Method and system for blocking, shielding and adaptively identifying mobile phone signals

Abstract

The present invention provides a method and system for blocking, shielding and adaptively identifying mobile phone signals in areas where communication control and signal blocking are required, and relates to the technical field of mobile phone signal identification and shielding, including collecting signals through distributed signal collectors, obtaining signal characteristics and positions through time-frequency domain analysis and triangulation positioning, and inputting convolutional neural networks to extract communication protocols and behavior pattern characteristics, and predicting signal movement trajectories and communication behaviors. A multi-objective optimization algorithm is used to iteratively optimize interference power, and the beam direction and phase delay are calculated according to the optimal power configuration parameters and signal movement trajectory prediction data to generate a blocking control signal. Digital pre-distortion technology is used to compensate for directional interference beams, and the interference effect is monitored in real time. The evaluation data is fed back to the convolutional neural network to update the signal identification data, forming a closed-loop control, thereby achieving accurate and efficient mobile phone signal blocking, reducing interference power, and improving anti-interference capabilities.

Description

Technical Field

[0001] The present invention relates to a mobile phone signal identification and shielding technology, and in particular to a method and system for blocking, shielding and adaptively identifying mobile phone signals in an area where communication control and signal blocking are required. Background Art

[0002] With the development of mobile communication technology, the demand for the application of mobile communication control and signal blocking technology in specific places is growing. The existing mobile phone signal blocking technology mainly adopts the method of full-power interference in a fixed frequency band. This method has problems such as high power consumption, uncontrollable interference range, and easy adjacent frequency interference. In addition, the existing technology cannot adaptively identify and accurately block different communication formats and protocol characteristics, and lacks the ability to intelligently perceive the location and behavior characteristics of mobile terminals. At the same time, due to the complexity of the wireless channel environment and the nonlinear characteristics of the interference equipment itself, the interference effect of traditional blocking equipment is difficult to guarantee. The existing technology also has problems such as unreasonable interference power configuration and low energy utilization efficiency. In addition, there is a lack of real-time evaluation and feedback mechanism for the interference effect, and it is impossible to achieve dynamic optimization and closed-loop control of interference parameters.

[0003] Therefore, there is an urgent need for a method for blocking, shielding and adaptively identifying mobile phone signals in areas where communication control and signal blocking are required, which can realize adaptive signal feature recognition, precise directional blocking, nonlinear compensation and closed-loop control. Summary of the invention

[0004] The embodiment of the present invention provides a method and system for blocking, shielding and adaptively identifying mobile phone signals in an area where communication control and signal blocking are required, which can solve the problems in the prior art.

[0005] According to a first aspect of the embodiments of the present invention,

[0006] Provided is a method for blocking, shielding and adaptively identifying mobile phone signals in an area where communication control and signal blocking are required, comprising:

[0007] Collect electromagnetic signals in the target area through a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, construct a target signal characteristic vector with the first signal characteristic parameters and the spatial position coordinates, input the target signal characteristic vector into a preset convolutional neural network, extract communication protocol features and behavior pattern features in the target signal characteristic vector, and generate signal recognition data including signal movement trajectory prediction and communication behavior prediction;

[0008] According to the signal identification data, the minimum interference power threshold is calculated based on the channel capacity theory, the interference power is iteratively optimized using a multi-objective optimization algorithm, the optimization result of each iteration is compared with the minimum interference power threshold, and the iteration is stopped when the optimization result is less than the minimum interference power threshold, and the optimal power configuration parameters are obtained. The beam direction and phase delay are calculated according to the optimal power configuration parameters and the signal movement trajectory prediction data, and a blocking control signal is generated, wherein the blocking control signal includes a power parameter, a direction parameter, and a phase parameter;

[0009] A directional interference beam is generated according to the blocking control signal, and the power, beam direction and phase delay of the directional interference beam are adjusted. The digital pre-distortion technology is used to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam. The interference effect of the compensated directional interference beam in the target area is monitored in real time to obtain interference effect evaluation data. The interference effect evaluation data is input into the convolutional neural network to update the signal recognition data to obtain updated signal recognition data. The blocking control signal is recalculated based on the updated signal recognition data to form a closed-loop control of signal shielding.

[0010] In an optional embodiment,

[0011] Collect electromagnetic signals in the target area by using a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, and construct a target signal characteristic vector using the first signal characteristic parameters and the spatial position coordinates, including:

[0012] At least three signal collectors are set up in the target area to build a triangulation positioning network, each signal collector includes an antenna array, a radio frequency front-end circuit and a digital processing unit, the antenna array is used to collect electromagnetic signals, the radio frequency front-end circuit amplifies and mixes the electromagnetic signals, and the digital processing unit converts the processed electromagnetic signals into discrete signal sequences;

[0013] Performing bandpass filtering on the discrete signal sequence to obtain a filtered signal, performing time-frequency decomposition on the filtered signal using a multi-scale wavelet transform to obtain time-frequency decomposition data, and extracting a carrier frequency based on an energy distribution of the time-frequency decomposition data;

[0014] Performing cyclic spectrum analysis on the time-frequency decomposition data, calculating a cyclic autocorrelation function, performing Fourier transform on the cyclic autocorrelation function to obtain a spectral correlation function, identifying a modulation mode according to a peak characteristic of the spectral correlation function and an instantaneous phase characteristic of a signal, and constructing a first signal characteristic parameter using a carrier frequency and a modulation mode;

[0015] Collecting time difference data and signal strength data of the electromagnetic signal arriving at each signal collector, calculating a delay spread parameter based on the time difference data, and building a path loss model in combination with the signal strength data;

[0016] Calculate the signal-to-noise ratio of each signal collector according to the path loss model and determine the weight coefficient, establish a hyperbolic equation group for triangulation positioning based on the time difference data, substitute the weight coefficient into the weighted least square method to solve the hyperbolic equation group to obtain initial position data, correct the initial position data according to the path loss model, input the corrected initial position data into the extended Kalman filter, dynamically adjust the prediction noise and observation noise using an adaptive attenuation factor, and obtain the spatial position coordinates of the signal source through iterative calculation;

[0017] The first signal characteristic parameter is feature-fused with the spatial position coordinates to construct a target signal characteristic vector, where the target signal characteristic vector includes a carrier frequency, a modulation mode, a three-dimensional coordinate position and a moving speed.

[0018] In an optional embodiment,

[0019] Establishing a hyperbolic equation group for triangulation positioning based on the time difference data, substituting the weight coefficient into a weighted least square method to solve the hyperbolic equation group to obtain initial position data, and correcting the initial position data according to the path loss model includes:

[0020] Collecting time difference measurement data of electromagnetic signals arriving at multiple signal collectors, and preprocessing the time difference measurement data using a wavelet denoising method to obtain denoised time difference data;

[0021] Constructing a set of hyperbolic equations based on the denoised time difference data, the position coordinates of the signal source to be solved, the known position coordinates of the signal collector, and the signal propagation delay, wherein the set of hyperbolic equations represents the distance difference relationship from the signal source to each signal collector;

[0022] Performing Taylor series expansion on the hyperbolic equation group at the initial estimation point to obtain a linearized equation group, and combining the weight coefficient of the signal collector with the linearized equation group to construct a weighted least squares objective function;

[0023] Iteratively solving the weighted least squares objective function, updating the position estimate value in each iteration, stopping the iteration when the change in the position estimate values ​​between two adjacent iterations is less than a convergence threshold, and using the final position estimate value as the initial position data;

[0024] The initial position data is input into a propagation path loss model, and the theoretical receiving power at each signal collector is calculated based on an equivalent path loss index in the propagation path loss model, and the theoretical receiving power is compared with the actually measured received signal power to obtain a power difference value;

[0025] When the power difference value is greater than a preset threshold, the initial position data is taken as the search center point, and within the search range determined by the system time synchronization error, the initial position data is corrected by minimizing the power difference value, and the corrected position data is used as the final position coordinates of the signal source.

[0026] In an optional embodiment,

[0027] According to the signal identification data, the minimum interference power threshold is calculated based on the channel capacity theory, the interference power is iteratively optimized using a multi-objective optimization algorithm, the optimization result of each iteration is compared with the minimum interference power threshold, and the iteration is stopped when the optimization result is less than the minimum interference power threshold. The optimal power configuration parameters include:

[0028] Extracting the carrier frequency, modulation mode, communication protocol characteristics and predicted signal movement trajectory of each signal source from the signal identification data, determining the modulation order according to the modulation mode, and determining the signal bandwidth based on the communication protocol characteristics;

[0029] The actual transmission rate is calculated by multiplying the signal bandwidth by the modulation order, a channel capacity model is established according to Shannon's channel capacity theory, and the actual transmission rate is substituted into the channel capacity model to solve for a minimum signal-to-noise ratio;

[0030] Collecting noise power measurement data of the distributed signal collector and performing sliding average filtering to obtain system noise power, and multiplying the minimum signal-to-noise ratio by the system noise power to obtain a minimum interference power threshold;

[0031] Determine the transmission power limit range of each signal source according to the predicted signal movement trajectory, and generate an initial population including multiple power configuration schemes within the transmission power limit range, each power configuration scheme including power configuration parameters of multiple signal sources;

[0032] Based on the spatial position coordinates of the signal source, a multi-objective optimization function including a total interference power function, an inter-channel interference function and a spectrum utilization efficiency function is constructed, and the fitness value of each power configuration scheme in the initial population is calculated; the power configuration schemes are stratified using a non-dominated sorting algorithm, the congestion distance value is calculated and the parent power configuration scheme is selected, and the child power configuration scheme is generated by simulating a binary crossover operation;

[0033] Performing mutation processing on the child power configuration scheme to obtain a mutated power configuration scheme, calculating the interference power corresponding to the mutated power configuration scheme, comparing the interference power with the minimum interference power threshold, and stopping iteration when the interference power is less than the minimum interference power threshold;

[0034] The power configuration scheme at the end of the iteration is normalized to obtain a normalized power value, the normalized power value is weighted and summed to obtain a comprehensive score value, and the power configuration scheme with the highest comprehensive score value is selected as the final power configuration scheme, and finally the optimal power configuration parameters are obtained.

[0035] In an optional embodiment,

[0036] The calculation formula of the multi-objective optimization function including the total interference power function, the inter-channel interference function and the spectrum utilization efficiency function based on the spatial position coordinates of the signal source is as follows:

[0037]

[0038] Among them, F opt represents the multi-objective optimization function, λ1 is the interference power weight coefficient, N is the number of signal sources, and w ij represents the path loss coefficient of signal source i and signal source j, P i represents the transmission power of signal source i, P j represents the transmission power of signal source j, p i represents the position coordinates of the signal source i, p j represents the position coordinates of the signal source j, δ represents the regularization parameter, λ2 represents the inter-channel interference weight coefficient, and f max represents the upper limit of the spectrum range of the signal source operation, f min Indicates the lower limit of the spectrum range of the signal source operation, I i (f) represents the interference power spectral density of signal source i at frequency f, S i (f) represents the target signal power spectrum density of signal source i at frequency f, λ3 represents the spectrum utilization efficiency weight coefficient, B i represents the spectrum bandwidth occupied by signal source i, M i Indicates the order corresponding to the modulation method of signal source i.

[0039] In an optional embodiment,

[0040] Generating a directional interference beam according to the blocking control signal, adjusting the power, beam direction and phase delay of the directional interference beam, and using digital pre-distortion technology to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam includes:

[0041] Acquire the amplitude and phase information of the blocking control signal, calculate the Bessel function value according to the amplitude and phase information, multiply the Bessel function value by the harmonic suppression coefficient to obtain the harmonic suppression amount, and perform weighted combination of the harmonic suppression amount and the array element excitation value to obtain the array beam pattern;

[0042] The array beam pattern is divided into a near-field region and a far-field region, the beam power in the near-field region is attenuated by a power function to obtain a near-field power distribution, the beam power in the far-field region is attenuated by an exponential function to obtain a far-field power distribution, and the near-field power distribution and the far-field power distribution are synthesized to obtain an overall power distribution;

[0043] Obtaining a target direction angle of the beam, calculating a deviation between the current beam direction and the target direction, constructing a compensation function including a direction error and a sidelobe level according to the deviation, iteratively optimizing the compensation function using a gradient descent method to obtain a direction compensation amount, and combining the direction compensation amount with the overall power distribution to obtain an optimized beam;

[0044] Detecting spatial position information and timing information of the optimized beam, calculating a spatial delay according to the spatial position information, calculating a time-varying delay according to the timing information, converting the spatial delay and the time-varying delay into frequency domain compensation coefficients, and processing the frequency domain compensation coefficients using a group delay equalizer to obtain a compensation transfer function;

[0045] The input signal is filtered through the compensation transfer function to obtain a compensation signal sequence, the compensation signal sequence is time-delay extended to obtain an extended signal sequence, the extended signal sequence is combined with a nonlinear basis function to obtain a predistortion basis function, and the coefficients of the predistortion basis function are optimized by a variational inference algorithm to obtain an optimal predistortion coefficient;

[0046] The optimal predistortion coefficient is distributed to the parallel processing units according to the signal subband, each processing unit simultaneously performs predistortion calculation to obtain a compensation signal, detects the error between the compensation signal and the expected signal, updates the predistortion parameter in real time according to the error, and finally outputs the compensated directional interference beam signal.

[0047] In an optional embodiment,

[0048] The input signal is filtered through the compensation transfer function to obtain a compensation signal sequence, the compensation signal sequence is time-delay extended to obtain an extended signal sequence, the extended signal sequence is combined with a nonlinear basis function to obtain a predistortion basis function, and the coefficients of the predistortion basis function are optimized using a variational inference algorithm to obtain the optimal predistortion coefficients, including:

[0049] Performing a fast Fourier transform on the input signal to obtain a frequency domain signal, multiplying the frequency domain signal by a compensation transfer function to obtain a compensated frequency domain signal, and performing an inverse fast Fourier transform on the compensated frequency domain signal to obtain a compensated signal sequence;

[0050] Determine the delay quantity based on the preset memory depth, determine the minimum interval between adjacent delay quantities according to the system coherence time, perform multi-stage delay processing on the compensation signal sequence according to the minimum interval to obtain multiple delay signals, and splice the multiple delay signals to form a first delay matrix;

[0051] Extracting amplitude information and phase information of the compensation signal sequence, constructing an odd-power nonlinear basis function according to the amplitude information and the phase information, wherein the amplitude of the odd-power nonlinear basis function is determined by the odd power of the amplitude information, and the phase is determined by superposition of the phase information and the phase rotation angle;

[0052] Combining the first time delay matrix with the odd-power nonlinear basis function to obtain a predistortion basis function matrix;

[0053] A complex normal distribution function including a mean parameter and a covariance parameter is constructed as a variational posterior distribution function, and an optimization objective function is constructed according to the variational posterior distribution function and the predistortion basis function matrix;

[0054] Deriving the optimization objective function with respect to a mean parameter and a covariance parameter respectively to obtain a first gradient and a second gradient, and combining the first gradient and the second gradient to obtain a total gradient;

[0055] The mean parameter and the covariance parameter are iteratively updated according to the total gradient, and the iteration is stopped when the modulus value of the total gradient is less than a preset threshold, so as to obtain an optimal predistortion coefficient.

[0056] According to a second aspect of the embodiments of the present invention,

[0057] Provided is a mobile phone signal blocking shielding and adaptive identification system in an area where communication control and signal blocking are required, comprising:

[0058] The first unit is used to collect electromagnetic signals in a target area through a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, construct a target signal characteristic vector with the first signal characteristic parameters and the spatial position coordinates, input the target signal characteristic vector into a preset convolutional neural network, extract communication protocol features and behavior pattern features in the target signal characteristic vector, and generate signal recognition data including signal movement trajectory prediction and communication behavior prediction;

[0059] The second unit is used to calculate the minimum interference power threshold according to the signal identification data based on the channel capacity theory, iteratively optimize the interference power using a multi-objective optimization algorithm, compare the optimization result of each iteration with the minimum interference power threshold, stop the iteration when the optimization result is less than the minimum interference power threshold, obtain the optimal power configuration parameter, calculate the beam direction and phase delay according to the optimal power configuration parameter and the signal movement trajectory prediction data, and generate a blocking control signal, wherein the blocking control signal includes a power parameter, a direction parameter and a phase parameter;

[0060] The third unit is used to generate a directional interference beam according to the blocking control signal, adjust the power, beam direction and phase delay of the directional interference beam, use digital pre-distortion technology to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam, monitor the interference effect of the compensated directional interference beam in the target area in real time to obtain interference effect evaluation data, input the interference effect evaluation data into the convolutional neural network to update the signal recognition data to obtain updated signal recognition data, and recalculate the blocking control signal based on the updated signal recognition data to form a closed-loop control of signal shielding.

[0061] According to a third aspect of the embodiments of the present invention,

[0062] An electronic device is provided, comprising:

[0063] processor;

[0064] a memory for storing processor-executable instructions;

[0065] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0066] A fourth aspect of the embodiments of the present invention is:

[0067] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0068] In this embodiment, by collecting electromagnetic signals, analyzing time-frequency domain features, and using triangulation and convolutional neural networks, the communication protocol, behavior pattern, and movement trajectory of the target signal can be accurately identified, thereby achieving accurate positioning and identification of specific signals and avoiding interference with other legitimate signals. Based on channel capacity theory and multi-objective optimization algorithm, this method can calculate the minimum interference power threshold, iteratively optimize the interference power, and finally obtain the optimal power configuration parameters. At the same time, the system will adjust the interference strategy in real time according to the interference effect evaluation data to form a closed-loop control, thereby achieving adaptive interference to the target signal, improving interference efficiency, and reducing energy consumption. The use of digital pre-distortion technology to compensate for nonlinear distortion of directional interference beams can effectively suppress the nonlinear distortion of interference signals, improve the quality of interference signals, and enhance the interference effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A flow chart of a method for blocking, shielding and adaptively identifying mobile phone signals in an area where communication control and signal blocking are required according to an embodiment of the present invention;

[0070] Figure 2 The present invention is a schematic structural diagram of a mobile phone signal blocking, shielding and adaptive identification system in an area where communication control and signal blocking are required. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0073] Figure 1 FIG. 1 is a flow chart of a method for blocking and adaptively identifying mobile phone signals in an area where communication control and signal blocking are required according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0074] S101. Collect electromagnetic signals in the target area through a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, construct a target signal feature vector with the first signal characteristic parameters and the spatial position coordinates, input the target signal feature vector into a preset convolutional neural network, extract communication protocol features and behavior pattern features in the target signal feature vector, and generate signal recognition data including signal movement trajectory prediction and communication behavior prediction;

[0075] S102. According to the signal identification data, the minimum interference power threshold is calculated based on the channel capacity theory, and the interference power is iteratively optimized using a multi-objective optimization algorithm. The optimization result of each iteration is compared with the minimum interference power threshold. When the optimization result is less than the minimum interference power threshold, the iteration is stopped to obtain the optimal power configuration parameters, and the beam direction and phase delay are calculated according to the optimal power configuration parameters and the signal movement trajectory prediction data, and a blocking control signal is generated, wherein the blocking control signal includes a power parameter, a direction parameter and a phase parameter;

[0076] S103. Generate a directional interference beam according to the blocking control signal, adjust the power, beam direction and phase delay of the directional interference beam, use digital pre-distortion technology to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam, monitor the interference effect of the compensated directional interference beam in the target area in real time to obtain interference effect evaluation data, input the interference effect evaluation data into the convolutional neural network to update the signal recognition data to obtain updated signal recognition data, recalculate the blocking control signal based on the updated signal recognition data, and form a closed-loop control of signal shielding.

[0077] In an optional implementation, electromagnetic signals in a target area are collected by a distributed signal collector, and time-frequency domain analysis is performed on the electromagnetic signals to obtain first signal characteristic parameters, where the first signal characteristic parameters include carrier frequency and modulation mode, and the spatial position coordinates of the signal source are calculated based on a triangulation positioning algorithm. The target signal characteristic vector is constructed by using the first signal characteristic parameters and the spatial position coordinates, including:

[0078] At least three signal collectors are set up in the target area to build a triangulation positioning network, each signal collector includes an antenna array, a radio frequency front-end circuit and a digital processing unit, the antenna array is used to collect electromagnetic signals, the radio frequency front-end circuit amplifies and mixes the electromagnetic signals, and the digital processing unit converts the processed electromagnetic signals into discrete signal sequences;

[0079] Performing bandpass filtering on the discrete signal sequence to obtain a filtered signal, performing time-frequency decomposition on the filtered signal using a multi-scale wavelet transform to obtain time-frequency decomposition data, and extracting a carrier frequency based on an energy distribution of the time-frequency decomposition data;

[0080] Performing cyclic spectrum analysis on the time-frequency decomposition data, calculating a cyclic autocorrelation function, performing Fourier transform on the cyclic autocorrelation function to obtain a spectral correlation function, identifying a modulation mode according to a peak characteristic of the spectral correlation function and an instantaneous phase characteristic of a signal, and constructing a first signal characteristic parameter using a carrier frequency and a modulation mode;

[0081] Collecting time difference data and signal strength data of the electromagnetic signal arriving at each signal collector, calculating a delay spread parameter based on the time difference data, and building a path loss model in combination with the signal strength data;

[0082] Calculate the signal-to-noise ratio of each signal collector according to the path loss model and determine the weight coefficient, establish a hyperbolic equation group for triangulation positioning based on the time difference data, substitute the weight coefficient into the weighted least square method to solve the hyperbolic equation group to obtain initial position data, correct the initial position data according to the path loss model, input the corrected initial position data into the extended Kalman filter, dynamically adjust the prediction noise and observation noise using an adaptive attenuation factor, and obtain the spatial position coordinates of the signal source through iterative calculation;

[0083] The first signal characteristic parameter is feature-fused with the spatial position coordinates to construct a target signal characteristic vector, where the target signal characteristic vector includes a carrier frequency, a modulation mode, a three-dimensional coordinate position and a moving speed.

[0084] Exemplarily, first, at least three signal collectors are reasonably arranged in the target area to form a triangulation positioning network. Each signal collector includes an antenna array, a radio frequency front-end circuit, and a digital processing unit. The antenna array is responsible for collecting electromagnetic signals in the target area. The radio frequency front-end circuit amplifies and mixes the collected electromagnetic signals, and converts the signal frequency to a range that is easier to process. The digital processing unit converts the analog signal processed by the radio frequency front end into a discrete digital signal sequence for subsequent digital signal processing. For example, a network consisting of four evenly distributed signal collectors can be used, each collector is equipped with an eight-element linear array antenna, a low-noise amplifier, and a high-speed analog-to-digital converter.

[0085] Next, the collected discrete digital signal sequence is bandpass filtered to remove noise and interference signals outside the target frequency band. This can be achieved using digital filters such as finite impulse response (FIR) or infinite impulse response (IIR) filters. The parameters of the filter are set according to the frequency band of the target signal. For example, if the frequency band of the target signal is 1 GHz to 1.5 GHz, a bandpass filter with a center frequency of 1.25 GHz and a bandwidth of 500 MHz can be designed.

[0086] Then, the filtered signal is subjected to time-frequency analysis. The filtered signal is decomposed using multi-scale wavelet transform to obtain time-frequency decomposition data at different scales. By analyzing the distribution of energy in the time-frequency decomposition data, the carrier frequency of the signal can be extracted. For example, it can be observed that the wavelet coefficient energy near 1.3 GHz is the strongest, so it can be determined that the carrier frequency of the target signal is 1.3 GHz.

[0087] At the same time, the time-frequency decomposition data is subjected to cyclic spectrum analysis to identify the modulation mode of the signal. First, the cyclic autocorrelation function of the signal is calculated, and then the cyclic autocorrelation function is Fourier transformed to obtain the spectral correlation function. By analyzing the peak characteristics of the spectral correlation function and the instantaneous phase characteristics of the signal, the modulation mode of the signal can be identified, for example, amplitude shift keying (ASK), frequency shift keying (FSK), phase shift keying (PSK), etc. Assuming that the spectral correlation function has a peak at a specific frequency and the instantaneous phase of the signal shows a regular jump, it can be determined that the signal uses the FSK modulation mode. The extracted carrier frequency and the identified modulation mode are combined to construct the first signal characteristic parameter.

[0088] In order to determine the spatial position of the signal source, it is necessary to collect the time difference data and signal strength data of the electromagnetic signal arriving at each signal collector. Using the arrival time difference data, the delay spread parameter of the signal can be calculated. Combined with the signal strength data, a path loss model can be constructed, which describes the attenuation law of signal strength with propagation distance. For example, a free space propagation model or a logarithmic distance path loss model can be used. Assuming that the measured signal strength data is inversely proportional to the square of the distance, it can be determined that the free space propagation model is used.

[0089] According to the path loss model, the signal-to-noise ratio of each signal collector can be calculated, and the weight coefficient of each signal collector in the positioning process can be determined based on this. The higher the signal-to-noise ratio of the signal collector, the greater its weight coefficient. At the same time, based on the time difference data, a hyperbolic equation group for triangulation positioning can be established. Substitute the calculated weight coefficient into the weighted least squares method, solve the hyperbolic equation group, and obtain the initial position data of the signal source. Then, the initial position data is corrected according to the path loss model to reduce the impact of environmental factors on positioning accuracy.

[0090] The corrected initial position data is input into the extended Kalman filter, and the prediction noise and observation noise are dynamically adjusted using an adaptive attenuation factor. Through iterative calculation, the precise spatial position coordinates of the signal source are finally obtained, including three-dimensional coordinates and moving speed. For example, the output result after Kalman filtering is: the target is located at 120 degrees east longitude, 30 degrees north latitude, 100 meters above sea level, and the speed is 10 meters / second.

[0091] Finally, the first signal characteristic parameters (carrier frequency and modulation mode) are fused with the spatial position coordinates (three-dimensional coordinates and moving speed) of the signal source to construct the target signal feature vector.

[0092] In this embodiment, by constructing a path loss model and using weighted least squares and extended Kalman filtering, the impact of environmental factors and noise on positioning accuracy can be effectively reduced, and positioning accuracy can be improved. Through signal processing technologies such as multi-scale wavelet transform and cyclic spectrum analysis, the characteristic parameters of the target signal can be effectively extracted from noise and interference, and the anti-interference ability of the system can be enhanced. By extracting multi-dimensional characteristic information such as carrier frequency, modulation mode, spatial position and moving speed of the target signal, a more comprehensive target signal feature vector can be constructed to provide richer information for subsequent target recognition and tracking applications.

[0093] In an optional implementation, establishing a hyperbolic equation group for triangulation positioning based on the time difference data, substituting the weight coefficient into a weighted least squares method to solve the hyperbolic equation group to obtain initial position data, and correcting the initial position data according to the path loss model includes:

[0094] Collecting time difference measurement data of electromagnetic signals arriving at multiple signal collectors, and preprocessing the time difference measurement data using a wavelet denoising method to obtain denoised time difference data;

[0095] Constructing a set of hyperbolic equations based on the denoised time difference data, the position coordinates of the signal source to be solved, the known position coordinates of the signal collector, and the signal propagation delay, wherein the set of hyperbolic equations represents the distance difference relationship from the signal source to each signal collector;

[0096] Performing Taylor series expansion on the hyperbolic equation group at the initial estimation point to obtain a linearized equation group, and combining the weight coefficient of the signal collector with the linearized equation group to construct a weighted least squares objective function;

[0097] Iteratively solving the weighted least squares objective function, updating the position estimate value in each iteration, stopping the iteration when the change in the position estimate values ​​between two adjacent iterations is less than a convergence threshold, and using the final position estimate value as the initial position data;

[0098] The initial position data is input into a propagation path loss model, and the theoretical receiving power at each signal collector is calculated based on an equivalent path loss index in the propagation path loss model, and the theoretical receiving power is compared with the actually measured received signal power to obtain a power difference value;

[0099] When the power difference value is greater than a preset threshold, the initial position data is taken as the search center point, and within the search range determined by the system time synchronization error, the initial position data is corrected by minimizing the power difference value, and the corrected position data is used as the final position coordinates of the signal source.

[0100] Exemplarily, first, the collected time difference measurement data is preprocessed. Specifically, the wavelet denoising method is used to remove the noise interference in the time difference data. For example, the db4 wavelet basis is selected, a 4-layer decomposition is performed, and then the detail coefficient is threshold processed, and finally the signal is reconstructed to obtain the denoised time difference data. Assuming that the time difference data collected from the three signal collectors are 1.2 microseconds, 1.5 microseconds and 0.8 microseconds respectively, after wavelet denoising, the obtained time difference data are 1.22 microseconds, 1.48 microseconds and 0.79 microseconds respectively.

[0101] Next, a set of hyperbolic equations is established based on the denoised time difference data. Given the coordinates of the signal collector and the speed of signal propagation in the air (approximately the speed of light), the distance difference between the signal source and different signal collectors can be calculated based on the time difference. Assume that the coordinates of the signal source are (x, y), and the coordinates of the three signal collectors are (x1, y1), (x2, y2), and (x3, y3). Then an equation like "the distance from the signal source to the first collector minus the distance from the signal source to the second collector equals the time difference between the first collector and the second collector multiplied by the speed of light" can be established. By analogy, three hyperbolic equations consisting of three collectors can be obtained.

[0102] In order to solve the hyperbolic equations, the nonlinear equations need to be linearized. Select an initial estimation point, perform Taylor series expansion on the hyperbolic equations near this point, ignore the higher-order terms, and obtain the linearized equations. Considering that different signal collectors may have different effects on positioning accuracy, weight coefficients are introduced. For example, collectors closer to the signal source can be given a larger weight. Combine the weight coefficients with the linearized equations to construct a weighted least squares objective function.

[0103] Then, an iterative method is used to solve the weighted least squares objective function. Each iteration updates the position estimate and calculates the change in the position estimate between two adjacent iterations. When the change is less than the preset convergence threshold, the iteration is stopped and the final position estimate is used as the initial position data. For example, the convergence threshold is set to 0.1 meters. After multiple iterations, the initial position data is (100, 200).

[0104] Finally, the propagation path loss model is used to correct the initial position data. According to the known signal collector position and initial position data, combined with the equivalent path loss index, the theoretical received power at each signal collector is calculated. The theoretical received power is compared with the actual measured received signal power to calculate the power difference value. Assuming that the preset threshold is 1dBm, if the power difference value is greater than the threshold, the search is performed within the search range determined by the system time synchronization error with the initial position data as the center. The goal of the search is to minimize the power difference value. For example, if the system time synchronization error is 100 nanoseconds and the corresponding distance error is about 30 meters, the search is performed within a radius of 30 meters with the initial position (100, 200) as the center. After the search is completed, the position with the smallest power difference value is used as the final position coordinate of the signal source.

[0105] In this embodiment, the influence of noise and environmental factors on positioning accuracy is effectively reduced through wavelet denoising and path loss model correction, achieving higher positioning accuracy. The weighted least squares method and iterative solution strategy are adopted to improve the stability and robustness of the positioning algorithm, and reliable positioning results can be obtained even in the presence of certain errors. This method is suitable for various complex electromagnetic environments and can be widely used in indoor and outdoor positioning, target tracking and other fields.

[0106] In an optional implementation, according to the signal identification data, the minimum interference power threshold is calculated based on the channel capacity theory, the interference power is iteratively optimized using a multi-objective optimization algorithm, the optimization result of each iteration is compared with the minimum interference power threshold, and the iteration is stopped when the optimization result is less than the minimum interference power threshold, and the optimal power configuration parameters are obtained, including:

[0107] Extracting the carrier frequency, modulation mode, communication protocol characteristics and predicted signal movement trajectory of each signal source from the signal identification data, determining the modulation order according to the modulation mode, and determining the signal bandwidth based on the communication protocol characteristics;

[0108] The actual transmission rate is calculated by multiplying the signal bandwidth by the modulation order, a channel capacity model is established according to Shannon's channel capacity theory, and the actual transmission rate is substituted into the channel capacity model to solve for a minimum signal-to-noise ratio;

[0109] Collecting noise power measurement data of the distributed signal collector and performing sliding average filtering to obtain system noise power, and multiplying the minimum signal-to-noise ratio by the system noise power to obtain a minimum interference power threshold;

[0110] Determine the transmission power limit range of each signal source according to the predicted signal movement trajectory, and generate an initial population including multiple power configuration schemes within the transmission power limit range, each power configuration scheme including power configuration parameters of multiple signal sources;

[0111] Based on the spatial position coordinates of the signal source, a multi-objective optimization function including a total interference power function, an inter-channel interference function and a spectrum utilization efficiency function is constructed, and the fitness value of each power configuration scheme in the initial population is calculated; the power configuration schemes are stratified using a non-dominated sorting algorithm, the congestion distance value is calculated and the parent power configuration scheme is selected, and the child power configuration scheme is generated by simulating a binary crossover operation;

[0112] Performing mutation processing on the child power configuration scheme to obtain a mutated power configuration scheme, calculating the interference power corresponding to the mutated power configuration scheme, comparing the interference power with the minimum interference power threshold, and stopping iteration when the interference power is less than the minimum interference power threshold;

[0113] The power configuration scheme at the end of the iteration is normalized to obtain a normalized power value, the normalized power value is weighted and summed to obtain a comprehensive score value, and the power configuration scheme with the highest comprehensive score value is selected as the final power configuration scheme, and finally the optimal power configuration parameters are obtained.

[0114] Exemplarily, first, extract key information from the signal identification data. This information includes the carrier frequency, modulation mode, communication protocol characteristics (WiFi, Bluetooth, Zigbee, etc.) of each signal source, and the predicted signal movement trajectory. The modulation order is determined according to the modulation mode of the signal, and the signal bandwidth is determined based on the communication protocol characteristics (the bandwidth of WiFi is 20MHz and the bandwidth of Bluetooth is 1MHz).

[0115] Next, calculate the actual transmission rate and the minimum signal-to-noise ratio. Multiply the signal bandwidth by the modulation order to get the actual transmission rate (the bandwidth is 20MHz, the modulation order is 2, and the actual transmission rate is 40Mbps). Then, establish a channel capacity model based on Shannon's channel capacity theory, substitute the actual transmission rate into the model, and solve for the minimum signal-to-noise ratio. For example, assuming the actual transmission rate is 40Mbps, the minimum signal-to-noise ratio calculated by the channel capacity model is 10dB.

[0116] Subsequently, the minimum interference power threshold is determined. The noise power measurement data of the distributed signal collector is collected. For example, the collected noise power data is -90dBm. The collected noise power data is subjected to sliding average filtering to reduce noise fluctuations. For example, the noise power data is filtered using a sliding average filter with a window size of 10 to obtain the system noise power, which is assumed to be -85dBm. The minimum signal-to-noise ratio (10dB) is converted to a linear value (10), and then multiplied by the system noise power (-85dBm) to obtain the minimum interference power threshold.

[0117] Then, perform iterative optimization of the power configuration. Determine the transmission power limit range of each signal source based on the predicted signal movement trajectory. For example, the power limit range of signal source 1 is 0dBm to 10dBm, and the power limit range of signal source 2 is -5dBm to 5dBm. Generate an initial population containing multiple power configuration schemes within the transmission power limit range, and each power configuration scheme contains power configuration parameters of multiple signal sources. For example, a power configuration scheme can be {signal source 1: 5dBm, signal source 2: 0dBm}.

[0118] A multi-objective optimization function is constructed based on the spatial position coordinates of the signal source, which includes the total interference power function, the inter-channel interference function, and the spectrum utilization efficiency function. The fitness value of each power configuration scheme in the initial population is calculated. For example, the total interference power, inter-channel interference, and spectrum utilization efficiency of each scheme are calculated, and these values ​​are combined into a fitness value. The power configuration schemes are stratified using a non-dominated sorting algorithm, and the congestion distance value is calculated to select the parent power configuration scheme. The child power configuration scheme is generated by simulating a binary crossover operation. For example, some power parameters of two parent schemes are exchanged to generate a new child scheme.

[0119] Perform mutation processing on the offspring power configuration scheme, for example, randomly change the power value of a signal source to obtain the mutated power configuration scheme. Calculate the interference power corresponding to the mutated scheme and compare the interference power with the minimum interference power threshold. For example, if the calculated interference power is -76dBm, which is less than the minimum interference power threshold of -75dBm, stop the iteration.

[0120] After the iteration is terminated, the final power configuration scheme is normalized to obtain the normalized power value. The normalized power values ​​are weighted and summed to obtain the comprehensive score value. For example, the normalized power value of each signal source is multiplied by different weights, and then summed to obtain the comprehensive score value. The power configuration scheme with the highest comprehensive score value is selected as the final power configuration scheme, and the optimal power configuration parameters are finally obtained. For example, the final scheme is {signal source 1: 4dBm, signal source 2: 2dBm}.

[0121] In this embodiment, by calculating the minimum interference power threshold based on the channel capacity theory and using a multi-objective optimization algorithm to iteratively optimize the interference power, the total interference power of the system can be effectively reduced and the overall performance of the system can be improved. By optimizing the power configuration scheme, the interference between channels can be minimized, the spectrum utilization efficiency can be improved, and more signal sources can be allowed to coexist in the same frequency band. By accurately controlling the interference power, the communication quality of each signal source can be ensured, the bit error rate can be reduced, and the reliability of data transmission can be improved.

[0122] In an optional implementation, a calculation formula for constructing a multi-objective optimization function including a total interference power function, an inter-channel interference function, and a spectrum utilization efficiency function based on the spatial position coordinates of the signal source is as follows:

[0123]

[0124] Among them, F opt represents the multi-objective optimization function, λ1 is the interference power weight coefficient, N is the number of signal sources, and w ij represents the path loss coefficient of signal source i and signal source j, P i represents the transmission power of signal source i, P j represents the transmission power of signal source j, p i represents the position coordinates of the signal source i, p j represents the position coordinates of the signal source j, δ represents the regularization parameter, λ2 represents the inter-channel interference weight coefficient, and f max represents the upper limit of the spectrum range of the signal source operation, f min Indicates the lower limit of the spectrum range of the signal source operation, I i (f) represents the interference power spectral density of signal source i at frequency f, S i (f) represents the target signal power spectrum density of signal source i at frequency f, λ3 represents the spectrum utilization efficiency weight coefficient, B i represents the spectrum bandwidth occupied by signal source i, M i Indicates the order corresponding to the modulation method of signal source i.

[0125] Exemplarily, first, obtain the location coordinate information and other related parameters of all signal sources. These parameters include the transmission power of each signal source, the path loss between signal sources, the operating frequency band of each signal source, the interference power spectrum density, the target signal power spectrum density, the signal bandwidth, and the order corresponding to the modulation method. For example, suppose there are three signal sources, their location coordinates are (10, 20), (30, 40) and (50, 60), and the transmission powers are 10mW, 20mW and 30mW respectively. The path loss coefficient can be expressed as the inverse square of the distance between the signal sources. Assuming that the distance between signal source 1 and signal source 2 is 22.36, the path loss coefficient between them is approximately 0.002. Other parameters can be set according to actual conditions.

[0126] Then, based on the acquired parameters, calculate the total interference power, inter-channel interference, and spectrum utilization efficiency. The total interference power is calculated by adding the interference power between all signal sources. The inter-channel interference is calculated by adding the interference power generated by each signal source to other signal sources. The spectrum utilization efficiency is calculated by dividing the target signal power of each signal source by the spectrum bandwidth it occupies, and then multiplying it by the order corresponding to its modulation method. For example, assuming that the interference power spectrum density of signal source 1 is 1mW / MHz, the target signal power spectrum density of signal source 2 is 10mW / MHz, the signal bandwidth is 10MHz, and the order corresponding to the modulation method is 4, then the interference power of signal source 1 to signal source 2 is 1mW / MHz*10MHz=10mW, and the spectrum utilization efficiency of signal source 2 is (10mW / MHz*10MHz)*4 / 10MHz=40.

[0127] Next, construct a multi-objective optimization function. The function contains three parts: total interference power, inter-channel interference, and spectrum utilization efficiency. Each part is multiplied by a weight coefficient to reflect its importance. For example, if you are more concerned about reducing the total interference power, you can set the weight coefficient of the total interference power to be larger. Assume that the weight coefficient of the total interference power is 0.5, the weight coefficient of the inter-channel interference is 0.3, and the weight coefficient of the spectrum utilization efficiency is 0.2. Then the value of the multi-objective optimization function is 0.5*(total interference power)+0.3*(inter-channel interference)-0.2*(spectrum utilization efficiency). Note that there is a negative sign in front of the spectrum utilization efficiency because the goal is to maximize the spectrum utilization efficiency, that is, to minimize the negative spectrum utilization efficiency.

[0128] Finally, the position coordinates of the signal source are adjusted by an optimization algorithm, such as a gradient descent algorithm, to minimize the value of the multi-objective optimization function. For example, the position coordinates of each signal source can be gradually adjusted, and the new value of the multi-objective optimization function can be calculated. If the new value is less than the old value, the new position coordinate is accepted; otherwise, the new position coordinate is rejected. This process is repeated until the value of the multi-objective optimization function converges to a minimum value.

[0129] In this embodiment, by optimizing the spatial position of the signal source, the mutual interference between the signal sources can be effectively reduced, thereby reducing the total interference power of the system and improving the overall performance of the system. By optimizing the spatial position of the signal source, the interference between the signal sources can be better controlled, thereby improving the interference suppression capability between channels, reducing the crosstalk between different signal sources, and improving the communication quality. By optimizing the spatial position of the signal source, the spectrum resources can be more effectively utilized, thereby improving the spectrum utilization efficiency of the system, supporting more users to access, and improving the system capacity.

[0130] In an optional implementation, a directional interference beam is generated according to the blocking control signal, the power, beam direction and phase delay of the directional interference beam are adjusted, and the nonlinear distortion compensation is performed on the adjusted directional interference beam using digital predistortion technology to obtain the compensated directional interference beam, including:

[0131] Acquire the amplitude and phase information of the blocking control signal, calculate the Bessel function value according to the amplitude and phase information, multiply the Bessel function value by the harmonic suppression coefficient to obtain the harmonic suppression amount, and perform weighted combination of the harmonic suppression amount and the array element excitation value to obtain the array beam pattern;

[0132] The array beam pattern is divided into a near-field region and a far-field region, the beam power in the near-field region is attenuated by a power function to obtain a near-field power distribution, the beam power in the far-field region is attenuated by an exponential function to obtain a far-field power distribution, and the near-field power distribution and the far-field power distribution are synthesized to obtain an overall power distribution;

[0133] Obtaining a target direction angle of the beam, calculating a deviation between the current beam direction and the target direction, constructing a compensation function including a direction error and a sidelobe level according to the deviation, iteratively optimizing the compensation function using a gradient descent method to obtain a direction compensation amount, and combining the direction compensation amount with the overall power distribution to obtain an optimized beam;

[0134] Detecting spatial position information and timing information of the optimized beam, calculating a spatial delay according to the spatial position information, calculating a time-varying delay according to the timing information, converting the spatial delay and the time-varying delay into frequency domain compensation coefficients, and processing the frequency domain compensation coefficients using a group delay equalizer to obtain a compensation transfer function;

[0135] The input signal is filtered through the compensation transfer function to obtain a compensation signal sequence, the compensation signal sequence is time-delay extended to obtain an extended signal sequence, the extended signal sequence is combined with a nonlinear basis function to obtain a predistortion basis function, and the coefficients of the predistortion basis function are optimized by a variational inference algorithm to obtain an optimal predistortion coefficient;

[0136] The optimal predistortion coefficient is distributed to the parallel processing units according to the signal subband, each processing unit simultaneously performs predistortion calculation to obtain a compensation signal, detects the error between the compensation signal and the expected signal, updates the predistortion parameter in real time according to the error, and finally outputs the compensated directional interference beam signal.

[0137] Exemplarily, first, a blocking control signal is received. The signal contains amplitude and phase information. For example, the amplitude is 0.8 and the phase is 30 degrees. Using this information, a pre-calculated Bessel function table is queried to find the corresponding Bessel function value. Assume that the value obtained from the table is 0.9. Then, this value is multiplied by the preset harmonic suppression coefficient to obtain the harmonic suppression amount. Assuming that the harmonic suppression coefficient is 0.5, the harmonic suppression amount is 0.9*0.5=0.45. Next, this harmonic suppression amount is weightedly combined with the excitation value of each antenna unit. Assuming that there are two antenna units with excitation values ​​of 1 and 0.5 respectively, the weighted excitation value is 1*0.45+0.5*0.45=0.675. Finally, based on these weighted excitation values, the array beam pattern is calculated.

[0138] Next, the generated array beam pattern is divided into the near field area and the far field area. Assuming 10 meters as the boundary, less than 10 meters is the near field area, and greater than 10 meters is the far field area. For the near field area, a power function is used for attenuation processing. For example, if the distance is 5 meters and the attenuation coefficient is 0.8, the near field power is power*5^0.8. For the far field area, an exponential function is used for attenuation processing. For example, if the distance is 20 meters and the attenuation coefficient is 0.2, the far field power is power*e^(-0.2*20). The near field power distribution and the far field power distribution are combined to obtain the overall power distribution.

[0139] Then, obtain the target direction angle of the beam. Assume that the target direction angle is 45 degrees. Calculate the deviation value between the current beam direction and the target direction. Assuming that the current beam direction is 40 degrees, the deviation value is 5 degrees. According to this deviation value, construct a compensation function including directional error and sidelobe level. The function can be a pre-designed lookup table or an empirical formula. For example, the compensation function can be defined as compensation amount = 0.1 * deviation value + 0.05 * sidelobe level. In this embodiment, the sidelobe level mainly refers to the attenuation degree of the intensity of the radiation intensity of the interference beam (directional interference beam) in the non-target direction compared with the target direction (i.e., the main lobe direction). Specifically, in a directional interference system, the goal is to interfere with the signal source through a directional interference beam. In order to ensure that the interference beam can accurately and effectively point to the target signal source and minimize interference to other areas, it is necessary to finely optimize the radiation pattern of the beam. The gradient descent method is used to iteratively optimize the compensation function to obtain the directional compensation amount. Assume that after optimization, the directional compensation amount is 0.3. Combining this compensation amount with the previously obtained overall power distribution, an optimized beam is obtained.

[0140] After that, the spatial position information and timing information of the optimized beam are detected. Assume that the spatial position is (10, 20, 30) and the time is 10ms. Calculate the spatial delay according to the spatial position information. For example, the delay can be calculated according to the distance, assuming that the delay is 0.01*distance. Calculate the time-varying delay according to the timing information. Assume that the delay is 0.001*time difference. Convert the spatial delay and the time-varying delay into frequency domain compensation coefficients. Use the group delay equalizer to process these coefficients to obtain the compensation transfer function.

[0141] Next, the input signal is filtered through the compensation transfer function to obtain a compensation signal sequence. The compensation signal sequence is time-delay extended to obtain an extended signal sequence. The extended signal sequence is combined with a nonlinear basis function (such as a polynomial function) to obtain a predistortion basis function. The coefficients of the predistortion basis function are optimized using a variational inference algorithm to obtain the optimal predistortion coefficients.

[0142] Finally, the optimal predistortion coefficients are distributed to the parallel processing units according to the signal subbands. Each processing unit performs predistortion calculations at the same time to obtain a compensated signal. The error between the compensated signal and the desired signal is detected. The predistortion parameters are updated in real time according to the error. Finally, the compensated directional interference beam signal is output.

[0143] In this embodiment, by calculating the Bessel function value and combining the harmonic suppression coefficient, the harmonic components in the interference signal are effectively suppressed, thereby improving the quality of the interference signal. By compensating and optimizing the beam direction, the direction of the interference beam can be accurately controlled to point to the target area and avoid interference to the non-target area. Through digital pre-distortion technology and variational inference algorithm, nonlinear distortion in the system can be compensated in real time to improve the stability and robustness of the interference signal.

[0144] In an optional implementation, filtering the input signal through the compensation transfer function to obtain a compensation signal sequence, delay extending the compensation signal sequence to obtain an extended signal sequence, combining the extended signal sequence with a nonlinear basis function to obtain a predistortion basis function, and optimizing the coefficients of the predistortion basis function using a variational inference algorithm to obtain the optimal predistortion coefficient includes:

[0145] Performing a fast Fourier transform on the input signal to obtain a frequency domain signal, multiplying the frequency domain signal by a compensation transfer function to obtain a compensated frequency domain signal, and performing an inverse fast Fourier transform on the compensated frequency domain signal to obtain a compensated signal sequence;

[0146] Determine the delay quantity based on the preset memory depth, determine the minimum interval between adjacent delay quantities according to the system coherence time, perform multi-stage delay processing on the compensation signal sequence according to the minimum interval to obtain multiple delay signals, and splice the multiple delay signals to form a first delay matrix;

[0147] Extracting amplitude information and phase information of the compensation signal sequence, constructing an odd-power nonlinear basis function according to the amplitude information and the phase information, wherein the amplitude of the odd-power nonlinear basis function is determined by the odd power of the amplitude information, and the phase is determined by superposition of the phase information and the phase rotation angle;

[0148] Combining the first time delay matrix with the odd-power nonlinear basis function to obtain a predistortion basis function matrix;

[0149] A complex normal distribution function including a mean parameter and a covariance parameter is constructed as a variational posterior distribution function, and an optimization objective function is constructed according to the variational posterior distribution function and the predistortion basis function matrix;

[0150] Deriving the optimization objective function with respect to a mean parameter and a covariance parameter respectively to obtain a first gradient and a second gradient, and combining the first gradient and the second gradient to obtain a total gradient;

[0151] The mean parameter and the covariance parameter are iteratively updated according to the total gradient, and the iteration is stopped when the modulus value of the total gradient is less than a preset threshold, so as to obtain an optimal predistortion coefficient.

[0152] Exemplarily, first, after receiving the input signal, the system filters the signal through the compensation transfer function. The compensation transfer function is obtained by the previous interference optimization process, which aims to correct various non-ideal conditions that the signal may encounter during the transmission process. After the signal is processed by the compensation transfer function, a compensation signal sequence is obtained. This sequence represents the form of the signal after preliminary compensation, but it is not the final compensation result.

[0153] Next, the compensation signal sequence is delayed extended in order to increase the time length of the signal and enhance the system's ability to capture signal features. In this process, we determine the number of delays based on the preset memory depth. The memory depth reflects the length of the signal's historical information, that is, the time span that the system needs to consider when calculating the delay. Then, based on the system's coherence time (i.e., the time scale for the signal to remain consistent), the minimum interval between adjacent delays is calculated. The minimum interval determines the granularity of the signal delay and also affects the accuracy of the system's delay processing. According to this interval, the compensation signal sequence is subjected to multi-level delay processing to obtain multiple delay signals. These delay signals are spliced ​​into a first delay matrix, which contains a collection of multiple delay signals and provides basic data for subsequent nonlinear pre-distortion.

[0154] Next, the amplitude information and phase information of the compensation signal sequence are extracted. Amplitude information refers to the intensity distribution of the signal, while phase information represents the phase change of the signal. Based on the amplitude information and phase information, an odd-power nonlinear basis function is constructed. This basis function determines the amplitude part by the odd power of the amplitude of the signal, and determines the phase part by superimposing the phase information and a phase rotation angle. This basis function can effectively capture the nonlinear characteristics of the signal, and then be used for subsequent pre-distortion processing.

[0155] Then, the first time delay matrix is ​​combined with the odd-power nonlinear basis function to obtain the predistortion basis function matrix. The function of this matrix is ​​to combine the time delay signal and the nonlinear characteristics to form a composite basis function set for further optimization of the variational inference algorithm.

[0156] In the next step, we need to construct a complex normal distribution function with mean and covariance parameters as the variational posterior distribution function. The complex normal distribution function is a statistical distribution that describes the probabilistic characteristics of complex domain signals. The mean parameter represents the expected value of the signal, while the covariance parameter describes the correlation between different parts of the signal. In variational inference, the posterior distribution function is used to estimate the unknown variables that may exist in the signal.

[0157] Based on the complex normal distribution function and the predistortion basis function matrix, an optimization objective function is constructed. The objective function is designed to minimize the distortion error of the signal and optimize the compensation effect of the signal by adjusting the parameters of the predistortion basis function. Specifically, the objective function calculates the error of the signal after processing and guides the subsequent optimization process based on the error value.

[0158] During the optimization process, the objective function is gradient-derived to calculate the gradients of the mean parameter and the covariance parameter. The first gradient represents the rate of change of the objective function to the mean parameter, and the second gradient represents the rate of change to the covariance parameter. After the two gradients are combined, a total gradient is obtained to guide the algorithm on how to adjust the parameters to achieve the optimal effect.

[0159] Then, based on the total gradient, the system iteratively updates the mean parameter and covariance parameter. Each update adjusts the parameter value according to the direction of the total gradient, thereby gradually approaching the optimal solution. When the modulus of the total gradient is less than the preset threshold, it means that the optimization has converged, and the iteration is stopped at this time to obtain the optimal predistortion coefficient.

[0160] Ultimately, the optimal pre-distortion coefficients obtained through the variational inference optimization process will be used to adjust and compensate for the characteristics of the directional interference beam, ensuring that the signal can accurately interfere with the target signal source during the interference process while minimizing interference to other areas.

[0161] In this embodiment, the accuracy and stability of the directional interference beam are effectively improved through precise signal delay expansion and nonlinear pre-distortion processing, so that it can point to the target signal source more accurately and reduce unintentional interference to other signals. The variational inference algorithm is used to optimize the pre-distortion coefficient, so that the system can adaptively adjust the parameters to further improve the effectiveness of the interference effect, ensure that the beam can be focused on the target area to the maximum extent, and avoid energy waste. Through the filtering and delay expansion of the compensation transfer function, the signal capture capability is enhanced, so that the system can better cope with signal changes in complex environments. The optimized interference beam not only improves the accuracy of the interference, but also effectively reduces the sidelobe radiation intensity, thereby reducing interference to non-target areas and improving the overall performance and safety of the system.

[0162] Figure 2 FIG. 1 is a schematic diagram of a structure of a mobile phone signal blocking shielding and adaptive identification system in an area where communication control and signal blocking are required in an embodiment of the present invention. Figure 2 As shown, the system comprises:

[0163] The first unit is used to collect electromagnetic signals in a target area through a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, construct a target signal characteristic vector with the first signal characteristic parameters and the spatial position coordinates, input the target signal characteristic vector into a preset convolutional neural network, extract communication protocol features and behavior pattern features in the target signal characteristic vector, and generate signal recognition data including signal movement trajectory prediction and communication behavior prediction;

[0164] The second unit is used to calculate the minimum interference power threshold according to the signal identification data based on the channel capacity theory, iteratively optimize the interference power using a multi-objective optimization algorithm, compare the optimization result of each iteration with the minimum interference power threshold, stop the iteration when the optimization result is less than the minimum interference power threshold, obtain the optimal power configuration parameter, calculate the beam direction and phase delay according to the optimal power configuration parameter and the signal movement trajectory prediction data, and generate a blocking control signal, wherein the blocking control signal includes a power parameter, a direction parameter and a phase parameter;

[0165] The third unit is used to generate a directional interference beam according to the blocking control signal, adjust the power, beam direction and phase delay of the directional interference beam, use digital pre-distortion technology to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam, monitor the interference effect of the compensated directional interference beam in the target area in real time to obtain interference effect evaluation data, input the interference effect evaluation data into the convolutional neural network to update the signal recognition data to obtain updated signal recognition data, and recalculate the blocking control signal based on the updated signal recognition data to form a closed-loop control of signal shielding.

[0166] According to a third aspect of the embodiments of the present invention,

[0167] An electronic device is provided, comprising:

[0168] processor;

[0169] a memory for storing processor-executable instructions;

[0170] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0171] A fourth aspect of the embodiments of the present invention is:

[0172] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0173] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0174] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for blocking, shielding and adaptively identifying mobile phone signals in areas where communication control and signal blocking are required, characterized in that: include: Collect electromagnetic signals in the target area through a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, construct a target signal characteristic vector with the first signal characteristic parameters and the spatial position coordinates, input the target signal characteristic vector into a preset convolutional neural network, extract communication protocol features and behavior pattern features in the target signal characteristic vector, and generate signal recognition data including signal movement trajectory prediction and communication behavior prediction; According to the signal identification data, the minimum interference power threshold is calculated based on the channel capacity theory, the interference power is iteratively optimized using a multi-objective optimization algorithm, the optimization result of each iteration is compared with the minimum interference power threshold, and the iteration is stopped when the optimization result is less than the minimum interference power threshold, and the optimal power configuration parameters are obtained. The beam direction and phase delay are calculated according to the optimal power configuration parameters and the signal movement trajectory prediction data, and a blocking control signal is generated, wherein the blocking control signal includes a power parameter, a direction parameter, and a phase parameter; A directional interference beam is generated according to the blocking control signal, and the power, beam direction and phase delay of the directional interference beam are adjusted. The digital pre-distortion technology is used to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam. The interference effect of the compensated directional interference beam in the target area is monitored in real time to obtain interference effect evaluation data. The interference effect evaluation data is input into the convolutional neural network to update the signal recognition data to obtain updated signal recognition data. The blocking control signal is recalculated based on the updated signal recognition data to form a closed-loop control of signal shielding.

2. The method according to claim 1, characterized in that Collect electromagnetic signals in the target area by using a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, and construct a target signal characteristic vector using the first signal characteristic parameters and the spatial position coordinates, including: At least three signal collectors are set up in the target area to build a triangulation positioning network, each signal collector includes an antenna array, a radio frequency front-end circuit and a digital processing unit, the antenna array is used to collect electromagnetic signals, the radio frequency front-end circuit amplifies and mixes the electromagnetic signals, and the digital processing unit converts the processed electromagnetic signals into discrete signal sequences; Performing bandpass filtering on the discrete signal sequence to obtain a filtered signal, performing time-frequency decomposition on the filtered signal using a multi-scale wavelet transform to obtain time-frequency decomposition data, and extracting a carrier frequency based on an energy distribution of the time-frequency decomposition data; Performing cyclic spectrum analysis on the time-frequency decomposition data, calculating a cyclic autocorrelation function, performing Fourier transform on the cyclic autocorrelation function to obtain a spectral correlation function, identifying a modulation mode according to a peak characteristic of the spectral correlation function and an instantaneous phase characteristic of a signal, and constructing a first signal characteristic parameter using a carrier frequency and a modulation mode; Collecting time difference data and signal strength data of the electromagnetic signal arriving at each signal collector, calculating a delay spread parameter based on the time difference data, and building a path loss model in combination with the signal strength data; Calculate the signal-to-noise ratio of each signal collector according to the path loss model and determine the weight coefficient, establish a hyperbolic equation group for triangulation positioning based on the time difference data, substitute the weight coefficient into the weighted least square method to solve the hyperbolic equation group to obtain initial position data, correct the initial position data according to the path loss model, input the corrected initial position data into the extended Kalman filter, dynamically adjust the prediction noise and observation noise using an adaptive attenuation factor, and obtain the spatial position coordinates of the signal source through iterative calculation; The first signal characteristic parameter is feature-fused with the spatial position coordinates to construct a target signal characteristic vector, where the target signal characteristic vector includes a carrier frequency, a modulation mode, a three-dimensional coordinate position and a moving speed.

3. The method according to claim 2, characterized in that Establishing a hyperbolic equation group for triangulation positioning based on the time difference data, substituting the weight coefficient into a weighted least square method to solve the hyperbolic equation group to obtain initial position data, and correcting the initial position data according to the path loss model includes: Collecting time difference measurement data of electromagnetic signals arriving at multiple signal collectors, and preprocessing the time difference measurement data using a wavelet denoising method to obtain denoised time difference data; Constructing a set of hyperbolic equations based on the denoised time difference data, the position coordinates of the signal source to be solved, the known position coordinates of the signal collector, and the signal propagation delay, wherein the set of hyperbolic equations represents the distance difference relationship from the signal source to each signal collector; Performing Taylor series expansion on the hyperbolic equation group at the initial estimation point to obtain a linearized equation group, and combining the weight coefficient of the signal collector with the linearized equation group to construct a weighted least squares objective function; Iteratively solving the weighted least squares objective function, updating the position estimate value in each iteration, stopping the iteration when the change in the position estimate values ​​between two adjacent iterations is less than a convergence threshold, and using the final position estimate value as the initial position data; The initial position data is input into a propagation path loss model, and the theoretical receiving power at each signal collector is calculated based on an equivalent path loss index in the propagation path loss model, and the theoretical receiving power is compared with the actually measured received signal power to obtain a power difference value; When the power difference value is greater than a preset threshold, the initial position data is taken as the search center point, and within the search range determined by the system time synchronization error, the initial position data is corrected by minimizing the power difference value, and the corrected position data is used as the final position coordinates of the signal source.

4. The method according to claim 1, characterized in that: According to the signal identification data, the minimum interference power threshold is calculated based on the channel capacity theory, and the interference power is iteratively optimized using a multi-objective optimization algorithm. The optimization result of each iteration is compared with the minimum interference power threshold. When the optimization result is less than the minimum interference power threshold, the iteration is stopped. The optimal power configuration parameters include: Extracting the carrier frequency, modulation mode, communication protocol characteristics and predicted signal movement trajectory of each signal source from the signal identification data, determining the modulation order according to the modulation mode, and determining the signal bandwidth based on the communication protocol characteristics; The actual transmission rate is calculated by multiplying the signal bandwidth by the modulation order, a channel capacity model is established according to Shannon's channel capacity theory, and the actual transmission rate is substituted into the channel capacity model to solve for a minimum signal-to-noise ratio; Collecting noise power measurement data of the distributed signal collector and performing sliding average filtering to obtain system noise power, and multiplying the minimum signal-to-noise ratio by the system noise power to obtain a minimum interference power threshold; Determine the transmission power limit range of each signal source according to the predicted signal movement trajectory, and generate an initial population including multiple power configuration schemes within the transmission power limit range, each power configuration scheme including power configuration parameters of multiple signal sources; Based on the spatial position coordinates of the signal source, a multi-objective optimization function including a total interference power function, an inter-channel interference function and a spectrum utilization efficiency function is constructed, and the fitness value of each power configuration scheme in the initial population is calculated; the power configuration schemes are stratified using a non-dominated sorting algorithm, the congestion distance value is calculated and the parent power configuration scheme is selected, and the child power configuration scheme is generated by simulating a binary crossover operation; Performing mutation processing on the child power configuration scheme to obtain a mutated power configuration scheme, calculating the interference power corresponding to the mutated power configuration scheme, comparing the interference power with the minimum interference power threshold, and stopping iteration when the interference power is less than the minimum interference power threshold; The power configuration scheme at the end of the iteration is normalized to obtain a normalized power value, the normalized power value is weighted and summed to obtain a comprehensive score value, and the power configuration scheme with the highest comprehensive score value is selected as the final power configuration scheme, and finally the optimal power configuration parameters are obtained.

5. The method according to claim 4, characterized in that The calculation formula of the multi-objective optimization function including the total interference power function, the inter-channel interference function and the spectrum utilization efficiency function based on the spatial position coordinates of the signal source is as follows: Among them, F opt represents the multi-objective optimization function, λ1 represents the interference power weight coefficient, N represents the number of signal sources, and w ij represents the path loss coefficient of signal source i and signal source j, P i represents the transmission power of signal source i, P j represents the transmission power of signal source j, p i represents the position coordinates of the signal source i, p j represents the position coordinates of the signal source j, δ represents the regularization parameter, λ2 represents the inter-channel interference weight coefficient, and f max represents the upper limit of the spectrum range of the signal source operation, f min Indicates the lower limit of the spectrum range of the signal source operation, I i (f) represents the interference power spectral density of signal source i at frequency f, S i (f) represents the target signal power spectrum density of signal source i at frequency f, λ3 represents the spectrum utilization efficiency weight coefficient, B i represents the spectrum bandwidth occupied by signal source i, M i Indicates the order corresponding to the modulation method of signal source i.

6. The method according to claim 1, characterized in that Generating a directional interference beam according to the blocking control signal, adjusting the power, beam direction and phase delay of the directional interference beam, and using digital pre-distortion technology to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam includes: Acquire the amplitude and phase information of the blocking control signal, calculate the Bessel function value according to the amplitude and phase information, multiply the Bessel function value by the harmonic suppression coefficient to obtain the harmonic suppression amount, and perform weighted combination of the harmonic suppression amount and the array element excitation value to obtain the array beam pattern; The array beam pattern is divided into a near-field region and a far-field region, the beam power in the near-field region is attenuated by a power function to obtain a near-field power distribution, the beam power in the far-field region is attenuated by an exponential function to obtain a far-field power distribution, and the near-field power distribution and the far-field power distribution are synthesized to obtain an overall power distribution; Obtaining a target direction angle of the beam, calculating a deviation between the current beam direction and the target direction, constructing a compensation function including a direction error and a sidelobe level according to the deviation, iteratively optimizing the compensation function using a gradient descent method to obtain a direction compensation amount, and combining the direction compensation amount with the overall power distribution to obtain an optimized beam; Detecting spatial position information and timing information of the optimized beam, calculating a spatial delay according to the spatial position information, calculating a time-varying delay according to the timing information, converting the spatial delay and the time-varying delay into frequency domain compensation coefficients, and processing the frequency domain compensation coefficients using a group delay equalizer to obtain a compensation transfer function; The input signal is filtered through the compensation transfer function to obtain a compensation signal sequence, the compensation signal sequence is time-delay extended to obtain an extended signal sequence, the extended signal sequence is combined with a nonlinear basis function to obtain a predistortion basis function, and the coefficients of the predistortion basis function are optimized by a variational inference algorithm to obtain an optimal predistortion coefficient; The optimal predistortion coefficient is distributed to the parallel processing units according to the signal subband, each processing unit simultaneously performs predistortion calculation to obtain a compensation signal, detects the error between the compensation signal and the expected signal, updates the predistortion parameter in real time according to the error, and finally outputs the compensated directional interference beam signal.

7. The method according to claim 6, characterized in that The input signal is filtered through the compensation transfer function to obtain a compensation signal sequence, the compensation signal sequence is time-delay extended to obtain an extended signal sequence, the extended signal sequence is combined with a nonlinear basis function to obtain a predistortion basis function, and the coefficients of the predistortion basis function are optimized using a variational inference algorithm to obtain the optimal predistortion coefficients, including: Performing a fast Fourier transform on the input signal to obtain a frequency domain signal, multiplying the frequency domain signal by a compensation transfer function to obtain a compensated frequency domain signal, and performing an inverse fast Fourier transform on the compensated frequency domain signal to obtain a compensated signal sequence; Determine the delay quantity based on the preset memory depth, determine the minimum interval between adjacent delay quantities according to the system coherence time, perform multi-stage delay processing on the compensation signal sequence according to the minimum interval to obtain multiple delay signals, and splice the multiple delay signals to form a first delay matrix; Extracting amplitude information and phase information of the compensation signal sequence, constructing an odd-power nonlinear basis function according to the amplitude information and the phase information, wherein the amplitude of the odd-power nonlinear basis function is determined by the odd power of the amplitude information, and the phase is determined by superposition of the phase information and the phase rotation angle; Combining the first time delay matrix with the odd-power nonlinear basis function to obtain a predistortion basis function matrix; A complex normal distribution function including a mean parameter and a covariance parameter is constructed as a variational posterior distribution function, and an optimization objective function is constructed according to the variational posterior distribution function and the predistortion basis function matrix; Deriving the optimization objective function with respect to a mean parameter and a covariance parameter respectively to obtain a first gradient and a second gradient, and combining the first gradient and the second gradient to obtain a total gradient; The mean parameter and the covariance parameter are iteratively updated according to the total gradient, and the iteration is stopped when the modulus value of the total gradient is less than a preset threshold, so as to obtain an optimal predistortion coefficient.

8. A mobile phone signal blocking shielding and adaptive identification system in an area where communication control and signal blocking are required, used to implement the method described in any one of claims 1 to 7, characterized in that: include: The first unit is used to collect electromagnetic signals in a target area through a distributed signal collector, perform time-frequency domain analysis on the electromagnetic signals, obtain first signal characteristic parameters, the first signal characteristic parameters include carrier frequency and modulation mode, calculate the spatial position coordinates of the signal source based on a triangulation positioning algorithm, construct a target signal characteristic vector with the first signal characteristic parameters and the spatial position coordinates, input the target signal characteristic vector into a preset convolutional neural network, extract communication protocol features and behavior pattern features in the target signal characteristic vector, and generate signal recognition data including signal movement trajectory prediction and communication behavior prediction; The second unit is used to calculate the minimum interference power threshold according to the signal identification data based on the channel capacity theory, iteratively optimize the interference power using a multi-objective optimization algorithm, compare the optimization result of each iteration with the minimum interference power threshold, stop the iteration when the optimization result is less than the minimum interference power threshold, obtain the optimal power configuration parameter, calculate the beam direction and phase delay according to the optimal power configuration parameter and the signal movement trajectory prediction data, and generate a blocking control signal, wherein the blocking control signal includes a power parameter, a direction parameter and a phase parameter; The third unit is used to generate a directional interference beam according to the blocking control signal, adjust the power, beam direction and phase delay of the directional interference beam, use digital pre-distortion technology to perform nonlinear distortion compensation on the adjusted directional interference beam to obtain a compensated directional interference beam, monitor the interference effect of the compensated directional interference beam in the target area in real time to obtain interference effect evaluation data, input the interference effect evaluation data into the convolutional neural network to update the signal recognition data to obtain updated signal recognition data, and recalculate the blocking control signal based on the updated signal recognition data to form a closed-loop control of signal shielding.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images