A method and system for three-dimensional positioning of a thunder sound source

Abstract

The application discloses a three-dimensional positioning method and system of a thunder sound source, and relates to the technical field of thunder sound source positioning.The method comprises the following steps: collecting original signals in real time through a distributed sound wave sensing interrogation unit connected with one end of a communication optical cable; performing denoising preprocessing on the original signals through filtering to obtain filtered signals; automatically obtaining signal arrival times of different sensing channels based on the filtered signals by using a hybrid strategy; constructing an atmospheric layered sound velocity propagation model to correct a propagation time error to obtain a signal theoretical arrival time; performing three-dimensional space inversion by using a preset algorithm based on the signal arrival time and the signal theoretical arrival time to obtain time-space parameters of the thunder sound source; and performing thunder sound source visual reconstruction in a three-dimensional coordinate system based on optimal solution data and the communication optical cable. The method can realize high-precision three-dimensional positioning of the thunder sound source while ensuring calculation efficiency.

Description

Technical Field

[0001] This application relates to the field of lightning sound source localization technology, and in particular to a three-dimensional localization method and system for lightning sound sources. Background Technology

[0002] Currently, mainstream lightning location technologies are mainly divided into two categories: electromagnetic detection and acoustic detection. Electromagnetic detection technologies, such as very low frequency / low frequency (VLF / LF) location systems or very high frequency (VHF) lightning mapping arrays (LMA), while possessing high temporal resolution, have limitations in close-range detection blind spots, detailed depiction of lightning within clouds, and detection capabilities in electromagnetically shielded environments. In contrast, thunder, as a shock wave generated by the rapid heating of the lightning discharge channel, carries rich channel geometry information in its acoustic signal. Traditional acoustic lightning location primarily relies on point microphone arrays. These systems typically consist of several airborne acoustic sensors (microphones) arranged in a specific geometry, retrieving the sound source location by measuring the time difference of arrival (TDOA) of the thunder at each sensor or using beamforming algorithms. However, with the acceleration of urbanization and the increasing requirements for monitoring accuracy, traditional microphone arrays have gradually shown their shortcomings in terms of deployment density, maintenance costs, environmental adaptability, and the degree of automation in data processing, making it difficult to meet the needs of large-scale, high-precision, and refined lightning monitoring.

[0003] The invention patent with authorization announcement number CN103792513B provides a thunder sound location system and method. The system includes a detection device, a data acquisition module, and a data processing module. The detection device uses a 19-element cross-array microphone array to detect thunder signals and lightning electromagnetic signals, which are then sent to the data acquisition module. After preprocessing by the data acquisition module, the data is output to the data processing module. The data processing module processes the thunder signals obtained through a frequency domain beamforming algorithm in stages to determine the azimuth and elevation angles of the thunder signals. Combined with the time difference of arrival (TDOA) of the sound and electricity signals, the distance to the thunder source is determined, achieving a three-dimensional location result for the lightning occurrence location and channel. However, this system uses a small array physical aperture, has low data inversion efficiency, lacks an automation mechanism, and has poor accuracy in locating the lightning source. Summary of the Invention

[0004] Therefore, this application addresses the problem of poor localization accuracy of lightning sound sources in existing technologies by providing a three-dimensional localization method and system for lightning sound sources, which can improve the accuracy of lightning sound source localization. The specific technical solution is as follows: Firstly, this application proposes a three-dimensional localization method for lightning sound sources, comprising the following steps: S1: The original signals along the communication optical cable are collected in real time by a distributed acoustic wave sensing interrogation unit connected to one end of the communication optical cable. S2: The original signal is preprocessed by filtering to remove noise, resulting in a filtered signal; S3: Based on the filtered signal, a hybrid strategy is adopted to automatically obtain the signal arrival time of different sensing channels, wherein the sensing channel is a virtual sensing channel that discretizes the communication optical cable. S4: Construct an atmospheric stratification sound speed propagation model to correct the propagation time error caused by atmospheric refraction and sound speed changes in the signal arrival time, and obtain the theoretical arrival time of the signal; S5: Based on the arrival time of the signal and the theoretical arrival time of the signal, a preset algorithm is used to perform three-dimensional spatial inversion to obtain the spatiotemporal parameters of the lightning source; S6: Based on the spatiotemporal parameters of the lightning sound source and the distribution of the communication optical cable, perform a visual reconstruction of the lightning sound source in a three-dimensional coordinate system.

[0005] Furthermore, the raw signal mentioned in step S1 includes raw acoustic wave field data of thunder signals and environmental noise.

[0006] Furthermore, the specific content of the noise reduction preprocessing by filtering described in step S2 is as follows: First, the raw signal is filtered using a bandpass filter to remove high-frequency electronic noise and extremely low-frequency temperature drift. Second, a visual velocity cutoff threshold is set, and a frequency-domain-wavenumber domain filter is used to filter out ground noise with visual velocities below the cutoff threshold. The bandpass filter includes a Butterworth bandpass filter.

[0007] Furthermore, the specific steps of automatically acquiring the signal arrival times of different sensing channels using a hybrid strategy in step S3 include: S301: Apply the short-time-long-time average ratio algorithm to all sensor channels to calculate the approximate signal arrival time for each sensor channel; S302: Calculate the signal-to-noise ratio of the signal in each sensing channel, take the sensing channel that meets the preset signal-to-noise ratio and has the earliest coarse signal arrival time as the reference channel, and take the coarse signal arrival time of the reference channel as the absolute time reference. S303: Using a generalized cross-correlation phase transform algorithm, the signal of the i-th sensing channel is cross-correlated with the signal of the reference channel. By searching for the peak value of the cross-correlation function within the maximum allowable physical delay window, the time difference between the i-th sensing channel and the reference channel is accurately calculated. ; S304: The absolute time reference is superimposed with the time difference to obtain the signal arrival time of the i-th sensing channel.

[0008] The specific details of the short-time-long-time average ratio algorithm are as follows: Set a short-time window and a long-time window. When the ratio of the energy of the short-time window to the energy of the long-time window exceeds a preset threshold, the time of the short-time window is determined to be the approximate signal arrival time.

[0009] Furthermore, the atmospheric stratified sound speed propagation model described in step S4 includes a sound speed profile model, and the theoretical arrival time of the signal is calculated using a stratified integration method along the propagation path.

[0010] Further, the specific process of the three-dimensional spatial inversion in step S5 is as follows: A four-dimensional unknown parameter set containing the thunder occurrence time and three-dimensional spatial coordinates is constructed; a residual objective function is constructed based on the signal arrival time and the theoretical signal arrival time; a parameter search boundary is set; and a preset algorithm is used to solve the unknown parameter set until the residual objective function converges to its minimum value. The solution at this point is used as the spatiotemporal parameters of the lightning source. The preset algorithm includes the trust region reflection algorithm and the Levenberg-Marquardt algorithm, both of which belong to the nonlinear least squares method.

[0011] Secondly, this application proposes a three-dimensional positioning system for a lightning sound source, including a memory and a processor; the memory is used to store instructions; the processor is used to operate according to the instructions to execute the three-dimensional positioning system for a lightning sound source proposed in this application.

[0012] This application provides a three-dimensional localization method and system for lightning sound sources. The method utilizes communication optical cables as a sensor array and combines signal processing and nonlinear inversion algorithms to achieve high-precision localization of the three-dimensional position of lightning sound sources while ensuring computational efficiency. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating a three-dimensional localization method for a lightning sound source according to an embodiment of this application; Figure 2 This is a diagram of the experimental environment for an embodiment of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0015] Example 1 This embodiment provides a three-dimensional localization system for lightning sound sources, and the method flowchart is as follows. Figure 1 As shown, it includes the following specific content: S1: The original signals along the communication optical cable are collected in real time by a distributed acoustic wave sensing interrogation unit connected to one end of the communication optical cable. It should be noted that the raw signal mentioned in step S1 includes the raw acoustic wave field data of thunder signals and environmental noise.

[0016] In one specific embodiment, the communication optical cable is typically deployed underground in a city, and the raw acoustic wave field data is directly reflected in the Rayleigh scattering phase or strain rate data of the communication optical cable. In this case, the communication optical cable is constructed into a linear acoustic sensing array with an ultra-large aperture.

[0017] S2: The original signal is preprocessed by filtering to remove noise, resulting in a filtered signal; It should be noted that the specific content of the noise reduction preprocessing by filtering described in step S2 is as follows: First, the original signal is filtered using a bandpass filter. Based on the spectral characteristics of the thunder signal, a preset frequency band where thunder energy is concentrated is retained, while high-frequency electronic noise and extremely low-frequency temperature drift are filtered out. Second, an apparent velocity cutoff threshold is set, and a frequency-domain-wavenumber domain filter is used to filter out ground noise with an apparent velocity below the cutoff threshold. The bandpass filter includes a Butterworth bandpass filter.

[0018] In one specific embodiment, the preset frequency band is 10Hz-1000Hz, the apparent velocity cutoff threshold is set to the speed of air sound, i.e., 340m / s, the signal components with an apparent velocity greater than or equal to the speed of air sound are retained, and ground noise such as vehicle noise with a lower apparent velocity is filtered out, thereby improving the signal-to-noise ratio (SNR) of the original signal, highlighting the waveform characteristics of the thunder signal, and providing a high-quality data foundation for subsequent automatic acquisition.

[0019] S3: Based on the filtered signal, a hybrid strategy is adopted to automatically obtain the signal arrival time of different sensing channels, wherein the sensing channel is a virtual sensing channel that discretizes the communication optical cable. It should be noted that the specific steps of automatically acquiring the signal arrival times of different sensing channels using a hybrid strategy in step S3 include: S301: Apply the short-time-long-time average ratio algorithm to all sensor channels to calculate the approximate signal arrival time for each sensor channel; S302: Calculate the signal-to-noise ratio of the signal in each sensing channel, take the sensing channel that meets the preset signal-to-noise ratio and has the earliest coarse signal arrival time as the reference channel, and take the coarse signal arrival time of the reference channel as the absolute time reference. S303: Using a generalized cross-correlation phase transform algorithm, the signal of the i-th sensing channel is cross-correlated with the signal of the reference channel. By searching for the peak value of the cross-correlation function within the maximum allowable physical delay window, the time difference between the i-th sensing channel and the reference channel is accurately calculated. ; S304: Superimpose the absolute time reference with the time difference to obtain the signal arrival time of the i-th sensing channel, expressed as follows: .

[0020] The specific details of the short-time-long-time average ratio algorithm are as follows: Set a short-time window and a long-time window. When the ratio of the energy of the short-time window to the energy of the long-time window exceeds a preset threshold, the time of the short-time window is determined to be the approximate signal arrival time.

[0021] Step S3 utilizes the spatial coherence of the original signal to accurately extract the arrival time information of weak signals in a strong noise environment, solving the problems of low efficiency and poor consistency of manual pickup, and recovering the time-distance curve (Move-out) that can reflect the characteristics of sound wave propagation.

[0022] S4: Construct an atmospheric stratification sound speed propagation model to correct the propagation time error caused by atmospheric refraction and sound speed changes in the signal arrival time, and obtain the theoretical arrival time of the signal; It should be noted that the atmospheric stratified sound speed propagation model described in step S4 includes a sound speed profile model, and the theoretical arrival time of the signal is calculated using a stratified integration method along the propagation path. The sound speed profile model incorporates the physical law that atmospheric temperature decreases linearly with altitude, and sets a reference altitude and surface temperature, making it more consistent with the theoretical travel time calculation basis of the real atmospheric environment compared to the constant sound speed model.

[0023] S5: Based on the arrival time of the signal and the theoretical arrival time of the signal, a preset algorithm is used to perform three-dimensional spatial inversion to obtain the spatiotemporal parameters of the lightning source; It should be noted that the specific process of the three-dimensional spatial inversion described in step S5 is as follows: constructing a system that includes the time of thunder occurrence. and three-dimensional spatial coordinates The four-dimensional unknown parameter set is used to construct a residual objective function based on the signal arrival time and the theoretical signal arrival time. A parameter search boundary is set, and a preset algorithm is used to solve the unknown parameter set until the residual objective function converges to its minimum value. The corresponding solution at this point represents the spatiotemporal parameters of the lightning source. The preset algorithm includes the Trust Region Reflective algorithm and the Levenberg-Marquardt algorithm, both of which are nonlinear least squares methods.

[0024] In one specific embodiment, the residual objective function is used to characterize the total error between the signal arrival time and the theoretical signal arrival time. It is usually the sum of squared residuals. By replacing the traditional discrete grid search method with the nonlinear least squares method, the mathematically optimal solution is directly and quickly approximated in the continuous solution space, realizing three-dimensional lightning positioning with sub-meter level computational accuracy, while significantly reducing the computation time.

[0025] S6: Based on the spatiotemporal parameters of the lightning sound source and the distribution of the communication optical cable, perform a visual reconstruction of the lightning sound source in a three-dimensional coordinate system.

[0026] The method utilizes existing underground communication fiber optic cable resources in cities to construct a linear acoustic array with a kilometer-scale ultra-large aperture and meter-scale high-density sampling through distributed optical fiber sensing (DAS) technology. This stealth observation network requires no additional hardware or power supply deployment in the field, and utilizes the underground buried environment to naturally shield against surface wind noise and high-frequency urban noise. Compared with traditional small-aperture microphone arrays, it has stronger anti-interference capabilities and higher azimuth resolution. Meanwhile, the method innovatively proposes a fully automatic signal arrival time (TOA) detection mechanism. This mechanism utilizes the energy maximization principle to select the reference channel, combines the short-long-time average ratio (STA / LTA) algorithm for initial absolute time detection, and applies the generalized cross-correlation phase transform (GCC-PHAT) algorithm to process massive array data. By leveraging the spatial coherence of the signal, it accurately calculates the relative time difference of each sensing channel in a low signal-to-noise ratio environment, achieving millisecond-level automatic acquisition of kilometer-scale array signals. Finally, the method constructs a four-dimensional objective function inversion model based on continuous space. By employing optimization algorithms such as nonlinear least squares and trust region reflection, the spatiotemporal optimal solution of the lightning sound source is directly approximated iteratively in the continuous solution space. This theoretically improves the positioning accuracy to sub-meter level while significantly reducing the time complexity, thus meeting the requirements for rapid three-dimensional reconstruction.

[0027] Example 2 This embodiment will verify the effectiveness of the method proposed in this application through experiments, the specific content of which is as follows: 1. Experimental scenario and data acquisition (corresponding method step S1) The experimental environment in this embodiment is as follows: Figure 2 As shown, an existing underground optical fiber communication cable located on the Guangzhou University Town outbound expressway section (near the newly built Pearl River Bridge) was selected as the sensing carrier. The DAS demodulation equipment was placed in the computer room of the Guangzhou Supercomputing Center and connected to one end of the optical fiber cable.

[0028] Sensor array configuration: One core of the optical cable is used as the sensing channel. The spatial sampling interval (Channel Spacing) of the DAS system is set to 2.0 meters, and the sampling rate (fs) is set to 5000Hz.

[0029] Data Selection: The experiment captured a severe thunderstorm process between 17:10 and 18:10 Beijing time on September 17, 2025. This embodiment extracts a data segment containing obvious thunder signals for analysis, and selects the section of the optical cable with a relatively good signal-to-noise ratio, namely channel index 4500 to 6500 (corresponding to a sensing length of approximately 4 kilometers), as the effective detection array.

[0030] Coordinate calibration: The precise geographic coordinates (longitude, latitude, and altitude) of each channel from 4500 to 6500 were obtained through optical cable routing map and GPS mapping, and the longitude and latitude were converted into a local Cartesian coordinate system (x, y, z) with the array center as the origin.

[0031] 2. Signal preprocessing and noise reduction (corresponding to step S2) Because the optical cable is laid along the newly built Pearl River Bridge, road traffic noise (vehicle vibration) constitutes a major interference to the extraction of thunder signals. This embodiment adopts a combined noise reduction strategy integrated into the code: First, the original signal is subjected to Butterworth bandpass filtering, with the passband frequency set to 10Hz to 1000Hz. This frequency band covers the main energy range of thunder, while filtering out extremely low-frequency temperature drift and high-frequency electronic noise. For the strong traffic noise in highway environments, FK filtering technology is applied. The apparent speed cutoff threshold (v_cutoff) is set to 340m / s.

[0032] Function and Effect: Thunder, as an airborne sound wave, typically has an apparent velocity greater than or equal to the speed of sound in air (approximately 340 m / s) when sweeping across the fiber optic array; while the apparent velocity of ground vibration waves caused by vehicles is usually lower. By filtering out signal components with apparent velocities below 340 m / s, the thunder signal and traffic background noise were successfully separated, significantly improving the signal-to-noise ratio and making the faint thunder wavefront clearly visible on the "waterfall diagram".

[0033] 3. Automatic detection and acquisition of signal arrival time (TOA) (corresponding to method step S3) To address the inefficiency of manually picking up 2000 channels (channels 4500-6500), this embodiment employs an automatic picking algorithm combining "coarse inspection + fine-tuning": STA / LTA Coarse Detection: The Short-Term Averaging Time (STA / LTA) algorithm is applied to all sensor channels. The short window (STA) length is set to 0.002 seconds, the long window (LTA) length to 0.2 seconds, and the trigger threshold to 3.0. The coarse signal arrival time (SAT) of all sensor channels is calculated, which is the coarse lightning arrival time. The signal-to-noise ratio (SNR) of the signals from all sensor channels is calculated, and the sensor channel that meets the preset SNR and has the earliest coarse signal arrival time is selected as the reference channel.

[0034] GCC-PHAT Refinement: Using the reference channel signal as a benchmark, the Generalized Cross-Correlation Phase Transform (GCC-PHAT) algorithm is used to calculate the time difference of all other sensing channels relative to the reference channel. The maximum allowable delay window is set to 0.05 seconds.

[0035] TOA synthesis: The reference time is superimposed on the time differences of all other sensing channels relative to the reference channel to obtain the final observation arrival time vector containing all channels.

[0036] Function and Effect: Compared to traditional manual acquisition, which takes several hours to process data from 2000 channels, the automated algorithm in this embodiment can complete the extraction of all channels within seconds. The GCC-PHAT algorithm, through phase whitening processing, effectively resists multipath reverberation interference in urban environments, achieving millisecond-level acquisition accuracy.

[0037] 4. Physical model construction and 3D inversion (corresponding to steps S4 and S5) Atmospheric sound propagation model: This embodiment abandons the constant speed of sound linear propagation model and constructs a sound speed profile model that varies with altitude. The ground surface reference temperature is set to 25℃ (298.15K), and the sound speed varies with altitude according to the adiabatic lapse rate. When calculating the theoretical propagation time from the sound source to each channel of the optical cable, the integrated method along the path is used to correct for errors caused by sound ray bending.

[0038] Nonlinear least squares inversion: Construct the objective function: .

[0039] Solution process: The solution uses the scipy.optimize.least_squares solver in Python and employs the Trust Region Reflective (TRF) algorithm.

[0040] Initial values ​​and boundaries: The initial search values ​​for the vibration time are set based on the camera trigger time, and the spatial search range is set to the area surrounding the communication optical cable. meters and high altitude Rice area.

[0041] Output: After iterative convergence, the algorithm outputs the optimal estimated location of the lightning source. and precise earthquake timing .

[0042] Based on the obtained data, a visual reconstruction is performed to intuitively display the three-dimensional spatial location of the lightning sound source.

[0043] This embodiment utilizes a communication range of approximately 4 kilometers as a sensing array. Compared to traditional microphone arrays with apertures of a few meters, it significantly improves the directional capability and depth resolution for distant sound sources. On a busy highway section (the newly built Zhujiang Grand Bridge), by combining buried sensors with FK filtering, it successfully extracted thunder signals from strong traffic noise, demonstrating the robustness of this technology in complex urban noise environments.

[0044] The entire process, from data reading, filtering, picking to inversion, is fully automated without human intervention, verifying the feasibility of applying this technology to a real-time lightning monitoring network. This embodiment verifies that the DAS-based lightning source localization technology is not only theoretically sound, but can also utilize existing urban communication fiber optic networks to achieve low-cost and high-precision fine three-dimensional imaging of lightning activity over urban areas.

[0045] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A three-dimensional localization method for a lightning sound source, characterized in that, Includes the following steps: S1: The original signals along the communication optical cable are collected in real time by a distributed acoustic wave sensing interrogation unit connected to one end of the communication optical cable. S2: The original signal is preprocessed by filtering to remove noise, resulting in a filtered signal; S3: Based on the filtered signal, a hybrid strategy is used to obtain the signal arrival time of different sensing channels, wherein the sensing channel is a virtual sensing channel that discretizes the communication optical cable. S4: Construct an atmospheric stratification sound speed propagation model to correct the propagation time error caused by atmospheric refraction and sound speed changes in the signal arrival time, and obtain the theoretical arrival time of the signal; S5: Based on the arrival time of the signal and the theoretical arrival time of the signal, a preset algorithm is used to perform three-dimensional spatial inversion to obtain the spatiotemporal parameters of the lightning source; S6: Based on the spatiotemporal parameters of the lightning sound source and the distribution of the communication optical cable, perform a visual reconstruction of the lightning sound source in a three-dimensional coordinate system.

2. The three-dimensional localization method for a lightning sound source according to claim 1, characterized in that, The raw signal mentioned in step S1 includes raw acoustic wave field data of thunder signals and environmental noise.

3. The three-dimensional localization method for a lightning sound source according to claim 1, characterized in that, The specific process of noise reduction preprocessing by filtering described in step S2 is as follows: First, the original signal is filtered by a bandpass filter to remove high-frequency electronic noise and extremely low-frequency temperature drift. Second, a visual velocity cutoff threshold is set, and a frequency domain-wavenumber domain filter is used to remove ground noise with visual velocity below the cutoff threshold to obtain a filtered signal.

4. The three-dimensional localization method for a lightning sound source according to claim 3, characterized in that, The bandpass filter includes a Butterworth bandpass filter.

5. The three-dimensional localization method for a lightning sound source according to claim 1, characterized in that, Step S3, which describes the specific steps for obtaining the signal arrival times of different sensing channels using a hybrid strategy, includes: S301: Apply the short-time-long-time average ratio algorithm to all sensor channels to calculate the approximate signal arrival time for each sensor channel; S302: Calculate the signal-to-noise ratio of the signal in each sensing channel, take the sensing channel that meets the preset signal-to-noise ratio and has the earliest coarse signal arrival time as the reference channel, and take the coarse signal arrival time of the reference channel as the absolute time reference. S303: Using a generalized cross-correlation phase transform algorithm, the signal of the i-th sensing channel is cross-correlated with the signal of the reference channel. By searching for the peak value of the cross-correlation function within the maximum allowable physical delay window, the time difference between the i-th sensing channel and the reference channel is accurately calculated. ; S304: The absolute time reference is superimposed with the time difference to obtain the signal arrival time of the i-th sensing channel.

6. The three-dimensional localization method for a lightning sound source according to claim 5, characterized in that, The specific details of the short-time-long-time average ratio algorithm are as follows: Set a short-time window and a long-time window. When the ratio of the energy of the short-time window to the energy of the long-time window exceeds a preset threshold, the time of the short-time window is determined to be the approximate signal arrival time.

7. The three-dimensional localization method for a lightning sound source according to claim 1, characterized in that, The atmospheric stratified sound speed propagation model described in step S4 includes a sound speed profile model, and the theoretical arrival time of the signal is calculated using a stratified integration method along the propagation path.

8. The three-dimensional localization method for a lightning sound source according to claim 1, characterized in that, The specific process of the three-dimensional spatial inversion described in step S5 is as follows: A four-dimensional unknown parameter set containing the time of thunder occurrence and three-dimensional spatial coordinates is constructed. A residual objective function is constructed based on the signal arrival time and the theoretical signal arrival time. The parameter search boundary is set, and a preset algorithm is used to solve the unknown parameter set until the residual objective function converges to the minimum value. The solution of the unknown parameter set at this time is used as the spatiotemporal parameters of the lightning source.

9. A three-dimensional localization method for a lightning sound source according to claim 8, characterized in that, The preset algorithms include the trust region reflection algorithm and the Levenberg-Marquardt algorithm.

10. A three-dimensional positioning system for a lightning sound source, characterized in that, It includes a memory and a processor; the memory is used to store instructions; the processor is used to operate according to the instructions to perform the method according to any one of claims 1-9.

Images