A method for identification and suppression of multiple simultaneous satellite navigation spoofing jamming

Abstract

This invention discloses a method for identifying and suppressing multiple synchronous satellite navigation spoofing interferences, comprising: constructing a denoised cross-correlation matrix using received data vectors and reference data vectors; calculating the spatial power spectrum using an observation function based on the constructed denoised cross-correlation matrix; adaptively classifying the estimated power and direction of arrival of the observed pairs of signals using a maximum entropy threshold based on the spatial power spectrum to identify the power and direction of arrival of the incident navigation signal; reconstructing the signal autocorrelation matrix and the interference plus noise covariance matrix based on the identification results; calculating the optimal vector based on the signal autocorrelation matrix and the interference plus noise covariance matrix; and then suppressing the spatial spoofing interference based on the optimal vector. This method identifies and suppresses multiple synchronous satellite navigation spoofing interferences.

Description

Technical Field

[0001] This invention relates to a method for identification and suppression, specifically a method for identifying and suppressing multiple synchronous satellite navigation deception interferences. Background Technology

[0002] Currently, multiple self-synchronizing deception jammers are sending false signals from different directions in concert, dragging the tracking loop of the target receiver and causing the navigation receiver to veer off course without its knowledge. To counter this coordinated deception jamming, the receiver needs to be able to identify the directions from which the enemy is sending false signals, enabling it to detect the deception signals as early as possible during the synchronization process, distinguish between genuine signals and deception jamming in the direction of arrival, and effectively suppress the deception jamming before it is dragged, ensuring that the receiver is correctly synchronized with the real satellite navigation signal.

[0003] Numerous scholars and research institutions both domestically and internationally have conducted research on the detection, classification, identification, and suppression of deception interference. Currently, it is generally categorized into five types: navigation signal encryption, advanced receiver autonomous integrity detection using additional signal characteristics, adding navigation message authentication bits, and reconstruction methods based on signal data vectors that utilize the spatial geometric differences between deception and real signals. (Publicly available materials include JOVANOVIC A, BOTTERON C, ...) P A. Multi-testdetection and protection algorithm against spoofing attacks on GNSS receivers [EB / OL]) proposes a spoofing detection scheme based on multiple statistical characteristics of navigation signals. This scheme detects the presence of spoofing signals based on carrier-to-noise ratio (C / NO) estimation, Doppler frequency shift, and the consistency of information such as position, velocity, and time. Public information (BROUMANDAN A, JAFARNIA-JAHROMI A, LACHAPELLE G. Spoofing detection, classification and cancelation (SDCC) receiver architecture for a moving GNSS receiver [J]. GPS Solutions, 2015, 19(3): 475-487. doi: 10.1007 / s10291-014-0407-3.) proposes a mobile receiver consisting of spoofing detection, real / spoofing signal classification, and spoofing suppression, which detects and suppresses spoofing signals from the acquisition layer, tracking layer, and positioning layer. However, this method requires adjustments to the receiver's internal structure, which involves significant expertise and technical difficulty, and also incurs substantial promotion costs.

[0004] Utilizing antenna arrays that capture the spatial geometric differences between deceptive interference and real signals allows for front-end detection, identification, and elimination of interference without requiring receiver modifications, offering significant advantages. Early multi-antenna detection typically considered the scenario where multiple false signals emitted simultaneously from the same location had roughly the same direction of arrival. The paper (DANESHMAND S, JAFARNIA-JAHROMI A, BROUMANDAN A, et al. GNSS spoofing mitigation in multipath environments using space-time processing [EB / OL]) proposed a method for anti-spoofing using antenna arrays in multipath environments, estimating the spoofing channel coefficients using spatial and temporal processing methods. This method is executed before despreading, has low algorithmic complexity, and does not require array calibration. However, these methods all assume that the spoofing interference is incident from a single direction with significantly higher power than the real signal. Self-synchronizing spoofing interference was first proposed by T. Humphreys' team, transmitting false signals with code phases and carrier phases similar to those received at the receiver antenna. The initial power of the spoofing signal is low, but the transmission power gradually increases during the acquisition and tracking loop process. Finally, upon synchronization, the code delay of the spoofing signal is altered, successfully dragging away the correlation peak. During this receiver synchronization process, the spoofing slowly increases its power until it exceeds the real signal by 1.5-2 dB, synchronizing the receiver with the spoofing interference upon acquisition. It can be seen that this type of interference can easily synchronize the receiver with only a slight power advantage. The mobile receiver spoofing interference detection method based on power variation considers the difference in signal power changes between the real signal and the spoofing signal when the terminal moves at the same distance, achieving effective detection of near-range spoofing interference sources. However, relying solely on a single signal feature is insufficient for effective identification of this type of spoofing. Combining the direction of arrival information for classifying spoofing interference and the real signal has higher reliability and provides crucial incident angle information necessary for interference suppression. The document (G Fan, X Gan, B Yu, Q Rong, C Sheng. Adaptive Spoofing Suppression Algorithm for GNSS Based on Multiple Antennas Array[J]. Sensors, 2020, 20(4): 1-19. doi:10.3390 / s20041115) uses array antenna reception to estimate DOA through despreading and the MUSIC method, and then performs interference suppression. However, it does not address the selection principles of the DOA prior information required for interference suppression, nor does it clearly explain how to distinguish the incident direction of the real signal from the deceptive interference.The paper (Y Zhao,FShen,G Xu,G Wang.A Spatial-Temporal Approach Based on Antenna Array for GNSS Anti-Spoofing[J].Sensors,2021,21(3):1-21.doi:1O.339O / s2103O929.) proposes a method to distinguish whether the incident wave is a real signal, a weak deception interference, or a multipath from a single direction by using joint power estimation and the number of correlation peaks. However, this method only examines low-power deception interference incident from a single direction and cannot be applied to cooperative deception interference from multiple directions. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for identifying and suppressing multiple synchronous satellite navigation spoofing interferences.

[0006] To achieve the above objectives, the method for identifying and suppressing multiple synchronous satellite navigation spoofing interferences described in this invention includes:

[0007] A denoising cross-correlation matrix is ​​constructed using the received data vector and the reference data vector;

[0008] Based on the denoised cross-correlation matrix obtained by construction, the spatial power spectrum is calculated through the observation function;

[0009] Based on the spatial power spectrum, the power and direction of arrival of the observed pairs of signals are adaptively classified using the maximum entropy threshold to identify the power and direction of arrival of the incident navigation signal.

[0010] Based on the identification results, the signal autocorrelation matrix and the interference plus noise covariance matrix are reconstructed. The optimal vector is calculated based on the signal autocorrelation matrix and the interference plus noise covariance matrix. Then, the spatial deception interference is suppressed based on the optimal vector.

[0011] The cross-correlation matrix of the noise is:

[0012]

[0013] Among them, the t-th observation from θ u Navigation signal power σ in direction 2 (θ u ), u=1,2,...,U, where U is the number of incident signals, A=[A au A sp A au =[a(θ1),a(θ2),…a(θ)] K [A] is the steering vector matrix of the M×K dimensional real signal.sp =[a(θ K+1 ),a(θ k+2 ),…a(θ U )] is the M×R dimensional deception interference steering vector matrix.

[0014] The spatial power spectrum is:

[0015]

[0016] in, Let θ represent the direction of incoming wave that maximizes the function observed in the t-th observation. + (θ)=a H (θ)(a H (θ)a(θ)) -1 , where a(θ) is an M×1 dimensional guiding vector.

[0017] The optimal weight vector is calculated according to equation (18), where,

[0018]

[0019] Where Γ{} is the eigenvector corresponding to the largest eigenvalue of the matrix. The self-conversion matrix for the array to receive the real signal. To reconstruct the interference plus noise covariance moment.

[0020] Reconstruct the interference plus noise covariance moment for:

[0021]

[0022] Among them, C sp Let A(Θsp) be the diagonal matrix representing the power estimate of the deception interference, and let A(Θsp) be the steering matrix of the deception interference obtained based on the estimated direction of the incoming wave and the known array configuration. It is noise.

[0023] noise for:

[0024]

[0025] in, Let tr{} be the orthogonal projection matrix of the steering matrix A(Θ) of U signals, and let tr{} denote the trace of the matrix.

[0026] The array receives the self-inverting matrix of the real signal. for:

[0027]

[0028] Among them, Cau The U received signals are classified into two categories based on their power according to the maximum entropy threshold δ.

[0029] The SINR corresponding to the optimal weight vector is:

[0030]

[0031] Where e1 is the eigenvector corresponding to the largest eigenvalue, Cau is a diagonal matrix composed of the estimated true signal power values, and C sp A(Θ) is a diagonal matrix composed of estimated interference power values ​​used to deceive the system. au A(Θsp) is the steering matrix using the estimated direction of the real signal arrival, and A(Θsp) is the steering matrix using the estimated direction of the deceptive signal arrival.

[0032] The present invention has the following beneficial effects:

[0033] The method for identifying and suppressing multiple synchronous satellite navigation deception interferences described in this invention, in specific operation, uses the maximum entropy threshold to adaptively classify the estimated power and direction of arrival of the observed pairs of signals to identify the power and direction of arrival of the incident navigation signal. Based on the identification results, the signal autocorrelation matrix and the interference plus noise covariance matrix are reconstructed. The optimal vector is calculated based on the signal autocorrelation matrix and the interference plus noise covariance matrix, and then the spatial deception interference is suppressed based on the optimal vector to eliminate the influence of deception on the receiver. Experiments have verified that this invention can identify and suppress synchronous deception interferences with multiple different incident directions from the incident direction, even if the deception interference is more than 1 dB higher than the real signal. Attached Figure Description

[0034] Figure 1 Spatial power spectrum diagram;

[0035] Figure 2 A schematic diagram of spatial power-DOA observation results and maximum entropy adaptive threshold;

[0036] Figure 3 The result of monitoring power in the 50s spatial power spectrum;

[0037] Figure 4 Beam pattern for deception and interference suppression;

[0038] Figure 5 The output SINR results are shown for different input SNR values.

[0039] Figure 6a A comparison of the results of suppressing correlation between the code before and after deception interference;

[0040] Figure 6b A comparison chart of the correlation results of normalized codes before and after deception interference suppression; Detailed Implementation

[0041] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, not all embodiments, and are not intended to limit the scope of the present invention. Furthermore, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion regarding the concepts disclosed in the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0042] The accompanying drawings show structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not drawn to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0043] The method for identifying and suppressing multiple synchronous satellite navigation spoofing interferences described in this invention includes the following steps:

[0044] 1) Establish a signal model

[0045] Against a backdrop of multiple synchronous spoofing interferences, the array receiver front end receives in real time visible satellite signals and multiple synchronous spoofing interferences from different directions, as well as statistically independent Gaussian white noise. It is assumed that all signals are stationary within the observation time slot. M uniformly spaced linear arrays are used for reception, and the antenna array at the receiver front end is a uniform linear array composed of M elements. K far-field narrowband incoherent real satellite navigation signals are received from (θ1, θ2, ..., θ...). K An incident signal is emitted into the array, and R uncorrelated far-field deception interferences originate from (θ). K+1 ,θ K+2 ,…,θ U An incident signal is emitted into the array, and the array receives a signal vector comprising the true signal vector, a deception interference vector, and Gaussian white noise, wherein the array receives a signal vector X(n) as follows:

[0046] X(n)=A au S(n)+A sp W(n)+V(n) (1)

[0047] Where S(n)=[s1(n),s2(n),…,s K (n)]T The true navigation signal matrix is ​​K×1 dimensional; W(n)=[w1(t),w2(t),…,w R (t)] T Let V(n) be an R×1 dimensional deception interference matrix; V(n) = [v1(n), v2(t), ..., v M (t)] T Let M×1 be a noise vector with a mean of 0 and a variance of Additive white Gaussian noise, independent of the signal; A au =[a(θ1),a(θ2),…a(θ)] K [A] is the steering vector matrix of the M×K dimensional real signal. sp =[a(θ K+1 ),a(θ k+2 ),…a(θ U Let θ be an M×R dimensional steering vector matrix for deception interference, and let a(θ) be an M×1 dimensional steering vector, where the phase delay of the m-th element relative to the reference element is α. m (θ) is:

[0048]

[0049] Where, d m The position of the m-th array element relative to the reference point is represented by ω, where ω is the center angular frequency of the narrowband signal, c is the speed of light, and s is the i-th real navigation signal. i (n)=σ i (n)b(n)c i (n), σ(n) represents the actual signal amplitude of the nth snapshot, c i (n) represents the Pseudo Random Noise (PRN) chip sequence of the i = 1, 2, ..., K satellite, which contains 20 groups of C / A codes of length 1023 in 1 ms. i b(n) represents navigation data bits, and the deception jamming transmits a false j-th synchronization deception J(n) = σ j ′(n)b′(n)c j (n-τ), where σ j (n) represents the deception interference amplitude of the nth snapshot, c j (n-τ) is the PRN sequence sent by the j-th spoofing interference, delayed by τ chips, b′(n) is the fake data bit sent by the spoofing interference, and V(n) is the additive white Gaussian noise vector.

[0050] Assume there may be a correlation between the real signal and the spoofing interference, while the noise is independent of the desired GPS signal and the spoofing interference. Let A = [A au A spThen the covariance matrix R of the array received data vector is:

[0051]

[0052] Wherein, the diagonal element σ 2 (θ1), σ 2 (θ2), ...,σ 2 (θ U The power values ​​of the U incident signals are respectively, where, Let ρ be the variance of the noise. When the spoofing jammer begins transmitting false signals but is uncorrelated with the real signal, its off-diagonal element is 0; when the spoofing jammer is correlated with the real signal, its off-diagonal element ρ is 0. ij (i≠j) is not zero. During the process of deceiving and forcing the receiver to synchronize with the false PRN sequence, the signal and interference become correlated until they are fully synchronized. At this point, the signal and interference are coherent. Angle estimation methods for correlated sources often require decorrelation preprocessing, which leads to a loss of array degrees of freedom and increased computational complexity. The power-DOA joint estimation method of this invention is unaffected by the correlation between the signal and interference and does not require decorrelation preprocessing.

[0053] 2) Deception interference with incident direction identification and suppression

[0054] Conventional spoofing interference suppression methods assume that the spoofing jammer transmits multiple false signals from a single incident direction. This invention considers the complex situation where the receiver's receiving environment contains multiple self-synchronizing spoofing interferences from different directions. In this case, during the synchronization of multiple spoofing jammers with the receiver, they transmit false signals from multiple incident directions. Initially, the received power of each spoofing interference signal is very close to the power of the real signal; therefore, the incident signal with the highest power at the start of synchronization cannot be directly selected as the spoofing interference. This invention introduces an observation function to observe the power and changes of each incident signal over a synchronization period. This provides a basis for judging and distinguishing the incident directions of spoofing interference and the real signal, and also provides important prior information for the array antenna to successfully suppress multiple spoofing interferences. Specifically:

[0055] 21) Observation function of spatial power spectrum

[0056] A single Global Positioning System (GPS) data symbol has its C / A code repeated 20 times within 1 bit, i.e. To construct a reference signal using this signal structure characteristic, the specific method is as follows: use data x(n-jK) with a delay of any integer multiple of 1023 chips as its reference signal.

[0057]

[0058] Where jK chips represent the delay between the reference signal and the data, K is the length of the C / A code (1023 chips), and j is the multiple of the C / A code spreading gain, generally taken as 1≤j<20. Both the real signal and the deception interference simulating the real signal in the received data X(n) exhibit strong autocorrelation. However, the noise is independent of the real signal and the deception interference; the received noise of each array element is statistically independent Gaussian white noise. Since s(n-jK)=s(n), cross-correlation is performed on the received data vector and the reference signal, and the cross-correlation matrix of the data vector and the reference signal in the t-th observation is... for:

[0059]

[0060] Among them, the t-th observation from θ u Navigation signal power σ in direction 2 (θ u ), u=1,2,...,U。 Compared with the array received signal covariance matrix shown in equation (3), this cross-correlation matrix no longer contains the noise term in the array received signal covariance matrix.

[0061] This invention proposes a spatial power spectrum capable of simultaneously observing power and DOA pair information. Without prior information about the incident angle, by substituting the cross-correlation matrix of the data vector and the reference signal from the t-th observation into the spatial power spectrum observation function shown in equation (6), and performing peak search, the pair of characteristic parameters—the DOA and power values—of each navigation signal can be obtained, i.e.:

[0062]

[0063] θ is traversed within the observation space. This represents the direction of incoming wave θ that maximizes the function observed in the t-th observation. The peak value corresponding to this direction is the estimated signal power from that direction. [] + Denotes the pseudo-inverse of a matrix, i.e., a + (θ)=a H (θ)(a H (θ)a(θ)) -1 When θ = θ u Then, we get from θ u Power estimate of navigation signal for:

[0064]

[0065] This is because any two distinct incident directions satisfy a + (θ i )a(θ u)0(i≠u), while when θ traverses to the u-th incident direction (such as θ i =θ u If a), then a + (θ u )a(θ u ) = 1, so δ u = [0,…,1,0,…,0] is a 1×M dimensional vector, where the u-th element is 1 and the rest are 0. Then the power of the u-th source is:

[0066]

[0067] As shown in equation (8), extracting the diagonal element yields the received signal power from the direction θ of the incoming wave observed in the t-th observation, which is σ. 2 (θ), which is not affected by off-diagonal ρ during the extraction process. ij The effect of (i≠j) means that even if the deception is coherent with the signal, it will not affect the estimation, and this advantage depends on the removal of the noise term.

[0068] In summary, the observation function σ of the spatial power spectrum 2 (θ) represents the direction of the incoming wave as the independent variable of the power, and the power is a function of the direction of the incoming wave. Therefore, this observed function is called the spatial power spectrum, and its spectral peak corresponds to the joint estimation result of power-DOA. It can be seen that this process does not require decorrelation and eigenvalue decomposition, and the search vector a + (θ) can be pre-calculated and stored. Once the cross-correlation matrix of an observation is available, a full-space power spectrum can be traversed.

[0069] 32) Classification method of incident signals based on maximum entropy threshold

[0070] The power of the real navigation signal received by the receiver remains essentially constant, while the synchronization spoofing gradually increases the transmission power during receiver synchronization. To facilitate observation of power changes in each incoming wave direction, the power value is... Arranged in a specific order according to the direction of arrival and matched one-to-one, after a period of continuous observation, when it was found that the received power of signals from multiple different incident directions was continuously increasing, a maximum entropy threshold adaptive method was used to divide each power and DOA value pair into two groups. One group with power below the threshold δ was considered the real signal, and the power and incident angle of this group were used as the power and direction of arrival of the real signal, respectively. The other group with power above the threshold δ was considered the spoofed signal transmitted by deception interference, and the power and direction of this group were used as the power and direction of arrival of the spoofed signal transmitted by deception interference, respectively.

[0071] The array receives signals consisting of K real navigation signals and R synchronization spoofing signals, for a total of U signals. The estimated power... From minimum to maximum value The intervals between them are divided into L segments with the same spacing [ξ]. l ,ξ l+1 Let q(ξ), l=0,…,L-1, l ) is power The quantity, P(ξ) l ) represents the probability that the estimated power occurs in the l-th interval, i.e.

[0072]

[0073] satisfy Let δ be the threshold for classifying real signals and deceptive interference based on power, and let P be the posterior probability of the real navigation signal. au The posterior probability P of deception interference sp They are respectively:

[0074]

[0075] The posterior probability P of the true navigation signal au The posterior probability P of deception interference sp The entropies of the parts are respectively:

[0076]

[0077] l δ This represents the index of the starting point containing the δ-interval, maximizing P. sp and P au The sum of these two entropies gives the threshold δ for threshold segmentation:

[0078]

[0079] Based on the maximum entropy threshold δ, the U received signals are divided into two categories according to their power: deceptive signals and genuine signals, represented as follows:

[0080]

[0081] A power diagonal array for both the deception interference group and the real signal group. The corresponding incoming wave directions Θ = Θ for these two groups. au ∪Θ sp for:

[0082]

[0083] By deceiving the set of incoming wave directions of the interference group and the set of incoming wave directions of the real signal group, these two sets can be used as prior information for interference suppression.

[0084] 23) Calculate the robust optimal weight vector.

[0085] When the code phase difference between the deception interference and the real signal is within 1 chip, the two are correlated. Traditional adaptive beamforming methods cannot accurately suppress this correlation. A robust beamforming method is adopted. By reconstructing the covariance matrix of the deception interference plus noise and the array autocorrelation matrix of the real satellite navigation signal, the optimal weight vector for maximum output SINR beamforming is calculated. This ensures that multiple main lobes are aligned with the desired real navigation signal, while multiple nulls counteract multiple deception interferences. The specific process is as follows:

[0086] Based on the deception interference power and direction of arrival estimation set C obtained from step 23), sp and Θ sp The interference plus noise covariance matrix is ​​reconstructed as follows:

[0087]

[0088] Among them, C sp For the power estimate of the deception interference, the diagonal matrix, A(Θ) sp To obtain the steering matrix for deception interference based on the estimated direction of incoming interference and the known array configuration, noise... for:

[0089]

[0090] in, Let tr{} be the orthogonal projection matrix of the steering matrix A(Θ) of U signals, where tr{} denotes the trace of the matrix, and tr{} is the self-conversion matrix of the array receiving the real signals. for:

[0091]

[0092] The autocorrelation matrix of the real signal and deception interference plus noise covariance matrix Substituting the minimum variance distortionless response beamformer, we get:

[0093]

[0094] Where Γ{} represents the eigenvector corresponding to the largest eigenvalue of the matrix, and this optimal weight vector guarantees the maximum output SINR.

[0095]

[0096] Where e1 is the eigenvector corresponding to the largest eigenvalue, C au C is a diagonal matrix composed of estimated values ​​of the true signal power. sp A(Θ) is a diagonal matrix composed of estimated interference power values ​​used to deceive the system. auA(Θ) is the steering matrix used to estimate the direction of the incoming wave from the real signal. sp This method employs a steering matrix estimated based on the direction of arrival of the deceptive signal. This optimal beamforming can form multiple main lobes, simultaneously pointing towards all directions of the received real signal. Multiple nulls align with all deceptive interference received. This beamforming is unaffected by the high correlation or even coherence between deceptive interference and the real signal, making it a robust method for suppressing deceptive interference. This method effectively avoids the rank recovery and degree-of-freedom loss of the covariance matrix caused by using smoothed decorrelation.

[0097] Simulation Experiment

[0098] To verify the effectiveness of this invention, four aspects were mainly examined: spatial power spectrum, beam pattern for suppressing spoofing interference, GPS receiver synchronization, and output signal-to-noise ratio (SNR). The GPS receiver array was assumed to be a 13-element linear array with uniform half-wavelength spacing. The receiver's time-domain sampling frequency was set to the Nyquist sampling frequency, and the received power of the real signal was -158.5 dBW. The PRN sequences used in the experiment were all coarse / intercepted (C / A) codes, and the observation duration was set to 50 seconds. This period was the synchronization phase between the spoofing interference release signal and the real signal. The power of the false navigation signal was assumed to increase from -158.5 dBW to -157.1 dBW, exceeding the real signal by 1.4 dB.

[0099] Assume the array receives asterisks including real PRN#1, real PRN#2, real PRN#3, real PRN#4, spurious PRN#2, spurious PRN#3, and spurious PRN#4. The arrival direction and power values ​​of each navigation signal within 50 seconds are set as shown in Table 1. The received noise power is -141 dBW Gaussian white noise, and the specific parameters of each sequence are shown in Table 1.

[0100] Table 1

[0101]

[0102] Experiment 1 Spatial Power Spectrum

[0103] The experimental simulation parameters are as shown above. By taking 1 bit of data every 10 seconds and observing once, the incoming direction-power results of each received navigation signal can be effectively estimated. Using the spatial power spectrum results proposed in this paper, such as... Figure 1 As shown, the vertical axis of the seven peaks represents the estimated power value, while the horizontal axis represents the estimated direction of arrival. It can be seen that this spatial power spectrum can accurately estimate the direction of arrival and incident power of each PRN sequence. Since there are no noise terms in the cross-correlation matrix, weak navigation signals and deception interference can be accurately estimated from the power of each incident direction using the spatial power spectrum.

[0104] Experiment 2 Maximum Entropy Adaptive Threshold

[0105] Six observations were conducted within 50 seconds, recording the power value of the incident PRN sequence in its incident direction for each observation. Figure 2 As shown, the adaptive threshold formed based on the estimated values ​​obtained from observations is approximately -156.7 dBW. The spoofing interference is indicated when the horizontal axis crosses the red line to the right in the sixth observation, specifically in the directions of -16°, 20°, and 60°, consistent with the experimental settings in Table 1. By observing the spatial power spectrum, the power-DOA joint estimation results are obtained. Introducing an observation period during the spoofing synchronization process allows for a direct observation of the direction from which the spoofing interference is incident.

[0106] Experiment 3 Beam Pattern

[0107] The experimental simulation parameters are as shown above. The beam pattern for suppressing deception interference using this invention is as follows. Figure 4 As shown, the null positions of this invention correspond to the directions of arrival of the three deceptive interference signals, while the directions of maximum and maximum gain are aligned with the incident directions of the four desired satellites. The interference suppression method employs a robust multi-beamformer, unaffected by the correlation between incident signals. The beam pattern of this invention only requires one set of optimal weights to enhance the navigation signals of multiple satellites.

[0108] Comparison of input and output SNR results in Experiment 4

[0109] After classifying signals from different incident directions, robust optimal beam weight vectors are calculated using two categories of estimates (including DOA / power for navigation signals and spoofing interference). The input SNR of the real signal and the spoofing signal at the receiver is compared with the beamforming output SNR. Specifically, the array output signal SNR is as follows:

[0110]

[0111] Among them, R S Given the autocorrelation matrix of a signal (real or spoof) received by the array, the beamforming output SNR of the real or spoof signal can be calculated. This experiment uses 5000 Monte Carlo simulations. Figure 5 As shown, with the increase of spoofing interference power, the power of the output true signal after interference suppression is 17dB higher than that before interference suppression. When the power of the spoofing signal reaches approximately -157.5dBW (approximately 1dB higher than the true signal power), the output spoofing signal SNR is equal to the average power of the true signal, and the output spoofing interference SNR decreases significantly with the increase of input interference power. This is because the stronger the interference, the more accurate the estimation and classification of interference power, and the better the interference suppression performance in removing interference data from the received data. Therefore, this invention can effectively suppress spoofing interference with a synchronization period exceeding 1dB higher than the true signal.

[0112] Synchronization results before and after interference suppression in Experiment 5

[0113] The experiment examined the code correlation results of the receiver before and after using this invention. Code correlation was performed on the PRN#2 satellite before and after spoofing interference suppression, as shown below. Figure 6a and Figure 6b As shown, observing the changes in the locking point on the code correlation curve before and after interference suppression reveals a decrease in the peak value, which is due to the amplitude reduction caused by the suppression of spoofing. The code delay occurs at sampling point 10231, a delay of only one sampling point. This invention enables the receiver to avoid the influence of spoofing during the synchronization process with the real signal, allowing the receiver to obtain correct synchronization information.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for identification and mitigation of multiple simultaneous spoofing jamming of satellite navigation, comprising: The method comprises the following steps: constructing a de-noised cross-correlation matrix by using the received data vector and a reference data vector; calculating a spatial power spectrum by using an observation function according to the constructed de-noised cross-correlation matrix; performing adaptive classification on the pair of power and direction of arrival estimation values obtained by observation by using a maximum entropy threshold according to the spatial power spectrum, so as to identify the power and direction of arrival of the incident navigation signal; reconstructing a signal autocorrelation matrix and an interference plus noise covariance matrix according to the identification result, calculating an optimal vector according to the signal autocorrelation matrix and the interference plus noise covariance matrix, and suppressing spatial deception jamming according to the optimal vector; the spatial power spectrum is: (6) wherein, represents the selected wave direction that maximizes the t , , is a steering vector of dimension is the denoised cross-correlation matrix.​​ 2. The method of claim 1, wherein, the de-noised cross-correlation matrix is: (5) Among them, the t The second observation from Navigation signal power in direction , The number of incident signals. , for The steering vector matrix of the real signal in dimensionality. for The deceptive interference of the dimension is guided by a vector matrix. Non-diagonal elements .

3. The method of claim 1, wherein: the optimal weight vector is calculated according to formula (18), wherein, (18) wherein, is an eigenvector corresponding to the largest eigenvalue of the matrix, is the autocorrelation matrix of the array receiving the real signal, is the reconstructed interference plus noise covariance matrix.

4. The method of claim 3, wherein, Reconstructing interference-plus-noise covariance matrices is: (15) wherein is a diagonal matrix of power estimates of the jammer, is a steering matrix of the jammer obtained from the estimated direction of arrival of the jammer and the known array configuration, is noise.

5. The method of claim 4, wherein, Noise is: (16) wherein is U the steering matrix of the signals the orthogonal projection matrix of denotes the trace of a matrix R is the number of synchronous deception signals.

6. The method of claim 5, wherein, The array receives an autocorrelation matrix of the real signal Is: (17) wherein, is according to a maximum entropy threshold will be U received signals are divided into two categories according to the size of the power of the real signal, is a steering matrix that estimates the direction of arrival of the real signal.

7. The method of claim 6, wherein, the SINR corresponding to the optimal weight vector is: (19) wherein, is an eigenvector corresponding to the largest eigenvalue, is a diagonal matrix composed of real signal power estimation values, is a diagonal matrix composed of spoofing signal power estimation values, is a steering matrix using estimated real signal to wave direction, is a steering matrix using estimated spoofing signal to wave direction.

Images