Communication method, apparatus, computer readable storage medium and program product

Abstract

Embodiments of this application provide a communication method, apparatus, computer-readable storage medium, and computer program product. In the method, a communication apparatus determines a set of nodes that will transmit data based on a first multiple access method. The node set includes a first node and a second node. The communication apparatus sends indication information to instruct the nodes in the node set to transmit data using the first multiple access method. The communication apparatus receives first superimposed data and processes the first superimposed data using a third AI model to obtain first data corresponding to the first AI model and second data corresponding to a second AI model. The data transmitted by the first node using the first multiple access method is the output data of the first AI model, and the data transmitted by the second node using the first multiple access method is the output data of the second AI model. This allows for the selection of suitable nodes for pairing, improving the spectrum efficiency and capacity of the communication system, thereby increasing the efficiency of multiple access.

Description

Technical Field

[0001] This application relates to the field of communications, and more specifically to a communication method, apparatus, computer-readable storage medium, and computer program product. Background Technology

[0002] Multiple access technology refers to techniques that allow multiple nodes (such as multiple terminals or multiple network devices) to access the same network or system for data transmission or interaction, enabling multiple nodes to simultaneously obtain communication services without signal interference between them. Common multiple access technologies include Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Code Division Multiple Access (CDMA), Orthogonal Frequency Division Multiple Access (OFDMA), and Non-Orthogonal Multiple Access (NOMA). However, there is still room for further optimization and improvement in multiple access technologies in the field of communication technology. Summary of the Invention

[0003] Embodiments of this application provide a communication method, apparatus, computer-readable storage medium, and computer program product that can select suitable nodes for pairing, improve the spectrum efficiency and capacity of the communication system, and thereby improve the efficiency of multiple access.

[0004] Firstly, embodiments of this application provide a communication method that can be applied to a receiving end for data transmission, such as a network device or terminal, a communication module / processing module in the network device or terminal, circuits or chips responsible for communication functions in the network device or terminal (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), and circuits or chips responsible for processing functions in the network device or terminal (such as a graphics processing unit (GPU), artificial intelligence (AI) processor, neural network processing unit (NPU), or application-specific integrated circuit (ASIC)). This method can also be implemented by logical nodes, logical modules, or software capable of implementing all or part of the receiving end's functions. For example, the method is applied to a network device. In this method, the network device determines a set of nodes that will transmit data based on a first multiple access method, the set of nodes including a first node and a second node. Further, the network device sends indication information, which instructs the nodes in the set of nodes to transmit data using the first multiple access method. The network device receives the first superimposed data and processes it using a third AI model to obtain first data corresponding to the first AI model and second data corresponding to the second AI model. The data transmitted by the first node using a first multiple access method is the output data of the first AI model, and the data transmitted by the second node using the first multiple access method is the output data of the second AI model. This method allows for the selection of suitable nodes for pairing, improving the spectrum efficiency and capacity of the communication system, thereby enhancing the efficiency of multiple access.

[0005] In one possible design, during the process of determining the node set, the network device determines that the first node and the second node belong to the node set based on the quality information of the first recovered data from the test data of the first node and the quality information of the second recovered data from the test data of the second node. The first and second recovered data are obtained by processing the second superimposed data using a third AI model. The second superimposed data includes data obtained by superimposing the third and fourth data. The third data is obtained by processing the test data of the first node using the first AI model, and the fourth data is obtained by processing the test data of the second node using the second AI model. In this way, nodes whose data recovery quality meets the requirements can access the network via a first multiple access method, improving the efficiency of multiple access.

[0006] In one possible design, the network device can also receive third data from the first node, or generate test data for the first node, wherein the generated test data for the first node is used to process the third data through the first AI model. Additionally or alternatively, the network device can also receive fourth data from the second node, or generate test data for the second node, wherein the generated test data for the second node is used to process the fourth data through the second AI model. This approach enables node pairing in scenarios where AI models are deployed or not, offering good adaptability and improving the efficiency of multi-access.

[0007] In one possible design, the network device can also send the first recovery data to the first node and receive quality information of the first recovery data from the first node. Alternatively, the network device can also determine the quality information of the first recovery data based on the first recovery data and the generated test data of the first node. In this way, node pairing can be performed in scenarios where nodes have deployed AI models or not, providing good adaptability and improving the efficiency of multi-access.

[0008] In one possible design, the network device can also send the second recovery data to the second node and receive quality information about the second recovery data from the second node. Alternatively, the network device can determine the quality information of the second recovery data based on the second recovery data and the generated test data of the second node. This approach allows for node pairing in scenarios where AI models are deployed or not, offering good adaptability and improving the efficiency of multi-access.

[0009] In one possible design, the network device may also send a first test command to the first node, instructing the first node to process its test data using a first AI model. Additionally or alternatively, the network device may also send a second test command to the second node, instructing the second node to process the first node's test data using a second AI model. In this way, testing and verification can be performed using test data selected by the nodes based on their own circumstances, which helps to pair suitable nodes and thus improves the efficiency of multiple access.

[0010] In one possible design, the network device can also receive instruction information from AI models of multiple nodes, including a first node and a second node. This approach is applicable to scenarios where nodes deploy one or more AI models, making it easier for the receiving end to determine the model currently being used by the node.

[0011] In one possible design, the network device can also receive input and output description information from the AI ​​models of multiple nodes. This allows nodes with matching input and output descriptions to be paired, increasing the success rate of user pairing and thus improving the efficiency of multiple access.

[0012] In one possible design, the network device can also receive processing capability information from multiple nodes, including a first node and a second node. The processing capability information of the first node is used to determine a first AI model, and the processing capability information of the second node is used to determine a second AI model. In this way, a node-specific AI model can be selected based on the node's processing capability, improving the compatibility between the node and the AI ​​model, enhancing the node's data processing efficiency, and thus improving the efficiency of multiple access.

[0013] In one possible design, the network device can also send a first AI model to the first node and a second AI model to the second node. In this way, AI models suitable for the processing capabilities of the nodes can be distributed to the corresponding nodes, improving the data processing efficiency of the nodes and thus improving the efficiency of multiple access.

[0014] In one possible design, the network device can further determine whether the first node and the second node belong to the node set, provided that the first node and the second node belong to the candidate node set. The nodes in the candidate node set are determined based on their data type information and the input-output description information of their AI models. This approach enables initial node pairing, eliminating nodes with low pairing probability, thereby improving node pairing efficiency and reducing the complexity of node pairing.

[0015] In one possible design, the network device can also query a historical information record table to see if any node in the node set belongs to that node set, and if such a historical record exists, determine the node set based on that record. This approach can improve node pairing efficiency and reduce the complexity of node pairing.

[0016] In one possible design, the network device can further determine that the first node and the second node belong to the node set if the historical record includes the following information: identification information of the first node, identification information of the second node, first indication information of the first AI model, second indication information of the second AI model, and third indication information of the third AI model used to process the second overlay data. The second overlay data includes data obtained by overlaying the third data and the fourth data, where the third data is obtained by processing the test data of the first node through the first AI model, and the fourth data is obtained by processing the test data of the second node through the second AI model. Alternatively, the network device can further determine that the first node and the second node belong to the node set if the historical record includes the following information: first indication information, second indication information, and third indication information. This approach facilitates the retrieval of historical records of node pairings, thereby improving node pairing efficiency and reducing the complexity of node pairing.

[0017] In one possible design, the network device can also generate records about which nodes in the node set belong to the node set, even without historical records, and store these records in a historical information record table. This allows for dynamic updating and maintenance of the historical information record table, improving node pairing efficiency and reducing the complexity of node pairing.

[0018] In one possible design, the network device can also receive data type information from multiple nodes, including a first node and a second node. This facilitates the pairing of appropriate nodes, thereby improving the efficiency of multiple access.

[0019] In one possible design, the first set of overlay data includes one or more tokens. This reduces communication overhead and improves transmission robustness.

[0020] Secondly, embodiments of this application provide a communication method that can be applied to a data transmission transmitter, such as a network device or terminal, a communication module / processing module in the network device or terminal, a circuit or chip responsible for communication functions in the network device or terminal (such as a modem chip, also known as a baseband chip, or a system-on-a-chip containing a modem core or a system-in-package (SIP) chip), or a circuit or chip responsible for processing functions in the network device or terminal (such as a graphics processor, AI processor, neural network processor, or application-specific integrated circuit). This method can also be implemented by a logical node, logical module, or software capable of implementing all or part of the receiver's functions. For example, the transmitter may include multiple nodes. Taking the application of this method to a terminal as an example, the number of terminals can be multiple, and a terminal can be an example of a node. In this method, a first terminal (an example of a first node) and a second terminal (an example of a second node) at the transmitter respectively receive indication information, which instructs nodes in the node set (including the first node and the second node) to transmit data using a first multiple access method. Then, the first terminal at the transmitter processes the data of the first terminal through a first AI model. The second terminal at the transmitter processes the data of the second terminal through a second AI model. The first terminal and the second terminal respectively send their respective AI model's processed output data. The output data of the first AI model sent by the first terminal and the output data of the second AI model sent by the second terminal will be superimposed during transmission, thus forming the first superimposed data. Using this method, only paired nodes send data using the first multiple access method based on the indication information, improving the spectrum efficiency and capacity of the communication system, thereby increasing the efficiency of multiple access.

[0021] Thirdly, 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. These modules, units, or means can be implemented by software, hardware, or a combination of software and hardware.

[0022] Fourthly, this application provides a communication device that has the functions of the second aspect above. For example, the communication device includes a module, unit, or means for performing the operations involved in the second aspect above. The module, unit, or means can be implemented by software, hardware, or a combination of software and hardware.

[0023] Fifthly, 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 stores part or all of the necessary computer program or instructions for implementing the functions described in the first aspect. The one or more processors can execute the computer program or instructions, causing the communication device to implement the methods in any possible design or implementation of the first aspect. The interface circuit is used to implement the communication functions within the communication device and / or the communication functions between the communication device and other devices or components.

[0024] In one possible design, the processor is used to communicate with other devices or components through the interface circuit.

[0025] In one possible design, the communication device may also include the memory.

[0026] The aforementioned communication device may be a receiving end for data transmission, such as a network device or terminal, or a communication and / or computing module in a network device or terminal, or a chip in a network device or 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 network device or terminal responsible for communication and / or computing functions (such as a GPU, AI processor, NPU, or ASIC), or a logical node or logical module that can implement all or part of the functions of a network device or terminal.

[0027] 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 stores part or all of the necessary computer program or instructions for implementing the functions described in the second aspect above. The one or more processors are executable to carry out the computer program or instructions, causing the communication device to implement the methods in any possible design or implementation of the second aspect above. The interface circuit is used to implement the communication functions within the communication device and / or the communication functions between the communication device and other devices or components.

[0028] In one possible design, the processor is used to communicate with other devices or components through the interface circuit.

[0029] In one possible design, the communication device may also include the memory.

[0030] The aforementioned communication device may be a transmitting end for data transmission, such as a network device or terminal, or a communication and / or computing module in a network device or terminal, or a chip in a network device or 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 network device or terminal responsible for communication and / or computing functions (such as a GPU, AI processor, NPU, or ASIC), or a logical node or logical module that can implement all or part of the functions of a network device or terminal.

[0031] In a seventh aspect, 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 second aspects described above.

[0032] Eighthly, 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 second aspects described above.

[0033] The beneficial effects of the third to eighth aspects mentioned above can be found in the descriptions of the first and second aspects. Attached Figure Description

[0034] Figure 1A A schematic diagram of a communication system according to some embodiments of this application is shown.

[0035] Figure 1B and Figure 1C A schematic diagram illustrating an example application framework of a communication system according to some embodiments of this application is shown.

[0036] Figure 1D The diagram illustrates a multiple access process applicable to some embodiments of this application.

[0037] Figure 2A – Figure 2D A schematic diagram of the communication flow of some embodiments of this application is shown.

[0038] Figure 3 and Figure 4 A schematic diagram of a user pairing process for multiple access is shown, representing some embodiments of this application.

[0039] Figure 5 The diagram illustrates the usage flow of the pairing history information table in some embodiments of this application.

[0040] Figure 6 This is a schematic diagram of the structure of an apparatus according to some embodiments of this application.

[0041] Figure 7 This is a schematic diagram of the terminal structure according to this application. Detailed Implementation

[0042] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that embodiments of this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0043] In the description of embodiments of this application, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0044] The embodiments of this application can be implemented according to any suitable communication protocol, including but not limited to cellular communication protocols such as fourth-generation (4G), fifth-generation (5G), and future communication protocols, and wireless local area network communication protocols such as IEEE 802.11. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems.

[0045] Figure 1A A schematic diagram of a communication system according to some embodiments of this application is shown. Figure 1A This is a schematic diagram illustrating one possible, non-limiting system. For example... Figure 1A As shown, the communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. RAN 100 includes at least one RAN node (e.g., ...). Figure 1A110a and 110b (collectively referred to as 110) and at least one terminal (such as Figure 1A RAN 100, denoted as RAN 120a-120j, is collectively referred to as RAN 120. RAN 100 may also include other RAN nodes, such as wireless relay equipment and / or wireless backhaul equipment. Figure 1A (Not shown in the image). Terminal 120 is connected to RAN node 110 wirelessly. RAN node 110 is connected to core network 200 wirelessly or via wired connection. The core network elements in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions.

[0046] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as 4G, 5G mobile communication systems, or future-oriented evolution systems. RAN 100 can also be an open access network, cloud radio access network (CRAN), virtualized radio access network (vRAN), artificial intelligence radio access network (AI RAN), or wireless fidelity (WiFi) system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0047] RAN node 110, sometimes also referred to as access network equipment, RAN entity, or access node, constitutes part of the communication system and is used to help terminals achieve wireless access. Multiple RAN nodes 110 in communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative, for example... Figure 1A Network element 120i can be a helicopter or a drone, and it can be configured as a mobile base station. For terminals 120j that access RAN 100 through network element 120i, network element 120i is a base station; however, for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes referred to as communication devices, for example... Figure 1A Network elements 110a and 110b can be understood as communication devices with base station functions, while network elements 120a-120j can be understood as communication devices with terminal functions.

[0048] In one possible scenario, a RAN node can be a base station, an evolved NodeB (eNodeB, eNB), 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, etc. A RAN node can also be a macro base station (such as...) Figure 1A 110a), micro base stations (also known as small stations) or indoor stations (such as...) Figure 1A The RAN node can be a relay station, relay node, or donor node (as described in section 110b), or a wireless controller in a CRAN scenario. Optionally, the RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of the RAN node in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node can also be equipped with communication modules, circuits, or chips that perform corresponding communication functions. The RAN node can 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 that can implement all or part of the access node functions, or a circuit or chip (such as a GPU, AI processor, NPU, or ASIC) responsible for communication and / or computing functions in the access node.

[0049] 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).

[0050] 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.

[0051] A terminal can be a device or module that accesses the aforementioned communication system and has corresponding communication functions. A terminal 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), V2X communication, MTC, IoT, virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, etc. A terminal can be a mobile phone, tablet computer, computer with wireless transceiver capabilities, wearable device, vehicle, drone, helicopter, airplane, ship, robot, robotic arm, smart home device, transportation vehicle with wireless communication capabilities, communication module, etc. The embodiments of this application do not limit the device form of the terminal. A terminal typically contains a communication module, circuit, or chip that performs corresponding communication functions. Furthermore, it can also contain modules, circuits, or chips (such as GPUs, AI processors, NPUs, or ASICs) that perform corresponding communication and / or computing functions. The terminal can also be configured with program instructions for performing corresponding communication and / or computing functions.

[0052] To support artificial intelligence (AI) technology in wireless networks, AI nodes may be introduced into the network. AI nodes can be deployed in one or more of the following locations within the communication system: access network nodes (RAN nodes), terminal devices, or core network elements, or they can be deployed independently, for example, in a location other than any of the above devices, such as a host in an over-the-top (OTT) system or a cloud server. AI nodes can communicate with other devices in the communication system, which may be one or more of the following: network devices, terminal devices, or core network elements. It is understood that this application does not limit the number of AI nodes. For example, when there are multiple AI nodes, they can be divided based on function, such as different AI nodes being responsible for different functions. It is also understood that AI nodes can be independent devices, integrated into the same device to implement different functions, or they can be network elements in hardware devices, software functions running on dedicated hardware, or virtualized functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of the AI ​​nodes. AI nodes can be AI network elements or AI modules.

[0053] Figure 1B A schematic diagram illustrating an example application framework of a communication system according to some embodiments of this application is shown. For example... Figure 1BAs shown, network elements in a communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These network element nodes, such as core network elements (or core network equipment), access network nodes (RAN nodes), terminals, or one or more devices in operations administration and maintenance (OAM), are equipped with one or more AI modules (for clarity, ...). Figure 1B (Only one is shown in the image). An access network node can be a single RAN node or can include multiple RAN nodes, such as CU and DU. CU and / or DU can also be configured with one or more AI modules. CU can also be split into CU-CP and CU-UP, with one or more AI modules configured in CU-CP and / or CU-UP.

[0054] The AI ​​modules deployed in different network elements can be the same or different. The models of the AI ​​modules can achieve different functions depending on the parameter configurations. The models of the AI ​​modules 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 function, or biases in the activation function), input parameters (e.g., the type and / or dimension of the input parameters), or output parameters (e.g., the type and / or dimension of the output parameters). The biases in the activation function can also be referred to as the biases of the neural network.

[0055] In one example, the neural network mentioned above can be a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), or a generative adversarial network (GAN).

[0056] Deep Neural Networks (DNNs) are artificial neural network architectures with multiple layers of nonlinear transformation units stacked in a hierarchical structure to form deep computational models. Compared to shallow neural networks, deep neural networks have more hidden layers, allowing the network model to capture more complex data structures and higher-level abstract features.

[0057] A CNN is a deep neural network with a convolutional structure. A CNN contains a feature extractor consisting of convolutional layers and subsampling layers. This feature extractor can be viewed as a filter, and the convolution process can be seen as performing convolution between a trainable filter and an input image or a convolutional feature map.

[0058] RNN is a type of recursive neural network that takes sequence data as input, recursively moves along the direction of sequence evolution, and connects all nodes (recurrent units) in a chain-like manner.

[0059] GAN is a deep learning model. It consists of a generator and a discriminator, and is trained through adversarial learning. Its purpose is to estimate the potential distribution of data samples and generate new data samples.

[0060] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0061] Figure 1C A schematic diagram illustrating an example application framework of a communication system according to some embodiments of this application is shown. For example... Figure 1C As shown, the communication system includes a RAN intelligent controller (RIC). For example, the RIC can be an AI module used to implement AI-related functions. RICs include near-real-time RICs (near-RT RICs) and non-real-time RICs (non-RT RICs). Non-real-time RICs primarily process non-real-time information, such as data that is not sensitive to latency, with latency in the order of seconds. Real-time RICs primarily process near-real-time information, such as data that is relatively sensitive to latency, with latency in the order of tens of milliseconds.

[0062] Near real-time (NRT) RICs are used for model training and inference. For example, they are used to train AI models (sometimes referred to as AI functions) and then use those models for inference. NRT RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, compute nodes, and / or RUs) and / or terminals. This information can be used as training data or inference data. NRT RICs can deliver inference results to RAN nodes and / or terminals. Inference results can be exchanged between CUs and DUs, and / or between DUs and RUs. For example, a NRT RIC delivers an inference result to a DU, which then sends it to an RU. Non-real-time (NRT) RICs are also used for model training and inference. For example, they are used to train AI models and then use those models for inference. Non-real-time (NRT) RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, compute nodes, and / or RUs) and / or terminals. This information can be used as training data or inference data, and inference results can be delivered to RAN nodes and / or terminals. Inference results can be exchanged between CU and DU, and / or between DU and RU. For example, a non-real-time RIC can submit its inference result to a DU, which then forwards it to a RU. Near-real-time and non-real-time RICs can also be configured as separate network elements. Furthermore, near-real-time and non-real-time RICs can be integrated into other devices. For instance, near-real-time RICs can be located in RAN nodes (e.g., CU, DU, compute nodes), while non-real-time RICs can be located in OAM, cloud servers, core network elements, or other network devices.

[0063] In this application, "sending information" can be understood as one device sending information to another device, or it can also be understood as one logic module within a device sending information to another logic module. For example, "sending device sending information" can be understood as the sending device sending information to another device (such as the receiving device), or it can be understood as logic module 1 in the sending device sending information to logic module 2 in the sending device.

[0064] 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, "receiving device receiving information" can be understood as the receiving device receiving information from another device (such as the sending device), or it can be understood as logical module 1 in the receiving device receiving information from logical module 2 in the receiving device.

[0065] In this application, phrases such as "sending information to... (e.g., a network device)" or related illustrations in the accompanying drawings can be understood as the destination of the information being 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 the source of the information being 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.

[0066] The communication system 10 in this application embodiment includes, but is not limited to: narrow band-internet of things (NB-IoT), long term evolution (LTE), long term evolution advanced (LTE-A), and 5G mobile communication system, etc.

[0067] This application's embodiments can be applied to 5G or future communication systems, satellite communication, and short-range wireless communication systems. Wireless communication systems can consist of cells, each containing a network-side device (such as a base station), which provides services to multiple terminals. Wireless communication systems can also perform point-to-point communication, such as communication between multiple terminals. A network-side device refers to a device deployed in a wireless access network that provides wireless communication functions for a mobile station (MS), such as a base station. In systems employing different wireless access technologies, the names of devices with base station functions may differ. The MS involved in this application's embodiments can include various handheld devices, vehicle-mounted devices, wearable devices, computing devices, or other processing devices connected to a wireless modem with wireless communication functions. An MS can also be called a terminal, and can also be a subscriber unit, cellular phone, smartphone, wireless data card, personal digital assistant (PDA) computer, tablet computer, wireless modem, handset, laptop computer, MTC terminal, etc.

[0068] The network device in this application embodiment can be the aforementioned RAN node (or access network device, radio access network device). The network device can also be a core network element or a device connected to the core network. Core network elements can be used to provide core network services to access network devices and terminal devices. Core network elements can correspond to different devices in different systems. For example, in 4G, core network elements can correspond to a mobility management entity (MME) and / or a serving gateway (S-GW). In 5G, core network elements can correspond to access and mobility management functions (AMF), session management functions (SMF), or user plane functions (UPF).

[0069] Network devices and terminal devices can be deployed on the ground, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on airplanes, balloons, and satellites. This application does not limit the scenario in which the network devices and terminal devices are located.

[0070] The communication method and apparatus of the present application embodiments will be further described below with reference to the accompanying drawings. It is understood that the present application uses a sending end (including at least a first node and a second node) and a receiving end as examples of the execution entities in the interaction illustration, but the present application does not limit the execution entities in the interaction illustration. For example, the method executed by the receiving end in the present application can also be implemented by a module (e.g., a circuit, chip, or chip system) in the receiving end, or by a logic node, logic module, or software that can implement all or part of the receiving end's functions; the method executed by the sending end in the present application can also be implemented by a communication / processing module in the sending end or by a circuit or chip responsible for communication / processing functions in the terminal (such as a modem chip (also known as a baseband chip), or a SoC chip / SIP chip containing a modem core, or a GPU / NPU / AI processor / ASIC), or by a logic node, logic module, or software that can implement all or part of the sending end's functions.

[0071] Figure 1DThis document illustrates a multiple access process applicable to some embodiments of this application. This multiple access process can involve multiple nodes, with a first node and a second node exemplarily shown. The first AI model and the second AI model are local AI models of the first and second nodes, respectively, for example, denoted as W1 and W2. The local AI models deployed on different nodes of the sending end are usually different and can be called personalized local models or personalized models. The number of personalized models on the same node can be one or more. The personalized models of a node can be distinguished by model IDs. A third AI model, such as a common model (denoted as W), is deployed on the receiving end side. Taking uplink multiple access as an example, the first node is, for example, UE1, the second node is, for example, UE2, and the receiving end side is, for example, the network (NW) side.

[0072] The first node processes data X using a local personalized model W1, and the second node processes data Y using a local personalized model W2. Data X processed by personalized model W1 becomes data X'. Data Y processed by personalized model W2 becomes data Y'. The first and second nodes respectively send their respective results (i.e., data X' and data Y') processed by their local personalized models to the sending end. The input data to a node's local personalized model is the data that the node needs to upload to the sending end, such as X or Y as mentioned above. Its output is a high-dimensional vector representation of the corresponding data projected into a high-dimensional space, such as data X' or Y'. The vector representations of data from different nodes after corresponding personalized processing will be orthogonal in the high-dimensional space. In some examples, multiple nodes can reuse the same time-frequency resources to send the output data of their respective local personalized models to the receiving end.

[0073] In some examples, data can also be referred to as signals. The receiver receives signals from nodes, which are data X' and Y' superimposed together, and can be called superimposed data, for example, it can be represented as X'+Y'. Superimposed data is, for example, the superposition result of high-dimensional orthogonal high-dimensional vector representations of data from multiple nodes. Superimposed data can also be understood as data formed by the waveform aliasing of multiple data (such as X' and Y' mentioned above) transmitted on the same resource during air interface transmission. The receiver receives the superimposed data and inputs the received superimposed data into a common model W to complete signal detection. The output of the common model W is the estimated value of the data reported by the transmitting node to the receiver, such as data X'" and Y'". Signal detection is the process by which the receiver obtains data X'" and Y'" from the superimposed data through the common model W.

[0074] Sending nodes, such as the first and second nodes, can obtain local AI models by acquiring data and training them, or they can configure local AI models via the network. The first and second AI models may be referred to as node-specific AI models, node-specific AI models, node-personalized AI models, node-local AI models, or node-local personalized models in some examples. For example, in some examples where the node is a UE, according to the above naming rules, the first AI model deployed on UE1 can be called the UE1-specific AI model, the second AI model deployed on UE2 can be called the UE2-specific AI model, and so on. The third AI model in the above process may be referred to as the common model in some examples. The common model is used to extract data sent by different nodes from multi-access data.

[0075] In some examples, the AI ​​models deployed at the transmitting and receiving ends in this application embodiment can refer to the low-rank adaptation of large language models (LoRA) technology to achieve source encoding and decoding for multi-access communication transmission. LoRA is a low-rank fine-tuning method. Its core idea is to freeze the weights of the pre-trained model and then inject the trainable low-rank decomposition matrix into the layers of the transformer architecture to reduce the number of trainable parameters for downstream tasks. By using different LoRA parameters, the same pre-trained model can be fine-tuned for different downstream tasks. In such examples, the multiple access in this application embodiment can be called LoRA multiple access, and the AI ​​model can be called a low-rank model accordingly. For example, in an uplink multiple access example where the transmitting end is a UE and the receiving end is a NW, a personalized low-rank model (i.e., different LoRA parameters) is deployed on the UE side. The same common model can be combined with different LoRA parameters to complete different tasks, thereby enabling multiple UEs to achieve uplink multiple access through different LoRA parameters. Alternatively, the above-mentioned LoRA multiple access can also be applied to downlink multiple access or sidelink communication scenarios. In uplink multiple access (MLA) processes, the transmitting end is the terminal side (e.g., UE side) and the receiving end is the NW (e.g., base station) side. In downlink multiple access (MLA) processes, the transmitting end is the NW (e.g., base station) side and the receiving end is the terminal side (e.g., UE side). In sidelink communication, both the transmitting end and the receiving end are terminal sides (e.g., UE). Although the multiple access processes illustrated in the following examples are described using uplink multiple access, they can also be applied to downlink multiple access or sidelink communication.

[0076] In multi-user multiple-input multiple-output (MU-MIMO) systems, user pairing can improve the system's spectral efficiency and uplink / downlink capacity. When multi-user MIMO is enabled, the base station schedules each UE through a pairing strategy and selects a suitable UE for pairing and transmission.

[0077] Currently, there is no user pairing scheme designed based on the characteristics of multiple access. Therefore, some embodiments of this application propose a user pairing scheme for multiple access, designing pairing principles based on the principles of multiple access. After pairing is completed, users in the same group access the system using a multiple access method (e.g., the first multiple access method), while users in different groups can access the system orthogonally. The multiple access scheme in the embodiments of this application can support multiple access that is orthogonal to the model domain. In some examples, the user pairing process is the process of determining the node set.

[0078] Figure 2A The diagram illustrates the communication flow of some embodiments of this application. For example... Figure 2A As shown, process 201 involves a receiver 210 and a sender. The sender may include one or more nodes. Figure 2A Exemplary examples show a first node 220 and a second node 230 at the transmitting end. In different examples, the data transmission from the transmitting node to the receiving end 210 can be uplink data transmission, downlink data transmission, or sidelink data transmission. In the uplink data transmission scenario, an example of the transmitting node can be a terminal (e.g., a UE), and an example of the receiving end 210 can be a network device (e.g., a base station). In the downlink data transmission scenario, an example of the transmitting node can be a network device (e.g., a base station), and an example of the receiving end 210 can be a terminal (e.g., a UE). In the sidelink data transmission scenario, examples of both the transmitting end and the receiving end 210 can be terminals (e.g., a UE or other terminals). The communication flow between the receiving end 210 and the transmitting end is illustrated below using examples of the receiving end 210 communicating with the first node 220 and the second node 230, respectively. It is understood that the implementation of flow 201 is not limited to the examples shown, and the transmitting node is not limited to the nodes shown.

[0079] At step 202, receiver 210 determines the set of nodes that will transmit data based on the first multiple access method. The node set may include multiple nodes; the following example illustrates the scenario where the node set includes the first node 220 and the second node 230. In some examples, the first multiple access method may be a model domain orthogonal multiple access method, such as the LoRA multiple access method.

[0080] After determining the node set, the receiving end 210 can send indication information to instruct the nodes in the node set to send data using a first multiple access method. For example, at 204a, the receiving end 210 can send indication information 205 (or fourth indication information 205) to the first node 220, and the first node 220 receives the fourth indication information 205. At 204b, the receiving end 210 can send indication information 215 (or fifth indication information 215) to the second node 230, and the second node 230 receives the fifth indication information 215.

[0081] At 206, the first node 220 and the second node 230 send data 225 and 235 using a first multiple access method. Data 225 sent by the first node using the first multiple access method is the output data of the first AI model, and data 235 sent by the second node using the first multiple access method is the output data of the second AI model. The first node 220 and the second node 230 can use the same resource to send data 225 and 235. Because data 225 and 235 are sent through the same resource, they will be superimposed during transmission, resulting in superimposed data (referred to as the first superimposed data) received by the receiving end 210. In some examples, the first superimposed data may include one or more tokens. For example, the sending end and the receiving end 210 can communicate using tokens. A token can be understood as a basic unit of text, data, or features in the processing of a model, such as an AI model. For example, text, data, or features in the processing of an AI model can be represented as tokens for processing and transmission. For example, the input and output of an AI model may contain a series of numbers or vectors, which can be called tokens, or these numbers or vectors can be represented as tokens. Alternatively, a token may also be called a mark, term, or token, etc. The naming of the token is not limited in this application embodiment.

[0082] At step 208, the receiving end 210 processes the first overlay data using the third AI model to obtain first data corresponding to the first AI model and second data corresponding to the second AI model. The first data corresponding to the first AI model can be recovered data from the data of the first node, and the second data corresponding to the second AI model can be recovered data from the data of the second node. This process can be referenced... Figure 1D To understand this, data X can represent the data of the first node, i.e., the input data of the first AI model. Data Y can represent the data of the second node, i.e., the input data of the second AI model. Superimposed data X'+Y' can represent the first superimposed data. Data X" can represent the first data corresponding to the first AI model. Data Y" can represent the second data corresponding to the second AI model.

[0083] In some examples, to determine the node set, the receiver 210 can determine that the first node and the second node belong to the node set based on the quality information of the first recovered data of the test data of the first node 220 and the quality information of the second recovered data of the test data of the second node 230. The first and second recovered data are obtained by processing the second superimposed data using a third AI model. The second superimposed data includes data obtained by superimposing the third and fourth data. The third data is obtained by processing the test data of the first node using the first AI model, and the fourth data is obtained by processing the test data of the second node using the second AI model. Similar to the example of the first superimposed data above, the second superimposed data may also include one or more tokens as an example.

[0084] In some examples, the first node 220 deploys a first AI model, and the second node 230 deploys a second AI model. The receiving end 210 can receive third data from the first node 220. For example, the receiving end 210 can send a first test instruction to the first node 220, instructing the first node 220 to process its test data using the first AI model. In such an example, after receiving the first test instruction, the first node 220 will select test data, process the selected test data using the first AI model to output third data, and then send the third data to the receiving end 210, allowing the receiving end 210 to receive the third data. In some examples, the receiving end 210 can receive fourth data from the second node. For example, the receiving end 210 can send a second test instruction to the second node 230, instructing the second node to process its test data using the second AI model. In this example, after receiving the second test instruction, the second node 230 selects test data and processes it using the second AI model to output fourth data. This fourth data is then sent to the receiving end 210, allowing the receiving end 210 to receive it. After receiving the second overlay data, the receiving end 210 processes it using the third AI model to obtain the first and second recovery data. This process can also be referenced... Figure 1D To understand this, data X, the input data of the first AI model, can represent the test data of the first node 220. Data Y, the input data of the second AI model, can represent the test data of the second node 230. Data X' can represent the third data, data Y' can represent the fourth data, and the superimposed data X'+Y' can represent the second superimposed data. Data X” can represent the first recovered data, and data Y” can represent the second recovered data.

[0085] In some examples, the first and second AI models are trained by the receiver 210, while the first node 220 and the second node 230 have not yet deployed their respective AI models. In such examples, the receiver 210 can generate test data for the first node 220, which is then processed by the first AI model to obtain third data. For example, the receiver 210 inputs the generated test data of the first node into the first AI model trained on the receiver 210 side to output the third data. In some examples, the receiver 210 can generate test data for the second node 230, which is then processed by the second AI model to obtain fourth data. For example, the receiver 210 inputs the generated test data of the second node 230 into the second AI model trained on the receiver 210 side to output the fourth data. In such examples, it can be understood that the simulation is performed on the receiver 210 side. Figure 1D The process is shown.

[0086] In the example described above, where the first node 220 deploys the first AI model and the second node 230 deploys the second AI model, the selection or generation of test data is performed at the sending node. After receiving the first recovery data, the receiving end 210 can send the first recovery data to the first node 220. The first node 220 determines the quality information of the first recovery data based on its local test data and the first recovery data, and then sends the quality information of the first recovery data to the receiving end 210, thereby receiving the quality information of the first recovery data from the first node 220. Alternatively, after receiving the second recovery data, the receiving end 210 can send the second recovery data to the second node 230. The second node 230 determines the quality information of the second recovery data based on its local test data and the second recovery data, and then sends the quality information of the second recovery data to the receiving end 210, thereby receiving the quality information of the second recovery data from the second node 230.

[0087] In some examples, the quality information of the first recovered data can be the quality value of the first recovered data determined by the first node 220 based on local test data and the first recovered data. Taking the test data and the first recovered data as channel state information (CSI) as an example, the quality value of the first recovered data can be the squared generalized cosine similarity (SGCS) of the CSI recovered quality. Taking the test data and the first recovered data as an example, the quality value of the first recovered data can be the peak signal-to-noise ratio (PSNR) of the recovered image. Similarly, the quality information of the second recovered data can be the quality value of the second recovered data determined by the second node 230 based on local test data and the second recovered data. The quality values ​​of the first and second recovered data can be used by the receiver 210 to determine whether the quality of the first recovered data meets the quality requirements of the first node 220, and whether the quality of the second recovered data meets the quality requirements of the second node 230, respectively. For example, if the quality value of either the first or second recovered data is greater than a certain quality threshold, it is determined to meet the requirements; otherwise, it is determined to not meet the requirements.

[0088] In other examples, the quality information of the first recovered data can be the judgment result of the first node 220 on the quality of the first recovered data. Additionally or alternatively, the quality information of the second recovered data can be the judgment result of the second node 230 on the quality of the second recovered data. For example, after determining the quality value of the first recovered data based on local test data and the first recovered data, the first node 220 can further determine whether the quality of the first recovered data meets the quality requirements of the first node 220 based on the quality value. The specific judgment method is similar to that mentioned above; for example, if the quality value of the first recovered data is greater than a certain quality threshold, it is determined to meet the requirements; otherwise, it is determined to not meet the requirements. The process by which the second node 230 judges the quality of the second recovered data to obtain the judgment result is similar to the judgment process performed by the first node 220 described above. In some examples, if the judgment result of the quality of the first or second recovered data is 1, it indicates that it meets the quality requirements of the corresponding node; if it is 0, it indicates that it does not meet the quality requirements of the corresponding node.

[0089] In the example described above, where the first and second AI models are trained by the receiving end 210, and the first node 220 and the second node 230 have not yet deployed AI models, the test data for the first node 220 and the second node 230 are generated by the receiving end 210. After obtaining the first restored data, the receiving end 210 can determine the quality information of the first restored data based on the first restored data and the generated test data of the first node. After obtaining the second restored data, the receiving end 210 can determine the quality information of the second restored data based on the second restored data and the generated test data of the second node. Similar to the example mentioned above, the quality information of the first restored data can be the quality value of the first restored data, or a judgment result on the quality of the first restored data obtained based on the quality value of the first restored data. Similarly, the quality information of the second restored data can be the quality value of the second restored data, or a judgment result on the quality of the second restored data obtained based on the quality value of the second restored data. The receiving end 210 performs a quality judgment process on the first or second recovered data based on the quality value of the first or second recovered data. This is the same as the quality judgment process performed on the first or second recovered data by the sending end node (first or second node) based on the quality value of the first or second recovered data as described above, and will not be repeated here.

[0090] In some examples, receiver 210 may also receive indication information of AI models from multiple nodes. This indication information can be used to indicate the AI ​​model's identifier (i.e., model ID), functionality, or one or more elements from the training dataset. In some examples, receiver 210 may also receive input and output description information of AI models from multiple nodes. In some examples, receiver 210 may also receive data type information of multiple nodes. These multiple nodes include the first node 220 and the second node 230, and may also include other nodes from the sending end. In some examples, receiver 210 can determine a candidate node set based on the received node data type information and the input and output description information of the node's AI model. In some examples, the nodes in the candidate node set may refer to nodes that were initially successfully paired. If it is determined that the first node 220 and the second node 230 belong to the candidate node set, the operation of determining that the first node 220 and the second node 230 belong to the node set is further performed. In the process of determining the candidate node set (i.e., the initial node pairing process), for example, if the data type information of the first node 220 and the second node 230 both indicate the same data type, such as an image, and the input-output description information of the first AI model of the first node 220 and the input-output description information of the second AI model of the second node 230 both indicate that the output of the corresponding first or second AI model is of the same type, such as an image embedding representation, then it means that at the receiving end 210, the same AI model (the third AI model) can be used to process the output data of the AI ​​model of the sending node (e.g., the image embedding representation) to perform data recovery, such as reconstructing the image. In this case, the first node 220 and the second node 230 are nodes in the candidate node set. Node pairing can also be referred to as user pairing in some examples.

[0091] In some examples, receiver 210 can receive processing capability information from multiple nodes (e.g., first node 220 and second node 230, and may also include other nodes from the sender). The processing capability information of the first node is used to determine the first AI model, and the processing capability information of the second node is used to determine the second AI model. As mentioned in the example above, the first and second AI models are trained by receiver 210; therefore, the simulation is completed on the receiver 210 side. Figure 1DThe process shown requires selecting corresponding personalized models for the first node 220 and the second node 230. When selecting a personalized model for a sending node, the receiving end 210 may consider the processing capability information of the corresponding node. In some examples, the node's processing capability information may include the node's computing power information, specifically including the following capabilities of the node: whether the processor's computing power (e.g., whether the processor can perform one trillion (10^12) operations per second, TOPS) can support model inference, whether the memory / GPU memory can load the corresponding model and inference data, and whether the speed and latency of inference using the corresponding model under the current CPU / GPU utilization meet the system requirements, etc.

[0092] In some examples, such as when the first AI model and the second AI model are trained by the receiver 210 and the first node 220 and the second node 230 have not yet deployed AI models, the receiver 210 may also send the first AI model to the first node 220 and the second AI model to the second node 230.

[0093] In some examples, the receiver 210 can query the historical information record table for historical records regarding whether a node belongs to a node set. If such historical records exist, the node set can be determined based on these records. For example, if the historical information record table contains historical records indicating that the first node 220 and the second node 230 belong to the node set (e.g., pairing records), then the first node 220 and the second node 230 can be directly determined to belong to the node set based on these historical records. In some examples, the historical information record table may be referred to as a pairing history information table; for a detailed description of pairing history information tables, please refer to the relevant examples below.

[0094] In some examples, when the historical record includes the following information, it can be determined that the first node 220 and the second node 230 belong to the node set: the identification information of the first node 220, the identification information of the second node 230, the first indication information of the first AI model, the second indication information of the second AI model, and the third indication information of the third AI model used to process the second overlay data. As mentioned in the example above, the second overlay data includes data obtained by overlaying the third and fourth data. The third data is obtained by processing the test data of the first node using the first AI model, and the fourth data is obtained by processing the test data of the second node using the second AI model. The identification information of the first node 220 is, for example, the ID of the first node 220. The identification information of the second node 230 is, for example, the ID of the second node 230. The first indication information of the first AI model is, for example, the ID of the first AI model. The second indication information of the second AI model is, for example, the ID of the second AI model. The third indication information of the third AI model used to process the second overlay data is, for example, the ID of the third AI model. In such examples, the pairing history information table can be found in Table 1 of the related examples below.

[0095] In some examples, where the historical record includes first indication information, second indication information, and third indication information, it can be determined that the first node 220 and the second node 230 belong to the node set. In such examples, the pairing history information table can be found in Table 2 of the related examples below.

[0096] In some examples, when no historical records exist, the receiver 210 can generate records about which nodes in the node set belong to the node set, and store these records in the historical information record table. This operation enables dynamic updating and maintenance of the historical information record table.

[0097] The specific process by which the receiving end 210 determines the node set in process 201 above can be further referenced in processes 211 and 221. Process 211 applies to the example mentioned above where the first node 220 deploys the first AI model and the second node 230 deploys the second AI model. In such an example, the selection or generation of test data is performed at the sending node. During the process of determining the node set, the interaction between the receiving end 210 and the sending nodes (such as the first node 220 and the second node 230) is implemented. Figure 1D The process is illustrated. Procedure 221 applies to the example mentioned above where the first and second AI models are trained by the receiving end 210, and the first node 220 and the second node 230 have not yet deployed AI models. In such an example, the test data for the first node 220 and the test data for the second node 230 are generated at the receiving end 210, and the simulation is implemented on the receiving end 210 side. Figure 1DThe process is shown below. (Followed by...) Figure 2B and Figure 2C The communication flow shown is described in terms of flow 211 and flow 221 respectively.

[0098] like Figure 2B As shown, process 211 involves a receiver 210 and a sender, the sender may include one or more nodes, with a first node 220 and a second node 230 shown as examples. For a description of the scenarios in which process 211 is used and the nodes of the receiver 210 and the sender, see process 201. Process 211 can be understood in conjunction with some examples in process 201 regarding the determination of the node set.

[0099] At point 212, the first node 220 sends third data 207, and the second node 230 sends fourth data 217. The receiving end 210 receives second superimposed data, which includes the superimposed data of the third data 207 and the fourth data 217. The third data is obtained by processing test data selected or generated by the first node 220 through a first AI model, and the fourth data is obtained by processing test data selected or generated by the second node 230 through a second AI model.

[0100] At 214, receiver 210 processes the second overlay data using a third AI model to obtain first recovered data 227 and second recovered data 237. At 216a and 216b, receiver 210 sends the obtained first recovered data 227 and second recovered data 237 to first node 220 and second node 230, respectively. At 218a, first node 220 determines quality information 247 of the first recovered data 227 based on local test data and the first recovered data 227. At 218b, second node 230 determines quality information 257 of the second recovered data 237 based on local test data and the second recovered data 237.

[0101] In 222a and 222b, the first node 220 and the second node 230 respectively send quality information 247 of the first recovery data 227 and quality information 257 of the second recovery data 237 to the receiving end 210.

[0102] After receiving the quality information 247 of the first recovered data 227 and the quality information 257 of the second recovered data 237, at step 224, the receiving end 210 determines, based on the quality information 247 and 257 of the first recovered data 227 and the second recovered data 237, that the first node 220 and the second node 230 belong to the node set. Referring to the example in process 201, taking the determination of whether the first node 220 belongs to the node set as an example, if the quality information 247 of the first recovered data 227 received by the receiving end 210 is the quality value of the first recovered data 227, then the receiving end 210 further performs a quality judgment on the first recovered data 227 based on this quality value, that is, judges whether the quality value of the first recovered data 227 meets the quality requirements of the first node 230. The specific judgment method is described in process 201. Alternatively, if the quality information 247 received by the receiving end 210 of the first recovered data 227 is a judgment result on the quality of the first recovered data 227, such as receiving a "1" indicating that the quality value of the first recovered data 227 meets the quality requirements of the first node 230, then the receiving end 210 can directly determine that the first node 220 belongs to the node set. The process of determining whether the second node 230 belongs to the node set is similar to the process of determining whether the first node 220 belongs to the node set.

[0103] like Figure 2C As shown, process 221 involves a receiver 210 and a sender. The sender may include one or more nodes, with a first node 220 and a second node 230 exemplarily shown. For a description of the scenarios in which process 221 is used and the nodes of the receiver 210 and the sender, see process 201. Process 221 can be understood in conjunction with some examples in process 201 regarding the determination of the node set.

[0104] At 232, the first node 220 sends its processing capability information 209 to the receiving end 210. At 234, the second node 230 sends its processing capability information 219 to the receiving end 210. At 236, the receiving end 210 determines the first AI model of the first node 220 based on the processing capability information 209, and determines the second AI model of the second node 230 based on the processing capability information 219.

[0105] At 238, the receiver 210 inputs the test data generated by the first node 220 into the first AI model to obtain the third data, and inputs the test data generated by the second node 230 into the second AI model to obtain the fourth data. Through the simulated channel transmission process, the third data and the fourth data are superimposed to form the second superimposed data.

[0106] At 242, receiver 210 processes the second overlay data using a third AI model, and the output of the third AI model is the first recovered data and the second recovered data. At 244, receiver 210 determines the quality information of the first recovered data based on the obtained first recovered data and the test data of the generated first node 220, and determines the quality information of the second recovered data based on the obtained second recovered data and the test data of the generated second node 230. Then, at 246, receiver 210 determines that the first node 220 and the second node 230 belong to the node set based on the quality information 247 of the first recovered data 227 and the quality information 257 of the second recovered data 237.

[0107] The specific process by which the receiving end 210 determines the node set in process 201 above can be further referenced in process 231. For example... Figure 2D As shown, process 231 involves a receiver 210 and a sender, the sender may include one or more nodes, with a first node 220 and a second node 230 shown as examples. For a description of the scenario in which process 211 is used and the nodes of the receiver 210 and the sender, see process 201. Process 231 can be understood in conjunction with some examples in process 201 regarding the use of a historical information record table to determine the set of nodes.

[0108] At step 252, the receiving end 210 queries the historical information record table for historical records regarding whether the first node 220 and the second node 230 belong to the node set. At step 254, if a historical record indicating that the first node 220 and the second node 230 belong to the node set is found, the belonging of the first node 220 and the second node 230 to the node set is determined based on the found historical record. At step 256, if the historical record is not found, the node set is determined by referring to the method described in steps 211 or 221. The interaction between the receiving end 210 and the sending end nodes (first node 220, second node 230) during this determination of the node set can be referred to... Figure 2B or Figure 2C .

[0109] The following example of uplink LoRA multiple access will be used to illustrate the multiple access process in this application embodiment. In the following uplink multiple access example, the transmitting end can be multiple UEs, and the UEs can be nodes of the transmitting end, such as UE1 and UE2. The receiving end can be the NW side. It should be noted that the following examples are not limited to the uplink, but can also be applied to the downlink or sidelink scenarios. In addition, the multiple access method is not limited to LoRA multiple access, but can also be applied to other multiple access methods that support model domain orthogonality.

[0110] The receiving end, such as the NW side, can perform initial pairing. Initial pairing can be based on the UE's data type and the output of the personalized model. Specifically, the initial pairing principle is as follows: if the data and personalized model outputs of multiple UEs can be detected by the same NW side model, meaning the UE's data type and personalized model output meet the same model's output and input requirements (e.g., type, format, dimension, modality), then the corresponding UEs are initially paired to obtain candidate LoRA multiple access user groups. Candidate LoRA multiple access user groups can be examples of candidate node sets in some of the examples above, and candidate LoRA multiple access users in the candidate LoRA multiple access user group are examples of nodes in the candidate node set. The initial pairing process can be an example of the process of determining the candidate node set. In some examples, for instance, if the data type of UE1 and UE2 is both images, and the output of the personalized models of UE1 and UE2 is both an embedded representation of the image, then the embedded representation of the image can be processed and the image reconstructed using the same common model. Therefore, UE1 and UE2 can be paired, meaning UE1 and UE2 are determined to belong to the candidate node set.

[0111] After initial pairing, the receiving end, such as the NW side, can perform test verification. The test verification process is an example of determining the node set from the candidate node set. In some examples, the test verification process can be determined through interaction between the sending and receiving ends, such as the process described in procedure 211. In other examples, the test verification process is primarily performed by the receiving end 210, such as the process described in procedure 221. An example of the test verification process is that the NW side or the UE can select test data for each user in each group of candidate LoRA multiple access users, with the selection principle being that the data type and content are the same as those reported by the UE to the NW.

[0112] According to the principle of uplink LoRA multiple access, the NW or UE (UE1 and UE2) can sequentially input test data (generated by the NW or UE) into the personalized model and transmit the output data of the personalized model through the air interface (when the NW performs this step, the air interface transmission can be implemented by the NW through local simulation; when the UE performs this step, the air interface transmission direction is from the NW to the UE). The NW inputs the superimposed data obtained by superimposing the output data of the personalized models corresponding to UE1 and UE2 into the common model. The output result obtained is called the LoRA multiple access test response. The LoRA multiple access test response includes the first LoRA multiple access test response, that is, the recovered data of the test data of UE1, and the second LoRA multiple access test response, that is, the recovered data of the test data of UE2. The first LoRA multiple access test response can be used as an example of the first recovered data, and the second LoRA multiple access test response can be used as an example of the second recovered data.

[0113] The receiving end, such as the NW, can determine the data quality recovered by each user based on the LoRA multiple access test response, or send the LoRA multiple access test response to the corresponding UE so that the UE can determine the data quality recovered by each user. The metrics for measuring data recovery quality are related to the specific amount of data. For example, the recovery quality of CSI can be measured by SGCS, and the recovery quality of images can be measured by PSNR. If both meet the system's data quality requirements, the users in this group can access the network via LoRA multiple access, indicating successful UE pairing and that the node belongs to the node set. Conversely, if pairing fails, the user grouping can be adjusted (e.g., removing UEs whose recovery quality does not meet the requirements from the group, and the remaining UEs using LoRA multiple access) or the system can fall back to the multiple access method used in some other schemes. For specific procedures, please refer to [link to relevant documentation]. Figure 3 or Figure 4 .

[0114] Figure 3 This document illustrates a user pairing process for multiple access according to some embodiments of this application. Process 300 involves a receiver 210 and a transmitter, which may include multiple nodes, with a first node 220 and a second node 230 exemplarily shown. Taking an uplink LoRA multiple access scenario as an example, the transmitter can be the NW side. The first node 220 is, for example, UE1, and the second node 230 is, for example, UE2. A personalized model has been deployed on the UE side, which can be trained locally on the UE side or requested from the NW side. The main steps for user pairing for uplink LoRA multiple access are as follows. The dashed box 301 illustrates the process from initial pairing to determining the user pairing result. This process can serve as an example of determining the node set in some of the examples above. Initial pairing is an example of determining the candidate node set during the process of determining the node set.

[0115] In sections 302a and 302b, the first node 220 and the second node 230 respectively report their respective personalized model IDs (305a, 305b) and personalized model input / output description information (315a, 315b). This reporting can be performed proactively by the first node 220 and / or the third node 230 via data request signaling, or by the receiving end 210 issuing an instruction to instruct the first node 220 and / or the third node 230 to perform the reporting. Alternatively, if the personalized model has already been registered with the receiving end 210 (e.g., on the NW side) and the registration information includes the personalized model's input / output description information, the sending node only needs to report the personalized model ID during the reporting process.

[0116] The receiver 210 can perform node pairing based on the data type of the sending node (e.g., UE) and the output of the personalized model. In some examples, if pairing history information exists, the corresponding nodes are directly paired. For example, if there is historical pairing information between the first node 220 and the second node 230, it can be directly determined that the first node 220 and the second node 230 are paired, that is, these nodes belong to the node set.

[0117] If no pairing history information exists, user pairing is performed according to the process shown in box 301. Specifically, in 304, the receiving end 210 can first perform initial pairing to determine that the first node 220 and the second node 230 belong to the candidate LoRA multiple access user group (an example of a candidate node set), and select a matching public model based on the data type of the node and the output of the personalized model.

[0118] Next, at 306a and 306b, receiver 210 sends LoRA multiple access test commands 325a and 325b to nodes (e.g., UE1 and UE2) in the candidate LoRA multiple access user group. After receiving the LoRA multiple access test commands 325a and 325b, the nodes in the candidate LoRA multiple access user group (e.g., first node 220 and second node 230) select test data respectively.

[0119] As shown in 308a and 308b, the first node 220 and the second node 230 process the test data using their respective personalized models.

[0120] In 312a and 312b, the first node 220 and the second node 230 will process the test data (i.e. Figure 3 The LoRA multiple access test data 335a and 335b are sent to the receiving end 210. The receiving end 210 receives signals from multiple nodes (e.g., the first node 220 and the second node 230) (which are superimposed data obtained by superimposing the LoRA multiple access test data 335a and 335b).

[0121] At 314, the receiver 210 inputs the received superimposed data into the common model and runs the common model.

[0122] In steps 316a and 316b, receiver 210 uses the output of the common model as LoRA multiple access test responses 345a and 345b, and sends them to the corresponding sending nodes (first node 220 and second node 230). First node 220 and second node 230, respectively, determine the data recovery quality of their respective sending nodes based on the received LoRA multiple access test responses 345a and 345b and the actual values ​​of their local test data; that is, they determine whether the quality of the recovered test data meets the quality requirements of their respective sending nodes. Then, first node 220 or second node 230 generates LoRA multiple access test feedback 355a and 355b, respectively.

[0123] In steps 318a and 318b, the first node 220 and the second node 230 respectively inform the receiving end 210 of the generated LoRA multiple access test feedback (355a, 355b). In one feasible scheme, the transmitting node can provide feedback on the data recovery quality, such as CSI recovery quality (SGCS) or image recovery quality (PSNR). In another feasible scheme, the transmitting node can directly provide feedback on the judgment result, such as 1 or 0. For example, if the data recovery quality meets the node's quality requirements, the LoRA multiple access test feedback is 1; otherwise, it is 0.

[0124] At 322, the receiver 210 determines the final user pairing result based on the received LoRA multiple access test feedback 355a and 355b.

[0125] At 324a and 324b, the receiving end 210 sends LoRA multiple access indications 365a and 365b to the sending end nodes (first node 220 and second node 230).

[0126] In step 326, uplink LoRA multiple access communication is performed between the receiver 210 and the transmitter nodes (first node 220 and second node 230). The process of uplink LoRA multiple access communication can be referred to the descriptions of steps 206 and 208 in procedure 201.

[0127] Figure 4 This document illustrates a user pairing process for multiple access according to other embodiments of this application. Process 400 involves a receiver 210 and a transmitter, which may include multiple nodes, with a first node 220 and a second node 230 exemplarily shown. Taking an uplink LoRA multiple access scenario as an example, the transmitter can be the NW side. The first node 220 is, for example, UE1, and the second node 230 is, for example, UE2. In this case, the UE side does not deploy a personalized model; the personalized model on the UE side is trained and distributed by the NW side.

[0128] In section 402a, the first node 220 (e.g., UE1) reports computing power information 405a and data type description 415a. In section 402b, the second node 230 (e.g., UE2) reports computing power information 405b and data type description 415b. The computing power information 405a or 405b mainly includes TOPS, memory / video memory size, and CPU / GPU utilization. The data type descriptions 415a and 415b include images, text, CSI, etc. This information can be proactively reported by the node (first node 220 or second node 230) to the receiving end 210 via data request signaling, or the receiving end 210 can issue instructions to the node to instruct it to perform the reporting.

[0129] At 404, receiver 210 completes user pairing and determines the common model and the personalized model for each node. During user pairing, if historical pairing information for nodes exists, the corresponding nodes are directly paired and the relevant personalized model is determined. If no historical pairing information exists, receiver 210 first completes initial pairing and selects a matching common model based on the node's data type. Receiver 210 can complete user pairing and select the relevant personalized model locally through testing and verification. When selecting a node's personalized model, receiver 210 needs to consider the node's computing power information, such as whether TOPS can support model inference, whether memory / GPU memory can load the corresponding model and inference data, and whether the speed and latency of inference using the corresponding model under the current CPU / GPU utilization meet system requirements. The process of receiver 210 completing user pairing locally through testing and verification can be referenced in process 221.

[0130] In 406a and 406b, receiver 210 sends LoRA multiple access indications 425a and 425b to the corresponding nodes (first node 220 and second node 230) and sends personalized models 435a and 435b to the corresponding nodes (first node 220 and second node 230).

[0131] At 408, the receiver 210 performs uplink LoRA multiple access communication with the nodes that have completed user pairing (e.g., the first node 220 and the second node 230). The process of uplink LoRA multiple access communication can be referred to the introduction of 206 and 208 in process 201.

[0132] As mentioned in some examples above, node pairing can be performed based on a pairing history information table. Figure 5 The following diagram illustrates the usage flow of the pairing history information table in some embodiments of this application. Taking the receiver as NW and the sender as UE as an example, the process of constructing, updating, and using the pairing history information table is explained.

[0133] For the construction of the pairing history information table, each receiving end (e.g., NW) can build a pairing history information table to record historical pairing schemes. As shown in Table 1, in some examples, the main information in the pairing history information table includes the row ID, common model ID, UE ID, and LoRAID (UE personalized model ID). In other examples, if the UE's personalized model ID (i.e., LoRAID ID) is uniformly assigned by the NW, then the UE ID does not need to be recorded in the pairing history information table.

[0134] Table 1

[0135] Line ID Public Model ID UE1 ID & LoRA1 ID UE2 ID & LoRA2 ID …… 0 1 2&4 3&7 1 4 2&5 5&4 …… …… …… ……

[0136] Table 2

[0137] Line ID Public Model ID LoRA1 ID LoRA2 ID …… 0 1 4 7 1 4 5 1 …… …… …… ……

[0138] To update the pairing history information table, after completing user pairing (see some examples above for the user pairing process), NW can add / update records in the pairing history information table.

[0139] Regarding the use of the pairing history information table, at step 502, during user pairing, the table (i.e., the pairing history information table) can be looked up to determine if there are any reusable user pairing results. If there are records of reusable user pairing results in the pairing history information table, step 504 is executed to reuse the pairing history information in the pairing history information table. If there are no records in the pairing history information table, steps 506 and 508 are executed, that is, the initial pairing of the UE is performed first, followed by test verification. Such pairing operations can refer to process 211 or process 221 (the user pairing process is an example of the process of determining the set of nodes in process 211 or process 221). After the user pairing is completed, at step 512, the NW can update the pairing history information table.

[0140] In some examples, tables between NWs are shared periodically to further improve pairing efficiency. It's important to note that it's necessary to distinguish between upstream and downstream pairing history tables. This distinction can be achieved by, for example, adding an extra column to the table to record upstream / downstream pairings. Alternatively, or additionally, differentiation can be achieved using table IDs, such as different calculation methods for upstream / downstream pairing history table IDs, different seed values ​​using the same calculation method, different sequence lengths, or different first digits of the sequence.

[0141] Using a matching history table for user matching can further reduce the complexity of user matching.

[0142] Figure 6 A possible exemplary block diagram of the communication device involved in an embodiment of this application is shown. For example... Figure 6As shown, the communication device 600 may include modules or units for implementing the methods described in the embodiments above. In one possible design, the communication device 600 includes a processing unit 602 and a communication unit 603. Optionally, the communication device 600 may further include a storage unit 601 for storing device program code and / or data.

[0143] The communication device 600 can be the receiving end device or apparatus in the above embodiments. For example, the receiving end can be a network device or terminal, or a communication module / processing module in a network device or terminal, or a circuit or chip in a network device or terminal responsible for communication functions, etc.

[0144] For example, in one embodiment, the processing unit 602 is used to determine a set of nodes that will send data based on a first multiple access method, the set of nodes including a first node and a second node.

[0145] In one embodiment, the communication unit 603 is configured to send indication information, which instructs nodes in the node set to send data using a first multiple access method. The communication unit 603 is also configured to receive first overlay data.

[0146] In one embodiment, the processing unit 602 is further configured to process the first superimposed data through a third AI model to obtain first data corresponding to the first AI model and second data corresponding to the second AI model, wherein the data sent by the first node using a first multiple access method is the output data of the first AI model, and the data sent by the second node using a first multiple access method is the output data of the second AI model.

[0147] In one possible design, the processing unit 602 is used to determine the node set by: the processing unit 602 determining that the first node and the second node belong to the node set based on the quality information of the first recovered data of the test data of the first node and the quality information of the second recovered data of the test data of the second node, wherein the first recovered data and the second recovered data are obtained by processing the second superimposed data through a third AI model, and the second superimposed data includes data obtained by superimposing the third data and the fourth data, wherein the third data is obtained by processing the test data of the first node through the first AI model, and the fourth data is obtained by processing the test data of the second node through the second AI model.

[0148] In one possible design, the communication unit 603 is further configured to receive third data from the first node, or the processing unit 602 is further configured to generate test data for the first node, wherein the generated test data for the first node is used to process the third data through a first AI model. Additionally or alternatively, the communication unit 603 is further configured to receive fourth data from the second node, or the processing unit 602 is further configured to generate test data for the second node, wherein the generated test data for the second node is used to process the fourth data through a second AI model.

[0149] In one possible design, the communication unit 603 is further configured to send the first recovery data to the first node and receive quality information of the first recovery data from the first node. Alternatively, the processing unit 602 is further configured to determine the quality information of the first recovery data based on the first recovery data and the generated test data of the first node.

[0150] In one possible design, the communication unit 603 is further configured to send the second recovery data to the second node and receive quality information of the second recovery data from the second node. The processing unit 602 is further configured to determine the quality information of the second recovery data based on the second recovery data and the generated test data of the second node.

[0151] In one possible design, the communication unit 603 is further configured to send a first test instruction to the first node, the first test instruction instructing the first node to process its test data using a first AI model. Additionally or alternatively, the communication unit 603 is further configured to send a second test instruction to the second node, the second test instruction instructing the second node to process its test data using a second AI model.

[0152] In one possible design, the communication unit 603 is also used to receive instruction information from AI models of multiple nodes, including a first node and a second node.

[0153] In one possible design, the communication unit 603 is also used to receive input and output description information of AI models from multiple nodes.

[0154] In one possible design, the communication unit 603 is also used to receive processing capability information of multiple nodes, including a first node and a second node. The processing capability information of the first node is used to determine a first AI model, and the processing capability information of the second node is used to determine a second AI model.

[0155] In one possible design, the communication unit 603 is also used to send a first AI model to a first node and a second AI model to a second node.

[0156] In one possible design, the processing unit 602 is further configured to determine that the first node and the second node belong to a node set if the first node and the second node belong to a candidate node set, wherein the nodes in the candidate node set are determined based on the following: data type information of the nodes in the candidate node set, and input-output description information of the AI ​​model of the nodes in the candidate node set.

[0157] In one possible design, the processing unit 602 is also used to query the historical information record table for historical records of nodes belonging to the node set, and if historical records exist, to determine the node set based on the historical records.

[0158] In one possible design, the processing unit 602 is further configured to determine, when the historical record includes the following information, that the first node and the second node belong to the node set: the identification information of the first node, the identification information of the second node, the first indication information of the first AI model, the second indication information of the second AI model, and the third indication information of the third AI model used to process the second overlay data, wherein the second overlay data includes data obtained by overlaying the third data and the fourth data, wherein the third data is obtained by processing the test data of the first node through the first AI model, and the fourth data is obtained by processing the test data of the second node through the second AI model; or the first indication information, the second indication information, and the third indication information.

[0159] In one possible design, the processing unit 602 is also used to generate a record of which nodes in the node set belong to the node set in the absence of historical records, and to store the generated record in the historical information record table.

[0160] In one possible design, the communication unit 603 is also used to receive data type information from multiple nodes, including a first node and a second node.

[0161] In one possible design, the first overlay data includes one or more tokens.

[0162] In one possible design, when the communication device 600 is a communication module in a terminal or network device, the function of the processing unit 602 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 603 can be implemented by a transceiver circuit.

[0163] In one possible design, when the communication device 600 is a circuit or chip responsible for communication functions in a terminal or network device, 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 602 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 603 can be implemented by an interface circuit or data transceiver circuit on the aforementioned chip.

[0164] In one possible design, when the communication device 600 is a communication and / or computing module in a terminal or network device, the functionality of the processing unit 602 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. Alternatively, the processor may include an ASIC, or a SoC or SIP chip containing an ASIC. Alternatively, the processor may include an NPU, or a SoC or SIP chip containing an NPU. The functionality of the communication unit 603 can be implemented by transceiver circuitry.

[0165] In one possible design, when the communication device 600 is a circuit or chip in a terminal or network device 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, an NPU or a SoC or SIP chip containing an NPU, or an ASIC or a SoC or SIP chip containing an ASIC, the function of the processing unit 602 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 603 can be implemented by interface circuits or data transceiver circuits on the aforementioned chip.

[0166] 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.

[0167] 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, or one or more 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.

[0168] In one example, storage unit 601 may include random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory and / or registers, etc.

[0169] See Figure 7 This is a schematic diagram of the structure of a terminal 700 provided in an embodiment of this application. The terminal 700 can correspond to the sending or receiving end shown in the aforementioned embodiments, such as a first node or a second node. Figure 7 As shown, the terminal includes: one or more antennas 710, a radio frequency processing system 720, and a processor system 730.

[0170] In the downlink or sidelink direction, the RF processing system 720 receives RF signals through the antenna 710 and sends the RF-processed signals to the processor system 730 for further processing. In the uplink or sidelink direction, the processor system 730 processes the terminal-side information and sends it to the RF processing system 720, which then processes the signal and transmits it through the antenna 710.

[0171] In one example, the radio frequency (RF) processing system 720 serves as the communication interface for external communication of the terminal and may include an RF front end (RFFE) 721 and an RF transceiver 722. The RFFE 721 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 721 can be a circuit system composed of multiple discrete components or integrated into one or more chips. The RF transceiver 722 processes the RF signals received by the RFFE into baseband / IF signals for further processing by the processor system 730, and processes the baseband / IF signals provided by the processor system 730 into RF signals for transmission to the RFFE 721. The baseband / IF signals transmitted between the RF transceiver 722 and the processor system 730 can be digital or analog signals. The RF transceiver 722 can be implemented by one or more chips, which are commonly referred to as RF ICs.

[0172] In one example, the processor system 730 may include one or more processors for processing signals and executing one or more communication protocols. Optionally, the processor system 730 may also include a memory 736. In one example, the one or more processors include at least one baseband processor 731 (also known as a modem processor). The memory 736 is used to store data and / or computer program instructions. Optionally, the processor system 730 may also include one or more application processors 732 for implementing processing of the terminal operating system and application layer. The application processor 732 may include, for example, a GPU, NPU, AI processor, or ASIC. Optionally, the processor system 730 may also include one or more of a voice subsystem 733, a multimedia subsystem 734, or an interface circuit 735. The voice subsystem 733 is used to process voice signals, the multimedia subsystem 734 is used to handle multimedia-related operations, such as video encoding / decoding, image processing, etc., and the interface circuit 735 is used to implement communication with other terminal components, such as a display 740, an input device 750, a memory 760, etc. The above-mentioned components in the processor system 730 can communicate with each other via a bus or communication interface circuit.

[0173] In one example, the processor system 730 can be packaged as a single processor chip, such as a SoC chip or a SIP chip. In another example, the processor system 730 can be a system composed of multiple chips; for example, the baseband processor 731 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.

[0174] In one example, memory 736 can be on-chip memory, i.e., located on the processor system 730 chip. In another example, memory 760 can be off-chip memory, i.e. located outside the processor system 730 chip.

[0175] In one example, the baseband processor 731 may include one or more processor cores 7311 and interface circuitry 7314. The one or more processor cores 7311 are used to process signals and execute one or more communication protocols. Optionally, the baseband processor 731 may also include a memory 7312 for storing at least a portion of the corresponding computer program instructions and / or data. In one example, the one or more processor cores 7311 execute the computer program instructions stored in the memory 7312 to implement the relevant operations in the above method embodiments (such as determining a first transmission parameter based on experience quality parameters). In this disclosure, memory 7312 is used to store corresponding computer program instructions and / or data. This can mean that memory 7312 stores all corresponding computer program instructions and / or data for execution by processor core 7311; or it can mean that memory 7312 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 7311. Memory 7312 can store different portions of computer program instructions and / or data multiple times for execution by processor core 7311 to implement the relevant operations in the above method embodiments. Interface circuit 7314 serves as a communication interface for communication with other components, such as transmitting signals with radio frequency processing system 720, communicating with other subsystems and related components of processor system 730 via bus, such as transmitting data control signals with application processor 732, and transmitting data or computer program instructions with memory 736 or memory 760. Optionally, in order to reduce the load on the processor core, a baseband signal processing circuit 7313 can be set to perform at least some baseband signal processing, including one or more of signal demodulation, modulation, encoding or decoding.

[0176] In one example, the communication device provided in this application may be a terminal 700, a communication module including a processor system 730 and a radio frequency system 720, or a baseband processor 731.

[0177] 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: CPU, DSP, microprocessor unit (MPU), MCU, GPU, FPGA, application-specific integrated circuit, artificial intelligence processor (AI processor) or neural network processor.

[0178] The aforementioned memory may include one or more of the following storage media: random access memory (RAM), static random access memory (SRAM), DRAM, phase-change memory (PCM), resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), cache, register, 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 in non-volatile memory, such as at least a portion of the aforementioned memory 760 (e.g., one or more of ROM, flash memory, EPROM, or hard disk). When the terminal is running, 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 736 and / or memory 7312 (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.

[0179] In one example, the RF transceiver 722 and the RF front-end 721 can also be packaged in a single chip. In another example, the RF transceiver 722, the RF front-end 721, and the baseband processor 731 can also be packaged in a single chip.

[0180] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, or all or part of it, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, random access memory (RAM), magnetic disks, or optical disks.

[0181] Based on the above embodiments, this application also provides a computer program that, when run on a computer, causes the computer to execute any of the methods provided in the above embodiments.

[0182] Based on the above embodiments, this application also provides a computer-readable storage medium storing a computer program. When executed by a computer, the computer program causes the computer to perform any of the methods provided in the above embodiments. The storage medium can be any available medium that can be accessed by a computer. By way of example, but not limited to, a computer-readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code having the form of instructions or data structures and that can be accessed by a computer.

[0183] Based on the above embodiments, this application also provides a chip for reading a computer program stored in a memory and implementing any of the methods provided in the above embodiments.

[0184] Based on the above embodiments, this application provides a chip system including a processor for supporting a computer device in implementing the functions involved in the communication devices in the above embodiments. In one possible design, the chip system further includes a memory for storing necessary programs and data of the computer device.

[0185] 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, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0186] 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.

[0187] 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.

[0188] 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, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0189] 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 that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0190] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

Claims

1. A communication method, characterized in that, include: Determine a set of nodes that will transmit data based on a first multiple access method, the set of nodes including a first node and a second node; Send an instruction message, the instruction message being used to instruct the nodes in the node set to send data using the first multiple access method; Receive the first overlay data; as well as The first superimposed data is processed by the third AI model to obtain first data corresponding to the first AI model and second data corresponding to the second AI model, wherein the data sent by the first node using the first multiple access method is the output data of the first AI model, and the data sent by the second node using the first multiple access method is the output data of the second AI model.

2. The method according to claim 1, characterized in that, Determining the set of nodes includes: Based on the quality information of the first recovered data from the test data of the first node and the quality information of the second recovered data from the test data of the second node, it is determined that the first node and the second node belong to the node set, wherein, The first recovered data and the second recovered data are obtained by processing the second superimposed data through the third AI model. The second superimposed data includes data obtained by superimposing the third data and the fourth data. The third data is obtained by processing the test data of the first node through the first AI model, and the fourth data is obtained by processing the test data of the second node through the second AI model.

3. The method according to claim 2, characterized in that, It also includes performing at least one of the following: The third data is received from the first node, or test data for the first node is generated, wherein the generated test data for the first node is used to obtain the third data through processing by the first AI model; or The fourth data is received from the second node, or test data for the second node is generated, wherein the generated test data for the second node is used to process the fourth data through the second AI model.

4. The method according to claim 2 or 3, characterized in that, Also includes: Send the first recovered data to the first node, and receive the quality information of the first recovered data from the first node; or Based on the first recovered data and the generated test data of the first node, the quality information of the first recovered data is determined.

5. The method according to any one of claims 2-4, characterized in that, Also includes: The second recovered data is sent to the second node, and the quality information of the second recovered data is received from the second node; or Based on the second recovered data and the generated test data of the second node, the quality information of the second recovered data is determined.

6. The method according to any one of claims 1 to 5, characterized in that, It also includes performing at least one of the following: Send a first test instruction to the first node, the first test instruction being used to instruct the first node to process the test data of the first node through the first AI model; or Send a second test instruction to the second node, the second test instruction being used to instruct the second node to process the test data of the second node through the second AI model.

7. The method according to any one of claims 1 to 6, characterized in that, Also includes: The system receives instruction information from AI models of multiple nodes, including the first node and the second node.

8. The method according to claim 7, characterized in that, Also includes: Receive the input and output description information of the AI ​​models from the multiple nodes.

9. The method according to any one of claims 1 to 8, characterized in that, Also includes: The system receives processing capability information from multiple nodes, including the first node and the second node. The processing capability information of the first node is used to determine the first AI model, and the processing capability information of the second node is used to determine the second AI model.

10. The method according to claim 9, characterized in that, Also includes: Send the first AI model to the first node; and Send the second AI model to the second node.

11. The method according to any one of claims 1 to 10, characterized in that, Also includes: If it is determined that the first node and the second node belong to the candidate node set, then the nodes in the candidate node set are determined based on the following: The data type information of the nodes in the candidate node set; and The input and output description information of the AI ​​model of the nodes in the candidate node set.

12. The method according to claims 1 to 11, characterized in that, Also includes: Query the historical records table for the nodes in the node set that belong to the node set; If the historical records exist, the set of nodes is determined based on the historical records.

13. The method according to claim 12, characterized in that, Also includes: The first node and the second node are determined to belong to the node set if the historical record includes the following information: The identification information of the first node, the identification information of the second node, the first indication information of the first AI model, the second indication information of the second AI model, and the third indication information of the third AI model used to process the second overlay data. The second overlay data includes data obtained by overlaying the third data and the fourth data. The third data is obtained by processing the test data of the first node through the first AI model, and the fourth data is obtained by processing the test data of the second node through the second AI model; or The first instruction information, the second instruction information, and the third instruction information.

14. The method according to claim 12, characterized in that, Also includes: If the historical record does not exist, generate a record indicating that a node in the node set belongs to the node set; as well as The generated records are stored in the historical information record table.

15. The method according to any one of claims 1 to 14, characterized in that, Also includes: The system receives data type information from multiple nodes, including the first node and the second node.

16. The method according to any one of claims 1 to 15, characterized in that, The first overlay data includes one or more tokens.

17. A communication device, characterized in that, include: Units for performing the method according to any one of claims 1 to 16.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions that, when executed, cause the method according to any one of claims 1 to 16 to be performed.

19. A computer program product, characterized in that, It includes a computer program or instructions that, when executed, cause the method according to any one of claims 1 to 16 to be performed.

Images