GPS pseudo code signal interference detection method and system based on Alpha-Beta pruning

Abstract

The invention provides a GPS pseudo code signal interference detection method based on Alpha-Beta pruning, and the method comprises the steps: carrying out the preprocessing of an obtained original GPS signal, obtaining a baseband signal, calculating a double-entropy value according to the baseband signal in combination with a preset initial window size, and outputting a signal state corresponding to the baseband signal according to the double-entropy value; calculating a dynamic time-frequency segmentation parameter according to the acquired GPS pseudo code type, the acquired baseband signal sampling rate, the signal state and the double entropy value; performing time-frequency analysis on the baseband signal by using the dynamic time-frequency segmentation parameter to obtain an initial time-frequency block; constructing a decision tree based on the initial time-frequency block, and executing Alpha-Beta pruning search on the decision tree to obtain a final score of interference judgment; and judging whether GPS pseudo code signal interference exists or not according to the final score, and if the interference exists, backtracking the optimal path searched by the Alpha-Beta pruning, and identifying the interference type. The problem that the interference detection difficulty is large due to the fact that the GPS signal shows the remarkable non-stationary characteristic is solved.

Description

Technical Field

[0001] This application relates to the field of signal detection, specifically to a method and system for detecting GPS pseudocode signal interference based on Alpha-Beta pruning. Background Technology

[0002] As a core technology in the field of satellite navigation and communication, GPS has been widely used in many critical scenarios such as transportation, precision agriculture, aerospace, and emergency rescue. Its positioning accuracy, service availability, and reliability are directly related to the safety and efficiency of various applications. The core working principle of the GPS system is to receive pseudocodes transmitted by multiple satellites through a receiver, and use the periodicity and autocorrelation of the pseudocodes to distinguish satellite signals, synchronize time, and measure distance, and finally complete positioning and navigation calculations. As the core identifier of GPS signals, the integrity of the pseudocodes is the foundation for ensuring the normal operation of the system.

[0003] However, GPS signals are susceptible to various forms of human-induced or natural interference during space transmission. Among these, pseudo-code interference is one of the most serious threats to GPS systems. Pseudo-code interference maliciously interferes with receivers by artificially generating pseudo-random codes that resemble genuine GPS pseudo-code signals or by disrupting the signal characteristics of genuine pseudo-codes. It mainly includes three typical types of interference: First, suppression interference, which overwhelms the genuine GPS signal by transmitting high-power signals, causing a sharp drop in the receiver's signal-to-noise ratio and masking the pseudo-code correlation peaks, making signal acquisition and tracking impossible. Second, deception interference, which simulates the power level and transmission characteristics of genuine GPS signals, sending false pseudo-code information to the receiver and inducing it to calculate incorrect positioning results; this type is highly concealed and more harmful. Third, multipath interference, where signals are reflected and refracted by the ground and superimposed on the direct signal before reaching the receiver, causing secondary peaks or broadening distortions in the pseudo-code correlation peaks, affecting positioning accuracy. Furthermore, the relative high-speed motion between the satellite and the receiver causes abrupt changes in Doppler frequency shift, making the GPS signal exhibit significant non-stationary characteristics, further increasing the difficulty of interference detection. Summary of the Invention

[0004] To address the problem of significant non-stationary characteristics of GPS signals and the difficulty in detecting interference in practical applications of existing technologies, this application provides a GPS pseudocode signal interference detection method and system based on Alpha-Beta pruning.

[0005] The first aspect of this application provides a method for detecting GPS pseudocode signal interference based on Alpha-Beta pruning, comprising: The acquired raw GPS IQ dual-channel signals are preprocessed to obtain the baseband signal; The dual entropy value is calculated based on the baseband signal and a preset initial window size. Based on the dual entropy value, output the signal state corresponding to the baseband signal; Based on the obtained GPS pseudocode type, the obtained baseband signal sampling rate, the signal state, and the dual entropy value, calculate the dynamic time-frequency segmentation parameters; The baseband signal is analyzed using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block. A decision tree is constructed based on the initial time-frequency block, and an Alpha-Beta pruning search is performed on the decision tree to obtain the final score for interference determination; The final score is used to determine whether there is GPS pseudocode signal interference. If interference exists, the optimal path of the Alpha-Beta pruning search is traced back to identify the type of interference.

[0006] In a possible implementation, calculating the dual entropy value based on the baseband signal and a preset window size includes: The baseband signal is divided into segments according to a preset initial window size, and the pseudocode autocorrelation time-varying entropy of the baseband signal within each initial window is calculated. The baseband signal is divided according to the preset initial window size, and the Doppler differential spectral entropy of the baseband signal within each initial window is calculated.

[0007] In a possible implementation, generating the signal state corresponding to the baseband signal based on the dual entropy value includes: If the pseudocode autocorrelation time-varying entropy is not greater than the pseudocode autocorrelation time-varying entropy threshold and the Doppler differential spectrum entropy is not greater than the Doppler differential spectrum entropy threshold, a stable signal is output. If the pseudocode autocorrelation time-varying entropy is greater than the pseudocode autocorrelation time-varying entropy threshold or the Doppler differential spectrum entropy is greater than the Doppler differential spectrum entropy threshold, a non-stationary signal is output. If the pseudocode autocorrelation time-varying entropy is not less than twice the preset pseudocode autocorrelation time-varying entropy threshold or the Doppler differential spectrum entropy is not less than twice the preset Doppler differential spectrum entropy threshold, an extremely non-stationary signal is output.

[0008] In a possible implementation, the step of dividing the baseband signal according to a preset initial window size and calculating the pseudocode autocorrelation time-varying entropy of the baseband signal within each initial window includes: Based on the baseband signal of the sampling point corresponding to each delay value selected in the initial window and the baseband signal of the sampling point after delay, the first pseudocode autocorrelation function value corresponding to each delay value is generated. The probability density of the first pseudocode autocorrelation function is obtained by normalizing the probability density of the first pseudocode autocorrelation function. The time-varying entropy of the pseudocode autocorrelation is calculated based on the probability density of the autocorrelation function of the first pseudocode.

[0009] In a possible implementation, dividing the baseband signal according to the preset initial window size and calculating the Doppler differential spectral entropy of the baseband signal within each initial window includes: Perform a short-time Fourier transform on the baseband signal within a single initial window to obtain a two-dimensional short-time Fourier transform spectrum in time and frequency. Calculate the difference between the short-time Fourier transform spectra corresponding to two adjacent initial windows, and obtain the Doppler difference spectrum by taking the absolute value; The probability density of the Doppler differential spectrum is normalized to obtain the differential spectrum probability density. The Doppler differential spectral entropy is calculated based on the differential spectral probability density.

[0010] In a possible implementation, calculating the dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value includes: Calculate the rate of change of the dual entropy value based on the stated dual entropy value; Based on the calculated rate of change of the dual entropy value, a comprehensive rate of change is generated; Calculate the number of sampling points within the pseudocode period based on the obtained GPS pseudocode type and the obtained baseband signal sampling rate; The dynamic time-frequency segmentation parameters are calculated based on the number of sampling points, the signal state, and the overall rate of change.

[0011] In a possible implementation, the dynamic time-frequency segmentation parameters are calculated based on the number of sampling points, the signal state, and the overall rate of change. These dynamic time-frequency segmentation parameters include window size and overlap rate. The window size is dynamically adjusted based on the number of sampling points and the signal state. The overlap rate is dynamically adjusted based on the overall rate of change.

[0012] In a possible implementation, the step of performing time-frequency analysis on the baseband signal using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block includes: The baseband signal is analyzed in time and frequency according to the dynamic time and frequency segmentation parameters to obtain the first time and frequency block; Perform feature integrity verification on the first time-frequency block and output the initial time-frequency block.

[0013] In a possible implementation, performing feature integrity verification on the first time-frequency block and outputting an initial time-frequency block includes: Based on the first time-frequency block, determine the number of valid pseudocode autocorrelation peaks in the first time-frequency block; The pseudo-code autocorrelation peak loss rate is calculated based on the number of pseudo-code periods covered by the first time-frequency block and the number of valid pseudo-code autocorrelation peaks. If the pseudocode autocorrelation peak loss rate is less than or equal to a preset threshold, the current time-frequency block is output as the initial time-frequency block; otherwise, the dynamic time-frequency segmentation parameters are adjusted until the pseudocode autocorrelation peak loss rate of the generated time-frequency block is less than or equal to the preset threshold.

[0014] A second aspect of this application provides a GPS pseudocode signal interference detection system based on Alpha-Beta pruning, comprising: The signal preprocessing module is used to preprocess the acquired raw GPS IQ dual-channel signals to obtain the baseband signal; The entropy calculation module is used to calculate the dual entropy value based on the baseband signal and a preset initial window size. The signal state calculation module is used to output the signal state corresponding to the baseband signal based on the dual entropy value. The dynamic parameter adjustment module is used to calculate dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value. The time-frequency analysis module is used to perform time-frequency analysis on the baseband signal using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block; The scoring module is used to construct a decision tree based on the initial time-frequency block and perform Alpha-Beta pruning search on the decision tree to obtain the final score for interference determination; The identification module is used to determine whether there is GPS pseudocode signal interference based on the final score. If interference exists, it backtracks the optimal path of the Alpha-Beta pruning search to identify the type of interference.

[0015] This application first preprocesses the original signal, then identifies the stationary state of the signal based on joint verification of dual entropy values, calculates suitable time-frequency segmentation parameters by combining the entropy change rate, and obtains qualified time-frequency blocks after feature integrity verification. Finally, interference detection and type identification are achieved through decision tree construction and Alpha-Beta pruning. The segmentation window size and overlap rate of the GPS signal are not arbitrarily set, but rather designed to ensure that the time-frequency blocks can fully retain the key signal features required for interference detection while matching the rhythm of signal state changes. The GPS signal state determines the time scale and feature continuity requirements of the signal features, which is the fundamental basis for dynamically adjusting the window size and overlap rate. The dynamic adjustment principle of this application solves the defect that fixed parameters cannot adapt to the non-stationary characteristics of GPS signals. By allowing the time-frequency blocks to actively adapt to signal features, rather than allowing signal features to passively adapt to fixed parameters, it ensures that subsequent Alpha-Beta pruning can perform interference detection based on complete video features, ultimately improving the accuracy of interference detection. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a GPS pseudocode signal interference detection method based on Alpha-Beta pruning in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of a GPS pseudocode signal interference detection system based on Alpha-Beta pruning in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] Alpha-Beta pruning is a technique commonly used to optimize search algorithms. Originally applied in game theory and decision trees, it aims to reduce the computational complexity of the search space. Its core idea is to prune obviously ineffective branches and retain potentially optimal paths, thereby significantly improving efficiency while maintaining accuracy. The alpha value represents the known lower bound of the current path's optimum, and the beta value represents the known upper bound of the current path's optimum. When the evaluation value of a node exceeds the alpha-beta range, its subsequent branches can be pruned. During the search process, the algorithm continuously updates the alpha and beta values, gradually narrowing down the range of feasible solutions.

[0021] Based on this, this application provides an implementation method for GPS pseudocode signal interference detection based on Alpha-Beta pruning, such as... Figure 1 As shown, it includes: S101 preprocesses the acquired raw GPS IQ dual-channel signal to obtain the baseband signal.

[0022] S102, calculate the dual entropy value based on the baseband signal and the preset initial window size.

[0023] S103, based on the dual entropy value, output the signal state corresponding to the baseband signal.

[0024] S104, calculate the dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value.

[0025] S105, the baseband signal is analyzed using the dynamic time-frequency segmentation parameters to obtain the initial time-frequency block.

[0026] S106, construct a decision tree based on the initial time-frequency block, and perform Alpha-Beta pruning search on the decision tree to obtain the final score for interference determination.

[0027] S107. Based on the final score, determine whether there is GPS pseudocode signal interference. If interference exists, backtrack the optimal path of the Alpha-Beta pruning search to identify the type of interference.

[0028] It should be noted that preprocessing is a preliminary purification process of the original GPS IQ signal, including down-conversion and low-pass filtering. The original signal is a radio frequency (RF) signal and is mixed with high-frequency noise. The RF signal has a high frequency and contains irrelevant carrier components. Based on these characteristics, the original signal cannot be directly used for the analysis of pseudocode features and Doppler shift features. Furthermore, the baseband signal is a near-zero frequency signal obtained after preprocessing. By stripping the RF carrier and filtering out high-frequency noise, the core features of the GPS pseudocode and Doppler shift information are retained, serving as the core input data for subsequent dual-entropy value verification and time-frequency analysis. For example, this application down-converts the original RF signal in the original GPS IQ signal to baseband to obtain an initial baseband signal. The center frequency of the original RF signal is 1575.42MHz in the GPS L1 band. Subsequently, an FIR low-pass filter is used to filter the initial baseband signal to remove high-frequency noise, resulting in the baseband signal.

[0029] For example, the raw GPS signal is denoted as Sampling rate Typically 1MHz to 5MHz, corresponding to the sampling requirements of C / A code (pseudocode period T_p=1ms); Processing objective: Filter out out-of-band noise, downconvert to baseband, and ensure the accuracy of subsequent dual-entropy value calculation.

[0030] The downconversion process is as follows: the original radio frequency signal (center frequency 1575.42MHz) is downconverted to baseband, using the following formula: The symbols are explained as follows: The original radio frequency signal, This is the center frequency of the GPS L1 band. The imaginary unit, As a time variable, this step eliminates the interference of the carrier frequency on subsequent Doppler shift calculations.

[0031] The low-pass filtering process is as follows: an FIR low-pass filter is used to remove high-frequency noise from the baseband signal. The filter's impulse response is: ,in, The value of the filter tap index is... N=64; The cutoff frequency covers the GPS signal bandwidth; For Hanning Window , .

[0032] In traditional detection methods, time-frequency segmentation parameters are mostly fixed values, which cannot adapt to the dynamic characteristics of GPS signals. When the signal has non-stationary characteristics, fixed parameters easily lead to the loss of time-frequency block features or computational redundancy. This application first preprocesses the original signal, then identifies the stationary state of the signal based on joint verification of dual entropy values, calculates the appropriate time-frequency segmentation parameters based on the entropy change rate, obtains qualified time-frequency blocks after feature integrity verification, and finally achieves interference detection and type identification through decision tree construction and Alpha-Beta pruning. The segmentation window size and overlap rate of the GPS signal are not arbitrarily set, but rather ensure that the time-frequency blocks can fully retain the key signal features required for interference detection and match the rhythm of signal state changes. The GPS signal state determines the time scale and feature continuity requirements of the signal features, which is the fundamental basis for dynamically adjusting the window size and overlap rate. The dynamic adjustment principle of this application solves the defect that fixed parameters cannot adapt to the non-stationary characteristics of GPS signals. By allowing the time-frequency blocks to actively adapt to signal features, rather than allowing signal features to passively adapt to fixed parameters, it ensures that subsequent Alpha-Beta pruning can perform interference detection based on complete video features, ultimately improving the accuracy of interference detection.

[0033] In one embodiment of this application, calculating the dual entropy value based on the baseband signal and a preset window size includes: S201, the baseband signal is divided according to a preset initial window size, and the pseudocode autocorrelation time-varying entropy of the baseband signal in each initial window is calculated.

[0034] S202, the baseband signal is divided according to the preset initial window size, and the Doppler differential spectrum entropy of the baseband signal in each initial window is calculated.

[0035] It should be noted that the aforementioned dual entropy values ​​refer to the pseudocode autocorrelation time-varying entropy and the Doppler differential spectral entropy, which are quantitative indicators designed in this application to represent the stationary characteristics of GPS signals. The pseudocode autocorrelation time-varying entropy originates from the specific autocorrelation characteristics of the GPS signal's pseudocode periodicity. In a stationary state, the autocorrelation function exhibits a sharp single peak, while in a non-stationary state, the autocorrelation function is diffuse or multi-peaked. The pseudocode autocorrelation time-varying entropy value can quantify the degree of disorder in its distribution. The Doppler differential spectral entropy originates from the Doppler frequency shift characteristics of the GPS signal. The Doppler frequency shift is generated by the relative motion between the satellite and the receiver. In a stationary state, the frequency shift is stable, while in a non-stationary state, the frequency shift jumps. The frequency shift change is quantified through the Doppler differential entropy of the frequency domain spectrum.

[0036] It should be noted that, firstly, the baseband signal is segmented according to a preset initial window size. The size of the initial window is designed based on the local feature scale of the GPS pseudocode to ensure that each window can reflect the local characteristics of a continuous signal segment. It is necessary to avoid windows that are too large to capture instantaneous changes, or windows that are too small to be severely affected by noise interference. The size of the preset initial window can be adjusted according to the actual rate of change of the signal, and this application is not limited to this.

[0037] The time-varying entropy of pseudocode autocorrelation reflects the stability of the pseudocode structure. A lower entropy value indicates a more concentrated autocorrelation function, resulting in a more stable signal; a higher entropy value indicates a more diffuse autocorrelation function, resulting in a less stationary signal. The Doppler differential spectral entropy reflects the stability of the frequency domain characteristics. A lower entropy value indicates a smaller Doppler frequency shift, resulting in a more stable signal; a higher entropy value indicates a more drastic frequency shift, resulting in a less stationary signal.

[0038] This step in the application uses dual entropy values ​​to complement each other, avoiding the one-sidedness of a single indicator. Relying solely on a single entropy value may lead to misjudgment. The combined dual entropy values ​​can comprehensively cover the core characteristics of the signal and improve the accuracy of determining the steady state.

[0039] In one embodiment that can be implemented in this application, generating the signal state corresponding to the baseband signal based on the dual entropy value includes: S301, if the pseudocode autocorrelation time-varying entropy is not greater than the pseudocode autocorrelation time-varying entropy threshold and the Doppler differential spectrum entropy is not greater than the Doppler differential spectrum entropy threshold, output a stable signal.

[0040] S302, if the pseudocode autocorrelation time-varying entropy is greater than the pseudocode autocorrelation time-varying entropy threshold or the Doppler differential spectrum entropy is greater than the Doppler differential spectrum entropy threshold, output a non-stationary signal.

[0041] S303: If the pseudocode autocorrelation time-varying entropy is not less than twice the preset pseudocode autocorrelation time-varying entropy threshold or the Doppler differential spectrum entropy is not less than twice the preset Doppler differential spectrum entropy threshold, output an extreme non-stationary signal.

[0042] In one embodiment of this application, the step of dividing the baseband signal according to a preset initial window size and calculating the pseudocode autocorrelation time-varying entropy of the baseband signal within each initial window includes: S401, based on the baseband signal of the sampling point corresponding to each delay value selected in the initial window and the baseband signal of the sampling point after delay, generate the first pseudocode autocorrelation function value corresponding to each delay value.

[0043] S402, normalize the probability density of the first pseudocode autocorrelation function to obtain the probability density of the first pseudocode autocorrelation function.

[0044] S403, calculate the time-varying entropy of the pseudocode autocorrelation based on the probability density of the first pseudocode autocorrelation function.

[0045] For example, for a baseband signal within a single initial small window, its pseudocode autocorrelation function (ACF) is calculated, and the formula for calculating the ACF is: ,in This is a delayed index, where M is the number of valid sampling points. Let m be the baseband signal at the m-th sampling point. for conjugate, To extract the real part, the probability density of the ACF is normalized to obtain the ACF probability density. The normalization formula is: Where W0 is the number of points in the preset initial small window. Based on the ACF probability density The time-varying entropy (RTE) of the pseudocode autocorrelation is calculated using the following formula: .

[0046] In one embodiment of this application, dividing the baseband signal according to the preset initial window size and calculating the Doppler differential spectral entropy of the baseband signal within each initial window includes: S501 performs a short-time Fourier transform on the baseband signal within a single initial window to obtain a two-dimensional short-time Fourier transform spectrum in time and frequency.

[0047] S502, calculate the difference between the short-time Fourier transform spectra corresponding to two adjacent initial windows, and obtain the Doppler difference spectrum by taking the absolute value.

[0048] S503, normalize the probability density of the Doppler differential spectrum to obtain the differential spectrum probability density.

[0049] S504, Calculate the Doppler differential spectral entropy based on the differential spectral probability density.

[0050] For example, performing a short-time Fourier transform (STFT) on the baseband signal within a single initial small window yields a two-dimensional time-frequency STFT spectrum. Where t is the time window index and f is the frequency index, the first-order difference of the STFT spectrum between two adjacent time windows is calculated to obtain the Doppler difference spectrum. The difference calculation formula is: ,in To obtain the absolute value, the Doppler difference spectrum is... Normalize the probability density to obtain the difference spectral probability density. and based on The Doppler differential spectral entropy (DSE) is calculated using the following formula: ,in The sampling rate of the baseband signal. It is the Nyquist frequency.

[0051] In one embodiment of this application, calculating the dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value includes: S601, Calculate the rate of change of the dual entropy value based on the dual entropy value.

[0052] S602, Based on the calculated rate of change of the dual entropy value, a comprehensive rate of change is generated.

[0053] S603, calculates the number of sampling points within the pseudocode period based on the acquired GPS pseudocode type and the acquired baseband signal sampling rate.

[0054] S604, calculate the dynamic time-frequency segmentation parameters based on the number of sampling points, the signal state, and the overall rate of change.

[0055] It should be noted that the period of GPS pseudocode types is a fixed parameter and is a core inherent characteristic of the pseudocode. For example, for C / A code, which is the mainstream civilian pseudocode of GPS, its period is fixed at 1 millisecond, and for P code, which is the military pseudocode, its period is fixed at 7 days.

[0056] For example, the overall rate of change of the dual entropy values ​​is calculated. The Rate of change of RTE With DSE rate of change The average value of, where , , Index for the current time window, As an index of the previous time window, the window size W is calculated based on the signal's stationary state, and W is related to the GPS pseudocode period. Corresponding number of sampling points Related, ,in This is the sampling rate; if it is a stationary signal, then... If it is a non-stationary signal, then If it is an extremely non-stationary signal, then Based on the aforementioned comprehensive rate of change Calculate the overlap rate O: If ,but ;like ,but ;like ,but .

[0057] In one embodiment of this application, the method further includes, for example, obtaining the GPS pseudocode type, and if it is a C / A code, determining the pseudocode period. S322: Obtain the sampling rate of the baseband signal. The The value range is from 1MHz to 5MHz, according to The number of sampling points was calculated. .

[0058] In one embodiment of this application, the dynamic time-frequency segmentation parameters are calculated based on the number of sampling points, the signal state, and the overall rate of change. These dynamic time-frequency segmentation parameters include window size and overlap rate. S701, dynamically adjust the window size according to the number of sampling points and the signal state.

[0059] S702, dynamically adjust the overlap rate according to the comprehensive change rate.

[0060] It should be noted that the overlap rate mentioned is the overlap rate between two adjacent time-frequency blocks during the GPS signal time-frequency segmentation process. Specifically, it refers to the proportion of baseband signal sampling points shared by the later time-frequency block and the previous time-frequency block to the total sampling points of a single time-frequency block when the baseband signal is continuously segmented according to the dynamically adjusted window size. For example, assuming a non-stationary signal scenario, the baseband signal has 500 sampling points according to the window size. The first time-frequency block covers sampling points 1 to 500, the second time-frequency block covers sampling points 126 to 625, and the two time-frequency blocks share sampling points 126 to 500, for a total of 375 shared sampling points. In this case, the overlap rate is 375 / 500 = 75%.

[0061] In one embodiment of this application, the step of performing time-frequency analysis on the baseband signal using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block includes: S801, perform time-frequency analysis on the baseband signal according to the dynamic time-frequency segmentation parameters to obtain the first time-frequency block.

[0062] S802, perform feature integrity verification on the first time-frequency block and output the initial time-frequency block.

[0063] In one embodiment that can be implemented in this application, the step of performing feature integrity verification on the first time-frequency block and outputting an initial time-frequency block includes: S901, Based on the first time-frequency block, confirm the number of valid pseudocode autocorrelation peaks of the first time-frequency block.

[0064] S902, calculate the pseudo-code autocorrelation peak loss rate based on the number of pseudo-code periods covered by the first time-frequency block and the number of valid pseudo-code autocorrelation peaks.

[0065] S903, if the pseudocode autocorrelation peak loss rate is less than or equal to a preset threshold, then output the current time-frequency block as the initial time-frequency block; otherwise, adjust the dynamic time-frequency segmentation parameters until the pseudocode autocorrelation peak loss rate of the generated time-frequency block is less than or equal to the preset threshold.

[0066] From the first time-frequency block, the original time-domain baseband signal within the current time window is separated, i.e., the frequency dimension of time-frequency analysis is removed, restoring it to a continuous IQ sampling point sequence, ensuring that the signal retains the complete pseudocode periodicity. The pseudocode autocorrelation function of the original time-domain baseband signal is calculated. For example, a series of continuous delay values ​​are set, with the delay range covering one complete pseudocode period, such as a 1-millisecond delay for C / A code. For each delay value, the time-domain signal is multiplied by its conjugate with the signal delayed by that time to cancel out phase effects. The real part of each multiplication result is taken, and then the average of all real parts is calculated to obtain the autocorrelation function value corresponding to that delay value. The autocorrelation function values ​​corresponding to all delay values ​​form an autocorrelation function curve with the delay value on the horizontal axis and the autocorrelation function value on the vertical axis. Since the signal may have a small amount of residual noise, the autocorrelation function curve may have false small peaks. Noise reduction is required by means of the following method: use moving average smoothing. Use a window with a length equal to the number of sampling points corresponding to the width of the pseudocode chip to perform moving average on the autocorrelation function curve to smooth out the false small peaks. This ensures that after smoothing, the maximum peak value of the autocorrelation function and the periodic small peaks caused by the pseudocode characteristics will not be destroyed, and only the irregular small fluctuations caused by random noise will be eliminated.

[0067] A local maximum (MMR) is a point on the curve where the autocorrelation function value is greater than the values ​​of its two immediate neighbors. Specifically, this is achieved by comparing the function value of each point with its two immediate neighbors, from the start to the end of the smoothed autocorrelation curve. If the function value at a point is greater than both the preceding and following points, then that point is marked as a candidate peak with a probability of being a true pseudo-code peak. The position delay values ​​and corresponding autocorrelation function values ​​of all candidate peaks are recorded.

[0068] Valid peaks are selected from candidate peaks according to preset rules. Among all candidate peaks, the peak with the largest autocorrelation function value is found, which is the global maximum peak. A threshold for valid peaks is set, with the peak height of the global maximum peak as the benchmark. For example, the peak value of a valid peak must be ≥ 70% of the peak value of the global maximum peak. Low-amplitude noise peaks are excluded. The peak height of each candidate peak is checked one by one. If the peak height is ≥ the above threshold, it is determined to be a valid peak. If it is lower than the threshold, it is determined to be an invalid peak. All valid peaks that meet the conditions are counted, which is the number of pseudo-code autocorrelation valid peaks in the initial time-frequency block.

[0069] For example, the number of effective pseudo-code autocorrelation peaks in the initial time-frequency block is calculated, where an effective peak is a local maximum value whose peak value exceeds 0.7 times the global maximum peak. The pseudo-code autocorrelation peak loss rate L of the initial time-frequency block is then calculated, and the formula for L is: The total number of peaks is the number of GPS pseudocode cycles covered by the initial time-frequency block. Then the initial time-frequency block is a qualified time-frequency block; if Then increase the window size W by 25% and re-execute the time-frequency analysis in step S4 until the result is obtained. Qualified time-frequency blocks.

[0070] In one embodiment of this application, the application further includes: starting from the root node of the decision tree, selecting the child node with the largest Alpha value layer by layer to form the optimal path for interference determination; extracting the time-frequency block features corresponding to each child node in the optimal path, wherein the time-frequency block features include signal power, pseudo-code autocorrelation peak shape and Doppler frequency shift; matching a preset interference type feature library based on the time-frequency block features; identifying and outputting the interference type, wherein the interference type includes suppression interference, deception interference and multipath interference.

[0071] In the second aspect of this application, as Figure 2 As shown, a GPS pseudocode signal interference detection system based on Alpha-Beta pruning is provided, comprising: The signal preprocessing module 1001 is used to preprocess the acquired raw GPS IQ dual-channel signal to obtain the baseband signal.

[0072] The entropy calculation module 1002 is used to calculate the dual entropy value based on the baseband signal and a preset initial window size.

[0073] The signal state calculation module 1003 is used to output the signal state corresponding to the baseband signal based on the dual entropy value.

[0074] The dynamic parameter adjustment module 1004 is used to calculate dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value.

[0075] The time-frequency analysis module 1005 is used to perform time-frequency analysis on the baseband signal using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block.

[0076] The scoring module 1006 is used to construct a decision tree based on the initial time-frequency block and perform Alpha-Beta pruning search on the decision tree to obtain the final score for interference determination.

[0077] The identification module 1007 is used to determine whether there is GPS pseudocode signal interference based on the final score. If interference exists, it backtracks the optimal path of the Alpha-Beta pruning search to identify the type of interference.

[0078] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A GPS pseudocode signal interference detection method based on Alpha-Beta pruning, characterized in that, The method includes: The raw GPS signal is preprocessed to obtain the baseband signal; The dual entropy value is calculated based on the baseband signal and the preset initial window size. Based on the dual entropy value, output the signal state corresponding to the baseband signal; Based on the obtained GPS pseudocode type, the obtained baseband signal sampling rate, the signal state, and the dual entropy value, calculate the dynamic time-frequency segmentation parameters; The baseband signal is analyzed using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block. A decision tree is constructed based on the initial time-frequency block, and an Alpha-Beta pruning search is performed on the decision tree to obtain the final score for interference determination; The final score is used to determine whether there is GPS pseudocode signal interference. If interference exists, the optimal path of the Alpha-Beta pruning search is traced back to identify the type of interference.

2. The method according to claim 1, characterized in that, The step of calculating the dual entropy value based on the baseband signal and a preset window size includes: The baseband signal is divided into segments according to a preset initial window size, and the pseudocode autocorrelation time-varying entropy of the baseband signal within each initial window is calculated. The baseband signal is divided according to the preset initial window size, and the Doppler differential spectral entropy of the baseband signal within each initial window is calculated.

3. The method according to claim 2, characterized in that, The step of generating the signal state corresponding to the baseband signal based on the dual entropy value includes: If the pseudocode autocorrelation time-varying entropy is not greater than the pseudocode autocorrelation time-varying entropy threshold and the Doppler differential spectrum entropy is not greater than the Doppler differential spectrum entropy threshold, a stable signal is output. If the pseudocode autocorrelation time-varying entropy is greater than the pseudocode autocorrelation time-varying entropy threshold or the Doppler differential spectrum entropy is greater than the Doppler differential spectrum entropy threshold, a non-stationary signal is output. If the pseudocode autocorrelation time-varying entropy is not less than twice the preset pseudocode autocorrelation time-varying entropy threshold or the Doppler differential spectrum entropy is not less than twice the preset Doppler differential spectrum entropy threshold, an extremely non-stationary signal is output.

4. The method according to claim 2, characterized in that, The step of dividing the baseband signal according to a preset initial window size and calculating the pseudocode autocorrelation time-varying entropy of the baseband signal within each initial window includes: Based on the baseband signal of the sampling point corresponding to each delay value selected in the initial window and the baseband signal of the sampling point after delay, the first pseudocode autocorrelation function value corresponding to each delay value is generated. The probability density of the first pseudocode autocorrelation function is obtained by normalizing the probability density of the first pseudocode autocorrelation function. The time-varying entropy of the pseudocode autocorrelation is calculated based on the probability density of the autocorrelation function of the first pseudocode.

5. The method according to claim 2, characterized in that, The step of dividing the baseband signal according to the preset initial window size and calculating the Doppler differential spectral entropy of the baseband signal within each initial window includes: Perform a short-time Fourier transform on the baseband signal within a single initial window to obtain a two-dimensional short-time Fourier transform spectrum in time and frequency. Calculate the difference between the short-time Fourier transform spectra corresponding to two adjacent initial windows, and obtain the Doppler difference spectrum by taking the absolute value; The probability density of the Doppler differential spectrum is normalized to obtain the differential spectrum probability density. The Doppler differential spectral entropy is calculated based on the differential spectral probability density.

6. The method according to claim 1, characterized in that, The calculation of dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value includes: Calculate the rate of change of the dual entropy value based on the stated dual entropy value; Based on the calculated rate of change of the dual entropy value, a comprehensive rate of change is generated; Calculate the number of sampling points within the pseudocode period based on the obtained GPS pseudocode type and the obtained baseband signal sampling rate; The dynamic time-frequency segmentation parameters are calculated based on the number of sampling points, the signal state, and the overall rate of change.

7. The method according to claim 6, characterized in that, The dynamic time-frequency segmentation parameters are calculated based on the number of sampling points, the signal state, and the overall rate of change. These parameters include the window size and overlap rate. Adjust the window size according to the number of sampling points and the signal state; The overlap rate is adjusted based on the overall rate of change.

8. The method according to claim 1, characterized in that, The step of performing time-frequency analysis on the baseband signal using the dynamic time-frequency segmentation parameters to obtain the initial time-frequency block includes: The baseband signal is analyzed in time and frequency according to the dynamic time and frequency segmentation parameters to obtain the first time and frequency block; Perform feature integrity verification on the first time-frequency block and output the initial time-frequency block.

9. The method according to claim 8, characterized in that, The step of performing feature integrity verification on the first time-frequency block and outputting the initial time-frequency block includes: Based on the first time-frequency block, determine the number of valid pseudocode autocorrelation peaks in the first time-frequency block; The pseudo-code autocorrelation peak loss rate is calculated based on the number of pseudo-code periods covered by the first time-frequency block and the number of valid pseudo-code autocorrelation peaks. If the pseudocode autocorrelation peak loss rate is less than or equal to a preset threshold, the current time-frequency block is output as the initial time-frequency block; otherwise, the dynamic time-frequency segmentation parameters are adjusted until the pseudocode autocorrelation peak loss rate of the generated time-frequency block is less than or equal to the preset threshold.

10. A GPS pseudocode signal interference detection system based on Alpha-Beta pruning, characterized in that, The system includes: The signal preprocessing module is used to preprocess the acquired raw GPS signal to obtain the baseband signal; The entropy calculation module is used to calculate the dual entropy value based on the baseband signal and a preset initial window size. The signal state calculation module is used to output the signal state corresponding to the baseband signal based on the dual entropy value. The dynamic parameter adjustment module is used to calculate dynamic time-frequency segmentation parameters based on the acquired GPS pseudocode type, the acquired baseband signal sampling rate, the signal state, and the dual entropy value. The time-frequency analysis module is used to perform time-frequency analysis on the baseband signal using the dynamic time-frequency segmentation parameters to obtain an initial time-frequency block; The scoring module is used to construct a decision tree based on the initial time-frequency block and perform Alpha-Beta pruning search on the decision tree to obtain the final score for interference determination; The identification module is used to determine whether there is GPS pseudocode signal interference based on the final score. If interference exists, it backtracks the optimal path of the Alpha-Beta pruning search to identify the type of interference.

Images