Communication method and device

Abstract

The invention provides a communication method and device. A terminal device receives a reference signal and a data signal; obtaining first information before demodulation according to the reference signal and the data signal; inputting the first information into a neural network model, and outputting second information; and demodulating the second information to obtain first data. The input information of the neural network model comprises information before demodulation obtained according to non-orthogonal reference signals on the same transmission resource, and the output information of the neural network model comprises information obtained by processing the input information. In the embodiment of the invention, the neural network model is adopted to correct the information before demodulation, so that the information before demodulation is closer to the information after modulation, errors caused by non-orthogonal reference signals can be reduced, the obtained data are more accurate, and the error rate can be reduced.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a communication method and apparatus. Background Technology

[0002] In wireless communication systems, downlink data can be demodulated using a demodulation reference signal (DMRS). For example, a base station sends a DMRS and data signal to a terminal device. After receiving the DMRS and data signal, the terminal device performs channel estimation based on the DMRS, and then performs demodulation and other processing on the data signal based on the channel estimation result to obtain the downlink data.

[0003] To reduce interference between signals, the DMRS sent by the base station to different terminal devices are orthogonal. However, the number of orthogonal DMRS that the same transmission resource can carry is limited. As the number of terminal devices increases, in order to reduce interference, the base station can send orthogonal DMRS to more terminal devices using more transmission resources, resulting in excessive consumption of transmission resources. Summary of the Invention

[0004] This application provides a communication method and apparatus for saving transmission resources.

[0005] Firstly, this application provides a communication method that can be executed by a terminal device, or by other devices including the functions of a terminal device, or by a chip system (which can also be replaced by a chip) or other functional module, wherein the chip system or functional module is capable of realizing the functions of the terminal device, and the chip system or functional module is, for example, disposed in the terminal device. Taking the execution of the method by a terminal device as an example: the terminal device receives a first signal, the first signal including a first reference signal and a first data signal; the terminal device inputs first information into a first neural network model to obtain second information output by the first neural network model; the terminal device demodulates the second information to obtain first data; wherein the first information is information before demodulation obtained by detecting the first data signal based on a first channel estimation result, the first channel estimation result being obtained by channel estimation based on the first reference signal, the input information of the first neural network model including information before demodulation obtained based on non-orthogonal reference signals on the same transmission resource, and the output information of the first neural network model including information obtained by processing the input information.

[0006] In this embodiment, the reference signal received by the terminal device can be non-orthogonal. Compared to orthogonal reference signals carried on the same transmission resource, the number of non-orthogonal reference signals carried on the same transmission resource can be greater. Therefore, the network device can send non-orthogonal reference signals to more terminal devices with limited transmission resources, saving transmission resources while ensuring the transmission of reference signals as much as possible. For a single terminal device, channel estimation can be performed based on the received reference signal. The information before demodulation can be determined based on the channel estimation result, and a first neural network model can be used to correct the information before demodulation. Demodulation based on the corrected information before demodulation results in more accurate data. If the terminal device directly demodulates based on the information before demodulation obtained from the non-orthogonal reference signal, the demodulated data may be inaccurate. This embodiment, through the first neural network model, can reduce the error caused by the non-orthogonal reference signal, making the obtained data more accurate and thus reducing the bit error rate.

[0007] In one possible implementation, the difference between the first data and the original data is less than the difference between the second data and the original data, where the second data is data obtained by demodulating the first information. Alternatively, the difference between the second information and the modulated information is less than the difference between the first information and the modulated information; wherein, the modulated information is information obtained by modulating the original data corresponding to the first data.

[0008] In one possible implementation, the terminal device may also receive first indication information, which is used to indicate the first neural network model; or, the terminal device may determine multiple training samples and train the first neural network model based on the multiple training samples.

[0009] In this implementation, the terminal device can obtain the first neural network model from other devices without having to train it itself, thereby simplifying the implementation of the terminal device. Alternatively, the terminal device can also participate in the training of the first neural network model, making the first neural network model more suitable for the terminal device's situation, such as the channel conditions between the terminal device and the network device, which can further reduce the bit error rate.

[0010] In one possible implementation, determining a training sample includes: receiving a sample signal; the sample signal includes a sample reference signal and a sample data signal; determining predicted sample information based on the sample signal; wherein the predicted sample information is information before demodulation obtained by detecting the sample data signal based on the sample channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal; determining actual sample information corresponding to the sample data signal, the actual sample information being the modulated information; and determining the actual sample information and the predicted sample information as a training sample.

[0011] Alternatively, the predicted sample information is data obtained by demodulating the information before demodulation. The information before demodulation is obtained by detecting the sample data signal based on the channel estimation result. The sample channel estimation result is obtained by channel estimation based on the sample reference signal. The actual sample information is the original data.

[0012] In one possible implementation, the method further includes: a terminal device receiving a sample signal, the sample signal including a sample reference signal and a sample data signal; and sending predicted sample information, the predicted sample information being used to train the first neural network model; wherein the predicted sample information is information before demodulation obtained by detecting the sample data signal based on the sample channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal. Alternatively, the predicted sample information is data obtained by demodulating the information before demodulation, the information before demodulation being information before demodulation obtained by detecting the sample data signal based on the current channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal.

[0013] In this implementation, the terminal device determines the prediction sample information and sends it to the network device, which makes the first neural network model more suitable for the terminal device, such as the channel conditions between the terminal device and the network device, and can further reduce the bit error rate.

[0014] In one possible implementation, the method further includes: a terminal device receiving a second signal, the second signal including a second reference signal and a second data signal; the terminal device demodulating third information to obtain second data; and updating the parameters of the first neural network model based on the second data; wherein the third information is determined according to fourth information, the fourth information being information before demodulation obtained by detecting the second data signal based on a second channel estimation result, and the second channel estimation result being obtained by channel estimation based on the second reference signal.

[0015] In this implementation, updating the parameters of the first neural network model by the terminal device makes the updated neural network model more compatible with the channel conditions between the terminal device and the network device, thereby further reducing the bit error rate. For example, the parameters of the first neural network model can be updated periodically, or they can be updated when certain conditions are met, such as when the accuracy of the output of the first neural network model decreases, when the data obtained from the first neural network model differs significantly from the original data, or when the bit error rate corresponding to the first neural network model is high. In these cases, the parameters of the first neural network model can be updated to make the first neural network model more accurate.

[0016] In one possible implementation, the second signal further includes a control signal indicating that the data corresponding to the second data signal is a training sample, or indicating that the second signal is used to update the first neural network model. Alternatively, an indication message may be included in a signal other than the second signal, indicating that the data corresponding to the second data signal is a training sample or indicating that the second signal is used to update the first neural network model. Through the indication of the control signal, the first terminal device can accurately determine the parameters of the second signal / second data signal used to update the model.

[0017] In one possible implementation, the third information is the information output by the first neural network model after the fourth information is input into it.

[0018] In this implementation, the third information is processed by the first neural network model, making it closer to the information sent by the network device, which can improve the accuracy of the training samples.

[0019] In one possible implementation, the method further includes: the terminal device sending a first request, the first request being used to request an update of the parameters of the first neural network model.

[0020] In this implementation, when the terminal device determines that the update cycle has arrived, or when it determines that the bit error rate does not meet the requirements after demodulation based on the output information of the first neural network model, it can actively request to update the parameters of the neural network model, which can improve the timeliness of model correction.

[0021] In one possible implementation, the method further includes: a terminal device receiving second indication information, the second indication information being used to indicate the transmission resources corresponding to the first reference signal, wherein the frequency domain resources in the transmission resources do not belong to 5G mobile communication resources.

[0022] In this implementation, network devices configure non-5G mobile communication resources for terminal devices with first neural network model processing capabilities, and configure 5G mobile communication resources for terminal devices without first neural network model processing capabilities. This reduces interference between the two types of terminal devices. The first neural network model processing capability can be applied to future mobile communication systems, and the resources of future mobile communication systems may differ from, but are not limited to, 5G mobile communication resources.

[0023] Secondly, this application provides a communication method that can be executed by a network device, or by other devices including network device functions, or by a chip system (which can also be replaced by a chip) or other functional module, which can realize the functions of the network device, and is, for example, disposed in the network device. Taking the execution of this method by a network device as an example: the network device sends first indication information to multiple terminal devices, the first indication information being used to indicate a first neural network model; wherein, the input information of the first neural network model includes undemodulated information obtained based on non-orthogonal reference signals on the same transmission resource, and the output information of the first neural network model includes information obtained by processing the input information.

[0024] In this embodiment, the reference signal received by the terminal device can be non-orthogonal. Compared to orthogonal reference signals carried on the same transmission resource, the number of non-orthogonal reference signals carried on the same transmission resource can be greater. Therefore, the network device can send non-orthogonal reference signals to more terminal devices with limited transmission resources, saving transmission resources while ensuring the transmission of reference signals as much as possible. For a single terminal device, channel estimation can be performed based on the received reference signal. The information before demodulation can be determined based on the channel estimation result, and a first neural network model can be used to correct the information before demodulation. Demodulation based on the corrected information before demodulation results in more accurate data. If the terminal device directly demodulates based on the information before demodulation obtained from the non-orthogonal reference signal, the demodulated data may be inaccurate. This embodiment, through the first neural network model, can reduce the error caused by the non-orthogonal reference signal, making the obtained data more accurate and thus reducing the bit error rate.

[0025] In one possible implementation, the method further includes: a network device sending a third signal to the plurality of terminal devices, the third signal including a third reference signal and a third data signal; the plurality of third reference signals are non-orthogonal and occupy the same transmission resources.

[0026] In one possible implementation, the method further includes: the network device training the first neural network model based on multiple training samples; or, the network device receiving third indication information, the third indication information being used to indicate the first neural network model.

[0027] In this implementation, network devices obtain neural network models from other devices, which reduces the resource consumption of network devices. Training the neural network model themselves allows the trained model to better reflect the channel conditions between the terminal device and the network device, further reducing the bit error rate.

[0028] In one possible implementation, the method further includes: a network device sending sample signals to N training devices; wherein the sample signal sent to the i-th training device among the N training devices includes an i-th sample reference signal and an i-th sample data signal, the N sample reference signals sent to the N training devices are non-orthogonal and occupy the same transmission resources, the value of i traverses integers from 1 to N, and N is an integer greater than or equal to 2; the network device receiving the i-th predicted sample information from the i-th training device; wherein the i-th predicted sample information is: The i-th training device detects the unmodulated information of the i-th sample data signal based on the channel estimation result of the i-th sample, and the channel estimation result of the i-th sample is obtained by the i-th training device through channel estimation based on the reference signal of the i-th sample; the network device determines the i-th actual sample information corresponding to the i-th sample data signal, and the i-th actual sample information is the modulated information; the network device uses the i-th actual sample information and the i-th predicted sample information as a training sample to determine some or all of the training samples among the multiple training samples.

[0029] In one possible implementation, the method further includes: a network device sending a second signal to a first terminal device, the second signal including a second reference signal and a second data signal, the data corresponding to the second data signal being used to update the parameters of the first neural network model, and the first terminal device belonging to the plurality of terminal devices.

[0030] In this implementation, the terminal device updates the parameters of the first neural network model, which makes the updated neural network model more compatible with the channel conditions between the terminal device and the network device, thereby further reducing the bit error rate.

[0031] In one possible implementation, the second signal further includes a control signal indicating that the data corresponding to the second data signal is a training sample, or indicating that the second signal is used to update the first neural network model. Alternatively, an indication message may be included in a signal other than the second signal, indicating that the data corresponding to the second data signal is a training sample or indicating that the second signal is used to update the first neural network model. Through the indication of the control signal, the first terminal device can accurately determine the parameters of the second signal / second data signal used to update the model.

[0032] In one possible implementation, before sending the second signal to the first terminal device, the network device may also receive a first request from the first terminal device; wherein the first request is for requesting an update of the parameters of the first neural network model.

[0033] In this implementation, the parameters of the first neural network model can be updated periodically or when certain conditions are met. For example, when the accuracy of the output of the first neural network model decreases, when the data obtained from the first neural network model differs significantly from the original data, or when the bit error rate of the first neural network model is high, the parameters of the first neural network model can be updated to make the first neural network model more accurate. The terminal device can actively request updates to the parameters of the neural network model, which can improve the timeliness of model correction.

[0034] In one possible implementation, the method further includes: a network device sending second indication information to the plurality of terminal devices; wherein the second indication information is used to indicate the transmission resources corresponding to the third reference signal, and the frequency domain resources in the transmission resources do not belong to 5G mobile communication resources.

[0035] In this implementation, network devices configure non-5G mobile communication resources for terminal devices with first neural network model processing capabilities, and configure 5G mobile communication resources for terminal devices without first neural network model processing capabilities. This reduces interference between the two types of terminal devices. The first neural network model processing capability can be applied to future mobile communication systems, and the resources of future mobile communication systems may differ from, but are not limited to, 5G mobile communication resources.

[0036] Thirdly, this application provides a communication method that can be executed by a network device, or by other devices including network device functions, or by a chip system (which can also be replaced by a chip) or other functional modules. The chip system or functional module can implement the functions of the network device, and is, for example, disposed within the network device. Taking the execution of this method by a network device as an example: the network device sends sample signals to N training devices; wherein, the sample signal sent to the i-th training device among the N training devices includes an i-th sample reference signal and an i-th sample data signal, the N sample reference signals sent to the N training devices are non-orthogonal and occupy the same transmission resources, the value of i traverses integers from 1 to N, and N is an integer greater than or equal to 2; the network device receives the i-th predicted sample information from the i-th training device; wherein, the i-th predicted... The sample information is the unmodulated information obtained by the i-th training device from detecting the i-th sample data signal based on the channel estimation result of the i-th sample. The channel estimation result of the i-th sample is obtained by the i-th training device from channel estimation based on the reference signal of the i-th sample. The network device determines the i-th actual sample information corresponding to the i-th sample data signal. The i-th actual sample information is the modulated information. The network device uses the i-th actual sample information and the i-th predicted sample information as a training sample to train the neural network model.

[0037] Fourthly, this application provides a communication method that can be executed by a terminal device, or by other devices including the functions of a terminal device, or by a chip system (which can also be replaced by a chip) or other functional module, wherein the chip system or functional module is capable of realizing the functions of the terminal device, and the chip system or functional module is, for example, disposed in the terminal device. Taking the execution of the method by a terminal device as an example: the terminal device receives a sample signal; the sample signal includes a sample reference signal and a sample data signal; the terminal device determines predicted sample information based on the sample signal; wherein the predicted sample information is information before demodulation obtained by detecting the sample data signal according to the sample channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal; the terminal device determines the actual sample information corresponding to the sample data signal, the actual sample information being the modulated information; the terminal device uses the actual sample information and the predicted sample information as a training sample to train a neural network model.

[0038] Fifthly, a communication device is provided, which can be a network device as described in the preceding aspects. The communication device possesses the functions of the aforementioned network device. The communication device is, for example, a functional module within the network device, such as a baseband device or a chip system. Alternatively, the communication device can be a terminal device as described in the preceding aspects. The communication device possesses the functions of the aforementioned terminal device. The communication device is, for example, a functional module within the terminal device, such as a baseband device or a chip system.

[0039] In one optional implementation, the communication device includes a baseband device and a radio frequency device. In another optional implementation, the communication device includes a processing unit (sometimes also called a processing module) and a transceiver unit (sometimes also called a transceiver module). The transceiver unit is capable of transmitting and receiving functions. When the transceiver unit performs the transmitting function, it can be called a transmitting unit (sometimes also called a transmitting module), and when the transceiver unit performs the receiving function, it can be called a receiving unit (sometimes also called a receiving module). The transmitting unit and the receiving unit can be the same functional module, which is called the transceiver unit and can perform both transmitting and receiving functions; or, the transmitting unit and the receiving unit can be different functional modules, and the transceiver unit is a collective term for these functional modules.

[0040] In one possible implementation, the communication device further includes a storage unit (sometimes also called a storage module), and the processing unit is configured to couple with the storage unit and execute programs or instructions in the storage unit to enable the communication device to perform the functions of the network device described in the foregoing aspects, or to perform the functions of the terminal device described in the foregoing aspects.

[0041] Sixthly, a communication device is provided, including an interface circuit and a processor, and optionally, a memory. The memory stores a computer program, and the processor is coupled to the memory and the interface circuit. When the processor reads the computer program or instructions, it causes the communication device to execute the methods executed by the network device or the methods executed by the terminal device as described in the preceding aspects. For example, the interface circuit is used to receive signals from other communication devices besides the communication device and transmit them to the processor, or to send signals from the processor to other communication devices besides the communication device. The processor, through logic circuits or executable code instructions, implements the methods executed by the network device or the methods executed by the terminal device as described in the preceding aspects.

[0042] In one possible implementation, the communication device is a chip or chip system.

[0043] In a seventh aspect, a communication device is provided, including a processor, and optionally, a memory; the processor and the memory are coupled; the memory is used to store computer programs or instructions; the processor is used to execute part or all of the computer programs or instructions in the memory, and when the part or all of the computer programs or instructions are executed, to implement the functions of a network device in the above aspects, or to implement the functions of a terminal device in the above aspects.

[0044] In one possible implementation, the apparatus may further include a transceiver for transmitting signals processed by the processor or receiving signals input to the processor. The transceiver may perform the transmitting or receiving actions performed by the network device in the foregoing aspects, or the transmitting or receiving actions performed by the terminal device in the foregoing aspects.

[0045] In one possible implementation, the processing unit in the fifth aspect can be implemented by the processor, the storage unit in the fifth aspect can be implemented by the memory, and the transceiver unit in the fifth aspect can be implemented by the transceiver.

[0046] In one possible implementation, the communication device is a chip or chip system.

[0047] Eighthly, a computer-readable storage medium is provided for storing a computer program or instructions that, when executed, cause the methods of the preceding aspects to be implemented.

[0048] Ninthly, a computer program product containing instructions is provided, which, when run on a computer, enables the methods in the above aspects to be implemented.

[0049] A tenth aspect provides a communication system including a network device and a terminal device, wherein the terminal device is configured to perform the method described in the first aspect, and the network device is configured to perform the method described in the second aspect. For example, the network device and the terminal device can be implemented using the communication apparatus described in the fifth aspect. Attached Figure Description

[0050] Figure 1 A schematic diagram of the architecture of a communication system provided in this application;

[0051] Figure 2 A schematic diagram of a data transmission method provided in this application;

[0052] Figure 3 A schematic diagram of a communication method provided in this application;

[0053] Figure 4 This application provides a constellation diagram;

[0054] Figure 5 A schematic diagram of a communication method provided in this application;

[0055] Figure 6 A schematic diagram of a communication method provided in this application;

[0056] Figure 7 A schematic diagram of a communication method provided in this application;

[0057] Figure 8a , Figure 8b and Figure 8c These are schematic diagrams illustrating the changes in signal-to-noise ratio and bit error rate provided in this application;

[0058] Figure 8d A schematic diagram illustrating the changes in signal-to-noise ratio and throughput provided in this application;

[0059] Figure 9 A structural diagram of a communication device provided in this application;

[0060] Figure 10 A structural diagram of a communication device provided in this application. Detailed Implementation

[0061] The technical solution of this application can be applied to terrestrial networks (TN) and non-terrestrial networks (NTN), such as satellite networks. It can be applied to various wireless communication systems, including but not limited to fourth-generation (4G) systems (also known as Long Term Evolution, LTE), fifth-generation (5G) systems (also known as New Radio, NR), or next-generation mobile communication systems or other similar communication systems (such as sixth-generation (6G) systems), without specific limitations. Furthermore, the technical solution can be applied to device-to-device (D2D) scenarios, such as NR-D2D scenarios, or to V2X scenarios, such as NR-V2X scenarios. The technical solution can also be applied to fields such as intelligent driving, assisted driving, intelligent connected vehicles, or factory manufacturing scenarios.

[0062] Figure 1 This is a schematic diagram of the architecture of a communication system used in an embodiment of this application. Figure 1 The communication system 1000 shown includes a wireless access network 100 and a core network 200. Optionally, the communication system 1000 also includes an Internet 300. The wireless access network 100 may include at least one network device (such as...). Figure 1 110a and 110b in the above), may also include at least one terminal device (such as Figure 1 (Referring to 120a-120j in the original text). Terminal devices connect to network devices wirelessly, and network devices connect to the core network 200 wirelessly or via wired connection. Core network devices and network devices can be independent physical devices, or they can integrate the functions of core network devices and the logical functions of network devices onto the same physical device. Alternatively, a single physical device can integrate some core network device functions and some network device functions. Terminal devices and network devices can be interconnected via wired or wireless connections. Figure 1 This is just an illustration; the communication system may also include other network devices, such as wireless repeaters and wireless backhaul devices. Figure 1 It is not shown in the middle.

[0063] The radio access network 100 can be a cellular system related to the 3rd generation partnership project (3GPP), such as the 4th generation (4G) system (also known as Long Term Evolution, LTE), the 5th generation (5G) system (also known as New Radio, NR), or it can be applied to next-generation mobile communication systems or other similar communication systems (such as the 6th generation (6G) system), etc., without specific limitations. The radio access network 100 can also be an open radio access network (open RAN, O-RAN or ORAN) or a cloud radio access network (CRAN). The wireless access network 100 can also be a non-terrestrial network (NTN), a satellite communication network, a high altitude platform station (HAPS) communication network, an integrated access and backhaul (IAB) communication network, a reconfigurable intelligent surface (RIS) communication network, etc. The wireless access network 100 can also be a communication system that integrates two or more of the above systems.

[0064] Network devices are nodes in a radio access network (RAN), also known as access network devices or RAN nodes (or devices). Network devices help terminal devices achieve wireless access. Multiple network devices in the communication system 1000 can be nodes of the same type or different types.

[0065] In one possible scenario, network equipment can be a base station, an evolved NodeB (eNodeB), a transmitting and receiving point (TRP), a transmitting point (TP), a next-generation NodeB (gNB), a next-generation base station in a 6G mobile communication system, a base station in a future mobile communication system, an access point (AP) in a satellite, an integrated access and backhaul (IAB) node, or network equipment in a mobile switching center non-terrestrial network (NTN) communication system. This means it can be deployed on high-altitude platforms or satellites. Network equipment can also be a macro base station (such as...). Figure 1 110a), micro base stations or indoor stations (such as Figure 1 In CRAN scenarios, network devices can be 110b), relay nodes or donor nodes, or wireless controllers. Network devices can also function as base stations in device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, drone communication, and machine-to-machine (M2M) communication. Optionally, network devices can also be servers, wearable devices, vehicles, or in-vehicle equipment. For example, in vehicle-to-everything (V2X) technology, the access network device can be a roadside unit (RSU).

[0066] In another possible scenario, multiple network devices collaborate to assist terminal devices in achieving wireless access, with each network device performing a portion of the base station's functions. For example, network devices can be central units (CUs), distributed units (DUs), CU-control plane (CPs), CU-user plane (UPs), or radio units (RUs). CUs and DUs can be separate entities or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio equipment or radio units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs). It is understood that network devices can be CU nodes, DU nodes, or devices comprising both CU and DU nodes. Furthermore, CUs can be classified as network devices in the access network (RAN) or the core network (CN), without limitation.

[0067] 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. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. 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 and hardware modules.

[0068] A terminal device is a device with wireless transceiver capabilities, capable of sending signals to or receiving signals from network devices. Terminal devices include, but are not limited to, terminal equipment, user equipment (UE), mobile stations, and mobile terminals. Terminal devices can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), the Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, intelligent transportation, and smart cities. Specifically, terminal devices can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, aircraft, ships, robots, robotic arms, smart home devices, etc. The embodiments of this application do not limit the specific technologies or device forms used in the terminal devices.

[0069] Network devices and terminal devices can be fixed in location or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can be deployed on aircraft, balloons, and artificial satellites. The embodiments of this application do not limit the application scenarios of the network devices and terminal devices.

[0070] The roles of network devices and terminal devices can be relative. For example, Figure 1 The helicopter or drone 120i can be configured as a mobile network device. For terminal devices 120j that access the wireless access network 100 via 120i, terminal device 120i is a network device; however, for network device 110a, 120i is a terminal device, meaning that 110a and 120i communicate via a wireless air interface protocol. Of course, 110a and 120i can also communicate via a network device-to-network device interface protocol; in this case, 120i is also a network device relative to 110a. Therefore, both network devices and terminal devices can be collectively referred to as communication devices. Figure 1 110a and 110b can be referred to as communication devices with network equipment functions. Figure 1 The 120a-120j in the text can be referred to as communication devices with terminal equipment functions.

[0071] In the embodiments of this application, the functions of the network device can be executed by modules (such as chips) within the network device, or by a control subsystem that includes network device functions. This control subsystem, including network device functions, can be a control center in the aforementioned application scenarios such as smart grids, industrial control, intelligent transportation, and smart cities. Similarly, the functions of the terminal device can be executed by modules (such as chips or modems) within the terminal device, or by a device that includes terminal device functions.

[0072] like Figure 2 As shown, a data transmission method is introduced.

[0073] Step 20: The transmitting end performs channel coding on the transport block to obtain multiple coded bits.

[0074] For example, the sending end performs forward error correction coding on the transport block, such as low-density parity-check code (LDPC).

[0075] Step 21: The transmitting end modulates multiple bits to obtain multiple modulation symbols.

[0076] Modulation can be understood as transforming an input signal into a waveform suitable for transmission through a channel. Modulation methods include, but are not limited to: quadrature amplitude modulation (QAM), offset quadrature amplitude modulation (OQAM), binary phase shift keying (BPSK), pi / 2-BPSK, QPSK, pi / 4-QPSK, 16QAM, 64QAM, 256QAM, 1024QAM, or APSK. The modulation order can be 1, 2, 4, 6, or 8, etc., and is related to the modulation method. This application does not limit the modulation method or the modulation order.

[0077] Step 22: The transmitter maps the modulation symbols onto frequency domain resources (i.e., subcarrier mapping).

[0078] Optionally, before mapping, the transmitter can also perform a discrete Fourier transform (DFT) on the modulation symbols, in which case the modulation symbols used for subcarrier mapping in step 22 can be replaced with the DFT-transformed symbols.

[0079] Optionally, before mapping, the transmitting end can also perform non-codebook transmission or codebook transmission precoding on the modulation symbols or DFT symbols, in which case the modulation symbols in the subcarrier mapping in step 22 can be replaced with the precoded symbols.

[0080] Step 23: The transmitter inserts DMRS into the mapped frequency domain symbols.

[0081] Step 24: The transmitting end performs an inverse fast fourier transform (IFFT) on the frequency domain signal obtained in step 23 to obtain the signal to be transmitted, which includes DMRS and data signals (the data signals are signals obtained based on the transport block).

[0082] Then, the transmitting end sends out the IFFT-compressed signal. After transmission through the channel, the signal received by the receiving end is affected by noise and / or interference compared to the signal sent by the transmitting end.

[0083] Step 25: The receiving end performs a fast fourier transform (FFT) on the received signal.

[0084] Step 26: The receiver performs frequency domain demapping on the FFT signal to determine the positions of the DMRS and data signals in the frequency domain.

[0085] Step 27: The receiver performs channel estimation based on DMRS, and detects and demodulates the data signal based on the channel estimation results to obtain the demodulated signal.

[0086] Step 28: Perform channel decoding on the demodulated signal to obtain the data.

[0087] To reduce interference between terminal devices, the DMRSs sent by network devices to different terminal devices are orthogonal. However, the length of DMRSs is limited, and the number of orthogonal DMRSs that can be carried by the same transmission resource is also limited. As the number of terminal devices increases, in order to reduce interference between terminal devices, orthogonal DMRSs can be sent to more terminal devices using more transmission resources. However, the increased transmission resource usage of DMRSs leads to a decrease in spectrum efficiency.

[0088] Based on this, embodiments of this application provide a communication method that can reduce the transmission resources occupied by the reference signal.

[0089] The methods provided in the various embodiments of this application can all be applied to... Figure 1 The network architecture shown or other network architectures. To be applied to... Figure 1For example, the network device involved in various embodiments of this application may be 110a, and the multiple terminal devices involved in various embodiments of this application may include multiple of 120i, 120a, 120b, or 120c; as another example, the network device involved in various embodiments of this application may be 120f, and the multiple terminal devices involved in various embodiments of this application may include 120h and 120g.

[0090] The following explanations of some terms or concepts used in the embodiments of this application are provided to facilitate understanding by those skilled in the art.

[0091] 1) Constellation Diagram: This can be understood as a vector coordinate system, where the horizontal axis represents the real part and the vertical axis represents the imaginary part. A constellation diagram consists of multiple points, called constellation points. When the transmitting end performs modulation, it can map information bits (e.g., "0100") onto a specific constellation point in the vector coordinate system.

[0092] 2) Block Error Rate (BLER), also known as bit error rate, refers to the probability of errors in a transmitted block after cyclic redundancy check (CRC) in a communication system. BLER is the percentage of erroneous blocks out of all transmitted blocks. The higher the BLER, the worse the performance of the communication system.

[0093] 3) Signal to Interference Plus Noise Ratio (SINR), also known as signal-to-noise ratio, is the ratio of the strength of the received useful signal to the strength of the received interference signal (noise and interference). A higher SINR indicates better signal quality and higher transmission reliability; conversely, a low SINR can lead to data transmission errors or unreliability.

[0094] 4) Throughput: The amount of data transmitted per unit of time. It is an important indicator for measuring network performance and can be measured in bytes per second.

[0095] 5) Orthogonal / non-orthogonal among multiple reference signals refers to the orthogonal / non-orthogonal relationships between sequences of multiple reference signals.

[0096] 6) The “neural network model” mentioned in the embodiments of this application can also be replaced with “artificial intelligence (AI) model”.

[0097] To better illustrate the embodiments of this application, the methods provided by the embodiments of this application are described below with reference to the accompanying drawings. Unless otherwise specified below, the steps indicated by dashed lines in the accompanying drawings corresponding to the various embodiments of this application are optional steps. In various embodiments of this application, the reference signal may include a downlink reference signal, which may include, for example, DMRS, or other downlink reference signals that can be used for channel estimation.

[0098] Please refer to Figure 3 This is a flowchart illustrating a communication method provided in an embodiment of this application. Wherein, in Figure 3 The text primarily describes the interaction process between network devices and the first terminal device. In practical applications, network devices can send reference signals to two, three, or even more terminal devices on the same transmission resource. The execution process of these additional terminal devices is similar to... Figure 3 The process performed by the first terminal device, as described in the illustrated embodiment, can be similar.

[0099] Step 301a: The network device sends a third signal to multiple terminal devices. The third signal includes a third reference signal and a third data signal.

[0100] Among them, multiple terminal devices include the first terminal device.

[0101] Step 301b: The first terminal device receives a first signal, which includes a first reference signal and a first data signal.

[0102] The third reference signals sent by the network device to different terminal devices are not orthogonal and occupy the same transmission resources. This can cause the amplitude and / or phase of the reference signal received by the terminal device to differ from the amplitude and / or phase of the third reference signal sent by the network device. Furthermore, interference, noise, or other factors in the channel between the network device and the terminal device can also cause the amplitude and / or phase of the reference signal received by the terminal device to differ from the reference signal sent by the network device. In this embodiment, for ease of description, the signal received by the first terminal device is referred to as the first signal, which includes a first reference signal and a first data signal.

[0103] The transmission resources occupied by the third data signals sent by the network device to different terminal devices can be the same or different, or partially the same. The third data signals carry downlink data sent by the network device to the terminal devices, and the downlink data sent by the network device to multiple terminal devices can be the same or different. The transmission resources include time-domain resources and / or frequency-domain resources. Optionally, the frequency-domain resources may not belong to 5G mobile communication resources. This can be understood as follows: the frequency-domain resources configured by the network device for terminal devices with the processing capability of a first neural network model (the first neural network model is the neural network model in this embodiment that processes information before demodulation) may not belong to 5G mobile communication resources, while the frequency-domain resources configured for terminal devices without the processing capability of the first neural network model may belong to 5G mobile communication resources, thereby reducing interference between the two types of terminal devices. The processing capability of the first neural network model can be applied to future mobile communication systems, and the resources of future mobile communication systems may differ from those of 5G mobile communication resources.

[0104] Network devices can indicate the transmission resources corresponding to reference signals and data signals to terminal devices, allowing the terminal devices to receive the reference signals and data signals on the corresponding transmission resources. For example, a network device can indicate the corresponding transmission resources to each of multiple terminal devices. Taking the network device indicating the transmission resources corresponding to a first reference signal (which can also be replaced by a third reference signal) as an example, for instance, the network device sends second indication information to the terminal device, and the terminal device receives the second indication information, which indicates the transmission resources corresponding to the first reference signal (which can also be replaced by a third reference signal).

[0105] The transmission resources corresponding to the reference signal and data signal can be pre-configured by the network device for the terminal device. After pre-configuration, the network device can update the transmission resources, for example, by using semi-static or dynamic scheduling. The network device can inform the terminal device of the updated transmission resources, and the terminal device will then receive the reference signal and data signal on the updated transmission resources. The network device can proactively update the transmission resources for one or more terminal devices, or one or more terminal devices can request the network device to update the transmission resources. For example, when a terminal device determines that the bit error rate of the data is higher than a set threshold, it can request more transmission resources for the reference signal from the network device.

[0106] Step 302: The terminal device inputs the first information into the first neural network model and obtains the second information output by the first neural network model.

[0107] The terminal device can determine the first information based on the first signal, where the first information is the information before demodulation. For example, the terminal device performs channel estimation based on the first reference signal to obtain a first channel estimation result; by detecting the first data signal based on the first channel estimation result, the first information can be obtained. In other words, the first information can be the information obtained by detecting the first data signal based on the first channel estimation result.

[0108] The input information of the first neural network model includes: pre-demodulation information obtained from non-orthogonal reference signals on the same transmission resource. The output information of the first neural network model includes: information obtained by processing the input information; the output information is also pre-demodulation information. In other words, the first neural network model is used to: process the pre-demodulation information obtained from non-orthogonal reference signals on the same transmission resource to obtain new pre-demodulation information. If the first information is pre-demodulation information, then the second information is also pre-demodulation information.

[0109] The first neural network model can process the input information. Optionally, one processing method could be correction, modification, rectification, or updating. The first neural network model does not modify the type of information; it only corrects, modifies, rectifyes, or updates the specific content / value of the information. Taking the correction of input information by the first neural network model as an example, the second information can be the corrected first information. Through the correction by the first neural network model, the information output by the first neural network model becomes more accurate than the input information.

[0110] For example, after receiving a first signal, a terminal device can determine first information based on the first signal in the manner described above. Following a traditional processing flow, the terminal device can demodulate the first information to obtain second data. For instance, the terminal device maps the first information to a corresponding constellation diagram to obtain the demodulated result of the first information (i.e., the second data). However, since the third reference signals sent by network devices to different terminal devices are non-orthogonal and occupy the same transmission resources, there may be interference between different third reference signals. Therefore, the first channel estimation result obtained by the first terminal device based on the first reference signal may not be accurate enough, which in turn may lead to inaccurate first information obtained based on the first channel estimation result. If the terminal device directly demodulates the first information, the error between the demodulated data and the original data may be large. Therefore, this application provides a first neural network model that can correct the first information, improving the accuracy of the second information compared to the first information. This results in a smaller, or even zero, error between the demodulated data and the original data, thereby improving the accuracy of the demodulation result.

[0111] For example, the difference between the first data and the original data is smaller than the difference between the second data and the original data. The first data is the data obtained by demodulating the second information, the second data is the data obtained by demodulating the first information, and the original data is the data that the network device wants to send to the first terminal device. The first terminal device may or may not be able to obtain the original data.

[0112] To give another example, network devices use 16QAM modulation, and terminal devices also use this modulation for demodulation. The constellation diagram for 16QAM can be found in [reference needed]. Figure 4 For example, the information bits corresponding to the third data signal sent by the network device are "1011", which corresponds to the constellation point in the upper left corner of the constellation diagram. After receiving the first data signal and the first reference signal, the first terminal device obtains a first channel estimation result based on the first reference signal. Because the first channel estimation result is not accurate enough, the information bits included in the first information obtained by the terminal device based on the first channel estimation result may become "0011", which corresponds to the constellation point in the upper right corner of the constellation diagram. After the first terminal device inputs the first information into the first neural network model, the first neural network model corrects the first information and outputs second information, which includes the information bits "1011". This is equivalent to the first information being corrected to "1011" by the first neural network model. The terminal device demodulates the second information according to the constellation diagram, and the difference between the obtained data and the original data is small, improving the demodulation accuracy.

[0113] Step 303: The terminal device demodulates the second information to obtain the first data.

[0114] In this embodiment, the reference signal received by the terminal device can be non-orthogonal. Compared to orthogonal reference signals carried on the same transmission resource, non-orthogonal reference signals carried on the same transmission resource can carry more data. Therefore, the network device can send non-orthogonal reference signals to more terminal devices with limited transmission resources, saving the transmission resource occupation of the reference signal while ensuring the transmission of the reference signal as much as possible. For a single terminal device, channel estimation can be performed based on the received reference signal. The information before demodulation can be determined based on the channel estimation result, and a first neural network model can be used to correct the information before demodulation. Demodulation based on the corrected information before demodulation results in more accurate data. If the terminal device directly demodulates based on the information before demodulation obtained from the non-orthogonal reference signal, the demodulated data may be inaccurate. However, this embodiment uses a first neural network model to reduce the error caused by the non-orthogonal reference signal, making the obtained data more accurate and thus reducing the bit error rate.

[0115] In step 302, the terminal device inputs the first information into the first neural network model. The first neural network model can process the first information. In this embodiment of the application, the mode of "inputting the information before demodulation into the first neural network model and intelligently processing the information before demodulation through the first neural network model" is called the intelligent receiving mode.

[0116] Understandably, in real-world scenarios, some terminal devices possess the capability to process a first neural network model and can operate in intelligent receiving mode; others do not, and therefore cannot operate in intelligent receiving mode. In one possible implementation, before step 301b, the terminal device may send its first capability to the network device, indicating that it possesses the capability to process a first neural network model. This allows the network device to send non-orthogonal reference signals to multiple terminal devices with the capability to process a first neural network model on the same transmission resource. If multiple terminal devices do not possess the capability to process a first neural network model, the network device may send non-orthogonal or orthogonal reference signals to these devices on different transmission resources, or it may send orthogonal reference signals to multiple terminal devices on the same transmission resource.

[0117] In one possible implementation, prior to step 301a, the network device may further send indication information to the first terminal device (a terminal device with the capability to process a first neural network model), which indicates the activation of the intelligent reception mode. Upon receiving this indication information, the first terminal device activates the intelligent reception mode. Then, the first terminal device can execute step 302. Further, it is possible that after receiving information from the first terminal device indicating that it has the capability to process a first neural network model, the network device sends indication information to the first terminal device to indicate the activation of the intelligent reception mode.

[0118] Furthermore, network devices can pre-configure transmission resources and data processing parameters for terminal devices based on their capabilities. For example, the frequency domain resources configured by the network device for a terminal device with first neural network model processing capabilities are not 5G mobile communication resources. Data processing parameters include, but are not limited to: modulation and coding scheme (MCS), modulation method, and stream number.

[0119] In step 302 above, the first terminal device inputs the first information into the first neural network model. Therefore, before step 302, the first terminal device can determine the first neural network model. The following describes several possible ways in which the first terminal device determines the first neural network model:

[0120] Method 1: The second device instructs the first neural network model on the first terminal device. In this method, the first terminal device does not need to participate in the training of the first neural network model, which simplifies the implementation of the first terminal device. The second device may be, for example, a network device serving the first terminal device, or it may be other devices besides the network device, such as operation administration and maintenance (OAM) devices or other terminal devices besides the first terminal device.

[0121] For example, the second device sends first indication information to the first terminal device, and the first terminal device receives the first indication information, which is used to indicate a first neural network model. For example, the first indication information indicates the index of the first neural network model and / or the parameters of the first neural network model. The parameters of the first neural network model include one or more of the following: the number of layers included in the first neural network model, the number of neurons in each of at least one layer included in the first neural network model, the connection relationships of some or all of the neurons included in the first neural network model, or the weights of some or all of the neurons included in the first neural network model.

[0122] Optionally, before sending the first instruction information to the first terminal device, the first terminal device may send instruction information to the second device to instruct the acquisition of the first neural network model. That is, the second device may proactively instruct the first terminal device to acquire the first neural network model, or it may instruct the first terminal device to acquire the first neural network model based on a request from the first terminal device. Optionally, after activating the intelligent receiving mode, the first terminal device may send instruction information to the second device to instruct the acquisition of the first neural network model.

[0123] Optionally, if the first terminal device correctly receives the parameters of the first neural network model, it can load the parameters of the first neural network model and send a loading completion indication message to the second device. If the first terminal device does not correctly receive the parameters of the first neural network model, it can send a failure to receive indication message to the second device, and the second device can then instruct the first terminal device to retrieve the parameters of the first neural network model again.

[0124] The second device can train a first neural network model. Optionally, the second device can train the first neural network model independently, or it can train the first neural network model jointly with other devices (such as OAM, other network devices, or one or more terminal devices). Whether training independently or jointly, the second device can identify multiple training samples and train the first neural network model based on these multiple training samples.

[0125] Alternatively, the second device may not need to train the model itself, but instead obtain the first neural network model from another device (e.g., OAM, or other network devices). For example, the other device trains the first neural network model and then instructs the second device on it. For instance, the second device receives third instruction information from the other device, which is used to instruct the first neural network model. For example, the third instruction information indicates the index of the first neural network model and / or the parameters of the first neural network model; for details on how to instruct the parameters of the first neural network model, please refer to the preceding description.

[0126] Method 2: The first terminal device trains the first neural network model.

[0127] The first terminal device can independently train the first neural network model, or it can jointly train the first neural network model with other devices (such as OAM, network devices, or one or more other terminal devices). Whether training independently or jointly, the first terminal device can determine multiple training samples and use these samples to train the first neural network model. Optionally, some or all of these multiple training samples can be obtained from network devices or OAM, or they can be determined by the first terminal device.

[0128] The participation of the first terminal device in training the neural network model can make the trained neural network model more suitable for the first terminal device's situation. For example, the channel conditions between the first terminal device and the network device can further reduce the bit error rate.

[0129] As time goes by, the channel conditions between network devices and terminal devices will change. In order to reduce the bit error rate, the parameters of the first neural network model can be updated. For example, the parameters of the first neural network model can be updated periodically, or they can be updated when certain conditions are met. For example, when the accuracy of the output result of the first neural network model decreases, or when the difference between the data obtained from the first neural network model and the original data is large, or when the bit error rate corresponding to the first neural network model is high, the parameters of the first neural network model can be updated to make the first neural network model more accurate.

[0130] When updating the model's parameters, the update can be performed by the terminal device, or by the network device, or by OAM, and then the updated parameters can be sent to the terminal device.

[0131] Taking the updating of model parameters by a network device as an example: The network device updates the parameters of the first neural network model based on multiple training samples (these multiple training samples may include: training samples used when training the first neural network model and / or new training samples). The network device sends the updated parameter information of the first neural network model to the first terminal device. After receiving the updated parameter information, the first terminal device updates the parameters of the first neural network model. It is understood that, in order to ensure that the first terminal device can obtain accurate updated parameters, the network device uses orthogonal reference signals and / or sends the updated parameters to the first terminal device on different transmission resources to avoid the influence of interference. Further optionally, before sending the updated parameter information to the first terminal device, the terminal device may send a first request to the network device, and the network device receives the first request accordingly; wherein, the first request is used to request the update of the parameters of the first neural network model. When the terminal device determines that the update cycle has arrived, or when it determines that the bit error rate does not meet the requirements after demodulation based on the output information of the first neural network model, it can actively request to update the parameters of the neural network model, which can improve the timeliness of model correction.

[0132] The following is an example of how the parameters of the first terminal device update model are described. Figure 5 As shown.

[0133] Step 51: The first terminal device sends a first request to the network device, and the network device receives the first request accordingly. The first request is used to request an update to the parameters of the first neural network model.

[0134] Step 51 is optional. When the terminal device determines that the update cycle has been reached, or when it determines that the bit error rate does not meet the requirements after demodulation based on the output information of the first neural network model, it can actively request to update the parameters of the neural network model, which can improve the timeliness of model correction.

[0135] In one possible example, after receiving the first request from the terminal device, the network device initiates the smart receiver parameter update mode.

[0136] Step 52: The network device sends a second signal to the first terminal device, and the first terminal device receives the second signal accordingly; wherein, the second signal includes a second reference signal and a second data signal, and the data corresponding to the second data signal is used to update the parameters of the first neural network model, that is, the data carried in the second data signal is the training sample.

[0137] The frame structure corresponding to the second signal can be called a training frame, or the frame structure corresponding to the second data signal can be called a training frame.

[0138] Optionally, the second signal further includes a control signal, which indicates that the data corresponding to the second data signal is a training sample, or indicates that the second signal is used to update the first neural network model. Alternatively, an indication information may be included in other signals different from the second signal, indicating that the data corresponding to the second data signal is a training sample, or indicating that the second signal is used to update the first neural network model. Through the indication of the control signal, the first terminal device can accurately determine the parameters of the second signal / second data signal used to update the model.

[0139] Step 53: The first terminal device obtains the fourth information before demodulation based on the second signal.

[0140] For example, the second reference signal is used to perform channel estimation to obtain a second channel estimation result. Based on the second channel estimation result, the second data signal is detected to obtain the information before demodulation, i.e., the fourth information.

[0141] Step 54: The first terminal device can demodulate the fourth information to obtain the second data, i.e., the training samples.

[0142] Alternatively, step 54 can be replaced by the following process: The first terminal device inputs the fourth information into the first neural network model, and the first neural network model outputs third information, which is also the information before demodulation. The terminal device can demodulate the third information to obtain the second data. After processing by the first neural network model, the third information becomes closer to the information sent by the network device, which can improve the accuracy of the training samples. The terminal device updates the parameters of the first neural network model, which makes the updated neural network model more consistent with the channel conditions between the terminal device and the network device, which can further reduce the bit error rate.

[0143] Step 55: The first terminal device updates the parameters of the first neural network model based on the second data.

[0144] For example, the terminal device determines training samples based on training frames. The training samples include predicted sample information and actual sample information. The actual sample information is the training label of the predicted sample information. The parameters of the first neural network model are then updated.

[0145] After the update, the first terminal device saves the updated parameters.

[0146] Steps 56 through 59 are all optional.

[0147] Step 56: After updating the parameters of the first neural network model, the first terminal device sends an indication message to the network device indicating that the model parameter update is complete.

[0148] Step 57: After receiving the indication information that the model parameter update is complete from the first terminal device, the network device sends a second request to the first terminal device. The second request is used to request the updated parameters.

[0149] Step 58: After receiving the second request, the first terminal device sends the updated parameter information to the network device.

[0150] In another possible implementation, the updated parameter information can be sent to the network device along with the indication that the model parameter update is complete, in which case the network device does not need to execute step 57. Alternatively, after updating the parameters of the first neural network model, the terminal device sends the updated parameter information to the network device, using the updated parameter information to indicate that the model parameter update is complete, without needing to send the additional indication that the model parameter update is complete. Optionally, in this case, step 57 can also be omitted.

[0151] Step 59: After receiving the updated parameter information, the network device saves the updated parameters.

[0152] This is so that when other terminal devices connect to the network device later, the parameters of the updated neural network model can be sent to those other terminal devices.

[0153] Optionally, network devices can also aggregate updated parameters from multiple terminal devices.

[0154] The following describes an example of training a neural network model. The first neural network model mentioned above can be trained in the manner described below.

[0155] like Figure 6 The diagram illustrates the training process of a neural network model.

[0156] Step 601: Determine N sample signals; wherein the i-th sample signal includes the i-th sample reference signal and the i-th sample data signal, the N sample reference signals are non-orthogonal and occupy the same transmission resources, the value of i traverses the integers from 1 to N, and N is an integer greater than or equal to 2.

[0157] Step 602: Determine the information of the i-th predicted sample.

[0158] Step 603: Determine the information of the i-th actual sample.

[0159] In one example, the i-th predicted sample information is: information before demodulation obtained by detecting the i-th sample data signal based on the channel estimation result of the i-th sample, wherein the channel estimation result of the i-th sample is obtained by channel estimation based on the reference signal of the i-th sample; and the i-th actual sample information is information after modulation.

[0160] In another example, the i-th predicted sample information is: the i-th data obtained by demodulating the i-th information before demodulation; the i-th information before demodulation is obtained by detecting the i-th sample data signal based on the channel estimation result of the i-th sample, and the channel estimation result of the i-th sample is obtained by channel estimation based on the reference signal of the i-th sample. The i-th actual sample information is the original data.

[0161] Step 604: Use the i-th actual sample information and the i-th predicted sample information as a training sample; wherein, the i-th actual sample information can be the training label of the i-th predicted sample information.

[0162] Step 605: Train the neural network model based on multiple training samples.

[0163] The following describes the process of determining training samples for different scenarios:

[0164] Scenario 1: A network device and N training devices jointly determine training samples. The N training devices may include terminal devices dedicated to model training and / or... Figure 3 The first terminal device in the process.

[0165] In step 601, the network device determines N sample signals.

[0166] In step 602, the network device may determine the i-th prediction sample information in the following manner: for example, the network device sends sample signals to N training devices, wherein the sample signal sent to the i-th training device includes the i-th sample reference signal and the i-th sample data signal, the N sample reference signals are non-orthogonal and occupy the same transmission resources, the value of i traverses the integers from 1 to N, and N is an integer greater than or equal to 2.

[0167] The training device performs channel estimation based on the received sample reference signal to obtain a sample channel estimation result; then, based on the sample channel estimation result, it detects the received sample data to obtain information before demodulation; the training device sends this information before demodulation as predicted sample information to the network device; or, the training device demodulates the information before demodulation and sends the demodulated data as predicted sample information to the network device. For example, the i-th training device sends the i-th predicted sample information to the network device; correspondingly, the network device receives the i-th predicted sample information from the i-th training device. In one example, the i-th predicted sample information is the information before demodulation, and the i-th predicted sample information is: information obtained by the i-th training device detecting the i-th sample data signal based on the i-th sample channel estimation result, where the i-th sample channel estimation result is obtained by the i-th training device performing channel estimation based on the i-th sample reference signal. In another example, the i-th predicted sample information is: the i-th data obtained by demodulating the i-th information before demodulation; the i-th information before demodulation is obtained by detecting the i-th sample data signal based on the channel estimation result of the i-th sample, and the channel estimation result of the i-th sample is obtained by channel estimation based on the reference signal of the i-th sample.

[0168] In step 603, the network device can determine the i-th actual sample information corresponding to the i-th sample data signal. In one example, the i-th actual sample information is modulated information. For example, the network device determines the i-th actual sample information based on the original data corresponding to the i-th sample data signal; the original data may be data encoded by the network device for the transport block. In another example, the i-th actual sample information is the original data corresponding to the i-th sample data signal.

[0169] In step 605, the network device may train the neural network model based on multiple training samples; alternatively, the network device may send multiple training samples to other devices (such as other network devices, OAM, etc.) so that the other devices can train the neural network model based on the multiple training samples.

[0170] Scenario 2: The third device uses simulation to determine the training samples.

[0171] The execution entity for steps 601 to 604 is a third device. The process executed by the third device is similar to that in scenario 1, but it does not need to send sample signals to the training device or receive prediction sample information from the training device. Specific details will not be repeated. In step 605, the third device may train the neural network model based on multiple training samples; alternatively, the third device may send multiple training samples to other devices (such as network devices, OAM, terminal devices, etc.), and the other devices may train the neural network model based on the multiple training samples.

[0172] Scenario 3: The terminal device determines the training samples.

[0173] The terminal device can be a terminal device specifically designed for model training or Figure 3 The first terminal device in the process.

[0174] In step 601, the network device determines N sample signals. Then, the network device sends the sample signals to the N training devices, wherein the sample signal sent to the i-th training device includes the i-th sample reference signal and the i-th sample data signal. The N sample reference signals are non-orthogonal and occupy the same transmission resources. The value of i traverses an integer from 1 to N, and N is an integer greater than or equal to 2.

[0175] In step 602, the i-th terminal device can determine the i-th predicted sample information in the following manner: for example, the terminal device receives a sample signal from a network device; the sample signal includes a sample reference signal and a sample data signal; the terminal device determines the predicted sample information based on the sample signal. In one example, the predicted sample information is the information before demodulation obtained by detecting the sample data signal based on the sample channel estimation result, where the sample channel estimation result is obtained by channel estimation based on the sample reference signal. In another example, the predicted sample information is: data obtained by demodulating the information before demodulation; the information before demodulation is obtained by detecting the sample data signal based on the sample channel estimation result, where the sample channel estimation result is obtained by channel estimation based on the sample reference signal.

[0176] In step 603, the terminal device can determine the actual sample information corresponding to the sample data signal. In one example, the actual sample information is modulated information. For instance, the terminal device determines the actual sample information based on the original data corresponding to the sample data signal; the original data may be data encoded by the network device into transport blocks. In another example, the actual sample information is the original data corresponding to the sample data signal. The original data corresponding to the sample data signal is known to both the terminal device and the network device.

[0177] In step 605, the terminal device may train the neural network model based on multiple training samples; alternatively, the terminal device may send multiple training samples to other devices (such as other network devices, OAM, other terminal devices, etc.) so that the other devices can train the neural network model based on the multiple training samples.

[0178] Taking the training of neural network models using a dedicated terminal as an example, such as Figure 7 The diagram illustrates a communication process for training a neural network model. Specifically, in... Figure 7 This paper mainly introduces the interaction process between a network device and a dedicated training terminal. In practical applications, the network device can send sample reference signals to two, three, or even more dedicated training terminals on the same transmission resource. The execution process of these additional dedicated training terminals is similar to... Figure 7 The process performed by the terminal device, as mainly described in the illustrated embodiments, can be similar.

[0179] Step 71: The network device sends instruction information to multiple dedicated training terminals, which is used to indicate the start of initial training of the neural network model.

[0180] For example, the network device initiates the initial training mode of the intelligent receiver, deploys multiple dedicated training terminals within the site, and sends instruction information to these dedicated terminals.

[0181] Step 72: After receiving the instruction information, the dedicated training terminal loads the neural network model to be trained.

[0182] Step 73: After loading the neural network model to be trained, the dedicated training terminal sends an instruction message to the network device. This instruction message is used to instruct the dedicated training terminal to wait for model training.

[0183] Step 74: After receiving the instruction information from step 73, the network device sends sample signals to multiple dedicated training terminals. The sample signals include sample reference signals and sample data signals.

[0184] Among them, the sample reference signals sent to multiple training terminal devices occupy the same transmission resources, and the multiple sample reference signals are not orthogonal.

[0185] Step 75: After receiving the sample signal, the dedicated training terminal determines the training sample based on the sample signal.

[0186] The process of determining training samples based on sample signals can be referred to in steps 602, 603 and 604, and will not be repeated here.

[0187] Network devices can repeatedly send sample signals to multiple dedicated training terminals, so that a single dedicated training terminal can obtain multiple training samples.

[0188] Step 76: The dedicated training terminal trains the neural network model loaded in step 72 based on multiple training samples.

[0189] Steps 77 to 80 are optional.

[0190] Step 77: After training the neural network model, the dedicated training terminal sends a training completion indication message to the network device.

[0191] Furthermore, the dedicated training terminal saves the parameters of the trained neural network model.

[0192] Step 78: After receiving the instruction message that training is complete, the network device sends a request to the dedicated training terminal to obtain the model parameters.

[0193] Step 79: After receiving the request, the dedicated training terminal sends the parameter information of the trained neural network model to the network device.

[0194] Step 80: After receiving the parameter information of the trained neural network model, the network device stores the model parameters.

[0195] This allows the network device to send the parameters of the trained neural network model to any subsequent terminal devices that connect to the network. Optionally, the network device can also aggregate parameters from multiple dedicated training terminals.

[0196] The following diagram illustrates the changes in signal-to-noise ratio (SINR) and block error rate (BLER) in three scenarios.

[0197] Scenario 1: Ideal scenario, the channel between network devices and terminal devices is interference-free and noise-free. The network device sends orthogonal DMRS to the terminal device.

[0198] Scenario 2: In a typical 5G baseline scenario, there is interference and / or noise in the channel between the network device and the terminal device. The network device sends orthogonal DMRS to the terminal device, and the terminal device does not use a neural network model to process the information before demodulation.

[0199] Scenario 3: Intelligent processing scenario, the channel between network devices and terminal devices is subject to interference and / or noise, the network device sends non-orthogonal DMRS to multiple terminal devices on the same transmission resource, and the terminal devices use a neural network model to process the information before demodulation.

[0200] like Figure 8a , Figure 8b ,like Figure 8cAs shown, a schematic diagram illustrating the changes in signal-to-noise ratio and block error rate under different MCS is presented, with the horizontal axis representing signal-to-noise ratio and the vertical axis representing bit error rate. Figure 8a This is a schematic diagram for an MCS index of 22. Figure 8b This is a diagram illustrating an MCS index of 16. Figure 8c This is a schematic diagram of an MCS index of 8.

[0201] like Figure 8a , Figure 8b ,and Figure 8c As shown, when the bit error rate is 0, the signal-to-noise ratios (SNRs) of scenarios 3 and 1 are almost the same, while the SNR of scenario 2 is relatively high. When BLER = 0, the lower the SINR, the more reliable the communication system can be even under poor channel conditions and low SNR.

[0202] like Figure 8d The diagram illustrates the changes in signal-to-noise ratio (SNR) and throughput in these three scenarios. It can be seen that the throughput of scenario 3 is higher than that of scenario 1. Scenario 3 uses non-orthogonal DMRS, while scenario 1 uses orthogonal DMRS, allowing a limited number of users to connect. Therefore, scenario 3 allows more users to connect and has a higher throughput than scenario 1.

[0203] It is understood that, in order to achieve the functions in the above embodiments, the network device and terminal device include hardware structures and / or software modules corresponding to perform each function. Those skilled in the art should readily recognize that, based on the units and method steps of the various examples described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0204] Figure 9 and Figure 10 The diagram illustrates the possible communication devices provided in the embodiments of this application. These communication devices can be used to implement the functions of the network devices and terminal devices in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0205] like Figure 9 As shown, the communication device 900 includes a processing unit 910 and a transceiver unit 920.

[0206] For example, the communication device 900 is used to achieve the above. Figure 3 , Figures 5 to 7The methods illustrated in the embodiments show the functions of the network device and the terminal device. The transceiver unit 920 can perform the receiving and sending actions performed by the network device and the terminal device in the above method embodiments. The processing unit 910 can perform other actions performed by the network device and the terminal device in the above method embodiments besides the sending and receiving actions.

[0207] For example, when the communication device 900 is used to implement Figure 3 In the method embodiment shown, when the network device functions as described, the transceiver unit 920 is used to transmit a third signal. The processing unit 910 is used to generate the third signal.

[0208] For example, when the communication device 900 is used to implement Figure 3 In the method embodiment shown, when the terminal device functions as described, the transceiver unit 920 is used to receive a first signal. The processing unit 910 is used to execute steps 302 and 303.

[0209] For a more detailed description of the processing unit 910 and the transceiver unit 920, please refer to [link / reference needed]. Figure 3 , Figures 5 to 7 The relevant descriptions in the illustrated method embodiments are directly obtained and will not be repeated here. The processing unit 910 can be implemented by a processor, and the transceiver unit 920 can be implemented by a transceiver.

[0210] like Figure 10 As shown, the communication device 1000 includes a processor 1010 and an interface circuit 1020. The processor 1010 and the interface circuit 1020 are coupled to each other. It is understood that the interface circuit 1020 can be a transceiver or an input / output interface. Optionally, the communication device 1000 may also include a memory 1030 for storing instructions executed by the processor 1010, or storing input data required by the processor 1010 to execute instructions, or storing data generated after the processor 1010 executes instructions. Sometimes, the interface circuit 1020 can also be understood as part of the processor 1010, in which case the communication device 1000 includes the processor 1010.

[0211] When the communication device 1000 is used to achieve the above Figure 3 , Figures 5 to 7 In the method shown, the processor 1010 is used to implement the functions of the processing unit 910, and the interface circuit 1020 is used to implement the functions of the transceiver unit 920.

[0212] When the aforementioned communication device is a chip applied to a terminal device, the terminal device chip implements the functions of the terminal device in the above method embodiments. The terminal device chip receives information from a network device, which can be understood as the information being first received by other modules (such as an RF module or antenna) in the terminal device, and then sent to the terminal device chip by these modules. The terminal device chip sends information to a network device, which can be understood as the information being first sent to other modules (such as an RF module or antenna) in the terminal device, and then sent to the network device by these modules.

[0213] When the aforementioned communication device is a chip applied to a network device, the network device chip implements the functions of the network device in the above method embodiments. The network device chip receives information from the terminal device, which can be understood as the information being first received by other modules (such as radio frequency modules or antennas) in the network device, and then sent to the network device chip by these modules. The network device chip sends information to the terminal device, which can be understood as the information being sent down to other modules (such as radio frequency modules or antennas) in the network device, and then sent to the terminal device by these modules. Here, the network device module can be the baseband chip of the network device, or a DU (Digital Unit) or other modules. The DU here can be a DU under the Open Radio Access Network (O-RAN) architecture.

[0214] In this application, entity A sends information to entity B, either directly or indirectly through other entities. Similarly, entity B receives information from entity A, either directly or indirectly through other entities. Entities A and B can be network devices or terminal devices, or modules within network devices or terminal devices. The sending and receiving of information can be between network devices and terminal devices, between two network devices (e.g., CU and DU), or between different modules within a single device (e.g., a terminal device chip and other modules within the terminal device, or a network device chip and other modules within the network device).

[0215] 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.

[0216] This application also provides a computer-readable storage medium storing a computer program that, when executed by a computer, enables the computer to perform the aforementioned communication method. Alternatively, the computer program includes instructions for implementing the aforementioned communication.

[0217] This application also provides a computer program product, including: computer program code, which, when run on a computer, enables the computer to execute the communication method provided above.

[0218] This application also provides a communication system, which includes a network device and a terminal device that perform the above-described communication method.

[0219] The method steps in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, compact disc read-only memory (CD-ROM), or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a base station or terminal. Of course, the processor and storage medium can also exist as discrete components in the base station or terminal.

[0220] In the above embodiments, implementation can be achieved 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. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a first control plane network element, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer program or instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video optical disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both types of storage media.

[0221] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0222] In this application embodiment, the number of nouns, unless otherwise specified, refers to "singular nouns or plural nouns," that is, "one or more." "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 exists alone, A and B exist simultaneously, or B exists alone, where A or B can be singular or plural. The character " / " generally indicates that the related objects before and after are in an "or" relationship. For example, A / B means: A or B. Expressions such as "at least one of the following" or "one or more of them" 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, or one or more of a, b, or c, means: a, b, c, a and b, a and c, b and c, or a and b and c. Each of a, b, and c can be single or multiple.

[0223] The ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects, and are not used to limit the size, content, order, timing, priority, or importance of the multiple objects. Furthermore, such names do not indicate differences in the content, sending / receiving end, sending order, size, application scenario, priority, or importance of the two pieces of information. Additionally, the numbering of steps in the various embodiments described in this application is only to distinguish different steps and is not used to limit the order of steps.

Claims

1. A communication method, characterized in that, include: Receive a first signal, the first signal including a first reference signal and a first data signal; The first information is input into the first neural network model to obtain the second information output by the first neural network model. The first information is the information before demodulation obtained by detecting the first data signal based on the first channel estimation result. The first channel estimation result is obtained by channel estimation based on the first reference signal. The input information of the first neural network model includes the information before demodulation obtained based on the non-orthogonal reference signal on the same transmission resource. The output information of the first neural network model includes the information obtained by processing the input information. The second information is demodulated to obtain the first data.

2. The method as described in claim 1, characterized in that, The difference between the first data and the original data is less than the difference between the second data and the original data, where the second data is the data obtained by demodulating the first information.

3. The method as described in claim 1 or 2, characterized in that, The method further includes: Receive first indication information, which is used to instruct the first neural network model; or... Multiple training samples are determined, and the first neural network model is trained based on the multiple training samples.

4. The method as described in claim 3, characterized in that, Determine a training sample, including: Receive sample signals; the sample signals include sample reference signals and sample data signals; Predicted sample information is determined based on the sample signal; wherein, the predicted sample information is the information before demodulation obtained by detecting the sample data signal based on the sample channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal; Determine the actual sample information corresponding to the sample data signal, wherein the actual sample information is the modulated information; The actual sample information and the predicted sample information are used to determine a training sample.

5. The method according to any one of claims 1-3, characterized in that, The method further includes: Receive sample signals, the sample signals including sample reference signals and sample data signals; Sending predicted sample information, which is used to train the first neural network model; wherein, the predicted sample information is the information before demodulation obtained by detecting the sample data signal based on the sample channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal.

6. The method according to any one of claims 1-5, characterized in that, The method further includes: Receive a second signal, the second signal including a second reference signal and a second data signal; The third information is demodulated to obtain the second data; wherein the third information is determined based on the fourth information, which is the information before demodulation obtained by detecting the second data signal based on the second channel estimation result, and the second channel estimation result is obtained by channel estimation based on the second reference signal; Based on the second data, the parameters of the first neural network model are updated.

7. The method according to claim 6, characterized in that, The second signal also includes a control signal, which is used to indicate that the data corresponding to the second data signal is a training sample, or to indicate that the second signal is used to update the first neural network model.

8. The method as described in claim 6 or 7, characterized in that, The third information is the information output by the first neural network model after the fourth information is input into it.

9. The method according to any one of claims 1-8, characterized in that, The method further includes: Send a first request, which is used to request an update of the parameters of the first neural network model.

10. The method according to any one of claims 1-9, characterized in that, The method further includes: Receive second indication information, which is used to indicate the transmission resources corresponding to the first reference signal, wherein the frequency domain resources in the transmission resources do not belong to the 5G mobile communication resources.

11. A communication method, characterized in that, Performed by a network device or a chip used in a network device, including: Send first indication information to multiple terminal devices, the first indication information being used to indicate a first neural network model; wherein, the input information of the first neural network model includes undemodulated information obtained from non-orthogonal reference signals on the same transmission resource, and the output information of the first neural network model includes information obtained by processing the input information.

12. The method as described in claim 11, characterized in that, Also includes: A third signal is sent to the plurality of terminal devices, the third signal including a third reference signal and a third data signal; The multiple third reference signals are not orthogonal and occupy the same transmission resources.

13. The method as described in claim 11 or 12, characterized in that, The method further includes: The first neural network model is obtained by training multiple training samples; or... Receive third indication information, which is used to indicate the first neural network model.

14. The method as described in claim 13, characterized in that, The method further includes: Sample signals are sent to N training devices; wherein the sample signal sent to the i-th training device among the N training devices includes the i-th sample reference signal and the i-th sample data signal, the N sample reference signals sent to the N training devices are non-orthogonal and occupy the same transmission resources, the value of i is an integer from 1 to N, and N is an integer greater than or equal to 2; Receive the i-th predicted sample information from the i-th training device; wherein, the i-th predicted sample information is the information before demodulation obtained by the i-th training device detecting the i-th sample data signal based on the channel estimation result of the i-th sample, and the channel estimation result of the i-th sample is obtained by the i-th training device performing channel estimation based on the reference signal of the i-th sample; The information of the i-th actual sample is determined based on the i-th sample data signal, wherein the i-th actual sample information is the modulated information; Using the i-th actual sample information and the i-th predicted sample information as a training sample, determine some or all of the training samples among the multiple training samples.

15. The method according to any one of claims 11-14, characterized in that, The method further includes: A second signal is sent to a first terminal device. The second signal includes a second reference signal and a second data signal. The data corresponding to the second data signal is used to update the parameters of the first neural network model. The first terminal device belongs to the plurality of terminal devices.

16. The method as described in claim 15, characterized in that, The second signal also includes a control signal, which is used to indicate that the data corresponding to the second data signal is a training sample, or to indicate that the second signal is used to update the first neural network model.

17. The method as described in claim 15 or 16, characterized in that, Before sending the second signal to the first terminal device, it also includes: A first request is received from the first terminal device; wherein the first request is used to request an update of the parameters of the first neural network model.

18. The method according to any one of claims 11-17, characterized in that, The method further includes: Send a second indication message to the plurality of terminal devices; wherein the second indication message is used to indicate the transmission resources corresponding to the third reference signal, and the frequency domain resources in the transmission resources do not belong to the 5G mobile communication resources.

19. A communication method, characterized in that, include: Sample signals are sent to N training devices; wherein the sample signal sent to the i-th training device among the N training devices includes the i-th sample reference signal and the i-th sample data signal, the N sample reference signals sent to the N training devices are non-orthogonal and occupy the same transmission resources, the value of i is an integer from 1 to N, and N is an integer greater than or equal to 2; Receive the i-th predicted sample information from the i-th training device; wherein, the i-th predicted sample information is the information before demodulation obtained by the i-th training device detecting the i-th sample data signal based on the channel estimation result of the i-th sample, and the channel estimation result of the i-th sample is obtained by the i-th training device performing channel estimation based on the reference signal of the i-th sample; The information of the i-th actual sample is determined based on the i-th sample data signal, wherein the i-th actual sample information is the modulated information; The neural network model is trained using the information of the i-th actual sample and the information of the i-th predicted sample as a training sample.

20. A communication method, characterized in that, include: Receive sample signals; the sample signals include sample reference signals and sample data signals; Predicted sample information is determined based on the sample signal; wherein, the predicted sample information is the information before demodulation obtained by detecting the sample data signal based on the sample channel estimation result, and the sample channel estimation result is obtained by channel estimation based on the sample reference signal; The actual sample information is determined based on the sample data signal, and the actual sample information is the modulated information. The actual sample information and the predicted sample information are used as a training sample to train the neural network model.

21. A communication device, characterized in that, Includes a module for performing the method as described in any one of claims 1-20.

22. A communication device, characterized in that, Includes a processor, which is coupled to a memory; The memory is used to store computer programs or instructions; The processor is configured to execute some or all of the computer programs or instructions in the memory, and when the some or all of the computer programs or instructions are executed, to implement the method as described in any one of claims 1-20.

23. A communication device, characterized in that, Including processor and memory; The memory is used to store computer programs or instructions; The processor is configured to execute some or all of the computer programs or instructions in the memory, and when the some or all of the computer programs or instructions are executed, to implement the method as described in any one of claims 1-20.

24. A chip system, characterized in that, Includes a processor and interface circuitry, wherein the processor is coupled to a memory; The memory is used to store computer programs or instructions; The processor is configured to execute some or all of the computer programs or instructions in the memory, and when the some or all of the computer programs or instructions are executed, to implement the method as described in any one of claims 1-20.

25. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions, which, when executed by a communication device, implement the method as described in any one of claims 1-20.

26. A computer program product, characterized in that, The computer program product includes: computer instructions that, when executed on a computer, cause the method as described in any one of claims 1-20 to be implemented.

Images