A Beidou anti-jamming spoofing timing method and system based on multi-layer detection

Abstract

The present invention relates to the technical field of satellite timekeeping, and discloses a Beidou anti-jamming spoofing timekeeping method and system based on multi-layer detection, including: receiving multi-band signals of satellites and recording noise parameters; extracting time-frequency domain features of the multi-band signals, performing weighted fusion on the extracted time-frequency features to obtain a comprehensive feature vector; establishing a joint analysis model, detecting abnormal signals according to the comprehensive feature vector, performing time resolution on the true signals passing the detection, and outputting time information. The method of the present invention ensures the efficient discrimination of abnormal signals and accurate timekeeping output, significantly improves the timekeeping accuracy and stability of the Beidou system in complex interference scenarios, and provides a reliable time reference for key industries.

Description

Technical Field

[0001] The present invention relates to the technical field of satellite time synchronization, and specifically to a Beidou anti-jamming spoofing time synchronization method and system based on multi-layer detection. Background Art

[0002] The Beidou Satellite Navigation System (BDS), as a global satellite navigation system independently developed by China, not only provides high-precision positioning services but also plays an important role in the field of time synchronization. The time synchronization technology provides a unified and accurate time reference for ground devices through the propagation characteristics of satellite signals, and is widely used in fields such as power dispatching, communication networks, and financial transactions. However, with the widespread popularity of satellite navigation technology, the problem that its signals are vulnerable to interference and spoofing attacks has become increasingly prominent. In a complex electromagnetic environment, especially in high-dynamic or high-noise scenarios, interference signals (such as broadband interference and narrowband interference) and forged signals (such as spoofing attacks) can significantly affect the credibility of satellite signals, resulting in a decrease in time synchronization accuracy and even serious consequences such as time synchronization failure.

[0003] To address the above challenges, existing technologies have introduced various anti-jamming and anti-spoofing time synchronization methods. For example, the interference suppression method based on single-frequency signals realizes preliminary discrimination by monitoring the signal strength and noise ratio; the multi-band fusion technology attempts to enhance the anti-jamming ability by utilizing the complementarity between multi-frequency signals; and the anomaly detection method based on machine learning identifies abnormal signal features by training a model. These methods have improved the anti-jamming and anti-spoofing capabilities to a certain extent, but their effectiveness and adaptability in complex environments still have limitations. For example, the single-frequency signal technology cannot completely eliminate multipath interference, the existing multi-band methods lack in-depth fusion of time-frequency domain features, and the machine learning methods face problems such as single features and poor real-time performance. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the existing anti-jamming time synchronization technology has the problems of single signal feature analysis dimension and insufficient adaptability to environmental changes.

[0006] To solve the above technical problem, the present invention provides the following technical solution: A Beidou anti-jamming spoofing time synchronization method based on multi-layer detection, including: receiving multi-band signals of a satellite and recording noise parameters; extracting time-frequency domain features of the multi-band signals, weighting and fusing the extracted time-frequency features to obtain a comprehensive feature vector; establishing a joint analysis model, detecting abnormal signals according to the comprehensive feature vector, performing time calculation on the true signals that pass the detection, and outputting time information.

[0007] As a preferred solution of the Beidou anti-jamming spoofing timing method based on multi-layer detection according to the present invention, wherein: the multi-band signals include B1-band signals, B2-band signals and B3-band signals.

[0008] As a preferred solution of the Beidou anti-jamming spoofing timing method based on multi-layer detection according to the present invention, wherein: the noise parameters include, during the period without signal transmission, recording the noise parameters and calculating the noise power, expressed as:

[0009]

[0010] wherein, P noise,i represents the noise power on frequency band i; i represents the frequency band index; T represents the sampling time; n i (t) represents the noise signal intensity of frequency band i at time t.

[0011] As a preferred solution of the Beidou anti-jamming spoofing timing method based on multi-layer detection according to the present invention, wherein: the time-frequency domain feature extraction includes filtering, denoising and normalizing the multi-band signals, extracting the time-domain features of the signals in the time domain, including phase offset rate, power fluctuation and time delay characteristics; using the fast Fourier transform to map the signals from the time domain to the frequency domain and extracting the frequency-domain features.

[0012] As a preferred solution of the Beidou anti-jamming spoofing timing method based on multi-layer detection according to the present invention, wherein: the weighted fusion includes calculating the signal power of each frequency band, expressed as:

[0013]

[0014] wherein, P signal,i represents the signal power of frequency band i; x i (t) represents the original signal received on frequency band i; calculating the signal-to-noise ratio SNR i , expressed as:

[0015]

[0016] Using the characteristic that the signal-to-noise ratio reflects the signal quality, normalizing the signal-to-noise ratio to calculate the weights, which intuitively reflects the contribution ratio of the signals of each frequency band to the comprehensive feature vector, expressed as:

[0017]

[0018] wherein, w i represents the weighting coefficient of frequency band i; represents the sum of the signal-to-noise ratios of all frequency bands.

[0019] As a preferred solution of the Beidou anti-jamming spoofing timing method based on multi-layer detection according to the present invention, wherein: the weighted fusion further includes, according to the weighting coefficient w i weight the time-domain features and frequency-domain features, expressed as:

[0020] T weighted,i = w i ·T i

[0021] F weighted,i = w i ·F i

[0022] wherein, T weighted,i represents the time-domain weighted feature of frequency band i; T i represents the time-domain feature vector of frequency band i; F weighted,i represents the frequency-domain weighted feature of frequency band i; F i represents the frequency-domain feature vector of frequency band i; perform weighted fusion on the time-domain and frequency-domain features to obtain a comprehensive feature vector, expressed as:

[0023]

[0024] wherein, V represents the comprehensive feature vector; α i represents the fusion weight of the time-domain features; β i represents the fusion weight of the frequency-domain features; α i and β i are dynamically adjusted based on the signal-to-noise ratio.

[0025] As a preferred solution of the Beidou anti-jamming spoofing timing method based on multi-layer detection according to the present invention, wherein: the joint analysis model includes, based on the logistic regression model, using the gradient descent method to optimize the logistic regression loss function, establishing a joint analysis model, and performing anomaly detection on the comprehensive feature vector V to obtain the anomaly probability P.

[0026] A Beidou anti-jamming spoofing timing system adopting any method of the present invention, wherein: a collection module, receiving multi-band signals of satellites, recording noise parameters and calculating the noise power during the signal-free transmission period, and uploading them to the analysis module; an analysis module, extracting the time-frequency features of the multi-band signals and performing weighted fusion, dynamically adjusting according to the noise parameters by calculating the signal-to-noise ratio, and obtaining a comprehensive feature vector; a timing module, establishing a joint analysis model based on the logistic regression model, detecting abnormal signals according to the comprehensive feature vector, performing time resolution on the true signals passing the detection, and outputting time information.

[0027] A computer device, comprising: a memory and a processor; the memory stores a computer program, including: when the processor executes the computer program, the steps of the method described in any one of the present inventions are implemented.

[0028] A computer-readable storage medium, on which a computer program is stored, including: when the computer program is executed by a processor, the steps of the method described in any one of the present inventions are implemented.

[0029] Advantages of the present invention: By receiving multi-band signals and recording noise parameters, the method of the present invention ensures the accuracy and integrity of the basic data for anti-interference processing, which helps to dynamically adapt to interference changes in complex environments; by extracting time-frequency domain features and weighted fusion to generate a comprehensive feature vector, the method realizes the in-depth analysis and optimized expression of the multi-dimensional characteristics of the signal, enhancing the robustness of signal discrimination; finally, using the joint analysis model to perform anomaly detection and timing calculation on the comprehensive feature vector, the method ensures the efficient discrimination of abnormal signals and accurate timing output, significantly improving the timing accuracy and stability of the Beidou system in complex interference scenarios, and providing a reliable time reference for key industries. Description of the Drawings

[0030] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0031] Figure 1 It is the overall flowchart of a Beidou anti-interference spoofing timing method based on multi-layer detection provided by an embodiment of the present invention. Detailed Embodiments

[0032] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the detailed embodiments of the present invention with reference to the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0033] Embodiment 1, referring to Figure 1 , which is an embodiment of the present invention, provides a Beidou anti-interference spoofing timing method based on multi-layer detection, including:

[0034] S1: Receive multi-band signals of satellites and record noise parameters.

[0035] Further, the multi - band signal refers to the navigation signal sent by navigation satellites (such as the Beidou system) through radio waves of different frequencies. The multi - band signals usually work on different frequency bands, and each band has a specific use, such as positioning, timing, or enhancing navigation accuracy.

[0036] The signals of the Beidou navigation system are mainly distributed in the B1 band, B2 band, and B3 band. Specifically, the frequency range of the B1 band is 1559.052 MHz - 1563.144 MHz (the center frequency is 1575.42 MHz). Generally, the timing function is implemented based on the B1 - band signal. At the same time, the B1 signal is suitable for a wide range of daily use; the B2 signal performs better in high - dynamic scenarios (such as aviation and drones); the B3 signal provides additional enhancement functions and is suitable for high - precision and high - security requirements.

[0037] Different from general timing technologies, considering the different influences of signals in different frequency bands during propagation in the atmosphere, the method of the present invention not only uses the B1 band for timing, but combines the multi - band signals with weights to calculate the comprehensive signal characteristics, and performs timing calculation after excluding interference to obtain accurate time. The combined use can effectively reduce the multipath effect and ionospheric interference, and at the same time can provide more signal redundancy and choices, improving the anti - interference ability of the signal.

[0038] Furthermore, the noise parameter includes recording the noise parameter during the no - signal transmission period. The no - signal transmission period is judged based on the signal power. Specifically, during the signal reception process, the power value of the received signal is calculated in real - time, expressed as:

[0039]

[0040] where \(P(t)\) represents the power value of the received signal; \(T\) represents the sampling window time length; \(x(t)\) represents the received signal; compare \(P(t)\) with the noise reference power \(P\) noise If \(P(t)\) is close to \(P\) noise and within the preset threshold range, it is determined as the no - signal transmission period.

[0041] Calculate the noise power according to the recorded noise parameter, expressed as:

[0042]

[0043] where \(P\) noise,i represents the noise power on frequency band \(i\); \(i\) represents the frequency - band index; \(T\) represents the sampling time; \(n\) i (t) represents the noise signal intensity of frequency band \(i\) at time \(t\).

[0044] It should be noted that the noise parameters do not necessarily have to be collected only during the signal - free transmission period. However, collecting noise parameters during the signal - free transmission period is the most direct and accurate method because the signal collected at this time is completely composed of environmental noise and does not contain the components of the main signal or interference signal. In a dynamic environment or when it is impossible to determine the signal - free period, the noise parameters can also be estimated during the signal transmission period through signal separation or statistical methods.

[0045] S2: Extract the time - frequency domain features of the multi - band signal, perform weighted fusion on the extracted time - frequency features, and obtain the comprehensive feature vector.

[0046] Furthermore, the time - frequency domain feature extraction includes filtering, denoising, and normalizing the multi - band signal, and extracting the time - domain features of the signal in the time domain, including the phase shift rate, power fluctuation, and time - delay characteristic. Specifically, the phase shift rate reflects the speed of phase change. The instantaneous phase of the received Beidou signal is extracted, and using the phase shift rate formula:

[0047]

[0048] where φ(t) represents the instantaneous phase; ω(t) represents the instantaneous angular frequency.

[0049] The change in signal power reflects the stability of signal strength. Calculate the instantaneous power P(t) of the signal, perform a moving average on the instantaneous power, and obtain the power fluctuation:

[0050]

[0051] where represents the power fluctuation. The time - delay characteristic is obtained by measuring the pseudorange of the signal and comparing it with the true signal pseudorange.

[0052] Even further, using the fast Fourier transform, map the signal from the time domain to the frequency domain and extract the frequency - domain features. In this embodiment, the frequency - domain features include the main frequency position, bandwidth symmetry, and power spectral density. Perform a fast Fourier transform on the signal to obtain the spectrum:

[0053]

[0054] The main frequency position is expressed as:

[0055] f peak =argmax f |X(f)|

[0056] where X(f) represents the spectrum; x(t) represents the signal in the time domain; f represents the frequency variable; e -j2πft represents the kernel function of the Fourier transform; f peakIndicates the main frequency position; argmax f Indicates taking the parameter that makes f reach the maximum value.

[0057] The calculation of bandwidth symmetry and power spectral density is expressed as:

[0058] B = f high -f low

[0059]

[0060] where B represents bandwidth symmetry; f high represents the highest frequency in the signal spectrum; f low represents the lowest frequency in the signal spectrum; S(f) represents the power spectral density.

[0061] Furthermore, the extracted time-frequency features are weighted and fused to calculate the signal power of each frequency band, which is expressed as:

[0062]

[0063] where P signal,i represents the signal power of frequency band i; x i (t) represents the original received signal of frequency band i.

[0064] Calculate the signal-to-noise ratio SNR using the noise power and signal power i , which is expressed as:

[0065]

[0066] Utilize the characteristic that the signal-to-noise ratio reflects the signal quality to perform normalized calculation of the weights to ensure that the value range of w i is between 0 and 1. The normalized weights can intuitively reflect the contribution ratio of the signals in each frequency band to the comprehensive feature vector, which is expressed as:

[0067]

[0068] where w i represents the weighting coefficient of frequency band i; represents the sum of the signal-to-noise ratios of all frequency bands.

[0069] It should be noted that SNR i is the ratio of the signal strength to the noise strength, indicating the clarity and anti-interference ability of the signal. The higher the SNR i , the smaller the interference received by the signal in this frequency band, the better the quality, and the higher the credibility of its features.

[0070] The distribution of noise and interference in different scenarios is uneven, and certain frequency bands may be more interfered with in specific scenarios (such as urban environments or high-dynamic scenarios). According to the real-time SNR i Dynamically adjust w i , it can adapt to the current environment, preferentially utilize the signal frequency bands with higher quality, and dynamic weighting avoids the excessive influence of the interfered frequency bands on the comprehensive feature vector, improving the reliability of the system in complex environments.

[0071] According to the weighting coefficient w i Weight the time-domain features and frequency-domain features, expressed as:

[0072] T weighted,i = w i ·T i

[0073] F weighted,i = w i ·F i

[0074] Among them, T weighted,i represents the time-domain weighted feature of frequency band i; T i represents the time-domain feature vector of frequency band i; F weighted,i represents the frequency-domain weighted feature of frequency band i; F i represents the frequency-domain feature vector of frequency band i.

[0075] The time-domain feature T i directly reflects the dynamic behavior of the signal during reception, is suitable for capturing instantaneous interference and abnormal changes, and plays an important role in rapid interference detection and signal authenticity analysis; the frequency-domain feature F i reflects the overall frequency distribution state of the signal, is suitable for detecting forged signals and interference signals in the frequency domain, and the frequency-domain feature has high value for the analysis of long-term interference and spectrum anomalies.

[0076] By weighted fusion of the time-domain and frequency-domain features, the complementary advantages of the two features can be fully utilized to improve the accuracy of abnormal signal detection, expressed as:

[0077]

[0078] Among them, V represents the comprehensive feature vector; α i represents the fusion weight of the time-domain feature; β i represents the fusion weight of the frequency-domain feature.

[0079] α i and β i are adjusted according to the signal-to-noise ratio, and static α i and β ivalue. For example, increase the time-domain feature weight α in a high-speed dynamic scenario i ; increase the frequency-domain feature weight β in a stable environment i , in this embodiment, α i and β i are calculated as follows:

[0080]

[0081] where, Var(T i ) represents the change rate of the time-domain feature. The larger the change rate, the stronger the dynamic characteristics of the signal; Var(F i ) represents the change rate of the frequency-domain feature. The smaller the change rate, the more stable the signal spectrum characteristics.

[0082] It should be noted that w i is dynamically adjusted based on the signal-to-noise ratio, reflecting the importance of signal quality to the feature. α i and β i adjust the contribution ratio of the time-domain and frequency-domain features. This hierarchical design makes full use of the advantages of multi-band signals, and at the same time enhances the adaptability and accuracy of the comprehensive feature vector for anomaly detection and timing output.

[0083] S3: Establish a joint analysis model, detect abnormal signals according to the comprehensive feature vector, perform time resolution on the real signals that pass the detection, and output time information.

[0084] Furthermore, based on the logistic regression model, establish a joint analysis model to perform anomaly detection on the received signals, expressed as:

[0085]

[0086]

[0087] where, P(y = 1|V) represents the probability that the signal is abnormal; w represents the model weight vector, obtained by learning from the training data; e represents the mathematical constant; T represents the transpose matrix; b represents the bias term.

[0088]

[0089] where, N represents the total number of samples, L represents the loss function; y i represents the label of the i-th sample.

[0090] Set the anomaly threshold θ according to historical data, compare P(y = 1|V) with θ, and judge the degree of signal interference. θ can also be set to different ranges or multi-level thresholds. For example, θ1 is 0.4 and θ2 is 0.7. When P ≤ θ1, it is a normal signal; when θ1 < P ≤ θ2, it means the signal is slightly interfered, record and mark the signal as potentially abnormal, but continue to attempt timing calculation; when P > θ2, it means the signal is severely interfered, pause or reject the timing calculation, and optionally trigger an alarm.

[0091] Furthermore, perform time resolution on the signals determined to be normal by the joint analysis model and output high-precision time information. Calculate the pseudorange of the navigation signal and correct the pseudorange according to the clock error information broadcast by the satellite. Consider the ionospheric and tropospheric delays to further correct the pseudorange. According to the pseudorange and satellite position, use the Doppler effect to calculate the time offset between the receiver and the satellite and perform time resolution.

[0092] Integrate the time resolution results of all frequency band signals and use the weighted average method to output the final high-precision time:

[0093]

[0094] Among them, t final represents the finally output time; t receiver,i represents the time resolved by frequency band i.

[0095] This embodiment also provides a Beidou anti-jamming spoofing timing system based on multi-level detection, including an acquisition module that receives multi-frequency band signals of satellites, records noise parameters and calculates noise power during signal-free transmission periods, and uploads them to the analysis module; an analysis module that extracts the time-frequency characteristics of multi-frequency band signals and performs weighted fusion, dynamically adjusts according to the noise parameters by calculating the signal-to-noise ratio, and obtains a comprehensive feature vector; a timing module that establishes a joint analysis model based on a logistic regression model, detects abnormal signals according to the comprehensive feature vector, performs time resolution on the true signals that pass the detection, and outputs time information.

[0096] Embodiment 2, hereinafter, is an embodiment of the present invention, which provides a Beidou anti-jamming spoofing timing method based on multi-level detection. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0097] In order to verify the innovation and advantages of the method of the present invention, comprehensive analysis experiments were carried out on the B1, B2, and B3 frequency band signals of the Beidou navigation system. Use a high-sensitivity receiver to simultaneously receive the B1, B2, and B3 frequency band signals, monitor the signal-free transmission period through signal power, and record the noise parameters during this period. The noise power recorded in the experiment is -115 dBm, corresponding to the background noise level in an urban environment. The experimental data is shown in Table 1.

[0098] Table 1 Experimental data table

[0099]

[0100] As can be seen from the tabular data, the signal-to-noise ratio (SNR) directly reflects the quality of signals in each frequency band. The SNR of the B2 band is the highest (25 dB), so its weight www is the highest (0.5). This indicates that in a high-dynamic scenario, the contribution of the B2 signal is the greatest. Compared with B1 (20 dB) and B3 (23 dB), the B2 signal has significant advantages in anti-interference and enhancing navigation accuracy.

[0101] The time-domain characteristics and frequency-domain characteristics of each frequency band are significantly different. The comprehensive time-domain and frequency-domain characteristics of B2 are 0.8 and 0.9 respectively, both higher than those of other frequency bands. After weighted fusion, the comprehensive feature vector values are 0.68 (time domain) and 0.75 (frequency domain), close to the theoretical optimal value, indicating that the method of the present invention can significantly improve the accuracy of feature fusion.

[0102] After comprehensive processing of multi-band signals, the final time resolution accuracy reaches 15 ns, significantly better than that of using only single-band signals (theoretical value 20 ns), proving the anti-interference effect and the improvement of time accuracy of multi-band weighted fusion in complex environments.

[0103] The method of the present invention effectively reduces the multipath effect and ionospheric interference and improves the anti-interference ability by dynamically weighted fusion of multi-band signal features. Compared with the prior art, the present invention can adjust the weights of signals in each frequency band in real time, give priority to using high-quality signals, and ensure the stability of the timekeeping accuracy in complex environments.

[0104] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. And the aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0105] The logic and / or steps represented in the flowchart or otherwise described herein can be considered, for example, as a definitional sequence of executable instructions for implementing a logical function, and can be embodied specifically in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. As used in this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device.

[0106] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0107] It should be understood that the various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0108] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A Beidou anti-interference and deception timing method based on multi-layer detection, characterized in that: include: Receive multi-band signals from satellites and record noise parameters; Extract time-frequency domain features from multi-band signals, perform weighted fusion on the extracted time-frequency features, and obtain a comprehensive feature vector; Establish a joint analysis model, detect abnormal signals based on comprehensive feature vectors, perform time calculation on real signals that pass the detection, and output time information; The joint analysis model is based on the logistic regression model to establish a joint analysis model to perform anomaly detection on the received signal, which is expressed as: Among them, P(y=1|V) represents the probability that the signal is abnormal; w represents the model weight vector, which is learned through training data; e represents a mathematical constant; T represents the transposed matrix; b represents the bias term; Train the joint analysis model, collect labeled data sets containing normal and abnormal signals, normalize the input features, and use the gradient descent method to optimize the logistic regression loss function: Among them, N represents the total number of samples, L represents the loss function; y i Represents the label of the i-th sample; The multi-band signal includes a B1 frequency band signal, a B2 frequency band signal and a B3 frequency band signal; The noise parameters include recording noise parameters and calculating noise power during a period of no signal transmission, which is expressed as: Among them, P noise,i represents the noise power on frequency band i; i represents the frequency band index; T represents the sampling time; n i (t) represents the noise signal intensity of frequency band i at time t; The time-frequency domain feature extraction includes filtering, denoising and normalizing the multi-band signal, and extracting the time-domain features of the signal in the time domain, including phase shift rate, power fluctuation and time delay characteristics; Using fast Fourier transform, the signal is mapped from the time domain to the frequency domain and the frequency domain features are extracted; The weighted fusion includes calculating the signal power of each frequency band, which is expressed as: Among them, P signal,i represents the signal power of frequency band i; x i (t) represents the original received signal of frequency band i; Calculate the signal-to-noise ratio (SNR) using noise power and signal power i , expressed as: Using the characteristic of signal-to-noise ratio reflecting signal quality, the signal-to-noise ratio is normalized to calculate the weight, which intuitively reflects the contribution ratio of each frequency band signal to the comprehensive feature vector, expressed as: Among them, w i represents the weighting coefficient of frequency band i; Represents the sum of signal-to-noise ratios of all frequency bands; The weighted fusion also includes, according to the weighted coefficient w i The time domain features and frequency domain features are weighted and expressed as: T weighted,i =w i ·T i F weighted,i =w i ·F i Among them, T weighted,i represents the time domain weighted feature of frequency band i; T i represents the time domain feature vector of frequency band i; F weighted,i represents the frequency domain weighted feature of frequency band i; F i represents the frequency domain feature vector of frequency band i; The weighted fusion of time domain and frequency domain features is used to obtain a comprehensive feature vector, which is expressed as: Among them, V represents the comprehensive feature vector; α i represents the fusion weight of time domain features; β i Represents the fusion weight of frequency domain features; α i and β i Dynamic adjustment based on signal-to-noise ratio; Among them, Var(T i ) represents the rate of change of time domain characteristics; Var(F i ) represents the rate of change of frequency domain characteristics; The joint analysis model includes, based on the logistic regression model, using the gradient descent method to optimize the logistic regression loss function, establishing the joint analysis model, performing anomaly detection on the comprehensive feature vector V, and obtaining the anomaly probability P.

2. A system using the Beidou anti-interference and deception timing method based on multi-layer detection as claimed in claim 1, characterized in that: include, The acquisition module receives the multi-band signals from the satellite, records the noise parameters and calculates the noise power during the period without signal transmission, and uploads it to the analysis module; The analysis module extracts the time-frequency characteristics of multi-band signals and performs weighted fusion, calculates the signal-to-noise ratio according to the noise parameters, performs dynamic adjustment, and obtains the comprehensive feature vector; The timing module establishes a joint analysis model based on the logistic regression model, detects abnormal signals according to the comprehensive feature vector, performs time calculation on the real signals that pass the detection, and outputs the time information.

3. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the Beidou anti-interference and deception timing method based on multi-layer detection as described in claim 1 are implemented.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the Beidou anti-interference and deception timing method based on multi-layer detection as described in claim 1 are implemented.

Images