CN102176018B - Doppler wave beam sharpening rapid imaging method of mechanical scanning radar - Google Patents

Abstract

The invention discloses a fast imaging method of Doppler beam sharpening for mechanical scanning radar, which mainly solves the problem that the prior art cannot be used for mechanical scanning radar. The implementation process is: first determine the number of pulse accumulation M and the sharpening ratio N, arrange the echoes after pulse compression into a distance-azimuth matrix in the order of reception; in this matrix, the length of the azimuth is 2*M in the order of reception , and overlap the two small matrices of M, and estimate their Doppler centers respectively; use these two Doppler centers to calculate the interpolation frequency points, and use the latter Doppler center to carry out the azimuth of the latter small matrix Compensate to the Doppler center; then perform azimuth FFT, use the interpolation frequency points to extract a sub-image from the data after FFT, and stitch the sub-images in the order obtained to obtain a large image. Compared with the existing DBS imaging method, the invention has less calculation amount and lower requirement on radar performance, and can be used for traditional mechanical scanning radar DBS imaging.

Description

Fast Imaging Method of Doppler Beam Sharpening for Mechanical Scanning Radar

technical field

The invention belongs to the technical field of signal processing, relates to radar imaging, and can be used for real-time, rapid and large viewing angle ground imaging.

Background technique

Due to the different azimuths of the fixed targets on the ground in the beam irradiation area of the airborne radar, the included angles between the line of sight and the velocity vector of the radar are also different, that is, they have different radial velocities relative to the aircraft and produce different Doppler frequency shifts. Doppler beam sharpening DBS is to split the fixed target echo beam in the same beam into several narrow sub-beams. Since different sub-beams have different azimuth angles and correspond to different Doppler frequencies, it can be obtained by dividing multiple Puller frequency separation separates different azimuth angles, effectively improving the radar azimuth resolution.

The following relationship exists in the Doppler beam sharpening imaging algorithm:

$\frac{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}{\mathrm{<span\; class="notranslate">\; m</span>}}$

Where Δf _d is the Doppler bandwidth at a certain scan angle, f _r is the pulse repetition frequency, M is the number of pulses accumulated, and N is the sharpening ratio.

Since the Doppler bandwidth Δf _d varies with the scan angle, in order to ensure consistent resolution, the sharpening ratio N should generally remain constant. So how to coordinate the pulse repetition frequency f _r , the coherent accumulation number M, and the antenna rotation speed is the key to Doppler beam sharpening image mosaic under scanning work. For phased array radar antennas, it is easier to control the scanning speed and dwell time of the antenna beam, which can ensure a constant sharpening ratio under different scanning angles. However, for mechanical scanning antennas, antenna scanning works continuously, and due to reasons such as mechanical inertia, it is difficult to achieve non-uniform rotation. Theoretically, both the pulse repetition frequency and the number of pulses to be accumulated remain constant during coherent integration. But to keep the sharpening ratio constant, at least one of the two changes step by step with the scanning wave level.

To ensure that the sharpening ratio is constant when the antenna is scanned at a constant speed, there are usually two ideas:

Idea 1: Fix the accumulated number of pulses M, so that the pulse repetition frequency f _r changes with the scanning angle;

Idea 2: The pulse repetition frequency f _r is fixed so that the accumulated number of pulses M changes with the scanning angle.

Compare the above two ideas to ensure a constant sharpening ratio: Idea 1, the number M of pulse accumulation is constant, that is, the number of filters in the coherent narrowband filter bank is fixed. At this time, f _r needs to be consistent with the change of scanning angle and the bandwidth Δf _d of the main beam, so that in any case, the filter bank is always filled at Δf _d ; the second idea is that f _r is constant, and the number of accumulated pulses M varies with The scan angle varies, thereby ensuring that the frequency resolution does not change. Idea 1 often adopts f _r stepped jumps in engineering practice, which has high requirements for antenna design; idea 2 changes in M, which brings inconvenience to signal processing: DFT is often used in signal processing to achieve coherent accumulation, and DFT is used here , the length of the input data, that is M, changes, and the length of the output data should be equal to the sharpening ratio N. For DFT with different points, it is necessary to extract the twiddle factor table. The programming is cumbersome and the implementation efficiency is low, which is not conducive to real-time implementation.

According to the above two ideas, there are currently two commonly used methods: beam-by-beam sharpening method and whole-range FFT beam sharpening method within _fr .

These two methods are relatively mature in theory, but for most mechanical scanning radars, it is still relatively difficult in engineering implementation. In order to solve this problem, someone proposed a pre-filtering method that keeps the sharpening ratio N constant under the premise that the pulse repetition frequency f _r is constant and the number M of coherent accumulation is also constant. This method is to filter and down-sample the echo before FFT, but it is very difficult in engineering to design a filter whose bandwidth varies with Δf _d and has good stop-band characteristics and linear phase characteristics.

Contents of the invention

Aiming at the shortcoming that the current DBS imaging method cannot be applied to mechanical scanning radar, the present invention proposes a fast imaging method of Doppler beam sharpening for mechanical scanning radar, so as to improve the speed of signal processing and realize the real-time imaging function in engineering.

To achieve the above object, the realization steps of the present invention include as follows:

(1) Using the known operating parameters of the mechanical scanning radar, calculate the constant radar pulse accumulation number M:

$\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&Delta;\&theta;</span>}}{{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}$

where v _θ is the rotational speed of the antenna, Δθ is the beam width, f _r is the radar pulse repetition frequency;

(2) Calculate the sharpening ratio N of Doppler beam sharpening according to the accumulated number M of radar pulses obtained and the known operating parameters of the mechanical scanning radar:

$\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Delta;f</span>}}_{\mathrm{<span\; class="notranslate">\; dl</span>}}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; m</span>}}{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}$

Among them, Δf _dl represents the Doppler bandwidth at the minimum scanning angle of the antenna;

(3) The continuously received radar echoes are pulse-compressed as the distance direction, arranged in a distance-azimuth matrix according to the receiving order, and the number of received radar echoes is recorded at the same time. When the number of received radar echoes reaches 3*M When using a sliding window whose azimuth length is 3*M and whose range length is equal to the number of sampling points in the range direction of the radar echo, read data in the range-azimuth matrix that has been arranged;

(4) The small distance-azimuth matrix A that arranges the 1st to 2*M radar echoes in the read data, and the small distance that arranges the M+1 to 3*M radar echoes -Azimuth matrix B, use the energy balance method to estimate the Doppler center of the two range-azimuth matrices A and B respectively, and obtain the Doppler center f d1 of the small range-azimuth matrix A and the doppler center f _d1 of the small range-azimuth matrix B Doppler center f _d2 ;

(5) According to the estimated two Doppler centers f _d1 and f _d2 , determine the interpolation frequency point set f _n :

n＝1，2，…N，N是偶数；5a)

n=1, 2, ... N, N is an even number;

n＝1，2，…N，N是奇数；5b)

n=1, 2, ... N, N is an odd number;

(6) Perform azimuth Doppler center compensation on the small range-azimuth matrix B according to the estimated Doppler center f _d2 , so that the center frequency of the azimuth data is moved to 0 frequency;

(7) Perform azimuth FFT on the data after Doppler center compensation to complete signal energy accumulation. At this time, the distance-azimuth matrix after Doppler center compensation becomes a distance-frequency matrix, and its frequency range is:- f _r /2 to f _r /2;

(8) Extract the data located at the frequency corresponding to the interpolation frequency point set f _n from the distance-frequency matrix, and obtain a subgraph with consistent azimuth angular resolution and consistent data rate;

(9) The obtained sub-pictures are directly spliced along the frequency direction according to the obtained order to obtain an image with a large viewing angle range;

(10) After waiting for the radar to receive M radar echo data again, slide the sliding window down M along the azimuth, read the data again, and perform step (11);

(11) Repeat step (4) to step (10).

The present invention has the following advantages:

Due to the repeated use of radar echo data, the present invention not only ensures that the targets in the beam irradiation area have sufficient and identical accumulation, but also improves the allowable antenna scanning rate, thereby reducing the possibility of image distortion; at the same time, due to the present invention Adopt fixed coherent accumulation number M, apply FFT to carry out energy accumulation to azimuth echo data, thereby greatly improved the speed of radar signal processing; In addition, because the present invention adopts constant radar pulse repetition frequency f _r , to reduce the antenna System design requirements, so that it is convenient to add Doppler beam sharpening function to the existing mechanical scanning radar.

Experimental results show that the present invention can be applied to mechanical scanning radar, and can obtain a relatively ideal ground image with a large viewing angle range.

Description of drawings

Fig. 1 is the fast imaging flowchart of Doppler beam sharpening of mechanical scanning radar of the present invention;

Fig. 2 is a real-time imaging result diagram of a certain city with Doppler beam sharpening of the present invention.

Detailed ways

With reference to Fig. 1, the concrete implementation process of the present invention is as follows:

Step 1. Calculate the constant accumulated number of radar pulses.

Installing the mechanical scanning radar on a plane flying at a constant speed can obtain the antenna speed v _θ , the radar beam width Δθ and the radar pulse repetition frequency f _r , and use these mechanical scanning radar operating parameters to calculate the constant accumulated number of radar pulses M:

$\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&Delta;\&theta;</span>}}{{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; .</span>$

Step 2. Calculate the sharpening ratio of Doppler beam sharpening.

Since the Doppler bandwidth Δf _d of the antenna increases with the increase of the antenna scan angle, the number of frequency points in the Doppler bandwidth Δf _d of the antenna also increases thereupon. In order to obtain subgraphs with consistent data rates, in the present invention The number of frequency points within the Doppler bandwidth Δf _dl at the minimum antenna scanning angle is used as the uniform sharpening ratio of Doppler beam sharpening. According to the radar pulse repetition frequency f _r , the number of accumulated radar pulses M and the minimum antenna scanning angle For the Doppler bandwidth Δf _dl , calculate the sharpening ratio N of Doppler beam sharpening:

$\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Delta;\; f</span>}}_{\mathrm{<span\; class="notranslate">\; dl</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; m</span>}}{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}<span\; class="notranslate">\; .</span>$

Step 3. Read data.

The continuously received radar echoes are pulse-compressed as the range direction, arranged in a range-azimuth matrix in the order of reception, and the number of received radar echoes is recorded at the same time. When the number of received radar echoes reaches 3*M, use A sliding window whose azimuth length is 3*M and whose range length is equal to the sampling points of the radar echo range is used to read data in the range-azimuth matrix that has been arranged.

Step 4. Doppler centroid estimation.

The small range-azimuth matrix A that arranges the 1st to 2*M radar echoes in the read data, and the small range-azimuth matrix that arranges the M+1 to 3*M radar echoes B. Use the energy balance method to estimate the Doppler center of the two range-azimuth matrices A and B respectively, and obtain the Doppler center f _d1 of the small range-azimuth matrix A and the Doppler of the small range-azimuth matrix B Center f _d2 .

Step 5. Determine the interpolation frequency point set f _n .

The difference between the estimated two Doppler centers f _d1 and f _d2 is the Doppler bandwidth of the echo data in the overlapping part of the small range-azimuth matrix A and B, and the sharpening ratio is uniformly extracted within this bandwidth range N frequency values form the interpolation frequency point set f _n , and the interpolation frequency point set f _n is calculated according to the odd and even numbers of the N frequency values according to the following two formulas:

n＝1，2，…N，N是偶数；5a)

n=1, 2, ... N, N is an even number;

5b) n=1, 2, ... N, N is an odd number.

Step 6. Doppler Centroid Compensation.

Since the estimated Doppler center f _d2 is the azimuth Doppler center of the small range-azimuth matrix B, when compensating for the Doppler center, each azimuth data of the small range-azimuth matrix B is multiplied by Doppler center compensation function: exp(-j2πf _d2 k/f _r ), k=0, 1, ... 2M-1, the center frequency of the compensated data can be moved to 0 frequency, and replaced by the compensated data The corresponding azimuth data in the small distance-azimuth matrix B, at this time, the center frequency of each azimuth data in the small distance-azimuth matrix B is located at frequency 0.

Step 7. Azimuth FFT.

Perform FFT on each azimuth data of the small range-azimuth matrix B after Doppler center compensation, and replace the corresponding azimuth data with the data after FFT. At this time, the range-azimuth matrix B after Doppler center compensation becomes A distance-frequency matrix is formed, and its frequency range is: _-fr /2 to _fr /2. This step also realizes signal energy accumulation.

Step 8. Extract subgraphs.

的整数倍时，直接抽取出距离-频率矩阵中此频率对应的数据，反之，在距离-频率矩阵中采用线性插值的方法抽取出此频率对应的数据；将抽取出的数据按频率由小到大的顺序排列成一幅子图，此子图方位向角分辨率一致，且数据率一致。Read the frequency value from the interpolation frequency point set f _n in turn, when the read frequency value is the frequency resolution of the matrix

When the integer multiple of , the data corresponding to this frequency in the distance-frequency matrix is directly extracted, otherwise, the data corresponding to this frequency is extracted by linear interpolation in the distance-frequency matrix; the extracted data is sorted by frequency from small to The large order is arranged into a sub-picture, and the azimuth angle resolution of the sub-picture is consistent, and the data rate is consistent.

Step 9. Subgraph splicing.

Taking the first sub-image as the reference image, the obtained sub-images are directly arranged along the frequency direction according to the order obtained, and an image with a large viewing angle range is obtained.

Step 10. After waiting for the radar to receive M radar echo data again, slide the sliding window down M along the azimuth, read the data again, and execute step (11).

Step 11. With the flight of the aircraft and the rotation of the antenna, the irradiation area of the radar antenna on the ground changes continuously, and steps (4) to (10) are repeated to obtain the submap of the antenna irradiation area at this time.

Effect of the present invention can be further illustrated by following experiments:

1. Experimental environment and content

Experimental environment: MATLAB 7.5.0, Intel(R) Pentium(R) 2CPU 3.0GHz, Window XP Professional.

Experimental content: the echo data recorded by the airborne mechanical scanning radar is applied to imaging in a simulation environment using the present invention.

2. Experimental results

Applying the present invention to quickly image the echo data collected by the airborne mechanical scanning radar, and obtain a ground image with a large viewing angle range, the result is shown in FIG. 2 .

It can be seen from Fig. 2 that the airport, city and surrounding landforms, etc., have better imaging quality, which shows that the present invention can be applied to mechanical scanning radar.

Claims (4)

1. A fast imaging method for Doppler beam sharpening of mechanical scanning radar, comprising the steps: (1) Using the known working parameters of the mechanical scanning radar, calculate the constant accumulated number of radar pulses M: $\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&Delta;\&theta;</span>}}{{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}$ where v _θ is the rotational speed of the antenna, Δθ is the beam width, f _r is the radar pulse repetition frequency; (2) Calculate the sharpening ratio N of Doppler beam sharpening according to the accumulated number M of radar pulses obtained and the known operating parameters of the mechanical scanning radar: $\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; dl</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; m</span>}}{{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}$ Among them, Δf _dl represents the Doppler bandwidth at the minimum scanning angle of the antenna; (3) The continuously received radar echoes are pulse-compressed as the distance direction, arranged in a distance-azimuth matrix according to the receiving order, and the number of received radar echoes is recorded at the same time. When the number of received radar echoes reaches 3*M When using a sliding window whose azimuth length is 3*M and whose range length is equal to the number of sampling points in the range direction of the radar echo, read data in the range-azimuth matrix that has been arranged; (4) The small distance-azimuth matrix A that arranges the 1st to 2*M radar echoes in the read data, and the small distance that arranges the M+1 to 3*M radar echoes -Azimuth matrix B, use the energy balance method to estimate the Doppler center of the two range-azimuth matrices A and B respectively, and obtain the Doppler center f d1 of the small range-azimuth matrix A and the doppler center f _d1 of the small range-azimuth matrix B Doppler center f _d2 ; (5) According to the estimated two Doppler centers f _d1 and f _d2 , determine the interpolation frequency point set f _n : 5a)

n=1,2,...N, N is an even number; 5b)

n=1,2,...N, N is an odd number; (6) Perform azimuth Doppler center compensation on the small range-azimuth matrix B according to the estimated Doppler center f _d2 , so that the center frequency of the azimuth data is shifted to 0 frequency; (7) Perform azimuth FFT on the data after Doppler center compensation to complete signal energy accumulation. At this time, the distance-azimuth matrix after Doppler center compensation becomes a distance-frequency matrix, and its frequency range is:- f _r /2 to f _r /2; (8) Extract the data at the frequency corresponding to the interpolation frequency point set f _n from the distance-frequency matrix, and obtain a submap with the same azimuth angle resolution and the same data rate; (9) The obtained sub-images are directly spliced along the frequency direction according to the obtained order to obtain an image with a large viewing angle range; (10) After waiting for the radar to receive M radar echo data again, slide the sliding window down M along the azimuth, read the data again, and execute step (11); (11) Repeat step (4) to step (10). 2. The method according to claim 1, wherein said step (6) carries out azimuth compensation to the Doppler center of the small range-azimuth matrix B according to the estimated Doppler center f _d2 , which is to make the small range-azimuth Each azimuth data of matrix B is multiplied by the Doppler center compensation function: exp(-j2πf _d2 k/f _r ), k=0,1,...2M-1 for compensation one by one, and the compensated data is used to replace the corresponding The azimuth data realizes the azimuth Doppler center compensation for the small range-azimuth matrix B. 3. The method according to claim 1, wherein the step (8) extracts the data at the frequency corresponding to the interpolation frequency point set f _n from the distance-frequency matrix, which is sequentially from the interpolation frequency point set f _n Read the frequency value, when the read frequency value is the frequency resolution of the matrix

When the integer multiple of , the data corresponding to this frequency in the distance-frequency matrix is directly extracted, otherwise, the data corresponding to this frequency is extracted by linear interpolation in the distance-frequency matrix; the extracted data is sorted by frequency from small to The large order is arranged into a subgraph. 4. The method according to claim 1, wherein in the step (9), the obtained sub-pictures are directly spliced along the frequency direction according to the obtained order, which is to use the first sub-picture as the reference image for the obtained sub-pictures, Arrange directly along the frequency direction according to the obtained order to obtain an image with a large viewing angle range.

Images