generally speaking, a terminal device is more likely to initiate a random access attempt when the random access channel load is light.

[0719] The AI / ML algorithm or AI / ML model corresponding to the access network device can output at least one of the following: the desired random access configuration; the updated AI algorithm; the AI ​​algorithm input parameter modification strategy; the AI ​​algorithm output parameter modification strategy; and the random access configuration selection strategy.

[0720] Optionally, the same input parameters may produce completely different results for different AI algorithms. In order to achieve relatively better optimization goals, the terminal device or network device may need to update the algorithm itself based on historical calculations. For example, the updated AI algorithm may include the order in which each input parameter is used in the algorithm, the weight ratio of each parameter when used in the algorithm, and other factors.

[0721] The AI / ML algorithms or AI / ML models corresponding to one or more target AI / ML entities in the terminal device can be applied to cell selection / reselection scenarios. For example:

[0722] At least one of the following items can be input into the AI / ML algorithm or AI / ML model corresponding to the terminal device: user's desired destination information; user's desired service type information; user's desired slice type information; terminal device's geographic location information; serving cell measurement results; at least one neighboring cell measurement result; historical cell selection and reselection data of the terminal device; channel interference assessment results; Minimization of Drive Test (MDT) record report; and cell deployment related information.

[0723] Optionally, the above-mentioned cell deployment related information can be used to provide basic information of cells within a certain area, including at least one of the following information: area identification information, geographic coordinate information of each cell in the area, frequency resource related information used by each cell in the area, physical cell identification (PCI) information used by each cell in the area, global cell identification (CGI) information used by each cell in the area, coverage information of each cell in the area, historical load information of each cell in the area, service types supported by each cell in the area, and slice type information supported by each cell in the area.

[0724] Optionally, the above-mentioned cell deployment related information can be provided by the network device through public signaling and / or dedicated signaling; or provided by the second terminal device through at least one of unicast signaling, multicast signaling and broadcast signaling.

[0725] The AI / ML algorithm or AI / ML model corresponding to the terminal device can output at least one of the following: desired cell selection and reselection path information; updated AI algorithm; AI algorithm input parameter modification strategy; AI algorithm output parameter modification strategy; decision information for determining the target cell during cell selection and reselection.

[0726] Optionally, the application scenario identification information corresponding to the AI / ML algorithm or AI / ML model may indicate: the application scenario of the AI / ML algorithm or AI / ML model.

[0727] It should be noted that in any embodiment of the present application, the description of the target AI / ML entity, for example, the description of at least one of the intra-node AI / ML entity, the cross-node AI / ML entity, the central entity, the sub-entity, and the specific entity, can be applied to at least one of the management device, the core network device, the access network device, and the terminal device, unless otherwise specified.

[0728] FIG14 is a flow chart of a communication method provided in an embodiment of the present application. As shown in FIG14 , the communication method may include:

[0729] S1401. The communication device executes an AI / ML-related task corresponding to the communication device.

[0730] Optionally, the AI / ML-related tasks corresponding to the communication device include at least one of the following:

[0731] AI / ML related tasks generated by the communication device;

[0732] AI / ML-related tasks requested by a target AI / ML entity deployed on a device other than the communication device;

[0733] AI / ML-related tasks sent to a target AI / ML entity deployed on a device other than the communication device;

[0734] Training, reasoning, or monitoring the AI / ML model in the communication device.

[0735] Optionally, the communication device may be one of the following: a management device, a core network device, an access network device, or a terminal device.

[0736] Optionally, the AI / ML related tasks include at least one of the following: data management tasks; storage management tasks; computing power management tasks; model management tasks.

[0737] Optionally, the task includes at least one of the following:

[0738] Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0739] Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0740] Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery;

[0741] Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

[0742] FIG15 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application. As shown in FIG15 , the communication device 1500 may include:

[0743] The execution unit 1501 is configured to execute AI / ML related tasks corresponding to the communication device.

[0744] Optionally, the AI / ML-related tasks corresponding to the communication device include at least one of the following:

[0745] AI / ML related tasks generated by the communication device;

[0746] AI / ML-related tasks requested by a target AI / ML entity deployed on a device other than the communication device;

[0747] AI / ML-related tasks sent to a target AI / ML entity deployed on a device other than the communication device;

[0748] Training, reasoning, or monitoring the AI / ML model in the communication device.

[0749] Optionally, the communication device may include a communication unit, which is used to: receive a task requested by a target AI / ML entity deployed by a device outside the communication device, and / or send a task to a target AI / ML entity deployed by a device outside the communication device.

[0750] Optionally, the communication device may include a training unit, and the training unit is used to train the AI / ML model in the communication device.

[0751] Optionally, the communication device may include an inference unit, and the application unit is used to: perform inference on the AI / ML model in the communication device.

[0752] Optionally, the communication device may include a monitoring unit, and the application unit is used to monitor the AI / ML model in the communication device.

[0753] Optionally, the communication device can be applied to one of the following: management equipment, core network equipment, access network equipment, and terminal equipment.

[0754] Optionally, the task includes at least one of the following:

[0755] Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding;

[0756] Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting;

[0757] Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery;

[0758] Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

[0759] Figure 16 is a schematic structural diagram of a communication device provided in an embodiment of the present application. The communication device 1600 may include one of the following: a management device, a core network device, an access network device, or a terminal device. The communication device 1600 shown in Figure 16 may include a processor 1610 and a memory 1620. The memory 1620 stores a computer program executable on the processor 1610. When the processor 1610 executes the program, the communication method in any of the above embodiments is implemented.

[0760] Optionally, the memory 1620 may be a separate device independent of the processor 1610 , or may be integrated into the processor 1610 .

[0761] In some embodiments, as shown in FIG16 , the communication device 1600 may further include a transceiver 1630 , and the processor 1610 may control the transceiver 1630 to communicate with other devices. Specifically, the transceiver 1630 may send information or data to other devices, or receive information or data sent by other devices.

[0762] The transceiver 1630 may include a transmitter and a receiver. The transceiver 1630 may further include an antenna, and the number of antennas may be one or more.

[0763] An embodiment of the present application further provides a computer storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the communication method in any embodiment of the present application.

[0764] In some embodiments, the computer-readable storage medium can be applied to the management device, core network device, access network device, and terminal device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the management device, core network device, access network device, and terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0765] Figure 17 is a schematic structural diagram of a chip according to an embodiment of the present application. The chip 1700 shown in Figure 17 includes a processor 1710. The processor 1710 is used to call and run a computer program from a memory to implement the method in any embodiment of the present application.

[0766] In some embodiments, as shown in FIG17 , the chip 1700 may further include a memory 1720. The processor 1710 may call and execute a computer program from the memory 1720 to implement the method in the embodiment of the present application.

[0767] The memory 1720 may be a separate device independent of the processor 1710 , or may be integrated into the processor 1710 .

[0768] In some embodiments, the chip 1700 may further include an input interface 1730. The processor 1710 may control the input interface 1730 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

[0769] In some embodiments, the chip 1700 may further include an output interface 1740. The processor 1710 may control the output interface 1740 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

[0770] In some embodiments, the chip can be applied to at least one of the management device, core network device, access network device, and terminal device in the embodiments of the present application, and the chip can implement the corresponding processes in the various methods of the embodiments of the present application implemented by at least one of the management device, core network device, access network device, and terminal device. For the sake of brevity, they will not be repeated here.

[0771] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0772] An embodiment of the present application also provides a computer program product, which includes a computer storage medium, the computer storage medium storing a computer program, and the computer program including instructions that can be executed by at least one processor. When the instructions are executed by the at least one processor, the communication method in any embodiment of the present application is implemented.

[0773] In some embodiments, the computer program product can be applied to the management device, core network device, access network device, and terminal device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the management device, core network device, access network device, and terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0774] Optionally, the computer program product in the embodiments of the present application may also be referred to as a software product in other embodiments.

[0775] An embodiment of the present application further provides a computer program, which enables a computer to execute the communication method in any embodiment of the present application.

[0776] In some embodiments, the computer program can be applied to the management device, core network device, access network device, and terminal device in the embodiments of the present application. When the computer program runs on a computer, the computer executes the corresponding processes implemented by the management device, core network device, access network device, and terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0777] Those skilled in the art should understand that the relevant descriptions of the above-mentioned communication methods, communication devices, communication equipment, computer storage media, chips, computer program products and computer programs of the embodiments of the present application can be understood by referring to the relevant descriptions of the management equipment, core network equipment, access network equipment and terminal equipment of the embodiments of the present application.

[0778] The preferred embodiments of the present application are described in detail above in conjunction with the accompanying drawings. However, the present application is not limited to the specific details in the above embodiments. Within the technical concept of the present application, the technical solution of the present application can be subjected to a variety of simple modifications, and these simple modifications all fall within the scope of protection of the present application. For example, the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present application will no longer describe the various possible combinations separately. For another example, the various different embodiments of the present application can also be arbitrarily combined, as long as they do not violate the idea of ​​the present application, they should also be regarded as the contents disclosed in the present application. For another example, under the premise of no conflict, the various embodiments and / or the technical features in each embodiment described in the present application can be arbitrarily combined with the prior art, and the technical solution obtained after the combination should also fall within the scope of protection of the present application.

[0779] It should also be understood that in the embodiments of this application, the term "and / or" simply describes the relationship between associated objects, indicating that three possible relationships exist. Specifically, A and / or B can represent: A exists alone, A and B exists simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0780] Each target AI / ML entity, processor, communication device or chip of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of the hardware in the processor or an instruction in the form of software. The above-mentioned processor, communication device or chip may include any one or more integrations of the following: general-purpose processor, application-specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), central processing unit (CPU), graphics processing unit (GPU), embedded neural network processor (neural-network processing units, NPU), controller, microcontroller, microprocessor, programmable logic device, discrete gate or transistor logic device, discrete hardware component. The disclosed methods, steps and logic block diagrams in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0781] It is understood that the memory or computer storage medium in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a 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 dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory. It should be understood that the above-mentioned memory or computer storage medium is exemplary but not restrictive. For example, the memory in the embodiments of the present application may also be 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 dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memory in the embodiments of the present application is intended to include, but is not limited to, these and any other suitable types of memory.

[0782] Those skilled in the art will appreciate that the units 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 beyond the scope of this application.

[0783] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0784] In the several embodiments provided in this 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 merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another 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 units, which can be electrical, mechanical or other forms.

[0785] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0786] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0787] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0788] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A management device, comprising at least one target AI / ML entity; each target AI / ML entity is configured to perform an AI / ML-related task corresponding to the management device.

2. The management device according to claim 1, wherein the at least one target AI / ML entity comprises at least one intra-node AI / ML entity and / or at least one cross-node AI / ML entity; each cross-node AI / ML entity or each intra-node AI / ML entity is configured to process at least one of the following: tasks requested by non-target AI / ML entities in the management device; The tasks generated by the cross-node AI / ML entity or the intra-node AI / ML entity; Tasks requested by other cross-node AI / ML entities deployed in the management device; Tasks requested by AI / ML entities in other nodes deployed in the management device; Tasks requested by the target AI / ML entity deployed on devices other than the management device.

3. The management device according to claim 1 or 2, wherein any two cross-node AI / ML entities deployed in the management device communicate directly via an interface protocol; or The management device includes at least one central entity, and any two cross-node AI / ML entities deployed in the management device communicate indirectly through one or more of the central entities; or Any intra-node AI / ML entity deployed in the management device communicates directly with the cross-node AI / ML entity via an interface protocol; or The management device includes at least one central entity, and any intra-node AI / ML entity deployed in the management device communicates indirectly with the cross-node AI / ML entity through one or more of the central entities; or Direct communication between AI / ML entities in any two nodes deployed in the management device via an interface protocol; or The management device includes at least one central entity, and the AI / ML entities in any two nodes deployed in the management device communicate indirectly through one or more of the central entities.

4. The management device according to any one of claims 1 to 3, wherein the cross-node AI / ML entity deployed in the management device directly communicates with the target AI / ML entity deployed on a device outside the management device via an interface protocol; and / or The cross-node AI / ML entity deployed in the management device indirectly communicates with the target AI / ML entity deployed in a device other than the management device through the cross-node AI / ML entity deployed in a device other than the management device; and / or The AI / ML entity in the node deployed in the management device directly communicates with the target AI / ML entity deployed in a device outside the management device via an interface protocol; and / or, The intra-node AI / ML entity deployed in the management device communicates indirectly with the target AI / ML entity deployed on a device outside the management device through the cross-node AI / ML entity deployed on a device outside the management device.

5. The management device according to any one of claims 1 to 4, wherein the at least one target AI / ML entity comprises: At least one central entity and at least one sub-entity, each of the central entities is associated with one or more sub-entities; Wherein, any of the central entities is used for at least one of the following: Processing tasks requested by non-target AI / ML entities in the management device; Processing tasks requested by other central entities deployed in the management device; Processing tasks requested by sub-entities deployed in the management device; Forwarding tasks triggered or responded to by a target AI / ML entity deployed on a device other than the management device to one or more corresponding sub-entities; Forward tasks triggered or responded to by one or more sub-entities to the corresponding target AI / ML entity deployed on a device other than the management device.

6. The management device according to claim 5, wherein any one of the central entities is used for at least one of the following: Forwarding a task sent by an AI / ML entity in a node deployed by a device other than the management device to one or more sub-entities in the management device associated with the AI / ML entity in the node that sent the task; Forwarding tasks sent by a cross-node AI / ML entity deployed on a device other than the management device to one or more sub-entities in the management device associated with the cross-node AI / ML entity that sent the task; Forwarding tasks sent by AI / ML entities in nodes deployed by devices other than the management device to one or more sub-entities associated with the task type in the management device; Forward tasks sent by cross-node AI / ML entities deployed on devices other than the management device to one or more sub-entities associated with the task type in the management device.

7. The management device according to claim 5 or 6, wherein the central entity deployed in the management device directly communicates with the target AI / ML entity deployed on a device outside the management device through an interface protocol, and / or indirectly communicates with the target AI / ML entity deployed on a device outside the management device through a cross-node AI / ML entity deployed on a device outside the management device.

8. The management device according to any one of claims 5 to 7, wherein any two central entities deployed in the management device communicate directly with each other via an interface protocol; or Any two central entities deployed in the management device communicate indirectly through a specific entity; or, A part of the central entities deployed in the management device communicate directly with each other through an interface protocol, and another part of the central entities communicate indirectly with each other through specific entities.

9. The management device according to any one of claims 1 to 8, wherein any one intra-node AI / ML entity deployed in the management device has or does not have a communication connection with one or more cross-node AI / ML entities deployed in the management device; and / or The AI / ML entity in any node deployed in the management device has or does not have a communication connection with one or more central entities deployed in the management device; and / or The AI / ML entity in any node deployed in the management device has or does not have a communication connection with one or more sub-entities deployed in the management device; and / or, Any cross-node AI / ML entity deployed in the management device has or does not have a communication connection with one or more central entities deployed in the management device; and / or Any cross-node AI / ML entity deployed in the management device may or may not have a communication connection with one or more sub-entities deployed in the management device.

10. The management device according to any one of claims 1 to 9, wherein the intra-node AI / ML entity, cross-node AI / ML entity, central entity, sub-entity, or specific entity deployed in the management device comprises at least one of the following: A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling; A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching; The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

11. The management device according to any one of claims 1 to 10, wherein the node AI / ML entity deployed in the management device is used for at least one of the following: Sending, assigning, or forwarding a task to a first designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the first designated AI / ML entity; Forwarding tasks sent, assigned, or forwarded by the first designated AI / ML entity; in, The first designated AI / ML entity includes at least one of the following: other in-node AI / ML entities deployed in the management device, cross-node AI / ML entities deployed in the management device, central entities deployed in the management device, sub-entities deployed in the management device, in-node AI / ML entities deployed on devices other than the management device, cross-node AI / ML entities deployed on devices other than the management device, central entities deployed on devices other than the management device, and sub-entities deployed on devices other than the management device.

12. The management device according to any one of claims 1 to 11, wherein the cross-node AI / ML entity deployed in the management device is used for at least one of the following: Sending, assigning, or forwarding a task to a second designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the second designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by a second designated AI / ML entity; in, The second designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the management device, other cross-node AI / ML entities deployed in the management device, a central entity deployed in the management device, a sub-entity deployed in the management device, an intra-node AI / ML entity deployed on a device other than the management device, a cross-node AI / ML entity deployed on a device other than the management device, a central entity deployed on a device other than the management device, and a sub-entity deployed on a device other than the management device.

13. The management device according to any one of claims 1 to 12, wherein the central entity deployed in the management device is used for at least one of the following: Sending, assigning, or forwarding tasks to a third designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by a third designated AI / ML entity; Forwarding tasks sent, assigned, or forwarded by a third designated AI / ML entity; in, The third designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the management device, a cross-node AI / ML entity deployed in the management device, other central entities deployed in the management device, a sub-entity deployed in the management device, an intra-node AI / ML entity deployed on a device other than the management device, a cross-node AI / ML entity deployed on a device other than the management device, a central entity deployed on a device other than the management device, and a sub-entity deployed on a device other than the management device.

14. The management device according to any one of claims 1 to 13, wherein the sub-entity deployed in the management device is used for at least one of the following: Sending, assigning, or forwarding tasks to a fourth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the fourth designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by a fourth designated AI / ML entity; in, The fourth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the management device, a cross-node AI / ML entity deployed in the management device, a central entity deployed in the management device, other sub-entities deployed in the management device, an intra-node AI / ML entity deployed on a device other than the management device, a cross-node AI / ML entity deployed on a device other than the management device, a central entity deployed on a device other than the management device, and a sub-entity deployed on a device other than the management device.

15. The management device according to claim 1, wherein the specific entity deployed in the management device is used for at least one of the following: Sending, assigning or forwarding tasks to a central entity deployed in the management device; Receive and process tasks sent, assigned or forwarded by a central entity deployed in the management device; Forwarding tasks sent, assigned or forwarded by a central entity deployed in the management device.

16. The management device according to any one of claims 1 to 15, wherein the task comprises at least one of the following: Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery; Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

17. The management device according to any one of claims 1 to 16, wherein a newly joined or newly activated target AI / ML entity in the management device sends an initialization message to at least one deployed AI / ML entity; in, The initialization message includes at least one of the following of the newly joined or newly activated target AI / ML entity: communication address information, identification information, supported function range information, and deployment location information; The at least one deployed AI / ML entity includes: one or more target AI / ML entities deployed in the management device, and / or one or more target AI / ML entities deployed in a device other than the management device.

18. The management device according to claim 17, wherein any of the deployed AI / ML entities sends a first response message or a second response message to the newly joined or newly activated target AI / ML entity; in, The first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity; the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

19. The management device according to claim 18, wherein the first response message includes at least one of the following of any of the deployed AI / ML entities: communication address information, identification information, supported function scope information, and deployment location information; The second response message includes: Information about the reason for the rejection and / or the duration of the rejection.

20. The management device according to any one of claims 1 to 19, wherein any target AI / ML entity deployed in the management device sends a first notification message to at least one fifth designated AI / ML entity; the first notification message is used to instruct the any target AI / ML entity to perform a deactivation operation or a deletion operation on itself; When any one of the target AI / ML entities receives a confirmation message sent by each of the fifth designated AI / ML entities, the any one of the target AI / ML entities performs a deactivation operation or a deletion operation on itself; in, The fifth designated AI / ML entity includes: other target AI / ML entities deployed in the management device that are in communication connection with any target AI / ML entity, and / or target AI / ML entities deployed in a device other than the management device that are in communication connection with any target AI / ML entity.

21. The management device according to any one of claims 1 to 20, wherein any target AI / ML entity deployed in the management device receives a second notification message sent by a fifth designated AI / ML entity; the second notification message is used to instruct the any target AI / ML entity to be deactivated or deleted; Any target AI / ML entity performs a deactivation operation or a deletion operation on itself based on the second notification message; in, The fifth designated AI / ML entity includes: other target AI / ML entities deployed in the management device that are in communication connection with any target AI / ML entity, and / or target AI / ML entities deployed in a device other than the management device that are in communication connection with any target AI / ML entity.

22. The management device according to any one of claims 1 to 21, wherein the management device can send capability information to a communication node deployed by a device other than the management device, or the management device can receive capability information sent by a communication node deployed by a device other than the management device; in, The capability information indicates at least one of the following: Whether to support deployment of AI / ML entities within the node; Whether to support deployment of cross-node AI / ML entities; The maximum number of AI / ML entities supported within a node; The maximum number of supported cross-node AI / ML entities; The maximum number of sub-entities supported; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the nodes can be deployed; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the node cannot be deployed; Communication nodes and / or management devices where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously; Communication nodes and / or management devices where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously; Among them, the communication nodes deployed in devices other than the management device include at least one of the following: communication nodes in core network devices, communication nodes in access network devices, target units in access network devices, protocol layer entities in access network devices, terminal devices, and protocol layer entities in terminal devices.

23. A core network device, comprising at least one target AI / ML entity; each of the target AI / ML entities is configured to perform an AI / ML-related task corresponding to the core network device.

24. The core network device according to claim 23, wherein the target AI / ML entity comprises an intra-node AI / ML entity; and the core network device further comprises at least one communication node; One or more intra-node AI / ML entities are deployed in each of the communication nodes; The AI / ML entity in each node is used to perform AI / ML related tasks of the corresponding communication node.

25. The core network device according to claim 23 or 24, wherein the target AI / ML entity comprises a cross-node AI / ML entity; the core network device further comprises at least one communication node; Each of the cross-node AI / ML entities is communicatively connected to one or more of the communication nodes; Each of the cross-node AI / ML entities is configured to execute AI / ML-related tasks corresponding to one or more of the communication nodes.

26. The core network device according to any one of claims 23 to 25, further comprising at least one communication node; wherein the at least one target AI / ML entity comprises: at least one intra-node AI / ML entity and at least one cross-node AI / ML entity; Any of the communication nodes includes one or more intra-node AI / ML entities; each intra-node AI / ML entity is used to perform AI / ML related tasks of the corresponding communication node; Each of the cross-node AI / ML entities is in communication with one or more of the intra-node AI / ML entities; each of the cross-node AI / ML entities is configured to perform AI / ML-related tasks corresponding to one or more of the communication nodes.

27. The core network device according to claim 24 or 26, wherein the AI / ML entities deployed in any two communication nodes in the core network device communicate directly via an interface protocol; or Intra-node AI / ML entities deployed in any two communication nodes in the core network device communicate indirectly via one or more cross-node AI / ML entities; or The intra-node AI / ML entities deployed in some of the communication nodes in the core network device communicate directly with each other through an interface protocol, and the intra-node AI / ML entities deployed in another part of the communication nodes communicate indirectly with each other through one or more cross-node AI / ML entities.

28. The core network device according to claim 25 or 26, wherein any two cross-node AI / ML entities in the core network device communicate directly with each other via an interface protocol; or The core network device includes at least one central entity, and any two cross-node AI / ML entities deployed in the core network device communicate indirectly through one or more of the central entities; or The core network device includes at least one central entity, and some of the cross-node AI / ML entities in the core network device communicate directly with each other through an interface protocol, and another part of the cross-node AI / ML entities communicate indirectly with each other through one or more central entities.

29. The core network device according to any one of claims 24 to 28, wherein a cross-node AI / ML entity deployed in the core network device directly communicates with a target AI / ML entity deployed on a device outside the core network device via an interface protocol; and / or The cross-node AI / ML entity deployed in the core network device indirectly communicates with the target AI / ML entity deployed in the device outside the core network device through the cross-node AI / ML entity deployed in the device outside the core network device; and / or The AI / ML entity in the node deployed in the core network device directly communicates with the target AI / ML entity deployed in a device outside the core network device via an interface protocol; and / or, The intra-node AI / ML entity deployed in the core network device indirectly communicates with the target AI / ML entity deployed on a device outside the core network device through the cross-node AI / ML entity deployed in the core network device and / or the cross-node AI / ML entity deployed on a device outside the core network device.

30. The core network device according to any one of claims 24 to 29, wherein each of the cross-node AI / ML entities or each of the intra-node AI / ML entities is configured to process at least one of the following: a task requested by at least one communication node included in the core network device; The tasks generated by the cross-node AI / ML entity or the intra-node AI / ML entity; Tasks requested by other cross-node AI / ML entities deployed in the core network devices; Tasks requested by AI / ML entities in other nodes deployed in the core network device; Tasks requested by the target AI / ML entity deployed on devices other than the core network device.

31. The core network device according to any one of claims 23 to 30, wherein the at least one target AI / ML entity comprises: At least one central entity and at least one sub-entity, each of the central entities is associated with one or more sub-entities; Wherein, any of the central entities is used for at least one of the following: Processing a task requested by at least one communication node included in the core network device; Processing tasks requested by other central entities deployed in the core network equipment; Processing tasks requested by sub-entities deployed in the core network device; Forwarding tasks triggered or responded to by an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the core network device to one or more sub-entities deployed in the core network device, or to a target AI / ML entity deployed in a device outside the core network device; Forward tasks triggered or responded to by one or more sub-entities deployed in the core network device, or by a target AI / ML entity deployed on a device outside the core network device, to an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the core network device.

32. The core network device according to claim 31, wherein any one of the central entities is used for at least one of the following: Forwarding a task triggered or responded to by an intra-node AI / ML entity deployed in the core network device to one or more sub-entities in the core network device associated with the intra-node AI / ML entity that triggered or responded to the task; Forwarding a task triggered or responded to by a cross-node AI / ML entity deployed in the core network device to one or more sub-entities in the core network device associated with the cross-node AI / ML entity that triggered or responded to the task; Forwarding tasks triggered or responded to by AI / ML entities in nodes deployed in the core network device to one or more sub-entities associated with the task type in the core network device; Forwarding tasks triggered or responded to by the cross-node AI / ML entity deployed in the core network device to one or more sub-entities associated with the task type in the core network device; Forwarding a task sent by an AI / ML entity in a node deployed on a device other than the core network device to one or more sub-entities in the core network device associated with the AI / ML entity in the node that sent the task; Forwarding a task sent by a cross-node AI / ML entity deployed on a device other than the core network device to one or more sub-entities in the core network device associated with the cross-node AI / ML entity that sent the task; Forwarding tasks sent by the AI / ML entity in the node deployed by the device outside the core network device to one or more sub-entities associated with the task type in the core network device; Forwarding tasks sent by a cross-node AI / ML entity deployed on a device other than the core network device to one or more sub-entities associated with the task type in the core network device.

33. According to the core network device of claim 31 or 32, the central entity deployed in the core network device directly communicates with the target AI / ML entity deployed on a device outside the core network device through an interface protocol, and / or indirectly communicates with the target AI / ML entity deployed on a device outside the core network device through a cross-node AI / ML entity deployed on a device outside the core network device.

34. The core network device according to any one of claims 31 to 33, wherein any two central entities deployed in the core network device communicate directly with each other via an interface protocol; or Any two central entities deployed in the core network device communicate indirectly through a specific entity; or, A part of the central entities deployed in the core network device communicate directly with each other through an interface protocol, and another part of the central entities communicate indirectly with each other through a specific entity.

35. The core network device according to any one of claims 23 to 34, wherein the intra-node AI / ML entity, cross-node AI / ML entity, central entity, sub-entity, or specific entity deployed in the core network device comprises at least one of the following: A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling; A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching; The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

36. The core network device according to any one of claims 23 to 35, wherein the intra-node AI / ML entity deployed in the core network device is used for at least one of the following: Sending, assigning, or forwarding tasks to a sixth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the sixth designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by the sixth designated AI / ML entity; in, The sixth designated AI / ML entity includes at least one of the following: other node AI / ML entities deployed in the core network device, cross-node AI / ML entities deployed in the core network device, central entities deployed in the core network device, sub-entities deployed in the core network device, node AI / ML entities deployed by devices other than the core network device, cross-node AI / ML entities deployed by devices other than the core network device, central entities deployed by devices other than the core network device, and sub-entities deployed by devices other than the core network device.

37. The core network device according to any one of claims 23 to 36, wherein the cross-node AI / ML entity deployed in the core network device is used for at least one of the following: Sending, assigning, or forwarding tasks to a seventh designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the seventh designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by the seventh designated AI / ML entity; in, The seventh designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the core network device, other cross-node AI / ML entities deployed in the core network device, a central entity deployed in the core network device, a sub-entity deployed in the core network device, an intra-node AI / ML entity deployed on a device other than the core network device, a cross-node AI / ML entity deployed on a device other than the core network device, a central entity deployed on a device other than the core network device, and a sub-entity deployed on a device other than the core network device.

38. The core network device according to any one of claims 23 to 37, wherein the central entity deployed in the core network device is used for at least one of the following: Sending, assigning, or forwarding tasks to an eighth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the eighth designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by the eighth designated AI / ML entity; in, The eighth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the core network device, a cross-node AI / ML entity deployed in the core network device, other central entities deployed in the core network device, a sub-entity deployed in the core network device, an intra-node AI / ML entity deployed on a device other than the core network device, a cross-node AI / ML entity deployed on a device other than the core network device, a central entity deployed on a device other than the core network device, and a sub-entity deployed on a device other than the core network device.

39. The core network device according to any one of claims 23 to 38, wherein the sub-entity deployed in the core network device is used for at least one of the following: Sending, assigning, or forwarding tasks to a ninth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the ninth designated AI / ML entity; Forwarding tasks sent, assigned, or forwarded by the ninth designated AI / ML entity; in, The ninth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the core network device, a cross-node AI / ML entity deployed in the core network device, a central entity deployed in the core network device, other sub-entities deployed in the core network device, an intra-node AI / ML entity deployed by a device other than the core network device, a cross-node AI / ML entity deployed by a device other than the core network device, a central entity deployed by a device other than the core network device, and a sub-entity deployed by a device other than the core network device.

40. The core network device according to claims 23-39, wherein the specific entity deployed in the core network device is used for at least one of the following: Sending tasks, assigning tasks, or forwarding tasks to a central entity deployed in the core network device; Receive and process tasks sent, assigned, or forwarded by a central entity deployed in the core network device; Forwarding tasks sent, assigned or forwarded by a central entity deployed in the core network device.

41. The core network device according to any one of claims 23 to 40, wherein the task comprises at least one of the following: Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery; Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

42. The core network device according to any one of claims 23-41, wherein a newly joined or newly activated target AI / ML entity in the core network device sends an initialization message to at least one deployed AI / ML entity; in, The initialization message includes at least one of the following of the newly joined or newly activated target AI / ML entity: communication address information, identification information, supported function range information, and deployment location information; The at least one deployed AI / ML entity includes: one or more target AI / ML entities deployed in the core network device, and / or one or more target AI / ML entities deployed in a device other than the core network device.

43. The core network device according to claim 42, wherein any of the deployed AI / ML entities sends a first response message or a second response message to the newly joined or newly activated target AI / ML entity; in, The first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity; the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

44. The core network device according to claim 43, wherein the first response message includes at least one of the following of any of the deployed AI / ML entities: communication address information, identification information, supported function scope information, and deployment location information; The second response message includes: Information about the reason for the rejection and / or the duration of the rejection.

45. The core network device according to any one of claims 23-44, wherein any target AI / ML entity deployed in the core network device sends a first notification message to at least one tenth designated AI / ML entity; the first notification message is used to instruct the any target AI / ML entity to perform a deactivation operation or a deletion operation on itself; When any one of the target AI / ML entities receives a confirmation message sent by each of the tenth designated AI / ML entities, the any one of the target AI / ML entities performs a deactivation operation or a deletion operation on itself; in, The tenth designated AI / ML entity includes: other target AI / ML entities deployed in the core network device that have a communication connection with any target AI / ML entity, and / or target AI / ML entities deployed in a device other than the core network device that have a communication connection with any target AI / ML entity.

46. ​​The core network device according to any one of claims 23 to 45, wherein any target AI / ML entity deployed in the core network device receives a second notification message sent by a tenth designated AI / ML entity; the second notification message is used to instruct the deactivation or deletion of the any target AI / ML entity; Any target AI / ML entity performs a deactivation operation or a deletion operation on itself based on the second notification message; in, The tenth designated AI / ML entity includes: other target AI / ML entities deployed in the core network device that have a communication connection with any target AI / ML entity, and / or target AI / ML entities deployed in devices other than the core network device that have a communication connection with any target AI / ML entity.

47. The core network device according to any one of claims 23 to 46, wherein a communication node deployed in the core network device can send capability information to other communication nodes deployed in the core network device, or a communication node deployed in the core network device can send capability information to a communication node deployed in a device other than the core network device, or a communication node deployed in the core network device can receive capability information sent by a communication node deployed in a device other than the core network device; in, The capability information indicates at least one of the following: Whether to support deployment of AI / ML entities within the node; Whether to support deployment of cross-node AI / ML entities; The maximum number of AI / ML entities supported within a node; The maximum number of supported cross-node AI / ML entities; The maximum number of sub-entities supported; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the nodes can be deployed; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the node cannot be deployed; Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously; Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously; The communication nodes deployed in devices other than the core network device include at least one of the following: a management device, a communication node in an access network device, a target unit in an access network device, a protocol layer entity in an access network device, a terminal device, and a protocol layer entity in a terminal device; The communication nodes deployed in the core network device include at least one of the following: access and mobility management function AMF network element, session management function SMF network element, user plane function UPF network element, unified data management UDM network element, policy control function PCF network element, location management function LMF network element, and authentication server function AUSF network element.

48. An access network device, comprising at least one target AI / ML entity; each of the target AI / ML entities is configured to perform an AI / ML-related task corresponding to the access network device.

49. The access network device according to claim 48, wherein the target AI / ML entity comprises the intra-node AI / ML entity; the access network device further comprises at least one communication node; The intra-node AI / ML entity includes at least one of the following: an access network AI / ML entity, a target unit AI / ML entity, and a protocol layer AI / ML entity; the target unit includes at least one of the following: a centralized unit CU, a distributed unit DU, a centralized unit control plane CU-CP, and a centralized unit user plane CU-UP; One or more access network AI / ML entities are deployed in the access network device, and / or one or more target unit AI / ML entities are deployed in the target unit in the communication node, and / or one or more protocol layer AI / ML entities are deployed in the protocol layer entity in the communication node; in, The access network AI / ML entity is used to perform AI / ML related tasks corresponding to the access network device, the target unit AI / ML entity is used to perform AI / ML related tasks corresponding to the target unit, and the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer.

50. The access network device according to claim 48 or 49, wherein the target AI / ML entity comprises a cross-node AI / ML entity; the access network device further comprises at least one communication node; Each of the cross-node AI / ML entities is communicatively connected to one or more of the communication nodes; Each of the cross-node AI / ML entities is configured to execute AI / ML-related tasks corresponding to one or more of the communication nodes.

51. The access network device according to any one of claims 48 to 50, further comprising at least one communication node; wherein the at least one target AI / ML entity comprises: at least one intra-node AI / ML entity and at least one cross-node AI / ML entity; The intra-node AI / ML entity includes at least one of the following: an access network AI / ML entity, a target unit AI / ML entity, and a protocol layer AI / ML entity; the target unit includes at least one of the following: a centralized unit CU, a distributed unit DU, a centralized unit control plane CU-CP, and a centralized unit user plane CU-UP; One or more of the access network AI / ML entities are deployed in the access network device, and / or one or more of the target unit AI / ML entities are deployed in the target unit in the access network device, and / or one or more of the protocol layer AI / ML entities are deployed in the protocol layer entity in the access network device; The access network AI / ML entity is used to perform AI / ML related tasks corresponding to the access network device, the target unit AI / ML entity is used to perform AI / ML related tasks corresponding to the target unit, and the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer; Each of the cross-node AI / ML entities is in communication with one or more of the intra-node AI / ML entities, and each of the cross-node AI / ML entities is configured to perform AI / ML-related tasks corresponding to one or more of the communication nodes.

52. The access network device according to claim 49 or 51, wherein the intra-node AI / ML entities deployed in any two communication nodes in the access network device communicate directly via an interface protocol; or Intra-node AI / ML entities deployed in any two communication nodes in the access network device communicate indirectly via one or more cross-node AI / ML entities; or The intra-node AI / ML entities deployed in a part of the communication nodes in the access network device communicate directly with each other, and the intra-node AI / ML entities deployed in another part of the communication nodes communicate indirectly with each other through one or more cross-node AI / ML entities.

53. The access network device according to claim 49, 51 or 52, wherein any two target unit AI / ML entities deployed in the access network device communicate with each other via one or more of the access network AI / ML entities; and / or Any two protocol layer AI / ML entities communicate via one or more access network AI / ML entities; and / or, Any two protocol layer AI / ML entities communicate with each other through one or more target unit AI / ML entities.

54. The access network device according to claim 50 or 51, wherein any two cross-node AI / ML entities deployed in the access network device communicate directly with each other via an interface protocol; or The access network device includes at least one central entity, and any two cross-node AI / ML entities deployed in the access network device communicate indirectly through one or more of the central entities; or The access network device includes at least one central entity, and a part of the cross-node AI / ML entities deployed in the access network device communicate directly with each other through an interface protocol, and another part of the cross-node AI / ML entities deployed communicate indirectly with each other through one or more central entities.

55. The access network device according to any one of claims 49 to 54, wherein the cross-node AI / ML entity in the access network device directly communicates with the target AI / ML entity deployed on a device outside the access network device via an interface protocol; and / or The cross-node AI / ML entity deployed in the access network device indirectly communicates with a target AI / ML entity deployed in a device other than the access network device through the cross-node AI / ML entity deployed in a device other than the access network device; and / or The AI / ML entity in the node deployed in the access network device directly communicates with the target AI / ML entity deployed in a device outside the access network device via an interface protocol; and / or, The intra-node AI / ML entity deployed in the access network device indirectly communicates with the target AI / ML entity deployed on a device outside the access network device through the cross-node AI / ML entity deployed in the access network device and / or the cross-node AI / ML entity deployed on a device outside the access network device.

56. The access network device according to any one of claims 49 to 55, wherein each of the cross-node AI / ML entities or each of the intra-node AI / ML entities is configured to process at least one of the following: a task requested by at least one communication node included in the access network device; The tasks generated by the cross-node AI / ML entity or the intra-node AI / ML entity; Tasks requested by other cross-node AI / ML entities deployed in the access network device; Tasks requested by AI / ML entities in other nodes deployed in the access network device; Tasks requested by the target AI / ML entity deployed on devices other than the access network device.

57. The access network device according to any one of claims 48 to 56, wherein the at least one target AI / ML entity comprises: At least one central entity and at least one sub-entity, each of the central entities is associated with one or more sub-entities; Wherein, any of the central entities is used for at least one of the following: Processing the task of requesting at least one communication node included in the access network device Processing tasks requested by other central entities deployed in the access network equipment; Processing tasks requested by sub-entities deployed in the access network device; Forwarding tasks triggered or responded to by an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the access network device to one or more sub-entities deployed in the access network device, or to a target AI / ML entity deployed in a device outside the access network device; Forwarding tasks triggered or responded to by one or more sub-entities deployed in the access network device, or a target AI / ML entity deployed on a device outside the access network device, to an intra-node AI / ML entity or a cross-node AI / ML entity deployed in the access network device.

58. The access network device according to claim 57, wherein any one of the central entities is used for at least one of the following: Forwarding a task triggered or responded to by an intra-node AI / ML entity deployed in the access network device to one or more sub-entities in the access network device associated with the intra-node AI / ML entity that triggered or responded to the task; Forwarding a task triggered or responded to by a cross-node AI / ML entity deployed in the access network device to one or more sub-entities in the access network device associated with the cross-node AI / ML entity that triggered or responded to the task; Forwarding tasks triggered or responded to by an AI / ML entity in a node deployed in the access network device to one or more sub-entities associated with the task type in the access network device; Forwarding tasks triggered or responded to by the cross-node AI / ML entity deployed in the access network device to one or more sub-entities associated with the task type in the access network device; Forwarding a task sent by an AI / ML entity in a node deployed on a device other than the access network device to one or more sub-entities in the access network device associated with the AI / ML entity in the node that sent the task; Forwarding a task sent by a cross-node AI / ML entity deployed on a device other than the access network device to one or more sub-entities in the access network device associated with the cross-node AI / ML entity that sent the task; Forwarding tasks sent by the AI / ML entity in the node deployed by the device outside the access network device to one or more sub-entities associated with the task type in the access network device; Forwarding tasks sent by a cross-node AI / ML entity deployed on a device other than the access network device to one or more sub-entities associated with the task type in the access network device.

59. The access network device according to claim 57 or 58, wherein the central entity deployed in the access network device directly communicates with the target AI / ML entity deployed on a device outside the access network device through an interface protocol, and / or communicates with the target AI / ML entity deployed on a device outside the access network device through a cross-node AI / ML entity deployed on a device outside the access network device.

60. The access network device according to any one of claims 57 to 59, wherein any two central entities deployed in the access network device communicate directly with each other via an interface protocol; or Any two central entities deployed in the access network device communicate indirectly through a specific entity; or, A part of the central entities deployed in the access network device communicate directly with each other through an interface protocol, and another part of the central entities communicate indirectly with each other through specific entities.

61. The access network device according to any one of claims 48 to 60, wherein the intra-node AI / ML entity, cross-node AI / ML entity, central entity, sub-entity, or specific entity deployed in the access network device comprises at least one of the following: A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling; A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching; The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

62. The access network device according to any one of claims 48 to 61, wherein the intra-node AI / ML entity deployed in the access network device is used for at least one of the following: Sending, assigning, or forwarding tasks to the eleventh designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the eleventh designated AI / ML entity; Forwarding tasks sent, assigned, or forwarded by the eleventh designated AI / ML entity; in, The eleventh designated AI / ML entity includes at least one of the following: other intra-node AI / ML entities deployed in the access network device, cross-node AI / ML entities deployed in the access network device, central entities deployed in the access network device, sub-entities deployed in the access network device, intra-node AI / ML entities deployed in devices other than the access network device, cross-node AI / ML entities deployed in devices other than the access network device, central entities deployed in devices other than the access network device, and sub-entities deployed in devices other than the access network device.

63. The access network device according to any one of claims 48 to 62, wherein the cross-node AI / ML entity deployed in the access network device is used for at least one of the following: Sending, assigning, or forwarding tasks to a twelfth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the twelfth designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by the twelfth designated AI / ML entity; in, The twelfth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the access network device, other cross-node AI / ML entities deployed in the access network device, a central entity deployed in the access network device, a sub-entity deployed in the access network device, an intra-node AI / ML entity deployed on a device other than the access network device, a cross-node AI / ML entity deployed on a device other than the access network device, a central entity deployed on a device other than the access network device, and a sub-entity deployed on a device other than the access network device.

64. The access network device according to any one of claims 48 to 63, wherein the central entity deployed in the access network device is used for at least one of the following: Send, assign, or forward tasks to the 13th designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the 13th designated AI / ML entity; Forwarding tasks sent, assigned, or forwarded by the thirteenth designated AI / ML entity; in, The thirteenth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the access network device, a cross-node AI / ML entity deployed in the access network device, other central entities deployed in the access network device, a sub-entity deployed in the access network device, an intra-node AI / ML entity deployed on a device other than the access network device, a cross-node AI / ML entity deployed on a device other than the access network device, a central entity deployed on a device other than the access network device, and a sub-entity deployed on a device other than the access network device.

65. The access network device according to any one of claims 48 to 64, wherein the sub-entity deployed in the access network device is used for at least one of the following: Sending, assigning, or forwarding tasks to the fourteenth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the fourteenth designated AI / ML entity; Forwarding tasks sent, assigned, or forwarded by the fourteenth designated AI / ML entity; in, The fourteenth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in the access network device, a cross-node AI / ML entity deployed in the access network device, a central entity deployed in the access network device, other sub-entities deployed in the access network device, an intra-node AI / ML entity deployed by a device other than the access network device, a cross-node AI / ML entity deployed by a device other than the access network device, a central entity deployed by a device other than the access network device, and a sub-entity deployed by a device other than the access network device.

66. The access network device according to claims 48-65, wherein the specific entity deployed in the access network device is used for at least one of the following: Sending tasks, assigning tasks, or forwarding tasks to a central entity deployed in the access network device; Receive and process tasks sent, assigned or forwarded by a central entity deployed in the access network device; Forwarding tasks sent, assigned or forwarded by a central entity deployed in the access network device.

67. The access network device according to any one of claims 48 to 66, wherein the task comprises at least one of the following: Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery; Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

68. The access network device according to any one of claims 48 to 67, wherein a newly joined or newly activated target AI / ML entity in the access network device sends an initialization message to at least one deployed AI / ML entity; in, The initialization message includes at least one of the following of the newly joined or newly activated target AI / ML entity: communication address information, identification information, supported function range information, and deployment location information; The at least one deployed AI / ML entity includes: one or more target AI / ML entities deployed in the access network device, and / or one or more target AI / ML entities deployed in a device outside the access network device.

69. The access network device according to claim 68, wherein any of the deployed AI / ML entities sends a first response message or a second response message to the newly joined or newly activated target AI / ML entity; in, The first response message is used to indicate acceptance of the newly joined or newly activated target AI / ML entity; the second response message is used to indicate rejection of the newly joined or newly activated target AI / ML entity.

70. The access network device according to claim 69, wherein the first response message includes at least one of the following of any of the deployed AI / ML entities: communication address information, identification information, supported function scope information, and deployment location information; The second response message includes: Information about the reason for the rejection and / or the duration of the rejection.

71. The access network device according to any one of claims 48 to 70, wherein any target AI / ML entity deployed in the access network device sends a first notification message to at least one fifteenth designated AI / ML entity; the first notification message is used to instruct the any target AI / ML entity to perform a deactivation operation or a deletion operation on itself; When any one of the target AI / ML entities receives a confirmation message sent by each of the fifteenth designated AI / ML entities, the any one of the target AI / ML entities performs a deactivation operation or a deletion operation on itself; in, The fifteenth designated AI / ML entity includes: other target AI / ML entities deployed in the access network device that are communicatively connected to any of the target AI / ML entities, and / or target AI / ML entities deployed in devices other than the access network device that are communicatively connected to any of the target AI / ML entities.

72. The access network device according to any one of claims 48 to 71, wherein any target AI / ML entity deployed in the access network device receives a second notification message sent by a fifteenth designated AI / ML entity; the second notification message is used to instruct the deactivation or deletion of the any target AI / ML entity; Any target AI / ML entity performs a deactivation operation or a deletion operation on itself based on the second notification message; in, The fifteenth designated AI / ML entity includes: other target AI / ML entities deployed in the access network device that have a communication connection with any target AI / ML entity, and / or target AI / ML entities deployed in a device other than the access network device that have a communication connection with any target AI / ML entity.

73. The access network device according to any one of claims 48 to 72, wherein a communication node deployed in the access network device can send capability information to other communication nodes deployed in the access network device, or a communication node deployed in the access network device can send capability information to a communication node deployed in a device other than the access network device, or a communication node deployed in the access network device can receive capability information sent by a communication node deployed in a device other than the access network device; in, The capability information indicates at least one of the following: Whether to support deployment of AI / ML entities within the node; Whether to support deployment of cross-node AI / ML entities; The maximum number of AI / ML entities supported within a node; The maximum number of supported cross-node AI / ML entities; The maximum number of sub-entities supported; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the nodes can be deployed; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the node cannot be deployed; Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously; Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously; The communication node deployed in the device other than the access network device includes at least one of the following: a management device, a communication node in a core network device, a terminal device, and a protocol layer entity in a terminal device; The communication nodes deployed in the access network device include at least one of the following: a base station, a centralized unit CU, a distributed unit DU, a centralized unit control plane CU-CP, a centralized unit user plane CU-UP, and a protocol layer entity within the access network, wherein the protocol layer entity within the access network includes at least one of the following: a non-access layer NAS entity, a service data adaptation protocol SDAP entity, a radio resource control RRC entity, a packet data convergence protocol PDCP entity, a radio link control RLC entity, a backhaul adaptation protocol BAP entity, a media access control MAC entity, and a physical layer PHY entity.

74. A terminal device, comprising at least one target AI / ML entity; each of the target AI / ML entities is configured to perform an AI / ML-related task corresponding to the terminal device.

75. The terminal device according to claim 74, wherein the target AI / ML entity comprises the intra-node AI / ML entity; the intra-node AI / ML entity comprises at least one of the following: a terminal device AI / ML entity, a protocol layer AI / ML entity; One or more terminal device AI / ML entities are deployed in the terminal device, and / or one or more protocol layer AI / ML entities are deployed in the protocol layer entity included in the terminal device; in, The terminal device AI / ML entity is used to perform AI / ML related tasks corresponding to the terminal device, and the protocol layer AI / ML entity is used to perform AI / ML related tasks corresponding to the protocol layer.

76. The terminal device according to claim 74 or 75, wherein any two protocol layer AI / ML entities deployed in the terminal device communicate directly via an interface protocol, or communicate indirectly via one or more terminal device AI / ML entities; and / or The terminal device AI / ML entity deployed in the terminal device directly communicates with the access network AI / ML entity deployed in the access network device via an air interface protocol, or directly communicates with the target unit AI / ML entity deployed in the access network device via an air interface protocol; and / or, The protocol layer AI / ML entity deployed in the terminal device communicates directly with the protocol layer AI / ML entity deployed in the access network device through an air interface protocol, or communicates indirectly with the protocol layer AI / ML entity deployed in the access network device through at least one of the terminal device AI / ML entity deployed in the terminal device, the access network AI / ML entity deployed in the access network device, and the target unit AI / ML entity deployed in the access network device; and / or, The terminal device AI / ML entity deployed in the terminal device and / or the protocol layer AI / ML entity deployed in the terminal device communicate indirectly with the intra-node AI / ML entity and / or cross-node AI / ML entity deployed in the core network device through the access network device.

77. The terminal device according to any one of claims 74 to 76, wherein the intra-node AI / ML entity deployed in the terminal device comprises at least one of the following: A data management unit, configured to perform at least one of the following operations: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; A storage management unit, configured to perform at least one of the following operations: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; A computing power management unit, configured to perform at least one of the following operations: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recycling; A model management unit, configured to perform at least one of the following operations: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching; The task management unit is used to perform at least one of the following operations: task generation, task acceptance, task rejection, task splitting, task allocation, task monitoring, task update, and task deletion.

78. The terminal device according to any one of claims 74 to 77, wherein the intra-node AI / ML entity deployed in the terminal device is used for at least one of the following: Sending, assigning, or forwarding tasks to the sixteenth designated AI / ML entity; Receive and process tasks sent, assigned, or forwarded by the sixteenth designated AI / ML entity; forwarding tasks sent, assigned, or forwarded by the sixteenth designated AI / ML entity; in, The sixteenth designated AI / ML entity includes at least one of the following: an intra-node AI / ML entity deployed in an access network device, a cross-node AI / ML entity deployed in an access network device, a central entity deployed in an access network device, a sub-entity deployed in an access network device, an intra-node AI / ML entity deployed in a core network device, a cross-node AI / ML entity deployed in a core network device, an intra-node AI / ML entity deployed in a management device, and a cross-node AI / ML entity deployed in a management device.

79. The terminal device according to any one of claims 74 to 78, wherein the task comprises at least one of the following: Data management tasks, including at least one of the following tasks: data collection, data storage, data modification, data update, data deletion, data replication, and data forwarding; Storage management tasks, the storage management tasks including at least one of the following tasks: remaining storage size indication, storage reservation, storage allocation, storage sharing, storage recycling, and storage formatting; Computing power management tasks, including at least one of the following tasks: remaining computing power indication, computing power reservation, computing power allocation, computing power sharing, and computing power recovery; Model management tasks, wherein the model management tasks include at least one of the following tasks: model training, model verification, model testing, model deployment, model replication, model forwarding, model reasoning, model monitoring, model update, model activation, model deactivation, model deletion, and model switching.

80. The terminal device according to any one of claims 74 to 79, wherein the terminal device receives first indication information sent by an access network device or a core network device, wherein the first indication information is used to indicate at least one of the following: Instructing the terminal device to newly deploy at least one intra-node AI / ML entity; Instructing the terminal device to delete at least one AI / ML entity in the node; instructing the terminal device to activate at least one intra-node AI / ML entity; Instructing the terminal device to deactivate at least one AI / ML entity within the node; Instruct the terminal device to modify at least one AI / ML entity in the node.

81. The terminal device according to any one of claims 74 to 80, wherein the terminal device is further configured to send second indication information to the access network device or the core network device, wherein the second indication information is used for at least one of the following: Indicate at least one intra-node AI / ML entity that the terminal device desires to add; Indicating to the terminal device at least one intra-node AI / ML entity that it desires to delete; Indicating at least one intra-node AI / ML entity that the terminal device desires to activate; Instructing the terminal device to deactivate at least one intra-node AI / ML entity; Indicates at least one intra-node AI / ML entity that has been added to the terminal device; Indicates that the terminal device has deleted at least one intra-node AI / ML entity; Indicates at least one intra-node AI / ML entity that has been activated on the terminal device; Indicates that the terminal device has deactivated at least one intra-node AI / ML entity; Indicates at least one intra-node AI / ML entity that has been modified by the terminal device.

82. The terminal device according to any one of claims 74 to 81, wherein a target AI / ML entity deployed in the terminal device can send capability information to other target AI / ML entities deployed in the terminal device, or the terminal device can send capability information to a communication node deployed in a device other than the terminal device, or the terminal device can receive capability information sent by a communication node deployed in a device other than the terminal device; in, The capability information indicates at least one of the following: Whether to support deployment of AI / ML entities within the node; Whether to support deployment of cross-node AI / ML entities; The maximum number of AI / ML entities supported within a node; The maximum number of supported cross-node AI / ML entities; The maximum number of sub-entities supported; Protocol layer entities that can deploy AI / ML entities in nodes in terminal devices; The protocol layer entity of the AI / ML entity in the terminal device cannot be deployed in the node; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the nodes can be deployed; Target units and / or protocol layer entities in communication nodes where AI / ML entities within the node cannot be deployed; Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities can be deployed simultaneously; Communication nodes where intra-node AI / ML entities and / or cross-node AI / ML entities cannot be deployed simultaneously; The communication node deployed in the device other than the terminal device includes at least one of the following: a management device, a communication node in a core network device, a communication node in an access network device, a target unit in an access network device, and a protocol layer entity in an access network device; The communication node deployed in the terminal device includes at least one of the following: a terminal device, a protocol layer entity within the terminal device, wherein the protocol layer entity within the terminal device includes at least one of the following: a non-access layer NAS entity, a service data adaptation protocol SDAP entity, a radio resource control RRC entity, a packet data convergence protocol PDCP entity, a radio link control RLC entity, a backhaul adaptation protocol BAP entity, a media access control MAC entity, and a physical layer PHY entity.

83. A communication method, the method comprising: The communication device performs an AI / ML-related task corresponding to the communication device; wherein the AI / ML-related task corresponding to the communication device includes at least one of the following: AI / ML related tasks generated by the communication device; AI / ML-related tasks requested by a target AI / ML entity deployed on a device other than the communication device; AI / ML-related tasks sent to a target AI / ML entity deployed on a device other than the communication device; Training, reasoning, or monitoring the AI / ML model in the communication device.

84. A communication device, comprising: an execution unit, configured to execute an AI / ML-related task corresponding to the communication device; wherein the AI / ML-related task corresponding to the communication device includes at least one of the following: AI / ML related tasks generated by the communication device; AI / ML-related tasks requested by a target AI / ML entity deployed on a device other than the communication device; AI / ML-related tasks sent to a target AI / ML entity deployed on a device other than the communication device; Training, reasoning, or monitoring the AI / ML model in the communication device.

85. A communication device comprising: processor and memory, The memory stores a computer program executable on the processor. When the processor executes the program, the method described in claim 83 is implemented.

86. A computer storage medium storing one or more programs, wherein the one or more programs can be executed by one or more processors to implement the method of claim 83.

87. A chip comprising: A processor, configured to call and execute a computer program from a memory to implement the method as claimed in claim 83.

88. A computer program product, comprising a computer storage medium storing a computer program, wherein the computer program comprises instructions executable by at least one processor, and wherein the method of claim 83 is implemented when the instructions are executed by the at least one processor.

89. A computer program, wherein the computer program causes a computer to execute the method according to claim 83.

Images