Recognition algorithm applied to weak Beidou deception signal

Abstract

The invention relates to the field of signal processing, in particular to an identification algorithm applied to weak Beidou deception signals, and the method comprises the steps: S1, carrying out the data downsampling, and generating a local ranging code according to a satellite number; s2, a correlation operation fast algorithm; s3, performing differential coherent accumulation; s4, determining a dynamic threshold according to the false alarm probability; and S5, deception signal judgment. According to the scheme, not only can strong deception signals be identified, but also weak Beidou deception signals can be identified.

Description

Technical Field

[0001] The present invention relates to the field of signal processing, and in particular to an identification algorithm applied to weak Beidou deception signals. Background Art

[0002] As the application expands, the interference risk and background of Beidou signals have gradually attracted attention. At present, portable jamming equipment is relatively easy to obtain, especially low-cost equipment sold on the market. These devices may be used maliciously, causing the failure of navigation equipment in key infrastructure such as airports and ports, affecting safety. Therefore, interference detection is particularly important, especially since the current jamming equipment is low-power and small, making it difficult to detect. Summary of the invention

[0003] The object of the present invention is to provide an identification algorithm for weak Beidou spoofing signals, which can detect interference when the interference signal is transmitted at a low power.

[0004] The present invention provides a basic solution 1: an identification algorithm for weak Beidou spoofing signals, including the following contents:

[0005] S1, data downsampling, and generating a local pseudo-random sequence based on the asterisk;

[0006] S2, fast algorithm for related operations;

[0007] S3, differential coherent accumulation;

[0008] S4. Determine a dynamic threshold based on the false alarm probability;

[0009] S5. Deception signal judgment.

[0010] Further, the S1 includes:

[0011] Get signal data, filter and downsample the signal data, and reduce the Beidou signal sampling rate to 4.096MHz, so that each chip has two sampling points. The Beidou signal is generated as follows: Figure 2 As shown, the initial phase is [01010101010], and the G1 sequence polynomial is g1=1+x+x^7+x^8+x^9+x^10+x^11,

[0013] The G2 sequence polynomial is g2=1+x+x^2+x^3+x^4+x^5+x^8+x^9+x^11

[0014] The phase of the G2 sequence can be found in the Beidou satellite navigation system space signal interface control file.

[0015] Further, the S2 includes:

[0016] The Doppler frequency deviation caused by the movement of the satellite needs to be compensated. The compensation range of the geosynchronous satellites from 1 to 5 satellites is set to -2KHz to 2KHz, and the compensation range of satellites from 6 to 37 is set to -7KHz to 7KHz, with a compensation interval of 500Hz. In order to increase the correlation speed, the time domain signal is correlated and transformed to the frequency domain for point multiplication. For the processing of Doppler frequency deviation, since the time domain multiplied by the compensated Doppler frequency deviation is equal to the shift in the frequency domain, the locally generated time domain pseudo-random sequence is converted to the frequency domain after using the maximum Doppler frequency deviation compensation. When other frequency deviation compensation is used later, only the corresponding cyclic shift in the frequency domain is required. Since the sampling rate is 4.096MHz, the data length of the 1ms code chip period is N=4096, and 2m of data is taken for FFT, then each point of movement is 500Hz. Assuming the number of searched stars is M, the number of Doppler compensation cycles is K, and the amount of FFT calculation is Nlog2(N), then the amount of calculation for a search of length N is M*(2*O(NlogN)+K*(O(NlogN)+N)), where 2 represents 2 FFTs, K represents K IFFTs, and N plus N represents N complex multiplications. If time domain correlation and time domain Doppler frequency offset compensation are used, M*((O(N^2)+N+N')*K is required, where N' represents the calculation of the frequency offset value in the complex domain. If 31 stars are searched, K=29, and N=4096, the amount of calculation is more than 200 times that of the fast algorithm.

[0017] Further, the S3 includes:

[0018] According to the frame structure characteristics of Beidou satellite, there is a 20ms NH code. Therefore, in order to improve the detection probability, multiple 1ms accumulation cannot be performed directly. Instead, according to the 20ms repetition characteristics of the NH code, differential coherence is performed at intervals of 20ms and accumulated for 20ms. The principle of coherent difference and accumulation is as follows: Figure 3 As shown, the Beidou data needs a total of 40ms of data to be processed. At the same time, when taking data, the data is most likely to be in the middle part of NH, so after 20ms of accumulation, the maximum value can only be in the first 10ms or the last 10ms. Therefore, when outputting the data, it is necessary to determine whether the maximum value is in the first 10ms or the last 10ms. For specific differential coherence and accumulation diagrams, see Figure 4 .

[0019] Further, the S4 comprises:

[0020] In order to ensure that the peak value is not detected due to the fluctuation of signal power, a dynamic threshold judgment method is adopted. The generation of the dynamic threshold uses the set false alarm probability to derive the actual dynamic threshold value, and dynamically adjusts it in combination with the actual environment. Since most interference signals meet the Rice distribution, the probability density function of the Rice distribution is used as a model. When the received signal only has noise, the sample of the received signal r(t) exceeds the threshold voltage V TThe probability is:

[0021]

[0022] The corresponding threshold is:

[0023] where ψ 2 is the variance of the noise

[0024] At the same time, in order to adapt to the front-end characteristics of different receivers, a proportional adjustment is required, and the final threshold value is:

[0025] Where β is the corresponding adaptation factor.

[0026] Further, the S5 comprises:

[0027] Bi-peak judgment, using the generated dynamic threshold to judge whether there are two correlation peaks outside the interval of one chip. Determine whether the maximum value of the accumulated value in 4096 points is greater than the threshold, subtract one chip distance from the current maximum chip position, and find the second largest peak within this range. The threshold uses an algorithm similar to S4.

[0028] In summary, this solution provides an identification algorithm for weak Beidou spoofing signals, which can detect interference spoofing signals in the case of weak interference signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of a recognition algorithm flow chart for weak Beidou spoofing signals according to the present invention;

[0030] Figure 2 Generate schematic diagram for ranging code;

[0031] Figure 3 This is a schematic diagram of the physical frame structure of the Beidou signal;

[0032] Figure 4 It is a schematic diagram of differential coherence and coherence result accumulation;

[0033] Figure 5 This is a schematic diagram of the correlation results between the main and secondary peak values ​​of the data;

[0034] Figure 6 This is a schematic diagram for comparing algorithm performance; DETAILED DESCRIPTION

[0035] The following is further described in detail through specific implementation methods:

[0036] The embodiment is basically as shown in the attached Figure 1 As shown: An identification algorithm for weak Beidou deception signals, including the following contents:

[0037] S1, data downsampling, and generating a local pseudo-random sequence based on the asterisk;

[0038] Specifically, the signal of Beidou satellite is expressed as:

[0039]

[0040] Wherein, j represents the satellite number; A I Indicates the amplitude of the ranging code modulated on the B1 frequency band carrier I branch; A Q Indicates the amplitude of the ranging code modulated on the Q branch of the B1 frequency band carrier; C I represents the I branch ranging code, that is, the pseudo-random code; C Q Indicates the Q branch ranging code; D I represents the data code modulated on the I branch ranging code; D Q represents the data code modulated on the Q branch ranging code; f represents the carrier frequency of the B1 band, Indicates the initial phase of the B1 frequency band carrier I branch; Indicates the initial phase of the Q branch of the B1 frequency band carrier. Each satellite has a B1I band, so the B1I signal is mainly processed. B1I is an open service, and B1Q is an authorized service. This article mainly focuses on the open service signal, so the signal is represented as:

[0041]

[0042] The physical frame structure of Beidou signal is as follows: Figure 3 .

[0043] The receiving end receives S j After that, down-conversion filtering is performed, and the down-converted signal is downsampled to 4.096MHz. The obtained signal is:

[0044]

[0045] S2, fast correlation operation of pseudo-random sequence;

[0046] Specifically, considering the need to compensate for the Doppler frequency deviation caused by movement, the compensation range of the geosynchronous satellites from satellites 1 to 5 is set to -2KHz to 2KHz, and the compensation range of satellites 6 to 37 is set to -7KHz to 7KHz, with a compensation interval of 500Hz. First, multiply the local sequence by the maximum frequency deviation f max ,

[0047]

[0048] Then perform a 2ms length N = 8192 point FFT on the locally generated signal to obtain:

[0049] P j(f) = fft(C j (t))

[0050] because Therefore, each left shift is equivalent to a frequency offset of fs / N, which is equivalent to a 500Hz Doppler frequency domain compensation. In this way, each shift at the positive maximum frequency offset position continuously compensates for the negative maximum frequency offset position. At the same time, the time domain code phase search and Doppler frequency offset compensation are performed simultaneously to reduce the amount of calculation.

[0051] Take the 2ms data length after downsampling and do 8192-point FFT:

[0052] Y j =fft(S j ),

[0053] Multiply P and F and do ifft:

[0054] m=[0,2f max / 500Hz-1] indicates the left shift size, k=0:4095.

[0055] S3, differential coherence and coherent accumulation;

[0056] Specifically, r(k) is differentially calculated every 20ms. Since the data bit of the NH code of 20ms is either 1 or -1, the positive and negative NH data of every 1ms within 20ms are the same, such as Figure 4 As shown, this can avoid the peak loss caused by the positive and negative flipping of the NH code. The differential coherence formula is as follows:

[0057] d(k)=r(k)*r(k-20) * Where 20 means the interval of 20ms, k=0:4096*20-1

[0058] At this time, the data of 10ms before and after are accumulated respectively.

[0059]

[0060] Since data values ​​cannot be Figure 4 As shown, the data is taken at the 20ms head position, so normally only one of the two Z values ​​has the maximum value at 10ms, so the maximum value of the two Z values ​​is compared as the output.

[0061] S4. Determine a dynamic threshold based on the false alarm probability;

[0062] Specifically, the maximum 10ms accumulated correlation value obtained by S3 is a two-dimensional signal, one dimension is the Doppler frequency deviation, and the other dimension is the code chip offset. This step is screened on the basis of S3, and the maximum value is the maximum value under the corresponding Doppler frequency deviation and code deviation, and it is judged whether the maximum value exceeds the threshold. The threshold is set using the dynamic threshold judgment method. The generation of the dynamic threshold uses the set false alarm probability to derive the actual dynamic threshold value, and dynamically adjusts it according to the actual environment. Since most interference signals satisfy the Rice distribution, the probability density function of the Rice distribution is used as a model. When the received signal only has noise, the sample of the received signal r(t) exceeds the threshold voltage V T The probability is:

[0063]

[0064] The corresponding threshold is derived as:

[0065] where ψ 2 is the variance of the noise, and its value is the variance of the correlation value outside the maximum value.

[0066] At the same time, in order to adapt to the front-end characteristics of different receivers, a proportional adjustment is required, and the final threshold value is:

[0067] Where β is the corresponding adaptation factor.

[0068] Figure 5 Results obtained using open source data provided online by the University of Texas at Austin.

[0069] S5, determining the main and secondary peaks according to the threshold to judge whether there is a deceptive signal;

[0070] Specifically, when the maximum value of S4 exceeds the threshold, it is determined whether there is a secondary peak. The distance between the position of the secondary peak and the position of the maximum value is at least greater than 1 code chip to eliminate the influence caused by sampling point fission and oversampling. The judgment method of the secondary peak is the same as S4. Interference with navigation satellite signals generally interferes with multiple signals. If only a single signal is interfered, the interference will be invalid because the equipment uses the better signal to calculate the navigation satellite pseudo-range equation. Therefore, after the secondary peak is detected, on the one hand, the interference caused by multipath is reduced, and on the other hand, the correctness of the judgment is improved. It is necessary to determine whether the number of satellites with the same main and secondary peak intervals that currently detect double peaks accounts for 80% of the total number of double peaks. If so, it is considered that a deceptive interference signal does exist. The final test performance is as follows Figure 6 As shown, the algorithm of the present invention is about 10db better than the traditional ranging code correlation which only performs 1ms.

[0071] This solution is based on the properties of Beidou satellite signals and can detect not only strong spoofing signals, but also weak spoofing signals.

[0072] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. An identification algorithm for weak Beidou spoofing signals, characterized in that: It includes the following: S1, data downsampling, and generating a local pseudo-random sequence based on the asterisk; S2, fast algorithm for related operations; S3, differential coherent accumulation; S4. Determine a dynamic threshold based on the false alarm probability; S5. Deception signal judgment.

2. The recognition algorithm for weak Beidou spoofing signals according to claim 1, characterized in that: Said S1 comprises: Acquire signal data, filter and downsample the signal data, reduce the Beidou signal sampling rate to 4.096MHz, and generate ranging codes for 37 Beidou satellites. The generation method is shown in Figure 2.

3. The recognition algorithm for weak Beidou spoofing signals according to claim 2 is characterized in that: The S2 comprises: Fast correlation and Doppler frequency offset compensation. The time domain signal and the locally generated satellite signal are transformed into the frequency domain by FFT, and the correlation in the time domain is replaced by the dot product in the frequency domain. At the same time, the Doppler frequency offset caused by the movement of the satellite is compensated. The compensation interval is set to 500Hz, and the compensation method is to shift the local frequency domain signal instead of the frequency offset compensation in the time domain.

4. The recognition algorithm for weak Beidou spoofing signals according to claim 3 is characterized in that: The S3 includes: According to the characteristic of Beidou satellite NH code repeating once every 20ms, the correlation results in S2 are differentially coherent at intervals of 20ms, and differential coherence is conjugate point multiplication at intervals of 20ms. Then, the 20ms point multiplication result is used to accumulate the coherent results of 10ms before and after to generate two 4096-point accumulation results. The specific process is shown in Figure 4.

5. The recognition algorithm for weak Beidou spoofing signals according to claim 4 is characterized in that: The S4 comprises: In order to ensure the accuracy of peak value judgment due to the influence of signal power, a dynamic threshold judgment method is adopted. The generation of the dynamic threshold uses the set false alarm probability to derive the actual dynamic threshold value, and makes dynamic adjustments based on the actual environment.

6. The recognition algorithm for weak Beidou spoofing signals according to claim 4 is characterized in that: The S5 comprises: Double peak judgment uses the generated dynamic threshold to judge whether there are two correlation peaks outside the interval of one chip. Normally, there must be deception interference with the timing position different from the actual Beidou signal during the movement of the device. Therefore, the judgment is made by the number of peaks. When there are multiple satellites with two positions and the interval between the two positions is the same, it can be judged that deception interference exists.

7. An identification algorithm for weak Beidou spoofing signals, characterized in that: An identification algorithm for weak Beidou spoofing signals as described in any one of claims 2 to 6 is adopted.

Images