CN105022038A - External radiation source radar wind field clutter suppression method based on morphological component analysis - Google Patents

Abstract

The invention discloses an external radiation source radar wind field clutter suppression method based on morphological component analysis. A situation that a target signal and wind field clutter in the echo received by radar have different morphological characteristics is utilized, and sparse representation can be performed in corresponding transform domains so that signal separation can be performed by utilizing the morphological component analysis method. Wind field clutter can be greatly suppressed by the method. Meanwhile, the method has advantages of being great in stability and easy and convenient in calculation so that external radiation source radar target detection performance can be enhanced.

Description

A kind of external illuminators-based radar wind field clutter suppression method based on form component analysis

Technical field

The invention belongs to external illuminators-based radar signal processing technology field, particularly relate to a kind of external illuminators-based radar wind field clutter suppression method based on form component analysis.

Background technology

External illuminators-based radar (also known as passive radar) is the Bistatic/Multistatic Radar System that electromagnetic signal that a kind of third party of utilization launches carries out target detection, the not direct radiated electromagnetic energy of this radar self, but by reception from the reflection wave formed after the direct wave of outside miscoordination radiation and irradiation target thereof or scattering wave, complete the detection to target, location and tracking.Because available foreign radiation sources (comprising broadcasting station, TV station and mobile communication base station etc.) has spatial domain, time domain wide coverage and base station configuring redundancy degree high, therefore, external illuminators-based radar is except possessing the advantage such as with low cost, good concealment, anti-lethality are strong, normal radar low-altitude detection blind area can also be reduced, realize the covering on a large scale to low latitude moving target, and be conducive to stealthy target.These particular advantages make it become current new system radar research in focus and emphasis.

In the practical application of radar, when target self or its parts exist fine motion, can produce frequency modulation (PFM) to radar echo signal, generate the Doppler sideband about target subject, this modulation phenomenon is called micro-Doppler effect.In recent years, whole world Wind Power Development is rapid, aerogenerator has long and wide blade, and have higher rotating speed, the radar return produced has stronger amplitude, and there is wide Doppler frequency spectrum on frequency domain, the Doppler sideband of Qiang Erkuan causes target to be covered, and forms interference to Radar Targets'Detection.External illuminators-based radar is a kind of new system radar, Present Domestic is outer mainly concentrates on target detection, location and tracking to the research of external illuminators-based radar, and it is less to wind field clutter Study on Problems, in addition, existing various external illuminators-based radar time domains, spatial domain Clutter suppression algorithm can only suppress static clutter, effectively cannot suppress the wind field clutter of non-zero-frequency composition.

Summary of the invention

The present invention, in order to solve above-mentioned technical matters, proposes a kind of external illuminators-based radar wind field clutter suppression method based on form component analysis, provides a kind of new solution for improving external illuminators-based radar target detection performance.

Technical scheme of the present invention is: a kind of external illuminators-based radar wind field clutter suppression method based on form component analysis, comprises the steps:

Step 1, carries out reference signal reconstruct respectively to reference channel, monitoring channel and multipath clutter suppresses, then computing reference passage and monitoring channel two-dimensional cross correlation function;

Step 2, whether exist along Doppler's axle and zero Doppler's symmetry, the multiple harmonic submaximum that occurs in the cycle; If existed, then need to carry out wind field clutter recognition; If there is no, then show without wind field clutter;

Step 3, when exist along Doppler's axle and zero Doppler's symmetry be the multiple harmonic submaximum occurred in the cycle time, obtain sparse coefficient by solving following constrained optimization problem with

$\{{\hat{c}}_1,{\hat{c}}_2\}=\arg \underset{c_1,c_2}{\min }{\mathrm{\λ}}_1\left|\right|c_1|{|}_1+{\mathrm{\λ}}_2|\left|c_2\right|{|}_1$

s.t.y＝A ₁c ₁+A c ₂

Wherein, c ₁and c ₂it is sparse coefficient; with c respectively ₁and c ₂optimum estimate; λ ₁, λ ₂for parameter, and λ ₁=1-λ ₂; Y is the composite signal that radar receives; A ₁, A ₂be respectively Fourier transform and Short Time Fourier Transform;

Then according to estimate obtain component of signal thus realize Signal separator.

Preferably, in described step 3, best sparse coefficient with solved by following sub-step:

Step 3.1, initializing variable: iterations initial value k=0, punishment parameter μ > 0, sparse coefficient initial value c to be solved _1,0, c _2,0, iterative process intermediate variable initial value d _1,0, d _2,0;

Step 3.2, more new variables:

v _i,k+1＝soft(c _i,k+d _i,k,λ _i/μ)-d _i,k，i＝1,2

g _k+1＝y-A ₁v _1,k+1-A ₂v _2,k+1

$d_{i,k+1}=\frac{1}{2}{A}_i^H{g}_{k+1},i=1,2$

c _i,k+1＝d _i,k+1+v _i,k+1，i＝1,2

Wherein, soft (.) is soft-threshold function, soft (x, T)=max (1-T/|x|, 0), the variable in x, T difference soft-threshold function, | x| is the absolute value of variable x; v _{i, k+1}, g _k+1, d _{i, k+1}, c _{i, k+1}represent the intermediate variable of i-th component of signal when carrying out kth+1 iteration respectively; A ₁, A ₂be respectively Fourier transform and Short Time Fourier Transform; H is transposed complex conjugate;

Step 3.3, judges whether to meet end condition, satisfied then stop; Otherwise, k=k+1.

The invention has the beneficial effects as follows: a kind of external illuminators-based radar wind field clutter suppression method based on form component analysis, utilize radar to receive echo signal in echo, from wind field clutter, there is different morphological characters, and all can carry out rarefaction representation at corresponding transform domain, thus the method for form component analysis can be utilized to be separated signal; Method of the present invention makes wind field clutter better be suppressed, and has good stability simultaneously, calculates the advantages such as easy, improves external illuminators-based radar target detection performance.

Accompanying drawing explanation

Fig. 1 is embodiment of the present invention Wind Field clutter recognition process the general frame;

Fig. 2 is based on multiple frequency network, MFN signal experiment sending and receiving site location layout in the embodiment of the present invention;

Fig. 3-1 is that in the embodiment of the present invention 1, multipath clutter suppresses front distance Doppler spectrogram;

Fig. 3-2 is that in the embodiment of the present invention 1, multipath clutter suppresses rear range Doppler spectrogram;

Fig. 3-3 is that in the embodiment of the present invention 1, multipath clutter suppresses rear range Doppler spectrum partial enlarged drawing;

Fig. 3-4 is the general figure of range Doppler after the embodiment of the present invention 1 fan blade clutter recognition;

Fig. 4 is based on multiple frequency network, MFN signal experiment sending and receiving site location layout in the embodiment of the present invention;

Fig. 5-1 is that in the embodiment of the present invention 2, multipath clutter suppresses front distance Doppler spectrogram;

Fig. 5-2 is that in the embodiment of the present invention 2, multipath clutter suppresses rear range Doppler spectrogram;

Fig. 5-3 is that in the embodiment of the present invention 2, multipath clutter suppresses rear range Doppler spectrum partial enlarged drawing;

Fig. 5-4 is the general figure of range Doppler after the embodiment of the present invention 2 fan blade clutter recognition.

Embodiment

Below in conjunction with drawings and Examples, the invention will be further described.

The present embodiment verifies validity of the present invention by outfield experiments.

Fig. 1 is embodiment of the present invention Wind Field clutter recognition process the general frame, is specially:

Based on an external illuminators-based radar wind field clutter suppression method for form component analysis, comprise the steps:

Step 1, carries out reference signal reconstruct respectively to reference channel, monitoring channel and multipath clutter suppresses, then computing reference passage and monitoring channel two-dimensional cross correlation function;

Step 2, whether exist along Doppler's axle and zero Doppler's symmetry, the multiple harmonic submaximum that occurs in the cycle; If existed, then need to carry out wind field clutter recognition; If there is no, then show without wind field clutter;

Step 3, when exist along Doppler's axle and zero Doppler's symmetry be the multiple harmonic submaximum occurred in the cycle time, obtain sparse coefficient by solving following constrained optimization problem with

$\{{\hat{c}}_1,{\hat{c}}_2\}=\arg \underset{c_1,c_2}{\min }{\mathrm{\λ}}_1\left|\right|c_1|{|}_1+{\mathrm{\λ}}_2|\left|c_2\right|{|}_1$

s.t.y＝A ₁c ₁+A c ₂

Wherein, c ₁and c ₂it is sparse coefficient; with c respectively ₁and c ₂optimum estimate; λ ₁, λ ₂for parameter, and λ ₁=1-λ ₂; Y is the composite signal that radar receives; A ₁, A ₂be respectively Fourier transform and Short Time Fourier Transform;

Then according to estimate obtain component of signal thus realize Signal separator.

In described step 3, best sparse coefficient with solved by following sub-step:

Step 3.1, initializing variable: iterations initial value k=0, punishment parameter μ > 0, sparse coefficient initial value c to be solved _1,0, c _2,0, iterative process intermediate variable initial value d _1,0, d _2,0;

Step 3.2, more new variables:

v _i,k+1＝soft(c _i,k+d _i,k,λ _i/μ)-d _i,k，i＝1,2

g _k+1＝y-A ₁v _1,k+1-A ₂v _2,k+1

$d_{i,k+1}=\frac{1}{2}{A}_i^H{g}_{k+1},i=1,2$

c _i,k+1＝d _i,k+1+v _i,k+1，i＝1,2

Wherein, soft (.) is soft-threshold function, soft (x, T)=max (1-T/|x|, 0), the variable in x, T difference soft-threshold function, | x| is the absolute value of variable x; v _{i, k+1}, g _k+1, d _{i, k+1}, c _{i, k+1}represent the intermediate variable of i-th component of signal when carrying out kth+1 iteration respectively; A ₁, A ₂be respectively Fourier transform and Short Time Fourier Transform; H is transposed complex conjugate;

Step 3.3, judges whether to meet end condition, satisfied then stop; Otherwise, k=k+1.

Test reception area scene in the embodiment of the present invention and be positioned at electronic information institute of Wuhan University experimental center roof, 3 blade fans of one vertical rotation in surface are in order to wind-driven generator simulation, fan leaf length of a film is 0.75m, during experiment, rotating speed is about 280-290rpm, and unit 8 Homogeneous Circular array configurations is monitoring channel.

Embodiment 1:

Fig. 2 is based on multiple frequency network, MFN signal experiment sending and receiving site location layout in the embodiment of the present invention.Cell site of Wuhan City is positioned at Guishan Mountain television tower, and sending and receiving site distance is from being 7.9km, and the centre frequency that transmits is 754MHz, bandwidth 8MHz, emissive power 3kW.

For the ease of performance evaluation, inject at the 3rd distance element (corresponding bistatic distance is 79.37m) of measured data, the 20th Doppler unit (corresponding bistatic Doppler is 39.84Hz) place the moving-target signal that signal to noise ratio is-45dB, ensure to inject the clutter district that target is positioned at fan blade generation.

Fig. 3-1 to Fig. 3-4 is the result of collected one group of data, and wherein, Fig. 3-1 is the range Doppler spectrum before multipath clutter suppression, visible, except zero Doppler locates to exist except strong multipath clutter in figure, cannot obtain other any effective information.Employing spatial domain method carries out the result after multipath clutter suppression as shown in figure 3-2, after the visible multipath clutter by zero Doppler place suppresses, echo spectrum substrate reduces, now there is multiple harmonic submaximum along Doppler's axle cycle around zero Doppler, this is because fan does not have translation motion, the fan blade rotated creates extra frequency modulation (PFM) in radar return, shows that the feature on frequency spectrum produces Harmonic lines exactly around zero Doppler.Fig. 3-3 is range Doppler spectrum partial enlarged drawing after multipath clutter suppression, and because fan blade produces strong and wide sideband, target is covered, and correctly cannot detect real goal.Utilize the inventive method to suppress the clutter of fan blade generation further, as shown in Figure 3-4, now fan blade clutter is eliminated result, and target displays clearly.

Embodiment 2:

Fig. 4 is based on multiple frequency network, MFN signal experiment sending and receiving site location layout in the embodiment of the present invention.Wuhan single frequency network signals centre frequency is 658MHz, signal bandwidth 8MHz, adopts vertical polarization, emissive power 1kW.For the ease of performance evaluation, inject at the 26th distance element (corresponding bistatic distance is 750m) of measured data, the 20th Doppler unit (corresponding bistatic Doppler is 39.35Hz) place the moving-target signal that signal to noise ratio is-60dB, ensure to inject the clutter district that target is positioned at fan blade generation.

Fig. 5-1 to Fig. 5-4 is the result of wherein one group of typical data, wherein, Fig. 5-1 is the range Doppler spectrum before multipath clutter suppression, and its characteristic feature locates to there is strong multipath clutter zero Doppler, and all the other information are submerged under the secondary lobe of strong multipath clutter interference.Spatial processing method is adopted to carry out multipath clutter suppression, result after suppression is as shown in Fig. 5-2, can see that the clutter of zero doppler position is effectively suppressed, but now about zero-frequency, there is multiple harmonic submaximum in the cycle along Doppler's axle, and also there is expansion along distance axis, this and the clutter feature utilizing multiple frequency network, MFN signal to produce as fan blade during irradiation source are very different.This is because under SFN structure, a target may be irradiated by multiple irradiation source simultaneously and produce multiple measuring value; Secondly, SFN is again the serious wireless environment of a multipath conditions, and the Doppler sideband being distributed in different distance cell position also may be caused by strong multipath component.Therefore, under SFN structure, blade of wind-driven generator can produce more serious noise jamming, and Fig. 5-3 shows the range Doppler spectrum partial enlarged drawing after multipath clutter suppression, the clutter that visual target and fan blade produce mixes, and cannot effectively distinguish.Adopt the inventive method to suppress fan clutter, result is as shown in Fig. 5-4, and now, fan blade clutter is eliminated, and target is high-visible.

Embodiment of the present invention centre field Data Processing in Experiment result demonstrates the validity of the inventive method.

Specific embodiment described in the present invention is only to the explanation for example of the present invention's spirit.Those skilled in the art can make various amendment or supplement or adopt similar mode to substitute to described specific embodiment, but can't depart from spirit of the present invention or surmount the scope that appended claims defines.

Claims (2)

1., based on an external illuminators-based radar wind field clutter suppression method for form component analysis, it is characterized in that, comprise the steps:

Step 1, carries out reference signal reconstruct respectively to reference channel, monitoring channel and multipath clutter suppresses, then computing reference passage and monitoring channel two-dimensional cross correlation function;

Step 2, whether exist along Doppler's axle and zero Doppler's symmetry, the multiple harmonic submaximum that occurs in the cycle; If existed, then need to carry out wind field clutter recognition; If there is no, then show without wind field clutter;

Step 3, when exist along Doppler's axle and zero Doppler's symmetry be the multiple harmonic submaximum occurred in the cycle time, obtain sparse coefficient by solving following constrained optimization problem with

$\{{\hat{c}}_1,{\hat{c}}_2\}=\arg \underset{c_1,c_2}{\min }{\mathrm{\λ}}_1\left|\right|c_1|{|}_1+{\mathrm{\λ}}_2|\left|c_2\right|{|}_1$

s.t.y＝A ₁c ₁+A c ₂

Wherein, c ₁and c ₂it is sparse coefficient; with c respectively ₁and c ₂optimum estimate; λ ₁, λ ₂for parameter, and λ ₁=1-λ ₂; Y is the composite signal that radar receives; A ₁, A ₂be respectively Fourier transform and Short Time Fourier Transform;

Then according to estimate obtain component of signal thus realize Signal separator.

2. a kind of external illuminators-based radar wind field clutter suppression method based on form component analysis according to claim 1, is characterized in that, in described step 3, and best sparse coefficient with solved by following sub-step:

Step 3.1, initializing variable: iterations initial value k=0, punishment parameter μ > 0, sparse coefficient initial value c to be solved _1,0, c _2,0, iterative process intermediate variable initial value d _1,0, d _2,0;

Step 3.2, more new variables:

v _i,k+1＝soft(c _i,k+d _i,k,λ _i/μ)-d _i,k，i＝1,2

g _k+1＝y-A ₁v _1,k+1-A ₂v _2,k+1

$d_{i,k+1}=\frac{1}{2}{A}_i^H{g}_{k+1},i=1,2$

c _i,k+1＝d _i,k+1+v _i,k+1，i＝1,2

Wherein, soft (.) is soft-threshold function, soft (x, T)=max (1-T/|x|, 0), the variable in x, T difference soft-threshold function, | x| is the absolute value of variable x; v _{i, k+1}, g _k+1, d _{i, k+1}, c _{i, k+1}represent the intermediate variable of i-th component of signal when carrying out kth+1 iteration respectively; A ₁, A ₂be respectively Fourier transform and Short Time Fourier Transform; H is transposed complex conjugate;

Step 3.3, judges whether to meet end condition, satisfied then stop; Otherwise, k=k+1.