Fractions of fourier transform-based fraud signal detection method and computer device

Abstract

The application relates to the field of deception signal detection, in particular to a deception signal detection method based on fractional Fourier transform and a computer device, which improves the accuracy and reliability of GNSS deception signal detection. coh The method comprises the following steps: acquiring an intermediate frequency signal sample; multiplying the acquired intermediate frequency signal sample with a locally generated carrier and a pseudo code, obtaining a correlation output result after stripping the carrier and the pseudo code; performing N coh point accumulation on the correlation output result to obtain a correlation integral result; processing the correlation integral result by using fractional Fourier transform to obtain output results in a correlation domain and a fractional Fourier transform domain; and determining the deception signal according to the output results in the correlation domain and the fractional Fourier transform domain. The application is suitable for GNSS deception signal detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of spoofing signal detection, and in particular to a spoofing signal detection method based on fractional Fourier transform and a computer device. BACKGROUND

[0002] GNSS (Global Navigation Satellite System) plays an important role in daily life. However, the open GNSS civilian signal structure and extremely low ground receiving power make it easy to be forged and vulnerable to spoofing. GNSS spoofing attacks are stealthy, flexible in strategy, easy to deploy and low in cost, which pose a serious threat to GNSS safety-critical applications. As the first step of anti-spoofing, spoofing detection usually identifies the direct or indirect subtle differences between spoofing signals and real signals to determine whether there is spoofing in the current environment. According to the different detection principles, the existing spoofing detection technologies can be divided into the following categories: spoofing detection methods based on navigation information encryption technology, spoofing detection methods based on spatial information processing technology, spoofing detection methods based on navigation result post-processing technology and spoofing detection methods based on GNSS receiver baseband signal processing technology.

[0003] The spoofing detection methods based on navigation information encryption technology need to modify the signal system at the national level, which is difficult to implement and difficult to achieve in the short term. The spoofing detection methods based on spatial information processing technology need to process data of different epochs, which is complex in operation and requires additional hardware resources. The spoofing detection methods based on navigation result post-processing technology also require additional hardware resources, and the detection effect is limited by the accuracy of motion sensors. The spoofing detection methods based on GNSS receiver baseband signal processing technology are relatively simple to implement, but the single detection method is limited in application scenarios and needs to be combined with multiple methods for detection.

[0004] The most common detection method currently uses a radio frequency front-end monitor with an automatic gain control mechanism to monitor the sudden jump in signal energy when GNSS spoofing occurs to determine the presence of spoofing signals, but this method only works when the spoofing attack is initiated. The correlation domain detection method that monitors the distortion of the correlation output also has similar limitations. If the spoofing attack is initiated before the GNSS receiver is powered on, the distortion of the coherent peak will not be monitored, and this method will fail. The correlation domain multi-peak detection algorithm based on the acquisition result does not have the detection capability for the retransmission type and high-end generation type spoofing interference with low signal-to-noise ratio, small delay and Doppler frequency difference.

[0005] In summary, the current spoofing detection technology has problems such as great implementation difficulty, high use cost, need for additional hardware resources, poor universality, weak generalization ability, limited action window, and the like. Moreover, with the diversification of spoofing attack forms and strategies, the existing spoofing detection technology has an unsatisfactory detection effect when dealing with medium and high-level spoofing strategies with good concealment. Therefore, how to ensure the accuracy, reliability and scene adaptability of the spoofing detection method is an important challenge currently faced. SUMMARY

[0006] The present application aims to overcome the shortcomings of the prior art and provide a spoofing signal detection method based on fractional Fourier transform and a computer device, which improves the accuracy and reliability of GNSS spoofing signal detection.

[0007] The present application achieves the above-mentioned purpose by adopting the following technical solutions. In a first aspect, the present application provides a spoofing signal detection method based on fractional Fourier transform, comprising:

[0008] S1, acquiring intermediate frequency signal samples;

[0009] S2, multiplying the acquired intermediate frequency signal samples with a locally generated carrier and a pseudo code to obtain a correlation output result after stripping the carrier and the pseudo code;

[0010] S3, performing N coh point accumulation on the correlation output result to obtain a correlation integral result;

[0011] S4, processing the correlation integral result using fractional Fourier transform to obtain an output result in the correlation domain and the fractional Fourier transform domain;

[0012] S5, determining spoofing signals according to the output result in the correlation domain and the fractional Fourier transform domain;

[0013] Peak value extraction is performed from the correlation domain result. If the peak value energy exceeds a set interference threshold, it is directly determined that there is a spoofing signal. If the interference threshold is not exceeded, it is determined whether the peak value energy exceeds an acquisition threshold. If the acquisition threshold is not exceeded, it indicates that there is no satellite signal corresponding to the PRN code, and the detection ends. If the peak value exceeds the acquisition threshold, it is determined whether there are two correlation peaks. If there are two correlation peaks, it is determined that there is a spoofing signal. If there is only one correlation peak, it is determined whether there is one peak in the fractional Fourier transform domain. If there is only one peak in the fractional Fourier transform domain, it is determined that there is no spoofing signal. If there are two peaks, it is determined that there is spoofing interference, and the detection ends.

[0014] Further, step S1 specifically comprises:

[0015] In a spoofing scenario, the signal y total (t) received by the receiver antenna is:

[0016] y total (t) = y(t) + y s (t) + v(t), v(t) represents the received noise, y(t) represents the real signal;

[0017] N is the number of satellites, A i is the real signal amplitude of the i-th satellite, D i (t) is the navigation message bit stream of the i-th satellite signal, C i (t) is the corresponding spreading code sequence, τ i (t) is the code phase corresponding to the real signal of the i-th satellite, φ i (t) is the carrier phase corresponding to the real signal, ω c is the nominal carrier angular frequency;

[0018] y S (t) represents the spoofing signal;

[0019] N s = N, that is, the number of satellite signals contained in the spoofing signal is equal to the corresponding number of satellite signals in the real signal, and each spoofing signal has the same spreading code C i (t) and the estimated navigation message bit stream A si is the spoofing signal amplitude of the i-th satellite, τ si (t) is the code phase corresponding to the spoofing signal of the i-th satellite, φsi(t) is the carrier phase corresponding to the spoofing signal;

[0020] The signal received by the receiver antenna is simplified, and the frequency variation is extracted from the phase variation, and then:

[0021]

[0022] In the formula, f c is the nominal frequency, f d is the Doppler frequency;

[0023] After the signal with the extracted frequency variation is down-converted and AD-sampled with t n as the period, the intermediate frequency signal sample is obtained:

[0024]

[0025] In the formula, A IF is the intermediate frequency signal sample amplitude, f IF is the frequency of the intermediate frequency signal sample, v IF (it n) is a noise component in the intermediate frequency signal sample.

[0026] Further, the correlation output result is:

[0027]

[0028] wherein f and f s respectively represent a real signal frequency difference and a spoofing signal frequency difference, and respectively represent a real signal phase difference and a spoofing signal phase difference.

[0029] Further, the correlation integral result is:

[0030]

[0031] wherein m represents the mth correlation integral result, T coh is a coherent integration time, v coh [m] is a noise item after integration,

[0032] Further, the step S4 specifically comprises:

[0033] The fractional Fourier transform is: K p (t, u) represents a transform kernel of the fractional Fourier transform;

[0034]

[0035] p represents an order of the fractional Fourier transform, which is an arbitrary real number and

[0036] The fractional Fourier transform result is obtained as:

[0037]

[0038] wherein

[0039] In a second aspect, the present application provides a computer device, comprising a memory, wherein the memory stores program instructions, and the program instructions are executed to perform the above-mentioned spoofing signal detection method based on the fractional Fourier transform.

[0040] The present application has the following beneficial effects:

[0041] The application discards the dependence on signal system changes such as encryption of navigation information, and cuts in from the perspective of GNSS baseband signal processing, so that effective deception detection can be realized under the existing signal structure. This design greatly reduces the difficulty and time cost of technical implementation, making the method more easily deployed and applied in existing GNSS systems.

[0042] The application optimizes the baseband signal processing algorithm of the GNSS receiver, avoiding the dependence on additional hardware resources. Compared with the existing complex spatial information processing and navigation result post-processing method, the implementation of the application is more simple, reduces power consumption and use cost, has wide application potential, and is especially suitable for single-antenna receiver application scenarios with limited power consumption and resources.

[0043] The application significantly enhances the feature extraction and analysis capability of the deception signal by introducing fractional Fourier transform. Not only can the deception signal features hidden in the traditional time domain and correlation domain signals be detected, but also the medium and high-end hidden deception signal sources can be accurately identified through the extended features of the time-frequency domain, thereby greatly improving the robustness and accuracy of the detection.

[0044] The method of the application can extract the dynamic characteristics of the signal and generate a time-frequency matrix or image, effectively expanding the application scenarios and time windows of the detection. Whether the deception attack occurs before or after the receiver is powered on, the application can maintain effective detection capability, ensuring accuracy and reliability in different attack times and scenarios.

[0045] The method proposed by the application is highly adaptable and can be applied to various complex and diversified deception attack strategies. Not only does it enhance the complement of existing detection technologies, but it also has a wider application prospect and can provide more efficient and reliable protection schemes for GNSS safety-critical applications. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is a GNSS deception detection conceptual diagram provided by an embodiment of the application;

[0047] Figure 2 is a GNSS deception detection signal processing block diagram provided by an embodiment of the application;

[0048] Figure 3 is a GNSS deception detection signal processing flowchart provided by an embodiment of the application;

[0049] Figure 4 is a deception interference detection strategy flowchart provided by an embodiment of the application. DETAILED DESCRIPTION

[0050] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0051] The spoofing scenario and the basic spoofing detection concept proposed in the present application are shown in Figure 1 In a typical spoofing attack scenario, the real GNSS signal and the spoofing attack signal exist at the same time, so the spoofing can be identified by finding the subtle relationship between them. It is very important for the target receiver to obtain the ability to distinguish the real signal from the spoofing signal. As shown in the figure, the spoofing signal source first receives the real GNSS signal and analyzes it, and generates a spoofing signal according to the satellite parameters at that time. For the victim receiver, the real signal received from the GNSS satellite is dynamically determined by the relative motion between the receiver and the satellite. On the other hand, the signal received from the spoofing signal source includes two parts: the relative motion between the spoofing signal source and the satellite and the relative motion between the spoofing signal source and the victim receiver. In daily civilian scenarios, the second part of the dynamic is unknown to the spoofing attacker. In addition, the spoofing signal source usually needs to induce the receiver to a false position or induce it to make unreasonable maneuvers, which will introduce additional dynamics in the spoofing signal. Therefore, we can use this characteristic to extract the GNSS signal dynamic feature by the method of fractional Fourier transform, and complete the spoofing signal detection by distinguishing the difference in dynamic characteristics between the real signal and the spoofing signal.

[0052] The signal processing block diagram of the GNSS spoofing detection based on the fractional Fourier transform is shown in Figure 2 The phase discriminator and loop filter in the traditional GNSS receiver are replaced by the FrFT (Fractional Fourier Transform) and the corresponding parameter estimation method. The baseband processing algorithm based on the fractional Fourier transform extends the traditional two-dimensional correlation domain method to three dimensions, and obtains the output results of the correlation domain and the FrFT domain. Compared with the correlation domain, the FrFT domain result has better resolution and contains potential signal dynamic information, and has excellent resolution capability for spoofing signals with low signal-to-noise ratio, small delay and Doppler frequency difference. The two domain results are then sent to the spoofing detection module for decision judgment, and the original observation information output by the baseband signal processing module can also be used as a detection feature for auxiliary judgment basis.

[0053] The present application provides a spoofing signal detection method based on the fractional Fourier transform, as shown in Figure 3 , comprising:

[0054] S1, acquiring intermediate frequency signal samples;

[0055] In a deception scenario, the signal y received by the receiver antenna total (t) is:

[0056] y total (t)=y(t)+y s y(t)+ν(t), where ν(t) represents the received noise and y(t) represents the real signal;

[0057] N is the number of satellites, A i Let D be the actual signal amplitude of the i-th satellite. i (t) represents the bit stream of the navigation message from the i-th satellite, C i (t) represents the corresponding spreading code sequence, τ i (t) is the code phase corresponding to the actual signal from the i-th satellite, φ i (t) represents the carrier phase corresponding to the real signal, ω c It is the nominal carrier angular frequency;

[0058] y S (t) represents a deception signal;

[0059] N s =N, meaning the number of satellite signals contained in the deception signal is equal to the corresponding number of satellite signals in the real signal, and each deception signal has the same spreading code C as the real signal. i (t) and the estimated navigation message bit stream A si Let τ be the amplitude of the spoofing signal from the i-th satellite. si φ(t) represents the code phase corresponding to the spoofing signal of the i-th satellite, and φsi(t) represents the carrier phase corresponding to the spoofing signal.

[0060] Simplifying the signal received by the receiver antenna and extracting the frequency change from the phase change, we get:

[0061]

[0062] In the formula, f c f is the nominal frequency. d The frequency is the Doppler frequency.

[0063] After down-converting the signal from which the frequency change was extracted, and then using t n By performing periodic A / D sampling, intermediate frequency signal samples are obtained:

[0064]

[0065] In the formula, A IF f is the amplitude of the intermediate frequency signal sample. IF ν is the frequency of the intermediate frequency signal sample.IF (it n ) is the noise component in the intermediate frequency signal sample.

[0066] S2, multiplying the obtained intermediate frequency signal sample with the locally generated carrier and pseudo code, and obtaining a correlation output result after stripping the carrier and pseudo code;

[0067] The correlation output result is:

[0068]

[0069] In the formula, f and f s respectively represent the real signal frequency difference and the spoofing signal frequency difference, and respectively represent the real signal phase difference and the spoofing signal phase difference.

[0070] S3, accumulating the correlation output result by N coh sampling points to obtain a correlation integral result;

[0071] The correlation integral result is:

[0072]

[0073] In the formula, m represents the mth correlation integral result, T coh is the coherent integration time, v coh [m] is the noise term after integration,

[0074] S4, processing the correlation integral result using fractional Fourier transform to obtain an output result in the correlation domain and the fractional Fourier transform domain;

[0075] The fractional Fourier transform is: K p (t, u) represents the transform kernel of the fractional Fourier transform;

[0076]

[0077] p represents the order of the fractional Fourier transform, which is any real number and

[0078] The fractional Fourier transform result is:

[0079]

[0080] Wherein

[0081] S5, determining the spoofing signal according to the output result in the correlation domain and the fractional Fourier transform domain;

[0082] The judgment process is as follows: Figure 4 As shown, the process begins by extracting peak values ​​from the correlation domain results. If the peak energy exceeds a set interference threshold, a spoofing signal is directly identified. If it does not exceed the interference threshold, the peak energy is then checked against the acquisition threshold. If it does not exceed the acquisition threshold, the satellite signal corresponding to the PRN code is not present, and the detection ends. If the peak exceeds the acquisition threshold, the number of correlated peaks is further determined. If two correlated peaks exist, a spoofing signal is identified. If only one correlated peak exists, the number of peaks in the FrFT domain is further determined. If only one peak exists in the FrFT domain, the possibility of a spoofing signal can be ruled out. If two peaks exist, spoofing interference is identified, and the detection ends.

[0083] This invention also provides a deception signal detection system, which mainly consists of the following parts:

[0084] GNSS receiver RF front-end module: used to receive GNSS satellite signals, including antenna, low noise amplifier, filter and automatic gain control unit, etc.

[0085] Baseband signal processing module: Receives and processes signals from the front-end module. This module integrates a fractional Fourier transform algorithm for time-frequency domain signal feature extraction while also enabling traditional signal acquisition and tracking functions. In baseband signal processing, this invention employs the fractional Fourier transform to expand the feature space of traditional methods. The fractional Fourier transform is a mathematical transform tool that provides a continuous transition between the time and frequency domains; by adjusting the transform order, richer signal features can be extracted in different transform domains.

[0086] Spoofing signal detection module: Analyzes the correlation and time-frequency feature outputs extracted by the baseband signal processing module to determine whether there are any abnormalities in the signal, and then detects the presence of spoofing signals.

[0087] This invention, starting from the perspective of GNSS baseband signal processing, does not rely on changes to the signal system and can achieve effective deception detection under the existing signal structure, thereby reducing the difficulty of technical implementation and the expected time.

[0088] Secondly, existing deception detection methods based on spatial information processing and navigation result post-processing are computationally complex, require additional hardware resources, and their detection effectiveness is limited by the accuracy of motion sensors. Furthermore, these post-processing methods are power-consuming and costly, making them unsuitable for widespread adoption. This invention, however, modifies the GNSS baseband signal processing algorithm from within the GNSS receiver itself. It requires no additional hardware resources, is low-cost, easy to implement, and has wide applicability.

[0089] Finally, the existing time domain and correlation domain detection methods based on GNSS receiver baseband signal processing rely on limited parameter types, which perform well when facing low-end GNSS spoofing signal sources, but when dealing with more concealed mid-to-high-end signal sources, the effect is greatly reduced. At the same time, the existing methods also have problems such as limited action window and application scenarios. The present application innovatively introduces fractional Fourier transform into the GNSS receiver baseband signal processing algorithm, improves and enhances the performance of spoofing signal feature analysis and extraction, while retaining the time domain and correlation domain signal features, it also expands the signal feature parameters in the time-frequency domain.

[0090] Moreover, the results of fractional Fourier transform can not only extract the Doppler frequency rate parameter in GNSS signals, but also because of the existence of Doppler frequency rate, the signal energy will be concentrated in a certain rotation angle of FrFT, so it has better Doppler frequency resolution in the FrFT domain. These characteristics make the spoofing signal and the real signal show more significant differences in the FrFT domain, thereby improving the sensitivity of spoofing signal detection.

[0091] In addition, the method proposed in the present application extracts signal dynamic characteristic information from the received signal and generates corresponding correlation domain matrix and time-frequency domain matrix respectively, and sends them to the spoofing signal detection module in the form of digital matrix or image, and judges whether there is a spoofing signal based on the two results by means of peak search or target detection method. The introduction of new features greatly expands the application scenarios and action windows of traditional methods, ensuring accurate and reliable spoofing signal detection at different attack initiation time points.

[0092] In summary, the present application significantly improves the accuracy, reliability and adaptability of spoofing signal detection while reducing the implementation difficulty and cost, and provides a new technical solution for the safe application of GNSS.

[0093] The above description is only the preferred embodiment of the present application, it should be understood that the present application is not limited to the form disclosed herein, should not be regarded as excluding other embodiments, and can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein, by the above-mentioned teaching or related art or knowledge. The modifications and changes made by those skilled in the art without departing from the spirit and scope of the present application shall be within the scope of protection of the appended claims of the present application.

Claims

1. A method of spoof signal detection based on fractional Fourier transform characterized by, The method comprises the following steps: S1, acquiring intermediate frequency signal samples; S2, multiplying the acquired intermediate frequency signal samples with a locally generated carrier and pseudo code, and obtaining a correlation output result after stripping the carrier and pseudo code; S3, N coh Sample points are accumulated to obtain correlation integral results. S4, processing the correlation integral result using a fractional Fourier transform to obtain an output result in a correlation domain and a fractional Fourier transform domain; S5, determining a spoofing signal according to the output result in the correlation domain and the fractional Fourier transform domain; Peak values are extracted from the correlation domain result. If the peak value energy exceeds a set interference threshold, it is directly determined that a spoofing signal exists. If the interference threshold is not exceeded, it is determined whether the peak value energy exceeds a capture threshold. If the capture threshold is not exceeded, it indicates that a satellite signal corresponding to a PRN code does not exist, and the detection ends. If the peak value exceeds the capture threshold, it is determined whether the number of correlation peaks exists. If two correlation peaks exist, it is determined that a spoofing signal exists. If only one correlation peak exists, it is determined whether the number of peaks in the fractional Fourier transform domain exists. If there is only one peak in the fractional Fourier transform domain, it is determined that a spoofing signal does not exist. If two peaks exist, it is determined that a spoofing interference exists, and the detection ends.

2. The fractional Fourier transform based spoofing signal detection method of claim 1, wherein, Step S1 specifically comprises: In a spoofing scenario, the signal y received by the receiver antennas total (t) is: y total (t) = y(t) + y s (t) + v(t), v(t) representing reception noise, y(t) representing the real signal; N is the number of satellites, A i D is the true signal amplitude of the ith satellite, D i C is the navigation message bit stream of the ith satellite, C i τ is the corresponding spreading code sequence, τ i φ is the code phase of the ith satellite true signal, φ i ω is the carrier phase of the true signal, ω c is the nominal carrier angular frequency; y S (t) indicates a spoofing signal; N s = N, i.e. the number of satellite signals contained in the spoofing signals equals the number of corresponding satellite signals in the real signals, and each spoofing signal has the same spreading code C i (t) and the estimated navigation message bit stream A si is the spoofing signal amplitude for the i-th satellite, τ si (t) is the code phase corresponding to the i-th satellite spoofing signal, and φsi(t) is the carrier phase corresponding to the spoofing signal. The signal received by the receiver antenna is simplified, and the frequency variation is extracted from the phase variation, and then the following equation is obtained: where f c is the nominal frequency, f d is the Doppler frequency; The signal of the extracted frequency variation is down-converted and AD sampled with t n as a period, and the intermediate frequency signal sample is obtained: where A IF is the amplitude of the intermediate frequency signal sample, f IF is the frequency of the intermediate frequency signal sample, v IF (it n ) is the noise component in the intermediate frequency signal sample.

3. The fractional Fourier transform based spoofing signal detection method of claim 2, wherein, The correlation output result is: wherein f and f s respectively represent the real signal frequency difference and the spoof signal frequency difference, and respectively represent the real signal phase difference and the spoof signal phase difference.

4. The fractional Fourier transform based spoofing signal detection method of claim 3, wherein, The correlation integral result is: where m represents the mth correlation integration result, T coh is the coherent integration time, v coh [m] is the noise term after integration, 5. The fractional Fourier transform based spoofing signal detection method of claim 4, wherein, Step S4 specifically comprises: The fractional Fourier transform is: K p (t, u) denotes the transform kernel of the fractional Fourier transform; p represents the order of the fractional Fourier transform, which is any real number and The fractional Fourier transform result is obtained as: wherein 6. A computer apparatus comprising a memory, said memory storing program instructions, characterised in that, When the program instructions are executed, the method for detecting a spoofing signal based on a fractional Fourier transform is executed, as claimed in any one of claims 1-5.

Images