Signal processing methods, first devices and second devices

Abstract

The present application relates to signal processing methods, first devices, second devices, a chip, a computer-readable storage medium, a computer program product, a computer program and a communication system. A method comprises: a first device uses a first AI model to perform pre-distortion processing on a first digital signal output by a baseband link, so as to obtain a second digital signal; and the first device processes the second digital signal into a first radio frequency signal. The embodiments of the present application can improve the compensation effect of pre-distortion processing, thus improving the communication performance.

Description

Signal processing method, first device and second device Technical Field

[0001] The present application relates to the field of communications, and more specifically, to a signal processing method, a first device, a second device, a chip, a computer-readable storage medium, a computer program product, a computer program, and a communication system. Background Art

[0002] Both the radio frequency (RF) link and the baseband link play a crucial role in modern wireless communication systems. The baseband link is primarily responsible for signal processing and protocol handling, including digital signal modulation and demodulation, channel encoding and decoding, source encoding and decoding, and some signaling processing. The RF link, on the other hand, is primarily responsible for converting the processed baseband link signal through digital-to-analog conversion (DAC) into an electromagnetic signal suitable for transmission, or receiving external signals and performing analog-to-digital conversion (ADC) before transmitting them to the baseband link.

[0003] However, communication transceivers have non-ideal RF characteristics, and RF links introduce various noises and nonlinear distortion into signals. As wireless communication frequencies increase and the bandwidths required grow, the degradation in communication performance caused by these non-ideal characteristics is becoming increasingly significant.

[0004] Summary of the Invention

[0005] Embodiments of the present application provide a signal processing method, a first device, a second device, a chip, a computer-readable storage medium, a computer program product, a computer program, and a communication system for improving communication performance.

[0006] The present invention provides a signal processing method, including:

[0007] The first device uses a first AI (Artificial Intelligence) model to perform predistortion processing on a first digital signal output by a baseband link to obtain a second digital signal;

[0008] The first device processes the second digital signal into a first radio frequency signal.

[0009] The present invention provides a signal processing method, including:

[0010] The second device receives a first RF signal sent by the first device; wherein the first RF signal is obtained by processing the second digital signal, and the second digital signal is obtained by pre-distorting the first digital signal output by the baseband link of the first device using the first AI model.

[0011] An embodiment of the present application provides a first device, including:

[0012] The first processing module is used to use the first AI model to perform pre-distortion processing on the first digital signal output by the baseband link to obtain a second digital signal, and process the second digital signal into a first RF signal.

[0013] An embodiment of the present application provides a second device, including:

[0014] The second communication module is used to receive a first RF signal sent by the first device; wherein the first RF signal is obtained by the first device processing the second digital signal, and the second digital signal is obtained by using the first AI model to pre-distort the first digital signal output by the baseband link of the first device.

[0015] An embodiment of the present application provides a first device, comprising: a transceiver, a processor, and a memory. The memory is configured to store a computer program, the transceiver is configured to communicate with other devices, and the processor is configured to call and execute the computer program stored in the memory, so that the first device performs the above-mentioned signal processing method.

[0016] An embodiment of the present application provides a second device, comprising: a transceiver, a processor, and a memory. The memory is configured to store a computer program, the transceiver is configured to communicate with other devices, and the processor is configured to call and execute the computer program stored in the memory, so that the second device performs the above-mentioned signal processing method.

[0017] An embodiment of the present application provides a chip for implementing the above-mentioned signal processing method.

[0018] Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip executes the above-mentioned signal processing method.

[0019] An embodiment of the present application provides a computer-readable storage medium for storing a computer program. When the computer program is executed by a device, the device executes the above-mentioned signal processing method.

[0020] An embodiment of the present application provides a computer program product, including computer program instructions, which enable a computer to execute the above-mentioned signal processing method.

[0021] An embodiment of the present application provides a computer program, which, when executed on a computer, enables the computer to execute the above-mentioned signal processing method.

[0022] An embodiment of the present application provides a communication system, including a first device and a second device for executing the above method.

[0023] The embodiment of the present application uses a first AI model to pre-distort the first digital signal output by the baseband link to obtain a second digital signal, that is, pre-distortion is performed through the AI ​​model to compensate for the noise and nonlinear distortion introduced by the non-ideal characteristics of the radio frequency link, which can improve the compensation effect of the pre-distortion processing and thus improve communication performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application.

[0025] FIG2 is a schematic diagram illustrating the nonlinear characteristics of a power amplifier.

[0026] FIG3 is a schematic diagram showing the impact of phase noise on a 16QAM constellation diagram.

[0027] FIG4A is a schematic diagram showing the connection between a predistorter and a power amplifier.

[0028] FIG4B is a schematic diagram of power amplifier characteristics after the predistorter is connected to the power amplifier.

[0029] FIG5 is a schematic flowchart of a signal processing method according to an embodiment of the present application.

[0030] FIG6 is a schematic diagram of a first AI model in an embodiment of the present application.

[0031] FIG7A is a schematic diagram of an application method of the second AI model in one embodiment of the present application.

[0032] FIG7B is a schematic diagram of another application method of the second AI model in one embodiment of the present application.

[0033] FIG8 is a schematic diagram of a second AI model in one embodiment of the present application.

[0034] FIG9A is a schematic diagram of a method for a network device to trigger monitoring of radio frequency performance indicators.

[0035] FIG9B is a schematic diagram of another method for a network device to trigger monitoring of a radio frequency performance indicator.

[0036] FIG10A is a schematic diagram of a method for conditionally triggering monitoring of radio frequency performance indicators.

[0037] FIG10B is a schematic diagram of another method of conditionally triggering the monitoring of radio frequency performance indicators.

[0038] FIG11 is a schematic diagram of fine-tuning the first AI model in one embodiment of the present application.

[0039] FIG12 is a schematic flowchart of a signal processing method according to another embodiment of the present application.

[0040] FIG13 is a schematic block diagram of a first device according to an embodiment of the present application.

[0041] FIG14 is a schematic block diagram of a first device according to another embodiment of the present application.

[0042] FIG15 is a schematic block diagram of a first device according to another embodiment of the present application.

[0043] FIG16 is a schematic block diagram of a first device according to another embodiment of the present application.

[0044] FIG17 is a schematic block diagram of a first device according to another embodiment of the present application.

[0045] FIG18 is a schematic block diagram of a first device according to another embodiment of the present application.

[0046] FIG19 is a schematic block diagram of a second device according to an embodiment of the present application.

[0047] FIG20 is a schematic block diagram of a second device according to another embodiment of the present application.

[0048] FIG21 is a schematic block diagram of a second device according to another embodiment of the present application.

[0049] Figure 22 is a schematic block diagram of a communication device according to an embodiment of the present application.

[0050] Figure 23 is a schematic block diagram of a chip according to an embodiment of the present application.

[0051] Figure 24 is a schematic block diagram of a communication system according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0053] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: Long Term Evolution (LTE) system, Advanced Long Term Evolution (LTE-A) system, New Radio (NR) system, NR system evolution system, LTE on unlicensed spectrum (LTE-U) system, NR on unlicensed spectrum (NR-based access to unlicensed spectrum, NR-U) system, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Wireless Fidelity (WiFi), Fifth Generation Communication (5G) system, Sixth Generation Communication (6G) system or other communication systems.

[0054] Generally speaking, traditional communication systems support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communications, but will also support, for example, device-to-device (D2D) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), vehicle-to-vehicle (V2V) communication, or vehicle-to-everything (V2X) communication, etc. The embodiments of the present application can also be applied to these communication systems.

[0055] In one embodiment, the communication system in the embodiment of the present application can be applied to a carrier aggregation (CA) scenario, a dual connectivity (DC) scenario, and a standalone (SA) networking scenario.

[0056] In one embodiment, the communication system in the embodiment of the present application can be applied to an unlicensed spectrum, wherein the unlicensed spectrum can also be considered as a shared spectrum; or, the communication system in the embodiment of the present application can also be applied to an authorized spectrum, wherein the authorized spectrum can also be considered as an unshared spectrum.

[0057] The embodiments of the present application describe various embodiments in conjunction with network devices and terminal devices, wherein the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.

[0058] The terminal device can be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a next-generation communication system such as an NR network, or a terminal device in a future evolved Public Land Mobile Network (PLMN) network, etc.

[0059] In an embodiment of the present application, the terminal device can be deployed on land, including indoors or outdoors, handheld, wearable or vehicle-mounted; it can also be deployed on the water surface (such as ships, etc.); it can also be deployed in the air (such as airplanes, balloons and satellites, etc.).

[0060] In an embodiment of the present application, the terminal device may be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city, or a wireless terminal device in a smart home, etc.

[0061] As an example and not a limitation, in the embodiment of the present application, the terminal device may also be a wearable device. Wearable devices may also be called wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0062] In an embodiment of the present application, the network device may be a device for communicating with a mobile device. The network device may be an access point (AP) in a WLAN, an evolved base station (eNB or eNodeB) in LTE, or a relay station or access point, or a vehicle-mounted device, a wearable device, and a network device (gNB) in an NR network, or a network device in a future evolved PLMN network or a network device in an NTN network, etc.

[0063] As an example and not a limitation, in an embodiment of the present application, the network device may have a mobile feature, for example, the network device may be a mobile device. Alternatively, the network device may be a satellite or a balloon station. For example, the satellite may be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, etc. Optionally, the network device may also be a base station set up in a location such as land or water.

[0064] In an embodiment of the present application, the network device can provide services for a cell, and the terminal device communicates with the network device through the transmission resources used by the cell (for example, frequency domain resources, or spectrum resources). The cell can be a cell corresponding to the network device (for example, a base station). The cell can belong to a macro base station or a base station corresponding to a small cell. The small cells here may include: metro cells, micro cells, pico cells, femto cells, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.

[0065] FIG1 exemplarily illustrates a communication system 100. The communication system includes a network device 110 and two terminal devices 120. In one embodiment, the communication system 100 may include multiple network devices 110, and each network device 110 may include a different number of terminal devices 120 within its coverage area, which is not limited in this embodiment of the present application.

[0066] It should be understood that in the embodiments of the present application, a device having a communication function in a network / system may be referred to as a communication device. Taking the communication system shown in Figure 1 as an example, the communication device may include a network device and a terminal device having a communication function. The network device and the terminal device may be specific devices in the embodiments of the present application and will not be described in detail here. The communication device may also include other devices in the communication system, such as a network controller, a mobility management entity, and other network entities, which are not limited in the embodiments of the present application.

[0067] It should be understood that the terms "system" and "network" are often used interchangeably herein. The term "and / or" is simply a description of an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " generally indicates that the related objects are in an "or" relationship.

[0068] It should be understood that the "indication" mentioned in the embodiments of this application can be a direct indication, an indirect indication, or an indication of an association. For example, "A indicates B" can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association between A and B.

[0069] In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship between indication and being indicated, configuration and being configured, etc.

[0070] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following relevant technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

[0071] (1) RF link

[0072] Both the RF link and the baseband link play a crucial role in modern wireless communication systems. The baseband link is primarily responsible for signal processing and protocol handling, including digital signal modulation and demodulation, channel encoding and decoding, source encoding and decoding, and some signaling processing. The RF link, on the other hand, is responsible for converting the processed baseband signal into an electromagnetic signal suitable for transmission after digital-to-analog conversion, or receiving external signals and transmitting them to the baseband through analog-to-digital conversion.

[0073] At the transmitter end, the baseband link transmits signals at a very low frequency, making them unsuitable for long-distance transmission. (Electromagnetic waves with frequencies below 100 kHz (kilohertz) are absorbed by the Earth's surface, preventing effective transmission. However, electromagnetic waves with frequencies above 100 kHz can propagate through the air and reflect off the ionosphere at the outer edge of the atmosphere, enabling long-distance transmission.) Therefore, the RF link must first up-convert the signal, modulating it from a low-frequency band to a high-frequency band (such as 900 MHz (megahertz), 3.5 GHz (gigahertz), or 28 GHz). In addition to frequency conversion, the RF link also requires a power amplifier (PA) to ensure sufficient transmit power before sending it to the antenna. After reaching the antenna, the signal often undergoes filtering to eliminate interference and clutter before being transmitted through the antenna element. At the receiver end, the RF link's processing is essentially the reverse of that at the transmitter end and will not be detailed here.

[0074] (2) Non-ideal characteristics of RF links

[0075] The design of wireless communication systems requires a balance between performance and cost. Cost constraints often make complex RF circuit structures and expensive components impractical for practical systems. Furthermore, targeted circuit optimization and adjustments are impossible for mass-produced equipment. These factors lead to RF non-idealities in communication transceivers, which introduce various noise and nonlinear distortion into the signal. As wireless communication utilizes higher frequency bands and requires ever-increasing bandwidth, the degradation in communication performance caused by RF non-idealities becomes increasingly significant.

[0076] Power amplifier nonlinearity: RF power amplifiers are core components in transmitters, amplifying analog signals. However, as typical nonlinear devices, power amplifiers face a trade-off between efficiency and linearity. As shown in Figure 2, operating in the high-efficiency region generates severe nonlinear distortion, increasing the bit error rate of the demodulated signal and causing out-of-band spectrum leakage, interfering with adjacent channel users. On the other hand, maintaining linearity requires high back-off, reducing power amplifier efficiency and output power, leading to cost and heat dissipation challenges. Therefore, traditional power amplifier research aims to strike a balance between efficiency and linearity.

[0077] I / Q imbalance (Inphase / Quadrature Imbalance): Signal processing in the RF link is implemented using analog circuits. However, analog circuits struggle to achieve the precise matching (equal amplitudes and a 90-degree phase difference) of digital signal processing when processing I / Q mixed carrier signals. This amplitude and phase mismatch, collectively known as IQ imbalance, causes image components in the signal spectrum and increases the difficulty of demodulating the signal.

[0078] Phase noise: Due to the non-ideal nature of the oscillator, the RF link introduces phase noise into the signal. As the name implies, phase noise can negatively impact communication systems that use phase-modulated signals. 4G / 5G cellular communication systems widely use phase- and amplitude-based signal modulation schemes, such as QPSK (Quadrature Phase Shift Keying), 16QAM (Quadrature Amplitude Modulation), 64QAM, and 256QAM. Phase noise can cause the constellation of the modulated signal to rotate. Signals with higher amplitude levels experience greater rotation, significantly increasing the difficulty of demodulating the signal. Figure 3 illustrates the impact of phase noise on the 16QAM constellation.

[0079] (3) Digital Pre-Distortion (DPD)

[0080] Simply put, digital predistortion is to artificially add a predistortion that is exactly opposite to the nonlinear distortion introduced by the RF link during the digital signal processing process to compensate for each other.

[0081] The most common digital pre-distortion scheme is used to offset the distortion generated by a power amplifier operating in its nonlinear region, ultimately significantly improving the power amplifier's efficiency and operating range, achieving a "negative plus negative equals positive" effect. As shown in Figure 4A, the pre-distorter is connected to the power amplifier, and the resulting power amplifier characteristics are shown in Figure 4B. Digital pre-distortion distorts the original signal in the opposite direction of the power amplifier, thereby compensating for the nonlinear effects of the power amplifier in the final output signal (the actual power amplifier output).

[0082] The challenge of digital pre-distortion technology lies in the fact that the noise intensity and nonlinear distortion characteristics introduced by RF devices vary dynamically due to numerous factors, such as temperature, power supply voltage, device aging, frequency band, and bandwidth. Therefore, while it is theoretically possible to design a correct pre-distortion algorithm for an RF device with defined characteristics, performing this process for every device is economically unfeasible.

[0083] To address these discrepancies, some traditional predistortion algorithms incorporate a feedback mechanism, sampling the RF output signal and using it to correct the predistortion algorithm. However, this approach is complex and computationally intensive. Furthermore, because the feedback channel is responsible for collecting nonlinear information from RF components, imperfect algorithms or hardware can directly affect the accuracy of the predistorter, causing spectrum spread in the nonlinear power amplifier. Considering that the sampling rate of the feedback channel ADC typically needs to be 3 to 5 times the signal bandwidth to capture sufficient nonlinear information, the increasing signal bandwidth and the adoption of massive MIMO (Multiple-Input Multiple-Output) transmitters have led to increasing cost pressures associated with high-speed, high-precision ADCs. Therefore, implementing low-complexity digital predistortion technology using low-precision or low-rate ADCs remains a pressing challenge.

[0084] FIG5 is a schematic flow chart of a signal processing method according to an embodiment of the present application. The method can optionally be applied to the system shown in FIG1 , but is not limited thereto. The method includes at least part of the following contents.

[0085] S510. The first device performs predistortion processing on a first digital signal output by a baseband link using a first AI model to obtain a second digital signal.

[0086] S520: The first device processes the second digital signal into a first radio frequency signal.

[0087] In the above signal processing method, the first device can be a terminal device or a network device. The above method can be applied to the first device processing a signal to be transmitted. That is, after receiving the first RF signal, the first device can send the first RF signal. Specifically, the first device sends the first RF signal to a second device, which can also be a terminal device or a network device.

[0088] That is, the embodiments of the present application can be applied to one of the following communication scenarios:

[0089] Scenario 1: The first device is a terminal device, and the second device is a network device;

[0090] Scenario 2: The first device is a network device, and the second device is a terminal device;

[0091] Scenario 3: The first device is a first terminal device, and the second device is a second terminal device;

[0092] Scenario 4: The first device is a first network device, and the second device is a second network device.

[0093] Since the implementation process of the embodiments of the present application mainly involves the processing of the signal to be transmitted by the first device, in some descriptions, the first device may also be referred to as a sending end, a transmitting end, a sending side, a sending device, etc., and the second device may also be referred to as a receiving end, a receiving side, a receiving device, etc.

[0094] In an embodiment of the present application, the first digital signal is a signal output by a baseband link. The baseband link is used to perform baseband processing on the input signal (i.e., the information bits to be transmitted), such as channel coding, scrambling, constellation modulation, precoding, resource element mapping, OFDM (Orthogonal Frequency Division Multiplexing) signal modulation, etc. Exemplarily, the first digital signal can be a complex signal used to carry the information bits to be transmitted.

[0095] According to the above step S510, the first AI model is used to perform pre-distortion processing on the first digital signal to obtain a second digital signal. That is, the first AI model is used to perform pre-distortion processing, and its input information includes the first digital signal, and its output information includes the second digital signal. Here, pre-distortion processing refers to processing used to compensate for noise and nonlinear distortion introduced by the radio frequency link, and does not limit the processing form. In practical applications, the first AI model can be a model based on an artificial neural network or a machine learning model. Among them, the model based on artificial neural network is, for example, DNN (Deep Neural Networks), CNN (Convolutional Neural Network), RNN (Recurrent Neural Network) or Transformer model, etc. Machine learning models are, for example, logistic regression models, decision tree models, etc.

[0096] In step S520, the first device may utilize a radio frequency link to process the second digital signal into a first radio frequency signal. In other words, the second digital signal is a signal input to the radio frequency link, which is used to process the input signal into a first radio frequency signal suitable for transmission. For example, the radio frequency link may be used to perform digital-to-analog conversion on the input second digital signal and then convert it into the first radio frequency signal. For example, the radio frequency link may include a digital-to-analog conversion module, an upconversion module, a power amplifier, and the like.

[0097] Since the above method performs pre-distortion through the first AI model before the signal to be transmitted enters the RF link to compensate for the noise and nonlinear distortion introduced by the non-ideal characteristics of the RF link, it is possible to utilize the nonlinear processing capabilities of the AI ​​model to improve the compensation effect of the pre-distortion processing, thereby improving communication performance. Specifically, the nonlinear distortion introduced by the RF link is often the combined result of multiple factors (such as power amplifier nonlinearity, phase noise, I / Q imbalance, etc.), and it is difficult to establish a complete theoretical model. The AI-based method can approach ideal performance in the absence of a theoretical model through a data-driven approach, thereby improving the compensation effect of the pre-distortion processing.

[0098] It should be noted that the above-mentioned first AI model can be provided by a network device. For example, the first device is a terminal device, and the first AI model is issued or configured by the network device. Optionally, the network device can retain one or more sets of models and corresponding model IDs, and issue or configure the first AI model to the terminal device through broadcast information, RRC (Radio Resource Control) signaling, MAC CE (MAC Control Element, media access control control unit), DCI (Downlink Control Information, downlink control information), PDSCH (Physical Downlink Shared Channel, physical downlink shared channel), etc. Alternatively, the first AI model can also be predefined by the device manufacturer, for example, the network device or terminal device manufacturer predefines one or more sets of models and corresponding model IDs (Identifiers). The first AI model can also be obtained through other channels, which is not limited in this embodiment of the present application.

[0099] In some embodiments, the input information of the first AI model includes a first digital signal and first auxiliary information related to the operating performance of the RF link. Accordingly, in step S510, the first device uses the first AI model to predistort the first digital signal output by the baseband link, including: the first device uses the first AI model to predistort the first digital signal based on the first auxiliary information. In other words, the first AI model is used to predistort the first digital signal based on the first auxiliary information.

[0100] Specifically, as shown in Figure 6, in an embodiment of the present application, the input information of the first AI model includes a first digital signal, and the output information includes a second digital signal; optionally, the input information of the first AI model may also include first auxiliary information, so that the first AI model can perform pre-distortion processing on the first digital signal based on the first auxiliary information.

[0101] In some embodiments, the first auxiliary information may include environmental information or device information of the first device, which affects the working performance of the RF link, or affects the requirements for the working performance of the RF link. In this way, the first AI model can dynamically adapt the corresponding pre-distortion processing results according to changes in the working performance of the RF link or related needs, thereby improving the ability of the first AI model to cope with different situations, thereby improving the compensation effect on the RF link. For example, assuming that the first auxiliary information is the temperature of the first device, the first AI model can adaptively perform pre-distortion processing based on the current temperature of the first device, so that the corresponding output can be dynamically adapted for different temperatures.

[0102] In some embodiments, the first assistance information may include at least one of the following:

[0103] (1) The temperature of the first device; specifically, the current temperature of the first device. The operating performance of the RF device will change at different temperatures.

[0104] (2) The operating voltage of the first device; specifically, the current operating voltage of the first device. The operating performance of the RF device will vary under different voltages.

[0105] (3) The operating frequency band of the first device; specifically, the current operating frequency band of the first device, such as 900 MHz, 4 GHz, 28 GHz, etc.

[0106] (4) The communication bandwidth of the first device; for example, 10 MHz, 50 MHz, 100 MHz, etc.

[0107] (5) Waveform information used by the first device to send signals; specifically, the waveform information indicates the waveform currently used for signal transmission, such as OFDM (Orthogonal Frequency Division Multiplexing), DFT-s-OFDM (Discrete Fourier Transform-Spread), etc. Different waveforms have different PAPR (Peak to Average Power Ratio) characteristics and therefore have different requirements for the operating performance of radio frequency devices.

[0108] (6) Modulation orders supported by the first device; specifically, this may include one or more modulation orders supported by the first device. For example, this may include the highest modulation order supported by the first device. Higher-order modulation methods require higher precision from RF components, and thus have different predistortion requirements.

[0109] (7) The continuous operating time of the first device; specifically, the operating performance of the radio frequency device may vary depending on the continuous operating time.

[0110] It is understandable that the first auxiliary information may include only one of the above information or multiple of the above information. For example, the first auxiliary information may include only the temperature, operating voltage, or operating frequency band of the first device. For another example, the first auxiliary information may include the temperature of the first device and the supported modulation order. For another example, the first auxiliary information may include the temperature and operating frequency band of the first device. In actual applications, the first auxiliary information can be set according to scenario requirements, protocol agreements, or system agreements. The implementation methods are not listed one by one here.

[0111] In some embodiments, the signal processing method further includes:

[0112] The first device uses the second AI model to process the third digital signal to obtain a radio frequency performance indicator; wherein the third digital signal is obtained by converting sampling information of at least one radio frequency signal sent by the first device, and the at least one radio frequency signal includes the above-mentioned first radio frequency signal; the radio frequency performance indicator is used to characterize the performance of the first AI model.

[0113] Specifically, the RF performance indicator can be used to characterize whether the pre-distortion processing performed by the first AI model effectively offsets the noise and nonlinear distortion introduced by the RF link.

[0114] In actual applications, the first device can sample the RF signal it sends to obtain sampling information. The sampling information is then processed as necessary to convert it into a digital signal form that can be processed by the second AI model, that is, a third digital signal. The first device uses the second AI model to process the third digital signal, and can use the processing capability of the second AI model to determine the RF performance indicators. In comparison, monitoring RF performance through traditional algorithms relies on a high-precision, high-sampling-rate RF signal feedback path. The complexity and cost issues it brings are difficult to accept in many cases. The use of the second AI model can achieve effective performance monitoring and evaluation under a low-precision, low-sampling-rate RF feedback path.

[0115] Exemplarily, the at least one RF signal may be a RF signal within a specific time period. For example, a sampling window is pre-configured, and the first device sends at least one RF signal within the first sampling window. Sampling information may then be obtained based on the at least one RF signal.

[0116] Exemplarily, the third digital signal can be obtained by down-converting and analog-to-digital converting sampling information of at least one RF signal, where the at least one RF signal includes the first RF signal. As shown in FIG7A , the RF link of the transmitter in the first device includes a digital-to-analog conversion module, an up-conversion module, and a power amplifier. Accordingly, the first device can sample the at least one transmitted RF signal to obtain sampling information, down-convert and analog-to-digital convert the sampling information to obtain a third digital signal, and then process the third digital signal using the second AI model to obtain RF performance indicators.

[0117] Optionally, the second AI model may be a model based on an artificial neural network or a machine learning model. Examples of artificial neural network models include DNN, CNN, RNN, or Transformer models. Examples of machine learning models include logistic regression models and decision tree models.

[0118] Optionally, the second AI model may be provided by a network device. For example, the first device is a terminal device, and the second AI model is issued or configured by the network device. Optionally, the network device may retain one or more models and corresponding model IDs, and issue or configure the second AI model to the terminal device through broadcast information, RRC signaling, MAC CE, DCI, PDSCH, etc. Alternatively, the second AI model may be predefined by the device manufacturer, for example, the network device or terminal device manufacturer predefines one or more models and corresponding model IDs. The second AI model may also be obtained through other channels, which is not limited in this embodiment of the present application.

[0119] It should be noted that the first AI model and the second AI model can be obtained in pairs, or can be obtained through different channels respectively.

[0120] In some embodiments, the RF performance indicator is related to the management policy of the first AI model. For example, the RF performance indicator can be used to determine the management policy of the first AI model, that is, the first device or another device can determine the management policy of the first AI model based on the RF performance indicator.

[0121] Exemplarily, the management strategy of the first AI model includes model fine-tuning, model updating, or model switching, so that the first AI model can be dynamically adapted to the current RF device.

[0122] Optionally, the first device may determine whether to fine-tune, update, or switch the first AI model based on the RF performance indicator. Alternatively, the first device may send the first RF performance indicator to the second device, which then determines whether to fine-tune, update, or switch the first AI model based on the RF performance indicator.

[0123] In some embodiments, the RF performance indicator is used to input the first AI model; the first AI model is used to perform pre-distortion processing on the fourth digital signal output by the baseband link based on the RF performance indicator. The fourth digital signal may be another digital signal output after the baseband link outputs the first digital signal. In other words, the input information of the first AI model may also include the RF performance indicator to adjust the output of the first AI model in real time according to the RF performance indicator. It can be understood that the processing of the first digital signal by the first AI model may also be based on the latest RF performance indicator determined when the baseband link outputs the first digital signal.

[0124] Exemplarily, as shown in FIG7B , a radio frequency performance indicator is introduced as feedback and input into the first AI model to improve the performance of digital predistortion.

[0125] In some embodiments, the sampling information includes a second RF signal determined from the at least one RF signal based on the sampling parameters, or statistical information of the at least one RF signal. In other words, the sampling information can be an instantaneous sampling result at a certain moment, or a statistical sampling result, such as a statistical sampling result of the at least one RF signal within a specific time period.

[0126] In some embodiments, at least one sampling parameter of a radio frequency signal is configured by a second device. The second device may be a network device. For example, the sampling parameter of the radio frequency signal in the terminal device may be configured by the network device, for example, via RRC, MAC CE, DCI, or the like.

[0127] In some embodiments, the sampling parameters include a sampling period and / or a sampling interval.

[0128] Exemplarily, the first device may periodically sample the radio frequency signal, and the sampling period may be pre-configured to determine how long the sampling is performed continuously. For example, if the sampling period is N time units, then every N time units is used as a sampling window, and the first device performs sampling and obtains sampling information in every N time units. Wherein, N is an integer greater than or equal to 1, and the time unit may be a time slot, a subframe, a radio frame, milliseconds, seconds, minutes, or hours, etc. The sampling method may be to sample one or more instantaneous radio frequency signals in every N time units, or to obtain a statistical sampling result in every N time units.

[0129] Exemplarily, the sampling interval may refer to the interval at which RF signal sampling is performed within a sampling period or within a non-periodic sampling window. For example, the sampling interval may be M time units, where M is an integer greater than or equal to 1, and a time unit may be a time slot, a subframe, a radio frame, milliseconds, seconds, minutes, or hours. For example, within the aforementioned period of N time units, the RF signal is sampled once every M time units.

[0130] In some embodiments, the input information of the second AI model may include a third digital signal and second auxiliary information. The second AI model is configured to process the third digital signal based on the second auxiliary information. That is, the first device using the second AI model to process the third digital signal may include: the first device using the second AI model to process the third digital signal based on the second auxiliary information.

[0131] Specifically, as shown in Figure 8, in an embodiment of the present application, the input information of the second AI model includes a third digital signal obtained by processing sampling information of at least one RF signal, and the output information includes RF performance indicators; optionally, the input information of the second AI model may also include second auxiliary information, so that the second AI model can process the third digital signal based on the second auxiliary information to obtain the RF performance indicators.

[0132] In some embodiments, the second auxiliary information includes the following information A and / or B:

[0133] A. Sampling information of at least one digital signal corresponding one-to-one to at least one RF signal; wherein each digital signal in the at least one digital signal is processed by the first AI model and the RF link to obtain a corresponding RF signal.

[0134] It is understood that each of the at least one digital signal is a digital signal before predistortion processing. That is, the first device samples the at least one RF signal and, accordingly, samples the digital signal before predistortion processing corresponding to the at least one RF signal. The third digital signal converted based on the sampling information of the at least one RF signal and the sampling information of the at least one digital signal before predistortion processing are input into the second AI model, so that the second AI model processes the third digital signal based on the digital signal before predistortion processing to obtain the RF performance indicator.

[0135] Optionally, the sampling method for the at least one digital signal may be the same as the sampling method for the at least one RF signal. For example, if the sampling information of the at least one RF signal is the RF signal at an instant, then the sampling information of the at least one digital signal is the instantaneous digital signal corresponding to the RF signal at that instant. For another example, if the sampling information of the at least one RF signal is the statistical sampling result of the at least one RF signal, then the sampling information of the at least one digital signal is the statistical sampling result of the at least one digital signal.

[0136] Since the RF performance index is used to evaluate whether the pre-distortion processing performed by the first AI model effectively offsets the noise and nonlinear distortion introduced by the RF link, it essentially evaluates whether there is obvious noise and nonlinear distortion between the RF signal and the digital signal before processing. Therefore, the digital signal before pre-distortion can be used as an effective reference benchmark to improve the accuracy of the RF performance index.

[0137] B. A precoding matrix for obtaining at least one digital signal.

[0138] Theoretically, when the RF link introduces no significant noise or distortion, the transmitted signal or bits can be demodulated nearly perfectly based on RF signal sampling. Therefore, when the second auxiliary information includes the precoding matrix, the second AI model can demodulate the third digital signal corresponding to the sampled RF signal based on the precoding matrix to obtain the corresponding information bits, thereby using these information bits to improve the accuracy of the output RF performance indicators.

[0139] Optionally, in combination with the foregoing embodiment, the input information of the second AI model may include the third digital signal, the second auxiliary information, and the RF performance indicator. The second AI model is configured to process the third digital signal based on the second auxiliary information and the RF performance indicator.

[0140] In some embodiments, the radio frequency performance indicator includes at least one of the following:

[0141] (1) RF performance level information;

[0142] The RF performance level information may be an overall rating of RF performance rather than an actual physical indicator. For example, the RF performance level may range from 1 to 5; a level of 1 indicates the worst RF performance, while a level of 5 indicates the best RF performance.

[0143] (2) error information of the first radio frequency signal;

[0144] This error information can be used to characterize the error between the actual sampled RF signal and the ideal RF signal. Optionally, when training the second AI model, the digital signal can be processed using an ideal RF device to obtain an ideal RF signal; at the same time, the digital signal can be processed using a non-ideal RF device to obtain a RF signal sample; the error between the ideal RF signal and the RF signal sample is used as a label for the RF performance indicator output by the second AI model for the RF signal sample, and based on this, the second AI model is trained to have the ability to output error information.

[0145] Exemplarily, the error information may include at least one of the following:

[0146] Error Vector Magnitude (EVM);

[0147] Frequency error;

[0148] Phase error.

[0149] Exemplarily, the vector magnitude error may be a mean square error of a ratio of an average error vector signal power of an ideal radio frequency signal and an actual radio frequency signal to an average reference signal power.

[0150] For example, the frequency error may be a difference between the frequency of an ideal radio frequency signal and the frequency of an actual radio frequency signal.

[0151] Exemplarily, the phase error may be a difference between the phase of an ideal radio frequency signal and the phase of an actual radio frequency signal.

[0152] (3) ACLR (Adjacent Channel Leakage Ratio);

[0153] ACLR can be used to measure the interference caused by transmitted RF signals to the same or similar communication systems.

[0154] (4) An accuracy index of information bits obtained by demodulating the third digital signal.

[0155] Exemplarily, the accuracy index may include BLER (Block Error Rate), BER (Bit Error Rate), SER (Symbol Error Rate), etc. As described above, the second AI model can demodulate the third digital signal based on the precoding matrix to obtain the corresponding information bits. Specifically, the second AI model can also infer the accuracy index based on the information bits, that is, the RF performance index may include the accuracy index to measure whether the RF signal can accurately demodulate the information bits corresponding to the transmitted signal, thereby measuring the RF performance.

[0156] By evaluating one or more of the RF performance indicators listed above, the device can understand whether the first AI model effectively compensates for problems such as nonlinear distortion introduced by the RF link, thereby determining whether the first AI model needs to be fine-tuned, updated, or switched.

[0157] In practical applications, there are many ways to trigger the monitoring of RF performance indicators. The following are three optional methods.

[0158] Method 1: Triggered by the second device.

[0159] Specifically, the above-mentioned signal processing method may further include: the first device receives trigger signaling from the second device; wherein the trigger signaling is used to trigger the first device to obtain the radio frequency performance indicator.

[0160] Exemplarily, when the first device is a terminal device and the second device is a network device, the network device triggers the terminal device to perform single monitoring or periodic monitoring by sending control signaling.

[0161] Exemplarily, the trigger signaling can be carried by a broadcast message, RRC signaling, MAC CE, DCI, etc.

[0162] Optionally, the trigger signaling may further include other information related to the monitoring of the radio frequency performance indicator. Exemplarily, the trigger signaling includes at least one of the following:

[0163] Type information of radio frequency performance indicators;

[0164] Monitoring time information of radio frequency performance indicators;

[0165] Threshold of the radio frequency performance indicator.

[0166] Exemplarily, the type information of the RF performance indicator can be used to indicate one or more types of information specifically included in the RF performance indicator, such as RF performance level information, error information of the first RF signal, ACLR, accuracy index of the information bit obtained by demodulation, etc.

[0167] Exemplarily, the monitoring time information of the radio frequency performance indicator may include a monitoring period or indication information of a specific time.

[0168] Exemplarily, the threshold of the radio frequency performance indicator is used to determine whether the radio frequency performance indicator meets the requirements. For example, the threshold is the maximum value of the error information, the maximum value of ACLR, etc.

[0169] Method 2: Condition triggering.

[0170] Specifically, in the above signal processing method, the first device uses the second AI model to process the third digital signal to obtain a radio frequency performance indicator, including:

[0171] When the first condition is met, the first device uses the second AI model to process the third digital signal to obtain a radio frequency performance indicator.

[0172] Optionally, the first condition may be configured by the second device.

[0173] For example, the first condition may include information related to the operating performance of the radio frequency link or a change in the information being higher or lower than a threshold. The type of the information and the related threshold may be configured by the second device to the first device, for example, by the network device to the terminal device. Specifically, the configuration may be performed via a broadcast message, RRC, MAC CE, DCI, or the like.

[0174] Exemplarily, the first condition is related to at least one of the following:

[0175] the temperature of the first device;

[0176] an operating voltage of the first device;

[0177] an operating frequency band of the first device;

[0178] a communication bandwidth of the first device;

[0179] Waveform information used by the first device to send a signal;

[0180] the modulation order supported by the first device;

[0181] The continuous operating time of the first device.

[0182] It can be understood that the various information listed above are related to the working performance of the radio frequency link, for example, they will affect the working performance of the radio frequency link or bring higher requirements to the working performance of the radio frequency link. Based on this, the first condition may include that the size relationship between one or more of the above information and the preset threshold meets the requirements.

[0183] Method 3: Triggered by the first device.

[0184] Specifically, the first device may trigger the monitoring of the radio frequency performance indicator by itself.

[0185] Optionally, when the first device triggers the monitoring of the RF performance indicator, it may send a request to the second device and monitor the RF performance indicator after obtaining the consent of the second device. Specifically, in the above signal processing method, the first device uses the second AI model to process the third digital signal to obtain the RF performance indicator, including:

[0186] The first device sends a first request to the second device, where the first request is used to request monitoring of a radio frequency performance indicator;

[0187] When the first device receives the consent information sent by the second device in response to the first request, the first device uses the second AI model to process the third digital signal to obtain a radio frequency performance indicator.

[0188] Exemplarily, the first request may be carried by a broadcast message, RRC signaling, MAC CE, or UCI.

[0189] In actual applications, after the first device completes monitoring of the RF performance indicator, it can process the RF performance indicator in a variety of ways. For example, the first device independently determines the management strategy of the first AI model based on the RF performance indicator, such as whether to fine-tune, update, or switch (including switching to another model or using a traditional digital pre-distortion solution). For another example, the first device inputs the RF performance indicator as feedback into the first AI model to adjust the output of the first AI model.

[0190] Optionally, the first device may also report the RF performance indicator to the second device, and the second device may determine a management policy for the first AI model based on the RF performance indicator, such as whether to fine-tune, update, or switch. Specifically, the above-mentioned signal processing method may further include: the first device sending reporting information to the second device; wherein the reporting information includes the RF performance indicator, and the RF performance indicator is used by the second device to determine whether to fine-tune, update, or switch the first AI model.

[0191] In some embodiments, the above-mentioned signal processing method may further include: the first device receives indication information sent by the second device, wherein the indication information is used to indicate the management strategy of the first AI model.

[0192] For example, the first device is a terminal device and the second device is a network device. The terminal device reports the radio frequency performance indicator by sending reporting information. The network device can determine to fine-tune, update or switch the first AI model based on the radio frequency performance indicator and instruct the terminal device.

[0193] For ease of understanding, taking the first device as a terminal device (UE) and the second device as a network device (NW) as an example, Figures 9A and 9B illustrate schematic diagrams of a network device triggering monitoring of a radio frequency performance indicator. In Figure 9A , the UE makes an autonomous decision based on the radio frequency performance indicator, while in Figure 9B , the UE reports the radio frequency performance indicator.

[0194] As shown in Figure 9A, after the network device triggers the UE to monitor RF performance indicators and configures the monitoring period, indicator type, threshold, and other information, the UE monitors the RF performance indicators and, based on the monitoring results, independently determines whether to perform a fallback or model switch. After completing the fallback or model switch, the UE reports the results of the action to the network device.

[0195] As shown in Figure 9B, after the network device triggers the UE to monitor the RF performance indicators and configures the monitoring period, indicator type, threshold and other information, the UE monitors the RF performance indicators and can report the monitoring results. The network device determines whether to roll back the solution or switch the model based on the monitoring results and sends instructions to the UE.

[0196] Figures 10A and 10B are schematic diagrams showing conditional triggering of monitoring of radio frequency performance indicators, wherein in Figure 10A the UE makes autonomous decisions based on the radio frequency performance indicators, and in Figure 10B the UE reports the radio frequency performance indicators.

[0197] As shown in Figure 10A, the network device configures trigger conditions for the UE to monitor RF performance indicators. When the trigger conditions are met, the UE monitors the RF performance indicators and, based on the monitoring results, determines whether to perform a fallback or model switch. After completing the fallback or model switch, the UE reports the results of the action to the network device.

[0198] As shown in Figure 10B, the network device configures trigger conditions for the UE to monitor RF performance indicators. When the trigger conditions are met, the UE monitors the RF performance indicators and reports the monitoring results. The network device then determines whether to perform a solution fallback or model switch based on the monitoring results and issues instructions to the UE.

[0199] In actual applications, the RF characteristics of different devices are different, and the RF characteristics of the same device will also change with changes in some conditions (such as temperature, voltage, device aging, etc.), and digital pre-distortion needs to dynamically adapt to these changes to ensure performance. Therefore, taking advantage of the flexibility of the AI ​​model and updating and fine-tuning it based on actual conditions will help to ensure RF performance. Among them, fine-tuning, as an inherent technical term in AI, refers to the use of new data to adjust the coefficients / weights in the model through training on the basis of an existing model, so that the model can better adapt to the new data. In some embodiments, fine-tuning can be used to correct or optimize the current first AI model.

[0200] There are many ways to fine-tune the first AI model. An exemplary method is provided below.

[0201] Method A: The first device fine-tunes the first AI model.

[0202] Exemplarily, the above signal processing method may further include:

[0203] The first device obtains a loss function based on the second digital signal and the fifth digital signal; wherein the fifth digital signal is obtained by processing the sixth digital signal by the first AI model, and the sixth digital signal is obtained by converting the first radio frequency signal;

[0204] The first device fine-tunes the first AI model based on the loss function.

[0205] Specifically, the first device can use the first RF signal to convert and obtain a sixth digital signal, for example, by down-converting and analog-to-digital conversion to convert the first RF signal into a sixth digital signal that can be processed by the first AI model. Afterwards, the first device processes the sixth digital signal using the first AI model to obtain a fifth digital signal. The first device obtains a loss function based on the second digital signal and the fifth digital signal, and fine-tunes the first AI model based on the loss function. Among them, the fifth digital signal is obtained after the second digital signal passes through the RF link and down-conversion and analog-to-digital conversion (this step introduces nonlinear distortion caused by RF devices into the digital signal), and then is processed by the first AI model. Therefore, the loss between the second digital signal and the fifth digital signal can be gradually reduced through fine-tuning, thereby gradually improving the compensation effect of AI pre-distortion.

[0206] For ease of understanding, assuming that the effect of the first AI model is denoted as f(·), the input signal of the RF link (the second digital signal) is x, and the nonlinear distortion introduced by the RF link is denoted as g(·), then the output of the RF link after preprocessing (the sixth digital signal) is g(x), where the preprocessing is, for example, down-conversion, analog-to-digital conversion, etc. The effect of the ideal AI predistortion model that can compensate should be g(·)-1, which is the inverse process of g(·). Therefore, the model training objective can be expressed as follows:

[0207] Among them, f FT represents the parameters of the final fine-tuned first AI model, and L(·,·) represents a loss function, such as the mean squared error (MSE) loss function. It can be understood that x is the second digital signal, f(g(x)) is the fifth digital signal, and the above training objective is to minimize the loss between the second digital signal and the fifth digital signal.

[0208] In actual applications, the above method enables the first device to fine-tune the first AI model based on the newly generated data, thereby correcting or optimizing the first AI model.

[0209] In some embodiments, the above-mentioned signal processing method may further include: when the radio frequency performance indicator meets the second condition, the first device determines that fine-tuning of the first AI model is completed.

[0210] That is, during the model fine-tuning process, the output of the RF performance monitoring AI model can be used as an indicator to determine whether the pre-distortion model fine-tuning is complete. For example, the second condition can be that the RF performance indicator is greater than a certain threshold, or the RF performance indicator is less than a certain threshold.

[0211] Method B: The second device fine-tunes the first AI model.

[0212] Specifically, the above signal processing method further includes:

[0213] The first device sends a sample data set to the second device; wherein the sample data set includes a second digital signal and a sixth digital signal; wherein the sixth digital signal is converted from the first radio frequency signal; the sample data set is used to fine-tune the first AI model.

[0214] In this approach, a first device can collect the data set required for tuning and send it to a second device, which then fine-tunes the first AI model. The second device can then reconfigure or send the updated, fine-tuned first AI model to the first device.

[0215] Optionally, the second device may be configured with a first AI model. Alternatively, the first device may send the first AI model or related information (e.g., model parameters) to the second device, so that the second device can determine the first AI model. In this way, the second device can use the first AI model to process the sixth digital signal in the sample data set to obtain a fifth digital signal; then calculate a loss function based on the second digital signal and the fifth digital signal, and fine-tune the first AI model based on the loss function.

[0216] In some embodiments, parameters related to the sample data set are configured by the second device.

[0217] In some embodiments, the relevant parameters include at least one of a data volume of the sample data set, a collection period, and a time interval between different samples in the sample data set.

[0218] Taking the first device as a terminal device and the second device as a network device as an example, since model training and fine-tuning require high computing power, fine-tuning by the network device can reduce the computing power requirements of the terminal device. Figure 11 shows a schematic diagram of fine-tuning the first AI model. As shown in Figure 11, the network device (NW) configures the relevant parameters of the data set for the UE. The UE collects the data set required for tuning, such as x and g(x) in the aforementioned example, and uploads it to the NW. Optionally, it can be uploaded through RRC messages, UCI, PUSCH, etc. After the NW completes the model fine-tuning, the updated model is sent to the UE. Optionally, the model can be sent through RRC messages, MAC CE, DCI or PDSCH. The parameters related to the data set, such as data volume, collection period, time interval between data samples, etc., can be configured by the NW.

[0219] It can be seen that the technical solution of the embodiment of the present application can reflect performance advantages in the following aspects:

[0220] (1) Nonlinear distortion introduced by the RF link is often the result of the combined effects of multiple factors (such as power amplifier nonlinearity, phase noise, and I / Q imbalance), making it difficult to establish a comprehensive theoretical model. Traditional algorithms rely on theoretical models, while AI-based predistortion can approach ideal performance through a data-driven approach even in the absence of theoretical models.

[0221] (2) Different devices have different RF characteristics. Designing and adjusting traditional algorithms for different devices is time-consuming and labor-intensive, and therefore impractical. However, the learning capability of AI solutions avoids the need for complex manual tuning and can learn model coefficients that are suitable for specific devices based on a certain amount of data collection.

[0222] (3) Even the RF characteristics of the same device will change with changes in certain conditions (such as temperature, voltage, device aging, etc.). Traditional methods have limited adjustment flexibility, while AI solutions can update / fine-tune model parameters based on rapid real-time or near-real-time data acquisition to adapt to changing nonlinear RF characteristics.

[0223] (4) To fully leverage the flexible adjustment and adaptability of the AI ​​pre-distortion model, effective monitoring of RF performance is crucial. Monitoring RF performance using traditional algorithms relies on high-precision, high-sampling-rate RF signal feedback paths, which in many cases present unacceptable complexity and cost. AI solutions, however, enable effective performance monitoring and evaluation using low-precision, low-sampling-rate RF feedback paths.

[0224] In a multi-antenna (MIMO) system, there are multiple independent RF links serving different antenna units or antenna unit groups. In this case, different RF links may have different nonlinear characteristics. Based on this, on the basis of the aforementioned embodiment, when the first device includes multiple RF links, one or more first AI models and one or more second AI models can be set.

[0225] In one embodiment, multiple first AI models corresponding to multiple RF links and multiple second AI models corresponding to multiple RF links can be set. Each of the multiple first AI models is used to compensate for the corresponding RF link, that is, the second digital signal output by the first AI model is used to input into the corresponding RF link; thus, the multiple first AI models compensate for the multiple RF links respectively. Each of the multiple second AI models is used to monitor its corresponding RF link; thus, the multiple second AI models monitor the multiple RF links respectively.

[0226] In one embodiment, multiple first AI models corresponding to multiple RF links can be set, and a single second AI model corresponding to multiple RF links (as a whole) can be set. Each first AI model is used to compensate for the corresponding RF link, that is, the second digital signal output by the first AI model is used to input into the corresponding RF link; in this way, the multiple first AI models compensate for the multiple RF links respectively. A single second AI model is used to monitor multiple RF links.

[0227] In one embodiment, a single first AI model corresponding to multiple RF links (as a whole) and a single second AI model corresponding to multiple RF links (as a whole) can be set. The single first AI model is used to compensate for the multiple RF links. The single second AI model is used to monitor the multiple RF links.

[0228] In one embodiment, a single first AI model corresponding to multiple RF links (as a whole) and multiple second AI models corresponding to the multiple RF links can be set. The single first AI model is used to compensate for the multiple RF links. Each of the multiple second AI models is used to monitor its corresponding RF link; in this way, the multiple second AI models monitor the multiple RF links separately.

[0229] When the first device is provided with multiple first AI models, operations such as fine-tuning / rollback / switching of the models may be performed on all models or only on one or more models.

[0230] Corresponding to the above method, an embodiment of the present application also provides a signal processing method performed by a second device. Figure 12 is a schematic flow chart of a signal processing method according to another embodiment of the present application. This method can optionally be applied to the system shown in Figure 1, but is not limited thereto. The method includes at least part of the following content.

[0231] S1210. The second device receives a first RF signal sent by the first device; wherein the first RF signal is obtained by the first device processing a second digital signal, and the second digital signal is obtained by pre-distorting the first digital signal output by the baseband link of the first device using the first AI model.

[0232] In some embodiments, the signal processing method further includes:

[0233] The second device sends first configuration information to the first device; wherein the first configuration information is used to configure sampling parameters, and the sampling parameters are used by the first device to determine sampling information of at least one RF signal, and the at least one RF signal includes the above-mentioned first RF signal; the sampling information is used to convert to obtain a third digital signal, and the third digital signal is processed by the second AI model to obtain a RF performance indicator, and the RF performance indicator is used to characterize the performance of the first AI model.

[0234] That is, the sampling parameters for radio frequency performance monitoring are configured by the second device.

[0235] In some embodiments, the sampling parameters include a sampling period and / or a sampling interval.

[0236] In some embodiments, the radio frequency performance monitoring may be triggered by the second device, that is, the above signal processing method further includes:

[0237] The second device sends a trigger signaling to the first device; wherein the trigger signaling is used to trigger the first device to obtain a radio frequency performance indicator for characterizing the performance of the first AI model.

[0238] In some embodiments, the trigger signaling includes at least one of the following:

[0239] Type information of radio frequency performance indicators;

[0240] Monitoring time information of radio frequency performance indicators;

[0241] Threshold of the radio frequency performance indicator.

[0242] In some embodiments, the radio frequency performance monitoring may be triggered by the first device and approved by the second device. The above signal processing method further includes:

[0243] The second device receives a first request sent by the first device, wherein the first request is used to request monitoring of a radio frequency performance indicator of the first device;

[0244] The second device sends consent information for the first request to the first device; wherein the consent information is used to instruct the first device to use the second AI model to obtain the radio frequency performance indicator of the first device.

[0245] In some embodiments, the signal processing method further includes:

[0246] The second device receives the reporting information sent by the first device, wherein the reporting information includes the radio frequency performance indicator of the first device;

[0247] The second device determines a management strategy for the first AI model based on the radio frequency performance indicator.

[0248] In some embodiments, the signal processing method further includes:

[0249] The second device sends indication information to the first device, where the indication information is used to indicate a management policy of the first AI model.

[0250] In some embodiments, the signal processing method further includes:

[0251] The second device receives a sample data set sent by the first device; wherein the sample data set includes a second digital signal and a sixth digital signal, wherein the sixth digital signal is converted from the first radio frequency signal;

[0252] The second device fine-tunes the first AI model based on the sample data set.

[0253] Optionally, the second device may be configured with a first AI model. Alternatively, the first device may send the first AI model or related information (e.g., model parameters) to the second device, so that the second device can determine the first AI model. In this way, the second device can use the first AI model to process the sixth digital signal in the sample data set to obtain a fifth digital signal; then calculate a loss function based on the second digital signal and the fifth digital signal, and fine-tune the first AI model based on the loss function.

[0254] In some embodiments, parameters related to the sample data set are configured by the second device.

[0255] In some embodiments, the relevant parameters include at least one of a data volume of the sample data set, a collection period, and a time interval between different samples in the sample data set.

[0256] For specific examples of the method executed by the second device of this embodiment, please refer to the relevant description of the second device in the aforementioned embodiment. For the sake of brevity, they will not be repeated here.

[0257] FIG13 is a schematic block diagram of a first device 1300 according to an embodiment of the present application. The first device 1300 may include:

[0258] The first processing module 1310 is configured to perform predistortion processing on the first digital signal output by the baseband link using the first AI model to obtain a second digital signal, and process the second digital signal into a first RF signal.

[0259] In one embodiment, the first processing module 1310 is further configured to:

[0260] The third digital signal is processed using the second AI model to obtain a radio frequency performance indicator; wherein the third digital signal is obtained by converting sampling information of at least one radio frequency signal sent by the first device, and the at least one radio frequency signal includes the first radio frequency signal; the radio frequency performance indicator is used to characterize the performance of the first AI model.

[0261] In one embodiment, the performance of the first AI model is related to a management policy of the first AI model.

[0262] In one embodiment, the management strategy of the first AI model includes model fine-tuning, model updating, or model switching.

[0263] In one embodiment, the radio frequency performance indicator is used to input into the first AI model; the first AI model is used to perform pre-distortion processing on the fourth digital signal output by the baseband link based on the radio frequency performance indicator.

[0264] In one embodiment, the sampling information includes a second radio frequency signal determined in at least one radio frequency signal based on the sampling parameter, or statistical information of at least one radio frequency signal.

[0265] In one embodiment, a sampling parameter of at least one radio frequency signal is configured by the second device.

[0266] In one embodiment, the sampling parameters include a sampling period and / or a sampling interval.

[0267] In one embodiment, the input information of the second AI model includes a third digital signal and second auxiliary information, where the second auxiliary information includes sampling information of at least one digital signal corresponding one-to-one to at least one RF signal and / or a precoding matrix used to obtain the at least one digital signal; wherein each digital signal of the at least one digital signal is processed by the first AI model and the RF link to obtain a corresponding RF signal;

[0268] The second AI model is used to process the third digital signal based on the second auxiliary information.

[0269] In one embodiment, the radio frequency performance indicator includes at least one of the following: radio frequency performance level information; error information of the first radio frequency signal; ACLR; and an accuracy rate indicator of information bits obtained by demodulating the third digital signal.

[0270] In one embodiment, the error information includes at least one of the following: vector magnitude error; frequency error; and phase error.

[0271] In one embodiment, as shown in FIG14 , the first device 1300 further includes:

[0272] The first communication module 1410 is configured to receive a trigger signaling from a second device, wherein the trigger signaling is used to trigger the first device to obtain a radio frequency performance indicator.

[0273] In one embodiment, the trigger signaling includes at least one of the following: type information of the radio frequency performance indicator; monitoring time information of the radio frequency performance indicator; and a threshold value of the radio frequency performance indicator.

[0274] In one embodiment, the first processing module 1310 is configured to:

[0275] When the first condition is met, the second AI model is used to process the third digital signal to obtain the RF performance index.

[0276] In one embodiment, the first condition is related to at least one of the following: the temperature of the first device; the operating voltage of the first device; the operating frequency band of the first device; the communication bandwidth of the first device; the waveform information used by the first device to send signals; the modulation order supported by the first device; and the continuous working time of the first device.

[0277] In one embodiment, as shown in FIG15 , the first device 1300 further includes a second communication module 1510 configured to send a first request to the second device; wherein the first request is configured to request monitoring of a radio frequency performance indicator;

[0278] The first processing module 1310 is configured to process the third digital signal using the second AI model to obtain a radio frequency performance indicator upon receiving consent information sent by the second device in response to the first request.

[0279] In one embodiment, as shown in FIG16 , the first device 1300 further includes:

[0280] The third communication module 1610 is used to send reporting information to the second device; wherein the reporting information includes radio frequency performance indicators, and the radio frequency performance indicators are used by the second device to determine the management strategy of the first AI model.

[0281] In one embodiment, the input information of the first AI model includes a first digital signal and first auxiliary information related to the operating performance of the RF link; the first processing module 1310 is used to use the first AI model to perform pre-distortion processing on the first digital signal based on the first auxiliary information.

[0282] In one embodiment, the first auxiliary information includes at least one of the following: the temperature of the first device; the operating voltage of the first device; the operating frequency band of the first device; the communication bandwidth of the first device; the waveform information used by the first device to send signals; the modulation order supported by the first device; and the continuous working time of the first device.

[0283] In one embodiment, as shown in FIG17 , the first device 1300 further includes:

[0284] The fourth communication module 1710 is used to receive indication information sent by the second device, where the indication information is used to indicate the management policy of the first AI model.

[0285] In one embodiment, the first processing module 1310 is further configured to:

[0286] A loss function is obtained based on the second digital signal and the fifth digital signal; wherein the fifth digital signal is obtained by processing the sixth digital signal by the first AI model, and the sixth digital signal is obtained by converting the first radio frequency signal;

[0287] Based on the loss function, the first AI model is fine-tuned.

[0288] In one embodiment, the first processing module 1310 is further configured to:

[0289] When the radio frequency performance indicator meets the second condition, it is determined that the fine-tuning of the first AI model is completed.

[0290] In one embodiment, as shown in FIG18 , the first device 1300 further includes:

[0291] The fifth communication module 1810 is used to send a sample data set to the second device; wherein the sample data set includes a second digital signal and a sixth digital signal; wherein the sixth digital signal is converted from the first RF signal; the sample data set is used to fine-tune the first AI model.

[0292] In one embodiment, relevant parameters of the sample data set are configured by the second device.

[0293] In one embodiment, the relevant parameters include at least one of the data volume of the sample data set, the collection period, and the time interval between different samples in the sample data set.

[0294] The first device 1300 of the embodiment of the present application can implement the corresponding functions of the first device in the aforementioned method embodiment. The processes, functions, implementation methods and beneficial effects corresponding to the various modules (sub-modules, units or components, etc.) in the first device 1300 can be found in the corresponding descriptions in the above-mentioned method embodiments, which will not be repeated here. It should be noted that the functions described in the various modules (sub-modules, units or components, etc.) in the first device 1300 of the application embodiment can be implemented by different modules (sub-modules, units or components, etc.) or by the same module (sub-module, unit or component, etc.).

[0295] FIG19 is a schematic block diagram of a second device 1900 according to an embodiment of the present application. The second device 1900 may include:

[0296] The sixth communication module 1910 is used to receive a first RF signal sent by a first device; wherein the first RF signal is obtained by the first device processing a second digital signal, and the second digital signal is obtained by pre-distorting the first digital signal output by the baseband link of the first device using the first AI model.

[0297] In one embodiment, the sixth communication module 1910 is further configured to:

[0298] Send first configuration information to the first device; wherein the first configuration information is used to configure sampling parameters, and the sampling parameters are used by the first device to determine sampling information of at least one RF signal, and the at least one RF signal includes the first RF signal; the sampling information is used to convert to obtain a third digital signal, and the third digital signal is processed by the second AI model to obtain a RF performance indicator, and the RF performance indicator is used to characterize the performance of the first AI model.

[0299] In one embodiment, the sampling parameters include a sampling period and / or a sampling interval.

[0300] In one embodiment, the sixth communication module 1910 is further configured to:

[0301] Send a trigger signaling to the first device; wherein the trigger signaling is used to trigger the first device to obtain a radio frequency performance indicator used to characterize the performance of the first AI model.

[0302] In one embodiment, the trigger signaling includes at least one of the following: type information of the radio frequency performance indicator; monitoring time information of the radio frequency performance indicator; and a threshold value of the radio frequency performance indicator.

[0303] In one embodiment, the sixth communication module 1910 is further configured to:

[0304] Receive a first request sent by a first device; wherein the first request is used to request monitoring of a radio frequency performance indicator of the first device;

[0305] Sending consent information for the first request to the first device; wherein the consent information is used to instruct the first device to obtain the radio frequency performance indicator.

[0306] In one implementation, the sixth communication module 1910 is further configured to receive reporting information sent by the first device, where the reporting information includes a radio frequency performance indicator of the first device.

[0307] Accordingly, as shown in FIG20 , the second device 1900 further includes:

[0308] The second processing module 2010 is used to determine the management strategy of the first AI model based on the radio frequency performance indicator.

[0309] In one embodiment, the sixth communication module 1910 is further configured to:

[0310] Send indication information to the first device, where the indication information is used to indicate a management policy of the first AI model.

[0311] In one embodiment, the sixth communication module 1910 is further configured to receive a sample data set sent by the first device; wherein the sample data set includes a second digital signal and a sixth digital signal, wherein the sixth digital signal is converted from the first radio frequency signal.

[0312] Accordingly, as shown in FIG21 , the second device 1900 further includes:

[0313] The third processing module 2110 is used to fine-tune the first AI model based on the sample data set.

[0314] In one embodiment, relevant parameters of the sample data set are configured by the second device.

[0315] In one embodiment, the relevant parameters include at least one of the data volume of the sample data set, a collection period, and a time interval between different samples.

[0316] The second device 1900 of the embodiment of the present application can implement the corresponding functions of the second device in the aforementioned method embodiment. The processes, functions, implementation methods and beneficial effects corresponding to the various modules (sub-modules, units or components, etc.) in the second device 1900 can be found in the corresponding descriptions in the above-mentioned method embodiments, which will not be repeated here. It should be noted that the functions described in the various modules (sub-modules, units or components, etc.) in the second device 1900 of the application embodiment can be implemented by different modules (sub-modules, units or components, etc.) or by the same module (sub-module, unit or component, etc.).

[0317] Figure 22 is a schematic structural diagram of a communication device 2200 according to an embodiment of the present application. The communication device 2200 includes a processor 2210, which can call and run a computer program from a memory to enable the communication device 2200 to implement the method in the embodiment of the present application.

[0318] In one embodiment, the communication device 2200 may further include a memory 2220. The processor 2210 may call and execute a computer program from the memory 2220 to enable the communication device 2200 to implement the method in the embodiment of the present application.

[0319] The memory 2220 may be a separate device independent of the processor 2210 , or may be integrated into the processor 2210 .

[0320] In one embodiment, the communication device 2200 may further include a transceiver 2230 , and the processor 2210 may control the transceiver 2230 to communicate with other devices. Specifically, the transceiver 2230 may send information or data to other devices, or receive information or data sent by other devices.

[0321] The transceiver 2230 may include a transmitter and a receiver. The transceiver 2230 may further include an antenna, and the number of antennas may be one or more.

[0322] In one embodiment, the communication device 2200 may be the first device of the embodiment of the present application, and the communication device 2200 may implement the corresponding processes implemented by the first device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0323] In one embodiment, the communication device 2200 may be the second device of the embodiment of the present application, and the communication device 2200 may implement the corresponding processes implemented by the second device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0324] 23 is a schematic structural diagram of a chip 2300 according to an embodiment of the present application. The chip 2300 includes a processor 2310, which can call and execute a computer program from a memory to implement the method according to the embodiment of the present application.

[0325] In one embodiment, the chip 2300 may further include a memory 2320. The processor 2310 may call and execute a computer program from the memory 2320 to implement the method performed by the first device or the second device in the embodiment of the present application.

[0326] The memory 2320 may be a separate device independent of the processor 2310 or may be integrated into the processor 2310 .

[0327] In one embodiment, the chip 2300 may further include an input interface 2330. The processor 2310 may control the input interface 2330 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

[0328] In one embodiment, the chip 2300 may further include an output interface 2340. The processor 2310 may control the output interface 2340 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

[0329] In one embodiment, the chip can be applied to the first device in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the first device in each method of the embodiment of the present application. For the sake of brevity, it will not be repeated here.

[0330] In one embodiment, the chip can be applied to the second device in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the second device in each method of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0331] The chips used in the first device and the second device may be the same chip or different chips.

[0332] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0333] The processor mentioned above may be a general-purpose processor, a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or other programmable logic devices, transistor logic devices, discrete hardware components, etc. The general-purpose processor mentioned above may be a microprocessor or any conventional processor, etc.

[0334] The memory mentioned above may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM).

[0335] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0336] FIG24 is a schematic block diagram of a communication system 2400 according to an embodiment of the present application. The communication system 2400 includes a first device 1300 and a second device 1900 .

[0337] The first device 1300 uses the first AI model to perform predistortion processing on the first digital signal output by the baseband link to obtain a second digital signal; the first device 1300 processes the second digital signal into a first RF signal.

[0338] The second device 1900 receives the first radio frequency signal sent by the first device 1300 .

[0339] The first device 1300 can be used to implement the corresponding functions implemented by the first device in the above method, and the second device 1900 can be used to implement the corresponding functions implemented by the second device in the above method. For the sake of brevity, they are not described here in detail.

[0340] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function in accordance with the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center by wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode to another website, computer, server or data center. 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 includes one or more available media integrations. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0341] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0342] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0343] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A signal processing method, comprising: The first device uses a first artificial intelligence AI model to perform predistortion processing on a first digital signal output by a baseband link to obtain a second digital signal; The first device processes the second digital signal into a first radio frequency signal.

2. The method according to claim 1, wherein: The method further comprises: The first device uses the second AI model to process the third digital signal to obtain a RF performance indicator; wherein the third digital signal is converted from sampling information of at least one RF signal sent by the first device, and the at least one RF signal includes the first RF signal; the RF performance indicator is used to characterize the performance of the first AI model.

3. The method according to claim 2, wherein: The performance of the first AI model is related to the management strategy of the first AI model.

4. The method according to claim 3, wherein: The management strategy of the first AI model includes model fine-tuning, model updating or model switching.

5. The method according to any one of claims 2 to 4, wherein: The RF performance indicator is used to input the first AI model; the first AI model is used to perform pre-distortion processing on the fourth digital signal output by the baseband link based on the RF performance indicator.

6. The method according to any one of claims 2 to 5, wherein: The sampling information includes a second radio frequency signal determined in the at least one radio frequency signal based on a sampling parameter, or statistical information of the at least one radio frequency signal.

7. The method according to any one of claims 2 to 6, wherein: The sampling parameters of the at least one radio frequency signal are configured by the second device.

8. The method according to claim 7, wherein: The sampling parameters include a sampling period and / or a sampling interval.

9. The method according to any one of claims 2 to 8, wherein: The input information of the second AI model includes the third digital signal and the second auxiliary information, and the second auxiliary information includes sampling information of at least one digital signal corresponding to the at least one RF signal and / or a precoding matrix for obtaining the at least one digital signal; wherein each digital signal in the at least one digital signal is processed by the first AI model and the RF link to obtain a corresponding RF signal; The second AI model is used to process the third digital signal based on the second auxiliary information.

10. The method according to any one of claims 2 to 9, wherein: The radio frequency performance index includes at least one of the following: RF performance level information; error information of the first radio frequency signal; Adjacent channel leakage ratio ACLR; An accuracy index of information bits obtained by demodulating the third digital signal.

11. The method according to claim 10, wherein: The error information includes at least one of the following: Vector magnitude error; Frequency error; Phase error.

12. The method according to any one of claims 2 to 11, wherein: The method further comprises: The first device receives a trigger signaling from the second device; wherein the trigger signaling is used to trigger the first device to obtain the radio frequency performance indicator.

13. The method according to claim 12, wherein: The trigger signaling includes at least one of the following: Type information of the radio frequency performance indicator; Monitoring time information of the radio frequency performance indicator; The threshold of the radio frequency performance indicator.

14. The method according to any one of claims 2 to 13, wherein: The first device processes the third digital signal using the second AI model to obtain radio frequency performance indicators, including: When the first condition is met, the first device uses the second AI model to process the third digital signal to obtain the RF performance indicator.

15. The method according to claim 14, wherein: The first condition is related to at least one of the following: the temperature of the first device; an operating voltage of the first device; an operating frequency band of the first device; the communication bandwidth of the first device; Waveform information used by the first device to send a signal; a modulation order supported by the first device; The continuous working time of the first device.

16. The method according to any one of claims 2 to 15, wherein: The first device processes the third digital signal using the second AI model to obtain radio frequency performance indicators, including: The first device sends a first request to the second device; wherein the first request is used to request monitoring of the radio frequency performance indicator; When the first device receives the consent information sent by the second device in response to the first request, the first device uses the second AI model to process the third digital signal to obtain the radio frequency performance indicator.

17. The method according to any one of claims 2 to 16, wherein: The method further comprises: The first device sends reporting information to the second device; wherein the reporting information includes the radio frequency performance indicator, and the radio frequency performance indicator is used by the second device to determine the management strategy of the first AI model.

18. The method according to any one of claims 1 to 17, wherein: The input information of the first AI model includes the first digital signal and first auxiliary information related to the working performance of the radio frequency link; The first device uses a first artificial AI model to perform predistortion processing on a first digital signal output by a baseband link, including: The first device uses the first AI model to perform pre-distortion processing on the first digital signal based on the first auxiliary information.

19. The method according to claim 18, wherein: The first auxiliary information includes at least one of the following: the temperature of the first device; an operating voltage of the first device; an operating frequency band of the first device; the communication bandwidth of the first device; Waveform information used by the first device to send a signal; a modulation order supported by the first device; The continuous working time of the first device.

20. The method according to any one of claims 1 to 19, wherein: The method further comprises: The first device receives indication information sent by the second device, wherein the indication information is used to indicate a management policy of the first AI model.

21. The method according to any one of claims 1 to 20, wherein: The method further comprises: The first device obtains a loss function based on the second digital signal and the fifth digital signal; wherein the fifth digital signal is obtained by processing the sixth digital signal by the first AI model, and the sixth digital signal is obtained by converting the first radio frequency signal; The first device fine-tunes the first AI model based on the loss function.

22. The method according to claim 21, wherein: The method further comprises: When the radio frequency performance indicator meets the second condition, the first device determines that fine-tuning of the first AI model is completed.

23. The method according to any one of claims 1 to 20, wherein: The method further comprises: The first device sends a sample data set to the second device; wherein the sample data set includes the second digital signal and a sixth digital signal; wherein the sixth digital signal is converted from the first RF signal; and the sample data set is used to fine-tune the first AI model.

24. The method according to claim 23, wherein: The relevant parameters of the sample data set are configured by the second device.

25. The method according to claim 24, wherein: The relevant parameters include at least one of the data volume of the sample data set, the collection period, and the time interval between different samples in the sample data set.

26. A signal processing method, comprising: The second device receives a first RF signal sent by the first device; wherein the first RF signal is obtained by the first device processing a second digital signal, and the second digital signal is obtained by pre-distorting the first digital signal output by the baseband link of the first device using a first AI model.

27. The method according to claim 26, wherein: The method further comprises: The second device sends first configuration information to the first device; wherein the first configuration information is used to configure sampling parameters, and the sampling parameters are used by the first device to determine sampling information of at least one RF signal, and the at least one RF signal includes the first RF signal; the sampling information is used to convert to obtain a third digital signal, and the third digital signal is processed by the second AI model to obtain an RF performance indicator, and the RF performance indicator is used to characterize the performance of the first AI model.

28. The method according to claim 27, wherein: The sampling parameters include a sampling period and / or a sampling interval.

29. The method according to any one of claims 26 to 28, wherein: The method further comprises: The second device sends a trigger signaling to the first device; wherein the trigger signaling is used to trigger the first device to obtain a radio frequency performance indicator used to characterize the performance of the first AI model.

30. The method of claim 29, wherein: The trigger signaling includes at least one of the following: Type information of the radio frequency performance indicator; Monitoring time information of the radio frequency performance indicator; The threshold of the radio frequency performance indicator.

31. The method according to any one of claims 26 to 28, wherein: The method further comprises: The second device receives a first request sent by the first device; wherein the first request is used to request monitoring of a radio frequency performance indicator of the first device; The second device sends consent information for the first request to the first device; wherein the consent information is used to instruct the first device to obtain the radio frequency performance indicator.

32. The method according to any one of claims 26 to 31, wherein: The method further comprises: The second device receives the reporting information sent by the first device, wherein the reporting information includes a radio frequency performance indicator of the first device; The second device determines a management strategy of the first AI model based on the radio frequency performance indicator.

33. The method according to any one of claims 26 to 32, wherein: The method further comprises: The second device sends indication information to the first device, wherein the indication information is used to indicate a management policy of the first AI model.

34. The method according to any one of claims 26 to 33, wherein: The method further comprises: The second device receives a sample data set sent by the first device; wherein the sample data set includes the second digital signal and a sixth digital signal, wherein the sixth digital signal is converted from the first radio frequency signal; The second device fine-tunes the first AI model based on the sample data set.

35. The method of claim 34, wherein: The relevant parameters of the sample data set are configured by the second device.

36. The method of claim 35, wherein: The relevant parameters include at least one of the data volume of the sample data set, the collection period, and the time interval between different samples in the sample data set.

37. A first device, comprising: The first processing module is used to use a first artificial intelligence AI model to perform pre-distortion processing on a first digital signal output by a baseband link to obtain a second digital signal, and process the second digital signal into a first radio frequency signal.

38. The first device according to claim 37, wherein: The first processing module is also used for: The third digital signal is processed using the second AI model to obtain a radio frequency performance indicator; wherein the third digital signal is converted from sampling information of at least one radio frequency signal sent by the first device, and the at least one radio frequency signal includes the first radio frequency signal; the radio frequency performance indicator is used to characterize the performance of the first AI model.

39. The first device according to claim 38, wherein: The performance of the first AI model is related to the management strategy of the first AI model.

40. The first device according to claim 39, wherein: The management strategy of the first AI model includes model fine-tuning, model updating or model switching.

41. A first device according to any one of claims 38 to 40, wherein: The RF performance indicator is used to input the first AI model; the first AI model is used to perform pre-distortion processing on the fourth digital signal output by the baseband link based on the RF performance indicator.

42. A first device according to any one of claims 38 to 41, wherein: The sampling information includes a second radio frequency signal determined in the at least one radio frequency signal based on a sampling parameter, or statistical information of the at least one radio frequency signal.

43. A first device according to any one of claims 38 to 42, wherein: The sampling parameters of the at least one radio frequency signal are configured by the second device.

44. The first device according to claim 43, wherein The sampling parameters include a sampling period and / or a sampling interval.

45. A first device according to any one of claims 38 to 44, wherein: The input information of the second AI model includes the third digital signal and the second auxiliary information, and the second auxiliary information includes sampling information of at least one digital signal corresponding to the at least one RF signal and / or a precoding matrix for obtaining the at least one digital signal; wherein each digital signal in the at least one digital signal is processed by the first AI model and the RF link to obtain a corresponding RF signal; The second AI model is used to process the third digital signal based on the second auxiliary information.

46. ​​A first device according to any one of claims 38 to 45, wherein: The radio frequency performance index includes at least one of the following: RF performance level information; error information of the first radio frequency signal; ACLR; An accuracy index of information bits obtained by demodulating the third digital signal.

47. The first device according to claim 46, wherein The error information includes at least one of the following: Vector magnitude error; Frequency error; Phase error.

48. A first device according to any one of claims 38 to 47, wherein: The first device also includes: The first communication module is used to receive a trigger signaling from a second device; wherein the trigger signaling is used to trigger the first device to obtain the radio frequency performance indicator.

49. The first device according to claim 48, wherein The trigger signaling includes at least one of the following: Type information of the radio frequency performance indicator; Monitoring time information of the radio frequency performance indicator; The threshold of the radio frequency performance indicator.

50. The first device according to any one of claims 38-49, wherein: The first processing module is used for: When the first condition is met, the second AI model is used to process the third digital signal to obtain the RF performance indicator.

51. The first device according to claim 50, wherein The first condition is related to at least one of the following: the temperature of the first device; an operating voltage of the first device; an operating frequency band of the first device; the communication bandwidth of the first device; Waveform information used by the first device to send a signal; a modulation order supported by the first device; The continuous working time of the first device.

52. A first device according to any one of claims 38 to 47, wherein: The first device also includes a second communication module, configured to send a first request to the second device; wherein the first request is used to request monitoring of the radio frequency performance indicator; The first processing module is used to use the second AI model to process the third digital signal to obtain the RF performance indicator when receiving the consent information sent by the second device in response to the first request.

53. A first device according to any one of claims 38 to 52, wherein: The first device also includes: The third communication module is used to send reporting information to the second device; wherein the reporting information includes the radio frequency performance indicator, and the radio frequency performance indicator is used by the second device to determine the management strategy of the first AI model.

54. A first device according to any one of claims 37 to 53, wherein: The input information of the first AI model includes the first digital signal and first auxiliary information related to the working performance of the radio frequency link; The first processing module is used for: The first AI model is used to perform pre-distortion processing on the first digital signal based on the first auxiliary information.

55. The first device according to claim 54, wherein The first auxiliary information includes at least one of the following: the temperature of the first device; an operating voltage of the first device; an operating frequency band of the first device; the communication bandwidth of the first device; Waveform information used by the first device to send a signal; a modulation order supported by the first device; The continuous working time of the first device.

56. A first device according to any one of claims 37 to 55, wherein: The first device also includes: A fourth communication module is used to receive indication information sent by the second device, wherein the indication information is used to indicate the management strategy of the first AI model.

57. A first device according to any one of claims 37 to 56, wherein: The first processing module is also used for: A loss function is obtained based on the second digital signal and the fifth digital signal; wherein the fifth digital signal is obtained by processing the sixth digital signal by the first AI model, and the sixth digital signal is obtained by converting the first radio frequency signal; Based on the loss function, fine-tune the first AI model.

58. The first device according to claim 57, wherein The first processing module is also used for: When the radio frequency performance indicator meets the second condition, it is determined that the fine-tuning of the first AI model is completed.

59. The first device according to any one of claims 37-56, wherein: The first device also includes: a fifth communication module, configured to send a sample data set to the second device; wherein the sample data set includes the second digital signal and a sixth digital signal; wherein the sixth digital signal is obtained by converting the first radio frequency signal; and the sample data set is used to The first AI model is fine-tuned.

60. The first device according to claim 59, wherein The relevant parameters of the sample data set are configured by the second device.

61. The first device according to claim 60, wherein: The relevant parameters include at least one of the data volume of the sample data set, the collection period, and the time interval between different samples in the sample data set.

62. A second device, comprising: The sixth communication module is used to receive a first RF signal sent by a first device; wherein the first RF signal is obtained by the first device processing a second digital signal, and the second digital signal is obtained by pre-distorting the first digital signal output by the baseband link of the first device using a first AI model.

63. The second device according to claim 62, wherein: The sixth communication module is also used for: Sending first configuration information to the first device; wherein the first configuration information is used to configure sampling parameters, and the sampling parameters are used by the first device to determine sampling information of at least one RF signal, and the at least one RF signal includes the first RF signal; the sampling information is used to convert to obtain a third digital signal, and the third digital signal is processed by the second AI model to obtain a RF performance indicator, and the RF performance indicator is used to characterize the performance of the first AI model.

64. The second device according to claim 63, wherein The sampling parameters include a sampling period and / or a sampling interval.

65. The second device according to any one of claims 62 to 64, wherein: The sixth communication module is also used for: Send a trigger signaling to the first device; wherein the trigger signaling is used to trigger the first device to obtain a radio frequency performance indicator used to characterize the performance of the first AI model.

66. The second device according to claim 65, wherein The trigger signaling includes at least one of the following: Type information of the radio frequency performance indicator; Monitoring time information of the radio frequency performance indicator; The threshold of the radio frequency performance indicator.

67. The second device according to any one of claims 62-64, wherein The sixth communication module is also used for: Receive a first request sent by the first device; wherein the first request is used to request monitoring of a radio frequency performance indicator of the first device; Sending consent information for the first request to the first device; wherein the consent information is used to instruct the first device to obtain the radio frequency performance indicator.

68. The second device according to any one of claims 62-67, wherein The sixth communication module is further configured to receive reporting information sent by the first device, wherein the reporting information includes a radio frequency performance indicator of the first device; The second device also includes: The second processing module is used to determine the management strategy of the first AI model based on the radio frequency performance indicator.

69. The second device according to any one of claims 62-68, wherein The sixth communication module is also used for: Send indication information to the first device, wherein the indication information is used to indicate a management policy of the first AI model.

70. The second device according to any one of claims 62-69, wherein The sixth communication module is further used to receive a sample data set sent by the first device; wherein the sample data set includes the second digital signal and a sixth digital signal, wherein the sixth digital signal is obtained by converting the first radio frequency signal; The second device also includes: The third processing module is used to fine-tune the first AI model based on the sample data set.

71. The second device according to claim 70, wherein: The relevant parameters of the sample data set are configured by the second device.

72. The second device according to claim 71, wherein The relevant parameters include at least one of the data volume of the sample data set, the collection period, and the time interval between different samples in the sample data set.

73. A first device, comprising: A transceiver, a processor and a memory, wherein the memory is used to store a computer program, the transceiver is used to communicate with other devices, and the processor is used to call and run the computer program stored in the memory so that the first device executes the method as described in any one of claims 1 to 25.

74. A second device, comprising: A transceiver, a processor and a memory, wherein the memory is used to store a computer program, the transceiver is used to communicate with other devices, and the processor is used to call and run the computer program stored in the memory so that the second device executes the method as described in any one of claims 26 to 36.

75. A chip, comprising: A processor, configured to call and run a computer program from a memory, so that a device equipped with the chip executes a method as claimed in any one of claims 1 to 25.

76. A chip, comprising: A processor, configured to call and run a computer program from a memory, so that a device equipped with the chip executes a method as claimed in any one of claims 26 to 36.

77. A computer-readable storage medium for storing a computer program, which, when executed by a device, causes the device to perform the method as claimed in any one of claims 1 to 25.

78. A computer-readable storage medium for storing a computer program, which, when executed by a device, causes the device to perform the method as claimed in any one of claims 26 to 36.

79. A computer program product comprising computer program instructions, the computer program instructions causing a computer to perform the method as claimed in any one of claims 1 to 25.

80. A computer program product comprising computer program instructions for causing a computer to perform the method of any one of claims 26 to 36.

81. A computer program, the computer program causing a computer to execute the method according to any one of claims 1 to 25.

82. A computer program causing a computer to execute the method of any one of claims 26 to 36.

83. A communication system comprising: A first device, configured to perform the method according to any one of claims 1 to 25; A second device, configured to execute the method according to any one of claims 26 to 36.

Images