WO2023246267A1 - 通信方法、通信装置和系统 - Google Patents

通信方法、通信装置和系统 Download PDF

Info

Publication number
WO2023246267A1
WO2023246267A1 PCT/CN2023/089586 CN2023089586W WO2023246267A1 WO 2023246267 A1 WO2023246267 A1 WO 2023246267A1 CN 2023089586 W CN2023089586 W CN 2023089586W WO 2023246267 A1 WO2023246267 A1 WO 2023246267A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
connection
communication device
information
task
Prior art date
Application number
PCT/CN2023/089586
Other languages
English (en)
French (fr)
Inventor
乔云飞
张公正
王坚
李榕
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2023246267A1 publication Critical patent/WO2023246267A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W48/00—Access restriction; Network selection; Access point selection
    • H04W48/16—Discovering, processing access restriction or access information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/12—Wireless traffic scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W76/00—Connection management
    • H04W76/10—Connection setup

Definitions

  • the present application relates to the field of wireless communication, specifically to wireless communication technology using intelligent networks, and in particular to a communication method, communication device and system.
  • AI artificial intelligence
  • the present application provides a communication method, communication device and system.
  • the communication device in the wireless network can be used to perform AI tasks.
  • a communication method is provided, which method can be executed by a communication device.
  • the communication device may be a communication device (such as a terminal device, a network device), or a chip or circuit used in the communication device, which is not limited in this application.
  • a communication device such as a terminal device, a network device
  • a chip or circuit used in the communication device which is not limited in this application.
  • the following description takes a communication device as an example.
  • the method may include: the communication device receives artificial intelligence AI connection information, and the communication device establishes an AI connection with the AI node according to the AI connection information.
  • the AI connection information includes at least one of the following configured for the AI connection: model, data set, and computing resource.
  • a communication method is provided, which method can be executed by a communication device.
  • the communication device may be a communication device (such as a terminal device, a network device), or a chip or circuit used in the communication device, which is not limited in this application.
  • a communication device such as a terminal device, a network device
  • a chip or circuit used in the communication device which is not limited in this application.
  • the following description takes a communication device as an example.
  • the method may include: the communication device receives artificial intelligence AI connection information, the AI connection information includes at least one of the following configured for the AI connection: model, data set, computing resource; the communication device establishes AI with the AI node according to the AI connection information connect.
  • the communication device can receive the AI connection information, and establish an AI connection with the AI node based on the AI connection information, so that the AI node and the communication device can communicate through the AI connection. , such as sending and/or receiving AI tasks, and sending and/or receiving processing results of AI tasks, etc., thereby realizing the integration of AI and wireless networks, and enabling the communication devices in the wireless network to perform AI-related tasks.
  • the method further includes: the communication device is connected through AI Send an AI message to the AI node, and/or the communication device receives the AI message from the AI node through the AI connection; wherein the AI message indicates at least one of the following information: the encryption method used for the AI message, the compression method used for the AI message, the AI The message type of the message, the AI message carries control information or data, the verification code for integrity verification of the AI message, and the content carried by the AI message.
  • the method further includes: the communication device receives updated AI connection information through the AI connection.
  • the AI connection information can be dynamically updated, and the communication device can receive the updated AI connection information, so that the AI connection information can be dynamically updated according to the communication situation.
  • the method further includes at least one of the following: the information of the AI connection includes a model, and the communication device processes the model; the information of the AI connection includes a data set, and the communication device performs operations based on the data set. Measurement; AI-connected information includes computing resources, and the communication device uses computing resources to perform AI tasks.
  • the method further includes: the communication device releases the first AI task to the AI node through the AI connection; and/or the communication device receives the second AI task from the AI node through the AI connection. AI tasks.
  • the communication device can issue AI tasks to the AI node through the AI connection between the communication device and the AI node, or can also receive tasks issued by the AI node.
  • the communication device releases the first AI task to the AI node through the AI connection, including: the communication device releases the first AI task to the AI node through the AI connection to use the AI node to Other communication devices release the first AI mission.
  • other communication devices may be other AI nodes or network devices, or may also be terminal devices.
  • the communication device can issue AI tasks to other communication devices through the AI node through the AI connection between the communication device and the AI node, which can be applied to more scenarios.
  • the AI node can be used to issue the AI task to other communication devices; for another example, if the communication device cannot communicate directly with other communication devices, the AI node can also be used to issue the AI task to other communication devices. .
  • the method further includes: when a preset condition is met, the communication device sends the processing result of the second AI task.
  • the communication device can report the processing result of the task when the preset conditions are met.
  • the second AI task includes instruction information of preset conditions.
  • the method further includes: the communication device sends a request message to the AI node, where the request message is used to request the establishment of an AI connection; or, the communication device receives instruction information from the AI node, The instruction information is used to trigger the communication device to establish an AI connection with the AI node.
  • the communication device before the communication device establishes an AI connection with the AI node, the communication device can establish an AI connection with the AI node by initiating a request, or establish an AI connection with the communication device by triggering the AI node. Therefore, the communication device or AI node can establish an AI connection when it is necessary to perform relevant operations through the AI connection, such as when releasing an AI task, to avoid the waste of resources caused by not using the AI connection after the AI connection is established.
  • the request message also includes indication information of the request type.
  • the request type is used to indicate the purpose of requesting to establish an AI connection, and the data set is determined according to the request type.
  • the AI connection information also includes at least one of the following: an identifier of the AI connection, and information about a wireless bearer associated with the AI connection.
  • the AI node is deployed in any of the following: network equipment and terminal equipment.
  • the method further includes: the communication device transmits the AI capability of the communication device.
  • the communication device sends the AI capability of the communication device to the core network.
  • the AI capabilities of the communication device include at least one of the following: AI task types supported by the communication device, hardware capabilities of the communication device, computing power capabilities of the communication device, priority.
  • the third aspect provides a communication method that can be performed by the AI node.
  • the AI node may be a communication device, or a chip or circuit used in the communication device, which is not limited in this application. For ease of description, the following takes the AI node as an example.
  • the method may include: the artificial intelligence AI node sends AI connection information, and the AI node establishes an AI connection with the communication device according to the AI connection information.
  • the AI connection information includes at least one of the following configured for the AI connection: model, data set, and computing resource.
  • the fourth aspect provides a communication method that can be performed by an AI node.
  • the AI node may be a communication device, or a chip or circuit used in the communication device, which is not limited in this application.
  • the following takes the AI node as an example.
  • the method may include: the artificial intelligence AI node sends AI connection information, and the AI connection information includes at least one of the following configured for the AI connection: model, data set, computing resource; the AI node establishes AI with the communication device according to the AI connection information. connect.
  • the method further includes: the AI node sends an AI message to the communication device through the AI connection, and/or the AI node receives the AI message from the communication device through the AI connection; wherein , the AI message indicates at least one of the following information: the encryption method used for the AI message, the compression method used for the AI message, the message type of the AI message, whether the AI message carries control information or data, and the verification of the integrity of the AI message. Code and content carried by AI messages.
  • the method further includes: the AI node sends updated AI connection information to the communication device through the AI connection.
  • the method also includes at least one of the following: the information of the AI connection includes a model, and the AI node processes the data sent by the communication device according to the model; the information of the AI connection includes data Set, AI node manages data set; AI connection information includes computing resources, and AI nodes use computing resources to perform AI tasks.
  • the method further includes: the AI node determines the AI connection information according to the AI capability of the communication device.
  • the method further includes: the AI node receiving the first message issued by the at least one communication device through an AI connection between the AI node and each of the at least one communication device. AI task; and/or, the AI node issues a second AI task to at least one communication device through an AI connection between the AI node and each of the at least one communication device; wherein at least one communication device includes a communication device.
  • the AI node releases the second AI task to the communication device through the AI connection between the AI node and the communication device, including: the AI node passes between the AI node and the communication device AI The connection is to release the second AI task to the communication device to release the second AI task to other communication devices through the communication device.
  • the method further includes: when the preset conditions are met, the AI node communicates to the AI node through the AI connection between the AI node and each communication device in the at least one communication device. At least one communication device sends the processing result of the first AI task.
  • the first AI task includes instruction information of preset conditions.
  • the method further includes: the AI node receives a request message from the communication device, the request message is used to request the establishment of the AI connection; or, the AI node sends instruction information to the communication device, The instruction information is used to trigger the communication device to establish an AI connection with the AI node.
  • the request message also includes indication information of the request type.
  • the request type is used to indicate the purpose of requesting to establish an AI connection, and the data set is determined according to the request type.
  • the AI connection information also includes at least one of the following: an identifier of the AI connection, and information about a wireless bearer associated with the AI connection.
  • the AI node includes a first AI node and a second AI node, and the first AI node and/or the second AI node satisfies any of the following: First AI node Used to transmit signaling, the second AI node is used to process AI tasks; the second AI node is used to process AI operations instructed by the first AI node; the first AI node is used to process the first type of AI tasks, and the second AI node is used to process AI tasks.
  • the first type of AI tasks are not exactly the same as the second type of AI tasks; the second AI node is used to save and/or send the processing results of the AI tasks.
  • the method further includes: the first AI node sends notification information to the second AI node, and the notification information is used to notify the second AI node of at least one of the following: executing an AI task , save the processing results of the AI task, and send the processing results of the AI task.
  • the AI node is deployed in any of the following: network equipment and terminal equipment.
  • the AI node is deployed on a network device, and the method further includes: the network device determines the information of the wireless bearer associated with the AI connection based on at least one of the following: a model, a data set, Computing resources.
  • a communication method is provided, which method can be executed by a communication device.
  • the communication device may be a communication device (such as a terminal device, a network device), or a chip or circuit used in the communication device, which is not limited in this application.
  • a communication device such as a terminal device, a network device
  • a chip or circuit used in the communication device which is not limited in this application.
  • the following description takes a communication device as an example.
  • the method may include: the communication device releases an AI task to the AI node through an AI connection between the communication device and the artificial intelligence AI node; and the communication device receives a processing result of the AI task.
  • the communication device can issue AI tasks to the AI node through the AI connection between the communication device and the AI node, thereby realizing the integration of AI and wireless networks, and realizing the integration of AI and wireless networks by devices in the wireless network (such as AI nodes, and such as other communication devices) to perform AI tasks.
  • the communication device releases the AI task to the AI node through the AI connection between the communication device and the artificial intelligence AI node, including: the communication device releases the AI task to the AI node through the AI connection Tasks to publish AI tasks to other communication devices through AI nodes.
  • the communication device can use the AI connection between the communication device and the AI node.
  • the node publishes AI tasks to other communication devices, which can be applied to more scenarios. For example, when the AI node is unable to perform the AI task, the AI node can be used to issue the AI task to other communication devices; for another example, if the communication device cannot communicate directly with other communication devices, the AI node can also be used to issue the AI task to other communication devices. .
  • the method further includes: the communication device sends a request message to the AI node, where the request message is used to request the establishment of the AI connection; or, the communication device receives instruction information from the AI node, The instruction information is used to trigger the communication device to establish an AI connection with the AI node.
  • the AI node is deployed in any of the following: network equipment and terminal equipment.
  • a sixth aspect provides a communication method that can be performed by an AI node.
  • the AI node may be a communication device, or a chip or circuit used in the communication device, which is not limited in this application.
  • the following takes the AI node as an example.
  • the method may include: the artificial intelligence AI node receives an AI task issued from at least one communication device through an AI connection between the AI node and each of the at least one communication device; the AI node performs the AI task, or the AI node sends a request to other communication devices.
  • the communication device issues AI tasks.
  • At least one communication device includes a first communication device, and the method further includes: the AI node transmits data to the first communication device through an AI connection between the AI node and the first communication device. The device sends the processing results of the AI task.
  • the AI node sends the processing result of the AI task to the first communication device through the AI connection between the AI node and the first communication device, including: If the conditions are met, the AI node sends the processing result of the AI task to the first communication device through the AI connection between the AI node and the first communication device.
  • the AI task includes instruction information of preset conditions.
  • the AI node issues AI tasks to other communication devices, and the method further includes: the AI node receives processing results of the AI tasks sent from other communication devices.
  • At least one communication device includes a second communication device
  • the method further includes: the AI node receiving a request message from the second communication device, the request message being used to request the establishment of the AI node AI connection with the second communication device; or, the AI node sends instruction information to the second communication device, and the instruction information is used to trigger the second communication device and the AI node to establish an AI connection between the AI node and the second communication device.
  • the AI node is deployed in any of the following: network equipment and terminal equipment.
  • a communication method is provided, which method can be executed by a communication device.
  • the communication device may be a communication device (such as a terminal device, a network device), or a chip or circuit used in the communication device, which is not limited in this application.
  • a communication device such as a terminal device, a network device
  • a chip or circuit used in the communication device which is not limited in this application.
  • the following description takes a communication device as an example.
  • the method may include: the communication device receives an AI task from the AI node through an AI connection between the communication device and the artificial intelligence AI node; the communication device performs the AI task, or the communication device issues the AI task to other communication devices.
  • the AI node can issue AI tasks to the communication device or to other communication devices through the AI connection between the communication device and the communication device.
  • the AI node can publish the AI task to the communication device, and then the communication device processes the AI task.
  • idle computing resources such as the computing power of the communication device itself
  • the method further includes: the communication device sends the processing result of the AI task to the AI node through the AI connection.
  • the communication device sends the processing result of the AI task to the AI node through the AI connection, including: when the preset conditions are met, the communication device sends the AI node to the AI node through the AI connection. Send the processing results of the AI task.
  • the AI task includes instruction information of preset conditions.
  • the communication device issues an AI task to other communication devices
  • the method further includes: the communication device receives a processing result of the AI task sent from the other communication device.
  • the method further includes: the communication device sends a request message to the AI node, and the request message is used to request the establishment of the AI connection; or, the communication device receives instruction information from the AI node, The instruction information is used to trigger the communication device to establish an AI connection with the AI node.
  • the AI node is deployed in any of the following: network equipment and terminal equipment.
  • An eighth aspect provides a communication method that can be performed by an AI node.
  • the AI node may be a communication device, or may be a chip or circuit execution used in the communication device, which is not limited in this application.
  • the following takes the AI node as an example.
  • the method may include: the artificial intelligence AI node releases an AI task to at least one communication device through an AI connection between the AI node and each of the at least one communication device; the AI node receives a processing result of the AI task.
  • At least one communication device includes a first communication device, and the AI node issues an AI task to the first communication device through an AI connection between the AI node and the first communication device. , including: the AI node releases the AI task to the first communication device to release the AI task to other communication devices through the first communication device.
  • At least one communication device includes a second communication device, and the method further includes: the AI node receiving a request message from the second communication device, the request message being used to request the establishment of the AI node AI connection with the second communication device; or, the AI node sends instruction information to the second communication device, and the instruction information is used to trigger the second communication device and the AI node to establish an AI connection between the AI node and the second communication device.
  • the AI node is deployed in any of the following: network equipment and terminal equipment.
  • a ninth aspect provides a communication device configured to perform the method provided in any one of the above-mentioned first to eighth aspects.
  • the device may include units and/or modules, such as a processing unit and/or a communication unit, for performing the method provided by any of the above implementations of any one of the first to eighth aspects.
  • the device is a communication device (such as a terminal device, an AI node, or a network device).
  • the communication unit may be a transceiver, or an input/output interface; the processing unit may be at least one processor.
  • the transceiver may be a transceiver circuit.
  • the input/output interface may be an input/output circuit.
  • the device is a chip, chip system or circuit used in a communication device.
  • the communication unit may be an input/output interface, interface circuit, output circuit, input circuit, pin or related circuit on the chip, chip system or circuit etc;
  • the processing unit can It is at least one processor, processing circuit or logic circuit, etc.
  • a communication device which device includes: a memory for storing a program; and at least one processor for executing the computer program or instructions stored in the memory to execute any one of the above-mentioned first to eighth aspects. Methods provided by any of the above implementations of aspects.
  • the device is a communication device (such as a terminal device, an AI node, or a network device).
  • a communication device such as a terminal device, an AI node, or a network device.
  • the device is a chip, chip system or circuit used in a communication device.
  • this application provides a processor for executing the methods provided in the above aspects.
  • a computer-readable storage medium stores program code for device execution.
  • the program code includes any of the above-mentioned steps for executing any one of the above-mentioned first to eighth aspects.
  • a computer program product containing instructions is provided.
  • the computer program product When the computer program product is run on a computer, it causes the computer to execute the method provided by any of the above-mentioned implementations of any one of the above-mentioned first to eighth aspects. method.
  • a fourteenth aspect provides a chip.
  • the chip includes a processor and a communication interface.
  • the processor reads instructions stored in the memory through the communication interface and executes any of the above implementations of any one of the first to eighth aspects. method provided.
  • the chip also includes a memory, in which computer programs or instructions are stored.
  • the processor is used to execute the computer programs or instructions stored in the memory.
  • the processor is used to execute The method provided by any one of the above implementations of any one of the above first to eighth aspects.
  • a communication system including the above communication device and an AI node.
  • FIG. 1 shows a schematic diagram of a wireless communication system 100 suitable for embodiments of the present application.
  • Figure 2 shows a schematic diagram of a communication system according to an embodiment of the present application.
  • Figure 3 shows a schematic diagram of a communication system according to another embodiment of the present application.
  • Figure 4 is a schematic diagram of a communication method 400 provided by an embodiment of the present application.
  • Figure 5 shows a schematic diagram of a protocol stack for communication between a terminal device and AI node #1.
  • Figure 6 shows a schematic diagram of the cell format of the AI message.
  • Figure 7 shows a schematic diagram of the protocol stack for communication between the UE and AI node #2.
  • Figure 8 shows a schematic flowchart of a communication method 800 provided according to an embodiment of the present application.
  • Figure 9 shows a schematic diagram of a radio bearer.
  • Figure 10 shows a schematic flowchart of a communication method 1000 provided according to another embodiment of the present application.
  • Figure 11 shows a schematic flowchart of a communication method 1100 provided according to another embodiment of the present application.
  • Figure 12 shows a schematic flowchart of a communication method 1200 provided according to another embodiment of the present application.
  • Figure 13 is a schematic block diagram of a communication device 1300 provided by an embodiment of the present application.
  • Figure 14 is a schematic block diagram of a communication device 1400 provided by an embodiment of the present application.
  • Figure 15 is a schematic block diagram of a chip system 1500 provided by an embodiment of the present application.
  • the technical solutions provided by this application can be applied to various communication systems, such as fifth generation (5th generation, 5G) or new radio (NR) systems, long term evolution (LTE) systems, LTE frequency division Duplex (frequency division duplex, FDD) system, LTE time division duplex (TDD) system, etc.
  • the technical solution provided by this application can also be applied to future communication systems, such as the sixth generation mobile communication system.
  • the technical solution provided by this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine type Communication (machine type communication, MTC), and Internet of Things (Internet of things, IoT) communication systems or other communication systems.
  • D2D device-to-device
  • V2X vehicle-to-everything
  • M2M machine-to-machine
  • MTC machine type Communication
  • Internet of Things Internet of things, IoT
  • Terminal devices in the embodiments of this application include various devices with wireless communication functions, which can be used to connect people, objects, machines, etc.
  • Terminal equipment can be widely used in various scenarios, such as: cellular communication, D2D, V2X, peer to peer (P2P), M2M, MTC, IoT, virtual reality (VR), augmented reality (augmented reality) , AR), industrial control, autonomous driving, telemedicine, smart grid, smart furniture, smart office, smart wearables, smart transportation, smart city drones, robots, remote sensing, passive sensing, positioning, navigation and tracking, autonomous delivery Wait for the scene.
  • the terminal device can be a terminal in any of the above scenarios, such as MTC terminal, IoT terminal, etc.
  • the terminal device can be a 3rd generation partnership project (3GPP) standard user equipment (UE), terminal (terminal), fixed equipment, mobile station (mobile station) equipment, or mobile equipment, user Subscriber unit, handheld device, vehicle-mounted device, wearable device, cellular phone, smart phone, SIP phone, wireless data card, personal digital assistant (PDA), computer, Tablet computers, notebook computers, wireless modems, handheld devices (handset), laptop computers (laptop computers), computers with wireless transceiver functions, smart books, vehicles, satellites, global positioning system (GPS) equipment, Target tracking equipment, aircraft (such as drones, helicopters, multi-helicopters, quad-helicopters, or aircraft, etc.), ships, remote control devices, smart home equipment, industrial equipment, or devices built into the above equipment (for example, in the above equipment) communication module, modem or chip, etc.), or other processing equipment connected to the wireless modem.
  • 3GPP 3rd generation partnership project
  • the UE can also be used to act as a base station.
  • a UE may act as a scheduling entity that provides sidelink signals between UEs in scenarios such as V2X, D2D or P2P.
  • the device used to implement the functions of the terminal device may be a terminal device, or may be a device capable of supporting the terminal device to implement the function, such as a chip system or a chip, and the device may be installed in the terminal device.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the network device in the embodiment of the present application may be a device used to communicate with a terminal device.
  • the network device may also be called an access network device or a wireless access network device.
  • the network device may be a base station.
  • the network device in the embodiment of this application may refer to a radio access network (radio access network, RAN) node (or device) that connects the terminal device to the wireless network.
  • radio access network radio access network, RAN node (or device) that connects the terminal device to the wireless network.
  • a base station can broadly cover various names as follows, or be replaced with the following names, such as: Node B (NodeB), evolved base station (evolved NodeB, eNB), next generation base station (next generation NodeB, gNB), relay station, Access point, transmission point (transmitting and receiving point, TRP), transmitting point (TP), primary station, secondary station, multi-standard wireless (motor slide retainer, MSR) node, home base station, network Network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit , AAU), radio head (remote radio head, RRH), central unit (central unit, CU), distributed unit (distributed unit, DU), positioning node, etc.
  • NodeB Node B
  • eNB evolved base station
  • next generation NodeB next generation NodeB, gNB
  • relay station Access point
  • transmission point transmitting and receiving point
  • TP
  • the base station may be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof.
  • a base station may also refer to a communication module, modem or chip used in the aforementioned equipment or devices.
  • the base station can also be a mobile switching center and equipment that performs base station functions in D2D, V2X, and M2M communications, network-side equipment in 6G networks, equipment that performs base station functions in future communication systems, etc.
  • Base stations can support networks with the same or different access technologies. The embodiments of this application do not limit the specific technology and specific equipment form used by the network equipment.
  • Base stations can be fixed or mobile.
  • a helicopter or drone may be configured to act as a mobile base station, and one or more cells may move based on the mobile base station's location.
  • a helicopter or drone may be configured to serve as a device that communicates with another base station.
  • Network equipment and terminal equipment can be deployed on land, indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; they can also be deployed on aircraft, balloons and satellites in the sky. In the embodiments of this application, the scenarios in which network devices and terminal devices are located are not limited.
  • FIG. 1 shows a schematic diagram of a wireless communication system 100 suitable for embodiments of the present application.
  • the wireless communication system 100 may include at least one network device, such as the network device 110 shown in FIG. 1 , and the wireless communication system 100 may further include at least one terminal device, such as the terminal device 120 shown in FIG. 1 and terminal equipment 130.
  • Both network equipment and terminal equipment can be configured with multiple antennas, and network equipment and terminal equipment can communicate using multi-antenna technology. Terminal devices can communicate directly with each other.
  • the network device When the network device communicates with the terminal device, the network device can manage one or more cells, and there can be an integer number of terminal devices in one cell.
  • the network device 110 and the terminal device 120 form a single-cell communication system, and without loss of generality, the cell is called cell #1.
  • the network device 110 may be a network device in cell #1, or the network device 110 may serve a terminal device (eg, terminal device 120) in cell #1.
  • a cell can be understood as an area within the wireless signal coverage of a network device.
  • FIG. 1 is a simplified schematic diagram for ease of understanding.
  • the wireless communication system 100 may also include other network devices or other terminal devices, which are not shown in FIG. 1 .
  • the embodiments of this application can be applied to any communication scenario in which the sending device and the receiving device communicate.
  • AI artificial intelligence
  • wireless network architecture level In order to cope with the vision of future intelligent inclusiveness, intelligence will further evolve at the wireless network architecture level, and artificial intelligence (AI) will be further integrated with wireless networks to achieve endogenous intelligence in the network and intelligence in terminals, thereby Can cope with some possible new needs and scenarios.
  • terminal types are diversified and terminal connections are more flexible and intelligent.
  • Terminal types are diverse, super IoT (such as Internet of Things, connected cars, industry, medical, etc.), massive connections, terminal connections are more flexible, and the terminal itself has certain AI capabilities.
  • Another example is a possible demand for endogenous intelligence in the network.
  • the network may also provide computing and AI services to better support inclusive, real-time and high-security AI services.
  • NWDAF network data analytics function
  • the main functions of NWDAF include: supporting data collection from other network functions (NF) and application functions (AF), and supporting data collection from network operation and maintenance systems (such as operation administration and maintenance (OAM)) collect data and may provide NF may provide metadata open services, data analysis services, etc. to AF.
  • the main goals of the introduction of NWDAF include: automation and intelligence of network operation and maintenance, optimization of network performance and business experience, end-to-end service level agreement (SLA) guarantee, etc.
  • SLA service level agreement
  • the AI model trained by NWDAF can be applied to the network's own fields such as mobility management, session management, and network automation, using AI methods to replace methods based on numerical formulas in the original network functions.
  • NWDAF is deployed in the core network and is a plug-in AI unit. It is not designed to be strongly coupled with the communication network, and its performance is limited.
  • this application takes into account: on the one hand, AI-related tasks require high computing power, and the load and energy consumption of AI network construction are high; on the other hand, there may be many terminal devices in the network with computing power idle. Therefore, this application proposes to create endogenous AI functions in the communication network, so that the communication network and AI can be more closely integrated to provide better AI services.
  • instruction may include direct instruction, indirect instruction, explicit instruction, and implicit instruction.
  • indication information When it is described that certain indication information is used to indicate A, it can be understood that the indication information carries A, indicates A directly, or indicates A indirectly.
  • the information indicated by the indication information is called information to be indicated.
  • the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated, etc.
  • the information to be indicated may also be indirectly indicated by indicating other information, where there is an association relationship between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while other parts of the information to be indicated are known or agreed in advance.
  • the indication of specific information can also be achieved by means of a pre-agreed (for example, protocol stipulated) arrangement order of each piece of information, thereby reducing the indication overhead to a certain extent.
  • the information to be instructed can be sent together as a whole, or can be divided into multiple sub-information and sent separately, and the sending period and/or sending timing of these sub-information can be the same or different.
  • the specific sending method is not limited in this application.
  • the sending period and/or sending timing of these sub-information may be predefined, for example, according to a protocol, or may be configured by the transmitting device by sending configuration information to the receiving device.
  • the configuration information may include, for example but not limited to, one or a combination of at least two of radio resource control signaling, media access control (media access control, MAC) layer signaling and physical layer signaling.
  • radio resource control signaling includes, for example, radio resource control (RRC) signaling;
  • MAC layer signaling for example, includes MAC control element (CE);
  • physical layer signaling for example, includes downlink control information (downlink control information, DCI).
  • FIG. 2 shows a schematic diagram of a communication system according to an embodiment of the present application.
  • the communication system may include a terminal device and AI node #1.
  • the terminal device and AI node #1 can communicate through the AI-1 interface, that is, the AI-1 interface can be used to send and/or receive AI-related signaling.
  • the terminal device can issue AI tasks to AI node #1 through the AI-1 interface; for another example, AI node #1 can issue AI tasks to the terminal device through the AI-1 interface; for another example, the terminal device can issue AI tasks to the terminal device through the AI-1 interface.
  • AI node #1 Send and/or receive the processing results of the AI task to AI node #1; for another example, AI node #1 can send and/or receive the processing results of the AI task to the terminal device through the AI-1 interface; for another example, the terminal device can send and/or receive the processing results of the AI task to the terminal device through the AI-1 interface.
  • the AI-1 interface requests AI node #1 to release AI tasks to other devices (such as other AI nodes, other terminal devices, and network devices); for another example, AI node #1 can request the terminal device to issue AI tasks to other devices through the AI-1 interface.
  • other Devices (such as other AI nodes, other terminal equipment, and network equipment) issue AI tasks.
  • AI task means a task related to AI.
  • AI tasks may include: model failure testing, model performance testing, model training testing, data collection, etc.
  • the terminal device communicates with the AI node #1 through the AI-1 interface, or alternatively, the terminal device communicates with the AI node #1 through the AI connection (AI connection).
  • AI connection represents the connection between the terminal device and the AI node #1.
  • the terminal device and the AI node #1 can transmit AI-related signaling and perform AI-related operations.
  • AI connection represents a connection relationship between the terminal device and AI node #1, and is a logical concept rather than a physical entity, which will not be described in detail below.
  • the AI-1 interface may be a logical interface.
  • the AI-1 interface can be implemented through the connection between the terminal device and the network device, and the connection between the network device and AI node #1.
  • the terminal device sends a message to AI node #1 through the AI-1 interface, which may include: the terminal device sends the message to the network device, and the network device forwards the message to AI node #1, that is, the network device can be similar to the terminal Relay device between device and AI node #1.
  • AI node #1 manages the AI-1 interface.
  • AI node #1 is responsible for establishing, maintaining, and releasing the AI-1 interface.
  • AI node #1 is responsible for configuring, updating, and releasing resources for the AI-1 interface.
  • AI node #1 is deployed on any of the following: network equipment, terminal equipment. It can be understood that AI node #1 can also be deployed separately, that is, deployed at a location other than the network device and the terminal device, such as a location close to the network device or a location close to the terminal device, without limitation. In addition, there is no limit on the number of AI nodes #1.
  • the communication system also includes network equipment. Communication is possible between the network device and the terminal device, and communication is also possible between the network device and AI node #1. As shown in Figure 3, the network device and the terminal device can communicate through the UE-universal terrestrial radio access network (Uu) interface, and the AI node #1 and the network device can communicate through the AI- 2 interface communication.
  • Uu UE-universal terrestrial radio access network
  • the communication system also includes a core network. Communication is possible between the core network and network equipment, and communication is also possible between the core network and AI node #1. As shown in Figure 3, the core network and network equipment can communicate through the next generation (NG) interface, and the core network and AI node #1 can communicate through the AI-3 interface.
  • NG next generation
  • the core network may be a device and/or software system deployed in a wireless network.
  • the core network may include one or more core network nodes to provide core network functions, such as UE connection management, mobility management, policy management and other functions.
  • the core network can also provide user plane gateway functions to external networks such as the Internet.
  • the core network node may be a next-generation (e.g., 6G or higher) core network node, or a legacy (e.g., 5G, 4th generation (4G), 3th generation (3G) or second generation) (2th generation, 2G)) core network node.
  • the core network can include network elements such as access and mobility management function (AMF) and unified data management (UDM).
  • AMF access and mobility management function
  • UDM unified data management
  • the core network may also include one or more network nodes in the public switched telephone network (PSTN), packet data network, optical network, Internet Protocol (IP) network, wide area network (WAN) ), local area network (LAN), wireless local area network (WLAN), wired network, wireless network, metropolitan area network and other networks to enable communication between terminal devices and/or network devices.
  • PSTN public switched telephone network
  • IP Internet Protocol
  • WAN wide area network
  • LAN local area network
  • WLAN wireless local area network
  • wired network wireless network
  • wireless network metropolitan area network and other networks to enable communication between terminal devices and/or network devices.
  • the names of devices with core network functions and the network functions contained therein may be different.
  • the 4G core network is generally called the evolved packet core (EPC);
  • the 5G core network is generally called the 5G core network (5GC or 5GCN).
  • the AI capabilities of the terminal device are stored in the core network (such as UDM).
  • the core network stores the contract information of the terminal device, and the contract information of the terminal device includes the AI capability of the terminal device. If AI node #1 wants to query If the AI capability of the terminal device is determined, AI node #1 can query the core network for the AI capability of the terminal device through the AI-3 interface.
  • the AI capabilities of the terminal device can also be called the AI-related parameters of the terminal device.
  • the AI capabilities of the terminal device are uniformly described below.
  • the AI capabilities of the terminal device may include, for example, at least one of the following: the priority of the terminal device, the maximum computing power supported by the terminal device, the hardware capabilities of the terminal device, and the AI task types supported by the terminal device.
  • the priority of the terminal device can be determined based on the historical response of the terminal device. For example, if the terminal device participates in collaborative processing of AI tasks more times, the priority of the terminal device is higher; if the terminal device participates in collaborative processing of AI tasks less often, the priority of the terminal device is lower.
  • the priority of the terminal device can be determined based on the capabilities of the terminal device (such as the maximum computing power supported, or the hardware capabilities of the terminal device itself, etc.). For example, if the capability of the terminal device is higher, the priority of the terminal device is higher; if the capability of the terminal device is lower, the priority of the terminal device is lower.
  • the AI capabilities of the terminal device may also include security requirements of the terminal device.
  • the AI capability of the terminal device stored in the core network is an exemplary description, and the present application is not limited thereto.
  • the AI node (such as AI node #1) stores the AI capability of the terminal device. If AI node #1 wants to query the AI capability of the terminal device, it can directly determine the AI capability of the terminal device locally, or obtain the terminal device from other AI nodes. The device’s AI capabilities.
  • FIG 3 shows a schematic diagram of a communication system according to another embodiment of the present application.
  • the communication system shown in Figure 3 also includes AI node #2.
  • Figure 3 is only an example.
  • the AI nodes are not limited to two, and there can be multiple AI nodes.
  • the AI node #2 can communicate with other devices in the communication system.
  • the AI node #2 can communicate with the terminal device through the AI-6 interface
  • the AI node #2 and the AI node #1 can communicate through the AI-4 interface
  • the AI node #2 and the network device can communicate with each other through the AI-4 interface.
  • AI node #2 is deployed on any of the following: network equipment, terminal equipment.
  • AI node #2 can also be deployed separately, that is, deployed at a location other than the network device and the terminal device, such as a location close to the network device or a location close to the terminal device, without limitation. In addition, there is no limit on the number of AI nodes #2.
  • the terminal device communicates with the AI node #2 through the AI-6 interface, or alternatively, the terminal device communicates with the AI node #2 through the AI connection.
  • the AI connection represents the AI connection between the terminal device and the AI node #2. Through the AI connection, the terminal device and the AI node #2 can transmit AI-related signaling and perform AI-related operations.
  • the AI-6 interface may be a logical interface.
  • the AI-6 interface can be implemented through the connection between the terminal device and the network device, and the connection between the network device and AI node #2.
  • the terminal device sends a message to AI node #2 through the AI-6 interface, which may include: the terminal device sends the message to the network device, and the network device forwards the message to AI node #2, that is, the network device can be similar to the terminal Relay device between device and AI Node #2.
  • AI node #1 is responsible for managing the AI-6 interface.
  • AI node #1 is responsible for establishing, maintaining, and releasing the AI-6 interface.
  • AI node #1 is responsible for resources used for the AI-6 interface, such as configuring, updating, and releasing resources used for the AI-6 interface.
  • AI node #1 and AI node #2 may be AI nodes responsible for different functions.
  • AI node #1 is responsible for signaling transmission
  • AI node #2 is responsible for performing processing operations.
  • AI node #2 is responsible for performing processing operations, which may include at least one of the following: AI node #2 processes AI tasks, AI node #2 maintains the processing results of AI tasks, AI node #2 maintains AI-related data sets, AI Node #2 maintains AI phase Off model.
  • AI node #2 is responsible for performing various operations instructed by AI node #1.
  • AI node #1 instructs AI node #2 to process an AI task.
  • AI node #2 processes the corresponding AI task.
  • AI node #2 is also used to send the processing results of the AI task to the terminal device (such as through the AI-6 interface).
  • AI node #1 is responsible for processing the first type of AI tasks
  • AI node #2 is responsible for processing the second type of AI tasks.
  • the first type of AI tasks and the second type of AI tasks are not exactly the same.
  • the overhead incurred by processing the first type of AI tasks is less than the overhead caused by processing the second type of AI tasks.
  • AI node #1 and AI node #2 are equivalent to two different AI nodes in the communication system, and the two AI nodes can be responsible for different AI tasks respectively.
  • AI node #1 and AI node #2 can be independent devices, or they can be integrated into the same device to implement different functions, or they can be network elements in hardware devices, or they can be on dedicated hardware.
  • the running software function, or the virtualization function instantiated on the platform for example, cloud platform, this application does not limit the specific forms of the above-mentioned AI node #1 and AI node #2.
  • the AI nodes can be divided into more fine details, or a larger number of AI nodes can be included.
  • the communication system includes at least two AI nodes (such as AI node #1, AI node #2, and AI node #3), and each AI node is responsible for different functions.
  • both of the at least two AI nodes can establish AI connections with the terminal device, or some of the at least two AI nodes can establish AI connections with the terminal device.
  • FIG. 2 and FIG. 3 are exemplary illustrations, and the present application is not limited thereto.
  • the communication systems shown in Figures 2 and 3 can also include a greater number of devices, such as a greater number of terminal devices, a greater number of AI nodes, a greater number of network devices, and so on.
  • terminal devices can also publish AI tasks to each other, such as terminal devices directly publishing AI tasks to each other, or terminal devices publishing AI tasks to other terminal devices through AI nodes (such as AI node #1, or AI node #2). AI tasks.
  • the communication system provided according to the embodiment of the present application is briefly introduced above with reference to FIG. 2 and FIG. 3 .
  • the communication method provided by the embodiment of this application is introduced below. The method described below can be used with the system shown in Figure 2 or Figure 3.
  • Figure 4 is a schematic diagram of a communication method 400 provided by an embodiment of the present application.
  • Method 400 may include the following steps.
  • the communication device receives AI connection information.
  • the communication device may be a terminal device, or a chip, chip system or circuit used in the terminal device, without limitation.
  • AI connection refers to the connection between the communication device and the AI node. Through the AI connection, the communication device and the AI node can transmit AI-related signaling and perform AI-related operations.
  • the AI connection between the communication device and the AI node can be realized through the connection between the communication device and the network device, and the connection between the network device and the AI node.
  • One possible implementation method is that the AI connection between the communication device and the AI node can be through a radio bearer (RB) (also called an air interface bearer, or an air interface radio bearer) established between the communication device and the network equipment. ), and the connection implementation between network equipment and AI nodes.
  • RB radio bearer
  • the communication device sends a message to the AI node through the AI connection, which may include: the communication device sends the message to the network device, and the network device forwards the message to the AI node, that is, the network device can be similar to the communication between the communication device and the AI node. relay equipment.
  • the communication device sends a message to the network device based on the communication device and the network device.
  • the wireless bearer established between them sends the message; the network device forwards the message to the AI node, and the message can be forwarded based on the connection between the network device and the AI node (such as the AI-2 interface).
  • the AI connection information represents information related to the AI connection.
  • the communication device and/or the AI node can process AI-related operations. Information about AI connection will be described in detail later.
  • the communication device establishes an AI connection with the AI node according to the AI connection information.
  • an AI connection can be established between the communication device and the AI node, and then communication can be carried out through the AI connection.
  • the communication device can directly establish an AI connection with the AI node based on the AI connection information, or the communication device can also receive/send instructions or triggers, or when preset conditions are met. Establishing an AI connection with an AI node based on the AI connection information is not limited by this application.
  • the method 400 further includes step 430: the communication device communicates with the AI node through the AI connection.
  • the communication device can receive AI connection information, and establish an AI connection with the AI node based on the AI connection information, so that the AI node and the communication device can communicate through the AI connection, such as sending and/or receiving AI Tasks, such as sending and/or receiving the processing results of AI tasks, etc., thereby realizing the integration of AI and wireless networks.
  • the method 400 further includes: the communication device sending a request message, the request message being used to request the establishment of an AI connection between the communication device and the AI node. Further optionally, the communication device receives a response message, the response message including the AI connection information.
  • the request message may also be called an AI connection establishment request message, for example.
  • the request message also includes indication information of the request type, which can be used to indicate the purpose of establishing the AI connection. For example, if the request message directly carries the request type, it can directly indicate the request type; for another example, if the request message carries certain indication information, the indication information can indicate the request type.
  • the request type includes at least one of the following: model training, data collection, task release, etc.
  • model training it means that the purpose of the UE requesting to establish an AI connection is to conduct model training.
  • request type includes task release it means that the purpose of the UE requesting to establish an AI connection is to release an AI task through the AI connection.
  • the communication device may initiate a request to establish an AI connection with the AI node. Therefore, when the communication device needs to perform related operations through the AI connection, such as when releasing an AI task, the communication device can actively initiate a request to establish an AI connection, thereby avoiding the waste of resources caused by not using the AI connection after the AI connection is established.
  • the method 400 further includes: the communication device receiving indication information from the AI node, the indication information being used to trigger the communication device to establish an AI connection with the AI node. Further optionally, after receiving the indication information, the communication device may send a request message to the AI node to request the establishment of an AI connection between the communication device and the AI node.
  • the AI connection can be established with the communication device by triggering the AI node. Therefore, when the AI node needs to perform relevant operations through the AI connection, such as when releasing an AI task, the AI node can actively trigger the communication device to send an AI establishment request, thereby avoiding the waste of resources caused by not using the AI connection after the AI connection is established.
  • the AI connection information includes: an identification of the AI connection, AI connection resource information, or wireless bearer information.
  • the identifier of the AI connection (identifier, ID): It can be provided by the AI node. Identity connected through this AI, The AI connection can be identified as an AI connection between the communication device and the AI node. For example, after receiving the request message from the communication device, the AI node learns that the communication device requests to establish an AI connection. Therefore, the AI node allocates an identifier of the AI connection.
  • AI connection resource information It can be provided by the AI node. For example, after receiving the request message from the communication device, the AI node learns that the communication device requests to establish an AI connection. Therefore, the AI node allocates resources for the AI connection.
  • the resource information of the AI connection includes at least one of the following configured for the AI connection: a model, a data set, or a computing resource.
  • AI-connected resource information includes models. That is, the AI node can configure a model for the AI connection, and the model can also be called a preset model.
  • the AI node can process data uploaded by the communication device based on the model.
  • the communication device can process the model, such as the communication device performing model training, model testing, model fusion, etc. based on the model.
  • AI connection resource information includes data sets. That is, the AI node can configure a data set for the AI connection, and the data set can also be called a preset data set. For an AI node, as an example, the AI node can manage the data set. For a communication device, as an example, the communication device may perform AI measurements based on the data set.
  • the AI node can allocate a data set according to the request type, that is, the data set can be a UE-specific data set. Or it can be a mission-specific data set.
  • the AI connection resource information includes computing resources. That is, the AI node can configure computing resources for the AI connection, and the computing resources can also be called preset computing resources.
  • the computing resources can be used by the communication device or AI node to perform AI-related operations. For example, the communication device or AI node uses the computing resources to process AI tasks.
  • the AI node can dynamically maintain and update the resource information of the AI connection during the validity period of the AI connection. For example, the AI node can update the resources of the AI connection. In addition, if the resource information of the AI connection is updated, the AI node can send the updated resource information of the AI connection to the communication device through the AI connection, or send it to the communication device on demand. The AI node and the communication device can re-establish the AI connection based on the updated resource information of the AI connection.
  • the above mainly uses the AI node to configure the identification of the AI connection and the resource information of the AI connection for the AI connection, and the present application is not limited thereto.
  • the identifier of the AI connection and the resource information of the AI connection may also be configured by other devices.
  • other devices send the configured ID of the AI connection and the resource information of the AI connection to the AI node, and then the AI node sends the configured ID of the AI connection and the resource information of the AI connection to the communication device.
  • other devices send the configured identifier of the AI connection and the resource information of the AI connection to the AI node, and the other device sends the configured identifier of the AI connection and the resource information of the AI connection to the communication device.
  • the AI connection information provided by the AI node (such as the identification of the AI connection and/or the resource information of the AI connection) is called the first information of the AI connection.
  • the wireless bearer information can be provided by network equipment.
  • the AI connection between the communication device and the AI node can be realized through the wireless bearer established between the communication device and the network device, and the connection between the network device and the AI node. Therefore, the wireless connection can be configured by the network device.
  • the wirelessly carried information is associated with the AI connection, so that the communication device and the AI node can communicate through the AI connection.
  • the communication device sends a request message to the AI node through the network device, and the AI node sends the first information of the AI connection to the communication device through the network device.
  • the network device learns that the AI node accepts (or (called consent) request of the communication device, so the network device configures the information of the radio bearer and combines the information of the radio bearer with the identification of the AI connection in the first information of the AI connection (if the first information does not include the identification of the AI connection knowledge, then the association is based on obtaining the identity of the AI connection through other means.
  • the network device determines the wireless bearer information based on the resource information of the AI connection.
  • the resource information of the AI connection includes a model configured for the AI connection.
  • the network device configures the wireless bearer according to the model, such as configuring the protocol layers included in the protocol stack, and/or configuring parameters of each protocol layer, etc.
  • the model configured for AI connection inputs environmental parameters, such as the remaining computing power of the transceiver and the receiver, and the channel parameters between the transceiver and receiver.
  • the output is the configuration parameters of each protocol layer, such as the frame structure of the physical layer and the medium.
  • Segmentation conditions of the access control (medium access control, MAC) layer, parameters involved in the scheduling algorithm, retransmission conditions of the radio link control (RLC) layer, packet data convergence protocol (PDCP) ) layer's header compression algorithm involves parameters, etc.
  • RLC radio link control
  • PDCP packet data convergence protocol
  • the AI node in the embodiment of the present application can be AI node #1 or AI node #2, without limitation.
  • AI node #1 or AI node #2, without limitation.
  • two possible situations are introduced based on the AI node.
  • the AI node is AI node #1.
  • the AI connection represents the connection between the communication device and AI node #1.
  • the interface between the communication device and AI node #1 is called the AI-1 interface.
  • the communication device and AI node #1 communicate through the AI connection, which can be replaced For, the communication device communicates with AI node #1 through the AI-1 interface.
  • the above-mentioned AI node that configures the first information of the AI connection for the AI connection may be AI node #1.
  • the device that configures the first information of the AI connection for the AI connection can also be another device, and the other device sends the first information of the AI connection configured for the AI connection to the AI node #1, and the AI node #1 then configures the AI connection.
  • the first information of the AI connection is sent to the communication device, or other devices send the first information of the AI connection to the AI node #1 and the communication device.
  • AI node #1 configures the identifier of the AI connection
  • the other device sends the resource information of the AI connection configured for the AI connection to AI node #1
  • AI node #1 then The resource information of the AI connection and the identifier of the AI connection configured by itself are sent to the communication device.
  • the communication device communicates with the AI node #1 through the AI connection, including: the communication device sends an AI message to the AI node #1 through the AI connection, and/or the AI node #1 sends an AI message to the communication device through the AI connection.
  • FIG. 5 shows a schematic diagram of a protocol stack for communication between the communication device and AI node #1.
  • the communication device and AI node #1 can communicate through the AI layer (stratum).
  • messages can be transferred between the AI layer and the RRC layer.
  • the AI layer transfers the message sent by the communication device to the AI node to the RRC layer
  • the RRC layer transfers the message sent by the AI node to the communication device to the AI layer.
  • the communication device sends an RRC message to the network device, and the RRC message carries the AI message (or may also be called an AI stratum message); the network device sends the AI message to the AI node.
  • the AI node sends an AI message to the network device, and the network device sends an RRC message to the communication device, where the RRC message carries the AI message. That is, the network device can use the RRC message as a container to carry the AI message.
  • FIG. 5 is an exemplary illustration.
  • the protocol stack for communication between the communication device and AI node #1 can be flexibly configured and is not limited here.
  • FIG. 5 mainly takes the communication device as a terminal device, and the transmission between the terminal device and the AI node #1 through a network device as an example for illustrative explanation. The present application is not limited thereto.
  • the AI message indicates at least one of the following information: security type, compression type, message type, control indicator or traffic indicator, message authentication Code (message authentication code), message content.
  • the AI message may directly include at least one of the above information to directly indicate at least one of the above information; or the AI message may The information may include corresponding indication information of at least one of the above information, and may indirectly indicate at least one of the above information.
  • FIG. 6 shows a schematic diagram of the cell format of the AI message. Each information is briefly introduced below.
  • Security type used to indicate the encryption method used for AI messages. If the AI message includes a security type, the AI message can be decrypted based on the security type. If the AI message does not include a security type, a possible implementation method is to decrypt the AI message based on the default encryption method. That is, in this case, the encryption method used in the AI message can be pre-agreed or predefined. .
  • Compression type Used to indicate the compression method used by the compression type. If the AI message includes a compression type, the AI message can be decompressed based on the compression type. If the AI message does not include a compression type, a possible implementation method is to decompress the AI message based on the default compression method. That is, in this case, the compression method used for the AI message can be pre-agreed or predefined. of.
  • One possible implementation method is to distinguish the types of AI messages by using different bit values. For example, if the message type is 0000, it means that the AI message is used to describe model parameters; if the message type is 0001, it means that the AI message is used for model update; if the message type is 0010, it means that the AI message is used to describe the data set; if If the message type is 0011, it means that the AI message is used to update the data set; if the message type is 0100, it means that the AI message is used to describe the task; if the message type is 0101, it means that the AI message is used for AI node switching or reselection of parameters. etc. It can be understood that the above examples are illustrative, and the embodiments of the present application are not limited thereto.
  • the type of the AI message can be determined based on the message type. If the AI message does not include the message type, one possible implementation is to learn the type of AI message by parsing the AI message; or another possible implementation is that AI messages in different formats can correspond to different types of AI messages, That is, the type of the AI message can be determined based on the format of the AI message. For example, there are many formats of AI messages. If the format of the AI message is the first format, it means that the AI message is used to describe the model parameters; if the format of the AI message is the second format, it means the AI message is used for the model.
  • the format of the AI message is the third format, it means that the AI message is used to describe the data set; if the format of the AI message is the fourth format, it means that the AI message is used to update the data set; if the format of the AI message is the fifth format, it means that the AI message is used to describe the task; if the format of the AI message is the sixth format, it means that the AI message is used for AI node switching or parameter reselection.
  • Control indication or traffic indication used to indicate whether the AI message transmits control signaling or AI-related data. For example, if the AI message includes a control indication, it means that the AI message transmits control signaling; for another example, if the AI message includes a traffic indication, it means that the AI message transmits control AI-related data.
  • Message authentication code A verification code used to indicate integrity verification. If the AI message includes a message authentication code, integrity verification can be performed based on the message authentication code. If the AI message does not include a message authentication code, a possible implementation method can be to perform integrity verification based on the default verification code. That is, in this case, the verification code for integrity verification of the AI message can be Pre-agreed or predefined.
  • the communication device communicates with the AI node #1 through the AI connection, or the communication device and the AI node #1 transmit the AI message through the AI connection, which may include the following examples.
  • Example 1 the communication device issues the first AI task to AI node #1 through the AI connection. Accordingly, AI node #1 receives the first AI task.
  • the task issued by the communication device is called the first AI task.
  • the first possible implementation method is that after receiving the first AI task, AI node #1 processes the first AI task.
  • AI node #1 After AI node #1 completes the first AI task, it can send the processing of the first AI task to the communication device. result.
  • the second possible implementation method is that after receiving the first AI task, AI node #1 releases the first AI task to other communication devices (such as other terminal devices or other AI nodes), and the other communication devices process the first AI task.
  • AI tasks That is, the communication device may issue the first AI task to other communication devices through AI node #1.
  • other communication devices may be referred to as communication devices that cooperate to complete the AI task.
  • the other communication devices are referred to as cooperative communication devices.
  • the number of cooperative communication devices is at least one.
  • AI node #1 determines the cooperative communication device, and releases the first AI task to the cooperative communication device through the AI connection between AI node #1 and the cooperative communication device
  • the release of the first AI task by the AI node #1 to the collaborative communication device includes: a complete task release of the first AI task and/or a partial task release of the first AI task.
  • the communication device determines the cooperative communication device, and the communication device issues a first AI task to the cooperative communication device through AI node #1. After AI node #1 receives the first AI task, it can communicate with the cooperative communication device through AI node #1. The AI connection between the devices releases the first AI task to the collaborative communication device.
  • the communication device determines the cooperating communication device, and the communication device releases the first subtask of the first AI task to AI node #1, and releases the second subtask of the first AI task to the cooperating communication device through AI node #1, After AI node #1 receives the first AI task, it can release the second sub-task of the first AI task to the cooperation communication device through the AI connection between AI node #1 and the cooperation communication device.
  • the cooperative communication device after the cooperative communication device completes the first AI task, it can send the processing result of the first AI task to AI node #1, and then AI node #1 sends the processing result of the first AI task to the communication device.
  • AI node #1 sends the processing result of the first AI task to the communication device.
  • the cooperative communication device may directly send the processing result of the first AI task to the communication device.
  • the above-mentioned collaborative communication device or AI node #1 completing the first AI task includes: completing the complete first AI task, or completing the sub-tasks (or partial/decomposition) of the first AI task. Task).
  • the third possible implementation method is that after receiving the first AI task, AI node #1 sends an instruction to AI node #2, and AI node #2 processes the first AI task. That is, the communication device may issue the first AI task to AI node #2 through AI node #1.
  • AI Node #2 may be considered a cooperating communication device. It can be understood that the first AI task released by AI node #1 to AI node #2 includes: a complete task of the first AI task and/or a partial task of the first AI task.
  • AI node #1 can send instructions to AI node #2 through the AI-4 interface. In this manner, after AI node #2 completes the first AI task, it can send the processing result of the first AI task to AI node #1, and then AI node #1 sends the processing result of the first AI task to the communication device. Alternatively, after AI node #2 completes the first AI task, it can also directly send the processing result of the first AI task to the communication device. Taking Figure 3 as an example, AI node #2 can send the first AI task to the communication device through the AI-6 interface. The processing results of AI tasks.
  • the completion of the first AI task by the above-mentioned AI node #1 or AI node #2 includes: completing the complete first AI task, or completing the sub-task (or partial/partial) of the first AI task. decompose tasks).
  • AI node #1 releases the first AI task to the cooperative communication device or AI node #2
  • the cooperative communication device or AI node #2 can AI node #2 processes the first AI task
  • the cooperative communication device and AI node #1 can process the first AI task together
  • AI node #2 and AI node #1 can process the first AI task together.
  • the collaborative communication device and AI node #1 may each be responsible for a part of the first AI task.
  • AI node #1 issues a second AI task to the communication device through the AI connection.
  • the communication device receives the second AI task.
  • the AI node (such as AI node #1, and also such as AI node #2) is Bu's mission is called the second AI mission.
  • the communication device after receiving the second AI task, processes the second AI task. In this manner, after completing the second AI task, the communication device can send the processing result of the second AI task to AI node #1.
  • the communication device after receiving the second AI task, releases the second AI task to the cooperative communication device, and the cooperative communication device processes the second AI task. That is, AI node #1 can process the second AI task through the communication device. Issue the second AI task to the cooperating communication device. It can be understood that the release of the second AI task by the communication device to the cooperating communication device includes: the complete task release of the second AI task and/or the partial task release of the second AI task.
  • the communication device determines the cooperating communication device and issues the second AI task to the cooperating communication device.
  • AI node #1 determines the cooperating communication device, and AI node #1 issues a second AI task to the cooperating communication device through the AI connection between AI node #1 and the communication device, and the communication device receives the second AI task. Afterwards, the second AI task can be released to the cooperative communication device.
  • the cooperative communication device after the cooperative communication device completes the second AI task, it can send the processing result of the second AI task to AI node #1 through the AI connection between the cooperative communication device and AI node #1.
  • the processing result of the second AI task can be sent to the communication device, and the communication device can transmit the processing result of the second AI task through the AI connection between the communication device and AI node #1.
  • the result is sent to AI node #1.
  • the above-mentioned communication device or cooperative communication device completing the second AI task includes: completing the complete second AI task, or completing the sub-tasks (or partial/decomposed tasks) of the second AI task. .
  • the above-mentioned collaborative communication device completing the second AI task includes: completing the complete second AI task, or completing the sub-tasks (or partial/decomposed tasks) of the second AI task.
  • AI node #1 can issue AI tasks to at least one communication device.
  • AI node #1 can issue AI tasks to each communication device through an AI connection with each communication device.
  • the AI task can be processed and the processing result of the AI task can be directly fed back, or alternatively, the processing result of the AI task can be fed back when the preset conditions are met.
  • AI node #1 sends the processing result of the first AI task to the communication device when the preset conditions are met.
  • the communication device sends the processing result of the second AI task to AI node #1.
  • the preset conditions which can also be called trigger conditions or feedback conditions, are not limited.
  • the first possible implementation method is to predefine preset conditions, such as standard predefined preset conditions.
  • AI tasks have a corresponding relationship with preset conditions.
  • the preset conditions corresponding to the AI task are also determined. That is, the processing results of the AI task can be reported based on the preset conditions corresponding to the AI task. .
  • AI tasks and preset conditions can exist in a form similar to Table 1.
  • the preset conditions may be, for example, periodic reporting, location-triggered reporting, or model performance-triggered reporting, etc., without limitation.
  • the AI task is an AI-M1 task, that is, a model failure test
  • the processing results of the AI-M1 task can be reported when the model performance degrades to a threshold.
  • the AI task is an AI-M2 task, that is, a model performance test
  • the processing results of the AI-M2 task can be reported periodically, and further a defined period (that is, the reporting time interval) can be predetermined.
  • the AI task is an AI-M4 task, that is, data collection
  • it can be used when the model performance degrades (such as when the model performance degrades to a threshold, or when degradation occurs) or when it moves to a specific area. (if the position change exceeds the threshold), the processing result of the AI-M4 task will be reported.
  • the thresholds involved in Table 1 can be predefined, such as standard predefined ones.
  • Table 1 is an exemplary description, and the modifications belonging to Table 1 are applicable to this application.
  • Table 1 may also include a larger number of AI tasks.
  • “the performance degrades to the threshold” in Table 1 can also be replaced by "the performance reaches the preset condition”.
  • the second possible implementation method is to carry the instruction information of the preset conditions corresponding to the AI task when releasing the AI task. For example, when an AI task is released, the preset conditions corresponding to the AI task are directly carried; for another example, when an AI task is released, certain instruction information is carried, and the instruction information can indicate the preset conditions corresponding to the AI task.
  • the first AI task includes preset conditions corresponding to the first AI task.
  • the communication device sends a first AI message to AI node #1 through the AI connection.
  • the first AI message is used to issue a first AI task, and the first AI message includes a preset condition corresponding to the first AI task.
  • AI node #1 sends the processing result of the first AI task to the communication device, then AI node #1 sends the processing result of the first AI task to the communication device when the preset conditions corresponding to the first AI task are met.
  • the second AI task includes preset conditions corresponding to the second AI task.
  • AI node #1 sends a second AI message to the communication device through the AI connection.
  • the second AI message is used to issue a second AI task, and the second AI message includes a preset condition corresponding to the second AI task.
  • the communication device sends the processing result of the second AI task to AI node #1, then the communication device sends the processing result of the second AI task to AI node #1 when the preset conditions corresponding to the second AI task are met.
  • the processing results of the AI task can be reported periodically. If the AI task is a model failure test, the processing results of the AI task can be reported when the performance degrades to a threshold. If the AI task is data collection, it can be used when the model performance degrades (such as when the model performance degrades to a threshold, or when degradation occurs) or when it moves to a specific area (such as when the position change exceeds the threshold). ), reporting the processing results of the AI task.
  • different AI tasks may feedback different information.
  • Table 2 shows the correspondence between AI tasks and feedback information.
  • the processing results of the sent AI-M1 task may include: the location of the model failure, the error amount of the model, and the timestamp of the model.
  • the timestamp of the model may be, for example, the expiration time of the model.
  • the processing results of the sent AI-M3 task may include: the location of the model training and testing, the model's training convergence time, and the model's timestamp.
  • the timestamp of the model may be, for example, the generation time of the model.
  • Table 2 is an exemplary description, and the modifications belonging to Table 2 are applicable to this application.
  • Table 2 may also include a larger number of AI tasks.
  • Table 2 may also include a larger amount of feedback information.
  • the feedback information may also include the identification of the model, the reasoning performance of the model, and the identification of the AI node (such as the identification of the AI node that processes the AI task).
  • the AI node is AI node #2.
  • the AI connection represents the connection between the communication device and AI node #2.
  • the interface between the communication device and AI node #2 is called the AI-6 interface.
  • the communication device and AI node #2 communicate through the AI connection, which can be replaced by, communication The device communicates with AI Node #2 via the AI-6 interface.
  • the AI node that configures the first information of the AI connection for the AI connection may be AI node #1, and the AI node #1 sends the first information of the AI connection configured for the AI connection to AI node #2.
  • the device that configures the first information of the AI connection for the AI connection may also be another device, and the other device sends the first information of the AI connection configured for the AI connection to AI node #2.
  • other devices may configure the resource information of the AI connection, AI node #1 configures the identifier of the AI connection, and AI node #1 will send the identifier of the AI connection configured for the AI connection to AI node #2, and other devices will configure the AI connection identifier for the AI node.
  • the resource information of the AI connection configured by the connection is sent to AI node #2.
  • another device may configure the resource information of the AI connection, AI node #2 configures the identifier of the AI connection, and the other device sends the resource information of the AI connection configured for the AI connection to AI node #2.
  • AI node #2 configures the first information of the AI connection by itself.
  • the communication device communicates with the AI node #2 through the AI connection, including: the communication device sends an AI message to the AI node #2 through the AI connection, and/or the AI node #2 sends an AI message to the communication device through the AI connection.
  • AI messages please refer to the description in the first possible scenario, which will not be described again here.
  • FIG. 7 shows a schematic diagram of a protocol stack for communication between the communication device and AI node #2.
  • the communication device and AI node #2 can communicate through the protocol stack shown in Figure 7. As shown in Figure 7, taking the communication device as a UE and the network device as a 5G access network (AN) as an example, there is a corresponding protocol data unit (PDU) layer between the UE and AI node #2 , the UE and the AI node #2 can communicate through the 5G AN.
  • the 5G AN can, for example, serve as a relay device to forward the information sent by the UE to the AI node #2, and/or forward the information sent by the AI node #2. Information sent to UE.
  • FIG. 7 is an exemplary illustration, and the protocol stack for communication between the communication device and AI node #2 can be flexibly configured and is not limited here.
  • FIG. 7 mainly takes the transmission between the communication device and AI node #2 through network equipment as an example for illustrative explanation, and the present application is not limited thereto.
  • the communication device communicates with the AI node #2 through the AI connection, or the communication device and the AI node #2 transmit the AI message through the AI connection, which may include the following examples.
  • Example 1 the communication device issues the first AI task to AI node #2 through the AI connection. Accordingly, AI node #1 receives the first AI task.
  • AI node #1 is responsible for processing less expensive AI tasks
  • AI node #2 is responsible for processing more expensive AI tasks. Therefore, the communication device can issue different AI tasks to different AI nodes. For example, the communication device issues an AI task with a small processing overhead to the AI node #1, and the communication device issues an AI task with a large processing overhead to the AI node #2.
  • the first possible implementation method is that after receiving the first AI task, AI node #2 processes the first AI task.
  • AI node #2 After AI node #2 completes the first AI task, it can send the processing result of the first AI task to the communication device.
  • the second possible implementation method is that after receiving the first AI task, AI node #2 releases the first AI task to the collaborative communication device (including releasing all and/or part of the first AI task), and the collaborative communication device Handle the first AI task. That is, the communication device may issue the first AI task to other communication devices through AI node #2.
  • AI node #2 releases the first AI task to the collaborative communication device (including releasing all and/or part of the first AI task), and the collaborative communication device Handle the first AI task. That is, the communication device may issue the first AI task to other communication devices through AI node #2.
  • AI node #2 issues a second AI task to the communication device through the AI connection.
  • the communication device receives the second AI task.
  • the communication device after receiving the second AI task, processes the second AI task. In this manner, after completing the second AI task, the communication device can send the processing result of the second AI task to AI node #2.
  • the communication device after receiving the second AI task, releases the second AI task to the collaborative communication device (including issuing all and/or part of the second AI task), and the collaborative communication device processes the second AI task.
  • 2 AI tasks That is, AI node #2 may issue the second AI task to the cooperating communication device through the communication device.
  • AI node #2 may issue the second AI task to the cooperating communication device through the communication device.
  • AI node #2 can issue AI tasks to at least one communication device.
  • AI node #2 can issue AI tasks to each communication device through an AI connection with each communication device.
  • AI node #2 sends the processing result of the first AI task to the communication device when the preset conditions are met.
  • the description in the first possible scenario which will not be described again here.
  • UE is used to represent a communication device below, and the embodiments of the present application are exemplarily described with reference to FIGS. 8 to 12 .
  • the specific steps and terminology involved please refer to the above description.
  • the AI-1 interface and the AI-6 interface are used for description.
  • Figure 8 shows a schematic flowchart of a communication method 800 provided according to an embodiment of the present application.
  • the method 800 can be used to implement a solution such as the method 400.
  • the method 800 can be used in a scenario where the UE establishes an AI-1 interface with the AI node #1.
  • method 800 may include the following steps.
  • the UE initiates random access (RA) to the network device.
  • RA random access
  • the UE can first access the network device to communicate with the network device.
  • the UE may access the network device by initiating random access.
  • the UE sends AI connection setup request (AI connection setup request) information to the network device.
  • One possible implementation method is that during the process of establishing a radio resource control (RRC) connection between the UE and the network device, the UE sends AI connection establishment request information to the network device.
  • RRC radio resource control
  • the AI connection establishment request information is carried in the RRC connection establishment complete message.
  • the AI connection establishment request information is used to request the establishment of the AI-1 interface between the UE and the AI node #1, that is, the AI connection establishment request information is used to request the establishment of the AI connection between the UE and the AI node #1.
  • the AI connection establishment request information includes a request type, and the request type can be used to indicate the purpose of establishing the AI-1 interface.
  • the request type please refer to the relevant description in method 400, which will not be described again here.
  • the network device sends AI connection establishment request information to AI node #1.
  • the network device After receiving the AI connection establishment request information sent by the UE, the network device forwards the AI connection establishment request information to AI node #1, that is, the AI connection establishment request information can be transparently transmitted to AI node #1 through the network equipment.
  • One possible implementation method is that the network device sends an initial UE message (initial UE message) to AI node #1, and the initial UE message carries AI connection establishment request information.
  • AI node #1 determines the first information of the AI-1 interface according to the AI capability of the UE.
  • AI node #1 learns that the UE requests to establish the AI-1 interface.
  • AI node #1 may first determine the AI capability of the UE, and then determine the first information of the AI-1 interface based on the AI capability of the UE.
  • the AI capabilities of the UE may include, for example, at least one of the following: the priority of the UE, the maximum computing power supported by the UE, the type of AI tasks supported by the UE, the hardware capabilities of the UE, and the security requirements of the UE.
  • the core network stores the UE's subscription information
  • the AI node #1 queries the core network for the UE's subscription information.
  • the AI node #1 can learn the UE's AI capabilities based on the UE's subscription information.
  • the UE's subscription information includes the UE's AI capabilities.
  • AI node #1 can query the core network for the UE's subscription information through the AI-3 interface.
  • the first information of the AI-1 interface includes resource information of the AI-1 interface (that is, resource information of the AI connection between the UE and AI node #1).
  • resource information of the AI-1 interface that is, resource information of the AI connection between the UE and AI node #1
  • the identifier of the AI-1 interface (that is, the identifier of the AI connection between the UE and AI node #1) may also be included.
  • the first information of the AI-1 interface you may refer to the relevant description of the first information of the AI connection in method 400, which will not be described again here.
  • AI node #1 can determine the first information of the AI-1 interface according to the AI capability of the UE.
  • the AI node #1 may allocate more computing resources to the UE, or the AI node #1 may allocate resource information of the AI-1 interface to the UE first.
  • AI node #1 may allocate a model corresponding to the AI task type according to the AI task type supported by the UE.
  • AI node #1 may allocate more data sets to the UE.
  • AI node #1 sends the first information of the AI-1 interface to the network device.
  • AI node #1 sends AI connection establishment response information or AI connection establishment acceptance information to the network device,
  • the AI connection establishment response message or the AI connection establishment acceptance message includes the first information of the AI-1 interface. If AI node #1 sends AI connection establishment response information to the network device, the AI connection establishment response information is used to notify acceptance of the AI connection establishment request.
  • AI node #1 sends initial context setup request information to the network device, and the initial context setup request information includes the first information of the AI-1 interface.
  • the initial context establishment request information includes the first information of the AI-1 interface, which may indicate that AI node #1 accepts the AI connection establishment request; or the initial context establishment request information includes the first information of the AI-1 interface and the AI Connection establishment accepts information.
  • the network device establishes a wireless bearer.
  • the AI-1 interface can be implemented through the wireless bearer established between the UE and the network device, and the connection between the network device and AI node #1. Therefore, if the network device learns that AI node #1 accepts the establishment of the AI-1 interface for the UE, the network device can establish a radio bearer and associate the radio bearer with the AI-1 interface (such as associating the radio bearer with the identity of the AI connection) .
  • the wireless bearer can identify a set of configuration parameters of the air interface protocol stack.
  • Figure 9 shows a schematic diagram of a radio bearer. (1) in Figure 9 shows the situation of downlink channel mapping, and (2) in Figure 9 shows the situation of uplink channel mapping. The following takes the downlink channel mapping as an example and introduces it with several examples.
  • Example 1 The network device configures the logical channel AI-CH for the AI connection between the UE and AI node #1, that is, the radio bearer configured by the network device is transmitted through the logical channel AI-CH.
  • AI-CH the logical channel corresponding to the AI connection between the UE and AI node #1.
  • the information of the logical channel can be mapped to the downlink shared channel (DL-SCH), and the transmission channel can be mapped to the physical downlink shared channel (PDSCH), that is, transmitted by the PDSCH.
  • DL-SCH downlink shared channel
  • PDSCH physical downlink shared channel
  • Example 2 The network device configures the logical channel AI-CH and the transport channel DL-AI-CH for the AI connection between the UE and AI node #1, that is, the radio bearer configured by the network device passes the logical channel AI-CH and the transport channel DL. -AI-CH transmission.
  • the downlink transmission channel corresponding to the AI connection between the UE and AI node #1 is called DL-AI-CH.
  • the information of the logical channel can be mapped to the transport channel DL-AI-CH, and the transport channel DL-AI-CH can be mapped to the PDSCH, that is, transmitted by the PDSCH.
  • the network device indicates the location of the transmission resource through the physical downlink control channel (PDCCH). Since both the transmission channels DL-AI-CH and DL-SCH are mapped to PDSCH, the location of the transmission resource indicated by the PDCCH can be distinguished as DL-AI by scrambling the radio network temporary identifier (RNTI) of the PDCCH.
  • RNTI radio network temporary identifier
  • the location of the transmission resource is the resource location of DL-SCH.
  • Example 3 The network device configures the logical channel AI-CH, the transport channel DL-AI-CH, and the physical channel PD-AI-CH for the AI connection between the UE and AI node #1, that is, the radio bearer configured by the network device passes Logical channel AI-CH, transport channel DL-AI-CH, and physical channel PD-AI-CH are transmitted.
  • the downlink physical channel corresponding to the AI connection between the UE and AI node #1 is called PD-AI-CH.
  • the information of the logical channel can be mapped to the transport channel DL-AI-CH, and the transport channel DL-AI-CH can It is mapped to the physical channel PD-AI-CH, that is, transmitted by PD-AI-CH.
  • the network device indicates the resource location of the PD-AI-CH through a system message. If some physical channels PD-AI-CH carry control information, the network device can also indicate which PD-AI-CHs have control information through system messages. The control information indicates the locations of different DL-AI-CH messages carried on the PD-AI-CH.
  • the network device indicates the occupancy status of the PD-AI-CH resource through downlink control information (DCI) (for example, called DCI_AI).
  • DCI downlink control information
  • the DCI may be scrambled by an RNTI (called Z-RNTI for distinction).
  • the uplink channel is similar and will not be described again here for simplicity.
  • the network device sends the AI-1 interface information to the UE.
  • the information of the AI-1 interface includes first information of the AI-1 interface.
  • the AI-1 interface information also includes radio bearer information. It can be understood that the wireless bearer information can also be sent independently from the first information of the AI-1 interface (such as being carried in different signaling), or can be combined with the first information of the AI-1 interface and sent (such as being carried in the same signaling). Order), no restrictions will be imposed.
  • the network device sends AI connection establishment response information or AI connection establishment acceptance information to the UE, and the AI connection establishment response information or AI connection establishment acceptance information includes the information of the AI-1 interface. If the network device sends AI connection establishment response information to the UE, the AI connection establishment response information is used to notify acceptance of the AI connection establishment request.
  • the network device sends RRC reconfiguration information to the UE, and the RRC reconfiguration information includes the information of the AI-1 interface.
  • the RRC reconfiguration information includes the information of the AI-1 interface, which may indicate that the AI node #1 accepts the AI connection establishment request; or the RRC reconfiguration information includes the information of the AI-1 interface and the AI connection establishment acceptance information.
  • the UE sends RRC reconfiguration complete information to the network device.
  • the network device sends RRC reconfiguration information to the UE in step 807, then the UE sends RRC reconfiguration completion information to the network device in step 808.
  • the network device sends initial context setup response (initial context setup response) information to AI node #1.
  • AI node #1 sends initial context establishment request information to the network device, then in step 809, the network device sends initial context establishment response information to AI node #1.
  • the UE sends RRC reconfiguration completion information to the network device, and the network device sends the initial context establishment response information to AI node #1.
  • AI node #1 receives the initial context establishment response information
  • AI-1 is connected by default. The interface is to complete the AI-1 interface between the UE and AI node #1.
  • step 808 and/or step 809 are exemplary descriptions, and step 808 and/or step 809 may not be executed in practice.
  • the network device sends the information of the AI-1 interface to the UE.
  • the UE After the UE receives the information of the AI-1 interface, it opens the AI-1 interface by default, that is, the UE and the AI node are completed. AI-1 interface between #1.
  • step 808 there is no need to perform step 808, that is, after the network device sends the AI-1 interface information to the UE, the network device sends the initial context establishment response information to AI node #1, and after AI node #1 receives the initial context establishment response information , the AI-1 interface is opened by default, that is, the AI-1 interface between the UE and AI node #1 is completed.
  • step 809 does not need to be performed, that is, after the network device sends the AI-1 interface information to the UE through the RRC reconfiguration information, the UE sends the RRC reconfiguration completion information to the network device, and after the network device receives the RRC reconfiguration information, silent If the AI-1 interface is opened, the AI-1 interface between the UE and AI node #1 is completed.
  • the UE and AI node #1 establish an AI-1 interface, that is, the AI-1 interface between the UE and AI node #1 is opened. Therefore, the UE and AI node #1 can communicate through the AI-1 interface. For example, the UE sends an AI message to AI node #1 through the AI-1 interface. For another example, AI node #1 sends an AI message to the UE through the AI-1 interface. Regarding the communication between the UE and AI node #1 through the AI-1 interface, please refer to the relevant description in method 400, which will not be described again here.
  • the scenario where the UE establishes the AI-1 interface with the AI node #1 is exemplarily introduced above with reference to Figure 8.
  • the method 800 can be implemented through the architecture shown in Figure 2.
  • the UE establishes an AI-1 interface with AI node #1, and can then transmit AI messages through the established AI-1 interface, such as issuing AI tasks, receiving AI tasks, etc., thereby realizing the integration of wireless networks and AI.
  • Figure 10 shows a schematic flowchart of a communication method 1000 provided according to another embodiment of the present application.
  • This method 1000 can be used to implement a solution such as method 400.
  • method 1000 can be used in a scenario where the UE establishes an AI-6 interface with AI node #2.
  • method 1000 may include the following steps.
  • the UE establishes the AI-1 interface with AI node #1.
  • the UE sends AI connection establishment request information to AI node #1.
  • the AI connection establishment request information is used to request the establishment of an AI-6 interface between the UE and AI node #2, that is, the AI connection establishment request information is used to request the establishment of an AI connection between the UE and the AI node #2.
  • the AI connection establishment request information includes a request type, which may be used to indicate the purpose of establishing the AI-6 interface.
  • request type please refer to the relevant description in method 400, which will not be described again here.
  • AI node #1 selects AI node #2 and determines the first information of the AI-6 interface.
  • the first information of the AI-6 interface includes the resource information of the AI-6 interface (that is, the resource information of the AI connection between the UE and AI node #2).
  • it may also include the identification of the AI-6 interface (also the resource information of the AI-6 interface). That is, the identification of the AI connection between the UE and AI node #2).
  • the first information of the AI-6 interface you may refer to the relevant description of the first information of the AI connection in method 400, which will not be described again here.
  • AI node #1 After receiving the AI connection establishment request information, AI node #1 can first determine the AI capability of the UE, and then determine the first information of the AI-6 interface based on the AI capability of the UE.
  • the AI capabilities of the UE the way in which the AI node #1 learns the AI capabilities of the UE, and the way in which the AI node #1 determines the first information of the AI-6 interface based on the AI capabilities of the UE, please refer to the description in step 804, which will not be used here. Again.
  • AI node #1 selects AI node #2, which can be achieved in any of the following ways.
  • AI node #1 can determine AI node #2 based on the UE's AI connection establishment request information. For example, AI node #1 can learn the purpose of the UE establishing the AI-6 interface according to the request type in the UE's AI connection establishment request information, and AI node #1 selects the appropriate AI node #2 based on the purpose.
  • AI node #1 determines AI node #2 based on the AI capability of the UE. For example, AI node #1 can learn the AI tasks supported by the UE based on the AI capabilities of the UE, and then AI node #1 can select AI node #2 that can support the AI task. For another example, AI node #1 can learn that the priority of the UE is relatively high based on the AI capability of the UE. Then AI node #1 can select AI node #2 with higher computing power. For example, AI node #2 with computing power exceeding the threshold, to satisfy the UE as much as possible.
  • AI node #1 may select AI node #2 based on the request type and the UE's AI capabilities.
  • AI node #1 and AI node #2 configure the AI-6 interface.
  • AI node #1 After AI node #1 selects AI node #2, the configuration of the AI-6 interface on the AI node #2 side can be completed. For example, AI node #1 sends the first information of the AI-6 interface to AI node #2.
  • AI node #1 sends an AI connection configuration (AI connection configuration) to AI node #2.
  • AI connection configuration includes the first information of the AI-6 interface.
  • AI node #1 sends the first information of the AI-6 interface to the network device.
  • AI node #1 sends AI connection establishment response information or AI connection establishment acceptance information to the network device, and the AI connection establishment response information or AI connection establishment acceptance information includes the first information of the AI-6 interface. If AI node #1 sends AI connection establishment response information to the network device, the AI connection establishment response information is used to notify acceptance of the AI connection establishment request.
  • AI node #1 sends radio bearer setup request information to the network device, and the radio bearer setup request information includes the first information of the AI-6 interface.
  • the radio bearer establishment request information includes the first information of the AI-6 interface, which indicates that AI node #1 accepts the AI connection establishment request; or the radio bearer establishment request information includes the first information of the AI-6 interface and the AI connection. Establish acceptance information.
  • the network device establishes a wireless bearer.
  • the AI-6 interface can be implemented through the wireless bearer established between the UE and the network device, and the connection between the network device and AI node #2. Therefore, if the network device learns that AI node #2 accepts the establishment of an AI-6 interface for the UE, the network device can establish a radio bearer and associate the radio bearer with the AI-6 interface (such as associating the radio bearer with the identity of the AI connection) .
  • Step 1006 is similar to step 806 and will not be described again here.
  • the network device sends the AI-6 interface information to the UE.
  • the information of the AI-6 interface includes the first information of the AI-6 interface.
  • the AI-6 interface information also includes radio bearer information. It can be understood that the wireless bearer information can also be sent independently from the first information of the AI-6 interface (such as being carried in different signaling), or can be sent in combination with the first information of the AI-6 interface (such as being carried in the same signaling). Order), no restrictions will be imposed.
  • the network device sends AI connection establishment response information or AI connection establishment acceptance information to the UE, and the AI connection establishment response information or AI connection establishment acceptance information includes the information of the AI-6 interface. If the network device sends AI connection establishment response information to the UE, the AI connection establishment response information is used to notify the acceptance of the AI connection establishment request.
  • the network device sends RRC reconfiguration information to the UE, and the RRC reconfiguration information includes the information of the AI-6 interface.
  • the RRC reconfiguration information includes the information of the AI-6 interface, which may indicate that the AI node #2 accepts the AI connection establishment request; or the RRC reconfiguration information includes the information of the AI-6 interface and the AI connection establishment acceptance information.
  • the UE sends RRC reconfiguration completion information to the network device.
  • the network device sends RRC reconfiguration information to the UE in step 1007, then the UE sends RRC reconfiguration completion information to the network device in step 1008.
  • the network device sends radio bearer setup response (radio bearer setup response) information to AI node #1.
  • AI node #1 sends radio bearer establishment request information to the network device, then in step 1009, the network device sends radio bearer establishment response information to AI node #1.
  • the method 1000 further includes step 1010.
  • AI node #1 and AI node #2 update the configuration of the AI-6 interface.
  • AI node #1 and AI node #2 update the configuration of the AI-6 interface, which means that AI node #1 updates the configuration of the AI-6 interface, such as updating the resource information of the AI-6 interface, and updates it on the AI node #2 side.
  • AI node #1 sends an AI connection configuration or AI connection configuration update (AI connection configuration update) to AI node #2.
  • AI connection configuration or AI connection configuration update includes the updated AI-6 interface. configuration.
  • AI node #1 can update the configuration of the AI-6 interface based on the UE's request. For example, the UE sends a message to the AI-1 node through the AI-1 interface, and the message is used to update the configuration of the AI-6 interface; after receiving the message, the AI-1 node updates the configuration of the AI-6 interface based on the message. .
  • AI node #1 can autonomously decide to update the configuration of the AI-6 interface.
  • the UE sends RRC reconfiguration completion information to the network device, and the network device sends the radio bearer establishment response information to AI node #1.
  • AI-6 is opened by default. interface, that is, complete the AI-6 interface between the UE and AI node #2.
  • step 1008 and/or step 1009 is an exemplary description, and step 1008 and/or step 1009 may not be executed in practice.
  • the network device sends the information of the AI-6 interface to the UE.
  • the UE After the UE receives the information of the AI-6 interface, it opens the AI-6 interface by default, that is, the UE and the AI node are completed. AI-6 interface between #2.
  • step 1008 does not need to be performed. That is, after the network device sends the AI-6 interface information to the UE, the network device sends the radio bearer establishment response information to AI node #1, and AI node #1 forwards the wireless bearer establishment response information to AI node #2.
  • Bearer establishment response information after AI node #2 receives the radio bearer establishment response information, the AI-6 interface is opened by default, that is, the AI-6 interface between the UE and AI node #2 is completed.
  • step 1009 does not need to be performed, that is, after the network device sends the AI-6 interface information to the UE through the RRC reconfiguration information, the UE sends the RRC reconfiguration completion information to the network device, and after the network device receives the RRC reconfiguration information, The AI-6 interface is opened by default, that is, the AI-6 interface between the UE and AI node #2 is completed.
  • the UE establishes an AI connection with AI node #2, that is, the AI-6 interface between the UE and AI node #2 is opened. Therefore, the UE and AI node #2 can communicate through the AI-6 interface. For example, the UE sends an AI message to AI node #2 through the AI-6 interface. For another example, AI node #2 sends an AI message to the UE through the AI-6 interface.
  • the communication between the UE and AI node #2 through the AI-6 interface please refer to the relevant description in method 400, which will not be described again here.
  • the method 1000 is mainly explained by taking the AI connection between the UE and the AI node #2 established through the AI node #1 as an example, and this is not limited.
  • the AI connection between the UE and AI node #2 may also be established through a network device.
  • the UE may send AI connection establishment request information to the network device to request the establishment of the AI-6 interface between the UE and AI node #2.
  • the network device selects AI node #2. 2, and sends the AI connection establishment request information to AI node #2.
  • the UE selects AI node #2 and sends an AI connection establishment request message to the network device to request the establishment of an AI-6 interface between the UE and the AI node #2.
  • the network device receives the AI connection establishment request information. After requesting information, the AI connection establishment request information is sent to the AI node #2.
  • the specific implementation can be the establishment between the UE and AI node #1 in method 800, which will not be described again here.
  • the scenario where the UE establishes the AI-6 interface with AI node #2 is exemplarily introduced above with reference to Figure 10.
  • the method 1000 can be implemented through the architecture shown in Figure 3.
  • the UE establishes an AI-6 interface with AI node #2, and can then transmit AI messages through the established AI-6 interface, such as issuing AI tasks, receiving AI tasks, etc., thereby realizing the integration of wireless networks and AI.
  • Figure 11 shows a schematic flowchart of a communication method 1100 provided according to another embodiment of the present application.
  • this method 1100 can be used to implement a solution such as method 400.
  • method 1100 can be used in a scenario where the UE issues a task to AI node #1.
  • method 1100 may include the following steps.
  • UE1 establishes the AI-1 interface with AI node #1.
  • One possible implementation method is that UE1 establishes an AI-1 interface with AI node #1, and in the process, releases the first AI task to AI node #1.
  • the AI connection establishment request information carries the request type, and the request type is used to indicate that the purpose of establishing the AI connection is to publish a task.
  • Another possible implementation method is that UE1 establishes an AI-1 interface with AI node #1, and releases the first AI task to AI node #1 through the established AI-1 interface.
  • AI node #1 determines the cooperating UE based on the AI capabilities of the on-network UE.
  • the number of cooperating UEs is at least one.
  • AI node #1 can determine the cooperating UE according to the AI capabilities of the on-network UE, so that the cooperating UE completes the first AI task.
  • the AI capabilities of the UE include the first AI task that the UE can support (or the first AI task that it can undertake, or the first AI task that it can perform). Therefore, AI node #1 can query the AI capabilities of the UE on the network to Determine which UEs can support the first AI task.
  • AI node #1 can query the AI capabilities of the online UE and determine the collaborative UE that can support the model training and testing task, that is, the AI node #1 The determined collaborative UE can support model training and testing tasks.
  • method 1100 also includes: AI node #1 queries the context of UE1 and confirms the priority information of UE1. In this way, if UE1 has a higher priority, AI node #1 can prioritize determining the cooperating UE for the first AI task released by UE1, thereby allowing more UEs to participate in collaborative processing of the first AI task.
  • AI node #1 and the cooperating UE jointly process the first AI task.
  • AI node #1 is responsible for processing part of the first AI task
  • the cooperating UE is responsible for processing the remaining parts of the first AI task.
  • Another possible implementation method is to cooperate with the UE to process the first AI task.
  • AI node #1 can also determine the tasks that each cooperating UE needs to process. For example, AI node #1 divides the first AI task into at least two subtasks, and each cooperating UE can Process the subtasks of the first AI task separately, and finally complete the first AI task together. Or AI node #1 can also directly issue the first AI task to the cooperating UE.
  • AI node #1 sends AI paging (AI paging) to the cooperating UE.
  • AI node #1 After AI node #1 determines the cooperating UE, it can send an AI paging to the cooperating UE to trigger the cooperating UE to initiate an AI connection establishment request message and establish an AI-1 interface between the cooperating UE and AI node #1. It can be understood that AI paging is only a possible naming method, and its naming does not limit the protection scope of this application.
  • AI node #1 establishes the AI-1 interface with the cooperating UE.
  • steps 1103-1104 may not be performed. That is, after AI node #1 determines the cooperating UE, it directly distributes the first AI task to the cooperating UE. .
  • AI node #1 releases the first AI task to the cooperating UE.
  • AI node #1 After AI node #1 establishes an AI-1 interface with the cooperating UE, it can release AI tasks to the cooperating UE through the corresponding AI-1 interface. For example, AI node #1 issues a first AI task to the cooperating UE, and the first AI task is the same as the first AI task issued by UE1 in step 1101. It can be understood that Figure 11 only shows the situation where the cooperative UE processes the first AI task. The application is not limited to this. For example, as mentioned above, if the number of cooperating UEs is at least two, AI node #1 can also divide the first AI task into at least two subtasks, and each cooperating UE can process the subtasks of the first AI task respectively. Finally, Complete the first AI mission together. For another example, AI node #1 processes the first sub-task of the first AI task, and AI node #1 releases the second sub-task of the first AI task to the cooperating UE, and the cooperating UE processes the second sub-task of the first AI task. .
  • the cooperating UE sends the processing result of the first AI task to AI node #1.
  • the cooperating UE After the cooperating UE completes the first AI task, it sends the processing result of the first AI task to AI node #1.
  • the processing result of the first AI task can be sent to AI node #1 if the preset conditions are met.
  • AI node #1 For details, reference may be made to the relevant description in method 400, which will not be described again here.
  • AI node #1 sends the processing result of the first AI task to UE1.
  • AI node #1 After AI node #1 receives the processing result of the first AI task sent by the cooperating UE, it can directly forward the processing result of the first AI task to UE1; or, if AI node #1 receives the processing result of at least two first AI tasks, The AI node #1 can also select one to send to UE1, or it can further process the received processing results of at least two first AI tasks and then send them to UE1 without restriction.
  • AI node #1 updates the priority information of the cooperating UE in the context of the cooperating UE.
  • the priority information of the UE can be dynamically updated based on the historical response of the UE.
  • the method 1100 is mainly explained by taking the cooperative UE to process the first AI task as an example. It can be understood that in actual communication, after the AI node #1 receives the first AI task issued by UE1, it can also process the first AI task by itself. AI task, and after AI node #1 processes the first AI task, AI node #1 can send the processing result of the first AI task to UE1. Alternatively, after receiving the first AI task issued by UEl, AI node #1 can instruct AI node #2 to process the first AI task (including a complete task or a sub-task).
  • AI node #2 can send the processing result of the first AI task to UE1, or AI node #2 can send the processing result of the first AI task to AI node #1, by AI node #1 sends the processing result of the first AI task to UE1.
  • AI node #1 may process a part of the first AI task by itself, and instruct the remaining tasks to AI node #2 and/or the cooperating UE for processing. It can be understood that this application does not limit the number of AI nodes, and there can also be AI node #3 and AI node #4 to cooperate to complete the first AI task.
  • the method 1100 is mainly explained by taking the cooperating UE to send the processing result of the first AI task to AI node #1, and then the AI node #1 sends the processing result of the first AI task to UE1 as an example.
  • This application does not Limited to this.
  • the cooperating UE may also directly send the processing result of the first AI task to UE1.
  • the UE releases an AI task to AI node #1 is exemplarily introduced above with reference to Figure 11.
  • the UE establishes an AI-1 interface with AI node #1, and then the UE can release an AI task to AI node #1 through the established AI-1 interface.
  • AI node #1 cannot handle the AI task or AI node #1 needs the assistance of other AI nodes or cooperating UEs
  • the AI task can also be released to other AI nodes or other UEs, and then the other AI nodes or other UEs can Processing the AI task enables the use of idle computing resources (such as collaborating with the UE's own computing power) to complete the AI task and reduce the load and energy consumption of the AI network.
  • Figure 12 shows a schematic flowchart of a communication method 1200 provided according to another embodiment of the present application.
  • This method 1200 can be used to implement a solution such as method 400.
  • method 1200 can be used in a scenario where AI node #1 issues a task to the UE.
  • method 1200 may include the following steps.
  • the core network stores the subscription information of at least one UE.
  • the contract information includes the AI capabilities of the UE.
  • the AI capabilities of the UE may include, for example, at least one of the following: the priority of the UE, the maximum computing power supported by the UE, the second AI task type supported by the UE, the hardware capabilities of the UE, and the security requirements of the UE.
  • AI node #1 queries the core network for the contract information of the UE on the network and determines the cooperating UE.
  • the number of cooperating UEs is at least one.
  • AI node #1 may query the core network for the subscription information of the on-network UE to obtain the AI capabilities of the on-network UE.
  • the AI node #1 determines at least one UE to process the second AI task based on the AI capabilities of the on-network UEs (for distinction, the at least one UE that processes the second AI task is called a cooperative UE).
  • the second AI task to be released by AI node #1 is a model testing task
  • AI node #1 can determine the UE that can support the model testing task based on the AI capabilities of the UE on the network, that is, the UE determined by AI node #1 Collaborating UEs can support model testing tasks.
  • AI node #1 releases the second AI task to the cooperating UE.
  • AI node #1 releases the second AI task to the cooperating UE through the AI-1 interface. It can be understood that if the AI-1 interface has not been established between AI node #1 and the collaborating UE, before step 1203, the AI node #1 and the collaborating UE can first establish the AI-1 interface, and then the AI node #1 can The -1 interface releases the second AI task to the cooperating UE.
  • AI node #1 can also divide the second AI task into at least two subtasks, and each cooperating UE can process the subtasks of the second AI task respectively, and finally jointly complete the second AI task. 2 AI tasks.
  • AI node #1 processes the first sub-task of the first AI task, and AI node #1 releases the second sub-task of the first AI task to the cooperating UE, and the cooperating UE processes the second sub-task of the first AI task. .
  • the tasks configured by AI node #1 can include: Need to test Model parameters, preset conditions, and measurement reporting parameters.
  • the preset condition is that the model performance degrades to the threshold, that is, when the model performance degrades to the threshold, the processing results of the model failure test task are reported.
  • the preset condition is periodic, that is, the processing results of the model performance test task are reported periodically (or at intervals).
  • the preset condition is periodic, that is, the processing results of the model performance testing task are reported periodically (or at intervals).
  • the parameters reported for measurement may include the location of model failure, the error amount of the model, and the timestamp of the model.
  • the parameters reported for measurement may include the location of the model performance test, the error amount of the model, and the timestamp of the model.
  • the parameters reported by measurement may include the location of model failure, the error amount of the model, and the training convergence time of the model.
  • the cooperating UE sends the processing result of the second AI task to AI node #1.
  • the second AI task released by AI node #1 is a model test task (such as the model failure test task in Table 3, the model performance test task in Table 3, and the model training test task in Table 3), then
  • the collaborative UE handles the model testing task. If the model test task corresponds to the corresponding preset conditions, after the collaborative UE executes the model test task, the test results can be cached locally, and when the preset conditions are met, the processing results of the model test task can be cached locally. Report to AI node #1 through the AI-1 interface.
  • AI node #1 releases an AI task to the UE is exemplarily introduced above with reference to Figure 12.
  • the UE establishes an AI-1 interface with AI node #1, and then AI node #1 can issue AI tasks to the UE through the established AI-1 interface, thereby utilizing idle computing resources (such as the UE's own computing power). ) to complete AI tasks and reduce the load and energy consumption of the AI network.
  • idle computing resources such as the UE's own computing power.
  • FIGS. 8 to 12 in the embodiments of the present application are only to facilitate those skilled in the art to understand the embodiments of the present application, and are not intended to limit the embodiments of the present application to the specific illustrated scenarios.
  • Those skilled in the art can obviously make various equivalent modifications or changes based on the examples of FIGS. 8 to 12 , and such modifications or changes also fall within the scope of the embodiments of the present application.
  • AI node #1 in Figure 11 or Figure 12 can also be replaced with AI node #2, and accordingly, the AI-1 interface is replaced with the AI-6 interface.
  • AI connection and "AI interface” can be replaced with each other.
  • the terminal device communicates with the AI node #1 through the AI connection can also be replaced by “the terminal device communicates with the AI node #1 through the AI-1 interface.”
  • the terminal device and AI node #2 communicate through the AI connection can also be replaced by "the terminal device and the AI node #2 communicate through the AI-6 interface.”
  • the AI node #1 manages the AI-1 interface and the AI-6 interface is mainly used as an example for illustrative explanation, and the present application is not limited thereto.
  • AI node #1 manages the AI-1 interface
  • AI node #2 manages the AI-6 interface.
  • other devices manage the AI-1 interface and the AI-6 interface.
  • other devices configure, update, release resources for the AI-1 interface, and notify AI node #1.
  • other devices configure, update, and release resources for the AI-6 interface, and notify AI node #2; or first notify AI node #1, and then AI node #1 notifies AI node #2.
  • processing AI tasks is mentioned multiple times. It can be understood that processing AI tasks can also be replaced by executing AI tasks.
  • AI tasks between AI nodes and UEs are mainly used as an example.
  • the illustrative description is given, and the present application is not limited thereto.
  • AI tasks can also be released between AI nodes.
  • AI tasks can also be released between UEs.
  • the AI task released by A to B may be a complete task of the AI task or a partial task of the AI task. Accordingly, depending on the release of the AI task, B's completion of the AI task (or B's processing of the AI task) includes: completing the complete AI task, or completing part of the AI task.
  • the establishment of an AI connection may be initiated by the UE or the AI node.
  • the AI node may be a dedicated functional node, or a network device (such as a base station) with corresponding functions, or a terminal device.
  • sending messages is mentioned multiple times.
  • a sending a message to B may include A sending a message directly to B, or may include A sending a message to B through other devices, and there is no restriction on this.
  • the methods and operations implemented by devices can also be implemented by components of the device (such as chips or circuits). .
  • embodiments of the present application also provide corresponding devices, and the devices include modules for executing corresponding modules in each of the above method embodiments.
  • the module can be software, hardware, or a combination of software and hardware. It can be understood that the technical features described in the above method embodiments are also applicable to the following device embodiments.
  • FIG. 13 is a schematic block diagram of a communication device 1300 provided by an embodiment of the present application.
  • the device 1300 includes a transceiver unit 1310 and a processing unit 1320.
  • the transceiver unit 1310 may be used to implement corresponding communication functions.
  • the transceiver unit 1310 may also be called a communication interface or a communication unit.
  • the processing unit 1320 may be used for data processing.
  • the device 1300 is used to perform the steps or processes performed by the communication device in the embodiment shown in Figure 4, the steps or processes performed by the UE in the embodiments shown in Figures 8 and 10, and the steps or processes performed by the UE in the embodiment shown in Figure 11.
  • the steps or processes executed by UE1 are the steps or processes executed by the cooperating UE in the embodiment shown in Figure 12 .
  • the transceiver unit 1310 is used to receive artificial intelligence AI connection information.
  • the AI connection information includes at least one of the following configured for the AI connection: model, data set, computing resource; the processing unit 1320 is used to Establish an AI connection with the AI node based on the AI connection information.
  • the transceiver unit 1310 is also configured to send an AI message to an AI node through an AI connection, and/or receive an AI message from an AI node through an AI connection; where the AI message indicates at least one of the following information: the AI message uses Encryption method, compression method used for AI messages, message type of AI messages, control information or data carried by AI messages, verification code for integrity verification of AI messages, and content carried by AI messages.
  • the transceiver unit 1310 is also configured to receive updated AI connection information through the AI connection.
  • the processing unit 1320 is configured to perform at least one of the following: the information of the AI connection includes a model, and the model is processed; the information of the AI connection includes a data set, and measurements are made based on the data set; the information of the AI connection includes computing resources, and the use of computing Resources perform AI tasks.
  • the transceiver unit 1310 is also configured to issue a first AI task to an AI node through an AI connection; and/or receive a second AI task from an AI node through an AI connection.
  • the transceiver unit 1310 is specifically configured to issue a first AI task to an AI node through an AI connection and to issue the first AI task to other communication devices through the AI node.
  • the transceiver unit 1310 is also configured to send the processing result of the second AI task when the preset conditions are met.
  • the second AI task includes instruction information of preset conditions.
  • the transceiver unit 1310 is also configured to send a request message to the AI node, the request message is used to request the establishment of an AI connection; or, receive instruction information from the AI node, the instruction information is used to trigger the communication device to establish an AI connection with the AI node. .
  • the request message also includes indication information of the request type.
  • the request type is used to indicate the purpose of requesting to establish an AI connection.
  • the data set is determined according to the request type.
  • the information about the AI connection also includes at least one of the following: an identifier of the AI connection, and information about the radio bearer associated with the AI connection.
  • AI nodes are deployed on any of the following: network equipment, terminal equipment.
  • the device 1300 is used to execute the steps or processes executed by the AI node in the embodiment shown in Figure 4, the steps or processes executed by the AI node #1 in the embodiment shown in Figure 8, the embodiment shown in Figure 10
  • the steps or processes executed by AI node #1 and/or AI node #2 are the steps or processes executed by AI node #1 in the embodiments shown in Figures 11 and 12.
  • the transceiver unit 1310 is used to send AI connection information.
  • the AI connection information includes at least one of the following configured for the AI connection: model, data set, computing resource; the processing unit 1320 is used to configure the AI connection according to the AI connection information. Connected information and communication devices establish AI connections.
  • the transceiver unit 1310 is also configured to send an AI message to the communication device through the AI connection, and/or receive the AI message from the communication device through the AI connection; wherein the AI message indicates at least one of the following information: the AI message uses Encryption method, compression method used for AI messages, message type of AI messages, control information or data carried by AI messages, verification code for integrity verification of AI messages, and content carried by AI messages.
  • the transceiver unit 1310 is also configured to send updated AI connection information to the communication device through the AI connection.
  • the processing unit 1320 is configured to perform at least one of the following: the information of the AI connection includes a model, and the data sent by the communication device is processed according to the model; the information of the AI connection includes a data set, which is a management data set; the information of the AI connection Including computing resources, using computing resources to perform AI tasks.
  • processing unit 1320 is also configured to determine AI connection information according to the AI capability of the communication device.
  • the transceiver unit 1310 is also configured to receive the first AI task issued from at least one communication device through the AI connection between the AI node and each of the at least one communication device; and/or, through the AI node and each communication device.
  • the AI connection between the communication devices in the at least one communication device releases the second AI task to the at least one communication device; wherein the at least one communication device includes a communication device.
  • the transceiver unit 1310 is specifically used to communicate to the AI node through the AI connection between the AI node and the communication device.
  • the device issues the second AI task to issue the second AI task to other communication devices through the communication device.
  • the transceiver unit 1310 is also configured to send the processing of the first AI task to at least one communication device through the AI connection between the AI node and each of the at least one communication device when the preset conditions are met. result.
  • the first AI task includes instruction information of preset conditions.
  • the transceiver unit 1310 is also configured to receive a request message from the communication device, the request message is used to request the establishment of an AI connection; or, to send instruction information to the communication device, the instruction information is used to trigger the communication device to establish an AI connection with the AI node. .
  • the request message also includes indication information of the request type.
  • the request type is used to indicate the purpose of requesting to establish an AI connection.
  • the data set is determined according to the request type.
  • the information about the AI connection also includes at least one of the following: an identifier of the AI connection, and information about the radio bearer associated with the AI connection.
  • the apparatus 1300 includes a first AI node and a second AI node, and the first AI node and/or the second AI node satisfy any of the following: the first AI node is used for transmitting signaling, and the second AI node is used for Process AI tasks; the second AI node is used to process the AI operations indicated by the first AI node; the first AI node is used to process the first type of AI tasks, the second AI node is used to process the second type of AI tasks, and the first type of AI The task is not exactly the same as the second type of AI task; the second AI node is used to save and/or send the processing results of the AI task.
  • the transceiver unit 1310 is also used to send notification information to the second AI node.
  • the notification information is used to notify the second AI node of at least one of the following: executing the AI task, saving the processing result of the AI task, and sending the processing of the AI task. result.
  • AI nodes are deployed on any of the following: network equipment, terminal equipment.
  • the AI node is deployed on a network device, and the processing unit 1320 is further configured to determine information about wireless bearers associated with the AI connection based on at least one of the following: models, data sets, and computing resources.
  • the device 1300 here is embodied in the form of a functional unit.
  • the term "unit” as used herein may refer to an application specific integrated circuit (ASIC), an electronic circuit, a processor (such as a shared processor, a proprietary processor, or a group of processors) used to execute one or more software or firmware programs. processor, etc.) and memory, merged logic circuitry, and/or other suitable components to support the described functionality.
  • ASIC application specific integrated circuit
  • processor such as a shared processor, a proprietary processor, or a group of processors
  • memory merged logic circuitry, and/or other suitable components to support the described functionality.
  • the product implementation form of the device 1300 provided by the embodiment of this application is a program code that can be run on a computer.
  • the device 1300 provided in the embodiment of the present application may be a communication device, or may be a chip, chip system (for example: system on chip (SoC), SoC)) or circuit applied to the communication device.
  • the transceiver unit 1310 may be a transceiver, or an input/output interface
  • the processing unit 1320 may be a processor.
  • the transceiver unit 1310 may be an input/output interface, interface circuit, output circuit, input circuit, pin or Related circuits, etc.
  • the processing unit 1320 may be a processor, a processing circuit or a logic circuit, etc.
  • transceiver unit 1310 may also be a transceiver circuit (for example, it may include a receiving circuit and a transmitting circuit), and the processing unit may be a processing circuit.
  • Figure 14 is a schematic block diagram of a communication device 1400 provided by an embodiment of the present application.
  • the device 1400 includes A processor 1410 is coupled to a memory 1420.
  • a memory 1420 is also included for storing computer programs or instructions and/or data, and the processor 1410 is used to execute the computer programs or instructions stored in the memory 1420, or read the data stored in the memory 1420 to perform the above. Methods in Method Examples.
  • processors 1410 there are one or more processors 1410 .
  • the memory 1420 is integrated with the processor 1410, or is provided separately.
  • the device 1400 also includes a transceiver 1430, which is used for receiving and/or transmitting signals.
  • the processor 1410 is used to control the transceiver 1430 to receive and/or transmit signals.
  • the device 1400 is used to implement the operations performed by the communication device in each of the above method embodiments.
  • the processor 1410 is used to execute computer programs or instructions stored in the memory 1420 to implement related operations of the communication device in each of the above method embodiments.
  • the device 1400 is used to implement the operations performed by the AI node in each of the above method embodiments.
  • the processor 1410 is used to execute computer programs or instructions stored in the memory 1420 to implement related operations of the AI node in each of the above method embodiments.
  • the method executed by the AI node in the embodiment shown in Figure 4 or the method executed by the AI node #1 in the embodiment shown in Figure 8, or the AI node #1 and/or AI node #2 in the embodiment shown in Figure 10
  • each step of the above method can be completed by instructions in the form of hardware integrated logic circuits or software in the processor 1410 .
  • the method disclosed in conjunction with the embodiments of the present application can be directly implemented by a hardware processor for execution, or can be executed by a combination of hardware and software modules in the processor.
  • the software module can be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other mature storage media in this field.
  • the storage medium is located in the memory 1420.
  • the processor 1410 reads the information in the memory 1420 and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.
  • the processor may be one or more integrated circuits, used to execute relevant programs to execute the method embodiments of the present application.
  • a processor may include one or more processors and be implemented as a combination of computing devices.
  • the processor may include one or more of the following: microprocessor, microcontroller, digital signal processor (digital signal processor, DSP), digital signal processing device (digital signal processing device, DSPD), application specific integrated circuit (application specific integrated circuit (ASIC), field programmable gate array (FPGA), programmable logic device (PLD), gate logic, transistor logic, discrete hardware circuits, processing circuits or other suitable Hardware, firmware, and/or a combination of hardware and software to perform the various functions described in this disclosure.
  • the processor may be a general purpose processor or a special purpose processor.
  • processor 1410 may be a baseband processor or a central processing unit.
  • the baseband processor can be used to process communication protocols and communication data.
  • the central processing unit may be used to cause the device to execute software programs and process data in the software programs.
  • a portion of the processor may include non-volatile random access memory.
  • the processor may also store information about the device type.
  • Program in this application is used in a broad sense to mean software.
  • Non-limiting examples of software include: program code, program, subroutine, instruction, set of instructions, code, code segment, software module, application, or software application wait.
  • Programs can run on a processor and/or computer. To cause the device to perform various functions and/or processes described in this application.
  • Memory may store data required by a processor (eg, processor 1410) when executing software.
  • the memory can be implemented using any suitable storage technology.
  • memory may be any available storage medium accessible to the processor and/or computer.
  • Non-limiting examples of storage media include: random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EPROM, EEPROM), optical disks Read-only memory (Compact Disc-ROM, CD-ROM), static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM) , double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synchlink DRAM, SLDRAM) and Direct rambus RAM (DR RAM), removable media, optical disk storage, magnetic disk storage media, magnetic storage devices, flash memory, registers, state memory, remote
  • the memory eg, memory 1420
  • the processor eg, processor 1410
  • the memory may be used in connection with the processor to enable the processor to read information from the memory, store and/or write information in the memory.
  • the memory can be integrated into the processor.
  • the memory and processor may be provided in an integrated circuit (for example, the integrated circuit may be provided in a UE or BS or other network node).
  • FIG. 15 is a schematic block diagram of a chip system 1500 provided by an embodiment of the present application.
  • the chip system 1500 (or can also be called a processing system) includes a logic circuit 1510 and an input/output interface 1520.
  • the logic circuit 1510 may be a processing circuit in the chip system 1500 .
  • the logic circuit 1510 can be coupled to the memory unit and call instructions in the memory unit, so that the chip system 1500 can implement the methods and functions of various embodiments of the present application.
  • the input/output interface 1520 can be an input/output circuit in the chip system 1500, which outputs information processed by the chip system 1500, or inputs data or signaling information to be processed into the chip system 1500 for processing.
  • the chip system 1500 is used to implement the operations performed by the communication device in each of the above method embodiments.
  • the logic circuit 1510 is used to implement processing-related operations performed by the communication device in the above method embodiment, such as the processing-related operations performed by the communication device in the embodiment shown in Figure 4, or as shown in Figures 8 and 10
  • the input/output interface 1520 uses In order to implement the sending and/or receiving related operations performed by the communication device in the above method embodiment, such as the sending and/or receiving related operations performed by the communication device in the embodiment shown in FIG. 4, or FIG. 8 and FIG.
  • the sending and/or receiving related operations performed by the UE in the embodiment shown in Figure 10, or the sending and/or receiving related operations performed by UE1 in the embodiment shown in Figure 11, or the cooperating UE performed in the embodiment shown in Figure 12 Send and/or receive related operations.
  • the chip system 1500 is used to implement the operations performed by the AI node in each of the above method embodiments.
  • the logic circuit 1510 is used to implement processing-related operations performed by the AI node in the above method embodiment, such as the processing-related operations performed by the AI node in the embodiment shown in Figure 4, or in the embodiment shown in Figure 8 Processing-related operations performed by AI node #1, or processing-related operations performed by AI node #1 and/or AI node #2 in the embodiment shown in Figure 10, or AI nodes in the embodiments shown in Figures 11 and 12 Processing-related operations performed by #1; the input/output interface 1520 is used to implement the sending and/or receiving-related operations performed by the AI node in the above method embodiment, such as those performed by the AI node in the embodiment shown in Figure 4 Send and/or receive related operations, or send and/or receive related operations performed by AI node #1 in the embodiment shown in Figure 8, or AI node #1 and/or AI node # in the embodiment shown in Figure 10 2, or the sending and/or reception-related operations performed by AI node #1 in the embodiments shown in Figures 11 and 12.
  • Embodiments of the present application also provide a computer-readable storage medium on which are stored computer instructions for implementing the methods executed by the communication device or AI node in each of the above method embodiments.
  • Embodiments of the present application also provide a computer program product, which includes instructions.
  • the instructions are executed by a computer, the methods executed by the communication device or the AI node in each of the above method embodiments are implemented.
  • An embodiment of the present application also provides a communication system, which includes the communication device and AI node in the above embodiments.
  • the disclosed devices and methods can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the above units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or may be Integrated into another system, or some features can be ignored, or not implemented.
  • the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.
  • the units described above as separate components may or may not be physically separated.
  • the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to implement the solution provided by this application.
  • each functional unit in each embodiment of the present application can be integrated into one unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable device.
  • the computer may be a personal computer, a server, or a network device.
  • Computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, e.g., computer instructions may be transmitted from a website, computer, server or data center via a wired link (e.g. coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g. infrared, wireless, microwave, etc.) to another website, computer, server or data center.
  • a wired link e.g. coaxial cable, fiber optic, digital subscriber line (DSL)
  • wireless e.g. infrared, wireless, microwave, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请提供了一种通信方法、通信装置和系统。该方法包括：通信装置接收人工智能AI连接的信息，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；通信装置根据AI连接的信息与AI节点建立AI连接。通信装置基于包括模型、数据集、计算资源中至少一项的AI连接的信息与AI节点建立AI连接，从而通信装置和AI节点之间可基于该AI连接传输与AI相关的信令，执行与AI相关的操作，实现AI与无线网络的融合。

Description

通信方法、通信装置和系统
本申请要求于2022年06月24日提交中国专利局、申请号为202210725738.9、申请名称为“通信方法、通信装置和系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及无线通信领域，具体地，涉及应用了智能网络的无线通信技术，尤其涉及一种通信方法、通信装置和系统。
背景技术
为了应对未来智能普惠的愿景，智能化将在无线网络架构层面进一步演进，人工智能(artificial intelligence，AI)将与无线网络进一步深度的融合，实现网络内生的智能和终端的智能化，从而可以应对一些可能的新需求和新场景。如何实现AI与无线网络的融合，是亟需解决的问题。
发明内容
本申请提供一种通信方法、通信装置和系统，通过建立通信装置与AI节点之间的AI连接，从而可以实现使用无线网络中的通信装置执行AI任务。
第一方面，提供了一种通信方法，该方法可以由通信装置执行。该通信装置可以是通信设备(如终端设备、网络设备)，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以通信装置为例进行说明。
该方法可以包括：通信装置接收人工智能AI连接的信息，通信装置根据AI连接的信息与AI节点建立AI连接。
可选地，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源。
第二方面，提供了一种通信方法，该方法可以由通信装置执行。该通信装置可以是通信设备(如终端设备、网络设备)，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以通信装置为例进行说明。
该方法可以包括：通信装置接收人工智能AI连接的信息，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；通信装置根据AI连接的信息与AI节点建立AI连接。
基于上述第一方面或第二方面的技术方案，通信装置可接收AI连接的信息，并根据该AI连接的信息与AI节点建立AI连接，这样AI节点和通信装置之间可通过该AI连接通信，如发送和/或接收AI任务，又如发送和/或接收AI任务的处理结果等，从而实现AI与无线网络的融合，实现由无线网络中的通信装置执行AI相关的任务。
结合第一方面或第二方面，在某些实现方式中，方法还包括：通信装置通过AI连接 向AI节点发送AI消息，和/或，通信装置通过AI连接接收来自AI节点的AI消息；其中，AI消息指示以下至少一项信息：AI消息采用的加密方式、AI消息采用的压缩方式、AI消息的消息类型、AI消息承载的是控制信息或数据、对AI消息进行完整性校验的验证码、AI消息承载的内容。
结合第一方面或第二方面，在某些实现方式中，方法还包括：通信装置通过AI连接接收更新后的AI连接的信息。
基于上述技术方案，AI连接的信息可动态更新，且通信装置可接收更新后的AI连接的信息，这样可以根据通信情况动态更新AI连接的信息。
结合第一方面或第二方面，在某些实现方式中，方法还包括以下至少一项：AI连接的信息包括模型，通信装置处理模型；AI连接的信息包括数据集，通信装置根据数据集进行测量；AI连接的信息包括计算资源，通信装置使用计算资源执行AI任务。
结合第一方面或第二方面，在某些实现方式中，方法还包括：通信装置通过AI连接向AI节点发布第一AI任务；和/或，通信装置通过AI连接接收来自AI节点的第二AI任务。
基于上述技术方案，通信装置可以通过通信装置与AI节点之间的AI连接，向AI节点发布AI任务，或者也可以接收AI节点发布的任务。
结合第一方面或第二方面，在某些实现方式中，通信装置通过AI连接向AI节点发布第一AI任务，包括：通信装置通过AI连接向AI节点发布第一AI任务以通过AI节点向其他通信装置发布第一AI任务。
示例地，其他通信装置可以为其他AI节点或者网络设备，或者也可以是终端设备。
基于上述技术方案，通信装置可以通过通信装置与AI节点之间的AI连接，通过AI节点向其他通信装置发布AI任务，这样可以应用于更多的场景。例如，AI节点无法执行AI任务时，可通过AI节点向其他通信装置发布AI任务；再例如，若通信装置与其他通信装置之间无法直接通信，也可以通过AI节点向其他通信装置发布AI任务。
结合第一方面或第二方面，在某些实现方式中，方法还包括：在满足预设条件的情况下，通信装置发送第二AI任务的处理结果。
基于上述技术方案，通信装置可以在满足预设条件的情况下再上报任务的处理结果。
结合第一方面或第二方面，在某些实现方式中，第二AI任务中包括预设条件的指示信息。
结合第一方面或第二方面，在某些实现方式中，方法还包括：通信装置向AI节点发送请求消息，请求消息用于请求建立AI连接；或者，通信装置接收来自AI节点的指示信息，指示信息用于触发通信装置与AI节点建立AI连接。
基于上述技术方案，在通信装置与AI节点建立AI连接之前，可通过通信装置发起请求的方式与AI节点建立AI连接，或者通过AI节点触发的方式与通信装置建立AI连接。从而通信装置或AI节点可以在需要通过AI连接执行相关操作时，如发布AI任务时，再建立AI连接，避免建立了AI连接后，不使用该AI连接造成的资源浪费。
结合第一方面或第二方面，在某些实现方式中，请求消息还包括请求类型的指示信息，请求类型用于表示请求建立AI连接的目的，数据集是根据请求类型确定的。
结合第一方面或第二方面，在某些实现方式中，AI连接的信息还包括以下至少一项：AI连接的标识、与AI连接关联的无线承载的信息。
结合第一方面或第二方面，在某些实现方式中，AI节点部署于以下任一项：网络设备、终端设备。
结合第一方面或第二方面，在某些实现方式中，方法还包括：通信装置发送通信装置的AI能力。
示例地，通信装置向核心网发送通信装置的AI能力。
结合第一方面或第二方面，在某些实现方式中，通信装置的AI能力包括以下至少一项：通信装置支持的AI任务类型、通信装置的硬件能力、通信装置的算力能力、通信装置的优先级。
第三方面，提供了一种通信方法，该方法可以由AI节点执行。该AI节点可以是通信设备，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以AI节点为例进行说明。
该方法可以包括：人工智能AI节点发送AI连接的信息，AI节点根据AI连接的信息与通信装置建立AI连接。
可选地，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源。
第四方面，提供了一种通信方法，该方法可以由AI节点执行。该AI节点可以是通信设备，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以AI节点为例进行说明。
该方法可以包括：人工智能AI节点发送AI连接的信息，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；AI节点根据AI连接的信息与通信装置建立AI连接。
结合第三方面或第四方面，在某些实现方式中，方法还包括：AI节点通过AI连接向通信装置发送AI消息，和/或，AI节点通过AI连接接收来自通信装置的AI消息；其中，AI消息指示以下至少一项信息：AI消息采用的加密方式、AI消息采用的压缩方式、AI消息的消息类型、AI消息承载的是控制信息或数据、对AI消息进行完整性校验的验证码、AI消息承载的内容。
结合第三方面或第四方面，在某些实现方式中，方法还包括：AI节点通过AI连接向通信装置发送更新后的AI连接的信息。
结合第三方面或第四方面，在某些实现方式中，方法还包括以下至少一项：AI连接的信息包括模型，AI节点根据模型对通信装置发送的数据进行处理；AI连接的信息包括数据集，AI节点管理数据集；AI连接的信息包括计算资源，AI节点使用计算资源执行AI任务。
结合第三方面或第四方面，在某些实现方式中，方法还包括：AI节点根据通信装置的AI能力确定AI连接的信息。
结合第三方面或第四方面，在某些实现方式中，方法还包括：AI节点通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，接收来自至少一个通信装置发布的第一AI任务；和/或，AI节点通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，向至少一个通信装置发布第二AI任务；其中，至少一个通信装置包括通信装置。
结合第三方面或第四方面，在某些实现方式中，AI节点通过AI节点与通信装置之间的AI连接向通信装置发布第二AI任务，包括：AI节点通过AI节点与通信装置之间的AI 连接向通信装置发布第二AI任务以通过通信装置向其他通信装置发布第二AI任务。
结合第三方面或第四方面，在某些实现方式中，方法还包括：在满足预设条件的情况下，AI节点通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，向至少一个通信装置发送第一AI任务的处理结果。
结合第三方面或第四方面，在某些实现方式中，第一AI任务包括预设条件的指示信息。
结合第三方面或第四方面，在某些实现方式中，方法还包括：AI节点接收来自通信装置的请求消息，请求消息用于请求建立AI连接；或者，AI节点向通信装置发送指示信息，指示信息用于触发通信装置与AI节点建立AI连接。
结合第三方面或第四方面，在某些实现方式中，请求消息还包括请求类型的指示信息，请求类型用于表示请求建立AI连接的目的，数据集是根据请求类型确定的。
结合第三方面或第四方面，在某些实现方式中，AI连接的信息还包括以下至少一项：AI连接的标识、与AI连接关联的无线承载的信息。
结合第三方面或第四方面，在某些实现方式中，AI节点包括第一AI节点和第二AI节点，第一AI节点和/或第二AI节点满足以下任一项：第一AI节点用于传输信令，第二AI节点用于处理AI任务；第二AI节点用于处理第一AI节点指示的AI操作；第一AI节点用于处理第一类AI任务，第二AI节点用于处理第二类AI任务，第一类AI任务与第二类AI任务不完全相同；第二AI节点用于保存和/或发送AI任务的处理结果。
结合第三方面或第四方面，在某些实现方式中，方法还包括：第一AI节点向第二AI节点发送通知信息，通知信息用于通知第二AI节点以下至少一项：执行AI任务、保存AI任务的处理结果、发送AI任务的处理结果。
结合第三方面或第四方面，在某些实现方式中，AI节点部署于以下任一项：网络设备、终端设备。
结合第三方面或第四方面，在某些实现方式中，AI节点部署于网络设备，方法还包括：网络设备根据以下至少一项确定与AI连接关联的无线承载的信息：模型、数据集、计算资源。
第三方面或第四方面及各个可能的设计的有益效果可以参考第一方面或第二方面相关的描述，在此不予赘述。
第五方面，提供了一种通信方法，该方法可以由通信装置执行。该通信装置可以是通信设备(如终端设备、网络设备)，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以通信装置为例进行说明。
该方法可以包括：通信装置通过通信装置与人工智能AI节点之间的AI连接向AI节点发布AI任务；通信装置接收AI任务的处理结果。
基于上述技术方案，通信装置可以通过通信装置与AI节点之间的AI连接，向AI节点发布AI任务，从而可实现AI与无线网络的融合，实现由无线网络中的装置(如AI节点，又如其他通信装置)执行AI任务。
结合第五方面，在第五方面的某些实现方式中，通信装置通过通信装置与人工智能AI节点之间的AI连接向AI节点发布AI任务，包括：通信装置通过AI连接向AI节点发布AI任务以通过AI节点向其他通信装置发布AI任务。
基于上述技术方案，通信装置可以通过通信装置与AI节点之间的AI连接，通过AI 节点向其他通信装置发布AI任务，这样可以应用于更多的场景。例如，AI节点无法执行AI任务时，可通过AI节点向其他通信装置发布AI任务；再例如，若通信装置与其他通信装置之间无法直接通信，也可以通过AI节点向其他通信装置发布AI任务。
结合第五方面，在第五方面的某些实现方式中，方法还包括：通信装置向AI节点发送请求消息，请求消息用于请求建立AI连接；或者，通信装置接收来自AI节点的指示信息，指示信息用于触发通信装置与AI节点建立AI连接。
结合第五方面，在第五方面的某些实现方式中，AI节点部署于以下任一项：网络设备、终端设备。
第六方面，提供了一种通信方法，该方法可以由AI节点执行。该AI节点可以是通信设备，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以AI节点为例进行说明。
该方法可以包括：人工智能AI节点通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，接收来自至少一个通信装置发布的AI任务；AI节点执行AI任务，或者，AI节点向其他通信装置发布AI任务。
结合第六方面，在第六方面的某些实现方式中，至少一个通信装置包括第一通信装置，方法还包括：AI节点通过AI节点与第一通信装置之间的AI连接，向第一通信装置发送AI任务的处理结果。
结合第六方面，在第六方面的某些实现方式中，AI节点通过AI节点与第一通信装置之间的AI连接，向第一通信装置发送AI任务的处理结果，包括：在满足预设条件的情况下，AI节点通过AI节点与第一通信装置之间的AI连接，向第一通信装置发送AI任务的处理结果。
结合第六方面，在第六方面的某些实现方式中，AI任务包括预设条件的指示信息。
结合第六方面，在第六方面的某些实现方式中，AI节点向其他通信装置发布AI任务，方法还包括：AI节点接收来自其他通信装置发送的AI任务的处理结果。
结合第六方面，在第六方面的某些实现方式中，至少一个通信装置包括第二通信装置，方法还包括：AI节点接收来自第二通信装置的请求消息，请求消息用于请求建立AI节点与第二通信装置之间的AI连接；或者，AI节点向第二通信装置发送指示信息，指示信息用于触发第二通信装置与AI节点建立AI节点与第二通信装置之间的AI连接。
结合第六方面，在第六方面的某些实现方式中，AI节点部署于以下任一项：网络设备、终端设备。
第六方面及各个可能的设计的有益效果可以参考第五方面相关的描述，在此不予赘述。
第七方面，提供了一种通信方法，该方法可以由通信装置执行。该通信装置可以是通信设备(如终端设备、网络设备)，也可以是用于通信设备的芯片或电路，本申请对此不作限定。为了便于描述，下面以通信装置为例进行说明。
该方法可以包括：通信装置通过通信装置与人工智能AI节点之间的AI连接接收来自AI节点的AI任务；通信装置执行AI任务，或者，通信装置向其他通信装置发布AI任务。
基于上述技术方案，AI节点可以通过通信装置与通信装置之间的AI连接，向通信装置或向其他通信装置发布AI任务。AI节点可以将AI任务发布给通信装置，进而由通信装置处理该AI任务，这样可实现利用空闲的计算资源(如通信装置自身的算力)完成AI 任务，减轻AI网络的负载和能耗。
结合第七方面，在第七方面的某些实现方式中，方法还包括：通信装置通过AI连接向AI节点发送AI任务的处理结果。
结合第七方面，在第七方面的某些实现方式中，通信装置通过AI连接向AI节点发送AI任务的处理结果，包括：在满足预设条件的情况下，通信装置通过AI连接向AI节点发送AI任务的处理结果。
结合第七方面，在第七方面的某些实现方式中，AI任务包括预设条件的指示信息。
结合第七方面，在第七方面的某些实现方式中，通信装置向其他通信装置发布AI任务，方法还包括：通信装置接收来自其他通信装置发送的AI任务的处理结果。
结合第七方面，在第七方面的某些实现方式中，方法还包括：通信装置向AI节点发送请求消息，请求消息用于请求建立AI连接；或者，通信装置接收来自AI节点的指示信息，指示信息用于触发通信装置与AI节点建立AI连接。
结合第七方面，在第七方面的某些实现方式中，AI节点部署于以下任一项：网络设备、终端设备。
第八方面，提供了一种通信方法，该方法可以由AI节点执行。该AI节点可以是通信设备，也可以是用于通信设备的芯片或电路执行，本申请对此不作限定。为了便于描述，下面以AI节点为例进行说明。
该方法可以包括：人工智能AI节点通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，向至少一个通信装置发布AI任务；AI节点接收AI任务的处理结果。
结合第八方面，在第八方面的某些实现方式中，至少一个通信装置包括第一通信装置，AI节点通过AI节点与第一通信装置之间的AI连接，向第一通信装置发布AI任务，包括：AI节点向第一通信装置发布AI任务以通过第一通信装置向其他通信装置发布AI任务。
结合第八方面，在第八方面的某些实现方式中，至少一个通信装置包括第二通信装置，方法还包括：AI节点接收来自第二通信装置的请求消息，请求消息用于请求建立AI节点与第二通信装置之间的AI连接；或者，AI节点向第二通信装置发送指示信息，指示信息用于触发第二通信装置与AI节点建立AI节点与第二通信装置之间的AI连接。
结合第八方面，在第八方面的某些实现方式中，AI节点部署于以下任一项：网络设备、终端设备。
第八方面及各个可能的设计的有益效果可以参考第七方面相关的描述，在此不予赘述。
第九方面，提供一种通信装置，该装置用于执行上述第一方面至第八方面中任一方面提供的方法。具体地，该装置可以包括用于执行第一方面至第八方面中任一方面的上述任一种实现方式提供的方法的单元和/或模块，如处理单元和/或通信单元。
在一种实现方式中，该装置为通信设备(如终端设备，又如AI节点，又如网络设备)。当该装置为通信设备时，通信单元可以是收发器，或，输入/输出接口；处理单元可以是至少一个处理器。可选地，收发器可以为收发电路。可选地，输入/输出接口可以为输入/输出电路。
在另一种实现方式中，该装置为用于通信设备中的芯片、芯片系统或电路。当该装置为用于终端设备中的芯片、芯片系统或电路时，通信单元可以是该芯片、芯片系统或电路上的输入/输出接口、接口电路、输出电路、输入电路、管脚或相关电路等；处理单元可以 是至少一个处理器、处理电路或逻辑电路等。
第十方面，提供一种通信装置，该装置包括：存储器，用于存储程序；至少一个处理器，用于执行存储器存储的计算机程序或指令，以执行上述第一方面至第八方面中任一方面的上述任一种实现方式提供的方法。
在一种实现方式中，该装置为通信设备(如终端设备，又如AI节点，又如网络设备)。
在另一种实现方式中，该装置为用于通信设备中的芯片、芯片系统或电路。
第十一方面，本申请提供一种处理器，用于执行上述各方面提供的方法。
对于处理器所涉及的发送和获取/接收等操作，如果没有特殊说明，或者，如果未与其在相关描述中的实际作用或者内在逻辑相抵触，则可以理解为处理器输出和输入等操作，也可以理解为由射频电路和天线所进行的发送和接收操作，本申请对此不做限定。
第十二方面，提供一种计算机可读存储介质，该计算机可读介质存储用于设备执行的程序代码，该程序代码包括用于执行上述第一方面至第八方面中任一方面的上述任一种实现方式提供的方法。
第十三方面，提供一种包含指令的计算机程序产品，当该计算机程序产品在计算机上运行时，使得计算机执行上述第一方面至第八方面中任一方面的上述任一种实现方式提供的方法。
第十四方面，提供一种芯片，芯片包括处理器与通信接口，处理器通过通信接口读取存储器上存储的指令，执行上述第一方面至第八方面中任一方面的上述任一种实现方式提供的方法。
可选地，作为一种实现方式，芯片还包括存储器，存储器中存储有计算机程序或指令，处理器用于执行存储器上存储的计算机程序或指令，当计算机程序或指令被执行时，处理器用于执行上述第一方面至第八方面中任一方面的上述任意一种实现方式提供的方法。
第十五方面，提供一种通信系统，包括上文的通信装置和AI节点。
附图说明
图1示出了适用于本申请实施例的无线通信系统100的示意图。
图2示出了根据本申请一实施例的通信系统的示意图。
图3示出了根据本申请另一实施例的通信系统的示意图。
图4是本申请一实施例提供的通信方法400的示意图。
图5示出了终端设备与AI节点#1通信的协议栈的示意图。
图6示出了AI消息的信元格式的示意图。
图7示出了UE与AI节点#2通信的协议栈的示意图。
图8示出了根据本申请一实施例提供的通信方法800的示意性流程图。
图9示出了无线承载的示意图。
图10示出了根据本申请另一实施例提供的通信方法1000的示意性流程图。
图11示出了根据本申请另一实施例提供的通信方法1100的示意性流程图。
图12示出了根据本申请另一实施例提供的通信方法1200的示意性流程图。
图13是本申请实施例提供的一种通信装置1300的示意性框图。
图14是本申请实施例提供的一种通信装置1400的示意性框图。
图15是本申请实施例提供的一种芯片系统1500的示意性框图。
具体实施方式
下面将结合附图，对本申请实施例中的技术方案进行描述。
本申请提供的技术方案可以应用于各种通信系统，例如：第五代(5th generation，5G)或新无线(new radio，NR)系统、长期演进(long term evolution，LTE)系统、LTE频分双工(frequency division duplex，FDD)系统、LTE时分双工(time division duplex，TDD)系统等。本申请提供的技术方案还可以应用于未来的通信系统，如第六代移动通信系统。本申请提供的技术方案还可以应用于设备到设备(device to device，D2D)通信，车到万物(vehicle-to-everything，V2X)通信，机器到机器(machine to machine，M2M)通信，机器类型通信(machine type communication，MTC)，以及物联网(internet of things，IoT)通信系统或者其他通信系统。
本申请实施例中的终端设备包括各种具有无线通信功能的设备，其可用于连接人、物、机器等。终端设备可以广泛应用于各种场景，例如：蜂窝通信，D2D，V2X，端到端(peer to peer，P2P)，M2M，MTC，IoT，虚拟现实(virtual reality，VR)，增强现实(augmented reality，AR)，工业控制，自动驾驶，远程医疗，智能电网，智能家具，智能办公，智能穿戴，智能交通，智慧城市无人机，机器人，遥感，被动传感，定位，导航与跟踪，自主交付等场景。终端设备可以是上述任一场景下的终端，如MTC终端、IoT终端等。终端设备可以是第三代合作伙伴项目(3rd generation partnership project，3GPP)标准的用户设备(user equipment，UE)、终端(terminal)、固定设备、移动台(mobile station)设备或者说移动设备、用户单元(subscriber unit)、手持设备、车载设备、可穿戴设备、蜂窝电话(cellular phone)、智能电话(smart phone)、SIP电话、无线数据卡、个人数字助理(personal digital assistant，PDA)、电脑、平板电脑、笔记本电脑、无线调制解调器、手持设备(handset)、膝上型电脑(laptop computer)、具有无线收发功能的计算机、智能书、车辆、卫星、全球定位系统(global positioning system，GPS)设备、目标跟踪设备、飞行器(例如无人机、直升机、多直升机、四直升机、或飞机等)、船只、遥控设备智能家居设备、工业设备，或者内置于上述设备中的装置(例如，上述设备中的通信模块、调制解调器或芯片等)，或者连接到无线调制解调器的其它处理设备。为了描述方便，下文将终端设备以终端或UE为例来描述。
应理解，在某些场景下，UE还可以用于充当基站。例如，UE可以充当调度实体，其在V2X、D2D或P2P等场景中的UE之间提供侧行链路信号。
本申请实施例中，用于实现终端设备的功能的装置可以是终端设备，也可以是能够支持终端设备实现该功能的装置，例如芯片系统或芯片，该装置可以被安装在终端设备中。本申请实施例中，芯片系统可以由芯片构成，也可以包括芯片和其他分立器件。
本申请实施例中的网络设备可以是用于与终端设备通信的设备，该网络设备也可以称为接入网设备或无线接入网设备，如网络设备可以是基站。本申请实施例中的网络设备可以是指将终端设备接入到无线网络的无线接入网(radio access network，RAN)节点(或设备)。基站可以广义的覆盖如下中的各种名称，或与如下名称进行替换，比如：节点B(NodeB)、演进型基站(evolved NodeB，eNB)、下一代基站(next generation NodeB，gNB)、中继站、接入点、传输点(transmitting and receiving point，TRP)、发射点(transmitting point，TP)、主站、辅站、多制式无线(motor slide retainer，MSR)节点、家庭基站、网 络控制器、接入节点、无线节点、接入点(AP)、传输节点、收发节点、基带单元(BBU)、射频拉远单元(remote radio unit，RRU)、有源天线单元(active antenna unit，AAU)、射频头(remote radio head，RRH)、中心单元(central unit，CU)、分布式单元(distributed unit，DU)、定位节点等。基站可以是宏基站、微基站、中继节点、施主节点或类似物，或其组合。基站还可以指用于设置于前述设备或装置内的通信模块、调制解调器或芯片。基站还可以是移动交换中心以及D2D、V2X、M2M通信中承担基站功能的设备、6G网络中的网络侧设备、未来的通信系统中承担基站功能的设备等。基站可以支持相同或不同接入技术的网络。本申请的实施例对网络设备所采用的具体技术和具体设备形态不做限定。
基站可以是固定的，也可以是移动的。例如，直升机或无人机可以被配置成充当移动基站，一个或多个小区可以根据该移动基站的位置移动。在其他示例中，直升机或无人机可以被配置成用作与另一基站通信的设备。
网络设备和终端设备可以部署在陆地上，包括室内或室外、手持或车载；也可以部署在水面上；还可以部署在空中的飞机、气球和卫星上。本申请实施例中对网络设备和终端设备所处的场景不做限定。
首先简单介绍适用于本申请的网络架构，如下。
图1示出了适用于本申请实施例的无线通信系统100的示意图。如图1所示，该无线通信系统100可以包括至少一个网络设备，例如图1所示的网络设备110，该无线通信系统100还可以包括至少一个终端设备，例如图1所示的终端设备120和终端设备130。网络设备和终端设备均可配置多个天线，网络设备与终端设备可使用多天线技术通信。终端设备与终端设备之间可以直接进行通信。
其中，网络设备和终端设备通信时，网络设备可以管理一个或多个小区，一个小区中可以有整数个终端设备。可选地，网络设备110和终端设备120组成一个单小区通信系统，不失一般性，将小区称为小区#1。网络设备110可以是小区#1中的网络设备，或者，网络设备110可以为小区#1中的终端设备(例如终端设备120)服务。
需要说明的是，小区可以理解为网络设备的无线信号覆盖范围内的区域。
应理解，图1为便于理解而示例的简化示意图，该无线通信系统100中还可以包括其他网络设备或者还可以包括其他终端设备，图1中未予以画出。本申请实施例可以适用于发送端设备和接收端设备通信的任何通信场景。
为了应对未来智能普惠的愿景，智能化将在无线网络架构层面进一步演进，人工智能(artificial intelligence，AI)将与无线网络进一步深度的融合，实现网络内生的智能和终端的智能化，从而可以应对一些可能的新需求和新场景。例如，一可能的场景，终端类型多样化，终端连接更加灵活和智能。终端类型多样化，超级物联网(supper IoT)(如物联，车联，工业，医疗等等)，海量连接，终端连接更加灵活，终端本身具备一定的AI能力。又如，一可能的需求，网络内生智能。网络除了提供传统的通信连接服务，还可能会提供计算和AI服务，来更好的支持普惠性、实时性和高安全的AI服务。这些新需求和新场景，可能会带来无线网络架构和通信模式的变化。
目前，3GPP在5G网络中通过新增网络数据分析功能(network data analytics function，NWDAF)，引入了AI能力。NWDAF的主要功能包括：支持从其他网络功能(network function，NF)和应用功能(application function，AF)收集数据，支持从网络运维系统(如操作维护管理(operation administration and maintenance，OAM))中收集数据，并可以向 NF或向AF提供元数据开放服务、数据分析服务等。NWDAF的引入，主要目标包括：网络运维的自动化和智能化、网络性能和业务体验优化、端到端服务级别协议(service level agreement，SLA)保障等。NWDAF训练的AI模型可应用于移动性管理、会话管理和网络自动化等网络自身领域，使用AI的方法来替换原有网络功能中基于数值公式的方法。但是NWDAF部署于核心网，属于外挂式AI单元，并未与通信网络作强耦合的设计，性能有局限。
基于未来无线网络面对的可能的场景和需求，通信网络中终端设备的数量和种类可能也会急速增长，终端设备采集、处理、产生的大量数据，可以为AI技术的应用提供动力。在这种背景下，本申请考虑到：一方面，AI相关的任务对算力的要求高、AI网络建设的负载、能耗高；另一方面，网络中可能会有很多终端设备的算力闲置。因此，本申请提出在通信网络中打造内生AI的功能，使得通信网络与AI结合更加紧密，提供更好的AI服务。
需要说明的是，在本申请中，“指示”可以包括直接指示、间接指示、显示指示、隐式指示。当描述某一指示信息用于指示A时，可以理解为该指示信息携带A、直接指示A，或间接指示A。
本申请中，指示信息所指示的信息，称为待指示信息。在具体实现过程中，对待指示信息进行指示的方式有很多种，例如但不限于，可以直接指示待指示信息，如待指示信息本身或者该待指示信息的索引等。也可以通过指示其他信息来间接指示待指示信息，其中该其他信息与待指示信息之间存在关联关系。还可以仅仅指示待指示信息的一部分，而待指示信息的其他部分则是已知的或者提前约定的。例如，还可以借助预先约定(例如协议规定)的各个信息的排列顺序来实现对特定信息的指示，从而在一定程度上降低指示开销。
待指示信息可以作为一个整体一起发送，也可以分成多个子信息分开发送，而且这些子信息的发送周期和/或发送时机可以相同，也可以不同。具体发送方法本申请不进行限定。其中，这些子信息的发送周期和/或发送时机可以是预先定义的，例如根据协议预先定义的，也可以是发射端设备通过向接收端设备发送配置信息来配置的。其中，该配置信息可以例如但不限于包括无线资源控制信令、媒体接入控制(media access control，MAC)层信令和物理层信令中的一种或者至少两种的组合。其中，无线资源控制信令例如包无线资源控制(radio resource control，RRC)信令；MAC层信令例如包括MAC控制元素(control element，CE)；物理层信令例如包括下行控制信息(downlink control information，DCI)。
首先，结合图2和图3介绍根据本申请实施例提供的通信系统。需要说明的是，以下示例中所使用的接口均为示例，本申请还可以采用其他同等功能的接口，本申请对接口的使用并不限制。
图2示出了根据本申请一实施例的通信系统的示意图。如图2所示，该通信系统可以包括终端设备和AI节点#1。终端设备与AI节点#1之间可通过AI-1接口通信，也即该AI-1接口可用于发送和/或接收与AI相关的信令。例如，终端设备可通过AI-1接口向AI节点#1发布AI任务；再例如，AI节点#1可通过AI-1接口向终端设备发布AI任务；再例如，终端设备可通过AI-1接口向AI节点#1发送和/或接收AI任务的处理结果；再例如，AI节点#1可通过AI-1接口向终端设备发送和/或接收AI任务的处理结果；再例如，终端设备可通过AI-1接口请求AI节点#1向其他装置(如其他AI节点，又如其他终端设备，又如网络设备)发布AI任务；再例如，AI节点#1可通过AI-1接口请求终端设备向其他 装置(如其他AI节点，又如其他终端设备，又如网络设备)发布AI任务。
在本申请中，AI任务，表示与AI相关的任务。作为示例，AI任务例如可以包括：模型失效测试、模型性能测试、模型训练测试、数据采集等。
在本申请中，终端设备与AI节点#1之间通过AI-1接口通信，也可替换为，终端设备与AI节点#1之间通过AI连接(AI connection)通信。其中，该AI连接，表示终端设备与AI节点#1之间的连接，通过该AI连接，终端设备与AI节点#1可传输与AI相关的信令、执行与AI相关的操作。可以理解，AI连接表征了终端设备与AI节点#1间的一种连接关系，是一个逻辑概念，而非一个物理实体，对此下文不再赘述。
其中，AI-1接口可以是逻辑接口。AI-1接口可通过终端设备与网络设备之间的连接、以及网络设备与AI节点#1之间的连接实现。举例来说，终端设备通过AI-1接口向AI节点#1发送消息，可以包括：终端设备向网络设备发送该消息，网络设备向AI节点#1转发该消息，也即网络设备可类似于终端设备与AI节点#1之间的中继设备。
可选地，AI节点#1管理AI-1接口。例如，AI节点#1负责建立、维护、释放AI-1接口。再例如，AI节点#1负责配置、更新、释放用于AI-1接口的资源。
可选地，AI节点#1部署于以下任一项：网络设备、终端设备。可以理解，AI节点#1也可以单独部署，即部署于网络设备和终端设备之外的位置，如靠近网络设备的位置，又如靠近终端设备的位置，不予限制。此外，关于AI节点#1的数量不予限制。
可选地，该通信系统还包括网络设备。网络设备与终端设备之间可以通信，网络设备与AI节点#1之间也可以通信。如图3所示，网络设备与终端设备之间可通过UE-通用陆地无线接入网(UE-universal terrestrial radio access aetwork，Uu)接口通信，AI节点#1与网络设备之间可通过AI-2接口通信。
可选地，该通信系统还包括核心网。核心网与网络设备之间可以通信，核心网与AI节点#1之间也可以通信。如图3所示，核心网与网络设备之间可通过下一代(next generation，NG)接口通信，核心网与AI节点#1之间可通过AI-3接口通信。
核心网可以是一种部署在无线网络中的装置和/或软件系统。核心网可以包括一个或多个核心网节点，以提供核心网功能，如提供UE连接管理、移动性管理、策略管理等功能。核心网还可以提供到外部网络如互联网的用户面网关功能。核心网节点可以是下一代(例如，6G或更高版本)核心网节点，或传统(例如，5G、第四代(4th generation，4G)、第三代(3th generation，3G)或第二代(2th generation，2G))核心网节点。一可能设计，核心网可以包括接入和移动性管理功能(access and mobility management function，AMF)、统一数据管理(unified data management，UDM)等网元。核心网还可以包括公共交换电话网络(public switched telephone network，PSTN)、分组数据网络、光网络、网际协议(internet protocol，IP)网络中的一个或多个网络节点、广域网(wide area network，WAN)、局域网(local area network，LAN)、无线局域网(wireless local area network，WLAN)、有线网络、无线网络、城域网和其他网络，以便使终端设备和/或网络设备之间能够进行通信。
在采用不同的无线接入技术的系统中，具备核心网功能的设备的名称，以及其中包含的网络功能可能会有所不同。例如，4G核心网一般称为演进型分组核心网(evolved packet core，EPC)；5G核心网一般称为5G核心网(5G core network，5GC或5GCN)。
可选地，核心网(例如UDM)中存储终端设备的AI能力。例如，核心网中存储终端设备的签约信息，终端设备的签约信息中包括终端设备的AI能力。若AI节点#1要查询 终端设备的AI能力，则AI节点#1可通过AI-3接口向核心网查询终端设备的AI能力。终端设备的AI能力，也可以称为终端设备的AI相关参数，下文统一用终端设备的AI能力描述。
其中，终端设备的AI能力例如可以包括以下至少一项：终端设备的优先级、终端设备支持的最大算力、终端设备的硬件能力、终端设备支持的AI任务类型。作为一示例，终端设备的优先级，可以根据终端设备的历史响应情况确定。举例来说，若终端设备参与协作处理AI任务的次数较多，则终端设备的优先级较高；若终端设备参与协作处理AI任务的次数较少，则终端设备的优先级较低。作为另一示例，终端设备的优先级，可以根据终端设备的能力(如支持的最大算力，又如终端设备本身的硬件能力等)确定。举例来说，若终端设备的能力较高，则终端设备的优先级较高；若终端设备的能力较低，则终端设备的优先级较低。
可以理解，上述为示例性说明，对此不予限制，例如终端设备的AI能力还可包括终端设备的安全要求等。
还可以理解，核心网(例如UDM)中存储终端设备的AI能力为示例性说明，本申请并未限定于此。例如，AI节点(如AI节点#1)中存储终端设备的AI能力，若AI节点#1要查询终端设备的AI能力，则可直接本地确定终端设备的AI能力，或者从其他AI节点获取终端设备的AI能力。
图3示出了根据本申请另一实施例的通信系统的示意图。与图2所示的系统相比，图3所示的通信系统还包括AI节点#2，图3仅为示例，AI节点不限于两个，可以有多个。该AI节点#2与通信系统中的其它设备之间可以通信。例如，该AI节点#2与终端设备之间可通过AI-6接口通信，该AI节点#2与AI节点#1之间可通过AI-4接口通信，该AI节点#2与网络设备之间可通过AI-5接口通信。可选地，AI节点#2部署于以下任一项：网络设备、终端设备。可以理解，AI节点#2也可以单独部署，即部署于网络设备和终端设备之外的位置，如靠近网络设备的位置，又如靠近终端设备的位置，不予限制。此外，关于AI节点#2的数量不予限制。
在本申请中，终端设备与AI节点#2之间通过AI-6接口通信，也可替换为，终端设备与AI节点#2之间通过AI连接通信。其中，该AI连接，表示终端设备与AI节点#2之间的AI连接，通过该AI连接，终端设备与AI节点#2可传输与AI相关的信令、执行与AI相关的操作。
其中，该AI-6接口可以是逻辑接口。AI-6接口可通过终端设备与网络设备之间的连接、以及网络设备与AI节点#2之间的连接实现。举例来说，终端设备通过AI-6接口向AI节点#2发送消息，可以包括：终端设备向网络设备发送该消息，网络设备向AI节点#2转发该消息，也即网络设备可类似于终端设备与AI节点#2之间的中继设备。
可选地，AI节点#1负责管理AI-6接口。例如，AI节点#1负责建立、维护、释放AI-6接口。再例如，AI节点#1负责用于AI-6接口的资源，如配置、更新、释放用于AI-6接口的资源。
AI节点#1和AI节点#2可以是负责不同功能的AI节点。
第一种可能的设计，AI节点#1负责信令传输，AI节点#2负责执行处理操作。其中，AI节点#2负责执行处理操作，例如可以包括以下至少一项：AI节点#2处理AI任务、AI节点#2维护AI任务的处理结果、AI节点#2维护AI相关的数据集、AI节点#2维护AI相 关的模型。
第二种可能的设计，AI节点#2负责执行AI节点#1指示的各种操作。例如，AI节点#1指示AI节点#2处理AI任务，AI节点#2收到AI节点#1的指示后，处理相应的AI任务。可选地，AI节点#2还用于向终端设备(如通过AI-6接口)发送AI任务的处理结果。
第三种可能的设计，AI节点#1负责处理第一类AI任务，AI节点#2负责处理第二类AI任务，该第一类AI任务和第二类AI任务不完全相同。作为示例，处理第一类AI任务产生的开销小于处理第二类AI任务产生的开销。在该设计下，AI节点#1和AI节点#2相当于是通信系统中的两个不同的AI节点，该两个AI节点可分别负责不同的AI任务。
可以理解，上述关于AI节点#1和AI节点#2的功能划分为示例性说明，关于AI节点#1和AI节点#2的功能划分，本申请不予限制。
还可以理解，AI节点#1和AI节点#2，可以是各自独立的设备，也可以集成于同一设备中实现不同的功能，或者可以是硬件设备中的网络元件，也可以是在专用硬件上运行的软件功能，或者是平台(例如，云平台)上实例化的虚拟化功能，本申请对于上述AI节点#1和AI节点#2的具体形态不作限定。
还可以理解，在实际通信中，可以将AI节点进行更细的划分，或者说还可以包括更多数量的AI节点。例如，通信系统中包括至少两个AI节点(如AI节点#1、AI节点#2、AI节点#3)，各个AI节点负责不同的功能。示例地，该至少两个AI节点均可与终端设备建立AI连接，或者该至少两个AI节点中部分AI节点可与终端设备建立AI连接。
还可以理解，图2和图3中的各个设备之间的接口名称只是一个示例，具体实现中接口的名称可能为其他的名称，本申请对此不作具体限定。
还可以理解，图2和图3为示例性说明，本申请不限于此。例如，图2和图3所示的通信系统中还可以包括更多数量的设备，如更多数量的终端设备，又如更多数量的AI节点，又如更多数量的网络设备等等。再例如，终端设备之间也可以互相发布AI任务，如终端设备之间直接发布AI任务，又如终端设备通过AI节点(如AI节点#1，又如AI节点#2)向其他终端设备发布AI任务。
上文结合图2和图3简单介绍了根据本申请实施例提供的通信系统。下面介绍本申请实施例提供的通信方法。下文所述的方法可用于图2或图3所示的系统。
图4是本申请一实施例提供的通信方法400的示意图。方法400可以包括如下步骤。
410，通信装置接收AI连接的信息。
在本申请中，通信装置可以是终端设备，也可以是用于终端设备中的芯片、芯片系统或电路，不予限制。
AI连接，表示通信装置与AI节点之间的连接，通过该AI连接，通信装置与AI节点可传输与AI相关的信令、执行与AI相关的操作。
作为示例，通信装置与AI节点之间的AI连接，可通过通信装置与网络设备之间的连接、以及网络设备与AI节点之间的连接实现。一种可能的实现方式，通信装置与AI节点之间的AI连接，可通过通信装置与网络设备之间建立的无线承载(radio bearer，RB)(或者称为空口承载，或者称为空口无线承载)、以及网络设备与AI节点之间的连接实现。
举例来说，通信装置通过AI连接向AI节点发送消息，可以包括：通信装置向网络设备发送该消息，网络设备向AI节点转发该消息，也即网络设备可类似于通信装置与AI节点之间的中继设备。其中，通信装置向网络设备发送消息，可以基于通信装置与网络设备 之间建立的无线承载发送消息；网络设备向AI节点转发消息，可以基于网络设备与AI节点之间的连接(如AI-2接口)转发该消息。
其中，AI连接的信息，或者称AI连接的配置信息，表示与AI连接相关的信息。通过该AI连接的信息，通信装置和/或AI节点可处理与AI相关的操作。关于AI连接的信息，后面详细描述。
420，通信装置根据AI连接的信息与AI节点建立AI连接。
也即根据AI连接的信息，通信装置与AI节点之间可建立AI连接，进而可通过该AI连接通信。作为示例，通信装置收到AI连接的信息后可直接基于该AI连接的信息与AI节点建立AI连接，或者通信装置也可以在收到/发送指示或触发后、或者满足预设条件时，再基于该AI连接的信息与AI节点建立AI连接，本申请不予限制。
可选地，方法400还包括步骤430：通信装置与AI节点通过AI连接通信。
在本申请中，通信装置可接收AI连接的信息，并根据该AI连接的信息与AI节点建立AI连接，这样AI节点和通信装置之间可通过该AI连接通信，如发送和/或接收AI任务，又如发送和/或接收AI任务的处理结果等，从而实现AI与无线网络的融合。
可选地，作为一种可能的情况，在步骤410之前，方法400还包括：通信装置发送请求消息，该请求消息用于请求建立通信装置与AI节点之间的AI连接。进一步可选地，通信装置接收响应消息，该响应消息包括AI连接的信息。其中，该请求消息例如也可称为AI连接建立请求消息。可选地，请求消息中还包括请求类型(request type)的指示信息，该请求类型可用于指示建立AI连接的目的。如，该请求消息中直接携带请求类型，直接可以指示请求类型；又如，该请求消息中携带某指示信息，该指示信息可指示请求类型。
作为示例，请求类型包括以下至少一项：模型训练、数据采集、任务发布等。例如，若请求类型包括模型训练，则表示UE请求建立AI连接的目的是进行模型训练。再例如，若请求类型包括任务发布，则表示UE请求建立AI连接的目的是通过该AI连接发布AI任务。
这样，在通信装置与AI节点建立AI连接之前，可通过通信装置发起请求的方式，与AI节点建立AI连接。从而通信装置可以在需要通过AI连接执行相关操作时，如发布AI任务时，由通信装置主动发起建立AI连接的请求，避免建立了AI连接后，不使用该AI连接造成的资源浪费。
可选地，作为另一种可能的情况，至少在步骤420之前，方法400还包括：通信装置接收来自AI节点的指示信息，该指示信息用于触发通信装置与AI节点建立AI连接。进一步可选地，通信装置收到该指示信息后，可向AI节点发送请求消息，以请求建立通信装置与AI节点之间的AI连接。
这样，在通信装置与AI节点建立AI连接之前，可通过AI节点触发的方式，与通信装置建立AI连接。从而AI节点可以在需要通过AI连接执行相关操作时，如发布AI任务时，由AI节点主动触发通信装置发送AI建立请求，避免建立了AI连接后，不使用该AI连接造成的资源浪费。
下面详细介绍AI连接的信息。
可选地，AI连接的信息包括：AI连接的标识、AI连接的资源信息、或无线承载的信息。
1)AI连接的标识(identifier，ID)：可以是AI节点提供的。通过该AI连接的标识， 可识别该AI连接为通信装置与AI节点之间的AI连接。举例来说，AI节点收到来自通信装置的请求消息后，获知通信装置请求建立AI连接，因此，AI节点分配AI连接的标识。
2)AI连接的资源信息：可以是AI节点提供的。举例来说，AI节点收到来自通信装置的请求消息后，获知通信装置请求建立AI连接，因此，AI节点为AI连接配置资源。
作为示例，AI连接的资源信息包括为AI连接配置的以下至少一项：模型、数据集、或计算资源。
例如，AI连接的资源信息包括模型。也即，AI节点可为AI连接配置模型，该模型也可称为预置模型。对于AI节点来说，作为示例，AI节点可基于该模型对通信装置上传的数据进行处理。对于通信装置来说，作为示例，通信装置可处理该模型，如通信装置基于该模型进行模型训练、模型测试、模型融合等等。
再例如，AI连接资源信息包括数据集。也即，AI节点可为AI连接配置数据集，该数据集也可称为预置数据集。对于AI节点来说，作为示例，AI节点可管理该数据集。对于通信装置来说，作为示例，通信装置可基于该数据集进行AI测量。
一种可能的实现方式，若通信装置发送请求消息，且请求消息包括请求类型，则AI节点可根据该请求类型分配数据集，也即该数据集可以是特定UE(UE-specific)的数据集或可以是特定任务(mission-specific)的数据集。
再例如，AI连接资源信息包括计算资源。也即，AI节点可为AI连接配置计算资源，该计算资源也可称为预置计算资源。该计算资源，可以用于通信装置或AI节点执行AI相关的操作，如通信装置或AI节点使用该计算资源处理AI任务。
上述单独介绍了各个信息，可以理解，上述各个信息也可以结合使用。
AI节点在AI连接的有效期内，可以对AI连接的资源信息进行动态的维护、更新。例如，AI节点可更新AI连接的资源。此外，若AI连接的资源信息发生更新，则AI节点可以将更新后的AI连接的资源信息通过AI连接发送给通信装置，或者按需发送给通信装置。AI节点和通信装置可基于更新后的AI连接的资源信息重新建立AI连接。
可以理解，上述主要以AI节点为AI连接配置AI连接的标识和AI连接的资源信息，本申请不限于此。作为示例，也可以是其他装置配置AI连接的标识和AI连接的资源信息。在该示例下，例如其他装置将配置的AI连接的标识和AI连接的资源信息发送给AI节点，进而AI节点将配置的AI连接的标识和AI连接的资源信息发送给通信装置。再例如，其他装置将配置的AI连接的标识和AI连接的资源信息发送给AI节点，并且其他装置将配置的AI连接的标识和AI连接的资源信息发送给通信装置。
为区分，将AI节点提供的AI连接的信息(如AI连接的标识和/或AI连接的资源信息)称为AI连接的第一信息。
3)无线承载的信息，可以是网络设备提供的。如前所述，通信装置与AI节点之间的AI连接，可以通过通信装置与网络设备之间建立的无线承载、以及网络设备与AI节点之间的连接实现，因此，可以由网络设备配置无线承载的信息，并将该无线承载的信息与AI连接关联，这样可实现通信装置与AI节点通过AI连接通信。
举例来说，通信装置通过网络设备向AI节点发送请求消息，AI节点通过网络设备向通信装置发送AI连接的第一信息，网络设备收到AI连接的第一信息后，获知AI节点接受(或者称为同意)通信装置的请求，因此网络设备配置无线承载的信息，并将该无线承载的信息与AI连接的第一信息中的AI连接的标识(如果第一信息没有包括AI连接的标 识，那么根据通过其他方式获知AI连接的标识)关联。
一种可能的实现方式，网络设备根据AI连接的资源信息，确定无线承载的信息。例如，AI连接的资源信息包括为AI连接配置的模型，网络设备根据该模型对无线承载进行配置，如配置协议栈中包含的协议层，和/或，配置各协议层的参数等。举例来说，为AI连接配置的模型输入的是环境参数，如收发端各自剩余算力、收发端之间的信道参数，输出的是各协议层的配置参数，如物理层的帧结构、介质访问控制(medium access control，MAC)层的分段条件、调度算法涉及的参数、无线链路控制(radio link control，RLC)层的重传条件、分组数据汇聚层协议(packet data convergence protocol，PDCP)层的头压缩算法涉及的参数等等。上述为示例性说明，本申请不限于此。
关于网络设备配置无线承载的信息的方式，后面结合方法800详细介绍。
以图2或图3所示的架构为例，本申请实施例中的AI节点可以为AI节点#1，也可以为AI节点#2，不予限制。下面结合AI节点，介绍两种可能的情形。
第一种可能的情形，AI节点为AI节点#1。
在该情形下，AI连接表示通信装置与AI节点#1之间的连接。以上述图2或图3所示的系统为例，通信装置与AI节点#1之间的接口称为AI-1接口，步骤430中，通信装置与AI节点#1通过AI连接通信，可替换为，通信装置与AI节点#1通过AI-1接口通信。
在该情形下，上述为AI连接配置AI连接的第一信息的AI节点可以为AI节点#1。或者为AI连接配置AI连接的第一信息的装置也可以为其他装置，且其他装置将为AI连接配置的AI连接的第一信息发送给AI节点#1，AI节点#1再将该AI连接的第一信息发送给通信装置，或者其他装置将该AI连接的第一信息发送给AI节点#1和通信装置。或者也可以是其他装置配置AI连接的资源信息，AI节点#1配置AI连接的标识，且其他装置将为AI连接配置的AI连接的资源信息发送给AI节点#1，AI节点#1再将该AI连接的资源信息以及自身配置的AI连接的标识发送给通信装置。
可选地，通信装置与AI节点#1通过AI连接通信，包括：通信装置通过AI连接向AI节点#1发送AI消息，和/或，AI节点#1通过AI连接向通信装置发送AI消息。
图5示出了通信装置与AI节点#1通信的协议栈的示意图。
如图5所示，通信装置与AI节点#1之间可通过AI层(stratum)通信。对于通信装置来说，AI层与RRC层之间可传递消息，如AI层向RRC层传递通信装置发给AI节点的消息，又如RRC层向AI层传递AI节点发给通信装置的消息。
例如，通信装置向网络设备发送RRC消息，该RRC消息携带AI消息(或者也可称为AI stratum消息)；网络设备向AI节点发送AI消息。再例如，AI节点向网络设备发送AI消息，网络设备向通信装置发送RRC消息，该RRC消息携带AI消息。也即，网络设备可利用RRC消息作容器携带AI消息。
可以理解，图5为示例性说明，关于通信装置与AI节点#1之间通信的协议栈，可灵活配置，此处不予限制。此外，图5主要以通信装置为终端设备，终端设备与AI节点#1之间通过网络设备传输为例进行示例性说明，本申请不限于此。
可选地，AI消息指示以下至少一项信息：安全类型(security type)、压缩类型(compression type)、消息类型(message type)、控制指示(control indicator)或流量指示(traffic indicator)、消息认证码(message authentication code)、消息内容。作为示例，AI消息中可以直接包括上述至少一项信息，用于直接指示上述至少一项信息；或者AI消 息中可以包括上述至少一项信息的对应指示信息，可间接指示上述至少一项信息。
图6示出了AI消息的信元格式的示意图。下面简单介绍各信息。
1)安全类型：用于指示AI消息采用的加密方法。若AI消息中包括安全类型，则可基于该安全类型来解密AI消息。若AI消息中不包括安全类型，则一种可能的实现方式，可以基于默认的加密方法来解密AI消息，也即此情况下，AI消息采用的加密方法可以是预先约定的，或者预定义的。
2)压缩类型：用于指示压缩类型采用的压缩方法。若AI消息中包括压缩类型，则可基于该压缩类型来解压缩AI消息。若AI消息中不包括压缩类型，则一种可能的实现方式，可以基于默认的压缩方法来解压缩AI消息，也即此情况下，AI消息采用的压缩方法可以是预先约定的，或者预定义的。
3)消息类型：用于指示AI消息的类型。
一种可能的实现方式，通过比特的取值不同，区分AI消息的类型。例如，若消息类型为0000，则表示AI消息用于描述模型参数；若消息类型为0001，则表示AI消息用于模型更新；若消息类型为0010，则表示AI消息用于描述数据集；若消息类型为0011，则表示AI消息用于更新数据集；若消息类型为0100，则表示AI消息用于描述任务；若消息类型为0101，则表示AI消息用于AI节点切换或重选参数，等等。可以理解，上述例子为示例性说明，本申请实施例不限于此。
若AI消息中包括消息类型，则可基于该消息类型确定AI消息的类型。若AI消息中不包括消息类型，则一种可能的实现方式，可通过解析AI消息获知AI消息的类型；或者另一种可能的实现方式，不同格式的AI消息可对应AI消息的不同类型，也即基于AI消息的格式可确定该AI消息的类型。举例来说，AI消息的格式有多种，若AI消息的格式为第一格式，则表示AI消息用于描述模型参数述；若AI消息的格式为第二格式，则表示AI消息用于模型更新；若AI消息的格式为第三格式，则表示AI消息用于描述数据集；若AI消息的格式为第四格式，则表示AI消息用于更新数据集；若AI消息的格式为第五格式，则表示AI消息用于描述任务；若AI消息的格式为第六格式，则表示AI消息用于AI节点切换或重选参数。可以理解，上述例子为示例性说明，本申请实施例不限于此。
4)控制指示或流量指示：用于指示AI消息传输的是控制信令还是AI相关数据。例如，若AI消息中包括控制指示，则表示AI消息传输的是控制信令；再例如，若AI消息中包括流量指示，则表示AI消息传输的是控AI相关数据。
5)消息认证码：用于指示完整性校验的验证码。若AI消息中包括消息认证码，则可基于该消息认证码来进行完整性校验。若AI消息中不包括消息认证码，则一种可能的实现方式，可以基于默认的验证码来进行完整性校验，也即此情况下，对AI消息进行完整性校验的验证码可以是预先约定的，或者预定义的。
可以理解，上述关于AI消息包括的信息，仅是示例性说明，本申请不限于此。
通信装置与AI节点#1通过AI连接通信，或者通信装置与AI节点#1通过AI连接传输AI消息，可以包括以下示例。
示例1，通信装置通过AI连接向AI节点#1发布第一AI任务。相应地，AI节点#1接收该第一AI任务。在本申请中，为区分，将通信装置发布的任务称为第一AI任务。
第一种可能的实现方式，AI节点#1收到第一AI任务后，处理第一AI任务。
在该方式下，AI节点#1完成第一AI任务后，可向通信装置发送第一AI任务的处理 结果。
第二种可能的实现方式，AI节点#1收到第一AI任务后，将第一AI任务发布给其他通信装置(如其他终端设备，又如其他AI节点)，由其他通信装置处理第一AI任务。也即，通信装置可通过AI节点#1向其他通信装置发布第一AI任务。其中，其他通信装置可被称为协作完成AI任务的通信装置，为区分，将其他通信装置称为协作通信装置。协作通信装置的数量为至少一个。
例如，AI节点#1收到第一AI任务后，AI节点#1确定协作通信装置，并将第一AI任务通过AI节点#1与协作通信装置之间的AI连接，发布给协作通信装置，可以理解，AI节点#1向协作通信装置进行的第一AI任务的发布，包括：第一AI任务的完整任务发布和/或第一AI任务的部分任务发布。再例如，通信装置确定协作通信装置，并且通信装置通过AI节点#1向协作通信装置发布第一AI任务，AI节点#1收到该第一AI任务后，可通过AI节点#1与协作通信装置之间的AI连接，将第一AI任务发布给协作通信装置。再例如，通信装置确定协作通信装置，并且通信装置向AI节点#1发布第一AI任务的第一子任务，并通过AI节点#1向协作通信装置发布第一AI任务的第二子任务，AI节点#1收到该第一AI任务后，可通过AI节点#1与协作通信装置之间的AI连接，将第一AI任务的第二子任务发布给协作通信装置。
在该方式下，协作通信装置完成第一AI任务后，可向AI节点#1发送第一AI任务的处理结果，然后AI节点#1再向通信装置发送第一AI任务的处理结果。或者，协作通信装置完成第一AI任务后，可以直接向通信装置发送第一AI任务的处理结果。
需要说明的是，根据任务发布的情况，上述协作通信装置或AI节点#1完成第一AI任务包括：完成完整的第一AI任务，或者完成第一AI任务的子任务(或称部分/分解任务)。
第三种可能的实现方式，AI节点#1收到第一AI任务后向AI节点#2发送指示，由AI节点#2处理该第一AI任务。也即，通信装置可通过AI节点#1向AI节点#2发布第一AI任务。在该实现方式中，AI节点#2可认为是协作通信装置。可以理解，AI节点#1向AI节点#2发布的第一AI任务包括：第一AI任务的完整任务和/或第一AI任务的部分任务。
以图3为例，AI节点#1可通过AI-4接口向AI节点#2发送指示。在该方式下，AI节点#2完成第一AI任务后，可向AI节点#1发送第一AI任务的处理结果，然后AI节点#1再向通信装置发送第一AI任务的处理结果。或者，AI节点#2完成第一AI任务后，也可直接向通信装置发送第一AI任务的处理结果，以图3为例，AI节点#2可通过AI-6接口向通信装置发送第一AI任务的处理结果。
需要说明的是，根据任务发布的情况，上述AI节点#1或AI节点#2完成第一AI任务包括：完成完整的第一AI任务，或者完成第一AI任务的子任务(或称部分/分解任务)。
可以理解，在上述第二种可能的实现方式或第三种可能的实现方式中，若AI节点#1将第一AI任务发布给协作通信装置或者AI节点#2，则可以由协作通信装置或者AI节点#2来处理第一AI任务，或者也可以是协作通信装置和AI节点#1共同处理第一AI任务，或者AI节点#2和AI节点#1共同处理第一AI任务。以协作通信装置和AI节点#1共同处理第一AI任务为例，例如，协作通信装置和AI节点#1可以分别负责第一AI任务的一部分任务。
示例2，AI节点#1通过AI连接向通信装置发布第二AI任务。相应地，通信装置接收该第二AI任务。在本申请中，为区分，将AI节点(如AI节点#1，又如AI节点#2)发 布的任务称为第二AI任务。
一种可能的实现方式，通信装置收到第二AI任务后，处理第二AI任务。在该方式下，通信装置完成第二AI任务后，可向AI节点#1发送第二AI任务的处理结果。
另一种可能的实现方式，通信装置收到第二AI任务后，将第二AI任务发布给协作通信装置，由协作通信装置处理第二AI任务，也即，AI节点#1可通过通信装置向协作通信装置发布第二AI任务。可以理解，通信装置向协作通信装置进行的第二AI任务的发布，包括：第二AI任务的完整任务发布和/或第二AI任务的部分任务发布。
例如，通信装置收到第二AI任务后，通信装置确定协作通信装置，并将第二AI任务发布给协作通信装置。再例如，AI节点#1确定协作通信装置，并且AI节点#1通过AI节点#1与通信装置之间的AI连接，向协作通信装置发布第二AI任务，通信装置收到该第二AI任务后，可将第二AI任务发布给协作通信装置。
在该方式下，协作通信装置完成第二AI任务后，可通过协作通信装置与AI节点#1之间的AI连接向AI节点#1发送第二AI任务的处理结果。或者，协作通信装置完成第二AI任务后，可以将第二AI任务的处理结果发给通信装置，由通信装置通过通信装置与AI节点#1之间的AI连接，将第二AI任务的处理结果发给AI节点#1。需要说明的是，根据任务发布的情况，上述通信装置或协作通信装置完成第二AI任务包括：完成完整的第二AI任务，或者完成第二AI任务的子任务(或称部分/分解任务)。
需要说明的是，根据任务发布的情况，上述协作通信装置完成第二AI任务包括：完成完整的第二AI任务，或者完成第二AI任务的子任务(或称部分/分解任务)。
可以理解，AI节点#1可以向至少一个通信装置发布AI任务，例如，AI节点#1可通过与各个通信装置之间的AI连接，分别向各个通信装置发布AI任务。
由上述示例1和示例2可知，通信装置和AI节点#1均可以主动发布AI任务。
在反馈AI任务的处理结果时，可以处理完AI任务，直接反馈AI任务的处理结果，或者可选地，在满足预设条件的情况下，反馈AI任务的处理结果。以上述示例1为例，例如，AI节点#1在满足预设条件的情况下，向通信装置发送第一AI任务的处理结果。以上述示例2为例，例如，通信装置在满足预设条件的情况下，向AI节点#1发送第二AI任务的处理结果。其中，预设条件，也可称为触发条件或反馈条件，不予限制。
第一种可能的实现方式，预定义预设条件，如标准预定义预设条件。例如，AI任务与预设条件具有对应关系，当确定AI任务后，与该AI任务对应的预设条件也确定，也即可以根据该AI任务对应的预设条件，上报该AI任务的处理结果。
作为示例，AI任务与预设条件的对应关系，可以以类似于表1的形式存在。
表1

以表1为例，预设条件例如可以是周期性上报，也可以是位置触发上报，或者也可以是模型性能触发上报，等等，不予限制。例如，若AI任务为AI-M1任务，即模型失效测试，则可以在模型性能退化至阈值的情况下，再上报AI-M1任务的处理结果。再例如，若AI任务为AI-M2任务，即模型性能测试，则可以周期性的上报AI-M2任务的处理结果，进一步还可以预定定义周期(也即上报的时间间隔)。再例如，若AI任务为AI-M4任务，即数据采集，则可以在模型性能退化的情况下(如模型性能退化至阈值的情况下、或者发生退化的情况下)或者运动至特定区域的情况下(如位置的改变超过阈值)，再上报AI-M4任务的处理结果。其中，表1中涉及到的阈值，可以是预先定义的，如标准预定义的。
可以理解，表1为示例性说明，属于表1的变形都适用于本申请。例如，表1中还可能包括更多数量的AI任务。再例如，表1中“性能退化至阈值”也可以替换为“性能达到预设条件”。
第二种可能的实现方式，发布AI任务时，携带AI任务对应的预设条件的指示信息。如，发布AI任务时，直接携带AI任务对应的预设条件；又如，发布AI任务时携带某指示信息，该指示信息可指示该AI任务对应的预设条件。
以上述示例1为例，第一AI任务包括第一AI任务对应的预设条件。例如，通信装置通过AI连接向AI节点#1发送第一AI消息，该第一AI消息用于发布第一AI任务，且该第一AI消息包括第一AI任务对应的预设条件。假设AI节点#1向通信装置发送第一AI任务的处理结果，那么AI节点#1在满足第一AI任务对应的预设条件的情况下，向通信装置发送第一AI任务的处理结果。
以上述示例2为例，第二AI任务包括第二AI任务对应的预设条件。例如，AI节点#1通过AI连接向通信装置发送第二AI消息，该第二AI消息用于发布第二AI任务，且该第二AI消息包括第二AI任务对应的预设条件。假设通信装置向AI节点#1发送第二AI任务的处理结果，那么通信装置在满足第二AI任务对应的预设条件的情况下，向AI节点#1发送第二AI任务的处理结果。
关于AI任务对应的预设条件，例如可以参考表1的描述。如若AI任务为模型性能测试任务或模型训练测试任务，则可以周期性的上报AI任务的处理结果。再如若AI任务为模型失效测试，则可以在性能退化至阈值的情况下，上报AI任务的处理结果。再如若AI任务为数据采集，则可以在模型性能退化的情况下(如模型性能退化至阈值的情况下、或者发生退化的情况下)或者运动至特定区域的情况下(如位置的改变超过阈值)，上报AI任务的处理结果。
在反馈AI任务的处理结果时，不同的AI任务，反馈的信息可能不同。作为示例，表2示出了AI任务与反馈的信息的对应关系。
表2

以表2为例，例如，若AI任务为AI-M1任务，即模型失效测试，则发送的AI-M1任务的处理结果可以包括：模型失效的位置、模型的误差量、模型的时间戳。其中，该模型的时间戳例如可以是模型的失效时间。
再例如，若AI任务为AI-M3任务，即模型训练测试，则发送的AI-M3任务的处理结果可以包括：模型训练测试的位置、模型的训练收敛时间、模型的时间戳。其中，该模型的时间戳例如可以是模型的产生时间。
可以理解，表2为示例性说明，属于表2的变形都适用于本申请。例如，表2中还可能包括更多数量的AI任务。再例如，表2中还可以包括更多数量的反馈信息，如反馈的信息还可以包括模型的标识、模型的推理性能、AI节点的标识(如处理AI任务的AI节点的标识)。
第二种可能的情形，AI节点为AI节点#2。
在该情形下，AI连接表示通信装置与AI节点#2之间的连接。以上述图3所示的系统为例，通信装置与AI节点#2之间的接口称为AI-6接口，步骤430中，通信装置与AI节点#2通过AI连接通信，可替换为，通信装置与AI节点#2通过AI-6接口通信。
在该情形下，上述为AI连接配置AI连接的第一信息的AI节点可以为AI节点#1，且AI节点#1将为AI连接配置的AI连接的第一信息发送给AI节点#2。或者为AI连接配置AI连接的第一信息的装置也可以为其他装置，且其他装置将为AI连接配置的AI连接的第一信息发送给AI节点#2。或者也可以是其他装置配置AI连接的资源信息，AI节点#1配置AI连接的标识，且AI节点#1将为AI连接配置的AI连接的标识发送给AI节点#2，其他装置将为AI连接配置的AI连接的资源信息发送给AI节点#2。或者也可以是其他装置配置AI连接的资源信息，AI节点#2配置AI连接的标识，且其他装置将为AI连接配置的AI连接的资源信息发送给AI节点#2。或者也可以是AI节点#2自己配置AI连接的第一信息。
可选地，通信装置与AI节点#2通过AI连接通信，包括：通信装置通过AI连接向AI节点#2发送AI消息，和/或，AI节点#2通过AI连接向通信装置发送AI消息。关于AI消息，可参考第一种可能的情形中的描述，此处不再赘述。
图7示出了通信装置与AI节点#2通信的协议栈的示意图。
通信装置与AI节点#2之间可通过图7所示的协议栈通信。如图7所示，以通信装置为UE，网络设备为5G接入网(access network，AN)为例，UE与AI节点#2之间有对应的协议数据单元(protocol data unit，PDU)层，UE与AI节点#2之间可通过5G AN通信，5G AN例如可作为中继设备转发UE发送给AI节点#2的信息，和/或，转发AI节点#2发 送给UE的信息。
可以理解，图7为示例性说明，关于通信装置与AI节点#2之间通信的协议栈，可灵活配置，此处不予限制。此外，图7主要以通信装置与AI节点#2之间通过网络设备传输为例进行示例性说明，本申请不限于此。
通信装置与AI节点#2通过AI连接通信，或者通信装置与AI节点#2通过AI连接传输AI消息，可以包括以下示例。
示例1，通信装置通过AI连接向AI节点#2发布第一AI任务。相应地，AI节点#1接收该第一AI任务。
如上文所述，AI节点#1负责处理开销较小的AI任务，AI节点#2负责处理开销较大的AI任务。因此，通信装置可以向不同的AI节点发布不同的AI任务。例如，通信装置向AI节点#1发布处理开销较小的AI任务，通信装置向AI节点#2发布处理开销较大的AI任务。
第一种可能的实现方式，AI节点#2收到第一AI任务后，处理第一AI任务。
在该方式下，AI节点#2完成第一AI任务后，可向通信装置发送第一AI任务的处理结果。
第二种可能的实现方式，AI节点#2收到第一AI任务后，将第一AI任务发布给协作通信装置(包括发布第一AI任务的全部和/或部分任务)，由协作通信装置处理第一AI任务。也即，通信装置可通过AI节点#2向其他通信装置发布第一AI任务。具体可参考第一种可能的情形中的相关描述，此处不再赘述。
示例2，AI节点#2通过AI连接向通信装置发布第二AI任务。相应地，通信装置接收该第二AI任务。
一种可能的实现方式，通信装置收到第二AI任务后，处理第二AI任务。在该方式下，通信装置完成第二AI任务后，可向AI节点#2发送第二AI任务的处理结果。
另一种可能的实现方式，通信装置收到第二AI任务后，将第二AI任务发布给协作通信装置(包括发布第二AI任务的全部和/或部分任务)，由协作通信装置处理第二AI任务。也即，AI节点#2可通过通信装置向协作通信装置发布第二AI任务。具体可参考第一种可能的情形中的相关描述，此处不再赘述。
可以理解，AI节点#2可以向至少一个通信装置发布AI任务，例如，AI节点#2可通过与各个通信装置之间的AI连接，分别向各个通信装置发布AI任务。
由上述示例1和示例2可知，通信装置和AI节点#2均可以主动发布AI任务。
在反馈AI任务的处理结果时，可以处理完AI任务，直接反馈AI任务的处理结果，或者可选地，在满足预设条件的情况下，反馈AI任务的处理结果。以上述示例1为例，例如，AI节点#2在满足预设条件的情况下，向通信装置发送第一AI任务的处理结果。具体的，可以参考第一种可能的情形中的描述，此处不再赘述。
在反馈AI任务的处理结果时，不同的AI任务，反馈的信息可能不同。具体的，可以参考第一种可能的情形中的描述，此处不再赘述。
为了便于理解，下面用UE表示通信装置，结合图8至图12对本申请实施例进行示例性说明。其中涉及到的步骤以及术语具体可以可参考上文描述。在下文实施例中，为区分UE与AI节点#1之间的AI连接、以及UE与AI节点#2之间的AI连接，用AI-1接口和AI-6接口进行描述。
图8示出了根据本申请一实施例提供的通信方法800的示意性流程图。该方法800可以用于实现如方法400的方案，例如方法800可用于UE与AI节点#1建立AI-1接口的场景。作为示例，方法800可以包括如下步骤。
801，UE向网络设备发起随机接入(random access，RA)。
若UE未与网络设备建立通信连接，则UE可先接入网络设备，以便与网络设备通信。作为示例，UE可通过发起随机接入的方式接入网络设备。
802，UE向网络设备发送AI连接建立请求(AI connection setup request)信息。
一种可能的实现方式，在UE与网络设备建立无线资源控制(radio resource control，RRC)连接的过程中，UE向网络设备发送AI连接建立请求信息。作为示例，AI连接建立请求信息携带于RRC连接建立完成(RRC connection complete)消息。
其中，AI连接建立请求信息用于请求建立UE与AI节点#1之间的AI-1接口，也即AI连接建立请求信息用于请求建立UE与AI节点#1之间的AI连接。可选地，AI连接建立请求信息包括请求类型，该请求类型可用于指示建立AI-1接口的目的。关于请求类型，可参考方法400中的相关描述，此处不再赘述。
803，网络设备向AI节点#1发送AI连接建立请求信息。
网络设备收到UE发送的AI连接建立请求信息后，向AI节点#1转发该AI连接建立请求信息，也即该AI连接建立请求信息可通过网络设备透传给AI节点#1。
一种可能的实现方式，网络设备向AI节点#1发送初始UE消息(initial UE message)，该初始UE消息中携带AI连接建立请求信息。
804，AI节点#1根据UE的AI能力，确定AI-1接口的第一信息。
AI节点#1收到AI连接建立请求信息后，获知UE请求建立AI-1接口。AI节点#1可先确定UE的AI能力，进而根据该UE的AI能力确定AI-1接口的第一信息。UE的AI能力例如可以包括以下至少一项：UE的优先级、UE支持的最大算力、UE支持的AI任务类型、UE的硬件能力、UE的安全要求。
一种可能的实现方式，核心网存储UE的签约信息，AI节点#1向核心网查询UE的签约信息，AI节点#1根据UE的签约信息，可获知该UE的AI能力。其中，UE的签约信息包括UE的AI能力。以图2或图3所示的架构为例，AI节点#1可通过AI-3接口向核心网查询UE的签约信息。
AI-1接口的第一信息包括AI-1接口的资源信息(也即UE与AI节点#1之间的AI连接的资源信息)。可选地，还可以包括AI-1接口的标识(也即UE与AI节点#1之间的AI连接的标识)。关于AI-1接口的第一信息，可以参考方法400中AI连接的第一信息的相关描述，此处不再赘述。
AI节点#1可根据UE的AI能力，确定AI-1接口的第一信息。
例如，若UE的优先级高，则AI节点#1可为该UE分配较多的计算资源，或者则AI节点#1优先为该UE分配AI-1接口的资源信息。
再例如，AI节点#1可根据UE支持的AI任务类型，分配对应于该AI任务的模型。
再例如，若UE支持的最大算力大于或等于预设值，则AI节点#1可为该UE分配较多的数据集。
805，AI节点#1向网络设备发送AI-1接口的第一信息。
例如，AI节点#1向网络设备发送AI连接建立响应信息或者AI连接建立接受信息， 该AI连接建立响应信息或者AI连接建立接受信息中包括AI-1接口的第一信息。若AI节点#1向网络设备发送AI连接建立响应信息，则该AI连接建立响应信息用于通知接受AI连接建立请求。
再例如，AI节点#1向网络设备发送初始上下文建立请求(initial context setup request)信息，该初始上下文建立请求信息中包括AI-1接口的第一信息。该初始上下文建立请求信息中包括AI-1接口的第一信息，其可表示AI节点#1接受AI连接建立请求；或者，该初始上下文建立请求信息中包括AI-1接口的第一信息和AI连接建立接受信息。
806，网络设备建立无线承载。
AI-1接口可通过UE与网络设备之间建立的无线承载、以及网络设备与AI节点#1之间的连接实现。因此，若网络设备获知AI节点#1接受为UE建立AI-1接口，则网络设备可建立无线承载，并将该无线承载与AI-1接口关联(如将无线承载与AI连接的标识关联)。
其中，无线承载，可标识一套空口协议栈的配置参数。
图9示出了无线承载的示意图。图9中的(1)示出了下行信道映射的情况，图9中的(2)示出了上行信道映射的情况，下面以下行信道映射为例，结合几个示例介绍。
示例1，网络设备为UE与AI节点#1之间的AI连接配置逻辑信道AI-CH，也即网络设备配置的无线承载通过逻辑信道AI-CH传输。在本申请中，为区分，将UE与AI节点#1之间的AI连接对应的逻辑信道称为AI-CH。
在该示例下，逻辑信道的信息可映射至下行共享信道(downlink shared channel，DL-SCH)，传输信道可映射至物理下行共享信道(physical downlink shared channel，PDSCH)，也即由PDSCH传输。
示例2，网络设备为UE与AI节点#1之间的AI连接配置逻辑信道AI-CH和传输信道DL-AI-CH，也即网络设备配置的无线承载通过逻辑信道AI-CH和传输信道DL-AI-CH传输。在本申请中，为区分，将UE与AI节点#1之间的AI连接对应的下行传输信道称为DL-AI-CH。
在该示例下，逻辑信道的信息可映射至传输信道DL-AI-CH，传输信道DL-AI-CH可映射至PDSCH，也即由PDSCH传输。
一种可能的实现方式，网络设备通过物理下行控制信道(physical downlink control channel，PDCCH)指示传输资源的位置。由于传输信道DL-AI-CH和DL-SCH均映射至PDSCH，因此，可通过加扰PDCCH的无线网络临时标识(radio network temporary identify，RNTI)区分该PDCCH指示的传输资源的位置是DL-AI-CH的资源位置，还是DL-SCH的资源位置。
作为示例，若PDCCH通过X-RNTI加扰，则该PDCCH指示的传输资源的位置是DL-AI-CH的资源位置；若PDCCH通过其他RNTI(不同于X-RNTI)加扰，则该PDCCH指示的传输资源的位置是DL-SCH的资源位置。
示例3，网络设备为UE与AI节点#1之间的AI连接配置逻辑信道AI-CH、传输信道DL-AI-CH、和物理信道PD-AI-CH，也即网络设备配置的无线承载通过逻辑信道AI-CH、传输信道DL-AI-CH、和物理信道PD-AI-CH传输。在本申请中，为区分，将UE与AI节点#1之间的AI连接对应的下行物理信道称为PD-AI-CH。
在该示例下，逻辑信道的信息可映射至传输信道DL-AI-CH，传输信道DL-AI-CH可 映射至物理信道PD-AI-CH，也即由PD-AI-CH传输。
一种可能的实现方式，网络设备通过系统消息指示PD-AI-CH的资源位置。若有些物理信道PD-AI-CH上承载控制信息，则网络设备还可通过系统消息指示哪些PD-AI-CH上存在控制信息。其中，控制信息指示PD-AI-CH上承载的不同DL-AI-CH消息的位置。
可选地，网络设备通过下行控制信息(downlink control information，DCI)(如称为DCI_AI)指示PD-AI-CH资源的占用情况。作为示例，该DCI可通过RNTI(为区分，如称为Z-RNTI)加扰。
上述主要以下行信道映射为例进行了示例性说明，上行信道类似，此处为简洁，不再赘述。
807，网络设备向UE发送AI-1接口的信息。
其中，该AI-1接口的信息包括AI-1接口的第一信息。可选地，该AI-1接口的信息还包括无线承载的信息。可以理解，无线承载的信息也可以独立于AI-1接口的第一信息单独发送(如携带于不同信令中)，也可以和AI-1接口的第一信息合并发送(如携带于同一信令中)，不予限制。
例如，网络设备向UE发送AI连接建立响应信息或者AI连接建立接受信息，该AI连接建立响应信息或者AI连接建立接受信息中包括AI-1接口的信息。若网络设备向UE发送AI连接建立响应信息，则该AI连接建立响应信息用于通知接受AI连接建立请求。
再例如，网络设备向UE发送RRC重配置(RRC reconfiguration)信息，该RRC重配置信息中包括AI-1接口的信息。该RRC重配置信息中包括AI-1接口的信息，其可表示AI节点#1接受AI连接建立请求；或者，该RRC重配置信息中包括AI-1接口的信息和AI连接建立接受信息。
808，UE向网络设备发送RRC重配置完成(RRC reconfiguration complete)信息。
若步骤807中网络设备向UE发送RRC重配置信息，那么步骤808中UE向网络设备发送RRC重配置完成信息。
809，网络设备向AI节点#1发送初始上下文建立响应(initial context setup response)信息。
若步骤805中，AI节点#1向网络设备发送初始上下文建立请求信息，则步骤809中，网络设备向AI节点#1发送初始上下文建立响应信息。
810，建立AI-1接口。
作为一种可能的情况，UE向网络设备发送RRC重配置完成信息，网络设备向AI节点#1发送初始上下文建立响应信息，AI节点#1收到初始上下文建立响应信息后，默认打通AI-1接口，即完成UE与AI节点#1之间的AI-1接口。
可以理解，步骤808和/或步骤809为示例性说明，实际中也可以不执行步骤808和/或步骤809。例如，不需要执行步骤808和步骤809，也即网络设备向UE发送AI-1接口的信息，UE收到该AI-1接口的信息后，默认打通AI-1接口，即完成UE与AI节点#1之间的AI-1接口。再例如，不需要执行步骤808，也即网络设备向UE发送AI-1接口的信息后，网络设备向AI节点#1发送初始上下文建立响应信息，AI节点#1收到初始上下文建立响应信息后，默认打通AI-1接口，即完成UE与AI节点#1之间的AI-1接口。再例如，不需要执行步骤809，也即网络设备通过RRC重配置信息向UE发送AI-1接口的信息后，UE向网络设备发送RRC重配置完成信息，网络设备收到RRC重配置信息后，默 认打通AI-1接口，即完成UE与AI节点#1之间的AI-1接口。
经过上述步骤，UE与AI节点#1建立AI-1接口，也即UE与AI节点#1之间的AI-1接口打通，因此，UE与AI节点#1可通过AI-1接口进行通信。例如，UE通过AI-1接口向AI节点#1发送AI消息。再例如，AI节点#1通过AI-1接口向UE发送AI消息。关于UE与AI节点#1通过AI-1接口通信的相关内容，可参考方法400中的相关描述，此处不再赘述。
上文结合图8示例地介绍了UE与AI节点#1建立AI-1接口的场景，该方法800可通过如图2所示的架构实现。基于上述实施例，UE与AI节点#1建立AI-1接口，进而可通过建立的AI-1接口传输AI消息，如发布AI任务、接收AI任务等，从而可实现无线网络与AI的融合。
图10示出了根据本申请另一实施例提供的通信方法1000的示意性流程图。该方法1000可以用于实现如方法400的方案，例如方法1000可用于UE与AI节点#2建立AI-6接口的场景。作为示例，方法1000可以包括如下步骤。
1001，UE与AI节点#1建立AI-1接口。
关于UE与AI节点#1建立AI-1接口的方法可以参考方法800，此处不再赘述。
1002，UE向AI节点#1发送AI连接建立请求信息。
其中，AI连接建立请求信息用于请求建立UE与AI节点#2之间的AI-6接口，也即AI连接建立请求信息用于请求建立UE与AI节点#2之间的AI连接。可选地，AI连接建立请求信息包括请求类型，该请求类型可用于指示建立AI-6接口的目的。关于请求类型，可参考方法400中的相关描述，此处不再赘述。
1003，AI节点#1选择AI节点#2，并确定AI-6接口的第一信息。
AI-6接口的第一信息包括AI-6接口的资源信息(也即UE与AI节点#2之间的AI连接的资源信息)，可选地，还可以包括AI-6接口的标识(也即UE与AI节点#2之间的AI连接的标识)。关于AI-6接口的第一信息，可以参考方法400中AI连接的第一信息的相关描述，此处不再赘述。
AI节点#1收到AI连接建立请求信息后，可先确定UE的AI能力，进而根据该UE的AI能力确定AI-6接口的第一信息。关于UE的AI能力、AI节点#1获知UE的AI能力、以及AI节点#1根据该UE的AI能力确定AI-6接口的第一信息的方式，可参考步骤804中的描述，此处不再赘述。
AI节点#1选择AI节点#2，可通过以下任一方式实现。
一种可能的实现方式，AI节点#1可根据UE的AI连接建立请求信息，确定AI节点#2。例如，AI节点#1可根据UE的AI连接建立请求信息中的请求类型，获知UE建立AI-6接口的目的，AI节点#1基于该目的，选择合适的AI节点#2。
另一种可能的实现方式，AI节点#1根据UE的AI能力，确定AI节点#2。例如，AI节点#1可根据UE的AI能力，获知UE支持的AI任务，那么AI节点#1可选择能够支持该AI任务的AI节点#2。再例如，AI节点#1可根据UE的AI能力，获知UE的优先级比较高，那么AI节点#1可选择算力较高的AI节点#2，如算力超过阈值的AI节点#2，以尽可能地满足UE。
上述两种实现方式为示例性说明，本申请并未限定于此。例如，AI节点#1可以基于请求类型和UE的AI能力，选择AI节点#2。
1004，AI节点#1与AI节点#2进行AI-6接口的配置。
AI节点#1选择AI节点#2后，可完成AI-6接口在AI节点#2侧的配置。举例来说，AI节点#1向AI节点#2发送AI-6接口的第一信息。一种可能的实现方式，AI节点#1向AI节点#2发送AI连接配置(AI connection configuration)，该AI连接配置包括AI-6接口的第一信息。
1005，AI节点#1向网络设备发送AI-6接口的第一信息。
例如，AI节点#1向网络设备发送AI连接建立响应信息或者AI连接建立接受信息，该AI连接建立响应信息或者AI连接建立接受信息中包括AI-6接口的第一信息。若AI节点#1向网络设备发送AI连接建立响应信息，则该AI连接建立响应信息用于通知接受AI连接建立请求。
再例如，AI节点#1向网络设备发送无线承载建立请求(radio bearer setup request)信息，该无线承载建立请求信息中包括AI-6接口的第一信息。该无线承载建立请求信息中包括AI-6接口的第一信息，其表示AI节点#1接受AI连接建立请求；或者，该无线承载建立请求信息中包括AI-6接口的第一信息和AI连接建立接受信息。
1006，网络设备建立无线承载。
AI-6接口可通过UE与网络设备之间建立的无线承载、以及网络设备与AI节点#2之间的连接实现。因此，若网络设备获知AI节点#2接受为UE建立AI-6接口，则网络设备可建立无线承载，并将该无线承载与AI-6接口关联(如将无线承载与AI连接的标识关联)。
步骤1006与步骤806类似，此处不再赘述。
1007，网络设备向UE发送AI-6接口的信息。
其中，该AI-6接口的信息包括AI-6接口的第一信息。可选地，该AI-6接口的信息还包括无线承载的信息。可以理解，无线承载的信息也可以独立于AI-6接口的第一信息单独发送(如携带于不同信令中)，也可以和AI-6接口的第一信息合并发送(如携带于同一信令中)，不予限制。
例如，网络设备向UE发送AI连接建立响应信息或者AI连接建立接受信息，该AI连接建立响应信息或者AI连接建立接受信息中包括AI-6接口的信息。若网络设备向UE发送AI连接建立响应信息，则该AI连接建立响应信息用于通知接受AI连接建立请求。
再例如，网络设备向UE发送RRC重配置信息，该RRC重配置信息中包括AI-6接口的信息。该RRC重配置信息中包括AI-6接口的信息，其可表示AI节点#2接受AI连接建立请求；或者，该RRC重配置信息中包括AI-6接口的信息和AI连接建立接受信息。
1008，UE向网络设备发送RRC重配置完成信息。
若步骤1007中网络设备向UE发送RRC重配置信息，那么步骤1008中UE向网络设备发送RRC重配置完成信息。
1009，网络设备向AI节点#1发送无线承载建立响应(radio bearer setup response)信息。
若步骤1005中，AI节点#1向网络设备发送无线承载建立请求信息，则步骤1009中，网络设备向AI节点#1发送无线承载建立响应信息。
可选地，方法1000还包括步骤1010。
1010，AI节点#1与AI节点#2更新AI-6接口的配置。
AI节点#1与AI节点#2更新AI-6接口的配置，表示AI节点#1更新AI-6接口的配置，如更新AI-6接口的资源信息，并在AI节点#2侧进行更新。一种可能的实现方式，AI节点#1向AI节点#2发送AI连接配置或AI连接配置更新(AI connection configuration update)，该AI连接配置或AI连接配置更新包括更新后的AI-6接口的配置。
一种可能的情形，AI节点#1可以基于UE的请求，更新AI-6接口的配置。举例来说，UE通过AI-1接口向AI-1节点发送消息，该消息用于更新AI-6接口的配置；AI-1节点收到该消息后，基于该消息更新AI-6接口的配置。
另一种可能的情形，AI节点#1可以自主决定更新AI-6接口的配置。
1011，建立AI-6接口。
作为一种可能的情况，UE向网络设备发送RRC重配置完成信息，网络设备向AI节点#1发送无线承载建立响应信息，AI节点#1收到无线承载建立响应信息后，默认打通AI-6接口，即完成UE与AI节点#2之间的AI-6接口。
可以理解，步骤1008和/或步骤1009为示例性说明，实际中也可以不执行步骤1008和/或步骤1009。例如，不需要执行步骤1008和步骤1009，也即网络设备向UE发送AI-6接口的信息，UE收到该AI-6接口的信息后，默认打通AI-6接口，即完成UE与AI节点#2之间的AI-6接口。再例如，不需要执行步骤1008，也即网络设备向UE发送AI-6接口的信息后，网络设备向AI节点#1发送无线承载建立响应信息，AI节点#1向AI节点#2转发给无线承载建立响应信息，AI节点#2收到无线承载建立响应信息后，默认打通AI-6接口，即完成UE与AI节点#2之间的AI-6接口。再例如，不需要执行步骤1009，也即网络设备通过RRC重配置信息向UE发送AI-6接口的信息后，UE向网络设备发送RRC重配置完成信息，网络设备收到RRC重配置信息后，默认打通AI-6接口，即完成UE与AI节点#2之间的AI-6接口。
经过上述步骤，UE与AI节点#2建立AI连接，也即UE与AI节点#2之间的AI-6接口打通，因此，UE与AI节点#2可通过AI-6接口进行通信。例如，UE通过AI-6接口向AI节点#2发送AI消息。再例如，AI节点#2通过AI-6接口向UE发送AI消息。关于UE与AI节点#2通过AI-6接口通信的相关内容，可参考方法400中的相关描述，此处不再赘述。
方法1000主要以通过AI节点#1建立UE与AI节点#2之间的AI连接为例进行示例性说明，对此不予限制。例如，也可以通过网络设备建立UE与AI节点#2之间的AI连接。举例来说，UE可以向网络设备发送用于请求建立UE与AI节点#2之间的AI-6接口的AI连接建立请求信息，网络设备收到该AI连接建立请求信息后，选择AI节点#2，并且向AI节点#2发送该AI连接建立请求信息。再举例来说，UE选择AI节点#2，并且向网络设备发送用于请求建立UE与该AI节点#2之间的AI-6接口的AI连接建立请求信息，网络设备收到该AI连接建立请求信息后，向该AI节点#2发送该AI连接建立请求信息。具体的实现，可以方法800中的UE与AI节点#1之间的建立，此处不再赘述。
上文结合图10示例地介绍了UE与AI节点#2建立AI-6接口的场景，该方法1000可通过如图3所示的架构实现。基于上述实施例，UE与AI节点#2建立AI-6接口，进而可通过建立的AI-6接口传输AI消息，如发布AI任务、接收AI任务等，从而可实现无线网络与AI的融合。
图11示出了根据本申请另一实施例提供的通信方法1100的示意性流程图。该方法 1100可以用于实现如方法400的方案，例如方法1100可用于UE向AI节点#1发布任务的场景。作为示例，方法1100可以包括如下步骤。
1101，UE1与AI节点#1建立AI-1接口。
一种可能的实现方式，UE1与AI节点#1建立AI-1接口，并且在此过程中，向AI节点#1发布第一AI任务。作为示例，UE1发送AI连接建立请求信息时，在该AI连接建立请求信息中携带请求类型，且该请求类型用于指示建立AI连接的目的是发布任务。
另一种可能的实现方式，UE1与AI节点#1建立AI-1接口，并且通过建立的AI-1接口，向AI节点#1发布第一AI任务。
关于建立AI-1接口的方式，可以参考方法800中的描述，此处不再赘述。
1102，AI节点#1根据在网UE的AI能力，确定协作UE。
其中，协作UE的数量为至少一个。
AI节点#1可根据在网UE的AI能力，确定协作UE，以实现由协作UE完成第一AI任务。例如，UE的AI能力包括UE能够支持的第一AI任务(或者能够承担的第一AI任务，或者能够执行的第一AI任务)，因此AI节点#1可查询在网UE的AI能力，以确定哪些UE可支持第一AI任务。
举例来说，假设步骤1101中UE1发布的第一AI任务为模型训练测试任务，那么AI节点#1可查询在网UE的AI能力，确定能够支持模型训练测试任务的协作UE，也即AI节点#1确定的协作UE可支持模型训练测试任务。
可选地，在步骤1102之前，方法1100还包括：AI节点#1查询UE1的上下文，确认UE1的优先级信息。这样，若UE1的优先级较高，则AI节点#1可优先为UE1发布的第一AI任务确定协作UE，从而可以使得更多的UE参与协作处理第一AI任务。
一种可能的实现方式，AI节点#1与协作UE共同处理第一AI任务。例如，AI节点#1负责处理第一AI任务的部分任务，协作UE负责处理第一AI任务的其余部分任务。
另一种可能的实现方式，协作UE处理第一AI任务。
若协作UE的数量为至少两个，则作为示例，AI节点#1还可以确定各个协作UE需要处理的任务，如AI节点#1将第一AI任务分为至少两个子任务，各个协作UE可分别处理第一AI任务的子任务，最终共同完成第一AI任务。或者AI节点#1也可直接向协作UE发布第一AI任务。
1103，AI节点#1向协作UE发送AI寻呼(AI paging)。
AI节点#1确定协作UE后，可向协作UE发送AI寻呼，以触发协作UE发起AI连接建立请求信息，建立协作UE与AI节点#1之间的AI-1接口。可以理解，AI寻呼仅是一种可能的命名方式，其命名不对本申请的保护范围造成限定。
1104，AI节点#1与协作UE建立AI-1接口。
关于建立AI-1接口的方式，可以参考方法800中的描述，此处不再赘述。
可以理解，若AI节点#1与协作UE之间已经建立了AI-1接口，则也可以不执行步骤1103-1104，即AI节点#1确定协作UE后，直接向协作UE分布第一AI任务。
1105，AI节点#1向协作UE发布第一AI任务。
AI节点#1与协作UE建立AI-1接口后，可通过对应的AI-1接口，向协作UE发布AI任务。例如，AI节点#1向协作UE发布第一AI任务，该第一AI任务与步骤1101中UE1发布的第一AI任务相同。可以理解，图11仅示出了协作UE处理第一AI任务的情况， 本申请并不限于此。例如，如前所述，若协作UE的数量为至少两个，AI节点#1也可以将第一AI任务分为至少两个子任务，各个协作UE可分别处理第一AI任务的子任务，最终共同完成第一AI任务。再例如，AI节点#1处理第一AI任务的第一子任务，并且AI节点#1向协作UE发布第一AI任务的第二子任务，由协作UE处理第一AI任务的第二子任务。
1106，协作UE处理第一AI任务。
1107，协作UE向AI节点#1发送第一AI任务的处理结果。
协作UE完成第一AI任务后，向AI节点#1发送第一AI任务的处理结果。
进一步可选地，若第一AI任务具有对应的预设条件，则可在满足预设条件的情况下，向AI节点#1发送第一AI任务的处理结果。具体的，可参考方法400中的相关描述，此处不再赘述。
1108，AI节点#1向UE1发送第一AI任务的处理结果。
AI节点#1收到协作UE发送的第一AI任务的处理结果后，可以直接向UE1转发该第一AI任务的处理结果；或者，若AI节点#1收到至少两个第一AI任务的处理结果，则AI节点#1也可以选择一个发送给UE1，或者也可以对收到的至少两个第一AI任务的处理结果进一步处理后再发给UE1，不予限制。
可选地，AI节点#1在协作UE的上下文中更新协作UE的优先级信息。在本申请中，可以根据UE的历史响应情况，动态更新UE的优先级信息。
可以理解，方法1100主要以协作UE处理第一AI任务为例进行示例性说明，可以理解，在实际通信中，AI节点#1收到UE1发布的第一AI任务后，也可以自己处理第一AI任务，并且AI节点#1处理第一AI任务后，AI节点#1可将第一AI任务的处理结果发送给UE1。或者，AI节点#1收到UE1发布的第一AI任务后，可以指示AI节点#2处理该第一AI任务(包括完整任务或者子任务)。AI节点#2处理第一AI任务后，AI节点#2可将第一AI任务的处理结果发送给UE1，或者AI节点#2可将第一AI任务的处理结果发送给AI节点#1，由AI节点#1将第一AI任务的处理结果发送给UE1。或者，AI节点#1可以自己处理第一AI任务的一部分任务，把剩余任务指示给AI节点#2和/或协作UE处理。可以理解，本申请不限制AI节点的数量，还可以有AI节点#3、AI节点#4来协作完成第一AI任务。
还可以理解，方法1100主要以协作UE向AI节点#1发送第一AI任务的处理结果，然后AI节点#1向UE1发送第一AI任务的处理结果为例进行示例性说明，本申请并不限于此。例如，协作UE也可以直接向UE1发送第一AI任务的处理结果。
上文结合图11示例地介绍了UE向AI节点#1发布AI任务的场景。基于上述实施例，UE与AI节点#1建立AI-1接口，进而UE可通过建立的AI-1接口向AI节点#1发布AI任务。此外，若AI节点#1不能处理AI任务或者AI节点#1需要其他AI节点或者协作UE的协助，则也可以将该AI任务发布给其他AI节点或其他UE，进而由其他AI节点或其他UE处理该AI任务，这样可实现利用空闲的计算资源(如协作UE自身的算力)完成AI任务，减轻AI网络的负载和能耗。
图12示出了根据本申请另一实施例提供的通信方法1200的示意性流程图。该方法1200可以用于实现如方法400的方案，例如方法1200可用于AI节点#1向UE发布任务的场景。作为示例，方法1200可以包括如下步骤。
1201，核心网中存储至少一个UE的签约信息。
其中，签约信息包括UE的AI能力。UE的AI能力例如可以包括以下至少一项：UE的优先级、UE支持的最大算力、UE支持的第二AI任务类型、UE的硬件能力、UE的安全要求。
1202，AI节点#1向核心网查询在网UE的签约信息，确定协作UE。
其中，协作UE的数量为至少一个。
在步骤1202中，AI节点#1可以向核心网查询在网UE的签约信息，以获取在网UE的AI能力。AI节点#1根据在网UE的AI能力，确定要处理第二AI任务的至少一个UE(为区分，将处理第二AI任务的至少一个UE称为协作UE)。例如，假设AI节点#1要发布的第二AI任务为模型测试任务，那么AI节点#1可根据在网UE的AI能力，确定能够支持模型测试任务的UE，也即AI节点#1确定的协作UE可支持模型测试任务。
1203，AI节点#1向协作UE发布第二AI任务。
例如，AI节点#1通过AI-1接口向协作UE发布第二AI任务。可以理解，若AI节点#1与协作UE之间还未建立AI-1接口，则在步骤1203之前，AI节点#1与协作UE可以先建立AI-1接口，然后AI节点#1再通过AI-1接口向协作UE发布第二AI任务。
作为示例，若协作UE的数量为至少两个，则AI节点#1也可以将第二AI任务分为至少两个子任务，各个协作UE可分别处理第二AI任务的子任务，最终共同完成第二AI任务。再例如，AI节点#1处理第一AI任务的第一子任务，并且AI节点#1向协作UE发布第一AI任务的第二子任务，由协作UE处理第一AI任务的第二子任务。
下面结合表3，给出一具体示例。
表3
假设AI节点#1发布的第二AI任务为模型测试任务(如模型失效测试任务，又如模型性能测试任务，又如模型训练测试任务)，那么AI节点#1配置的任务可以包括：需要测试的模型的参数、预设条件、测量上报的参数。
1)需要测试的模型的参数，用于描述一个待测试的模型。
2)预设条件，表示测量上报的触发条件。例如，若第二AI任务为模型失效测试任务，则预设条件为模型性能退化至阈值，也即在模型性能退化至阈值的情况下，再上报模型失效测试任务的处理结果。再例如，若第二AI任务为模型性能测试任务，则预设条件为周期性，也即周期性上报(或者说间隔上报)模型性能测试任务的处理结果。再例如，若第二AI任务为模型训练测试任务，则预设条件为周期性，也即周期性上报(或者说间隔上报)模型性能测试任务的处理结果。
3)测量上报的参数，表示反馈的任务的处理结果。例如，若第二AI任务为模型失效测试任务，则测量上报的参数可以包括模型失效的位置、模型的误差量、模型的时间戳。再例如，若第二AI任务为模型性能测试任务，则测量上报的参数可以包括模型性能测试的位置、模型的误差量、模型的时间戳。再例如，若第二AI任务为模型训练测试任务，则测量上报的参数可以包括模型失效的位置、模型的误差量、模型的训练收敛时间。
1204，协作UE处理第二AI任务。
1205，协作UE向AI节点#1发送第二AI任务的处理结果。
假设AI节点#1发布的第二AI任务为模型测试任务(如表3中的模型失效测试任务，又如表3中的模型性能测试任务，又如表3中的模型训练测试任务)，那么协作UE处理该模型测试任务。若该模型测试任务对应有相应的预设条件，则协作UE执行模型测试任务后，可先将测试结果缓存在本地，并在满足预设条件的情况下，再将该模型测试任务的处理结果通过AI-1接口上报至AI节点#1。
上文结合图12示例地介绍了AI节点#1向UE发布AI任务的场景。基于上述实施例，UE与AI节点#1建立AI-1接口，进而AI节点#1可通过建立的AI-1接口向UE发布AI任务从而可实现利用空闲的计算资源(如UE自身的算力)完成AI任务，减轻AI网络的负载和能耗。
可以理解，本申请实施例中的图8至图12中的例子仅仅是为了便于本领域技术人员理解本申请实施例，并非要将本申请实施例限于例示的具体场景。本领域技术人员根据图8至图12的例子，显然可以进行各种等价的修改或变化，这样的修改或变化也落入本申请实施例的范围内。例如，图11或图12中的AI节点#1也可替换为AI节点#2，相应地，AI-1接口替换为AI-6接口。
还可以理解，在本申请的各实施例中，“AI连接”与“AI接口”可以相互替换。例如，“终端设备与AI节点#1通过AI连接通信”也可以替换为“终端设备与AI节点#1通过AI-1接口通信”。再例如，“终端设备与AI节点#2通过AI连接通信”也可以替换为“终端设备与AI节点#2通过AI-6接口通信”。
还可以理解，在本申请的各实施例中，主要以AI节点#1管理AI-1接口和AI-6接口为例进行示例性说明，本申请不限于此。作为一示例，AI节点#1管理AI-1接口，AI节点#2管理AI-6接口。作为另一示例，其他装置管理AI-1接口和AI-6接口。例如，其他装置配置、更新、释放用于AI-1接口的资源，并通知给AI节点#1。再例如，其他装置配置、更新、释放用于AI-6接口的资源，并通知给AI节点#2；或者先通知给AI节点#1，再由AI节点#1通知给AI节点#2。
还可以理解，在本申请的一些实施例中，多次提及处理AI任务，可以理解，处理AI任务，也可以替换为执行AI任务。
还可以理解，在本申请的各实施例中，主要以AI节点与UE之间发布AI任务为例进 行了示例性说明，本申请并未限定于此。例如，AI节点之间也可以发布AI任务。再例如，UE之间也可以发布AI任务。
还可以理解，在本申请的各实施例中，以A向B发布AI任务为例，A向B发布的AI任务可以是该AI任务的完整任务，也可以是该AI任务的部分任务。相应地，根据AI任务发布的情况，B完成AI任务(或者称B处理AI任务)包括：完成完整的AI任务，或者完成AI任务的部分任务。
还可以理解，在本申请的各实施例中，AI连接的建立可以由UE发起，也可以由AI节点发起。
还可以理解，在本申请的各实施例中，AI节点可以是专用的功能节点，也可以是具备相应功能的网络设备(如，基站)、或者终端设备。
还可以理解，在本申请的各实施例中涉及到一些消息或信息名称，其命名不对本申请实施例的保护范围造成限定。以A向B发送消息为例，只要可以用于A和B之间的消息都适用于本申请实施例。
还可以理解，在上述一些实施例中，多次提及发送消息。以A向B发送消息为例，A向B发送消息，可以包括A直接向B发送消息，也可以包括A通过其他装置向B发送消息，对此不予限制。
还可以理解，本申请的各实施例中的一些可选地特征，在某些场景下，可以不依赖于其他特征，也可以在某些场景下，与其他特征进行结合，不作限定。
还可以理解，本申请的各实施例中的方案可以进行合理的组合使用，并且实施例中出现的各个术语的解释或说明可以在各个实施例中互相参考或解释，对此不作限定。
还可以理解，上述各个方法实施例中，由设备(如终端设备，又如AI节点，又如网络设备)实现的方法和操作，也可以由可由设备的组成部件(例如芯片或者电路)来实现。
相应于上述各方法实施例给出的方法，本申请实施例还提供了相应的装置，所述装置包括用于执行上述各个方法实施例相应的模块。该模块可以是软件，也可以是硬件，或者是软件和硬件结合。可以理解的是，上述各方法实施例所描述的技术特征同样适用于以下装置实施例。
图13是本申请实施例提供的一种通信装置1300的示意性框图。该装置1300包括收发单元1310和处理单元1320。收发单元1310可以用于实现相应的通信功能。收发单元1310还可以称为通信接口或通信单元。处理单元1320可以用于进行数据处理。
作为一种设计，该装置1300用于执行图4所示实施例中通信装置执行的步骤或者流程，图8和图10所示实施例中UE执行的步骤或者流程，图11所示实施例中UE1执行的步骤或者流程，图12所示实施例中协作UE执行的步骤或者流程。
一种可能的实现方式，收发单元1310，用于接收人工智能AI连接的信息，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；处理单元1320，用于根据AI连接的信息与AI节点建立AI连接。
一示例，收发单元1310，还用于通过AI连接向AI节点发送AI消息，和/或，通过AI连接接收来自AI节点的AI消息；其中，AI消息指示以下至少一项信息：AI消息采用的加密方式、AI消息采用的压缩方式、AI消息的消息类型、AI消息承载的是控制信息或数据、对AI消息进行完整性校验的验证码、AI消息承载的内容。
又一示例，收发单元1310，还用于通过AI连接接收更新后的AI连接的信息。
又一示例，处理单元1320，用于执行以下至少一项：AI连接的信息包括模型，处理模型；AI连接的信息包括数据集，根据数据集进行测量；AI连接的信息包括计算资源，使用计算资源执行AI任务。
又一示例，收发单元1310，还用于通过AI连接向AI节点发布第一AI任务；和/或，通过AI连接接收来自AI节点的第二AI任务。
又一示例，收发单元1310，具体用于通过AI连接向AI节点发布第一AI任务以通过AI节点向其他通信装置发布第一AI任务。
又一示例，收发单元1310，还用于在满足预设条件的情况下，发送第二AI任务的处理结果。
又一示例，第二AI任务中包括预设条件的指示信息。
又一示例，收发单元1310，还用于向AI节点发送请求消息，请求消息用于请求建立AI连接；或者，接收来自AI节点的指示信息，指示信息用于触发通信装置与AI节点建立AI连接。
又一示例，请求消息还包括请求类型的指示信息，请求类型用于表示请求建立AI连接的目的，数据集是根据请求类型确定的。
又一示例，AI连接的信息还包括以下至少一项：AI连接的标识、与AI连接关联的无线承载的信息。
又一示例，AI节点部署于以下任一项：网络设备、终端设备。
作为另一种设计，该装置1300用于执行图4所示实施例中AI节点执行的步骤或者流程，图8所示实施例中AI节点#1执行的步骤或者流程，图10所示实施例中AI节点#1和/或AI节点#2执行的步骤或者流程，图11和图12所示实施例中AI节点#1执行的步骤或者流程。
一种可能的实现方式，收发单元1310，用于发送AI连接的信息，AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；处理单元1320，用于根据AI连接的信息与通信装置建立AI连接。
一示例，收发单元1310，还用于通过AI连接向通信装置发送AI消息，和/或，通过AI连接接收来自通信装置的AI消息；其中，AI消息指示以下至少一项信息：AI消息采用的加密方式、AI消息采用的压缩方式、AI消息的消息类型、AI消息承载的是控制信息或数据、对AI消息进行完整性校验的验证码、AI消息承载的内容。
又一示例，收发单元1310，还用于通过AI连接向通信装置发送更新后的AI连接的信息。
又一示例，处理单元1320，用于执行以下至少一项：AI连接的信息包括模型，根据模型对通信装置发送的数据进行处理；AI连接的信息包括数据集，管理数据集；AI连接的信息包括计算资源，使用计算资源执行AI任务。
又一示例，处理单元1320，还用于根据通信装置的AI能力确定AI连接的信息。
又一示例，收发单元1310，还用于通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，接收来自至少一个通信装置发布的第一AI任务；和/或，通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，向至少一个通信装置发布第二AI任务；其中，至少一个通信装置包括通信装置。
又一示例，收发单元1310，具体用于通过AI节点与通信装置之间的AI连接向通信 装置发布第二AI任务以通过通信装置向其他通信装置发布第二AI任务。
又一示例，收发单元1310，还用于在满足预设条件的情况下，通过AI节点与至少一个通信装置中各个通信装置之间的AI连接，向至少一个通信装置发送第一AI任务的处理结果。
又一示例，第一AI任务包括预设条件的指示信息。
又一示例，收发单元1310，还用于接收来自通信装置的请求消息，请求消息用于请求建立AI连接；或者，向通信装置发送指示信息，指示信息用于触发通信装置与AI节点建立AI连接。
又一示例，请求消息还包括请求类型的指示信息，请求类型用于表示请求建立AI连接的目的，数据集是根据请求类型确定的。
又一示例，AI连接的信息还包括以下至少一项：AI连接的标识、与AI连接关联的无线承载的信息。
又一示例，装置1300包括第一AI节点和第二AI节点，第一AI节点和/或第二AI节点满足以下任一项：第一AI节点用于传输信令，第二AI节点用于处理AI任务；第二AI节点用于处理第一AI节点指示的AI操作；第一AI节点用于处理第一类AI任务，第二AI节点用于处理第二类AI任务，第一类AI任务与第二类AI任务不完全相同；第二AI节点用于保存和/或发送AI任务的处理结果。
又一示例，收发单元1310，还用于向第二AI节点发送通知信息，通知信息用于通知第二AI节点以下至少一项：执行AI任务、保存AI任务的处理结果、发送AI任务的处理结果。
又一示例，AI节点部署于以下任一项：网络设备、终端设备。
又一示例，AI节点部署于网络设备，处理单元1320，还用于根据以下至少一项确定与AI连接关联的无线承载的信息：模型、数据集、计算资源。
应理解，各单元执行上述相应步骤的具体过程在上述各方法实施例中已经详细说明，为了简洁，在此不再赘述。
还应理解，这里的装置1300以功能单元的形式体现。这里的术语“单元”可以指应用特有集成电路(application specific integrated circuit，ASIC)、电子电路、用于执行一个或多个软件或固件程序的处理器(例如共享处理器、专有处理器或组处理器等)和存储器、合并逻辑电路和/或其它支持所描述的功能的合适组件。
示例地，本申请实施例提供的装置1300的产品实现形态是可以在计算机上运行的程序代码。
示例地，本申请实施例提供的装置1300可以是通信设备，也可以是应用于通信设备上的芯片、芯片系统(例如：片上系统(system on chip，SoC))或电路。当该装置1300为通信设备时，收发单元1310可以是收发器，或，输入/输出接口；处理单元1320可以是处理器。当该装置1300为用于通信设备中的芯片、芯片系统或电路时，收发单元1310可以是该芯片、芯片系统或电路上的输入/输出接口、接口电路、输出电路、输入电路、管脚或相关电路等；处理单元1320可以是处理器、处理电路或逻辑电路等。
此外，上述收发单元1310还可以是收发电路(例如可以包括接收电路和发送电路)，处理单元可以是处理电路。
图14是本申请实施例提供的一种通信装置1400的示意性框图。该装置1400包括处 理器1410，处理器1410与存储器1420耦合。可选地，还包括存储器1420，用于存储计算机程序或指令和/或数据，处理器1410用于执行存储器1420存储的计算机程序或指令，或读取存储器1420存储的数据，以执行上文各方法实施例中的方法。
可选地，处理器1410为一个或多个。
可选地，存储器1420为一个或多个。
可选地，该存储器1420与该处理器1410集成在一起，或者分离设置。
可选地，如图14所示，该装置1400还包括收发器1430，收发器1430用于信号的接收和/或发送。例如，处理器1410用于控制收发器1430进行信号的接收和/或发送。
作为一种方案，该装置1400用于实现上文各个方法实施例中由通信装置执行的操作。
例如，处理器1410用于执行存储器1420存储的计算机程序或指令，以实现上文各个方法实施例中通信装置的相关操作。例如，图4所示实施例中通信装置执行的方法，或图8和图10所示实施例中UE执行的方法，或图11所示实施例中UE1执行的方法，或图12所示实施例中协作UE执行的方法。
作为另一种方案，该装置1400用于实现上文各个方法实施例中由AI节点执行的操作。
例如，处理器1410用于执行存储器1420存储的计算机程序或指令，以实现上文各个方法实施例中AI节点的相关操作。例如，图4所示实施例中AI节点执行的方法，或图8所示实施例中AI节点#1执行的方法，或图10所示实施例中AI节点#1和/或AI节点#2执行的方法，或图11、图12所示实施例中AI节点#1执行的方法。
在实现过程中，上述方法的各步骤可以通过处理器1410中的硬件的集成逻辑电路或者软件形式的指令完成。结合本申请实施例所公开的方法可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器1420，处理器1410读取存储器1420中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
应理解，本申请实施例中，处理器可以为一个或多个集成电路，用于执行相关程序，以执行本申请方法实施例。
处理器(例如，处理器1410)可包括一个或多个处理器并实现为计算设备的组合。处理器可分别包括以下一种或多种：微处理器、微控制器、数字信号处理器(digital signal processor，DSP)、数字信号处理设备(digital signal processing device，DSPD)、专用集成电路(application specific integrated circuit，ASIC)、现场可编程门阵列(field programmable gate array，FPGA)、可编程逻辑器件(programmable logic device，PLD)、选通逻辑、晶体管逻辑、分立硬件电路、处理电路或其它合适的硬件、固件和/或硬件和软件的组合，用于执行本公开中所描述的各种功能。处理器可以是通用处理器或专用处理器。例如，处理器1410可以是基带处理器或中央处理器。基带处理器可用于处理通信协议和通信数据。中央处理器可用于使装置执行软件程序，并处理软件程序中的数据。此外，处理器的一部分还可以包括非易失性随机存取存储器。例如，处理器还可以存储设备类型的信息。
本申请中的程序在广义上用于表示软件。软件的非限制性示例包括：程序代码、程序、子程序、指令、指令集、代码、代码段、软件模块、应用程序、或软件应用程序 等。程序可以在处理器和/或计算机中运行。以使得装置执行本申请中描述的各种功能和/或过程。
存储器(例如，存储器1420)可存储供处理器(例如，处理器1410)在执行软件时所需的数据。存储器可以使用任何合适的存储技术实现。例如，存储器可以是处理器和/或计算机能够访问的任何可用存储介质。存储介质的非限制性示例包括：随机存取存储器(random access memory，RAM)、只读存储器(read-only memory，ROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)、光盘只读存储器(Compact Disc-ROM,CD-ROM)、静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)、可移动介质、光盘存储器、磁盘存储介质、磁存储设备、闪存、寄存器、状态存储器、远程挂载存储器、本地或远程存储器组件，或能够携带或存储软件、数据或信息并可由处理器/计算机访问的任何其它介质。需要说明的是，本文描述的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
存储器(例如，存储器1420)和处理器(例如，处理器1410)可以分开设置或集成在一起。存储器可以用于与处理器连接，使得处理器能够从存储器中读取信息，在存储器中存储和/或写入信息。存储器可以集成在处理器中。存储器和处理器可以设置在集成电路中(例如，该集成电路可以设置在UE或BS或其他网络节点中)。
图15是本申请实施例提供的一种芯片系统1500的示意性框图。该芯片系统1500(或者也可以称为处理系统)包括逻辑电路1510以及输入/输出接口(input/output interface)1520。
其中，逻辑电路1510可以为芯片系统1500中的处理电路。逻辑电路1510可以耦合连接存储单元，调用存储单元中的指令，使得芯片系统1500可以实现本申请各实施例的方法和功能。输入/输出接口1520，可以为芯片系统1500中的输入输出电路，将芯片系统1500处理好的信息输出，或将待处理的数据或信令信息输入芯片系统1500进行处理。
作为一种方案，该芯片系统1500用于实现上文各个方法实施例中由通信装置执行的操作。
例如，逻辑电路1510用于实现上文方法实施例中由通信装置执行的处理相关的操作，如，图4所示实施例中通信装置执行的处理相关的操作，或图8和图10所示实施例中UE执行的处理相关的操作，或图11所示实施例中UE1执行的处理相关的操作，或图12所示实施例中协作UE执行的处理相关的操作；输入/输出接口1520用于实现上文方法实施例中由通信装置执行的发送和/或接收相关的操作，如，图4所示实施例中的通信装置执行的发送和/或接收相关的操作，或图8和图10所示实施例中UE执行的发送和/或接收相关的操作，或图11所示实施例中UE1执行的发送和/或接收相关的操作，或图12所示实施例中协作UE执行的发送和/或接收相关的操作。
作为另一种方案，该芯片系统1500用于实现上文各个方法实施例中由AI节点执行的操作。
例如，逻辑电路1510用于实现上文方法实施例中由AI节点执行的处理相关的操作，如，图4所示实施例中AI节点执行的处理相关的操作，或图8所示实施例中AI节点#1执行的处理相关的操作，或图10所示实施例中AI节点#1和/或AI节点#2执行的处理相关的操作，或图11、图12所示实施例中AI节点#1执行的处理相关的操作；输入/输出接口1520用于实现上文方法实施例中由AI节点执行的发送和/或接收相关的操作，如，图4所示实施例中AI节点执行的发送和/或接收相关的操作，或图8所示实施例中AI节点#1执行的发送和/或接收相关的操作，或图10所示实施例中AI节点#1和/或AI节点#2执行的发送和/或接收相关的操作，或图11、图12所示实施例中AI节点#1执行的发送和/或接收相关的操作。
本申请实施例还提供一种计算机可读存储介质，其上存储有用于实现上述各方法实施例中由通信装置或AI节点执行的方法的计算机指令。
本申请实施例还提供一种计算机程序产品，包含指令，该指令被计算机执行时以实现上述各方法实施例中由通信装置或AI节点执行的方法。
本申请实施例还提供一种通信系统，该通信系统包括上文各实施例中的通信装置和AI节点。
上述提供的任一种装置中相关内容的解释及有益效果均可参考上文提供的对应的方法实施例，此处不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅是示意性的，例如，上述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。此外，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
上述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元实现本申请提供的方案。
另外，在本申请各个实施例中的各功能单元可以集成在一个单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。该计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。例如，计算机可以是个人计算机，服务器，或者网络设备等。计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、 无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。关于计算机可读存储介质，可以参考上文描述。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (35)

  1. 一种通信方法，其特征在于，包括：
    通信装置接收人工智能AI连接的信息，所述AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；
    所述通信装置根据所述AI连接的信息与AI节点建立所述AI连接。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    所述通信装置通过所述AI连接向所述AI节点发送AI消息，和/或，所述通信装置通过所述AI连接接收来自所述AI节点的AI消息；
    其中，所述AI消息指示以下至少一项信息：所述AI消息采用的加密方式、所述AI消息采用的压缩方式、所述AI消息的消息类型、所述AI消息承载的是控制信息或数据、对所述AI消息进行完整性校验的验证码、所述AI消息承载的内容。
  3. 根据权利要求1或2所述的方法，其特征在于，所述方法还包括：
    所述通信装置通过所述AI连接接收更新后的所述AI连接的信息。
  4. 根据权利要求1至3中任一项所述的方法，其特征在于，所述方法还包括以下至少一项：
    所述AI连接的信息包括所述模型，所述通信装置处理所述模型；
    所述AI连接的信息包括所述数据集，所述通信装置根据所述数据集进行测量；
    所述AI连接的信息包括所述计算资源，所述通信装置使用所述计算资源执行AI任务。
  5. 根据权利要求1至4中任一项所述的方法，其特征在于，所述方法还包括：
    所述通信装置通过所述AI连接向所述AI节点发布第一AI任务；和/或，
    所述通信装置通过所述AI连接接收来自所述AI节点的第二AI任务。
  6. 根据权利要求5所述的方法，其特征在于，所述通信装置通过所述AI连接向所述AI节点发布第一AI任务，包括：
    所述通信装置通过所述AI连接向所述AI节点发布第一AI任务以通过所述AI节点向其他通信装置发布所述第一AI任务。
  7. 根据权利要求5或6所述的方法，其特征在于，所述方法还包括：
    在满足预设条件的情况下，所述通信装置发送所述第二AI任务的处理结果。
  8. 根据权利要求7所述的方法，其特征在于，所述第二AI任务中包括所述预设条件的指示信息。
  9. 根据权利要求1至8中任一项所述的方法，其特征在于，所述方法还包括：
    所述通信装置向所述AI节点发送请求消息，所述请求消息用于请求建立所述AI连接；或者，
    所述通信装置接收来自所述AI节点的指示信息，所述指示信息用于触发所述通信装置与所述AI节点建立所述AI连接。
  10. 根据权利要求9所述的方法，其特征在于，所述请求消息还包括请求类型的指示信息，所述请求类型用于表示请求建立所述AI连接的目的，所述数据集是根据所述请求类型确定的。
  11. 根据权利要求1至10中任一项所述的方法，其特征在于，
    所述AI连接的信息还包括以下至少一项：所述AI连接的标识、与所述AI连接关联的无线承载的信息。
  12. 根据权利要求1至11中任一项所述的方法，其特征在于，所述AI节点部署于以下任一项：网络设备、终端设备。
  13. 一种通信方法，其特征在于，包括：
    人工智能AI节点发送AI连接的信息，所述AI连接的信息包括为AI连接配置的以下至少一项：模型、数据集、计算资源；
    所述AI节点根据所述AI连接的信息与通信装置建立所述AI连接。
  14. 根据权利要求13所述的方法，其特征在于，所述方法还包括：
    所述AI节点通过所述AI连接向所述通信装置发送AI消息，和/或，所述AI节点通过所述AI连接接收来自所述通信装置的AI消息；
    其中，所述AI消息指示以下至少一项信息：所述AI消息采用的加密方式、所述AI消息采用的压缩方式、所述AI消息的消息类型、所述AI消息承载的是控制信息或数据、对所述AI消息进行完整性校验的验证码、所述AI消息承载的内容。
  15. 根据权利要求13或14所述的方法，其特征在于，所述方法还包括：
    所述AI节点通过所述AI连接向所述通信装置发送更新后的所述AI连接的信息。
  16. 根据权利要求13至15中任一项所述的方法，其特征在于，所述方法还包括以下至少一项：
    所述AI连接的信息包括所述模型，所述AI节点根据所述模型对所述通信装置发送的数据进行处理；
    所述AI连接的信息包括所述数据集，所述AI节点管理所述数据集；
    所述AI连接的信息包括所述计算资源，所述AI节点使用所述计算资源执行AI任务。
  17. 根据权利要求13至16中任一项所述的方法，其特征在于，所述方法还包括：
    所述AI节点根据所述通信装置的AI能力确定所述AI连接的信息。
  18. 根据权利要求13至17中任一项所述的方法，其特征在于，所述方法还包括：
    所述AI节点通过所述AI节点与至少一个通信装置中各个通信装置之间的AI连接，接收来自所述至少一个通信装置发布的第一AI任务；和/或，
    所述AI节点通过所述AI节点与所述至少一个通信装置中各个通信装置之间的AI连接，向所述至少一个通信装置发布第二AI任务；
    其中，所述至少一个通信装置包括所述通信装置。
  19. 根据权利要求18所述的方法，其特征在于，所述AI节点通过所述AI节点与所述通信装置之间的所述AI连接向所述通信装置发布第二AI任务，包括：
    所述AI节点通过所述AI节点与所述通信装置之间的所述AI连接向所述通信装置发布第二AI任务以通过所述通信装置向其他通信装置发布所述第二AI任务。
  20. 根据权利要求18或19所述的方法，其特征在于，所述方法还包括：
    在满足预设条件的情况下，所述AI节点通过所述AI节点与所述至少一个通信装置中各个通信装置之间的AI连接，向所述至少一个通信装置发送所述第一AI任务的处理结果。
  21. 根据权利要求20所述的方法，其特征在于，所述第一AI任务包括所述预设条件 的指示信息。
  22. 根据权利要求13至21中任一项所述的方法，其特征在于，所述方法还包括：
    所述AI节点接收来自所述通信装置的请求消息，所述请求消息用于请求建立所述AI连接；或者，
    所述AI节点向所述通信装置发送指示信息，所述指示信息用于触发所述通信装置与所述AI节点建立所述AI连接。
  23. 根据权利要求22所述的方法，其特征在于，所述请求消息还包括请求类型的指示信息，所述请求类型用于表示请求建立所述AI连接的目的，所述数据集是根据所述请求类型确定的。
  24. 根据权利要求13至23中任一项所述的方法，其特征在于，
    所述AI连接的信息还包括以下至少一项：所述AI连接的标识、与所述AI连接关联的无线承载的信息。
  25. 根据权利要求13至24中任一项所述的方法，其特征在于，所述AI节点包括第一AI节点和第二AI节点，所述第一AI节点和/或所述第二AI节点满足以下任一项：
    所述第一AI节点用于传输信令，所述第二AI节点用于处理AI任务；
    所述第二AI节点用于处理所述第一AI节点指示的AI操作；
    所述第一AI节点用于处理第一类AI任务，所述第二AI节点用于处理第二类AI任务，所述第一类AI任务与所述第二类AI任务不完全相同；
    所述第二AI节点用于保存和/或发送AI任务的处理结果。
  26. 根据权利要求25所述的方法，其特征在于，所述方法还包括：
    所述第一AI节点向所述第二AI节点发送通知信息，所述通知信息用于通知所述第二AI节点以下至少一项：执行AI任务、保存所述AI任务的处理结果、发送所述AI任务的处理结果。
  27. 根据权利要求13至26中任一项所述的方法，其特征在于，所述AI节点部署于以下任一项：网络设备、终端设备。
  28. 根据权利要求27所述的方法，其特征在于，所述AI节点部署于所述网络设备，所述方法还包括：
    所述网络设备根据以下至少一项确定与所述AI连接关联的无线承载的信息：所述模型、所述数据集、所述计算资源。
  29. 一种通信装置，其特征在于，包括：用于执行如权利要求1至28中任一项所述的方法的单元。
  30. 一种通信装置，其特征在于，包括：
    通信接口，用于输入和/或输出信息；
    处理器，用于执行计算机程序，以使得所述装置实现如权利要求1至28中任一项所述的方法。
  31. 一种通信装置，其特征在于，包括：
    存储器，用于存储可执行指令；
    处理器，用于调用并运行所述存储器中的所述可执行指令，以执行权利要求1至28中任一项所述的方法。
  32. 根据权利要求29至31中任一项所述的装置，其特征在于，所述装置为以下任一 项：通信设备、芯片或芯片系统。
  33. 一种通信系统，其特征在于，所述系统包括如权利要求30至32中任一项所述的装置。
  34. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有程序指令，当所述程序指令由处理器运行时，实现权利要求1至28中任一项所述的方法。
  35. 一种计算机程序产品，其特征在于，所述计算机程序产品包括计算机程序代码，当所述计算机程序代码在计算机上运行时，实现权利要求1至28中任一项所述的方法。
PCT/CN2023/089586 2022-06-24 2023-04-20 通信方法、通信装置和系统 WO2023246267A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210725738.9 2022-06-24
CN202210725738.9A CN117336820A (zh) 2022-06-24 2022-06-24 通信方法、通信装置和系统

Publications (1)

Publication Number Publication Date
WO2023246267A1 true WO2023246267A1 (zh) 2023-12-28

Family

ID=89281683

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/089586 WO2023246267A1 (zh) 2022-06-24 2023-04-20 通信方法、通信装置和系统

Country Status (2)

Country Link
CN (1) CN117336820A (zh)
WO (1) WO2023246267A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111869303A (zh) * 2020-06-03 2020-10-30 北京小米移动软件有限公司 一种资源调度的方法、装置、通信设备及存储介质
CN112188547A (zh) * 2020-09-09 2021-01-05 中国联合网络通信集团有限公司 一种业务处理方法及装置
WO2022052059A1 (zh) * 2020-09-11 2022-03-17 华为技术有限公司 一种通信方法及装置
WO2022088826A1 (zh) * 2020-10-28 2022-05-05 华为技术有限公司 一种计算服务发现方法及通信装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111869303A (zh) * 2020-06-03 2020-10-30 北京小米移动软件有限公司 一种资源调度的方法、装置、通信设备及存储介质
CN112188547A (zh) * 2020-09-09 2021-01-05 中国联合网络通信集团有限公司 一种业务处理方法及装置
WO2022052059A1 (zh) * 2020-09-11 2022-03-17 华为技术有限公司 一种通信方法及装置
WO2022088826A1 (zh) * 2020-10-28 2022-05-05 华为技术有限公司 一种计算服务发现方法及通信装置

Also Published As

Publication number Publication date
CN117336820A (zh) 2024-01-02

Similar Documents

Publication Publication Date Title
US11606824B2 (en) Network-initiated PDU session connection update method between terminal and network
US20230209390A1 (en) Intelligent Radio Access Network
WO2020156127A1 (zh) 无线通信的方法和通信设备
WO2020156126A1 (zh) 无线通信的方法和通信设备
EP4266756A1 (en) Network resource selection method, and terminal device and network device
WO2019029466A1 (zh) 一种时间配置的方法、网络设备及ue
US20230199550A1 (en) Relay Management Method and Communication Apparatus
CN112584545B (zh) 数据传输方法及装置
EP4221005A1 (en) Multipath transmission method and communication apparatus
TWI775009B (zh) 用於行動通訊系統之基地台及其資料傳輸方法
US20240015534A1 (en) Model processing method, communication apparatus, and system
WO2021163894A1 (zh) 一种基于中继的通信方法及装置
WO2023185615A1 (zh) 通信方法和装置
WO2023088465A1 (zh) 一种模型训练方法及相关装置
CN113747605A (zh) 通信方法和通信装置
WO2022179367A1 (en) New method for external parameter provisioning for an af session
WO2023246267A1 (zh) 通信方法、通信装置和系统
CN116939512A (zh) 群组通信的方法、装置和系统
CN113938985A (zh) 一种通信方法及装置
WO2023070684A1 (zh) 无线通信的方法和设备
WO2024082274A1 (zh) Ai任务指示的方法、通信装置和系统
WO2023202503A1 (zh) 通信方法和装置
WO2023030328A1 (zh) 无线网络服务的配置方法、终端及网络侧设备
WO2023116560A1 (zh) 通信的方法与装置
WO2023205959A1 (zh) 通信方法及通信装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23825920

Country of ref document: EP

Kind code of ref document: A1