A broadband interference identification method based on real-time spectrum analysis
Through real-time spectrum analysis and short-time Fourier transform technology, the accurate identification of interference signals in complex electromagnetic environments is achieved, and the automation challenges of radar system in interference signal recognition is solved, and the anti-interference ability and combat effectiveness of radar system are improved. .
Patent Information
- Application Number
- CN202111158129.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-09-30
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2041-09-30
AI Technical Summary
It is difficult for existing radar systems to automatically identify and classify interfering signals in complex electromagnetic environments, resulting in the choice of anti-interference measures relying on manual experience, which limits the development of radar anti-interference technology.
Using a broadband interference identification method based on real-time spectrum analysis, the time-frequency diagram is obtained through short-time Fourier transform, spectrum statistics and probability density matrix analysis are carried out, and the interference detection classifier is designed to realize the scanning interference, broadband suppression interference and comb-shaped Accurate identification of spectral interference.
It improves the situational awareness capability of the radar system and the automatic selection of anti-interference measures in complex electromagnetic environments, enhances the survivability and combat effectiveness of the radar system, with small calculation volume, fast recognition speed and high accuracy.
Smart Images

Figure CN113917409B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electronic countermeasures. Background Art
[0002] The electromagnetic environment of modern battlefields is becoming increasingly complex. With the continuous development of technical research in the field of electronic countermeasures, many radar broadband interference patterns with special interference capabilities have emerged in an endless stream, posing a severe challenge to the perception and anti-interference capabilities of modern radars on battlefield interference situations. Effective cognition of the interference environment is a necessary condition for radars to accurately counter active interference. First, it is necessary to extract, analyze and classify the characteristics of interference signals to complete radar interference pattern recognition, and then provide a reference for the selection of anti-interference measures. Radar interference pattern recognition refers to the analysis of the interference signal received by the radar system when it receives an interference signal attack, and the interference pattern is determined according to the characteristics of the signal, so as to adopt corresponding anti-interference strategies to maintain the normal operation of the radar system. Although radars have relatively complete anti-interference functions and strategies, there is still room for development in the automatic real-time perception of specific interference patterns. The selection of anti-interference measures mainly relies on subjective judgment based on human experience. The inability of radars to automatically detect and identify interference has restricted the further development of radar anti-interference technology.
[0003] Therefore, it is of great significance to study radar interference signal recognition technology, provide accurate indication information for radar systems, promote optimal allocation of radar system resources, and enhance radar's situational awareness capabilities in complex electromagnetic environments. Summary of the invention
[0004] The present invention proposes a broadband interference identification method based on real-time spectrum analysis, which is suitable as an auxiliary decision-making means for an adaptive anti-interference radar system and can enhance the situational comprehensive intelligent perception capability of the radar system in a complex electromagnetic environment.
[0005] The technical solution to realize the present invention is as follows: interference pattern detection is performed on the intermediate frequency data of the broadband interference signal, the time-frequency diagram of the interference data over the detection threshold is obtained by using the short-time Fourier transform (STFT), the probability density matrix is obtained by performing spectrum statistics on the interference time-frequency diagram, and the broadband interference pattern recognition is realized by performing parameter measurement and feature extraction on the probability density matrix and completing the design of the interference detection classifier.
[0006] The specific steps are:
[0007] Step 1: Perform short-time Fourier transform on the intermediate frequency data of the broadband interference signal to obtain a time-frequency matrix containing time-frequency-amplitude information;
[0008] Step 2: Perform spectrum statistics on the time-frequency matrix and map it to obtain a probability density matrix containing frequency-amplitude-frequency information;
[0009] Step 3: Conduct an electromagnetic spectrum survey on the probability density matrix and count the frequency of interference signals at each frequency point of the probability density matrix within the ambient noise amplitude range;
[0010] Step 4: Set the interference detection threshold to detect each frequency point of the probability density matrix, and make a judgment based on the frequency of occurrence of the interference signal counted in step 3. If the frequency exceeds the interference detection threshold, it is determined that there is an interference signal at the current frequency point, and the number of frequency points with interference is counted. The bandwidth of the interference signal is calculated based on the number of frequency points with interference to determine whether it is broadband interference. If so, proceed to step 5, otherwise it is determined that there is no interference;
[0011] Step 5: Set the interference density threshold to detect each frequency point of the probability density matrix, and make a judgment based on the frequency of occurrence of the interference signal counted in step 3. If the frequency exceeds the interference density threshold, it is determined that there is an interference signal at the current frequency point, and the number of frequency points with interference is counted. The bandwidth of the interference signal is calculated based on the number of frequency points with interference to determine whether it is swept frequency interference. If so, it is determined to be swept frequency interference. If not, go to step 6;
[0012] Step 6: Set the interference interval threshold, traverse the frequency points with interference in detection step 5, calculate the frequency difference between adjacent interfered frequency points, record and update the maximum frequency difference. If the maximum frequency difference is greater than the interference interval threshold, the interference is determined to be comb spectrum interference. If the maximum frequency difference is less than the interference interval threshold, the interference is determined to be broadband noise interference.
[0013] The beneficial effects of the present invention are as follows: due to the adoption of the method described in the present invention, by adopting a broadband interference identification method based on real-time spectrum analysis, accurate classification of swept frequency interference, broadband suppression interference and comb spectrum interference can be achieved, the automatic selection capability of anti-interference measures is effectively enhanced, and the survivability and combat effectiveness of the radar system in a complex electromagnetic environment are enhanced. In addition, the method proposed in the present invention has small calculation amount, fast identification speed, high accuracy, simple process and is easy to implement in engineering.
[0014] The present invention is described in further detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 Flowchart of the broadband interference pattern recognition algorithm. DETAILED DESCRIPTION
[0016] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments, but the protection scope of the present invention is not limited by the implementing regulations. Figure 1 This is a flow chart of the broadband interference pattern recognition algorithm. The specific steps are:
[0017] Step 1: Perform short-time Fourier transform on the intermediate frequency data of the broadband radar jammer signal to obtain the time-frequency matrix STFT(τ,ω) containing time-frequency-amplitude information. The discrete expression of the lth frame of the time-frequency matrix can be described as:
[0018]
[0019] Where x represents the broadband interference signal, N STFT represents the short-time Fourier transform length, and H represents the jump step size between different frames.
[0020] Step 2: Perform spectrum statistics on the time-frequency matrix STFT(τ,ω) to obtain the probability density matrix P(Amp,f), P(Amp i ,f j ) indicates that the amplitude value in this interference detection is Amp i , the frequency value is f j The number of interference signals that occur, where Amp i Represents the i-th amplitude value on the amplitude dimension of the probability density matrix, 0≤i≤N Amp ,i∈N,N Amp =(Amp max -Amp min ) / ΔAmp,f j Represents the jth frequency value in the frequency dimension of the probability density matrix, where 0≤j≤N STFT ,j∈N,; when the spectrum is counted, if ΔAmp*i≤STFT(t i ,f j )≤ΔAmp*(i+1), then the probability density matrix is in (Amp i ,f j ) at the statistical value P(Amp i ,f j ) plus 1.
[0021] Step 3: The noise amplitude range of the electromagnetic spectrum survey is set to (N lower ,N upper ), traverse the probability density matrix P(Amp i ,f j ) at all frequencies f j (0≤j≤N f ,j∈N), each frequency point f j The frequency of occurrence of interference signals in the upper amplitude range belonging to the noise amplitude range is specifically described as:
[0022]
[0023] Step 4: Set the interference detection threshold to perform interference detection on all frequency points of the probability density matrix. The specific description of the interference detection threshold is:
[0024]
[0025] Where μ is the statistical mean of the noise background, P fa is the false alarm probability; when the frequency point f i The frequency of the interference signal L fj When it is greater than the detection threshold r1, the frequency point f i There is interference, count the number of frequencies where interference exists N r The number of false alarms within the observation time T is specifically described as
[0026] N fa =T·f s ·P fa (4)
[0027] In the formula, f s is the receiver sampling rate, if N r >N fa , it is determined that there is broadband interference, otherwise it is determined that there is no interference.
[0028] Step 5: Set the interference density detection threshold to perform interference detection on all frequency points of the probability density matrix. The interference density detection threshold is specifically described as:
[0029]
[0030] Where S is the statistical mean of the interference amplitude; if the frequency point f j The frequency of the interference signal L fj If the frequency point f is higher than the detection threshold r2, j There is interference, frequency f j The threshold mark R(f j ) is set to 1, otherwise it is set to 0, and all frequency point over-threshold flags constitute the over-threshold flag sequence R(j); the number of frequency points with interference N is counted r , compared with the number of false alarms within the measurement time T, if N r <N fa , it is determined to be a frequency sweep interference, otherwise, go to step 6 to determine the interference type.
[0031] Step 6: Traverse the threshold-crossing mark sequence R(j) obtained in step 5, and calculate the frequency difference between adjacent threshold-crossing mark sequences:
[0032] ΔR(j)=R(j)-R(j-1) (6)
[0033] And find the maximum frequency difference ΔR max=max(ΔR(j)); According to the effective signal bandwidth B, the comb spectrum interference frequency interval B w And the number of short-time Fourier transform points N STFT , we can get the interference interval threshold of comb spectrum interference, which is specifically described as
[0034]
[0035] If ΔR max >C, the interference is judged to be comb spectrum interference, otherwise it is judged to be broadband noise interference.
Claims
1. A broadband interference identification method based on real-time spectrum analysis, characterized in that: Step 1: Perform short-time Fourier transform on the intermediate frequency data of the broadband interference signal to obtain a time-frequency matrix containing time-frequency-amplitude information; Step 2: Perform spectrum statistics on the time-frequency matrix and map it to obtain a probability density matrix containing frequency-amplitude-frequency information; Step 3: Conduct an electromagnetic spectrum survey on the probability density matrix and count the frequency of interference signals at each frequency point of the probability density matrix within the ambient noise amplitude range; Step 4: Set the interference detection threshold to detect each frequency point of the probability density matrix. The interference detection threshold is specifically described as: Where μ is the statistical mean of the noise background, P fa is the false alarm probability; according to the frequency of the interference signal counted in step 3, if the frequency point f j The frequency of the interference signal L fj If the interference detection threshold r1 is exceeded, it is determined that there is an interference signal at the current frequency point, and the number of frequency points with interference N is counted. r , the number of false alarms within the observation time T is specifically described as: N fa =T·f s ·P fa (2) In the formula, f s is the receiver sampling rate; if N r <N fa , then it is judged as no interference, if N r >N fa , it is determined that there is broadband interference, and the process goes to step 5; Step 5: Set the interference density threshold to detect each frequency point of the probability density matrix. The interference density detection threshold is specifically described as: Where S is the statistical mean of the interference amplitude; According to the frequency of interference signal in step 3, the judgment is made. If the frequency point f j The frequency of the interference signal L fj If the interference density threshold r2 is exceeded, it is determined to be the current frequency point f j There is an interference signal, frequency f j The threshold mark R(f j ) is set to 1, otherwise the frequency point f j The threshold mark R(f j ) is set to 0, and all frequency point over-threshold flags constitute the over-threshold flag sequence R(j); the number of frequency points with interference N is counted r , and the number of false alarms N within the observation time T fa Compare, if N r <N fa , it is determined to be a sweep frequency interference, if not, go to step 6; Step 6: Traverse the threshold-crossing mark sequence R(j) obtained in step 5, and calculate the frequency difference between adjacent threshold-crossing mark sequences: ΔR(j)=R(j)-R(j-1) (4) And find the maximum frequency difference ΔR max =max(ΔR(j)); According to the effective signal bandwidth B, the comb spectrum interference frequency interval B w And the number of short-time Fourier transform points N STFT , we can get the interference interval threshold of comb spectrum interference as: If ΔR max >C, it is judged as comb spectrum interference, otherwise it is judged as broadband noise interference.
2. The method for identifying broadband interference based on real-time spectrum analysis according to claim 1, characterized in that: In step 1, the discrete expression of the lth frame of the time-frequency matrix STFT (τ, ω) is: Where x represents the broadband interference signal, N STFT represents the short-time Fourier transform length, and H represents the jump step size between different frames.
3. The method for identifying broadband interference based on real-time spectrum analysis according to claim 2, characterized in that: In step 2, the probability density matrix P(Amp,f) represents the amplitude value Amp in this interference detection. i , the frequency value is f j The number of interference signals that occur, where Amp i Represents the i-th amplitude value on the amplitude dimension of the probability density matrix, 0≤i≤N Amp ,i∈N,N Amp =(Amp max -Amp min ) / ΔAmp,f j Represents the jth frequency value in the frequency dimension of the probability density matrix, where 0≤j≤N STFT ,j∈N; when the spectrum is counted, if ΔAmp*i≤STFT(t i ,f j )≤ΔAmp*(i+1), then the probability density matrix is in (Amp i ,f j ) at the statistical value P(Amp i ,f j ) plus 1.
4. The method for identifying broadband interference based on real-time spectrum analysis according to claim 3, characterized in that: The noise amplitude interval of the electromagnetic spectrum survey in step 3 is set to (N lower ,N upper ), traverse the probability density matrix P(Amp i ,f j ) at all frequencies f j , where 0≤j≤N f ,j∈N, each frequency point f j The frequency of interference signals in the upper amplitude range belonging to the noise amplitude range is:
Citation Information
Patent Citations
MGCSTFT-based chirp signal parameter estimation method
CN105785324A
Recognizing and classifying method for broadband radar interference signals
CN109459732A
