Communication method, communication device and system

Abstract

The application provides a communication method, a communication device and a system. In the method, the first device can determine the power corresponding to each second power based on the computing power of each second device in N second devices, and can determine the data corresponding to each second device according to the first AI model corresponding to the N second devices. In this way, the first device can send a first signal to the second device based on the power and data corresponding to each second device in the N second devices. Wherein N is an integer greater than 1. In this way, even in a large-scale scenario, the first device can reasonably allocate power to the second device, which helps to improve the efficiency of multiple access.

Description

Technical Field

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

[0002] With the continuous evolution of modern communication networks and the explosive growth of user demand, multiple access technology (MIG) is becoming increasingly important as a key support for enabling numerous users to simultaneously access the network and enjoy communication services. The purpose of MIG is to allow multiple users to access the network simultaneously, enjoy the communication services provided by network equipment, and ensure that the signals of each user do not interfere with each other.

[0003] In current multiple access technology systems, the network side typically allocates power resources to each user according to pre-defined rules. However, these pre-defined rules are difficult to accurately adapt to different users. Especially in large-scale user access scenarios, as the number of users increases rapidly, the network side needs to allocate power to a large number of users simultaneously. Inappropriate power allocation methods (e.g., allocating too much or too little power to some users) prevent the efficient use of power resources, thus severely restricting the efficiency of multiple access. Summary of the Invention

[0004] This application provides a communication method, communication device, and system that are beneficial for improving the efficiency of multiple access.

[0005] In a first aspect, a communication method is provided. This method can be applied to a first device (e.g., a network device), and can be executed by the first device itself, or by components configured in the first device (e.g., a processor, chip, chip system, etc.), or by logic modules or software capable of implementing all or part of the functions of the first device. This application does not limit the scope of this method.

[0006] The method includes: a first device determining the power corresponding to each of N second devices, wherein the power corresponding to each second device is determined based on the computing power of the second device, and N is an integer greater than 1; and sending a first signal based on the power and data corresponding to each of the N second devices, wherein the data corresponding to each second device is determined based on a first artificial intelligence (AI) model.

[0007] The first AI model can be used to process the data to be transmitted for each of the N second devices. The data to be transmitted for each second device refers to the data that the first device wants to transmit to the second devices. Since the data to be transmitted for each of the N second devices is related (e.g., sharing the same type, format, dimension, or modality), the first device can process the data for each of the N second devices simultaneously using the same first AI model. After inputting the data to be transmitted for each of the N second devices into the first AI model, the first device outputs the data corresponding to each of the N second devices.

[0008] It should be understood that, in order to distinguish between the data input to the first AI model and the data output by the first AI model, the input data is referred to as the first data corresponding to each second device, and the output data is referred to as the second data corresponding to each second device.

[0009] The computing power of each of the N secondary devices varies, resulting in different power requirements for each device. If the relationship between the power of a secondary device and its computing power is negatively correlated, then the secondary device with stronger computing power will have lower power requirements. This is because a device with stronger computing power can process signals more effectively, performing signal processing and detection tasks well even with lower power. Conversely, a secondary device with weaker computing power will have higher power requirements, as it needs more power to compensate for its insufficient computing power and ensure it can process signals normally, thereby improving the overall system performance and stability.

[0010] Based on the above scheme, the first device can identify the N second devices performing multiple access as devices in the same multiple access user group, and process the data to be transmitted corresponding to each of the N second devices based on the same first AI model, such as extracting the data features of each second device to obtain the second data corresponding to each second device. Since the computing power of each of the N second devices differs, the first device can allocate corresponding power to each second device based on its computing power. In this way, the first device can send a first signal to each of the N second devices based on the first AI model. This first signal includes the data and power corresponding to each of the N second devices. Thus, even in large-scale access scenarios, the first device can reasonably allocate power to the second devices, which is beneficial to improving the efficiency of multiple access.

[0011] In one possible implementation, the signals corresponding to different second devices in the first signal are carried on the same resources. These same resources may include the same time domain, frequency domain, spatial domain, or codewords, etc., and this application does not limit this. Carrying the signals corresponding to different second devices in the first signal on the same resources can improve the utilization rate of communication resources.

[0012] In one possible implementation, the first signal includes one or more tokens. Tokens can also be called labels or terms, etc., and the naming of tokens is not limited in this application embodiment.

[0013] A token can be understood as the basic unit processed by the first AI model. For example, a token can be the basic unit of text, data, or features processed by the first AI model. Alternatively, the first AI model processes data using tokens as the basic unit; text, data, or features processed by the AI ​​model can be represented as tokens for processing and transmission. Or, the input and output of the first AI model consist of a series of numbers or vectors, which can be called tokens, or these numbers or vectors can be represented as tokens.

[0014] In one possible implementation, at most one of the N second devices is unable to run the second AI model due to insufficient computing power. The second AI model is either the first AI model or obtained by distillation based on the first AI model.

[0015] The model obtained by distillation based on the first AI model can also be called a distilled version of the first AI model, a lightweight version, etc., without any limitation.

[0016] The first AI model is a relatively complex model with a large computational load. The distilled version of the first AI model is a simplified version of the first AI model to reduce the computational requirements. Therefore, among the N second devices, the second device with the lowest computing power may not be able to run the first AI model or the distilled version of the first AI model.

[0017] In one possible implementation, each second device corresponds to the same third AI model, which is used to obtain the signal corresponding to the second device from the first signal.

[0018] In this application, the first signal output by the first device based on the first AI model can be input into the third AI model corresponding to each second device. Thus, each second device can acquire or detect the corresponding signal from the first signal based on the third AI model. The third AI model has a relatively small number of parameters and low computational complexity. For the second devices, using the same third model avoids excessive computational burden caused by employing a complex model.

[0019] In one possible implementation, the method further includes: determining a model corresponding to each second device based on the computing power of each second device, the model including: a third AI model, or a second AI model and a third AI model, the third AI model being used to obtain the signal corresponding to the second device from the first signal, the second AI model being the first AI model or obtained by distillation based on the first AI model; and sending first indication information, the first indication information being used to configure the model corresponding to each second device.

[0020] When the computing power of the second device is low, the first indication information is used to indicate that the model corresponding to the second device is the third AI model. On the one hand, the computing power of the second device is low, and the third AI model is relatively simple and does not require high computing power. On the other hand, the second device with lower computing power has higher power consumption, which helps it overcome interference factors such as noise in signal acquisition or detection to a certain extent.

[0021] When the second device has strong computing power, the first indication information is used to indicate that the model corresponding to the second device is the third AI model and the second AI model. Because the second device has strong computing power, it can perform more complex signal acquisition or detection operations. However, the second device has relatively low power, so it may not be able to overcome noise and other interference in signal acquisition or detection. Therefore, the second device can use the third AI model and the second AI model to reconstruct the interference signals from second devices with lower computing power (but higher power) to obtain the final signal it needs.

[0022] This application does not limit the method by which the first device obtains the computing power of each second device. In one implementation, the first device requests the second device to report; in another implementation, the second device actively reports.

[0023] In one possible implementation, the method further includes receiving second indication information from each second device, the second indication information indicating the computing power of the second device. Each second device has a different computing power, and the first device can allocate corresponding power to each second device based on its computing power.

[0024] In one possible implementation, the method further includes: sending third indication information, which is used to indicate the sending of second indication information. The second indication information is used to indicate the computing power of each of the N second devices, and the first device sends the indication information to instruct the N second devices to send computing power information, so that the first device can allocate power to each of the N second devices.

[0025] In one possible implementation, the second indication information includes at least one of the following: trillions of operations per second (TOPS); memory size; video memory size; central processing unit (CPU) utilization; and graphics processing unit (GPU) utilization. Based on the second indication information, the first device can gain a clearer understanding of the computing power of each of the N second devices.

[0026] In one possible implementation, the signal-to-interference-plus-noise ratio (SIR) of each second device when receiving a signal at a corresponding power is greater than or equal to a threshold, which is used to determine whether the second device can obtain the corresponding signal from the first signal through a third AI model.

[0027] The signal-to-interference-plus-noise ratio (SIR / NNR) reflects the relative relationship between signal strength and interference / noise intensity. When the SIR / NNR is greater than or equal to a threshold, it means that the interference and noise affecting the signal during transmission are within an acceptable range, and the third AI model can more clearly identify and analyze the feature information in the first signal. Although the third AI model has relatively fewer parameters, under good SIR / NNR conditions, it can still extract the signal corresponding to each second device from the first signal, thus ensuring the accuracy of signal acquisition.

[0028] In one possible implementation, the method further includes: determining a first AI model based on the data to be transmitted. When the first data corresponding to the second device matches the input of the first AI model, the first device can determine that the first AI model is the model for processing the first data corresponding to the second device. After processing the first data corresponding to each second device based on the first AI model, the first device can output the second data corresponding to each second device.

[0029] In one possible implementation, the method further includes: determining a third AI model based on the data to be transmitted and the computing power of the second device. The third AI model is used to obtain the signal corresponding to the second device from the first signal. The second device can obtain the corresponding signal from the first signal based on the AI ​​model; therefore, the computing power of the second device should meet the size requirements of the AI ​​model, and the first data corresponding to the second device should meet the output requirements of the AI ​​model. For example, the first device can determine the third AI model as the AI ​​model on the second device side based on the first data corresponding to the second device and the computing power of the second device, and the second device obtains the corresponding signal from the first signal based on the third AI model.

[0030] In one possible implementation, the method further includes: sending fourth indication information, which instructs the second device to receive the first signal based on the third AI model. Alternatively, the fourth information instructs the second device to perform first multiple access. This indicates that each of the N second devices can perform first multiple access. Here, first multiple access can be understood as sending a first signal when the first device allocates power to the second devices based on their computing power. The first signal is generated by the first device based on the power and data corresponding to each second device.

[0031] In one possible implementation, the method further includes: generating a mapping relationship corresponding to N second devices, the mapping relationship including at least one of the following: information of a first AI model; information of a third AI model; information of each second device; a power preset value, the power preset value being greater than or equal to the sum of the power of the N second devices.

[0032] When multiple second devices are accessing the network via multiple access points, the first device can first check the pairing history for relevant records of these second devices when pairing users with them. If records exist, the pairing history can be reused directly; otherwise, the first device will proceed with user pairing. This reduces the complexity of user pairing.

[0033] In one possible implementation, the method further includes: sending fifth indication information, which indicates the power corresponding to each second device, so that the second device can obtain the corresponding signal from the first signal based on the corresponding power.

[0034] Secondly, a communication method is provided, which can be applied to a second device (e.g., a terminal device). For example, it can be executed by the second device, or by components configured in the second device (e.g., processors, chips, chip systems, etc.), or by logic modules or software capable of implementing all or part of the functions of the second device. This application does not limit this aspect.

[0035] The method includes: receiving a first signal based on the power and data corresponding to each of the N second devices, wherein the power corresponding to each second device is determined based on the computing power of the second device, and the data corresponding to each second device is determined based on a first AI model, and N is an integer greater than 1; based on the first signal, obtaining the signal corresponding to the second device from the first signal using a third AI model, or obtaining the signal corresponding to the second device from the first signal using both a second AI model and a third AI model, wherein the second AI model is the first AI model or is obtained by distillation based on the first AI model.

[0036] In one possible implementation, the signals corresponding to different second devices in the first signal are carried on the same resource.

[0037] In one possible implementation, the first signal includes one or more tokens. In another possible implementation, at most one of the N second devices has insufficient computing power to run the second AI model.

[0038] In one possible implementation, each second device corresponds to the same third AI model.

[0039] In one possible implementation, the method further includes: receiving first indication information, the first indication information being used to configure a model corresponding to each second device, the model being determined based on the computing power of each second device, and the model including: the third AI model, or, the second AI model and the third AI model.

[0040] In one possible implementation, the method further includes sending a second indication message, which is used to indicate the computing power of the second device.

[0041] In one possible implementation, the method further includes: receiving third indication information, which is used to indicate the sending of second indication information.

[0042] In one possible implementation, the second indication information includes at least one of the following: TOPS; memory size; video memory size; CPU utilization; GPU utilization.

[0043] In one possible implementation, the signal-to-interference-plus-noise ratio (SIR) of each second device when receiving a signal at a corresponding power is greater than or equal to a threshold, which is used to determine whether the second device can obtain the corresponding signal from the first signal through a third AI model.

[0044] In one possible implementation, the first AI model is determined based on the data to be transmitted.

[0045] In one possible implementation, the third AI model is determined based on the data to be transmitted and the computing power of the second device.

[0046] In one possible implementation, the method further includes receiving fourth indication information, which instructs the second device to receive the first signal based on a third AI model.

[0047] In one possible implementation, the method further includes receiving fifth indication information, which indicates the power corresponding to each second device.

[0048] Thirdly, a communication device is provided that can implement the communication method described in any of the possible implementations of the first or second aspect. The device includes one or more functional units or modules for performing the described method. The functional units or modules included in the device can be implemented by software and / or hardware.

[0049] Fourthly, a communication device is provided, comprising at least one processor for executing the communication method described in any possible implementation of the first or second aspect.

[0050] Optionally, the apparatus may further include a memory for storing instructions and data. The memory is coupled to the processor, which, when executing the instructions stored in the memory, can implement the methods described in the foregoing aspects.

[0051] Optionally, the device may further include a communication interface for communicating with other devices. For example, the communication interface may be a transceiver, circuit, bus, module, or other type of communication interface.

[0052] Fifthly, a chip system is provided, the chip system including at least one processor for supporting the implementation of the functions involved in any possible implementation of the first or second aspect described above, such as receiving or processing data and / or information involved in the methods described above.

[0053] In one possible design, the chip system also includes a memory for storing program instructions and data, which may be located within or outside the processor.

[0054] In one possible design, the chip system further includes an interface circuit and / or a power supply circuit, wherein the interface circuit is used to transmit data and the power supply circuit is used to supply power to the chip system.

[0055] The chip system can consist of chips or include chips and other discrete components.

[0056] In a sixth aspect, a communication system is provided, which includes the aforementioned first device and second device.

[0057] In a seventh aspect, a computer-readable storage medium is provided, including a computer program that, when executed on a computer, causes the computer to implement the method in any of the possible implementations of the first or second aspect.

[0058] Eighthly, a computer program product is provided, the computer program product comprising: a computer program (also referred to as code or instructions), which, when run, causes a computer to perform the method in any possible implementation of the first or second aspect.

[0059] The beneficial effects of the second to eighth aspects and the possible implementations described above can be found in the first aspect and the beneficial effects of the various possible implementations of the first aspect, and will not be repeated here. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of a communication system provided in an embodiment of this application;

[0061] Figure 2 This is a schematic diagram of the application framework of the communication system provided in the embodiments of this application;

[0062] Figure 3 This is a schematic flowchart of a communication method provided in an embodiment of this application;

[0063] Figure 4 This is a schematic diagram illustrating the generation of the first signal provided in an embodiment of this application;

[0064] Figure 5 This is a schematic diagram illustrating the acquisition of signals corresponding to the second device provided in an embodiment of this application;

[0065] Figure 6 This is a schematic flowchart illustrating yet another communication method provided in an embodiment of this application;

[0066] Figure 7 This is a schematic diagram illustrating the use of the pairing history information table provided in the embodiments of this application;

[0067] Figure 8 This is a schematic block diagram of the communication device provided in the embodiments of this application;

[0068] Figure 9 This is another schematic block diagram of the communication device provided in the embodiments of this application. Detailed Implementation

[0069] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0070] First, in this application, the indication includes explicit indication (also known as direct indication) and implicit indication (also known as indirect indication). Explicit indication information A means including information A; implicit indication information A means indicating information A through the correspondence between information A and information B, and direct indication information B. The correspondence between information A and information B can be predefined, pre-stored, pre-burned, or pre-configured; or it can refer to indicating information A through information B and preset rules.

[0071] Second, in this application, information C is used to determine information D, which includes both determining information D based solely on information C and determining it based on information C and other information. Furthermore, information C can also be used to determine information D indirectly, for example, in the case where information D is determined based on information E, and information E is determined based on information C.

[0072] Third, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship; the specific meaning can be understood in context. "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 can mean: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0073] Fourth, the use of prefixes such as "first" and "second" in this application is solely for the purpose of distinguishing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first equipment" and "second equipment" are simply different pieces of equipment, and there is no temporal, size, or priority relationship between them. Similarly, "first instruction information," "second instruction information," "third instruction information," "fourth instruction information," and "fifth instruction information" are simply different instructions, and there is no temporal, size, or priority relationship among them.

[0074] Fifth, in the embodiments of this application, "when," "if," and "if" all refer to the device making corresponding processing under certain objective circumstances, and are not limited to a time, nor do they require the device to make a judgment action when it is implemented, nor do they mean that there are other limitations.

[0075] The technical solutions provided in this application can be applied to various communication systems, such as: Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, sidelink (SL) communication systems, Universal Mobile Telecommunication System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) communication systems, 5th Generation (5G) mobile communication systems, or new radio access technology (NR). Among these, 5G mobile communication systems can include non-standalone (NSA) and / or standalone (SA) networking. The technical solutions provided in this application can also be applied to future communication networks. This application does not limit the scope of the application in this regard.

[0076] Figure 1 A schematic diagram of a communication system 100 according to an embodiment of this application is shown, as follows: Figure 1 As shown, the communication system 100 may include: at least one network device (e.g., network device 110) and multiple terminal devices (e.g., terminal devices a to j). The network device 110 may be connected to... Figure 1 The terminal device shown performs uplink and downlink transmission.

[0077] It should be understood that the aforementioned network device or terminal device can be configured with multiple antennas, which may include at least one transmitting antenna for transmitting signals and at least one receiving antenna for receiving signals. Additionally, the network device or terminal device may also include transmitter chains and receiver chains, which, as will be understood by those skilled in the art, may include multiple components (e.g., processors, modulators, multiplexers, demodulators, demultiplexers, or antennas) related to signal transmission and reception. Therefore, the network device and the terminal device can communicate via multi-antenna technology.

[0078] It should be understood that Figure 1 The communication system shown is only a schematic diagram. The communication system may also include other terminal devices and network devices, such as wireless relay devices and wireless backhaul devices. Figure 1 The number of network devices and terminal devices included in the communication system is not shown in the embodiments of this application.

[0079] In the embodiments of this application, the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user apparatus.

[0080] Terminal devices can be devices that provide voice / data, such as handheld devices with wireless connectivity, in-vehicle devices, etc. Currently, examples of terminals include: mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, wearable devices, terminal devices in 5G networks, or future public land mobile communication networks. Terminal devices in a network (PLMN), etc., are not limited to this in the embodiments of this application.

[0081] By way of example and not limitation, in this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0082] In this embodiment, the device for implementing the functions of the terminal device can be the terminal device itself, or it can be any device capable of supporting the terminal device in implementing those functions, such as a chip system. This device can be installed in or used in conjunction with the terminal device. In this embodiment, the chip system can be composed of chips or may include chips and other discrete components. This embodiment only uses the terminal device as an example to illustrate the device for implementing the functions of the terminal device, and does not constitute a limitation on the solution of this embodiment.

[0083] The network device in this application embodiment can be a device for communicating with a terminal device. This network device can also be called an access network device or a wireless access network device, such as a base station. In this application embodiment, the network device can refer to a radio access network (RAN) node (or device) that connects the terminal device to the wireless network. A base station can broadly encompass, or be replaced by, various names including: NodeB, evolved NodeB (eNB), next-generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master station, auxiliary station, multi-standard radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, satellite base station, cellular base station, etc. A base station can be a macro base station, micro base station, relay node, donor node, or similar entities, or combinations thereof. A base station can also refer to a communication module, modem, or chip installed within the aforementioned equipment or apparatus. A base station can also be a mobile switching center, equipment performing base station functions in D2D, V2X, and M2M communications, network-side equipment in future communication networks, or equipment performing base station functions in future communication systems. A base station can support networks using the same or different access technologies. Optionally, a 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). The embodiments of this application do not limit the specific technologies or equipment forms used in the network equipment.

[0084] In some deployments, the network devices mentioned in the embodiments of this application may be devices including CU, DU, or CU and DU, or devices with control plane CU nodes (central unit-control plane (CU-CP)) and user plane CU nodes (central unit-user plane (CU-UP)) and DU nodes. For example, the network devices may include gNB-CU-CP, gNB-CU-UP, and gNB-DU.

[0085] In some deployments, 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 CUs, DUs, CU-CPs, CU-UPs, or RUs. CUs and DUs can be configured separately or included in the same network element, such as a BBU. RUs can be included in radio frequency equipment or radio frequency units, such as RRUs, AAUs, or RRHs.

[0086] RAN nodes can support one or more types of fronthaul interfaces, each corresponding to a DU and RU with different functions. If the fronthaul interface between the DU and RU is a common public radio interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and RU is another type of interface, relative to CPRI, some downlink and / or uplink baseband functions, such as, for downlink, precoding, digital beamforming (BF), or one or more of inverse fast Fourier transform (IFFT) / cyclic prefix addition (CP), are moved from the DU to the RU; and for uplink, digital beamforming (BF), or one or more of fast Fourier transform (FFT) / cyclic prefix removal (CP), are moved from the DU to the RU. In one possible implementation, the interface can be an enhanced common public radio interface (eCPRI). Under the eCPRI architecture, the segmentation between DU and RU differs, corresponding to different categories (Cat) of eCPRI, such as eCPRI Cat A, B, C, D, E, F.

[0087] Taking eCPRI Cat A as an example, for downlink transmission, layer mapping is used as the dividing line. DU is configured to implement one or more functions preceding layer mapping (i.e., coding, rate matching, scrambling, modulation, and layer mapping), while other functions following layer mapping (e.g., RE mapping, digital beamforming (BF), or one or more inverse fast Fourier transform (IFFT) / cyclic prefix (CP) addition) are moved to RU. For uplink transmission, deRE mapping is used as the dividing line. DU is configured to implement one or more functions preceding deRE mapping (i.e., decoding, rate matching de-matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and deRE mapping), while other functions following deRE mapping (e.g., digital BF or fast Fourier transform (FFT) / CP removal) are moved to RU. It is understandable that the functional descriptions of the DU and RU corresponding to various types of eCPRI can be found in the eCPRI protocol, and will not be elaborated here.

[0088] In one possible design, the processing unit in the BBU used to implement baseband functions is called the baseband high (BBH) unit, and the processing unit in the RRU / AAU / RRH used to implement baseband functions is called the baseband low (BBL) unit.

[0089] In different systems, CU (or CU-CP and CU-UP), DU, 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. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0090] For example, Figure 2 A schematic diagram of the application framework of the communication system is shown. For example... Figure 2As shown, the network elements in this communication system may include core network equipment, access network nodes (RAN nodes), terminals, and operations, administration, and maintenance (OAM) systems. The network elements in the communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These network element nodes, such as core network equipment, access network nodes, terminals, or one or more OAM devices, may be equipped with one or more AI modules (for clarity, ...). Figure 2 (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 a CU and a DU. A CU and / or DU can also have one or more AI modules configured. Optionally, a CU can also be split into a CU-CP and a CU-UP. One or more AI models are configured in the CU-CP and / or CU-UP.

[0091] The AI ​​module is used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the parameter configuration, the AI ​​module can achieve different functions. The AI ​​module model 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 bias in the activation function), input parameters (e.g., type and / or dimension of input parameters), or output parameters (e.g., type and / or dimension of output parameters). The bias in the activation function can also be referred to as the neural network bias.

[0092] 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.

[0093] In this embodiment, the apparatus for implementing the functions of a network device can be a network device itself; it can also be an apparatus capable of supporting the network device in implementing those functions, such as a chip system, hardware circuit, software module, or a hardware circuit plus a software module. This apparatus can be installed in the network device or used in conjunction with the network device. In this embodiment, the example of a network device being used to implement the functions of a network device is provided only and does not constitute a limitation on the solutions described in this embodiment.

[0094] Network devices and / or terminal devices can be deployed on land, 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. Furthermore, terminal devices and network devices can be hardware devices, or software functions running on dedicated hardware, or software functions running on general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities that include dedicated or general-purpose hardware devices and software functions. This application does not limit the specific form of the terminal devices and network devices.

[0095] Multiple access technology aims to enable multiple user equipment to efficiently share wireless resources by dividing the domain into time, frequency, code, or spatial domains to ensure that signals do not interfere with each other. Its core objective is to increase system capacity, reduce latency, and support diverse service requirements with limited resources.

[0096] Traditional 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). FDMA divides channels by frequency, with each user having exclusive access to a specific frequency band. TDMA allocates resources by time slot, with users taking turns occupying the same frequency band. CDMA distinguishes users using orthogonal codes, allowing simultaneous transmission on the same frequency. OFDMA divides the frequency band into orthogonal subcarriers and dynamically allocates subcarrier resources. NOMA uses power domain superposition of multiple user signals, combined with serial interference cancellation decoding.

[0097] As communication demands evolve towards high density, low latency, and semantic awareness, new technologies focus on resource reuse efficiency and semantic information fusion, with representative technologies including RSMA, MDMA, and SFMA. The core idea of ​​RSMA is to split the transmitted message into two parts at the transmitting end, called the "private part" and the "public part," and then merge the "public part" into a whole, transmitting it along with the "private part" within the same time-frequency resources. The core idea of ​​MDMA is that the network side uses different encoders to extract semantic information from multiple terminal devices. By comparing the semantic information from multiple terminal devices, the network side obtains common semantics and personalized semantics. The former is transmitted using the same time-frequency resources, while the latter is transmitted orthogonally. The terminal devices receive the common semantics and their respective personalized semantics and use decoders to reconstruct the signal. The core idea of ​​SFMA is that the network side uses different encoders to extract semantic information from multiple terminal devices. The network side superimposes the semantic information from multiple terminal devices and transmits it using the same time-frequency resources. The terminal devices receive the aliased semantic information and use decoders to reconstruct the signal.

[0098] In current multiple access technology systems, the network side typically allocates power resources to each user according to pre-defined rules. However, these pre-defined rules are difficult to accurately adapt to different users. Especially in large-scale user access scenarios, as the number of users increases rapidly, the network side needs to allocate power to a large number of users simultaneously. Inappropriate power allocation methods (e.g., allocating too much or too little power to some users) prevent the efficient use of power resources, thus severely restricting the efficiency of multiple access.

[0099] In view of this, this application proposes a communication method in which a network device can divide multiple terminal devices into a multiple access user group (MIBG), and preprocess the data to be transmitted corresponding to each terminal device in the MIBG based on the same network-side model, thereby outputting the data (or data features) corresponding to each terminal device. Since the computing power of each terminal device varies, the network device can allocate corresponding power to each terminal device based on its computing power. In this way, the network device can send the data and power corresponding to each terminal device to each terminal device based on the network-side model. Thus, even in large-scale scenarios, the network device can reasonably allocate power to terminal devices, which is beneficial to improving the efficiency of multiple access.

[0100] The method provided in this application will now be described in detail with reference to the accompanying drawings. It should be understood that the technical solution of this application can be applied to, for example... Figure 1 The communication system shown.

[0101] It should be understood that the following description is for ease of understanding and explanation only, and uses the interaction between the first device and the second device as an example to explain in detail the method provided in the embodiments of this application.

[0102] The first device may be, for example, a network device; the second device may be, for example, a terminal device. For example, the first device may correspond to... Figure 1 Network device 110 in the middle, the second device can correspond to Figure 1 Any one of the terminal devices 120a to 120j in the terminal device.

[0103] However, it should be understood that this should not limit the entity executing the methods provided in this application. Any entity capable of executing the methods provided in this application can do so by running a program containing code for the methods described in the embodiments of this application. For example, the first device shown in the following embodiments can be replaced by components of that first device, such as a chip, a chip system, or other functional modules capable of calling and executing programs. The second device can also be replaced by components of that second device, such as a chip, a chip system, or other functional modules capable of calling and executing programs.

[0104] The method provided in this application will now be described in detail with reference to the accompanying drawings.

[0105] Figure 3 A communication method 300 provided in an embodiment of this application is illustrated. The method 300 includes steps 310 to 330. The various steps in method 300 are described in detail below.

[0106] In step 310, the first device determines the power corresponding to each of the N second devices. The power corresponding to each second device is determined based on the computing power of the second device, where N is an integer greater than 1.

[0107] In this application, the computing power of each of the N second devices is different, so the power corresponding to each second device is different.

[0108] The first device can determine the power of each of the N second devices based on their computing power. For example, the relationship between the power of a second device and its computing power can be negatively correlated: the second device with stronger computing power will have lower power, because it can process signals more effectively and complete signal processing and detection tasks well even with lower power. Conversely, the second device with weaker computing power will have higher power, because it needs more power to compensate for its insufficient computing power to ensure it can process signals normally, thereby improving the overall system performance and stability.

[0109] For example, the first device can sort the N second devices in ascending order of computing power based on the computing power of the second devices. For instance, according to the arrangement order of second device #1, second device #2, second device #3, ..., second device #N, as the number after # of the second device gradually increases, the computing power also increases. There is no limitation on this.

[0110] The sum of the power of each of the above N second devices is less than or equal to the preset power value (i.e., p). max For example, the power p1 corresponding to the second device #1, the power p2 corresponding to the second device #2, the power p3 corresponding to the second device #3, ... and the power p corresponding to the second device #N. N The sum is less than or equal to p max And p1≥p1≥p1≥…≥p N .Right now Where k is 1, 2, 3, ..., N, p k The power corresponding to each second device, p max The maximum sum of power allocated to the first device from the N second devices.

[0111] The first device can process the data (or data to be transmitted) to each of the N second devices, obtaining the data corresponding to each second device. During the processing of this data to be transmitted, the data corresponding to the second device with lower computing power (e.g., second device #k) may be interfered with by the signal of the second device with higher computing power (e.g., second device #k+1, ..., second device #N). In this case, the signal-to-interference-plus-noise ratio (SINR) of each second device can be calculated as follows:

[0112]

[0113] Where, p k (e.g., k is 1, 2, 3, ..., N) is the power allocated to the second device #k, h k These are parameters such as channel gain related to the second device #k. This represents the total power of the interference signals from devices #k+1 to #N. It is the noise power on the #k side of the second device, SINR. k The higher the value, the better the signal transmission quality. Where h... k , These are the parameter values ​​determined by the first device through channel estimation.

[0114] As mentioned above, the first device determines or allocates power p to each of the second devices. kThe power allocation problem for each of the N second devices differs from the communication performance of the N second devices. To optimize the communication performance of the N second devices in a multiple access user group, the power allocation problem for each of the N second devices can be optimized using the following objective expression:

[0115]

[0116] Wherein, log(1+SINR) k The value reflects the user's rate, so the optimization objective expression is to maximize the sum of the rates of N users, i.e., the system throughput.

[0117] The optimization problem constraints satisfied by the above objective expression are as follows: SINR k ≥α, p1≥p1≥p1≥…≥p N SINR k ≥α indicates that the SINR of each of the N second devices when receiving data at the corresponding power is greater than or equal to the threshold α. This threshold is used to determine whether the second device can obtain the corresponding signal from the signal processed by the first device (e.g., the first signal mentioned in step 320). For example, whether the second device can obtain the corresponding signal from the first signal using the third AI model in step 330. For instance, the first device can obtain the value of the threshold α by testing the performance of the first AI model and the third AI model under different SINR values. Regarding p1≥p1≥p1≥…≥p N and p max As mentioned earlier, it will not be repeated here.

[0118] It should be understood that when the signal-to-interference-plus-noise ratio (SIR) is greater than or equal to the threshold, it means that the interference and noise affecting the signal during transmission are within an acceptable range, and the third AI model can more clearly identify and analyze the feature information in the first signal. Although the third AI model has relatively fewer parameters, under good SIR conditions, it can still extract the signal corresponding to each second device from the first signal, thus ensuring the accuracy of signal acquisition.

[0119] Based on the aforementioned optimization objective expression and optimization problem constraints, the first device can select the final power allocation scheme from a preset power allocation library. There can be multiple preset power allocation libraries, each with a different ID. For example, the preset power allocation libraries with IDs as shown in Table 1 are as follows:

[0120] Table 1

[0121] Solution ID Power distribution 0 <![CDATA[0.1p max ,0.9p max ]]> 1 <![CDATA[0.2p max ,0.8p max ]]> 2 <![CDATA[0.3p max 0.7p max ]]> 3 <![CDATA[0.4p max ,0.6p max ]]>

[0122] It should be understood that in some possible implementations, the first device can directly obtain the power allocation result corresponding to the second device based on the above optimization objective expression and optimization problem constraints.

[0123] In some embodiments, the first device may also determine the power of each of the N second devices based on the difficulty of the task performed by each of the second devices. The task of the second device may be, for example, localization, channel estimation, or image generation. This is not limited.

[0124] For example, the power of a second device and the difficulty of its task can be negatively correlated. For instance, among N second devices, the more difficult the task, the lower the power corresponding to that device. This is because a difficult task likely means the second device needs to process a large amount of data. Allocating less power to it is beneficial because even if a large power were allocated to a device with a difficult task, interference from other devices would hinder a significant increase in data processing speed. In this case, the device with a difficult task can gradually eliminate interference from other devices to acquire its corresponding data. Conversely, a second device with an easier task receives a higher power, as an easier task indicates a smaller amount of data to process. Allocating more power to it helps it overcome interference during data processing to some extent, allowing the device with an easier task to acquire its corresponding data more quickly.

[0125] In one possible scenario, if two of the N second devices have comparable computing power, the first device can determine the power of each second device based on the difficulty of the task performed by each of the N second devices.

[0126] In this application, the first device can obtain the computing power of the second device in either the second device actively reporting to the first device, or the first device instructing the second device to report to the first device; there is no limitation on this.

[0127] In some embodiments, when the second device actively reports its computing power information to the first device, the first device can directly receive the computing power information of the second device.

[0128] For example, the second device sends a second indication message to the first device, the second indication message being used to indicate the computing power of the second device. Correspondingly, the first device receives the second indication message from the second device.

[0129] Taking N second devices, including second device #1 and second device #2, as an example, Figure 6In steps 603 and 604, in step 603, the second device #1 sends second indication information to the first device. Correspondingly, the first device receives the second indication information from the second device #1. In step 604, the second device #2 sends second indication information to the first device. Correspondingly, the first device receives the second indication information from the second device #2. The second indication information may include the computing power information of the second device, which may include at least one of TOPS, memory size, video memory size, CPU utilization, and GPU utilization. No limitation is imposed on this.

[0130] In some other embodiments, when the first device sends an instruction to the second device, instructing the second device to report its computing power information, the second device, upon receiving the instruction, sends its computing power information back to the first device.

[0131] For example, the second device sends a third indication message to the first device, the third indication message being used to indicate the sending of the second indication message (i.e., computing power information) by the second device. Correspondingly, the first device receives the third indication message from the second device.

[0132] Taking N second devices, including second device #1 and second device #2, as an example, Figure 6 In steps 601 and 602, in step 601, the first device sends third indication information to the second device #1. Correspondingly, the second device #1 receives the third indication information from the first device. In step 602, the first device sends third indication information to the second device #2. Correspondingly, the second device #2 receives the third indication information from the first device.

[0133] Optionally, after the first device determines the computing power of each of the N second devices based on the computing power of the N second devices, it can also send the power corresponding to each second device to the second device so that the second device can obtain the corresponding signal from the first signal based on the corresponding power.

[0134] One implementation involves the first device sending a fifth indication message to each of the N second devices, whereby the fifth indication message indicates the power corresponding to each second device.

[0135] Taking N second devices, including second device #1 and second device #2, as an example, Figure 6 In steps 609 and 610, in step 609, the first device sends a fifth indication message to the second device #1. Correspondingly, the second device #1 receives the fifth indication message from the first device. In step 610, the first device sends a fifth indication message to the second device #2. Correspondingly, the second device #2 receives the fifth indication message from the first device.

[0136] It should be understood that the first device may carry the fifth indication information in the common downlink control information (DCI).

[0137] Another implementation method is that the first device sends the fifth instruction information to the second device with the stronger computing power among N second devices.

[0138] Taking N second devices, including second device #1 and second device #2, as an example, second device #2 has a higher computing power than second device #1. The more powerful second device #2 has a lower power consumption. In the subsequent signal acquisition process, when second device #2 needs to acquire its own signal, it must first acquire the signal from second device #1. Therefore, second device #2 needs to acquire the power consumption corresponding to second device #1 and the power consumption corresponding to second device #2. In this case, the first device only sends the fifth indication information to second device #2. This fifth indication information can be carried in the DCI sent by the first device to second device #2.

[0139] Another implementation involves the first device sending a fifth indication message to each of the N second devices. This fifth indication message is used to indicate the power allocation library ID and the scheme ID.

[0140] Taking N second devices, including second device #1 and second device #2, as an example, after second device #1 and second device #2 receive the fifth instruction information, they can find the corresponding scheme in the corresponding power allocation library.

[0141] In step 320, the first device sends a first signal to the second devices based on the power and data corresponding to each of the N second devices, where the data corresponding to each second device is determined based on a first AI model. Correspondingly, the second devices receive the first signal from the first device.

[0142] During the data transmission process from the first device to each of the N second devices, the data to be transmitted is input into the first AI model for processing to obtain output data. Then, based on the output data and the power determined in step 310 above, a first signal is generated and sent. To distinguish between the data input to the first AI model and the data output by the first AI model, the input data is referred to as the first data corresponding to each second device, and the output data is referred to as the second data corresponding to each second device.

[0143] It should be noted that this application is not limited to inputting the first data corresponding to each second device into the first AI model for processing to obtain the second data corresponding to that second device. For example, it is also possible to simultaneously input N first data corresponding to N second devices into the first AI model for processing to obtain N second data corresponding to N second devices.

[0144] In this application, the first device can send a first signal based on the power and data corresponding to each second device. This can include: the first device generating a first signal based on the power and data corresponding to each second device. In other words, the first signal can include the power and data corresponding to each of the N second devices. For example, the first signal = data corresponding to second device #1 * power corresponding to second device #1 + ... + data corresponding to second device #N * power corresponding to second device #N.

[0145] Taking N as 2, meaning that two second devices (e.g., second device #1 and second device #2) belong to the same multiple access user group, Figure 4 A schematic diagram of the generation of the first signal is shown, as follows. Figure 4 As shown, x represents the data to be transmitted corresponding to the second device #1, and y represents the data to be transmitted corresponding to the second device #2.

[0146] The first device can first preprocess x and y in the first AI model and output x' and y', where x' is the data corresponding to the second device #1 and y' is the data corresponding to the second device #2. Next, based on the computing power of the second devices #1 and #2 (the computing power of the second device #1 is lower than that of the second device #2), the first device allocates power to the second devices #1 and #2, i.e., the power corresponding to the second device #1 is p1, and the power corresponding to the second device #2 is p2, with p1 being greater than p2. Based on p1 and x' of the second device #1 and p2 and y' of the second device #2, the first device determines or derives p1x'+p2y', which is the first signal. Based on this, the first device can send p1x'+p2y' to the second devices #1 and #2.

[0147] The first AI model can extract data features of the data to be transmitted for each of the N second devices. For example, when the data to be transmitted for a second device is an image, the first AI model can extract features such as the image's edges and color distribution during the transmission of the image. This allows the model to represent the original image with a smaller amount of data, thereby reducing the amount of data transmitted, increasing transmission speed, and saving network bandwidth resources.

[0148] The power of each of the N second devices has been described in detail in step 310, and will not be repeated here.

[0149] The aforementioned first AI model can also be called a public model, without any limitation.

[0150] Optionally, the signals corresponding to different second devices in the first signal are carried on the same resource, which may include the same time domain, frequency domain, spatial domain, or codeword, etc. No limitation is imposed in this regard.

[0151] In this application, the first AI model is a large model, and its output can be defined as a token. The first device and the second device interact using the token. Therefore, the first signal can include one or more tokens. The token can also be called a mark, term, or token, etc. The naming of the token is not limited in the embodiments of this application.

[0152] Optionally, a token can be understood as the basic unit processed by the first AI model. For example, a token can be the basic unit of text, data, or features processed by the first AI model. Alternatively, the first AI model processes data using tokens as the basic unit; for example, text, data, or features processed by the AI ​​model can be represented as tokens for processing and transmission. Or, the input and output of the first AI model consist of a series of numbers or vectors, which can be called tokens, or these numbers or vectors can be represented as tokens.

[0153] Regarding the definition of a token, it can be understood as a basic unit of model processing. For example, a token can be a basic unit of AI model processing. For example, a token can be understood as a basic unit of text, data, or features in the AI ​​model processing process. Optionally, "token as a basic unit of model processing" can be understood as the model processing data in token form as the basic unit. For example, text, data, or features in the AI ​​model processing process can be processed and transmitted in token form. For example, the input and output of an AI model consist of a series of numbers or vectors, which can be called tokens, or these numbers or vectors can be represented in token form. Optionally, a token can also be called a mark, term, or token, etc. This application embodiment does not limit the naming of tokens.

[0154] Optionally, before the first device sends a first signal to the N second devices based on the power and data corresponding to each of the N second devices, the first device may first determine or select a model on its own side and a model on the side of the second device. The model on the first device's side is used to process the first data corresponding to each of the N second devices. The N second devices may include, for example, second device #1 and second device #2. This is not limited.

[0155] For example, Figure 6 In step 605, the first device can determine the first AI model and the third AI model.

[0156] The first AI model is the model on the first device side, and the third AI model is the model on the second device side. The third AI model can also be understood as a low-rank model. No specific limitations are imposed.

[0157] The first data corresponding to the second device #1 conforms to the input characteristics of the first AI model. Therefore, the first device can determine that the first AI model is the model for processing the first data corresponding to the second device #1. This first AI model is used to process the first data corresponding to the second device #1 and output the second data. Similarly, the first data corresponding to the second device #2 conforms to the input characteristics of the first AI model. Therefore, the first device can determine that the first AI model is the model for processing the first data corresponding to the second device #2. This first AI model is used to process the first data corresponding to the second device #2 and output the second data.

[0158] The second device can obtain the corresponding signal from the first signal based on the AI ​​model. Therefore, the computing power of the second device should meet the size requirements of the AI ​​model, and the first data corresponding to the second device should meet the output requirements of the AI ​​model. For example, the first device can determine the third AI model as the AI ​​model on the side of the second device #1 based on the first data corresponding to the second device #1 and the computing power of the second device #1. The second device #1 then obtains the corresponding signal from the first signal based on the third AI model. Similarly, the first device can determine the third AI model as the AI ​​model on the side of the second device #2 based on the first data corresponding to the second device #2 and the computing power of the second device #2. The second device #2 then obtains the corresponding signal from the first signal based on the third AI model.

[0159] Optionally, when the first device determines the power corresponding to each of the N second devices, the method further includes: the first device performing user pairing on the N second devices.

[0160] For example Figure 6 In step 606, the first device can perform user pairing and power pairing for the paired users based on computing power.

[0161] The aforementioned user pairing may include steps such as data pairing, model pairing, computing power pairing, and power pairing. This user may include N second devices, such as second device #1 and second device #2; of course, other second devices may also be included, without limitation.

[0162] The first data corresponding to the second device #1 and the first data corresponding to the second device #2 are related, such as having the same type, format, dimension, or modality. The first device can process both simultaneously. For example, if the first data corresponding to the second device #1 and the first data corresponding to the second device #2 conform to the input characteristics of the same first AI model, the first device can select the same first AI model to process both. For instance, if both the first data corresponding to the second device #1 and the first data corresponding to the second device #2 are image data and require image recognition processing, the first device will select a first AI model whose input adapts to the image format, whose output meets the recognition result requirements, whose functionality includes image recognition capability, and whose performance meets the processing speed requirements. In this way, there is no need to configure separate adapted models for the second device #1 and the second device #2, or in other words, there is no need to process and send the first data corresponding to the second device #1 and the first data corresponding to the second device #2 one by one, thereby significantly shortening the overall processing time and improving processing efficiency.

[0163] Based on this, the second device #1 and the second device #2 can form a data pairing user group.

[0164] In the above data pairing user group, if the third model used by the second device #1 and the third model used by the second device #2 are the same third AI model, then the second device #1 and the second device #2 further form a model pairing user group; if the third model used by the second device #1 and the second device #2 are not the same third AI model, then the second device #1 and the second device #2 can use other multiple access methods (such as TDMA, FDMA, CDMA, etc.) to access the network.

[0165] The first AI model is a relatively complex model with a large computational load. The distilled version of the first AI model is a simplified version to reduce computational requirements. Therefore, among the second devices #1 and #2, the second device #1, with the lowest computing power, may not be able to run the first AI model or its distilled version. Thus, in the above model-paired user group, at most one second device (such as the second device #1) may not have the computing power to run either the first AI model or its distilled version.

[0166] When both the computing power of device #1 and device #2 supports running the first AI model or a distilled version of the first AI model, or when only device #1 with lower computing power is unable to run the first AI model or a distilled version of the first AI model, device #1 and device #2 can form a computing power paired user group. When neither the computing power of device #1 nor device #2 can run the first AI model or a distilled version of the first AI model, device #1 and device #2 need to use other multiple access methods to access the network.

[0167] In the aforementioned power-paired user group, the first device can allocate corresponding power to the second device #1 and the second device #2 with different computing power based on the optimization objective expression and optimization problem constraints in step 310. In this way, the second device #1 and the second device #2 can form a power-paired user group and can perform first multiple access based on the power allocation. However, if the power allocation of the second device #1 and the second device #2 does not meet the optimization objective expression and optimization problem constraints in step 310, then the second device #1 and the second device #2 can use other multiple access methods (such as TDMA, FDMA, CDMA, etc.) to access the network, or the second device (second device #1 or second device #2) that does not meet the optimization objective expression and optimization problem constraints can be removed. No restrictions are imposed on this.

[0168] As mentioned above, the second device #1 and the second device #2 become devices in the same user group (or the same multiple access user group) after data pairing, model pairing, computing power pairing, and power pairing. Therefore, the first device can generate a mapping relationship corresponding to this user group. This mapping relationship can include information about the first AI model, information about the third AI model, information about each second device, and a power preset value p. max The mapping relationship between user groups can be referred to as group information or pairing history information for that user group.

[0169] The power preset value has been described in detail in step 310, which can be found in the relevant description in step 310. It will not be repeated here.

[0170] The aforementioned first AI model information may include the identification (ID) or index of the first AI model, and the information in the third AI model information may include the ID or index of the third AI model, without limitation.

[0171] For example, Table 2 shows the group information for the current version of the user group. As shown in Table 2, the group information includes the group ID, the first AI model ID, the third AI model ID, and p. max And the ID of the second device (such as second device #1, second device #2, etc.).

[0172] Table 2

[0173] Group ID First AI Model ID Third AI Model ID <![CDATA[p max ]]> Second device #1ID Second device #2ID … 0 1 4 5 2 5 1 4 5 2 1 2 … … … … … …

[0174] In some embodiments, the first device may pair multiple second devices that are simultaneously performing multiple access, such as pairing the multiple second devices into one or more multiple access user groups, and generating a mapping relationship for the one or more multiple access user groups. This is not limited to any particular method.

[0175] Figure 7 The diagram illustrates the use of the pairing history information table. When multiple second devices are accessing the network via multiple access points, and the first device is pairing users with these multiple second devices, the following steps can be followed:

[0176] In step 710, the first device can query the pairing history information table.

[0177] If the pairing history information table contains records for all the second devices in multiple second devices, proceed to step 720 to reuse the pairing history information.

[0178] If there are records for a portion of the second devices among multiple second devices in the pairing history information table, proceed to step 720 for the portion of second devices with records, and reuse the pairing history information.

[0179] If there are no relevant records for some or all of the multiple second devices in the pairing history information table, proceed to steps 730 to 770 for the second devices without records, as shown below.

[0180] In step 730, the first device performs data pairing with the second device.

[0181] In step 740, the first device performs model pairing with the second device.

[0182] In step 750, the first device performs computing power pairing with the second device.

[0183] In step 760, the first device performs power pairing with the second device.

[0184] Steps 730 to 760 have been described in detail in step 606 and will not be repeated here.

[0185] In step 770, the first device updates or adds a pairing history information table.

[0186] The first device can update or add records to the pairing history information table shown in Table 2. For example, it can add the second device #10 with ID 9 to the group with group ID 0, or add the group with group ID 12, which includes the first AI model with ID 6 and the third AI model with ID 10. max The ID is 5, the ID of the second device #1 is 5, the ID of the second device #2 is 6, and so on.

[0187] Optionally, after the first device completes user pairing for N second devices, it can also deploy the model corresponding to each of the N second devices to the corresponding second device. This model is determined by the first device based on the computing power of the N second devices.

[0188] One possible implementation is that the first device sends first instruction information to each of the N second devices. This first instruction information is used to configure the model corresponding to each second device. The model may include a third AI model, or a second AI model and a third AI model. The second model may be the first AI model or a distilled version of the first model. The N second devices include, for example, second device #1 and second device #2. This is not limited.

[0189] For example, Figure 6 In steps 611 and 612, in step 611, the first device sends first instruction information to the second device #1. The first instruction information is used to configure the model corresponding to the second device #1 as the third AI model. Correspondingly, the second device #1 receives the first instruction information from the first device.

[0190] On the one hand, compared to the second device #2, the second device #1 has lower computing power, and the third AI model is relatively simple, requiring less computing power. On the other hand, the second device #1 has higher power, which helps it overcome noise and other interference factors in signal acquisition or detection to some extent.

[0191] In step 612, the first device sends first instruction information to the second device #2. The first instruction information is used to configure the model corresponding to the second device #2 as a second AI model and a third AI model. Correspondingly, the second device #2 receives the first instruction information from the first device.

[0192] Compared to device #1, device #2 has stronger computing power and can perform more complex signal acquisition or detection operations. However, device #2 has lower power consumption, so it may not be able to overcome noise and other interference in signal acquisition or detection. Therefore, device #2 can use the third AI model and the second AI model to reconstruct the interference signals from device #1 that have lower computing power (but higher power). This means it can analyze and reconstruct the signal from device #1 using these two models to obtain the characteristics and information of the interference signal. Next, through a subtraction operation, the reconstructed interference signal is subtracted from the received signal to obtain its own signal, which is then output by the second AI model. Finally, the third AI model is used to detect the obtained signal to obtain the final signal it needs.

[0193] Optionally, after the first device completes user pairing with N second devices, it can also send a fourth indication message to each of the N second devices. This fourth indication message instructs the second device to receive the first signal based on the third AI model. Alternatively, the fourth message instructs the second device to perform first multiple access, indicating that each of the N second devices is capable of performing first multiple access. Here, first multiple access can be understood as sending a first signal when the first device allocates power to the second devices based on their computing power. The first signal is generated by the first device based on the power and data corresponding to each second device.

[0194] For example, Figure 6 In steps 607 and 608, in step 607, the first device sends a fourth indication message to the second device #1. Correspondingly, the second device #1 receives the fourth indication message from the first device. In step 608, the first device sends a fourth indication message to the second device #2. Correspondingly, the second device #2 receives the fourth indication message from the first device.

[0195] In step 330, the second device obtains the signal corresponding to the second device from the first signal using a third AI model based on the first signal, or obtains the signal corresponding to the second device from the first signal using both the second AI model and the third AI model. The second AI model is the first AI model or is obtained by distillation based on the first AI model.

[0196] It should be understood that the model obtained by distillation based on the first AI model can also be called a distilled version of the first AI model, a lightweight version, etc., without any limitation.

[0197] Example 1: The second device obtains its corresponding signal from the first signal based on the third AI model.

[0198] On the one hand, compared to the second device in the same multiple access user group, the second device in this example has lower computing power, and the third AI model is relatively simple, requiring less computing power. On the other hand, the second device in this example has higher power, which helps it overcome noise and other interference factors in signal acquisition or detection to some extent.

[0199] Example 2: The second device obtains its corresponding signal from the first signal based on the second AI model and the third AI model.

[0200] Compared to the second device in the same multiple access user group, the second device in this example has stronger computing power and can perform more complex signal acquisition or detection operations. However, the power of the second device in this example is relatively low, so it may not be able to overcome noise and other interference in signal acquisition or detection. The second device in this example can use the third AI model and the second AI model to reconstruct the interference signals of second devices with lower computing power (but higher power) than itself (i.e., the second devices in the same multiple access user group). This means that they analyze and reconstruct the signal of the second device in this example through these two models to obtain the characteristics and information of the interference signal. Secondly, through a subtraction operation, the reconstructed interference signal is subtracted from the received signal to obtain its own signal, which is then output by the second AI model. Finally, the third AI model is used to detect the obtained signal to obtain the final signal it needs.

[0201] This signal acquisition or detection method can be applied in scenarios involving multi-device communication or various types of signal interference. It works by sequentially detecting and eliminating signal interference from each device. Specifically, the receiving device first detects and recovers the strongest signal (or the signal from the device with the weakest computing power but the highest power). After recovery, this signal is subtracted from the received signal to eliminate its interference with other signals. This process is then repeated for the next strongest signal (or the signal from the device with the second weakest computing power but the second highest power), and so on, until all signals have been processed. This signal detection method effectively improves the receiving device's signal resolution capabilities, increasing system capacity and communication quality.

[0202] It should be understood that since the N second devices are devices in the same multi-access user group, the third AI model corresponding to each of the N second devices is the same.

[0203] Taking N as 2, meaning that two second devices (e.g., second device #1 and second device #2) belong to the same multiple access user group, Figure 5 A schematic diagram illustrating the acquisition of the signal corresponding to the second device is shown, such as... Figure 5As shown, after receiving p1x'+p2y', the power p1 corresponding to the second device #1 with lower computing power is higher than the power p2 corresponding to the second device #2 with higher computing power. Therefore, the second device #1 can obtain or detect its own signal from p1x'+p2y' based on the third AI model, and obtain... The second device #1 acquires or detects its own signal from p1x'+p2y' based on the second AI model and the third AI model. Specifically, the second device #1 first acquires or detects the signal using the second AI model, that is... Then Input into the second AI model to recover the interference signal Then Input into the third AI model to obtain

[0204] Based on the above scheme, the first device can identify the N second devices performing multiple access as devices in the same multiple access user group, and process the data to be transmitted corresponding to each of the N second devices based on the same first AI model, such as extracting the data features of each second device to obtain the second data corresponding to each second device. Since the computing power of each of the N second devices differs, the first device can allocate corresponding power to each second device based on its computing power. In this way, the first device can send a first signal to each of the N second devices based on the first AI model. This first signal includes the data and power corresponding to each of the N second devices. Thus, even in large-scale access scenarios, the first device can reasonably allocate power to the second devices, which is beneficial to improving the efficiency of multiple access.

[0205] Figure 3 The communication method 300 only illustrates some steps of the interaction between the first device and the second device, and the communication method may also include other steps.

[0206] For example, taking N as 2, such as the second device #1 and the second device #2. Figure 6 An embodiment of this application illustrates a communication method 600, which includes steps 601 to 617. The steps of method 600 are described in detail below.

[0207] In step 601, the first device sends a third instruction message to the second device #1, which instructs the second device #1 to report its computing power information. Correspondingly, the second device #1 receives the third instruction message from the first device.

[0208] In step 602, the first device sends a third instruction message to the second device #2, which instructs the second device #2 to report its computing power information. Correspondingly, the second device #2 receives the third instruction message from the first device.

[0209] In step 603, the second device #1 sends a second indication message to the first device, which indicates the computing power information of the second device #1. Correspondingly, the first device receives the second indication message from the second device #1.

[0210] In step 604, the second device #2 sends a second indication message to the first device, which indicates the computing power information of the second device #2. Correspondingly, the first device receives the second indication message from the second device #2.

[0211] Steps 601 to 605 have been described in detail in step 310 of the above method 300. Please refer to the relevant description of step 310. They will not be repeated here.

[0212] In step 605, the first device determines the first AI model and the third AI model.

[0213] In step 606, the first device performs user pairing, which involves power pairing for the users after the computing power is paired.

[0214] In step 607, the first device sends a fourth indication message to the second device #1, which instructs the second device #1 to receive the first signal based on the third model. Correspondingly, the second device #1 receives the fourth indication message from the first device.

[0215] In step 608, the first device sends a fourth indication message to the second device #2, which instructs the second device #2 to receive the first signal based on the second AI model and the third model. Correspondingly, the second device #2 receives the fourth indication message from the first device.

[0216] Steps 606 to 608 have been described in detail in step 320 of method 300 above. Please refer to the relevant description of step 320. They will not be repeated here.

[0217] In step 609, the first device sends a fifth indication message to the second device #1, which indicates the power corresponding to the second device #2. Correspondingly, the second device #2 receives the fifth indication message from the first device.

[0218] In step 610, the first device sends a fifth indication message to the second device #2, which is used to indicate the power corresponding to the second device #2.

[0219] Steps 609 to 610 have been described in detail in step 310 of the above method 300. Please refer to the relevant description of step 310. They will not be repeated here.

[0220] In step 611, the first device sends first instruction information to the second device #1, which is used to configure the model corresponding to the second device #2. Correspondingly, the second device #1 receives the first instruction information from the first device.

[0221] In step 612, the first device sends first instruction information to the second device #2, the first instruction information being used to configure the model corresponding to the second device #2. Correspondingly, the second device #2 receives the first instruction information from the first device.

[0222] Steps 611 to 612 have been described in detail in step 320 of the above method 300. Please refer to the relevant description of step 320. They will not be repeated here.

[0223] In step 613, the first device processes the data to be transmitted corresponding to the second device #1 and the data to be transmitted corresponding to the second device #2 based on the first AI model.

[0224] The process by which the first device processes the data to be transmitted corresponding to the second device (e.g., the second device #1 and the second device #2) based on the first AI model has been detailed in step 320 of the above method 300. Please refer to the relevant description in step 320. It will not be repeated here.

[0225] In step 614, the first device sends a first signal to the second device #1. Correspondingly, the second device #1 receives the first signal from the first device.

[0226] In step 615, the first device sends a first signal to the second device #2. Correspondingly, the second device #2 receives the first signal from the first device.

[0227] The transmission of the first signal from the first device to the second device (e.g., the second device #1 and the second device #2) has been described in detail in step 320 of the above method 300. Please refer to the relevant description in step 320. It will not be repeated here.

[0228] In step 616, the second device #1 obtains the signal corresponding to the second device #1 from the first signal based on the third AI model.

[0229] The process of obtaining the signal corresponding to the second device #1 from the first signal based on the third AI model has been detailed in step 330 of the above method 300. Please refer to the relevant description in step 330. It will not be repeated here.

[0230] In step 617, the second device #2 obtains the signal corresponding to the second device #2 from the first signal based on the second AI model and the third AI model.

[0231] The process of obtaining the signal corresponding to the second device #1 from the first signal based on the third AI model and the second AI model has been described in detail in step 330 of the above method 300. Please refer to the relevant description in step 330. It will not be repeated here.

[0232] It should be understood that Figure 3 or Figure 6 The processes shown are merely examples and should not be construed as limiting the scope of this application. In other embodiments, these processes may include more or fewer steps.

[0233] It should also be understood that the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0234] The channel state information reporting method provided in the embodiments of this application has been described in detail above with reference to the accompanying drawings. The apparatus provided in the embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0235] Figures 8 to 9 Schematic block diagrams of possible communication devices provided for embodiments of this application. These communication devices can be used to implement the functions of the first device or the second device in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0236] This application provides a communication device such as Figure 8 As shown, the communication device 800 includes a communication unit 810 and a processing unit 820. The communication unit 810 can be used to perform receiving or sending actions, while the processing unit 820 can be used to perform actions other than receiving and sending, such as generating information or messages, processing received information or messages, etc.

[0237] One possible design is that the communication device 800 is used to achieve the above. Figure 3 or Figure 6 The method embodiments shown illustrate the function of the first device in any one of the embodiments. For example, the communication device can be the first device, a component configured in the first device (such as a chip, chip system, processor, etc.), or a logic module or software capable of implementing some or all of the functions of the first device.

[0238] For example, when the communication device 800 is used to implement the function of the first device in method 300, the processing unit 820 is used to determine the power corresponding to each of the N second devices, the power corresponding to each second device is determined based on the computing power of the second device, and N is an integer greater than 1; the communication unit 810 is used to send a first signal based on the power and data corresponding to each of the N second devices, the data corresponding to each second device is determined based on a first artificial intelligence (AI) model.

[0239] Optionally, the signals corresponding to different second devices in the first signal are carried on the same resource.

[0240] Optionally, the first signal may include one or more tokens.

[0241] Optionally, at most one of the N second devices may be unable to run the second AI model due to insufficient computing power. The second AI model is either the first AI model or a model obtained by distillation based on the first AI model.

[0242] Optionally, each of the second devices corresponds to the same third AI model, which is used to obtain the signal corresponding to the second device from the first signal.

[0243] Optionally, the processing unit 820 is further configured to determine a model corresponding to each of the second devices based on the computing power of each of the second devices. The model includes: a third AI model, or a second AI model and a third AI model. The third AI model is used to obtain the signal corresponding to the second device from the first signal. The second AI model is the first AI model or is obtained by distillation based on the first AI model. The communication unit 810 is further configured to send first indication information, which is used to configure the model corresponding to each of the second devices.

[0244] Optionally, the communication unit 810 is further configured to receive second indication information from each of the second devices, the second indication information being used to indicate the computing power of the second device.

[0245] Optionally, the communication unit 810 is further configured to send a third indication information, the third indication information being used to indicate the sending of the second indication information.

[0246] Optionally, the second indication information includes at least one of the following: TOPS; memory size; video memory size; CPU utilization; GPU utilization.

[0247] Optionally, the signal-to-interference-plus-noise ratio (SIR) of each of the second devices when receiving a signal at the corresponding power is greater than or equal to a threshold, the threshold being used to determine whether the second device can obtain the corresponding signal from the first signal through the third AI model.

[0248] Optionally, the processing unit 820 is also configured to determine the first AI model based on the data to be transmitted.

[0249] Optionally, the processing unit 820 is further configured to determine a third AI model based on the data to be transmitted and the computing power of the second device, wherein the third AI model is used to obtain the signal corresponding to the second device from the first signal.

[0250] Optionally, the communication unit 810 is further configured to send a fourth indication message, which instructs the second device to receive the first signal based on a third AI model.

[0251] Optionally, the first AI model and the third AI model corresponding to each of the N second devices are the same. The processing unit 820 is further configured to generate group information corresponding to the N second devices. The group information includes at least one of the following: information of the first AI model; information of the third AI model; information of each second device; and a power preset value, wherein the power preset value is greater than or equal to the sum of the power of the N second devices.

[0252] Optionally, the communication unit 810 is further configured to send a fifth indication message, the fifth indication message being used to indicate the power corresponding to each of the second devices.

[0253] One possible design is that the communication device 800 is used to achieve the above. Figure 3 or Figure 6 The method embodiments shown illustrate the function of the second device in any one of the embodiments. For example, the communication device can be the second device, a component configured in the second device (such as a chip, chip system, processor, etc.), or a logic module or software capable of implementing some or all of the functions of the second device.

[0254] For example, when the communication device 800 is used to implement the function of the second device in method 300, the communication unit 810 is used to receive a first signal based on the power and data corresponding to each of the N second devices, wherein the power corresponding to each second device is determined based on the computing power of the second device, and the data corresponding to each second device is determined based on a first AI model, and N is an integer greater than 1; the processing unit 820 is used to obtain the signal corresponding to the second device from the first signal using a third AI model based on the first signal, or to obtain the signal corresponding to the second device from the first signal using a second AI model and a third AI model, wherein the second AI model is the first AI model or is obtained by distillation based on the first AI model.

[0255] Optionally, the signals corresponding to different second devices in the first signal are carried on the same resource.

[0256] Optionally, the first signal may include one or more tokens.

[0257] Optionally, at most one of the N second devices may be unable to run the second AI model due to insufficient computing power.

[0258] Optionally, each of the second devices corresponds to the same third AI model.

[0259] Optionally, the communication unit 810 is further configured to receive first indication information, the first indication information being used to configure a model corresponding to each of the second devices, the model being determined based on the computing power of each of the second devices, the model including: the third AI model, or, the second AI model and the third AI model.

[0260] Optionally, the communication unit 810 is also configured to send a second indication information, the second indication information being used to indicate the computing power of the second device.

[0261] Optionally, the communication unit 810 is further configured to receive third indication information, the third indication information being used to instruct the transmission of the second indication information.

[0262] Optionally, the second indication information includes at least one of the following: TOPS; memory size; video memory size; CPU utilization; GPU utilization.

[0263] Optionally, the signal-to-interference-plus-noise ratio (SIR) of each of the second devices when receiving a signal at the corresponding power is greater than or equal to a threshold, the threshold being used to determine whether the second device can obtain the corresponding signal from the first signal through the third AI model.

[0264] Optionally, the first AI model is determined based on the data to be transmitted.

[0265] Optionally, the third AI model is determined based on the data to be transmitted and the computing power of the second device.

[0266] Optionally, the communication unit 810 is further configured to receive fourth indication information, which instructs the second device to receive the first signal based on the third AI model.

[0267] Optionally, the communication unit 810 is further configured to receive fifth indication information, which indicates the power corresponding to each of the second devices.

[0268] It should also be understood that the communication unit 810 in the communication device 800 can also be called a transceiver unit. The communication unit 810 may include a transmitting module but not a receiving module. Alternatively, the communication unit 810 may include a receiving module but not a transmitting module. Specifically, it depends on whether the above-described scheme executed by the communication device 800 includes both transmitting and receiving actions. The receiving module can be used to perform the receiving action in the above-described scheme, and the transmitting module can be used to perform the transmitting action in the above-described scheme.

[0269] It is understood that the division of units in the above-described device is merely a logical functional division. Each function can correspond to a functional unit, or two or more functions can be integrated into one functional unit. In actual implementation, all or some units can be integrated into a single physical entity, or they can be 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 each specific application, but such implementation should not be considered beyond the scope of this application.

[0270] Another communication device provided in this application is such as Figure 9 As shown, the communication device 900 includes at least one processor 910. The at least one processor 910 can be used to execute computer programs or instructions stored in memory to achieve... Figure 3 or Figure 6 The steps performed by the first device or the steps performed by the second device in any of the embodiments of the method shown.

[0271] Optionally, the communication device 900 may further include at least one memory 920 for storing instructions executed by the processor 910, or storing input data required by the processor 910 to execute instructions, or storing data generated after the processor 910 executes instructions. The at least one processor 910 and the at least one memory 920 may be configured separately. For example, each memory may be connected to one or more processors, enabling the connected processors to read information from, store, and / or write information to the memory. Alternatively, the at least one processor 910 and the at least one memory 920 may be integrated together; for example, one or more memories may be integrated into a single processor.

[0272] Optionally, the communication device 900 further includes an interface circuit 930 for transmitting data and / or signaling. The at least one processor 910 and the interface circuit 930 are coupled to each other. It is understood that the interface circuit 930 can be a transceiver, input / output circuit, bus, module, pin, or other type of communication interface, wherein the input circuit in the input / output circuit can be used for receiving, and the output interface can be used for transmitting.

[0273] Optionally, the communication device 900 further includes a power supply circuit 940, which can be used to supply power to the communication device 900.

[0274] When the communication device 900 is used to achieve Figures 8 to 9 When the method is performed in any of the embodiments shown in the method examples, the processor 910 is used to execute the functions of the processing unit, and the interface circuit 920 is used to execute the functions of the receiving unit and / or the transmitting unit. Whether the interface circuit 920 is used for transmitting or receiving depends on whether the communication device 900 is performing a transmitting or receiving action in the execution scheme.

[0275] It is understood that when the communication device 900 is a communication device (e.g., a first device or a second device), the interface circuit 920 can be a transceiver, specifically including a transmitter and a receiver, with the transmitter used to send signals and the receiver used to receive signals. When the communication device 900 is a chip used in a communication device, the interface circuit 920 can be an input / output circuit, a bus, a module, a pin, or other type of communication interface, wherein the input circuit in the input / output circuit can be used for receiving, and the output interface can be used for sending.

[0276] It should be understood that Figure 9 In the communication device 900 shown, the processor 910 may correspond to the processing unit 820 in the aforementioned communication device 900, and the interface circuit 920 may correspond to the communication unit 810 in the aforementioned communication device 800.

[0277] It should also be understood that the coupling in the embodiments of this application is an indirect coupling or communication connection between devices, units, or modules, which can be electrical, mechanical, or other forms, used for information interaction between devices, units, or modules. The embodiments of this application do not limit the specific connection medium between the at least one processor 910, at least one memory 920, interface circuit 930, and power supply circuit 940. The embodiments of this application in... Figure 9 The processor 910, memory 920, interface circuit 930, and power supply circuit 940 are connected via bus 970. Bus 950 is... Figure 9The connections between other components are shown in bold lines only and are not intended to be limiting. The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0278] It is understood that the processor in the embodiments of this application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0279] The memory in this application embodiment can be volatile memory or non-volatile memory, or it can include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0280] This application also provides a communication system, which includes the aforementioned first device and second device.

[0281] This application also provides a computer program product comprising: a computer program (also referred to as code or instructions), which, when executed, causes the computer to perform actions such as... Figure 3 or Figure 6 The method executed by the first device or the method executed by the second device in the illustrated embodiment.

[0282] This application also provides a computer-readable storage medium storing a computer program (also referred to as code or instructions). When the computer program is executed, it causes the computer to perform actions such as... Figure 3 or Figure 6 The method executed by the first device or the method executed by the second device in the illustrated embodiment.

[0283] The terms “unit”, “module”, etc., used in this specification may be used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution.

[0284] Those skilled in the art will recognize that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented 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 each specific application, but such implementation should not be considered beyond the scope of this application. In the several embodiments provided in this application, it should be understood that the disclosed apparatus, devices, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0285] The unit described as a separate component may or may not be physically separate. The component shown as a unit may or may not be a physical unit; that is, it may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0286] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0287] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. This computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0288] If this function 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, in essence, or the part that contributes to the prior art, or a portion of the technical solution, 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.) 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, RAM, magnetic disks, or optical disks.

[0289] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A communication method applied to a first device, characterized in that, The method includes: Determine the power corresponding to each of the N second devices. The power corresponding to each second device is determined based on the computing power of the second device, where N is an integer greater than 1. Based on the power and data corresponding to each of the N second devices, a first signal is sent, wherein the data corresponding to each second device is determined based on a first artificial intelligence (AI) model.

2. The method according to claim 1, characterized in that, The signals corresponding to different second devices in the first signal are carried on the same resource.

3. The method according to claim 1 or 2, characterized in that, The first signal includes one or more tokens.

4. The method according to any one of claims 1 to 3, characterized in that, At most one of the N second devices is unable to run the second AI model due to insufficient computing power. The second AI model is the first AI model or is obtained by distillation based on the first AI model.

5. The method according to any one of claims 1 to 4, characterized in that, Each of the second devices corresponds to the same third AI model, which is used to obtain the signal corresponding to the second device from the first signal.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: Based on the computing power of each second device, a model corresponding to each second device is determined. The model includes: a third AI model, or a second AI model and a third AI model. The third AI model is used to obtain the signal corresponding to the second device from the first signal. The second AI model is the first AI model or is obtained by distillation based on the first AI model. Send a first instruction message, which is used to configure the model corresponding to each of the second devices.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Receive second indication information from each of the second devices, the second indication information being used to indicate the computing power of the second device.

8. The method according to claim 7, characterized in that, The method further includes: Send a third instruction message, which is used to instruct the sending of the second instruction message.

9. The method according to claim 7 or 8, characterized in that, The second instruction information includes at least one of the following: TOPS (trillion operations per second); Memory size; Video memory size; Central Processing Unit (CPU) utilization; Graphics processor (GPU) utilization.

10. The method according to any one of claims 1 to 9, characterized in that, The signal-to-interference-plus-noise ratio (SIR) of each of the second devices when receiving signals at the corresponding power is greater than or equal to a threshold, which is used to determine whether the second device can obtain the corresponding signal from the first signal through the third AI model.

11. The method according to any one of claims 1 to 10, characterized in that, The method further includes: The first AI model is determined based on the data to be transmitted.

12. The method according to any one of claims 1 to 11, characterized in that, The method further includes: A third AI model is determined based on the data to be transmitted and the computing power of the second device. The third AI model is used to obtain the signal corresponding to the second device from the first signal.

13. The method according to any one of claims 1 to 12, characterized in that, The method further includes: Generate a mapping relationship corresponding to the N second devices, wherein the mapping relationship includes at least one of the following: Information from the first AI model; Information from a third AI model, wherein the third AI model is used to obtain the signal corresponding to the second device from the first signal; Information about each of the second devices; A power preset value, wherein the power preset value is greater than or equal to the sum of the power of the N second devices.

14. A communication method applied to a second device, characterized in that, The method includes: Based on the power and data corresponding to each of the N second devices, a first signal is received. The power corresponding to each second device is determined based on the computing power of the second device, and the data corresponding to each second device is determined based on the first AI model. N is an integer greater than 1. Based on the first signal, a third AI model is used to obtain the signal corresponding to the second device from the first signal, or a second AI model and a third AI model are used to obtain the signal corresponding to the second device from the first signal, wherein the second AI model is the first AI model or is obtained by distillation based on the first AI model.

15. The method according to claim 14, characterized in that, The signals corresponding to different second devices in the first signal are carried on the same resource.

16. The method according to claim 14 or 15, characterized in that, The first signal includes one or more tokens.

17. The method according to any one of claims 14 to 16, characterized in that, At most one of the N second devices is unable to run the second AI model due to insufficient computing power.

18. The method according to any one of claims 14 to 17, characterized in that, Each of the second devices corresponds to the same third AI model.

19. The method according to any one of claims 14 to 18, characterized in that, The method further includes: Receive first instruction information, the first instruction information is used to configure the model corresponding to each second device, the model is determined based on the computing power of each second device, the model includes: the third AI model, or the second AI model and the third AI model.

20. The method according to any one of claims 14 to 19, characterized in that, The method further includes: Send a second indication message, which is used to indicate the computing power of the second device.

21. The method according to claim 20, characterized in that, The method further includes: Receive third indication information, which is used to instruct the sending of the second indication information.

22. The method according to claim 20 or 21, characterized in that, The second instruction information includes at least one of the following: TOPS; Memory size; Video memory size; CPU utilization; GPU utilization.

23. The method according to any one of claims 14 to 22, characterized in that, The signal-to-interference-plus-noise ratio (SIR) of each of the second devices when receiving signals at the corresponding power is greater than or equal to a threshold, the threshold being used to determine whether the second device can obtain the corresponding signal from the first signal through the third AI model.

24. The method according to any one of claims 14 to 23, characterized in that, The first AI model is determined based on the data to be transmitted.

25. The method according to any one of claims 14 to 24, characterized in that, The third AI model is determined based on the data to be transmitted and the computing power of the second device.

26. A communication device, characterized in that, It includes units for performing the method as described in any one of claims 1 to 13, or units for performing the method as described in any one of claims 14 to 25.

27. A communication device, characterized in that, The device includes one or more processors, which are configured to execute computer programs or instructions in memory to cause the communication device to perform the method as described in any one of claims 1 to 13, or to cause the communication device to perform the method as described in any one of claims 14 to 25.

28. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it causes the method as described in any one of claims 1 to 13 to be performed, or causes the method as described in any one of claims 14 to 25 to be performed.

29. A computer program product, characterized in that, Includes a computer program that, when run, causes the method as claimed in any one of claims 1 to 13 to be performed, or causes the method as claimed in any one of claims 14 to 25 to be performed.

Images