Robustness bilinear frequency modulation signal identification method under complex conditions

Abstract

The invention belongs to the technical field of radar radiation source signal intra-pulse analysis, and particularly relates to a robust bilinear frequency modulation signal identification method under complex conditions. Direct recognition of the bilinear frequency modulation signals is realized through signal time-frequency structure characteristic analysis, adaptive parameter setting, adaptive preprocessing, multi-feature-domain characteristic measurement, multi-index joint judgment and the like, the radar signal intra-pulse processing flow is greatly shortened, and influences of acquisition conditions, signal parameters, random interference and the like are avoided. The method can adapt to actual complex application scenes, and has the advantages of low calculation complexity, high robustness and the like.

Description

Technical Field

[0001] The invention belongs to the technical field of intra-pulse analysis of radar radiation source signals, and in particular relates to a robust dual linear frequency modulation signal recognition method under complex conditions. Background Art

[0002] In order to adapt to the complex electromagnetic environment, especially the increasingly severe electronic warfare signal environment, the current electronic reconnaissance processing system must have the ability to quickly analyze and process various signals in real time. In particular, it should have the ability to robustly and accurately identify the radar signal modulation type under complex conditions such as low signal-to-noise ratio caused by non-ideal factors in the transmission and reception links, the presence of interference signals, changes in transmission conditions, and changes in signal parameters.

[0003] However, existing methods are generally targeted at certain selected signal types, that is, the signal types in the candidate set are limited and cannot include all types. Therefore, when the possible signal type is unknown, the recognition results are often not completely reliable and cannot be directly extended to practical applications. Moreover, a single criterion is often used to adopt a serial process, that is, relying on a single indicator and combining this indicator to set up a special process to determine the type of signal one by one. However, a single indicator cannot completely distinguish all types, the algorithm has a limited scope of application, and the processing time is too long. The analysis ability of a certain modulation type signal is limited, and when faced with complex situations, especially when the signal acquisition is incomplete during reception and transmission, and there is noise, interference, etc., different signals need to be processed differently, which makes it difficult to apply directly.

[0004] In fact, we can adjust the processing ideas, determine the appropriate algorithm based on the signal type, and process specific types of signals specifically, that is, set up appropriate processes combined with signal structure characteristics, and perform multiple processing on a certain type of signal, with the ultimate goal of confirming whether the received signal is of a certain type. In this way, we can carry out targeted and detailed processing, and will not be affected by the increase in signal types, thus improving the robustness of the system.

[0005] Among many signal types, dual linear frequency modulation (DLFM) is a variation of linear frequency modulation. Its basic feature is that the signal can be divided into two segments of equal length. The first segment is a linear frequency modulation signal, and the second segment is a linear frequency modulation signal with an opposite frequency modulation slope to the previous segment. The signal frequency at the junction of the two segments is equal. Therefore, from the perspective of instantaneous frequency, the ideal dual linear frequency modulation signal is a V-shaped or inverted V-shaped signal. However, in actual applications, the phase and frequency will change to a certain extent after being affected by noise, interference, etc., and the instantaneous frequency has random fluctuations and glitches. It is difficult to identify directly according to the V-shaped signal, and it is easy to make a misjudgment. Therefore, in order to obtain accurate signal information under complex conditions and quickly and accurately determine whether the signal is a dual linear frequency modulation signal, it is necessary to carry out targeted processing and identification based on the modulation characteristics of the DLFM signal. Summary of the invention

[0006] In order to realize accurate and rapid recognition of dual linear frequency modulation signals in actual application scenarios, the present invention designs a robust dual linear frequency modulation signal recognition method under complex conditions in combination with the structural characteristics of DLFM signals in the time domain and frequency domain. Direct recognition of DLFM signals is realized by means of signal time-frequency structural characteristic analysis, adaptive parameter setting, adaptive preprocessing, multi-feature domain characteristic measurement, and multi-index joint judgment, which greatly shortens the radar signal pulse processing process and is not affected by acquisition conditions, signal parameters, random interference, etc. It can adapt to actual complex application scenarios and has the advantages of low computational complexity and high robustness.

[0007] The technical solution adopted by the present invention is: a robust dual linear frequency modulation signal recognition method under complex conditions, which is divided into the following steps:

[0008] S1. Signal reception

[0009] The received signal to be processed is represented by s0(n0), where n0 represents the index value of the sampling point, n0 = 1, 2, ..., N0, N0 represents the initial signal length, and the signal sampling rate is f s ;

[0010] S2. Adaptive filtering

[0011] S2.1 uses the real sinusoidal signal frequency estimation method based on DFT to estimate the peak frequency point f0 of the signal s0(n0) (for specific methods, see the document "Guo Luxiang. Research on real sinusoidal signal frequency estimation algorithm based on DFT [D], Southeast University, 2019.").

[0012] S2.2 estimates the bandwidth BW0 of the signal s0(n0), the specific steps are as follows:

[0013] S2.2.1 According to the non-parametric power spectrum estimation method, the power spectrum S(f) of the signal s0(n0) is obtained, where f represents the frequency index value of the power spectrum, and f = -f s / 2,...,f s / 2.

[0014] S2.2.2 Find the peak position f of S(f) max , and from f max Search to both sides and record the frequency value f when S(f) drops to 3dB of the peak value l and f r , f l Indicates the frequency value 3dB to the left of the peak, f r Indicates the frequency value 3dB to the right of the peak.

[0015] S2.2.3 Calculation of fl and f r The distance between them is obtained, and the bandwidth estimation value BW0 of the signal s0(n0) is obtained. r -f l .

[0016] S2.3 uses an FFT filter to adaptively filter the signal with f0 as the center frequency according to the estimated peak frequency position f0 and bandwidth BW0. The specific steps are as follows:

[0017] S2.3.1 Set the filter bandwidth factor δ and set the starting frequency of the frequency domain filter to f start =f0-δ·BW0, the cut-off frequency is f end =f0+δ·BW0. In practical applications, δ is generally set to 0.55-0.85.

[0018] S2.3.2 Perform N on s0(n0) FFT Fourier transform of the point, convert it to the frequency domain, and then keep f start and f end The frequency components between them are set to zero, and the remaining frequency components are all set to zero. Finally, an inverse Fourier transform is performed to obtain the filtered signal s1(n0), n0 = 1, 2, ..., N0, usually N FFT Take 1024.

[0019] S3. Pulse leading and trailing edge detection

[0020] S3.1 Detect the envelope starting point n of the filtered signal s1(n0) start and end point n end , the specific method is as follows:

[0021] S3.1.1 Calculate the signal envelope a0(n0) = s1(n0), smooth a0(n0) to obtain the smoothed envelope a(n0), and calculate the average value of the smoothed envelope

[0022] S3.1.2 Setting the signal detection threshold th A , the detection signal envelope exceeds the threshold a(n0)≥th A The starting point index value n corresponding to part A start and the end point index n end . Signal detection threshold th A Generally take th A =0.5.

[0023] S3.2 Using n start and n end Intercept the leading and trailing edges of signal s1(n0), that is, take the signal index value n0=n start ,...,n endThe corresponding part of the segment is taken as the intercepted signal s(n), where n represents the new signal index, n=1,2,...,N, N=n end -n start +1 indicates the new signal length after truncation.

[0024] S4. Instantaneous frequency calculation

[0025] S4.1 calculates the corresponding differential order L according to the signal length N. Different differential orders are set for signals of different lengths. According to engineering experience, the differential order L can generally be set adaptively according to the signal length:

[0026]

[0027] S4.2 estimates the instantaneous frequency of the signal s(n) according to the L-fold phase difference method. The specific calculation steps are as follows:

[0028] S4.2.1 Calculate the signal amplitude p0(n):

[0029]

[0030] Among them, s I (n), s Q (n) respectively represent the real and imaginary parts of the signal s(n): s(n) = s I (n)+i×s Q (n), i represents the imaginary unit, and Imag(·) represents the operation of taking an imaginary number.

[0031] S4.2.2 Use L-fold phase difference method to find the instantaneous frequency sequence p(n) to improve the adaptability to noise:

[0032]

[0033] S5. Basic characteristics judgment

[0034] Determine whether the signal has the basic characteristics of a dual-linear frequency modulation signal, that is, whether the instantaneous frequency presents an upper V-shape or a lower V-shape; the specific steps are as follows:

[0035] S5.1 performs summing and smoothing on the instantaneous frequency sequence p(n), with a smoothing order of λ, usually λ = 50, and calculates the smoothed sequence g(n) as follows:

[0036]

[0037] S5.2 Find the minimum and maximum values ​​of g(n), and replace the index corresponding to the maximum value with n max Indicates that the index corresponding to the minimum value is n min express;

[0038] S5.3 determines whether the signal is an upper V-shaped dual linear frequency modulation signal. If so, proceed to step S6; otherwise, proceed to step S5.4. The specific steps are as follows:

[0039] S5.3.1 Calculate the flag F1 as follows:

[0040]

[0041] Among them, ∧ represents the "AND" operation, and ∨ represents the "OR" operation.

[0042] If F1 is not zero, that is, the signal preliminarily meets the characteristics of an upper V-shaped dual linear frequency modulation signal, then proceed to S5.3.2; if F1 is zero, that indicates that the signal is not an upper V-shaped dual linear frequency modulation signal, then proceed to step S5.4.

[0043] S5.3.2 Preliminary identification of V-type dual linear frequency modulation signal:

[0044] S5.3.2.1 Compute the first-order difference of g(n):

[0045] S5.3.2.2 Take n = 2, ..., n min -1 corresponds to Part, calculation The number of elements less than 0 in

[0046] S5.3.2.3 Take n = n min ,...,N corresponds to Part, calculation The number of elements greater than 0 in

[0047] S5.3.2.4 Setting the bilinear threshold th N ,calculate and The sum of; In general, the bilinear threshold th N Set to N =0.65;

[0048] S5.3.2.5 If It is initially determined to be a dual-linear frequency modulation signal, and then the process goes to step S6 for subsequent judgment; if Then proceed to step S5.4.

[0049] S5.4 determines whether it is a lower V-type dual linear frequency modulation, and the specific steps are as follows:

[0050] S5.4.1 Calculate the flag F2 as follows:

[0051]

[0052] If F2 is not zero, that is, the signal preliminarily meets the characteristics of the lower V-type dual linear frequency modulation signal, then S5.4.2 is performed; if F1 is zero, it means that the signal is not an upper V-type dual linear frequency modulation signal, and the signal is determined to be other types, and the entire processing flow ends.

[0053] S5.4.2 Preliminary identification of V-type dual linear frequency modulation signal:

[0054] S5.4.2.1 If not Then calculate the first-order difference of g(n): If already exists Then

[0055] S5.4.2.2 Take n = 2, ..., n max -1 corresponds to Part, calculation The number of elements greater than 0 in

[0056] S5.4.2.3 Take n = n max ,...,N corresponds to Part, calculation The number of elements less than 0 in

[0057] S5.4.2.4 Setting the bilinear threshold th N ,calculate and The sum of; In general, the bilinear threshold th N Set to N =0.65;

[0058] S5.4.2.5 If It is initially determined to be a dual linear frequency modulation signal; if The signal is not a dual linear frequency modulation signal, but other types, and the entire processing flow ends.

[0059] S6. Segmented Line Fitting

[0060] The instantaneous frequency sequence p(n) is divided into two segments and linear fitting is performed based on the least squares estimation method. The specific steps are as follows:

[0061] S6.1 Divide the instantaneous frequency sequence p(n) into two equal parts, A and B, according to the index value. A (n) and p B (n):

[0062]

[0063] S6.2 Linear Fitting of Segment A: Use a straight line P A (n) is fitted and the fitting slope is obtained using the least squares estimation method. Fitting bias value and the fitted instantaneous frequency series For specific steps, please refer to "Steven M. Kay, Luo Pengfei. Fundamentals of Statistical Signal Processing [M]. Publishing House of Electronics Industry, 2014."

[0064] S6.3 Linear Fitting of Segment B: Use a straight line P B (n) is fitted and the fitting slope is obtained using the least squares estimation method. Fitting bias value and the fitted instantaneous frequency series

[0065] S6.4 Calculate the fitting error:

[0066]

[0067] S6.5 Set the error threshold th ε , th ε It is generally set to 0.015~0.1.

[0068] S6.6 Determine the error situation: If the fitting error is greater than the threshold th ε , then proceed to step S8; otherwise, perform time-frequency analysis in step S7 and update the fitting result.

[0069] S7. Time-frequency analysis

[0070] Perform time-frequency analysis on the signal s(n), and update the instantaneous frequency sequence p(n) based on the time-frequency analysis results to further reduce the impact of random factors such as noise interference and improve the robustness of the algorithm. The specific steps are as follows:

[0071] S7.1 Perform short-time Fourier transform on the signal s(n) to obtain the time-frequency matrix in Represents N×N FFT dimensional complex matrix;

[0072] S7.2 Calculate the energy of the time-frequency diagram matrix Μ0(n,f), that is, calculate the amplitude value point by point, and obtain Μ(n,f)=Μ0(n,f);

[0073] S7.3 Calculate the mean of M(n,f):

[0074] S7.4 extracts the time-frequency distribution. The specific steps are as follows:

[0075] S7.4.1 Setting the time-frequency energy threshold thM , generally set to 0.7~0.9.

[0076] S7.4.2 Compare M(n,f) at time n with The size of n will satisfy The total f-value is expressed as Indicates that at time n, The kth frequency value of , k = 1, 2, ..., K, K is the total number of frequencies that meet the conditions;

[0077] S7.4.3 Calculate the value of each column The mean of

[0078] S7.5 According to the method in step S6, the curve q(n) is fitted to obtain the corresponding fitting slope and Fitting bias value and Fitted instantaneous frequency series and The fitting error ε′ is further obtained;

[0079] S7.6 compares the size of ε′ and ε. If ε′<ε, update the fitting result in step S6, that is, update p(n) to q(n). Updated to Will Updated to Will Updated to Will Updated to Update ε to ε′;

[0080] S8. Overall judgment

[0081] S8.1 Calculation of the dual slope ratio

[0082] S8.2 Dual slope ratio judgment: If α≥0, the signal is not a dual linear frequency modulation signal, but should be other types, and the process ends; otherwise, proceed to step S8.3;

[0083] S8.3 Fitting error judgment: If ε>0.5·th ε , the signal is not a dual-linear frequency modulation signal, and the process ends; otherwise, proceed to step S8.4;

[0084] S8.4 Calculate the fit slope amplitude ratio

[0085] S8.5 Slope relationship judgment: If β-1<10, the signal is judged to be a bilinear frequency modulation signal; otherwise, the signal is of other types.

[0086] Technical effects:

[0087] The present invention proposes a robust dual-linear frequency modulation signal recognition method under complex conditions to adapt to the accurate recognition of dual-linear frequency modulation signals in actual non-ideal application scenarios. Specifically, it includes the design of a processing flow for dual-linear frequency modulation signal recognition, adaptive signal processing and adaptive parameter setting for complex transmission and reception conditions, and multiple judgment indicators and calculation methods based on the structural characteristics of dual-linear frequency modulation signals in the time domain and frequency domain, which shortens the processing flow of the existing radar intra-pulse analysis system and improves the accuracy and robustness of radar modulation type recognition.

[0088] The technical benefit of the present invention is to directly identify and confirm the type of dual-linear frequency modulation radar signal under complex conditions, and support the radar pulse analysis system to quickly and accurately identify radar targets in complex actual scenarios.

[0089] (1) A processing flow for directly identifying dual-linear frequency modulation signals under complex conditions is designed. The modulation slope estimation value, characteristic curve fitting error and other indicators are combined for comprehensive judgment. In order to further shorten the processing flow, the instantaneous frequency shape of the signal is firstly judged whether it meets the basic characteristics of the dual-linear frequency modulation signal. If it does not meet the requirements, it is directly judged as another type, which saves the subsequent calculation steps and shortens the processing time of signal judgment.

[0090] (2) The present invention designs a method for obtaining the instantaneous frequency domain curve of the signal based on time-frequency analysis, and uses the signal distribution in the time-frequency image to obtain the main trend of the instantaneous frequency of the signal to increase the robustness against noise interference and other situations;

[0091] (3) For two different types of dual-linear frequency modulation signals, a method is designed to determine whether the instantaneous frequency of the signal meets the basic characteristics of the dual-linear frequency modulation signal;

[0092] (4) Robust adaptive signal processing and adaptive parameter setting are specially performed for actual complex scenarios, including signal adaptive bandwidth filtering, leading and trailing edge detection, and time-frequency analysis combined with error conditions. In terms of threshold setting, adaptive threshold values ​​are set based on signal conditions, such as the signal modulation slope threshold th K ;

[0093] (5) In the final comprehensive judgment, a hierarchical judgment method is adopted, and the dual linear frequency modulation signal is finally judged step by step according to the conditions of dual slope ratio, fitting slope amplitude ratio, and slope relationship. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 It is a flow chart of the present invention;

[0095] Figure 2 is a flow chart of step S5 basic characteristic determination;

[0096] Figure 3 is a flowchart of step S6 piecewise straight line fitting;

[0097] Figure 4 is a flow chart of step S7 time-frequency analysis;

[0098] Figure 5 is a flow chart of the comprehensive decision of step S8;

[0099] Figure 6 It is the time domain waveform of DLFM signal when the signal-to-noise ratio is 10dB;

[0100] Figure 7 It is the frequency domain waveform of DLFM signal when the signal-to-noise ratio is 10dB;

[0101] Figure 8 It is the instantaneous frequency of the DLFM signal obtained by the first-order phase difference algorithm when the signal-to-noise ratio is 10dB;

[0102] Fig. 9 is the instantaneous frequency of the DLFM signal obtained by L-fold phase difference algorithm when the signal-to-noise ratio is 10dB;

[0103] Fig.10 It is the time-frequency diagram obtained by short-time Fourier transform of DLFM signal when the signal-to-noise ratio is 10dB;

[0104] Fig.11 It is the time domain waveform of DLFM signal when the signal-to-noise ratio is -5dB;

[0105] Fig.12 is the spectrum of the DLFM signal when the signal-to-noise ratio is -5dB;

[0106] Fig.13 is the instantaneous frequency of the DLFM signal obtained in step S4 when the signal-to-noise ratio is -5dB;

[0107] Fig.14 is the short-time Fourier transform time-frequency diagram of the DLFM signal after steps S2-S3 when the signal-to-noise ratio is -5dB;

[0108] Fig.15 is the time-frequency distribution of the filtered DLFM signal obtained by step S7.4 when the signal-to-noise ratio is -5dB;

[0109] Fig.16 is the instantaneous frequency of the filtered DLFM signal obtained in step S7 when the signal-to-noise ratio is -5dB;

[0110] Fig.17 It is the instantaneous frequency fitting result obtained by time-frequency analysis of the filtered DLFM signal when the signal-to-noise ratio is -5dB. DETAILED DESCRIPTION

[0111] The specific implementation modes of the present invention will be further described below in conjunction with the accompanying drawings.

[0112] Figure 1 : is a flow chart of the present invention. The present invention proposes a robust dual linear frequency modulation signal recognition method under complex conditions, which is divided into the following steps:

[0113] S1, signal reception, mainly refers to obtaining the signal to be analyzed;

[0114] S2, adaptive filtering, mainly estimates the frequency and bandwidth of the signal, and adaptively sets the filtering parameters according to the signal frequency and bandwidth to complete the signal filtering;

[0115] S3, pulse leading and trailing edge detection, mainly performs pulse detection on the signal obtained after filtering and intercepts the signal part;

[0116] S4, instantaneous frequency estimation, mainly calculates the instantaneous frequency of the signal based on the phase difference method;

[0117] S5, basic characteristic judgment, mainly preliminarily judges whether it is a dual linear frequency modulation signal based on the basic characteristics of the instantaneous frequency. If the basic conditions are not met, it is directly judged as other types and the processing flow is terminated in advance.

[0118] S6, piecewise straight line fitting, mainly fits the instantaneous frequency of the signal based on the least squares estimation method to obtain the fitting slope, fitting error, etc.;

[0119] S7, time-frequency analysis, mainly performs time-frequency analysis on the signal with high error in step S5, and updates the instantaneous frequency and its corresponding least square fitting result in combination with the characteristics of the time-frequency graph to improve the robustness of the algorithm;

[0120] S8, comprehensive judgment, final confirmation of the signal type, and determination of whether it is a dual-linear frequency modulation signal.

[0121] Figure 2 This is the flowchart of the basic characteristic judgment of step S5, which mainly examines whether the instantaneous frequency of the signal conforms to the basic characteristics of the dual-linear frequency modulation signal for judgment, that is, whether the instantaneous frequency of the signal observation signal generally presents an upper V shape or a lower V shape (steps S5.3.1 and S5.4.1). After basically meeting the requirements, it is further determined whether the two parts of the dual-linear frequency modulation corresponding to the signal are the main body of the signal (steps S5.3.2 and S5.4.2). If the requirements are met, the corresponding label F DLFM=1, otherwise the signal is of other types and the discrimination process ends directly.

[0122] Figure 3 This is a flowchart of the segmented straight line fitting of step S6. It is mainly used to divide the signal into two segments, A and B, and perform straight line fitting on these two segments based on the least squares estimation method to obtain the corresponding modulation slope and fitting error, so as to support subsequent signal discrimination. At the end of step S6, the fitting error is compared with the set threshold value. If the error is higher than the threshold value, it indicates that the signal fitting effect is not good, which may be caused by interference or low signal-to-noise ratio of the signal, so the time-frequency analysis of step S7 is entered to enhance the robustness of the system.

[0123] Figure 4 This is the flowchart of step S7 time-frequency analysis. In this step, the signal with a large fitting error in the previous piecewise straight line fitting process is further processed to increase the robustness to noise interference and other situations, that is, the corresponding instantaneous frequency characteristics are extracted using the time-frequency image. The specific process is to perform a short-time Fourier transform on the signal, calculate the representative frequency value at each moment based on the time-frequency matrix, combine to form a new instantaneous frequency, and then perform piecewise straight line fitting again according to the method of step S6, and compare the two fitting errors; if the fitting error corresponding to the instantaneous frequency curve obtained by the time-frequency analysis is small, the fitting instantaneous frequency, fitting parameters and fitting error obtained at this time are used in subsequent steps.

[0124] Figure 5 This is a flowchart of the comprehensive judgment of step S8. It mainly examines in turn whether the dual slope ratio, the fitting slope amplitude ratio, and the slope relationship meet the characteristics of the dual linear frequency modulation signal, and performs hierarchical and progressive judgment step by step, and finally identifies the dual linear frequency modulation signal.

[0125] Figure 6 This is the time domain waveform of the DLFM signal when the signal-to-noise ratio is 10dB. This figure shows the image of the signal amplitude of the simulated dual linear frequency modulation DLFM signal s0(n0) in the time domain changing with time when the signal-to-noise ratio is 10dB. The horizontal axis in the figure is time in seconds; the vertical axis is the amplitude value of the signal. There is a zero point before and after the signal.

[0126] Figure 7 This is the frequency domain waveform of the DLFM signal when the signal-to-noise ratio is 10dB. It can be seen that the signal includes multiple frequency points, and the amplitude values ​​of different frequency points are similar, and the spectrum is close to a rectangle, which means that the DLFM signal is composed of more continuous frequencies, and the energy at different frequency points is similar.

[0127] Figure 8It is the instantaneous frequency of the DLFM signal obtained by the first-order phase difference algorithm when the signal-to-noise ratio is 10dB. This figure shows the instantaneous frequency of the signal after the DLFM signal is directly processed by the first-order phase difference when the signal-to-noise ratio is 10dB. The horizontal axis represents the sampling point and the vertical axis represents the normalized instantaneous frequency. It can be seen that the curve in the figure can generally show the main trend of the instantaneous frequency of the signal, that is, the two inclined straight lines with opposite slopes basically have the basic characteristics of the dual linear frequency modulation signal. Affected by noise, there is a lot of noise on the instantaneous frequency.

[0128] Fig. 9 It is the instantaneous frequency of the DLFM signal obtained by the L-phase difference algorithm when the signal-to-noise ratio is 10dB. The horizontal axis represents the sample point index of the signal, and the vertical axis is the frequency value. The figure shows the instantaneous frequency curve image of the dual linear frequency modulation signal at 10dB calculated by the L-phase difference method in step S4. It should be noted that this signal is an ideal signal and has not undergone the adaptive filtering processing of step S2, nor the leading and trailing edge detection of step S3. Instead, the received signal is directly subjected to step S4 to obtain the result. It can be seen that the figure consists of two inclined straight lines with opposite slopes, and the horizontal segment at the end of the curve is caused by the continuous zero points after the signal.

[0129] Fig.10 It is a time-frequency diagram obtained by short-time Fourier transform of DLFM signal when the signal-to-noise ratio is 10dB. The horizontal axis of the figure is time, the unit is second, and the vertical axis is frequency value, the unit is Hz. Affected by noise, there are randomly scattered noise points in the figure, but in the time-frequency diagram, two inclined straight lines with opposite slopes can be clearly identified as dual linear frequency modulation signals. It should be noted that this figure does not perform adaptive narrowband filtering on the signal, but the original received signal is directly transformed by short-time Fourier transform in steps S3 and S7.2 to obtain M (n, f).

[0130] Fig.11 This is the time domain waveform of the DLFM signal when the signal-to-noise ratio is -5dB. The dual linear frequency modulation signal in this figure is accompanied by additive white Gaussian noise, and the signal-to-noise ratio is -5dB. Figure 5 Compared with the time domain waveform of the signal with a signal-to-noise ratio of 10dB, it can be seen that the signal edge changes greatly, and the signal and the zero point boundaries before and after are not obvious, indicating that there is little difference between the signal and the electromagnetic environment noise in the time domain, and it is difficult to directly detect and distinguish the signal.

[0131] Fig.12 This figure shows the spectrum of the dual linear frequency modulation signal when the signal-to-noise ratio is -5dB. It can be seen that the signal has a certain bandwidth. Figure 6Compared with the spectrum of the 10dB signal, the noise floor level of the signal has increased significantly. However, compared with the signal waveform in the time domain, the signal position can still be clearly located in the frequency domain, and the noise and signal can be distinguished.

[0132] Fig.13 is the instantaneous frequency of the DLFM signal obtained by step S4 when the signal-to-noise ratio is -5dB. The figure shows the instantaneous frequency curve image of the DLFM signal when the signal-to-noise ratio is -5dB, calculated by the L-phase difference method in step S4. The horizontal axis represents the sample point index of the signal, and the vertical axis is the frequency value. It should be noted that the signal here has not been processed by the adaptive filtering of step S2, nor has the leading and trailing edge detection of step S3 been performed. Instead, the result obtained by step S4 is directly performed on the received signal. It can be seen that due to the influence of noise, there are a lot of large fluctuations on the instantaneous frequency curve, which is different from the Fig. 9 Compared with the medium instantaneous frequency, the instantaneous frequency has changed significantly under low SNR and cannot be directly determined as a dual linear frequency modulation signal, indicating that the signal type cannot be determined directly based on the instantaneous phase under low SNR.

[0133] Fig.14 It is the short-time Fourier transform time-frequency diagram of the DLFM signal after steps S2-S3 when the signal-to-noise ratio is -5dB. This figure is the time-frequency image obtained by short-time Fourier transform of the signal after steps S2-S3. The horizontal axis is time in seconds, and the vertical axis is the frequency value in Hz. It can be seen that after filtering and detection, some noise points are left near the signal. These residual noise points can be eliminated later through step S7.4 to achieve adaptation to the noise environment.

[0134] Fig.15 is the time-frequency distribution of the filtered DLFM signal obtained by step S7.4 when the signal-to-noise ratio is -5dB. Specifically, the figure is a signal with a signal-to-noise ratio of -5dB directly proceeding to step S7 after steps S2 and S3. The corresponding value in the figure is the value that satisfies Μ(n,f)≥th in step S7.4.2 M All frequency points of M. It can be seen that compared with Fig.14 There are some snowflakes caused by noise left in the image. Fig.15 Only the frequency part corresponding to the signal is retained. The instantaneous frequency extracted on this basis can better suppress noise, thereby achieving a robust effect.

[0135] Fig.16 is the instantaneous frequency of the filtered DLFM signal obtained by step S7 when the signal-to-noise ratio is -5dB. Fig.15 The instantaneous frequency q(n) is obtained based on the result after step S7.4.3. It can be seen that compared with Fig. 9 and Fig.13 The instantaneous frequency image in is Fig.16The middle curve is two inclined straight lines with slight fluctuations, which have obvious characteristics of a bilinear frequency modulation signal and can be directly determined as a bilinear frequency modulation signal, indicating that this method has strong signal feature extraction capability and robustness.

[0136] Fig.17 This is the instantaneous frequency fitting result of the filtered DLFM signal obtained by time-frequency analysis when the signal-to-noise ratio is -5dB. Fig.16 The instantaneous frequency in is taken as p(n) in step S7, and then the fitting result is performed based on the least squares algorithm. The long and short alternating curved lines in the figure represent the original instantaneous frequency p(n), and the dotted line composed of a continuous single short line represents the fitting result of the least squares algorithm. It can be seen that the least squares algorithm can better fit the overall trend of the original instantaneous frequency curve, and is not affected by noise, and has good robustness, which shows the effectiveness of the present invention.

Claims

1. A robust dual linear frequency modulation signal recognition method under complex conditions, characterized in that: The method is divided into the following steps: S1. Signal reception The received signal to be processed is represented by s0(n0), where n0 represents the index value of the sampling point, n0 = 1, 2, ..., N0, N0 represents the initial signal length, and the signal sampling rate is f s ; S2. Adaptive filtering S2.1 uses the real sinusoidal signal frequency estimation method based on DFT to estimate the peak frequency position f0 of the signal s0(n0); S2.2 estimates the bandwidth BW0 of the signal s0(n0), the specific steps are as follows: S2.2.1 According to the non-parametric power spectrum estimation method, the power spectrum S(f) of the signal s0(n0) is obtained, where f represents the frequency index value of the power spectrum, and f = -f s / 2,...,f s / 2; S2.2.2 Find the peak position f of S(f) max , and from f max Search to both sides and record the frequency value f when S(f) drops to 3dB of the peak value l and f r , f l Indicates the frequency value 3dB to the left of the peak, f r Indicates the frequency value 3dB to the right of the peak; S2.2.3 Calculation of f l and f r The distance between them is obtained, and the bandwidth estimation value BW0 of the signal s0(n0) is obtained. r -f l ; S2.3 uses an FFT filter to adaptively filter the signal with f0 as the center frequency according to the estimated peak frequency position f0 and bandwidth BW0. The specific steps are as follows: S2.3.1 Set the filter bandwidth factor δ and set the starting frequency of the frequency domain filter to f start =f0-δ·BW0, the cut-off frequency is f end =f0+δ·BW0; S2.3.2 Perform N on s0(n0) FFT Fourier transform of the point, convert it to the frequency domain, and then keep f start and f end The frequency components between them are set to zero, and the remaining frequency components are all set to zero. Finally, an inverse Fourier transform is performed to obtain the filtered signal s1(n0), n0=1,2,...,N0; S3. Pulse leading and trailing edge detection S3.1 Detect the envelope starting point n of the filtered signal s1(n0) start and end point n end , the specific method is as follows: S3.1.1 Calculate the signal envelope a0(n0) = |s1(n0)|, smooth a0(n0) to obtain the smoothed envelope a(n0), and calculate the average value of the smoothed envelope S3.1.2 Setting the signal detection threshold th A , the detection signal envelope exceeds the threshold a(n0)≥th A The starting point index value n corresponding to part A start and the end point index n end ; S3.2 Using n start and n end Intercept the leading and trailing edges of signal s1(n0), that is, take the signal index value n0=n start ,...,n end The corresponding part of the segment is taken as the intercepted signal s(n), where n represents the new signal index, n=1,2,...,N, N=n end -n start +1 indicates the new signal length after interception; S4. Instantaneous frequency calculation S4.1 calculates the corresponding differential order L according to the signal length N. Different differential orders are set for signals of different lengths. S4.2 estimates the instantaneous frequency of the signal s(n) according to the L-fold phase difference method. The specific calculation steps are as follows: S4.2.1 Calculate the signal amplitude p0(n): Among them, s I (n), s Q (n) respectively represent the real and imaginary parts of the signal s(n): s(n) = s I (n)+i×s Q (n), i represents the imaginary unit, and Imag(·) represents the operation of taking imaginary numbers; S4.2.2 Use L-fold phase difference method to find the instantaneous frequency sequence p(n) to improve the adaptability to noise: S5. Basic characteristics judgment Determine whether the signal has the basic characteristics of a dual-linear frequency modulation signal, that is, whether the instantaneous frequency presents an upper V-shape or a lower V-shape; the specific steps are as follows: S5.1 performs summing and smoothing on the instantaneous frequency sequence p(n), with a smoothing order of λ, and obtains the smoothed sequence g(n) by the following calculation method: S5.2 Find the minimum and maximum values ​​of g(n), and replace the index corresponding to the maximum value with n max Indicates that the index corresponding to the minimum value is n min express; S5.3 determines whether the signal is an upper V-shaped dual linear frequency modulation signal. If so, proceed to step S6; otherwise, proceed to step S5.

4. The specific steps are as follows: S5.3.1 Calculate the flag F1 as follows: Among them, ∧ represents the "and" operation, and ∨ represents the "or" operation; If F1 is not zero, that is, the signal preliminarily meets the characteristics of an upper V-shaped dual linear frequency modulation signal, then proceed to step S5.3.2; if F1 is zero, then it means that the signal is not an upper V-shaped dual linear frequency modulation signal, then proceed to step S5.4; S5.3.2 Preliminary identification of V-type dual linear frequency modulation signal: S5.3.2.1 Compute the first-order difference of g(n): S5.3.2.2 Take n = 2, ..., n min -1 corresponds to Part, calculation The number of elements less than 0 in S5.3.2.3 Take n = n min ,...,N corresponds to Part, calculation The number of elements greater than 0 in S5.3.2.4 Setting the bilinear threshold th N ,calculate and of and; S5.3.2.5 If It is initially determined to be a dual-linear frequency modulation signal, and then the process goes to step S6 for subsequent judgment; if Then proceed to step S5.4; S5.4 determines whether it is a lower V-type dual linear frequency modulation, and the specific steps are as follows: S5.4.1 Calculate the flag F2 as follows: If F2 is not zero, that is, the signal preliminarily meets the characteristics of the lower V-type dual linear frequency modulation signal, then S5.4.2 is performed; if F1 is zero, it means that the signal is not an upper V-type dual linear frequency modulation signal, and the signal is determined to be other types, and the entire processing flow ends; S5.4.2 Preliminary identification of V-type dual linear frequency modulation signal: S5.4.2.1 If not Then calculate the first-order difference of g(n): If already exists Then S5.4.2.2 Take n = 2, ..., n max -1 corresponds to Part, calculation The number of elements greater than 0 in S5.4.2.3 Take n = n max ,...,N corresponds to Part, calculation The number of elements less than 0 in S5.4.2.4 Setting the bilinear threshold th N ,calculate and of and; S5.4.2.5 If It is initially determined to be a dual linear frequency modulation signal; if Then the signal is not a dual linear frequency modulation signal, but other types, and the whole processing flow ends; S6. Segmented Line Fitting The instantaneous frequency sequence p(n) is divided into two segments and linear fitting is performed based on the least squares estimation method. The specific steps are as follows: S6.1 Divide the instantaneous frequency sequence p(n) into two equal parts, A and B, according to the index value. A (n) and p B (n): S6.2 Linear Fitting of Segment A: Use a straight line P A (n) is fitted and the fitting slope is obtained using the least squares estimation method. Fitting bias value And the fitted instantaneous frequency series S6.3 Linear Fitting of Segment B: Use a straight line P B (n) is fitted and the fitting slope is obtained using the least squares estimation method. Fitting bias value And the fitted instantaneous frequency series S6.4 Calculate the fitting error: S6.5 Set the error threshold th ε ; S6.6 Determine the error situation: If the fitting error is greater than the threshold th ε , then proceed to step S8; otherwise, perform the time-frequency analysis of step S7 and update the fitting result; S7. Time-frequency analysis Perform time-frequency analysis on the signal s(n), and update the instantaneous frequency sequence p(n) based on the time-frequency analysis results to further reduce the influence of random factors such as noise interference and improve the robustness of the algorithm. The specific steps are as follows: S7.1 Perform short-time Fourier transform on the signal s(n) to obtain the time-frequency matrix in Represents N×N FFT dimensional complex matrix; S7.2 Calculate the energy of the time-frequency diagram matrix Μ0(n,f), that is, calculate the amplitude value point by point, and obtain Μ(n,f)=Μ0(n,f); S7.3 Calculate the mean of M(n,f): S7.4 extracts the time-frequency distribution. The specific steps are as follows: S7.4.1 Setting the time-frequency energy threshold th M ; S7.4.2 Compare M(n,f) at time n with The size of n will satisfy The total f-value is expressed as Indicates that at time n, The kth frequency value of , k = 1, 2, ..., K, K is the total number of frequencies that meet the conditions; S7.4.3 Calculate the value of each column The mean of S7.5 According to the method in step S6, the curve q(n) is fitted to obtain the corresponding fitting slope and Fitting bias value and Fitted instantaneous frequency series and The fitting error ε′ is further obtained; S7.6 compares the size of ε′ and ε. If ε′<ε, update the fitting result in step S6, that is, update p(n) to q(n). Updated to Will Updated to Will Updated to Will Updated to Update ε to ε′; S8. Overall judgment S8.1 Calculation of the dual slope ratio S8.2 Dual slope ratio judgment: If α≥0, the signal is not a dual linear frequency modulation signal, but should be other types, and the process ends; otherwise, proceed to step S8.3; S8.3 Fitting error judgment: If ε>0.5·th ε , the signal is not a dual-linear frequency modulation signal, and the process ends; otherwise, proceed to step S8.4; S8.4 Calculate the fit slope amplitude ratio S8.5 Slope relationship judgment: If |β-1|<10, the signal is judged to be a dual linear frequency modulation signal; otherwise, the signal is of other types.

2. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S2.3.1, the filter bandwidth factor δ is set to 0.55-0.

85.

3. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S2.3.2, N FFT Take 1024.

4. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S3.1.2, the signal detection threshold th A =0.

5.

5. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S4.1, the differential order L is adaptively set according to the signal length:

6. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S5.1, the smoothing order λ=50.

7. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S5.3.2.4, the bilinear threshold th N =0.

65.

8. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S5.4.2.4, the bilinear threshold th N =0.

65.

9. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S6.5, the error threshold th ε Set to 0.015~0.

1.

10. A robust bilinear frequency modulation signal recognition method under complex conditions according to claim 1, characterized in that: In S7.4.1, the time-frequency energy threshold th M Set to 0.7~0.9.

Images