A communication method and apparatus

Abstract

A communication method and device are used to implement a scheme in which a terminal hands over a computing task to a server for execution. In the present application, the terminal determines first information including an address of the server, and sends the first information to a RAN node, the first information being used by the RAN node to send data to the server. The RAN node receives the first information from the terminal, and sends data to the server according to the first information. The server executes a computing task according to the received data, so as to implement the terminal handing over the computing task to the server for execution.

Description

Technical Field

[0001] This application relates to the field of wireless communication, and more particularly to a communication method and apparatus. Background Technology

[0002] In communication systems, terminals need to perform computational tasks, such as artificial intelligence (AI) computational tasks. However, terminals typically have limited computing power, and performing computational tasks may affect their ability to perform other tasks, such as interfering with calls with other terminals. How to reduce the computational burden on terminals is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] This application provides a communication method and apparatus that enables the execution of computing tasks by a server, thereby improving the execution efficiency of computing tasks, saving computing resources of the terminal, and reducing the computing burden on the terminal.

[0004] In a first aspect, embodiments of this application provide a communication method that can be applied to the terminal side, such as a terminal or a communication module and / or computing module in the terminal, or a circuit or chip in the terminal responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core or a system-in-package (SIP) chip), or a circuit or chip in the terminal responsible for communication and / or computing functions (such as a graphics processing unit (GPU), an AI processor, or an application-specific integrated circuit (ASIC)), or a logic node, logic module, or software that can implement all or part of the terminal functions.

[0005] Taking the application of this method to a terminal as an example, in this method, the terminal determines first information, which includes the address of the server. The terminal sends the first information to a radio access network (RAN) node, and the first information is used by the RAN node to send data to the server. The server's address, for example, is the server's Internet Protocol (IP) address. The server is used to provide services to the terminal; the server can be referred to as the target server, and correspondingly, the server's address can also be referred to as the target server's address.

[0006] In the above technical solutions, the terminal can delegate computing tasks to the server. In this case, the server needs to obtain the data required to execute the computing task and then execute the task based on the obtained data. In this embodiment, the RAN node sends the data required for the computing task to the server. Considering that the RAN node does not know the server's address, the terminal can first determine the first information, which includes the server's address. The terminal sends the first information to the RAN node, allowing the RAN node to obtain the server's address from the first information and then send data to the server. This enables the server to obtain the data and execute the computing task based on it. Compared to the terminal executing the computing task itself, this method helps improve the execution efficiency of the computing task and saves the terminal's computing resources.

[0007] In one possible design, the first piece of information is used for data encapsulation.

[0008] In one possible design, the server's address is encapsulated in the packet header, and the packet itself includes the data. For example, the data is included in the packet's payload.

[0009] In the above technical solution, the RAN node can encapsulate data based on first information (such as the server's address) to obtain a data packet, and then send the data packet to the server. Furthermore, when the RAN node sends a data packet to the server, the routing device (e.g., the user plane function (UPF)) can perform routing based on the packet header, thereby enabling the server to obtain the data packet.

[0010] In one possible design, the first information may also include one or more of the following: the terminal's address, the terminal's port number, the server's port number (also known as the target server's port number), the transport layer protocol, the application layer protocol, and the network layer protocol.

[0011] In the above technical solution, the possibility of providing parameters included in the first information is provided so that when the RAN node sends a data packet to the server, the routing device can perform routing based on the packet header, thereby enabling the server to obtain the data packet.

[0012] In one possible example, the first information includes the terminal's address, the server's address, the terminal's port number, the server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol. The application layer protocol is used to encapsulate application layer data (i.e., data related to computational tasks) to obtain application layer messages (e.g., Hypertext Transfer Protocol (HTTP) messages); the terminal's port number, the server's port number, and the transport layer protocol are used to encapsulate application layer messages to obtain transport layer messages (e.g., Transmission Control Protocol (TCP) segments); and the terminal's address, the server's address, and the network layer protocol are used to encapsulate transport layer messages to obtain network layer messages (e.g., IP packets).

[0013] In the above technical solution, the first information includes encapsulation information for the RAN node to encapsulate data (or, in other words, the first information is encapsulation information), so that the RAN node can encapsulate data based on the first information to obtain data packets. Furthermore, the terminal instructs the RAN node on the encapsulation information, and can carry different encapsulation protocols in the encapsulation information based on different transmission requirements. For example, based on transmission requirements, the terminal determines that the application layer protocol is one of HTTP or Hypertext Transfer Protocol over Secure Socket Layer (HTTPS); or, based on transmission requirements, the terminal determines that the transport layer protocol is one of UDP or Transmission Control Protocol (TCP) or Quick UDP Internet Connections (QUIC); or, based on transmission requirements, the terminal determines that the network layer protocol is one of Address Resolution Protocol (ARP), IP, or Internet Control Message Protocol (ICMP). This helps to improve the flexibility of data transmission methods.

[0014] In one possible design, when the terminal determines the first information, it may specifically obtain second information from the core network equipment. The second information includes one or more of the following: the terminal's address, the terminal's port number, the server's address, the server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol. The terminal determines the first information based on the second information.

[0015] In the above technical solution, when the terminal cannot determine the first information on its own, it can request second information from the core network device. That is, it can request one or more of the following: the terminal's address, the terminal's port number, the server's address, the server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol. Based on the requested second information, the first information to be sent to the RAN node can then be determined. Furthermore, in this method, the core network device can allocate a server (i.e., the target server) that closely matches the terminal's computing task requirements.

[0016] In one possible example, when the terminal obtains the second information from the core network device, specifically, the terminal sends a first request to the core network device. The core network device receives the first request from the terminal and, based on the first request, sends the second information to the terminal; that is, the second information responds to the first request. The terminal receives the second information from the core network device. Here, the second information may include one or more of the following: the terminal's port number, the server's address, the server's port number, a transport layer protocol, an application layer protocol, and a network layer protocol. In another possible example, when the terminal obtains the second information from the core network device, specifically, it receives a Protocol Data Unit (PDU) session response from the core network device, the PDU session response including the terminal's address. That is, the terminal can obtain its address (e.g., the terminal's IP address) during the PDU session establishment process.

[0017] Core network equipment includes, for example, access and mobility management function (AMF) network elements or session management function (SMF) network elements.

[0018] The above technical solution provides a method for the terminal to obtain second information from the core network equipment.

[0019] In one possible design, before sending the first information to the RAN node, the terminal can also receive a second request from the RAN node, which requests the first information. This second request can be considered as the RAN node requesting "permission for the terminal to send data to the server on its behalf," and can also be referred to as a proxy permission request.

[0020] In the above technical solution, after the RAN node obtains proxy permissions, that is, after obtaining the first information, the RAN node can act as a proxy for the terminal and send data to the server.

[0021] In one possible design, data is used by the server to train a first model, and also by the RAN nodes to train a second model; the first and second models are paired (or used together), and the data includes training data and / or a third model. For example, the terminal also receives the first model from the server. The first model is used by the terminal, and the second model is used by the RAN nodes.

[0022] The third model is, for example, the initial model. The third model can be obtained by the RAN node based on the training data. The training data here can be the same as or different from the training data sent by the RAN node to the server.

[0023] In the above technical solution, the server can train the first model instead of the terminal, and the RAN node trains the second model. The server and RAN node use the same training data when training their respective models, thus achieving better results when the first and second models are used in combination. For example, after the server sends the trained first model to the terminal, the terminal can interact with the RAN node based on the first model, and the RAN node can interact with the terminal based on the second model, achieving good results.

[0024] In one possible design, the training data includes channel state information (CSI). The first model is a CSI compression (encode) model applied to the terminal side, and the second model is a CSI decompression (decode) model applied to the RAN node side.

[0025] The above technical solution is applicable to CSI compression scenarios. For example, the terminal can compress the CSI obtained from the measured channel state information-reference signal (CSI-RS) according to the CSI compression model to obtain compressed information, and then send the compressed information to the RAN node. The RAN node then decompresses the received compressed information according to the CSI decompression model. Compared with the terminal directly sending the CSI to the RAN node, this helps to save air interface overhead, improve spectrum efficiency, enhance system performance, and meet the needs of large-scale connections in multiple-input multiple-output (MIMO) systems.

[0026] In one possible design, the terminal also sends a Quality of Service Flow ID (QFI) to the RAN node. The QFI indicates a Quality of Service (QoS) flow, which is used by the RAN node to send data to the server via the QoS flow.

[0027] In the above technical solution, since the PDU session established between the terminal and the network may include multiple QoS streams, the terminal can indicate to the RAN node which QoS stream the data is transmitted by, that is, send a QFI to the RAN node so that the RAN node can transmit the data.

[0028] In one possible design, the terminal determines the QFI based on one or more of the following: the server's address, the server's port number, the terminal's address, the terminal's port number, and the transport layer protocol.

[0029] The above technical solution provides a way for the terminal to determine the QFI, so that the terminal can match the QoS stream suitable for data transmission requirements, which helps the QoS stream to transmit data more efficiently.

[0030] Secondly, this method can be applied to the RAN node side, such as RAN nodes, modules in RAN nodes (e.g., circuits, chips or chip systems), or logic nodes, logic modules or software that can implement all or part of the functions of RAN nodes, or circuits or chips in RAN nodes that are responsible for communication and / or computing functions (e.g., GPUs, AI processors, or ASICs).

[0031] Taking the application of this method to a RAN node as an example, in this method, the RAN node receives first information from the terminal, which includes the server's address. Based on the first information, the RAN node sends data to the server.

[0032] In one possible design, when a RAN node sends data to the server based on the first information, specifically, the RAN node encapsulates the data based on the first information to obtain a data packet. The RAN node then sends the data packet to the server.

[0033] In one possible design, the server's address is encapsulated in the packet header, and the packet contains the data. For example, the packet payload includes the data. For instance, the RAN node encapsulates the server's address in the packet header and the data in the packet payload.

[0034] In one possible design, the first information may also include one or more of the following: the terminal's address, the terminal's port number, the server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol.

[0035] In one possible design, the RAN node sends a second request to the terminal before receiving the first information from the terminal. This second request is used to request the first information. The second request can also be called a proxy authorization request.

[0036] In one possible design, the data is used to train a first model on the server, and the data is also used to train a second model on the RAN nodes; the first and second models are paired, and the data includes training data and / or a third model.

[0037] In one possible design, the training data includes CSI, the first model is a CSI compression model applied to the terminal side, and the second model is a CSI decompression model applied to the RAN node side.

[0038] In one possible design, the RAN node also receives a QFI from the terminal, which indicates a QoS flow. The RAN node sends data to the server, specifically via a QoS flow. For example, the RAN node sends data to the UPF via a QoS flow, and the UPF then sends data to the server. The data is specifically data packets encapsulated by the RAN node based on the first information.

[0039] Thirdly, this method can be applied to the server side, such as servers, modules in servers (e.g., circuits, chips or chip systems), or logical nodes, logical modules or software that can implement all or part of the server functions, or circuits or chips in servers that are responsible for communication and / or computing functions (e.g., GPUs, AI processors, or ASICs).

[0040] Taking the application of this method to a server as an example, in this method, the server receives data from the RAN node. The data is encapsulated by first information, which includes the server's address. Based on the data, the server trains a first model and sends the first model to the terminal.

[0041] In one possible design, the data includes training data and / or a third model.

[0042] Fourthly, this method can be applied to the core network equipment side, such as core network equipment, modules (e.g., circuits, chips or chip systems) in core network equipment, or logical nodes, logical modules or software that can implement all or part of the server functions, or circuits or chips (e.g., GPUs, AI processors, or ASICs) in servers that are responsible for communication and / or computing functions.

[0043] The core network equipment sends second information to the terminal, which includes one or more of the following: the terminal's address, the terminal's port number, the server's address, the server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol.

[0044] In one possible example, when the core network device sends the second information to the terminal, it may specifically be that the core network device receives a first request from the terminal. Based on the first request, the core network device then sends the second information to the terminal.

[0045] In one possible example, when the core network device sends the second information to the terminal, specifically, the core network device receives a PDU session request from the terminal. The core network device then sends a PDU session response to the terminal, which includes the terminal's address.

[0046] Core network equipment includes, for example, AMF network elements or SMF network elements.

[0047] Fifthly, this application provides a communication device that has the functions of the first aspect described above. For example, the communication device includes modules, units, or means that perform the operations involved in the first aspect, or modules, units, or means that perform the operations involved in the second aspect, or modules, units, or means that perform the operations involved in the third aspect, or modules, units, or means that perform the operations involved in the fourth aspect. These modules, units, or means can be implemented by software, by hardware, or by a combination of software and hardware.

[0048] Sixthly, this application provides a communication device including an interface circuit and one or more processors. The one or more processors are coupled to a memory. The memory is used to store part or all of the necessary computer programs or instructions for implementing the functions involved in the first aspect above. Alternatively, the memory is used to store part or all of the necessary computer programs or instructions for implementing the functions involved in the second aspect above. Alternatively, the memory is used to store part or all of the necessary computer programs or instructions for implementing the functions involved in the third aspect above. Alternatively, the memory is used to store part or all of the necessary computer programs or instructions for implementing the functions involved in the fourth aspect above.

[0049] The one or more processors can execute the computer program or instructions, which, when executed, cause the communication device to implement any possible design or implementation method in the first aspect described above. The interface circuit is used to implement communication functions within the communication device and / or communication functions between the communication device and other devices or components.

[0050] In one possible design, the processor is used to communicate with other devices or components through the interface circuit.

[0051] In one possible design, the communication device may also include the memory.

[0052] The aforementioned communication device may be a terminal, or a communication and / or computing module in a terminal, or a chip in a terminal responsible for communication functions such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a circuit or chip in a terminal responsible for communication and / or computing functions (such as a GPU, AI processor, or ASIC), or a logical node or logical module capable of implementing all or part of the terminal functions.

[0053] In a seventh aspect, this application provides a communication system including one or more of a terminal, a RAN node, a server, and a core network device, wherein the terminal is used to perform the method in the first aspect or any possible design of the first aspect, the RAN node is used to perform the method in the second aspect or any possible design of the second aspect, the server is used to perform the method in the third aspect or any possible design of the third aspect, and the core network device is used to perform the method in the fourth aspect or any possible design of the fourth aspect.

[0054] Eighthly, this application provides a computer-readable storage medium storing computer-readable instructions that, when read and executed by a computer, cause the computer to perform any of the possible designs in the first to fourth aspects described above.

[0055] Ninthly, this application provides a computer program product that, when read and executed by a computer, causes the computer to perform any of the possible designs in the first to fourth aspects described above.

[0056] The technical effects that can be achieved by any of the second to ninth aspects mentioned above can be referred to the descriptions of the beneficial effects in the other aspects. Attached Figure Description

[0057] Figure 1 This is a network architecture diagram of a communication system;

[0058] Figure 2 This is a schematic diagram of the structure of a communication chip for a terminal.

[0059] Figure 3 This is a network architecture diagram for yet another type of communication system;

[0060] Figure 4 This is a schematic diagram illustrating the relationship between a PDU session, radio bearer, and QoS flow.

[0061] Figure 5 A schematic diagram illustrating the matching of a terminal with a PDU session;

[0062] Figure 6 A flowchart illustrating a communication method provided in this application;

[0063] Figure 7 A schematic diagram of a CSI compression process provided in this application;

[0064] Figure 8 A schematic diagram of the structure of a dual-end model provided in this application;

[0065] Figure 9 A flowchart illustrating the first communication method for CSI compression scenarios provided in this application;

[0066] Figure 10 A schematic diagram illustrating the interaction between RAN nodes, terminals, and servers in the CSI compression scenario provided in this application;

[0067] Figure 11 A flowchart illustrating the communication method in the second CSI compression scenario provided in this application;

[0068] Figure 12 A flowchart illustrating the communication method in the third CSI compression scenario provided in this application;

[0069] Figure 13 A flowchart illustrating a capability negotiation process provided for this application;

[0070] Figure 14 A schematic diagram of the structure of a communication device provided in this application;

[0071] Figure 15 This is a schematic diagram of the structure of a communication device provided in this application. Detailed Implementation

[0072] The embodiments of this application can be applied to various communication systems, such as Universal Mobile Telecommunications System (UMTS), Wireless Local Area Network (WLAN), Wireless Fidelity (Wi-Fi) system, 4th generation (4G) mobile communication system (such as Long Term Evolution (LTE) system), 5th generation (5G) mobile communication system (such as New Radio (NR) system), and future communication networks, etc.

[0073] The 3rd Generation Partnership Project (3GPP) standards group defined the 5G network architecture. This architecture supports radio access technologies (such as LTE access technology and 5G radio access network (R)AN) access technology) to the 5G core network (CN). R)AN and RAN are interchangeable; for ease of description, RAN will be used as an example below.

[0074] For example, Figure 1 A schematic diagram of the architecture of a possible communication system applicable to embodiments of this application is shown.

[0075] Figure 1 The architecture shown is a 5G network architecture based on a service-oriented architecture. Figure 1 The document illustrates the interaction relationships between network functions and entities, as well as their corresponding interfaces. For example, terminals and access and mobility management function (AMF) network elements can interact through the N1 interface, and the interaction messages are called N1 messages.

[0076] Figure 1 Some of the interfaces in the document are implemented using a service-oriented approach.

[0077] The architecture of this communication system may include terminals, RAN, data network (DN), and core network.

[0078] 1. RAN

[0079] RAN can be a 3GPP-related cellular system, such as 4G, 5G mobile communication systems, or future-oriented evolution systems. RAN can also be an open RAN (O-RAN or ORAN), a cloud RAN (CRAN), a virtualized RAN (vRAN), an artificial intelligence radio access network (AI RAN), or a wireless fidelity (WiFi) system. RAN can also be a communication system that integrates two or more of the above systems.

[0080] The RAN includes RAN nodes. RAN nodes, sometimes also called access network nodes, access network equipment, RAN entities, or access nodes, are part of the communication system and are used to help terminals achieve wireless access.

[0081] In one possible scenario, a RAN node can be a base station, an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a base station in a future mobile communication system, or an access node in a WiFi system. A RAN node can be a macro base station, a micro base station, an indoor station, a relay node, a donor node, or a radio controller in a CRAN scenario. Optionally, a RAN node can also be a server, a wearable device, a vehicle, or in-vehicle equipment. For example, in vehicle-to-everything (V2X) technology, the RAN node can be a roadside unit (RSU).

[0082] All or part of the functions of the RAN node in this application can also be implemented through software functions executed on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node may also include communication modules, circuits, or chips that perform corresponding communication functions. The RAN node may also be configured with program instructions for performing corresponding communication functions and corresponding program instructions. The RAN node in this application can also be a logical node, logical module, or software capable of implementing all or part of the RAN node functions, or a circuit or chip (such as a GPU, AI processor, or ASIC) responsible for communication and / or computing functions in an access node.

[0083] In another possible scenario, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes each implementing some of the base station's functions. For example, RAN nodes can be central units (CUs), distributed units (DUs), CU-control plane (CPs), CU-user plane (UPs), or radio units (RUs), etc. CUs and DUs can be set up separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio frequency equipment or radio frequency units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs). Furthermore, RAN nodes can also be computing units, providing computing power for tasks (such as model inference and / or model training), and can also be used to implement one or more of the following: task partitioning, scheduling, and orchestration. The functionality of a computing unit can be implemented by a separate module independent of other units (e.g., CU, DU, RU), or by one or more other units (e.g., one or more of CU, DU, RU).

[0084] In different systems, CU (or CU-CP and CU-UP), DU, computing unit, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (Open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, computing unit, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, computing unit, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules. For example, the interface between DU and RU can be called a fronthaul interface.

[0085] When the RAN is O-RAN, it can also have AI capabilities. For example, O-RAN includes an intelligent controller. The intelligent controller can be a non-real-time RAN intelligent controller (RIC / non-RT RIC / NRT RIC) or a near-real-time RAN intelligent controller (RIC / near-RT RIC / nRT RIC). The non-real-time RIC can be used to implement non-real-time intelligent management of RAN functions, enabling workflows including model training and model updates, and guiding applications / functions in the nRT RIC based on policies. The near-real-time RIC can be used to implement near-real-time intelligent management of the RAN. For example, near-real-time control and optimization of O-RAN modules and resources can be achieved through data collection and related operations on the E2 interface.

[0086] 2. Terminal

[0087] The terminal can be a device or module that accesses the aforementioned communication system and has corresponding communication functions. It can also be called a terminal device, user equipment (UE, mobile station, mobile terminal, etc.). Terminals can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), the Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, and smart cities. Terminals can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, transportation vehicles with wireless communication capabilities, communication modules, etc.

[0088] When a terminal is used in V2X, it can also be called a V2X device, such as a smart car, an unmanned car, a driverless car, a pilotless car, or an automobile, or a roadside unit (RSU). The various terminals described above, if located on a vehicle (e.g., placed / installed inside the vehicle), can all be considered in-vehicle terminals. An in-vehicle terminal can be built into a vehicle's in-vehicle module, in-vehicle component, in-vehicle chip, or in-vehicle unit as one or more components or units. The vehicle can implement the methods of this application through the built-in in-vehicle module, in-vehicle component, in-vehicle chip, or in-vehicle unit. An in-vehicle terminal can be a complete vehicle device, an in-vehicle module, a vehicle, an on-board unit (OBU), an RSU, an in-vehicle infotainment system (or in-vehicle transmission unit) (telematics box, T-box), a chip, or a system-on-chip (SoC), etc. The aforementioned chip or SoC can be installed in the vehicle, OBU, RSU, or T-box.

[0089] Terminals can establish connections with the operator's network through interfaces provided by the operator's network (such as N1) and use data, voice, and video services provided by the operator's network. Terminals can also access the DN (Network Provider) through the operator's network and use operator services deployed on the DN, and / or services provided by third parties. These third parties can be service providers outside of the operator's network and the terminal, providing data, voice, and video / voice services to the terminal. The specific form of the third party can be determined based on the actual application scenario and is not limited here.

[0090] The embodiments of this application do not limit the device form of the terminal. The terminal typically includes a communication module, circuit, or chip that performs corresponding communication functions. Furthermore, it may also include modules, circuits, or chips (such as GPUs, AI processors, or ASICs) that perform corresponding communication and / or computing functions. The terminal may also be configured with program instructions for performing corresponding communication and / or computing functions.

[0091] For example, Figure 2The diagram illustrates the structure of the terminal's communication chip, which mainly consists of a baseband subsystem, a radio frequency (RF) subsystem, a power management subsystem, and peripherals. The baseband subsystem can implement one or more functions, including application layer processing, external interfaces, and layer 3 (L3) / layer 2 (L2) / L1 communication protocol processing. The RF subsystem may include an RF front-end and antenna, primarily implementing one or more functions such as spatial electromagnetic wave to electrical signal conversion, signal amplification, and signal filtering. The RF subsystem can also connect to the baseband subsystem to perform analog signal frequency conversion and nonlinear distortion correction. The power subsystem, also known as the power management subsystem, performs power management functions. For example, the power subsystem can connect to the baseband and RF subsystems to manage their power supply. Peripherals may include memory and / or external interfaces, enabling communication between the chip and external devices.

[0092] 3. Core Network

[0093] The core network may include: network slice selection function (NSSF) network elements, network exposure function (NEF) network elements, network function repository function (NRF) network elements, policy control function (PCF) network elements, unified data management (UDM) network elements, application function (AF) network elements, authentication server function (AUSF) network elements, AMF network elements, SMF network elements, UPF network elements, network data analytics function (NWDAF) network elements, network slice admission control function (NSACF) network elements, and network slice-specific authentication and authorization function (NSSAAF) network elements, etc.

[0094] The NSSF network element is used to select the target network element. The NSSF network element selects the target network element based on acquired information, such as parameters provided by the consumer or subscription information obtained from the subscription data management network element. It is understood that in future communication systems, the network element responsible for the above functions may still be an NSSF network element, or may have other names; this application does not impose any limitations.

[0095] AMF (Automatic Facilitation Function) network elements can be used to manage terminal access control and mobility. Specifically, they can be responsible for terminal registration, mobility management, tracking area update procedures, reachability detection, selection of session management function network elements, and mobility state transition management. It is understood that in future communication systems, the network elements responsible for the above functions may still be AMF network elements, or may have other names; this application does not impose any limitations.

[0096] SMF network elements can be used to manage terminal sessions (including session establishment, modification, and release), select and reselect UPF network elements, allocate terminal IP addresses, control QoS, collect billing data, and manage roaming. It is understood that in future communication systems, the network elements responsible for the above functions may still be SMF network elements, or may have other names; this application does not impose any limitations.

[0097] The UPF network element is responsible for forwarding and receiving user data in the terminal. It can receive user data from the DN and transmit it to the terminal through the RAN node; the UPF network element can also receive user data from the terminal through the RAN node and forward it to the DN. The transmission resources and scheduling functions providing services to the terminal in the UPF network element are managed and controlled by the SMF network element. It is understood that in future communication systems, the network element responsible for the above functions may still be a UPF network element, or may have other names; this application does not impose any limitations.

[0098] PCF network elements primarily support providing a unified policy framework to control network behavior, providing policy rules to control layer network functions, and acquiring user subscription information related to policy decisions. For example, they can provide rules based on service data flow and application detection, gating, QoS, and flow-based charging control. It is understood that in future communication systems, the network element responsible for the above functions may still be a PCF network element, or may have other names; this application does not impose any limitations.

[0099] UDM network elements are used to generate authentication credentials, process user identifiers (such as storing and managing permanent user identities), control access authorization, and manage subscription data. It is understood that in future communication systems, the network elements responsible for these functions may still be UDM network elements, or may have other names; this application does not impose any limitations.

[0100] AF (AF-based) network elements are used for data routing affected by applications, accessing network open functions, or interacting with policy frameworks for policy control. It is understood that in future communication systems, the network elements responsible for these functions may still be AF network elements, or may have other names; this application does not limit this.

[0101] NEF network elements are used to securely expose services and capabilities provided by 3GPP network functions to the outside world. It is understood that in future communication systems, the network element responsible for the above functions may still be an NEF network element, or may have other names; this application does not limit this.

[0102] NRF network elements are used to store network functional entities and their descriptions of the services they provide, as well as to support functions such as service discovery and network element entity discovery. It is understood that in future communication systems, the network elements responsible for the above functions may still be NRF network elements, or may have other names; this application does not impose any limitations.

[0103] NWDAF network elements can be used for big data analysis. It is understood that in future communication systems, the network elements responsible for the above functions may still be NWDAF network elements, or may have other names; this application does not limit this.

[0104] The NSACF network element is responsible for controlling the number of terminals registered in each network slice and the number of sessions corresponding to each network slice. It is understood that in future communication systems, the network element responsible for the above functions may still be an NSACF network element, or may have other names; this application does not limit this.

[0105] The NSSAAF network element is responsible for the authentication and authorization of network slices. In future communication systems, the network element responsible for the above functions may still be an NSSAAF network element, or may have other names; this application does not impose any restrictions.

[0106] 4. DN

[0107] A DN, also known as a packet data network (PDN), is a data network that provides services to users. Typically, the client is located at the terminal, and the server is located at the DN. A DN can be a private network, such as a local area network (LAN), an external network not controlled by the operator, such as the Internet, or a dedicated network jointly deployed by operators, such as the network providing IP multimedia core network subsystem (IMS) services.

[0108] It is understood that in practical applications, the above network architecture may also include other network elements, and this application does not limit this.

[0109] It should be noted that the names of the network elements in this application are merely examples, and this application does not preclude the possibility of using other names for the network elements in the future, or the merging of functions between the network elements. With the evolution of technology, any device or network element capable of implementing the functions of the aforementioned network elements is within the scope of protection of this application. Figure 1 The interface names between the various network elements are merely examples; in actual implementations, the interface names may differ, and this application does not impose any specific limitations on them. Furthermore, the names of the messages (or signaling) transmitted between the aforementioned network elements are also merely examples and do not constitute any limitation on the function of the messages themselves.

[0110] For ease of description, each network element will be referred to by its corresponding English abbreviation in the following text. For example, “AMF” represents the access and mobility management function network element, “SMF” represents the session management function network element, and so on.

[0111] See Figure 3 The diagram illustrates a network architecture diagram of another communication system to which embodiments of this application are applicable, which is a RAN-based service architecture. Figure 3 Only some network function (NF) elements are listed, such as Short Message Service Function (SMSF), PCF, SMF, AMF, UPF, etc. Figure 3 The dashed line in the middle indicates that the connection is not yet complete.

[0112] Among them, the functions of SMSF include: NAS short message processing function, which can simultaneously provide SMS services to terminals, that is, realize interconnection with the SMS center, etc.

[0113] from Figure 3As can be seen, in a RAN-based service architecture, terminals can communicate directly with NFs via the RAN without using the AMF as an anchor point. This reduces AMF congestion. Furthermore, since the AMF doesn't need to be an anchor point, data from the terminal to other NFs doesn't pass through the AMF, achieving data isolation and simplifying the process. For the RAN, the N2 interface can also be based on a service-oriented interface protocol, eliminating the need for the AMF as an anchor point. Therefore, the RAN only needs to maintain one set of protocols for each NF. Compared to using the AMF as an anchor point, where operators need to maintain one set of protocols between the RAN node and the AMF, and another set between the AMF and other NFs, the operator only needs to maintain one set of protocols between the RAN node and the NFs, resulting in fewer protocols and lower complexity. Additionally, during the process of the terminal sending messages to establish a connection with the AMF, access stratum (AS) security is activated, ensuring connection security. Moreover, radio resource control (RRC) connections and N2 interface connections can be established, ensuring reliable message transmission after the connection is established.

[0114] For ease of description, the embodiments of this application are applied to Figure 1 or Figure 3 The network architecture shown is an example. The system described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems. It should be noted that the above... Figure 1 , Figure 3 The network elements, functions, or services mentioned can be network components in hardware devices, software functions running on dedicated hardware, or virtualized functions instantiated on a platform (e.g., a cloud platform). Optionally, the aforementioned network elements, functions, or services can be implemented by a single device, multiple devices working together, or a functional module within a single device; this application does not specifically limit this. Furthermore, some English abbreviations used in this document to describe embodiments of this application using the current 5G network as an example may change as the network evolves; specific evolution can be found in the descriptions in the relevant standards.

[0115] The relevant technical features involved in the embodiments of this application will be explained below. It should be noted that these explanations are for the purpose of making the embodiments of this application easier to understand, and should not be regarded as a limitation on the scope of protection claimed by this application.

[0116] I. Quality of Service (QoS)

[0117] To provide different qualities of service (QoS) for different services, wireless networks offer QoS. QoS management is a control mechanism for wireless networks to meet the quality requirements of different services. It is an end-to-end process that requires the cooperation of all network nodes along the route from initiator to responder to ensure service quality. Air interface QoS management features provide different end-to-end QoS for various services and users with different needs.

[0118] Within a PDU session, a QoS flow is the smallest unit of granularity that distinguishes QoS. In 5G systems, QoS flows are identified by QFIs, and a QFI is unique for a QoS flow within a PDU session. This means that a PDU session can have multiple QoS flows; for example, a PDU session can have a maximum of 64 QoS flows, each with a different QFI. Within a PDU session, user plane data with the same QFI is transmitted through the same QoS flow, meaning it uses the same service forwarding processing method (such as scheduling).

[0119] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating the relationship between a PDU session, a data radiobearer (DRB), and a QoS stream, as disclosed in an embodiment of this application. Figure 4 As shown, at the configuration granularity, a PDU session can correspond to multiple DRBs, and services on the same DRB can also use different service levels, i.e., different QoS flows.

[0120] II. User Equipment Route Selection Policy (URSP)

[0121] The terminal selects a suitable PDU session for its uplink service flow based on the URSP. That is, some uplink services of the terminal have certain requirements for the data network name (DNN), slice, and session service continuity mode (SSC mode) of the PDU session used.

[0122] During the execution of URSP, the terminal may trigger the establishment or modification of a PDU session. For example, if the terminal determines that no suitable PDU session exists, it will initiate a PDU session establishment process; if the terminal determines that a suitable PDU session exists, it may directly use the existing PDU session. Combined with... Figure 5In the example, when the terminal is preparing to send data for application A, it determines that application A requires PDU sessions DNN1, S-NSSAI-a, and SSC2. Then the terminal can determine that application A will transmit data through PDU session 2.

[0123] For example, during the execution of URSP, the terminal may specifically evaluate whether a new application matches a traffic descriptor in the URSP rule according to rule precedence when it detects a new application. If a match is found, the terminal selects the route selection descriptor (RSD) in the URSP rule according to route selection descriptor precedence (RSD precedence). If the RSD is invalid (valid), the terminal moves to the next RSD; otherwise, it selects the current RSD. For example, a URSP rule's RSD should be considered valid only if all of the following conditions are met: Condition 1, if any S-NSSAI(s) exist, then the S-NSSAI(s) are among the allowed NSSAIs in non-roaming situations; in roaming situations, the S-NSSAI(s) are included in the mapping of allowed NSSAIs to HPLMN S-NSSAI(s). Condition 2: If any Data Network Neighborhood (DNN) exists, and the DNN is a Local Area Data Network (LADN) DNN, then the terminal is within the availability zone of this LADN. Condition 3: If a time window exists, and the time matches the time indicated within the time window. Condition 4: If a location standard exists, and the terminal's location matches the information indicated in the location standard.

[0124] When a PDU session that meets the requirements (with attributes identical to RSD) exists, the terminal routes the application's data packets to that PDU session (if multiple PDU sessions exist, the terminal selects one based on configuration, etc.). If no PDU session meets the requirements, the terminal initiates the establishment of a PDU session with the specified attributes. In short, the terminal first checks if a PDU session meets the application's requirements (DNN, network slicing, etc.). If so, the application is mapped to that session; otherwise, a PDU session meeting the application's requirements is established.

[0125] The attributes of a PDU session include network slices, DNN, and SSC modes. 3GPP defines the URSP as the mapping between applications and network slices, DNN, and SSC modes. For example, a URSP includes one or more URSP rules. A URSP rule mainly consists of two parts: a traffic descriptor and an RSD. The traffic descriptor includes the names or identifiers of multiple applications, and the RSD includes the NSSAI corresponding to each application—that is, the NSSAIs that the application can use but are not included in the traffic descriptor.

[0126] In communication systems, terminals need to perform computational tasks, such as AI computations. However, terminals typically have limited computing power, and performing computational tasks may interfere with the execution of other tasks, such as communication with other terminals. Currently, the idea is to offload computational tasks to servers, with a corresponding relationship between the terminal and the server. For example, if the terminal is manufactured by manufacturer A, the server may also be provided by manufacturer A; or, if the terminal is manufactured by manufacturer A, the chip installed in the terminal is manufactured by manufacturer B, and the server may also be provided by manufacturer B.

[0127] like Figure 6 This is a schematic flowchart illustrating a communication method exemplified by this application. The terminal sends first information (including the address of the target server) to the RAN node. The RAN node can send relevant data for the computing task to the target server based on the target server's address. The target server can then execute the computing task based on the relevant data. This achieves the method of the terminal entrusting the computing task to the target server for execution. The target server is a server used to provide services to the terminal.

[0128] Step 601: The terminal determines the first information.

[0129] The first piece of information includes the address of the target server. For example, the address of the target server is its IP address.

[0130] For example, the first information may also include one or more of the following: the terminal's address, the terminal's port number, the target server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol. For instance, the terminal's address is the terminal's IP address.

[0131] The application layer protocol included in the first piece of information can specifically be the application layer protocol used to encapsulate the data to be sent. This application layer protocol can be determined by the terminal itself or obtained from the core network equipment. Examples of application layer protocols in the first piece of information include HTTP and HTTPS.

[0132] The transport layer protocol included in the first piece of information can specifically be the transport layer protocol used to encapsulate the data to be sent. This transport layer protocol can be determined by the terminal itself or obtained from the core network equipment. Examples of transport layer protocols in the first piece of information include UDP, TCP, and QUIC.

[0133] The network layer protocol included in the first piece of information can specifically be a network layer protocol used to encapsulate the data to be sent. This network layer protocol can be determined by the terminal itself or obtained from the core network equipment. Examples of network layer protocols in the first piece of information include ARP, IP, or ICMP.

[0134] In the first possible example, the first information includes the terminal's address, the terminal's port number, the target server's address, the target server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol.

[0135] For example, the first piece of information includes the terminal's address, the terminal's port number, the target server's address, the target server's port number, the transport layer protocol (TCP), the application layer protocol (HTTPS), and the network layer protocol (IP).

[0136] In the second possible example, the first information includes the terminal's address, the terminal's port number, the target server's address, and the target server's port number.

[0137] In the third possible example, the first piece of information includes the address of the terminal and the address of the target server.

[0138] The following example provides a method for a first terminal to determine the first information:

[0139] In the first possible implementation, the terminal determines the first information based on its local configuration. For example, the first information is pre-configured in the terminal's memory before it leaves the factory. Alternatively, the memory may contain the first information, and the terminal's processor can read the first information from the memory.

[0140] In a second possible implementation, the terminal obtains second information from the core network equipment. The second information includes one or more of the following: the terminal's address, the terminal's port number, the target server's address, the target server's port number, a transport layer protocol, an application layer protocol, and a network layer protocol. The terminal determines the first information based on the second information. For example, the terminal parses the content of the second information and then determines the first information based on the parsed content. For instance, the first information may include all the content of the second information.

[0141] When a terminal obtains second information from a core network device, at least the following two examples are possible:

[0142] Example 1: The terminal sends a first request to the core network device, the first request being used to request second information. Correspondingly, the core network device receives the first request from the terminal, responds to the first request, determines the second information, and sends the second information to the terminal.

[0143] For example, the second information may include the terminal's address, the terminal's port number, the target server's address, the target server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol. Another example is that the second information may include the terminal's address, the terminal's port number, the target server's address, and the target server's port number. Yet another example is that the second information may include both the terminal's address and the target server's address.

[0144] Example 2: The second message can be a PDU session response.

[0145] Specifically, in the PDU session establishment process, the terminal can send a PDU session establishment request to the core network device. Correspondingly, the core network device receives the PDU session establishment request from the terminal, which is used to establish a PDU session in the network. The core network device then sends a PDU session response to the terminal, which includes the terminal's address.

[0146] In Example 2, the terminal can also receive third information from the core network device. The terminal can determine the first information based on the second and third information together. The first information can include all the contents of the second and third information. The third information may include, for example, the address of the target server and one or more of the following: the terminal's port number, the target server's port number, a transport layer protocol, an application layer protocol, and a network layer protocol. Optionally, the third information may be a response to a first request sent by the terminal to the core network device.

[0147] In Example 2, the order in which the second and third messages are received is not limited.

[0148] In one possible implementation, the core network device determines the target server, and then determines the target server's address and / or port number. For example, the core network device determines the target server after receiving the first request from the terminal.

[0149] When determining the target server, the core network equipment can have at least the following examples (1) to (4):

[0150] Example (1): The core network equipment uses the server provided by the terminal manufacturer or the terminal chip manufacturer as the target server.

[0151] For example, the server can send registration information to the core network device, including the server's manufacturer identifier. The core network device can then determine the server provided by the terminal's manufacturer and use that server as the target server; alternatively, the core network device can determine the server provided by the terminal chip's manufacturer and use that server as the target server. Here, the terminal chip is the chip installed within the terminal.

[0152] For example, a server from vendor A sends registration information 1 to the core network equipment, indicating that it can provide services to terminals from vendor A. A server from vendor B can send registration information 2 to the core network equipment, indicating that it can provide services to terminals equipped with / installed with chips from vendor B. Based on this, the core network equipment can designate a server from vendor A as the target server for terminals from vendor A, or designate a server from vendor B as the target server for terminals equipped with chips from vendor B.

[0153] In one possible implementation, the core network device receives a first request from the terminal and identifies the terminal based on the first request. For example, the first request may carry information about the terminal.

[0154] For example, the terminal information includes one or more of the terminal's manufacturer identifier and the manufacturer identifier of the chip it uses.

[0155] For example, when the core network device receives a first request from terminal 1, which carries the vendor identifier of terminal 1, the core network device determines that terminal 1 is a terminal of vendor A based on the vendor identifier of terminal 1, and then determines that the server of vendor A will provide services to terminal 1. When the core network device receives a first request from terminal 2, which carries the vendor identifier of terminal 2 and the vendor identifier of the chip installed in the terminal, the core network device determines that terminal 2 is a terminal of vendor A and that the chip installed in the terminal is a chip of vendor B based on the vendor identifier of terminal 2 and the vendor identifier of the chip installed in the terminal, and then determines that the server of vendor A will provide services to terminal 2, or determines that the server of vendor B will provide services to terminal 2.

[0156] Example (2): The core network equipment determines the target server based on the terminal type.

[0157] For example, the server may send registration information to the core network equipment, the registration information including the types of terminals that the server can provide services for. For example, the terminal types may include one or more of mobile phones, wearable devices, and in-vehicle controllers.

[0158] In one possible implementation, the core network device receives a first request from the terminal and identifies the type of the terminal based on the first request. For example, the type of the terminal may be carried in the first request.

[0159] For example, a core network device receives a first request from terminal 1, which carries the type of terminal 1 (e.g., a mobile phone). After determining that terminal 1 is a mobile phone, the core network device can identify a server capable of providing services to the mobile phone as the target server. Alternatively, the core network device receives a first request from terminal 2, which carries the type of terminal 2 (e.g., a wearable device). After determining that terminal 2 is a wearable device, the core network device can identify a server capable of providing services to the wearable device as the target server. Or, the core network device receives a first request from terminal 3, which carries the type of terminal 3 (e.g., an in-vehicle controller). After determining that terminal 3 is an in-vehicle controller, the core network device can identify a server capable of providing services to the in-vehicle controller as the target server.

[0160] Example (3) The core network device selects the target server from multiple servers based on the resource information of multiple servers.

[0161] For example, the multiple servers are provided by the terminal manufacturer and / or the terminal chip manufacturer. For instance, the terminal is manufactured by manufacturer A, and the multiple servers are provided by manufacturer A; or, the terminal is manufactured by manufacturer A, the chip installed in the terminal is manufactured by manufacturer B, and the multiple servers are provided by manufacturer A; or, the multiple servers are provided by manufacturer B; or, some of the multiple servers are provided by manufacturer A, while other servers are provided by manufacturer B.

[0162] The method by which the core network equipment determines multiple servers can be found in the description in example (1) above.

[0163] For example, the multiple servers are servers that provide services for this type of terminal. For instance, if the terminal is a mobile phone, the multiple servers are servers that provide services for the mobile phone; if the terminal is a wearable device, the multiple servers are servers that provide services for the wearable device; if the terminal is an in-vehicle controller, the multiple servers are servers that provide services for the in-vehicle controller, and so on.

[0164] The method by which the core network equipment determines multiple servers can be found in the description in Example (2) above.

[0165] In one possible implementation, the core network device selects a target server from multiple servers based on resource information from these servers. Specifically, it selects a server that meets the resource requirements, where resource requirements refer to the resources needed to perform the computing task. These resource requirements are included in the first request. The resource requirements include computing resource requirements and / or storage resource requirements. Computing resource requirements include one or more of the following: central processing unit (CPU) resource requirements and graphics processing unit (GPU) resource requirements. Storage resource requirements include memory resource requirements.

[0166] For example, the core network device may also send a resource allocation request to the selected target server. The resource allocation request is used to request the target server to reserve / allocate resources for computing tasks. For example, the resource allocation request may include resource requirements.

[0167] Example (4): If the core network device determines that it has already selected a target server for the terminal before receiving the first request, then the target server can continue to be the target server corresponding to the first request.

[0168] For example, the terminal sends a first request corresponding to a first computing task to the core network device, and the core network device determines the target server for executing the first computing task based on the methods in Example (1), Example (2) or Example (3); subsequently, the terminal sends a first request corresponding to a second computing task to the core network device, and the core network device can continue to use the target server as the target server for executing the second computing task.

[0169] In one possible implementation, the core network equipment is either an AMF or an SMF.

[0170] Step 602: The terminal sends the first information to the RAN node.

[0171] For example, the terminal sends the first information to the RAN node via control.

[0172] For example, the first information can be encrypted by the terminal and then sent to the RAN node.

[0173] For example, before sending the first information to the RAN node, the terminal can receive a second request from the RAN node, which requests the first information. Accordingly, the terminal responds to the second request and sends the first information to the RAN node. The second request can also be considered as the RAN node requesting "permission for the terminal to send data to the server on its behalf," and can also be referred to as a proxy permission request.

[0174] For example, after receiving a second request from the RAN node, the terminal can send first information to the RAN node based on its local authorization configuration. For instance, the local authorization configuration might be user-configured. For example, the user might pre-configure "Allow the terminal to delegate computing tasks to the server" in the authorization interface, and the terminal sends first information to the RAN node after receiving the second request. Alternatively, after receiving the second request from the RAN node, the terminal might display an authorization interface, where the user has configured "Allow the terminal to delegate computing tasks to the server," and the terminal sends first information to the RAN node based on the configuration instructions detected in the authorization interface.

[0175] This application does not limit the order of steps 601 and 602. It is permissible for the terminal to determine the first information and send it to the RAN node after receiving the second request from the RAN node; that is, the second request is the trigger condition for the terminal to determine the first information. Alternatively, the terminal may determine the first information first, and then send it to the RAN node after receiving the second request from the RAN node.

[0176] Step 603: The RAN node sends data to the server based on the first information.

[0177] In other words, the first piece of information is used by the RAN node to send data to the server.

[0178] For example, the first information is used for data encapsulation.

[0179] For example, the RAN node encapsulates data based on the first information to obtain a data packet, and then sends the data packet to the server.

[0180] For example, the first information includes the address of the target server, which is encapsulated in the header of the data packet. For instance, the header of the data packet carries the address of the target server. Alternatively, the first information may include not only the address of the target server but also the address of the terminal, the port number of the terminal, and the port number of the target server. Furthermore, the address of the terminal, the port number of the terminal, and the port number of the target server are also encapsulated in the header of the data packet. That is, the header of the data packet also carries the address of the terminal, the port number of the terminal, and the port number of the target server.

[0181] As another example, the data packet includes data. For instance, the data is included in the data packet's payload. Or, the data is the data packet's payload.

[0182] For example, the protocol stack of a RAN node includes an application layer, a transport layer, and a network layer. In the first piece of information, the application layer protocol is HTTP, the transport layer protocol is TCP, and the network layer protocol is IP. When the RAN node encapsulates data according to the first piece of information to obtain data packets, specifically, the application layer encapsulates the data into HTTP messages based on HTTP; the transport layer then encapsulates the HTTP messages into TCP segments based on TCP, adding the source port number (i.e., the terminal's port number) and the destination port number (i.e., the target server's port number), for example, the source port number is 5000 and the destination port number is 80; the network layer then encapsulates the TCP segments into IP packets based on IP, adding the source IP address (i.e., the terminal's address) and the destination IP address (i.e., the target server's address).

[0183] For example, RAN nodes can also update the packet header, such as updating the checksum and length of the header.

[0184] The data in this application is explained below, taking into account the type of computational task:

[0185] In one possible example, the computational task is an AI computational task, such as an inference task (e.g., model training), an image recognition task, or a natural language processing task; or, the computational task is a data processing task, such as a log file parsing and statistics task, or a database query task; or, the computational task is an encryption and security task, such as a security authentication task or a cryptographic authentication task; or, the computational task is a multimedia processing task, such as a graphics rendering task, an audio encoding and decoding task, or a real-time audio and video filtering task (e.g., beauty filters, noise reduction).

[0186] When the computational task is an AI computational task, the data may include training data and / or a third model. The data is used by the server to train a first model (or, in other words, the data is used by the server to train a first model, for example, the first model is a model that meets the accuracy requirements after training). For example, the RAN node sends training data and a third model to the server, and the server trains the first model based on the training data and the third model. Another example is that the server locally contains training data, the RAN node sends the third model to the server, and the server trains the first model based on the training data and the third model. Yet another example is that the server has a pre-configured (or standardized) model structure locally, the RAN node sends training data to the server (without sending the third model), and the server trains the first model based on the training data and the model structure.

[0187] The third model, also known as the initial model, can be obtained by the RAN node based on the training data. The training data used by the RAN node to train the third model can be the same as the training data used by the server to train the first model.

[0188] For example, the server may also send the first model to the terminal, which receives the first model from the server. For instance, the server sends the first model to the UPF, and the UPF sends the first model to the terminal based on QoS flow. For example, after sending the first information to the RAN node, the terminal may send a model request to the server, requesting the first model. In response to the model request, the server, after training the first model, sends it to the terminal. For example, the terminal uses the first model.

[0189] For example, in beam management (BM) scenarios, the first model can be used to discover the strongest transmit / receive beam pair. For instance, during the use of the first model, the terminal measures the downlink synchronization signal block (SSB) and CSI-RS to obtain the channel's angle domain information. This angle domain information is then input into the first model to infer the top-k beams from multiple beams (e.g., the top k beams in the reference signal received power (RSRP) ranking, where k is a positive integer). The identifiers of the inferred top-k beams are then fed back to the RAN node.

[0190] For example, in positioning accuracy enhancement scenarios, the first model can be used to improve positioning accuracy. For instance, in an indoor scenario (which can be termed a heavy NLOS scenario), the terminal can identify line-of-sight (LOS) / non-line-of-sight (NLOS) based on the first model, and improve positioning accuracy with a smaller number of transmission reception point (TRP) antennas depending on the LOS or NLOS.

[0191] For example, in a CSI prediction scenario, a first model can be used to predict the CSI at future times. Exemplarily, the input to the first model is the CSI at multiple historical times, and the output of the first model is the CSI at the target time (i.e., the inferred CSI). The first model could, for example, be a multi-layer perceptron (MLP).

[0192] In addition, RAN nodes can also train a second model based on the training data (or, in other words, RAN nodes can also train a second model based on the training data; for example, the second model is a model that meets the accuracy requirements after the model training is completed).

[0193] For example, when the RAN node does not send the third model to the server, the training data used by the RAN node to train the second model can be the same as the training data sent by the RAN node to the server to train the first model.

[0194] When a RAN node sends a third model to the server, the training data used by the RAN node to train the second model, the training data used by the RAN node to train the third model, and the training data sent by the RAN node to the server to train the first model can all be the same.

[0195] In this model, the first model and the second model are paired. That is, the first model and the second model are paired / pairing models, or the first model and the second model are used together, or the first model and the second model are two-ended models.

[0196] For example, in a CSI compression scenario, the training data includes CSI. The first model is a CSI compression model applied to the terminal side, and the second model is a CSI decompression model applied to the RAN node side. The CSI compression model and the CSI decompression model are two paired models.

[0197] When the computing task is a multimedia processing task, the data may include data to be processed. The server can perform multimedia processing based on the data to be processed, such as generating rendered images, obtaining encoded audio, obtaining decoded audio, etc.

[0198] In the first possible implementation, the PDU session established by the terminal with the UPF includes one or more QoS flows, each with different guarantees for data transmission (e.g., latency and packet loss rate). When the terminal needs to send data, it needs to determine which QoS flow the data should be sent through based on the data's packet filter. For example, a packet filter includes a 5-tuple: source address, destination address, protocol number, source port number, and destination port number. Another example is where the source address is the terminal's address, the destination address is the target server's address, the source port number is the terminal's port number, and the destination port number is the target server's port number.

[0199] For example, in step 602 above, the terminal may also send the target QFI to the RAN node.

[0200] In this application, the terminal can determine the identifier QFI of the QoS flow used to send data packets, and can record the QoS flow determined by the terminal as the target QoS flow, and record the identifier of the target QoS flow as the target QFI.

[0201] In one possible example, before determining the target QoS flow or the target QFI, the terminal may establish a PDU session according to the URSP rules, or select a PDU session that conforms to the URSP rules based on the URSP rules and the already established PDU sessions. For details on the implementation, please refer to the description in "II. User Equipment Route Selection Policy (URSP)" above.

[0202] For example, a terminal can determine the target QFI based on one or more of the following: the target server's address, the target server's port number, the terminal's address, the terminal's port number, and the transport layer protocol (also known as the protocol number). For instance, the terminal determines the target QFI based on the target server's address and the terminal's address. Another example is that the terminal determines the target QFI based on the target server's address, the terminal's address, the target server's port number, and the terminal's port number. Yet another example is that the terminal determines the target QFI based on the target server's address, the terminal's address, and the transport layer protocol.

[0203] The terminal also sends the target QFI to the RAN node.

[0204] The target QFI and the first information can be contained in the same message. For example, this message can be encrypted by the terminal and then sent to the RAN node. Alternatively, the target QFI and the first information can be contained in two different messages. For instance, the terminal first sends the first information to the RAN node, and then sends the target QFI to the RAN node. For example, the target QFI can be encrypted by the terminal and then sent to the RAN node.

[0205] The target QFI indicates the target QoS flow, which is used by the RAN node to send data to the server. That is, in step 603, the RAN node can also determine the target QoS flow based on the target QFI, and then send data (or data packets) to the server through the target QoS flow. This helps to implement traffic billing in the core network.

[0206] In the second possible implementation, after receiving the first information, the RAN node can determine the target QFI based on the first information. For example, the RAN node obtains the target server's address, the target server's port number, the terminal's address, the terminal's port number, and one or more of the transport layer protocols from the first information, and determines the target QoS flow based on the obtained information. Then, the RAN node can send data to the server through the target QoS flow.

[0207] In one possible example, when the RAN node sends data to the server via a target QoS flow, specifically, the RAN node sends data to the UPF via the target QoS flow. The UPF receives the data from the RAN node via the target QoS flow and then sends the data to the server.

[0208] The terminal obtains CSI by measuring the received CSI-RS, which includes parameters such as signal strength, signal quality, delay, and Doppler frequency shift. The terminal then reports the measured CSI to the RAN node using an appropriate method. Based on the received CSI, the RAN node performs operations such as downlink resource scheduling, precoding design, and link adaptation.

[0209] Furthermore, the combination of AI and CSI compression is an important research direction in the field of wireless communication, aiming to optimize CSI compression and transmission using AI technology, thereby improving the efficiency and performance of communication systems. Specifically, the motivation for CSI compression mainly stems from the needs of improving spectrum efficiency, reducing signaling overhead, enhancing system performance, and meeting the demands of large-scale connectivity. Especially in the case of massive MIMO systems, where the number of RAN node antennas is large and the amount of CSI data is massive, AI compression technology can significantly reduce feedback overhead. In millimeter-wave communication scenarios, where millimeter-wave channels change rapidly, AI technology can quickly compress and feed back CSI, supporting real-time beamforming. In resource-constrained narrowband Internet of Things (NB-IoT) devices, AI technology can reduce the energy consumption and bandwidth requirements of CSI transmission.

[0210] like Figure 7 A schematic diagram of a CSI compression process is provided as an example in this application:

[0211] Step 701: The RAN node sends CSI-RS to the terminal.

[0212] Before step 701, the process may also include: the RAN node sending CSI measurement configuration information to the terminal.

[0213] The CSI measurement configuration information may include, but is not limited to, one or more of the following parameters: CSI-RS resource location, CSI report configuration, and measurement timing. The CSI report configuration information may include, for example, the reported content (such as channel quality indicator (CQI), precoding matrix indicator (PMI), rank indicator (RI), and CSI-RS resource indicator (CRI)) and the reporting type (such as periodic reporting or non-periodic triggered reporting). Measurement timing may include, for example, the measurement time window, measurement period, and time slot offset.

[0214] Step 702: The terminal generates the first CSI based on the CSI-RS.

[0215] Step 703: The terminal obtains compression information based on the first CSI and the CSI compression model.

[0216] Step 704: The terminal sends compression information to the RAN node.

[0217] Step 705: The RAN node obtains the second CSI based on the CSI decompression model and compression information.

[0218] For example, the first CSI is the actual CSI generated by the terminal measurement CSI-RS. The terminal can input the first CSI into the CSI compression model, and the CSI compression model outputs compressed information. The terminal sends the compressed information to the RAN node, and the RAN node inputs the compressed information into the CSI decompression model, and the CSI decompression model outputs decompressed information (that is, the second CSI, or inference CSI).

[0219] The CSI compression model can be the first model in the above embodiments, and the CSI decompression model can be the second model in the above embodiments.

[0220] like Figure 8This diagram illustrates the structure of a dual-end model. The first model (CSI compression model) includes an encoder and a quantizer, while the second model (CSI decompression model) includes a decoder and a dequantizer. The encoder's input is the first CSI obtained from the terminal's CSI-RS measurement. This first CSI can be represented as the original channel matrix, a pre-encoded channel matrix, or an eigenvector after channel decomposition, etc. The encoder's output is a floating-point vector carrying the compressed CSI. Furthermore, the quantizer maps this floating-point vector carrying the compressed CSI to a quantized bit sequence (possible quantization methods include scalar quantization, vector quantization, etc.). The decoder and dequantizer function in the opposite way to the encoder and quantizer; the CSI output by the decoder can be considered the second CSI (i.e., inference CSI). For example, the encoder or decoder can be implemented using AI models such as convolutional neural networks (CNNs) or transformers.

[0221] like Figure 9 This is a flowchart illustrating the communication method in the first CSI compression scenario provided in this application. The terminal entrusts the server to perform the training task of the CSI compression model. Specifically, the first model is the CSI compression model, and the second model is the CSI decompression model.

[0222] The first piece of information includes the terminal's address, the terminal's port number, the server's address, the target server's port number, the transport layer protocol, the application layer protocol, and the network layer protocol. This first piece of information can also be referred to as the packet header information or complete encapsulation information. The following example uses complete encapsulation information.

[0223] For an explanation of each parameter in the first information, please refer to the description in step 601.

[0224] Step 901: The terminal determines the complete encapsulation information. See step 601 for details.

[0225] Step 902: The terminal sends complete encapsulation information and the target QFI to the RAN node. See step 602 for details.

[0226] Sending the target QFI to the RAN node is an optional step.

[0227] Step 903: The RAN node encapsulates the training data (including CSI) and / or the third model based on the complete encapsulation information to obtain the data packet.

[0228] In one possible example, the RAN node encapsulates the training data and the third model based on complete encapsulation information. For example, the data packet payload includes the training data and the third model. The third model can be obtained by the RAN node based on the training data, which may be the same as the training data in the data packet. It is understood that the server's computing power is higher than that of the RAN node, and the accuracy of the CSI compression model trained by the server is higher than the accuracy of the third model trained by the RAN node. The data interaction between the RAN node, terminal, and server, as well as the functions of each device, are as follows: Figure 10 As shown in (A).

[0229] In another possible example, the server and RAN nodes train the model based on a predefined model structure; that is, the RAN nodes encapsulate the training data according to complete encapsulation information. For example, the data packet payload includes the training data without carrying a third model. The data interaction between the RAN nodes, terminals, and servers, as well as the functions of each device, are as follows: Figure 10 As shown in (B).

[0230] For example, the model structure can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation functions, or biases in activation functions), input parameters (e.g., the type and / or dimension of input parameters), or output parameters (e.g., the type and / or dimension of output parameters). The biases in the activation functions can also be referred to as the neural network biases. That is, when a RAN node sends a third model, it specifically refers to sending the parameters of the third model; when it predefines a model structure, it specifically refers to predefining the parameters of the model structure.

[0231] For detailed instructions, please refer to the description in step 603. CSI can be obtained from the terminal's CSI-RS measurement. CSI can include, but is not limited to, one or more of the following parameters: CQI, PMI, RI, and CSI-RS resource indication.

[0232] In step 904, the RAN node sends a data packet to the UPF according to the target QFI and the target QoS flow. The target QFI indicates the target QoS flow. See step 603 for a detailed explanation.

[0233] Step 905: UPF sends a data packet to the server.

[0234] Step 906: The server trains the CSI compression model based on the training data.

[0235] Specifically, when the data packet includes training data and a third model, the server obtains the training data and the third model from the data packet, and trains the CSI compressed model based on the training data and the third model. The accuracy of the CSI compressed model obtained by the server is higher than that of the third model.

[0236] When the data packet includes training data but not the third model, the server obtains the training data from the data packet and trains the CSI compressed model based on the training data and the predefined model structure.

[0237] Step 907: The RAN node trains the CSI decompression model based on the training data.

[0238] It is understandable that the training data used by the RAN node to train the CSI decompression model and the training data used by the server to train the CSI compression model are the same.

[0239] For explanations of the CSI compression and decompression models, please refer to [link to relevant documentation]. Figure 7 and Figure 8 The description in the relevant embodiments.

[0240] Step 908: The server sends the CSI compression model to the terminal.

[0241] like Figure 11 This is a flowchart illustrating the communication method in the second CSI compression scenario provided in this application. The terminal entrusts the server to perform the training task of the CSI compression model. Specifically, the first model is the CSI compression model, and the second model is the CSI decompression model.

[0242] Figure 11 and Figure 9 The difference lies in the fact that both RAN nodes and servers use a standardized protocol stack, meaning they use the same protocols at each layer. Therefore, the first information no longer needs to carry the transport layer protocol, application layer protocol, and network layer protocol. For example, when a RAN node generates a data packet, it uses HTTP as the application layer protocol, TCP as the transport layer protocol, and IP as the network layer protocol; similarly, when the server parses the data packet, it also uses HTTP as the application layer protocol, TCP as the transport layer protocol, and IP as the network layer protocol.

[0243] The first piece of information includes the terminal's address, the terminal's port number, the target server's address, and the target server's port number. Here, the terminal's address and the target server's address can be collectively referred to as network layer encapsulation information, and the terminal's port number and the target server's port number can be collectively referred to as transport layer encapsulation information, etc. The following example uses the scenario where the first piece of information includes both network layer and transport layer encapsulation information.

[0244] For an explanation of each parameter in the first information, please refer to the description in step 601.

[0245] Step 1101: The terminal determines the network layer encapsulation information, the transport layer encapsulation information, and the target QFI. See step 601 for details.

[0246] In step 1102, the terminal sends network layer encapsulation information, transport layer encapsulation information, and the target QFI to the RAN node. For details, please refer to the description in step 602. The steps of the terminal determining the target QFI and sending the target QFI to the RAN node are both optional.

[0247] In step 1103, the RAN node encapsulates the training data (including CSI) and / or the third model according to the encapsulation information of the network layer, the encapsulation information of the transport layer, and the standardized protocols of each layer, to obtain the data packet. For details, please refer to the description in step 603.

[0248] CSI can be obtained from terminal measurement of CSI-RS. CSI can include, but is not limited to, one or more of the following parameters: CQI, PMI, RI, and CSI-RS resource indication.

[0249] In one possible example, the RAN node encapsulates the training data and the third model based on the encapsulation information of the network layer, the encapsulation information of the transport layer, and the standardized protocols of each layer. For example, the payload of a data packet includes both the training data and the third model.

[0250] In another possible example, the server and RAN nodes train the model based on a predefined model structure. That is, the RAN nodes encapsulate the training data according to the encapsulation information of the network layer, the encapsulation information of the transport layer, and the standardized protocols of each layer. For example, the training data is included in the payload of the data packet without carrying a third model.

[0251] For details not described in this step, please refer to the description in step 903.

[0252] In step 1104, the RAN node sends data packets to the UPF according to the target QFI and the target QoS flow. The target QFI indicates the target QoS flow. See step 603 for a detailed explanation.

[0253] Step 1105: UPF sends a data packet to the server.

[0254] Step 1106: The server trains the CSI compression model based on the training data. See step 906 for a detailed implementation description.

[0255] Step 1107: The RAN node trains the CSI decompression model based on the training data.

[0256] It is understandable that the training data used by the RAN node to train the CSI decompression model and the training data used by the server to train the CSI compression model are the same.

[0257] For explanations of the CSI compression and decompression models, please refer to [link to relevant documentation]. Figure 7 and Figure 8 The description in the relevant embodiments.

[0258] Step 1108: The server sends the CSI compression model to the terminal.

[0259] like Figure 12 This is a flowchart illustrating the communication method in the third CSI compression scenario provided in this application. The terminal entrusts the server with the training task of the CSI compression model. Specifically, the first model is the CSI compression model, and the second model is the CSI decompression model.

[0260] Figure 12 and Figure 11 The difference lies in the fact that both the terminal and the server use standardized port numbers. Consequently, the initial information no longer needs to include the port numbers of the terminal and the target server. For example, the source port number (i.e., the terminal's port number) is specified as 58, and the destination port number (i.e., the target server's port number) is specified as 47.

[0261] The first piece of information includes the terminal's address and the target server's address. This first piece of information can also be referred to as network layer encapsulation information.

[0262] For an explanation of each parameter in the first information, please refer to the description in step 601.

[0263] Step 1201: The terminal determines the network layer encapsulation information and the target QFI. See step 601 for details.

[0264] Step 1202: The terminal sends the network layer encapsulation information and the target QFI to the RAN node. See step 602 for details. The steps of determining the target QFI and sending the target QFI to the RAN node are optional.

[0265] In step 1203, the RAN node encapsulates training data (including CSI) and / or a third model based on the network layer encapsulation information, the standardized transport layer encapsulation information (i.e., the terminal's port number and the target server's port number), and the standardized protocols of each layer, to obtain the data packet. See step 603 for details. The CSI can be obtained by the terminal measuring CSI-RS. CSI can include, but is not limited to, one or more of the following parameters: CQI, PMI, RI, and CSI-RS resource indication.

[0266] In one possible example, the RAN node encapsulates the training data and the third model based on the encapsulation information of the network layer, the standardized encapsulation information of the transport layer, and the standardized protocols of each layer. For example, the payload of a data packet includes the training data and the third model.

[0267] In another possible example, the server and RAN nodes train the model based on a predefined model structure. That is, the RAN nodes encapsulate the training data according to the encapsulation information of the network layer, the standardized encapsulation information of the transport layer, and the standardized protocols of each layer. For example, the training data is included in the payload of the data packet without carrying a third model.

[0268] For details not described in this step, please refer to the description in step 903.

[0269] In step 1204, the RAN node sends data packets to the UPF according to the target QFI and the target QoS flow. The target QFI indicates the target QoS flow. See step 603 for a detailed explanation.

[0270] Step 1205: UPF sends a data packet to the server.

[0271] Step 1206: The server trains the CSI compression model based on the training data. See step 906 for a detailed implementation description.

[0272] Step 1207: The RAN node trains the CSI decompression model based on the training data.

[0273] It is understandable that the training data used by the RAN node to train the CSI decompression model and the training data used by the server to train the CSI compression model are the same.

[0274] For explanations of the CSI compression and decompression models, please refer to [link to relevant documentation]. Figure 7 and Figure 8 The description in the relevant embodiments.

[0275] Step 1208: The server sends the CSI compression model to the terminal.

[0276] For example, in the above Figure 9 , Figure 11 , Figure 12 Before the capability negotiation process, the terminal and RAN node can determine, through the capability negotiation process, whether to use compressed CSI when transmitting CSI over the air interface. After the capability negotiation process, the terminal can send first information (or first information and target QFI, as illustrated below using first information as an example) to the RAN node. Optionally, after the capability negotiation process, the RAN node sends a second request to the terminal, and the terminal responds to the second request by sending first information to the RAN node.

[0277] like Figure 13 The capability negotiation process diagram provided for this application may have at least the following methods A to D.

[0278] In Method A, the terminal sends first capability information to the RAN node. This first capability information indicates that the terminal has the capability to use the CSI compression model. Correspondingly, the RAN node receives the first capability information from the terminal, determines that the terminal has the capability to use the CSI compression model based on the first capability information, and then sends an instruction to the terminal to start model training.

[0279] In Method B, the terminal sends first capability information to the RAN node. Correspondingly, the RAN node receives the first capability information from the terminal, determines based on the first capability information that the terminal has the capability to use the CSI compression model, and then sends second capability information to the terminal. The second capability information can be used to instruct the RAN node that it has the capability to use the CSI decompression model.

[0280] In method C, the terminal sends first capability information to the RAN node. Correspondingly, the RAN node receives the first capability information from the terminal, determines that the terminal has the capability to use the CSI compression model based on the first capability information, and then sends second capability information to the terminal. The terminal receives the second capability information, determines that the RAN node has the capability to use the CSI decompression model, and then sends a request to the RAN node to use the CSI dual-end model. Correspondingly, the RAN node sends an instruction to the terminal to start model training.

[0281] In method D, the terminal sends a request to the RAN node to use the CSI two-ended model. After receiving the request from the terminal, the RAN node determines that the terminal has the capability to use the CSI compressed model and then sends an instruction to the terminal to start model training.

[0282] It is understood that the above methods A to D are all exemplary illustrations of capability negotiation. In other possible examples, the terminal and the RAN node may also exchange other messages, which will not be listed in this application.

[0283] The step numbers in the flowcharts described in this application are merely examples of the execution flow and do not constitute a restriction on the order in which the steps are executed. There is no strict execution order between steps in the embodiments that do not have temporal dependencies. Not all steps shown in the flowcharts are mandatory; some steps may be deleted or added as needed.

[0284] The above focuses on describing the differences between different embodiments. Apart from the differences, the embodiments can be referred to each other. In addition, different implementations or different examples in the same embodiment can also be referred to each other.

[0285] Figure 14 A possible exemplary block diagram of the communication device involved in an embodiment of this application is shown. For example... Figure 14 As shown, the communication device 1400 may include modules or units for implementing the methods described above.

[0286] In one possible design, the communication device 1400 includes a processing unit 1402 and a communication unit 1403. Optionally, the communication device 1400 may also include a storage unit 1401 for storing device program code and / or data.

[0287] The communication device 1400 can be a terminal-side device as described in the above embodiments, such as a terminal or a communication module in a terminal, or a circuit or chip in a terminal that is responsible for communication functions.

[0288] For example, in one embodiment, the processing unit 1402 is configured to: determine first information, the first information including the address of the target server; and the communication unit 1403 is configured to: send the first information to the RAN node, the first information being used by the RAN node to send data to the server.

[0289] In one possible design, when determining the first information, the processing unit 1402 is specifically used to: obtain second information from the core network device, the second information including one or more of the following: the address of the terminal, the port number of the terminal, the address of the target server, the port number of the target server, the transport layer protocol, the application layer protocol, and the network layer protocol; and determine the first information based on the second information.

[0290] In one possible design, when processing unit 1402 obtains the second information from the core network device, it is specifically configured to: control communication unit 1403 to send a first request to the core network device; and control communication unit 1403 to receive the second information from the core network device, the second information being a response to the first request. In another possible design, when processing unit 1402 obtains the second information from the core network device, it is specifically configured to: control communication unit 1403 to receive a PDU session response from the core network device, the PDU session response including the terminal's address.

[0291] In one possible design, before sending the first information to the RAN node, the communication unit 1403 is also used to: receive a second request from the RAN node, the second request being for requesting the first information.

[0292] In one possible design, the data is used to train a first model on the server, and the data is also used to train a second model on the RAN nodes. The first and second models are paired, and the data includes training data and / or a third model.

[0293] In one possible design, the communication unit 1403 is also used to receive a first model from the server.

[0294] In one possible design, the communication unit 1403 is also used to: send a QFI to the RAN node, the QFI indicating a QoS flow, the QoS flow being used by the RAN node to send data to the server via the QoS flow.

[0295] In one possible design, the processing unit 1402 is further configured to: determine the QFI based on one or more of the following: the address of the target server, the port number of the target server, the address of the terminal, the port number of the terminal, and the transport layer protocol.

[0296] In one possible design, when the communication device 1400 is a terminal or a communication module within a terminal, the function of the processing unit 1402 can be implemented by one or more processors. Specifically, the processor may include a modem chip, or a system-on-a-chip (SoC) chip or a SIP chip containing a modem core. The function of the communication unit 1403 can be implemented by transceiver circuitry.

[0297] In one possible design, when the communication device 1400 is a circuit or chip in a terminal responsible for communication functions, such as a modem chip or a system-on-a-chip (SoC) or SIP chip containing a modem core, the function of the processing unit 1402 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processor cores. The function of the communication unit 1403 can be implemented by the interface circuitry or data transceiver circuitry on the aforementioned chip.

[0298] In one possible design, when the communication device 1400 is a terminal or a communication and / or computing module within a terminal, the functionality of the processing unit 1402 can be implemented by one or more processors. Specifically, the processor may include a GPU, or a system-on-a-chip (SoC) or SIP chip containing a GPU. Alternatively, the processor may include an AI processor, or a SoC or SIP chip containing an AI processor. Or, the processor may include an ASIC, or a SoC or SIP chip containing an ASIC. The functionality of the communication unit 1403 can be implemented by transceiver circuitry.

[0299] In one possible design, when the communication device 1400 is a circuit or chip in a terminal responsible for communication and / or computing functions, such as a GPU or a system-on-a-chip (SoC) or SIP chip containing a GPU, an AI processor or a SoC or SIP chip containing an AI processor, or an ASIC or a SoC or SIP chip containing an ASIC, the function of the processing unit 1402 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processor cores. The function of the communication unit 1403 can be implemented by interface circuitry or data transceiver circuitry on the aforementioned chip.

[0300] The communication device 1400 can be a RAN node-side device in the above embodiments, such as a communication module in the RAN node, or a circuit or chip in the RAN node responsible for communication functions.

[0301] For example, in one embodiment, the communication unit 1403 is configured to: receive first information from the terminal, the first information including the address of the target server; and the processing unit 1402 is configured to: control the communication unit 1403 to send data to the server according to the first information.

[0302] In one possible design, when the processing unit 1402 sends data to the server based on the first information, it is specifically used to: encapsulate the data based on the first information to obtain a data packet; and control the communication unit 1403 to send the data packet to the server.

[0303] In one possible design, before receiving the first information from the terminal, the communication unit 1403 is also used to: send a second request to the terminal, the second request being used to request the first information.

[0304] In one possible design, the communication unit 1403 is also used to: receive a QFI from the terminal, wherein the QFI indicates a QoS flow; and when sending data to the server, the communication unit 1403 is specifically used to: send data to the server via a QoS flow.

[0305] The communication device 1400 can be a server-side device as described in the above embodiments, such as a communication module in a server, or a circuit or chip in a server responsible for communication functions. For example, in one embodiment, the communication unit 1403 is used to: receive data from a RAN node, the data being encapsulated by first information, the first information including the address of the target server; the processing unit 1402 is used to: train a first model based on the data; and the communication unit 1403 is also used to: send the first model to a terminal.

[0306] It is understood that the division of units in the above-described device is merely a logical functional division. One function can correspond to one functional unit, or two or more functions can be integrated into one functional unit. In actual implementation, all or some units can be integrated onto a single physical entity, or distributed across different physical entities. Furthermore, the aforementioned functional units can be implemented in hardware, software, or a combination of both. Whether a function is executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for specific applications, but such implementations should not be considered beyond the scope of this application.

[0307] In one example, the functional unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as: one or more application-specific integrated circuits (ASICs), or one or more central processing units (CPUs), one or more microcontroller units (MCUs), one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0308] In one example, storage unit 1401 may include random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory and / or registers, etc.

[0309] See Figure 15 This is a schematic diagram of the structure of a terminal 1500 provided in an embodiment of this application. The terminal 1500 can correspond to... Figures 1 to 3 The terminal shown is used to implement the operations of the terminal in the above embodiments.

[0310] like Figure 15As shown, the terminal includes: one or more antennas 1510, a radio frequency processing system 1520, and a processor system 1530.

[0311] In the downlink or sidelink direction, the RF processing system 1520 receives RF signals through the antenna 1510 and sends the RF-processed signals to the processor system 1530 for further processing. In the uplink or sidelink direction, the processor system 1530 processes the terminal-side information and sends it to the RF processing system 1520, which then processes the signal and transmits it through the antenna 1510.

[0312] In one example, the radio frequency (RF) processing system 1520 serves as the communication interface for external communication of the terminal and may include an RF front end (RFFE) 1521 and an RF transceiver 1522. The RFFE 1521 is primarily used for one or more processing operations, such as shaping, passband selection, or gain adjustment, on the RF signals received by the antenna or those to be transmitted through the antenna. It may include one or more components such as RF switches, duplexers, filters, power amplifiers, antenna tuners, and low-noise amplifiers. The RFFE 1521 can be a circuit system composed of multiple discrete components or integrated into one or more chips. The RF transceiver 1522 processes the RF signals received by the RFFE into baseband / IF signals for further processing by the processor system 1530, and processes the baseband / IF signals provided by the processor system 1530 into RF signals for transmission to the RFFE 1521. The baseband / IF signals transmitted between the RF transceiver 1522 and the processor system 1530 can be digital or analog signals. The radio frequency transceiver 1522 can be implemented by one or more chips, which are commonly referred to as radio frequency chips (RFICs).

[0313] In one example, processor system 1530 may include one or more processors for processing signals and executing one or more communication protocols. Optionally, processor system 1530 may also include memory 1536. In one example, the one or more processors include at least one baseband processor 1531 (also known as a modem processor). Memory 1536 is used to store data and / or computer program instructions. Optionally, processor system 1530 may also include one or more application processors 1532 for implementing processing of the terminal operating system and application layer. Application processor 1532 may include, for example, a GPU, AI processor, or ASIC. Optionally, processor system 1530 may also include one or more of a voice subsystem 1533, a multimedia subsystem 1534, or an interface circuit 1535. The voice subsystem 1533 is used to process voice signals, the multimedia subsystem 1534 is used to handle multimedia-related operations, such as video encoding / decoding, image processing, etc., and the interface circuit 1535 is used to implement communication with other terminal components, such as a display 1540, an input device 1550, memory 1560, etc. The aforementioned components in the processor system 1530 can communicate with each other via a bus or communication interface circuit.

[0314] In one example, the processor system 1530 can be packaged as a single processor chip, such as a SoC chip or a SIP chip. In another example, the processor system 1530 can be a system composed of multiple chips; for example, the baseband processor 1531 can be packaged as a single chip, or packaged with part or all of the circuitry of the radio frequency processing system into a single chip.

[0315] In one example, memory 1536 can be on-chip memory, i.e., located on the processor system 1530 chip. In another example, memory 1560 can be off-chip memory, i.e. located outside the processor system 1530 chip.

[0316] In one example, the baseband processor 1531 may include one or more processor cores 15311 and interface circuitry 15314. The one or more processor cores 15311 are used to process signals and execute one or more communication protocols. Optionally, the baseband processor 1531 may also include a memory 15315 for storing at least a portion of the corresponding computer program instructions and / or data. In one example, the one or more processor cores 15311 execute the computer program instructions stored in the memory 15315 to perform the relevant operations (such as determining first information, obtaining second information from core network equipment, etc.) in the above method embodiments.

[0317] In this disclosure, memory 15315 is used to store corresponding computer program instructions and / or data. This can mean that memory 15315 stores all corresponding computer program instructions and / or data for execution by processor core 15311; or it can mean that memory 15315 stores a portion of corresponding computer program instructions and / or data, including the computer program instructions and / or data currently required to be executed by processor core 15311. Memory 15315 can store different portions of computer program instructions and / or data multiple times for execution by processor core 15311 to implement the relevant operations in the above method embodiments. Interface circuit 15314 serves as a communication interface for communication with other components, such as transmitting signals with radio frequency processing system 1520, communicating with other subsystems and related components of processor system 1530 via a bus, such as transmitting data control signals with application processor 1532, and transmitting data or computer program instructions with memory 1536 or memory 1560. Optionally, in order to reduce the load on the processor core, a baseband signal processing circuit 15313 can be set to perform at least some baseband signal processing, including one or more of signal demodulation, modulation, encoding or decoding.

[0318] In one example, the communication device provided in this application may be a terminal 1500, a communication module including a processor system 1530 and a radio frequency system 1520, or a baseband processor 1531.

[0319] The processor, processor system, application processor, baseband processor, processor circuit, or processor core mentioned above can be collectively referred to as a processor. The processor may include one or more of the following: central processing unit (CPU), digital signal processor (DSP), microprocessor unit (MPU), microcontroller unit (MCU), graphics processing unit (GPU), field programmable gate array (FPGA), application-specific integrated circuit (ASIC), artificial intelligence processor (AI processor), or neural processing unit (NPU).

[0320] The aforementioned memory may include one or more of the following storage media: random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), phase-change memory (PCM), resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), cache, register, read-only memory (ROM), flash memory, erasable programmable read-only memory (EPROM), hard disk, etc. In one example, computer program instructions for executing the above embodiments may be stored on non-volatile memory, such as at least a portion of the aforementioned memory 1560 (e.g., one or more of ROM, flash memory, EPROM, or hard disk). When executed on the terminal, the corresponding computer program instructions may be partially or wholly loaded onto a memory with a faster transfer speed than the processor, such as at least a portion of memory 1536 and / or memory 15315 (e.g., one or more of RAM, SRAM, DRAM, PCM, RERAM, MRAM, FRAM, cache, or register), for the processor to execute in order to implement the steps in the above method embodiments.

[0321] In one example, the RF transceiver 1522 and the RF front-end 1521 can also be packaged in a single chip. In another example, the RF transceiver 1522, the RF front-end 1521, and the baseband processor 1531 can also be packaged in a single chip.

[0322] The terms "system" and "network" in this application embodiment are used interchangeably. "At least one" refers to one or more, and "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, "at least one of A, B, or C" includes A, B, C, AB, AC, BC, or ABC; "at least one of A, B, and C" can also be understood as including A, B, C, AB, AC, BC, or ABC. Furthermore, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in this application embodiment are used to distinguish multiple objects and are not used to limit the order, sequence, priority, or importance of multiple objects.

[0323] In this application, "sending information" can be understood as one device sending information to another device, or it can also be understood as one logical module within a device sending information to another logical module. For example, "RAN node sending information" can be understood as the RAN node sending information to another device (such as a terminal), or it can be understood as logical module 1 in the RAN node sending information to logical module 2 in the RAN node.

[0324] In this application, "receiving information" can be understood as one device receiving information from another device, or it can also be understood as a logical module within a device receiving information from another logical module. For example, "RAN node receiving information" can be understood as the RAN node receiving information from another device (such as a terminal), or it can be understood as logical module 1 in the RAN node receiving information from logical module 2 in the RAN node.

[0325] In this application, phrases such as "sending information to... (e.g., a terminal)" or related illustrations in the accompanying drawings can be understood as indicating that the destination of the information is a terminal. This can include sending information directly or indirectly to a terminal. Similarly, phrases such as "receiving information from... (e.g., a terminal)," "receiving information from... (e.g., a terminal)," or "receiving information sent by (e.g., a terminal)," or related illustrations in the accompanying drawings, can be understood as indicating that the source of the information is a terminal. This can include receiving information directly or indirectly from a terminal. Information may undergo necessary processing between the source and destination, such as format changes, but the destination can understand the valid information from the source. Similar expressions in this application can be interpreted similarly and will not be elaborated further here.

[0326] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0327] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0328] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0329] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

Claims

1. A communication method, characterized in that, Applied to the terminal side, including: Determine the first piece of information, which includes the server's address; The first information is sent to the Radio Access Network (RAN) node, and the first information is used by the RAN node to send data to the server.

2. The method as described in claim 1, characterized in that, The first information is used for encapsulating the data.

3. The method as described in claim 1 or 2, characterized in that, The server's address is encapsulated in the header of the data packet, and the data packet includes the data.

4. The method according to any one of claims 1-3, characterized in that, The first information also includes one or more of the following: the address of the terminal, the port number of the terminal, the port number of the server, the transport layer protocol, the application layer protocol, and the network layer protocol.

5. The method according to any one of claims 1-4, characterized in that, The determination of the first information includes: Obtain second information from the core network equipment, the second information including one or more of the following: the address of the terminal, the port number of the terminal, the address of the server, the port number of the server, the transport layer protocol, the application layer protocol, and the network layer protocol; The first information is determined based on the second information.

6. The method as described in claim 5, characterized in that, The process of obtaining the second information from the core network equipment includes: Send a first request to the core network device; The second information is received from the core network device, and the second information is in response to the first request.

7. The method as described in claim 5, characterized in that, The process of obtaining the second information from the core network equipment includes: The terminal receives a Protocol Data Unit (PDU) session response from the core network device, the PDU session response including the address of the terminal.

8. The method according to any one of claims 1-7, characterized in that, Before sending the first information to the RAN node, the method further includes: A second request is received from the RAN node, the second request being used to request the first information.

9. The method according to any one of claims 1-8, characterized in that, The data is used by the server to train a first model, and the data is also used by the RAN node to train a second model, wherein the first model and the second model are paired. The data includes training data and / or a third model.

10. The method as described in claim 9, characterized in that, The training data includes channel state information. The first model is a compression model of channel state information applied to the terminal side, and the second model is a decompression model of channel state information applied to the RAN node side.

11. The method as described in claim 9 or 10, characterized in that, Also includes: Receive the first model from the server.

12. The method according to any one of claims 1-11, characterized in that, Also includes: The RAN node sends a Quality of Service Flow Identifier (QFI), which indicates a Quality of Service Flow. This Quality of Service Flow is used by the RAN node to send the data to the server.

13. The method as described in claim 12, characterized in that, Also includes: The QFI is determined based on one or more of the following: the server address, the server port number, the terminal address, the terminal port number, and the transport layer protocol.

14. A communication method, characterized in that, Applied to the RAN node side of the radio access network, including: Receive first information from the terminal, the first information including the address of the server; Based on the first information, data is sent to the server.

15. The method as described in claim 14, characterized in that, Sending data to the server based on the first information includes: Based on the first information, the data is encapsulated to obtain a data packet; The data packet is sent to the server.

16. The method as described in claim 14 or 15, characterized in that, The server's address is encapsulated in the header of the data packet, and the data packet includes the data.

17. The method as described in claim 16, characterized in that, The first information also includes one or more of the following: the address of the terminal, the port number of the terminal, the port number of the server, the transport layer protocol, the application layer protocol, and the network layer protocol.

18. The method according to any one of claims 14-17, characterized in that, Before receiving the first information from the terminal, the method further includes: A second request is sent to the terminal, the second request being used to request the first information.

19. The method according to any one of claims 14-18, characterized in that, The data is used by the server to train a first model, and the data is also used by the RAN node to train a second model, wherein the first model and the second model are paired. The data includes training data and / or a third model.

20. The method as described in claim 18 or 19, characterized in that, The training data includes channel state information. The first model is a compression model of channel state information applied to the terminal side, and the second model is a decompression model of channel state information applied to the RAN node side.

21. The method according to any one of claims 14-20, characterized in that, Also includes: Receive a Quality of Service Flow Identifier (QFI) from the terminal, wherein the QFI indicates the Quality of Service Flow; Sending data to the server includes: The data is sent to the server via the service quality stream.

22. A communication method, characterized in that, Applied to the server side, including: Receive data from a Radio Access Network (RAN) node, the data being encapsulated in first information, the first information including the address of the server; Based on the data, train the first model; The first model is sent to the terminal.

23. The method as described in claim 22, characterized in that, The data includes training data and / or a third model.

24. A communication device, characterized in that, include: A module for performing the method as described in any one of claims 1 to 13, or, A module for performing the method as described in any one of claims 14 to 21, or, A module for performing the method as described in any one of claims 22 to 23.

25. A communication device, characterized in that, The device includes a processor and an interface circuit. The interface circuit is used to receive signals from other communication devices besides the communication device and transmit them to the processor, or to send signals from the processor to other communication devices besides the communication device. The processor is used to implement the method as described in any one of claims 1 to 13 through logic circuits or executable code instructions, or the processor is used to implement the method as described in any one of claims 14 to 21 through logic circuits or executable code instructions, or the processor is used to implement the method as described in any one of claims 22 to 23 through logic circuits or executable code instructions.

26. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions that, when executed by a communication device, implement the method as described in any one of claims 1 to 13, or the method as described in any one of claims 14 to 21, or the method as described in any one of claims 22 to 23.

27. A computer program product, characterized in that, The computer program product includes a computer program or instructions that, when executed by a communication device, implement the method as described in any one of claims 1 to 13, or implement the method as described in any one of claims 14 to 21, or implement the method as described in any one of claims 22 to 23.

Images