CN104808182A - High-resolution wide-swath onboard SAR (synthetic aperture radar) real-time motion compensation algorithm - Google Patents

Abstract

The invention discloses a high-resolution wide-swath onboard SAR (synthetic aperture radar) real-time motion compensation algorithm which mainly includes an envelop compensation portion and a phase compensation portion. A traditional phase compensation algorithm is improved, and phase compensation can be finished before envelop compensation. Envelop compensation approximate treatment needs to divide distance sub-bands for echo signals after distance compression, so that motion error space varying is reduced. Therefore, interpolation operation of a traditional envelop compensation algorithm is avoided, and envelop compensation is finished by directly multiplying distance frequency domain with linear phase. Finally, distance sub-band images are spliced to obtain an imaging result of a whole swath. By the aid of an improved phase compensation algorithm, signal phase precision is ensured, so that the interpolation operation with large calculating amount can be more effectively avoided by rapid envelop compensation based on the distance sub-bands, and the real-time treatment performance of the algorithm is improved without affecting resolution.

Description

A kind of high resolution wide swath carried SAR real-time motion compensation algorithm

Technical field

The present invention relates to SAR technical field of imaging, specifically a kind of high resolution wide swath carried SAR real-time motion compensation algorithm.

Background technology

Major issue during airborne synthetic aperture radar data processing is the compensation of kinematic error.Owing to being subject to the interference of the factors such as air-flow in carrier aircraft flight course, actual flight path is no longer a desirable straight line, introduces kinematic error in echo data.Kinematic error produces resolution, geometric linear degree, contrast etc. to be affected largely, causes picture quality degradation.Therefore, motion compensation is the committed step of carried SAR Data processing.

Movement compensating algorithm carries out in two-dimensional time-domain usually, is divided into envelope cancellation and phase compensation two parts.Conventional motion compensation algorithm needs first to carry out envelope cancellation, and echo signal is corrected to correct range unit, eliminates envelope delay error; Next carry out phase compensation, envelope cancellation is as the basis of phase compensation, and after echo signal is only corrected to correct range unit, phase error just matches with traditional error calculation formula, thus accurately can complete phase compensation.During carried SAR data processing, the envelope delay error of each target in imaging scene is also inconsistent, and desirable envelope cancellation algorithm corrects by range unit, needs to be realized by interpolation, this results in larger operand.Further, wide swath has become an important trend of SAR development, and this makes operand increase further.Conventional motion compensation algorithm is so limited, and is difficult to the requirement meeting process in real time, is difficult to apply in engineering practice well.

When distance is little to mapping band scope, the space-variant envelope error of each range unit can be approximately equal to the constant envelope error of sky at mapping band center.Therefore, in order to avoid the macrooperation amount that interpolation is brought, researchist once proposed a kind of fast envelope backoff algorithm without the need to interpolation according to above-mentioned character: the echoed signal after compression of adjusting the distance divides distance subband, reduce the space-variant of kinematic error, replace the envelope error of each range unit in subband by the envelope error of subband center, be directly multiplied by linear phase at distance frequency domain and complete envelope cancellation.This movement compensating algorithm based on distance subband brings another key issue: the echo signal of non-subband center is not corrected to correct range unit, creates remaining envelope delay error; If still traditionally phase error computation formula, can cause phase compensation reference function not mate with real phase error, reduce the precision of phase compensation, make final picture quality be affected.

Summary of the invention

The problem to be solved in the present invention is to provide a kind of high resolution wide swath carried SAR real-time motion compensation algorithm, and this algorithm improves the phase compensation section in traditional algorithm, have adjusted the processing sequence of envelope cancellation and phase compensation.This algorithm improves real-time handling property, ensure that the precision of phase compensation meets resolution requirement, and avoids a large amount of interpolation arithmetic amounts, and the picture quality of acquisition is high.

Technical scheme: a kind of high resolution wide swath carried SAR real-time motion compensation algorithm disclosed by the invention, basic thought improves the phase compensation section in traditional algorithm, have adjusted the processing sequence of envelope cancellation and phase compensation.The envelope cancellation part of this algorithm still adopts the disposal route of distance subband, avoids the interpolation operation that operand is larger; But before envelope cancellation approximate processing, first carry out the phase compensation operation improved, to make phase compensation not by the impact of remaining envelope delay error, ensure that the phase accuracy of signal.Algorithm employs the account form different from traditional wave beam direction of visual lines kinematic error computing formula, improves the treatment scheme of traditional algorithm, breaches the restricting relation of envelope cancellation to phase compensation; Under guarantee phase compensation precision meets the condition of resolution requirement, thus the fast envelope backoff algorithm without the need to interpolation can be combined better, improve the real-time handling property of algorithm.

Specifically comprise the following steps:

1) Range compress is carried out to original echoed signals; Distance carries out matched filtering to transforming to distance frequency domain, then switches back to time domain and complete Range compress.

2) echoed signal after compressing of adjusting the distance divides distance subband.

3) phase compensation improved is carried out.In i-th distance subband, wave beam direction of visual lines kinematic error computing formula is:

${\mathrm{\Δr}}_i=r_b\left(l\right)-\sqrt{{(\sqrt{r_b^2\left(l\right)-{(H+\mathrm{\Δz})}^2}+\mathrm{\Δy})}^2+H^2}---\left(1\right)$

i·(N _{r_sub}-N _{r_overlap})≤l≤i·(N _{r_sub}-N _{r_overlap})+N _{r_sub}-1 (2)

In formula, H is the carrier aircraft height of Desired Track, and Δ y is the site error in path in elevation direction, and Δ z is the site error of short transverse, N _{r_sub}for the range unit number of subband, N _{r_overlap}for range unit number multiplexing between adjacent sub-bands, l represents the index of range unit, Δ r _irepresent wave beam direction of visual lines kinematic error in i-th distance subband, r _brepresent the oblique distance that in mapping band, each range unit is corresponding

$r_b=r_0+n\·\frac{c}{2f_s},n=\mathrm{0,1,2},...,N_r-1---\left(3\right)$

In formula, r ₀represent the oblique distance that first range unit is corresponding, f _sfor distance is to the sampling rate of signal, c is the light velocity, N _rfor total range unit number.

4) fast envelope carried out without the need to interpolation compensates.

5) Omega-K algorithm process, transforms to the image that two-dimensional time-domain obtains distance subband.

6) to step 5) the distance sub-band images that obtains carries out Autofocus processing; Autofocus processing obtains focusing on good distance sub-band images further.

7) step 3 is repeated) ~ step 6), until all distance subband datas are disposed.

8) splice each sub-band images, obtain the imaging results of whole mapping band.

As the further improvement of technique scheme, described step 1) Range compress reference function is:

$H_{\mathrm{rc}}=\exp \left\{j\mathrm{\π}\frac{f_r^2}{k_r}\right\}---\left(4\right)$

In formula, f _rfor distance is to frequency, k _rfor the chirp rate that transmits.

As the further improvement of technique scheme, described step 2) distance subband number be:

N _subswath＝(N _r-N _{r_sub})/(N _{r_sub}-N _{r_overlap})+1 (5)

As the further improvement of technique scheme, described step 4) in when carrying out the phase compensation improved, the phase compensation reference function that i-th distance subband is corresponding is:

$H_{\mathrm{pc},i}=\exp \left\{j\frac{4\mathrm{\π}}{\mathrm{\λ}}{\mathrm{\Δr}}_i\right\}---\left(6\right)$

As the further improvement of technique scheme, described step 4) in when carrying out compensating without the need to the fast envelope of interpolation, the envelope cancellation reference function that i-th distance subband is corresponding is:

$H_{\mathrm{ec},i}=\exp \left\{j\frac{4\mathrm{\π}{f}_r}{c}{\mathrm{\Δr}}_{c,i}\right\}---\left(7\right)$

In formula, Δ r _c,irepresent the kinematic error that i-th distance subband center is corresponding.

As the further improvement of technique scheme, described step 5) in when carrying out Omega-K algorithm process, the consistent reference compression function that i-th distance subband is corresponding is:

$H_{\mathrm{rfm},i}=\exp \{j\frac{4\mathrm{\π}{r}_{c,i}}{c}\sqrt{{(f_c+f_r)}^2-\frac{c^2{f}_a^2}{4V_a^2}}-j\frac{4\mathrm{\π}{r}_{c,i}}{c}{f}_r\}---\left(8\right)$

In formula, f _cfor carrier frequency, f _afor orientation is to frequency, V _afor carrier aircraft speed, r _c,irepresent the shortest oblique distance of i-th distance subband center.

Below by with the comparing step 3 of traditional calculating formula) be described in detail.

Wave beam direction of visual lines kinematic error Δ r has distance space-variant, and movement compensating algorithm needs to calculate Δ r corresponding to each range unit according to inertial guidance data.As shown in Figure 2, make the shortest oblique distance (r+ Δ r) of target T and actual flight path corresponding to p range unit r+ Δ r=r in the Δ r computing formula of improvement _b(p), then

$y_T=\sqrt{r_b^2\left(p\right)-{(H+\mathrm{\Δz})}^2}+\mathrm{\Δy}---\left(9\right)$

So, the shortest oblique distance of target and Desired Track is

$r=\sqrt{y_T^2+H^2}---\left(10\right)$

The Δ r computing formula then improved is

$\mathrm{\Δr}=r_b\left(p\right)=\sqrt{{(\sqrt{r_b^2\left(p\right)-{(H+\mathrm{\Δz})}^2}+\mathrm{\Δy})}^2+H^2}---\left(11\right)$

In conventional motion compensation algorithm, after having carried out accurate envelope cancellation, echo signal has been corrected to correct range unit.Therefore, the shortest oblique distance r calculating Δ r target T in season and Desired Track in follow-up phase compensation corresponds to m range unit r=r _bm (), then its projected length on ground is

$y_T=\sqrt{r_b^2\left(m\right)-H^2}---\left(12\right)$

So, the shortest oblique distance (r+ Δ r) of target and actual flight path can be expressed as

$r+\mathrm{\Δr}=\sqrt{{(y_T-\mathrm{\Δy})}^2+{(H+\mathrm{\Δz})}^2}---\left(13\right)$

Then in traditional algorithm, the computing formula of Δ r is

$\mathrm{\Δr}=\sqrt{{(\sqrt{r_b^2\left(m\right)-H^2}-\mathrm{\Δy})}^2+{(H+\mathrm{\Δz})}^2}-r_b\left(m\right)---\left(14\right)$

Formula (11) compares can find with formula (14), formula (14) is the shortest oblique distance obtained according to the shortest oblique distance of target and Desired Track under actual flight path, and formula (11) is then the anti-the shortest oblique distance pushed away under Desired Track of the shortest oblique distance according to target and actual flight path.The shortest oblique distance under target Desired Track and the shortest oblique distance under actual flight path all should be corresponding to a certain unit in scale with formula (3) distance.Formula (14) is the shortest oblique distance under each range unit that assumption (3) represents is mapped as Desired Track, the phase error that each range unit calculated is corresponding is the signal after compensating for exact envelope, and at this moment envelope cancellation generation error will make phase compensation also not accurate enough.Formula (11) is the shortest oblique distance under each range unit that assumption (3) represents is mapped as actual flight path, phase error corresponding to each range unit calculated be for envelope cancellation before signal, the phase compensation function of each range unit and the real phase error of signal are accurate match, ensure that the precision of phase compensation is not by the impact of envelope cancellation error.

The high resolution wide swath carried SAR real-time motion compensation algorithm that the present invention proposes is compared with existing movement compensating algorithm, and core is to improve the phase compensation section in traditional algorithm, have adjusted the processing sequence of envelope cancellation and phase compensation.First the algorithm that the present invention proposes ensures the phase accuracy of signal, makes the effect of phase compensation no longer be subject to the restriction of envelope cancellation result; Under guarantee phase compensation precision meets the condition of resolution requirement, thus the fast envelope backoff algorithm without the need to interpolation can be combined better, improve the real-time handling property of algorithm.Owing to greatly reducing operand, reduce research cost, improve Efficiency, become the background of an important trend of SAR development at wide swath under, the research and development for SAR imaging technique provides an important research direction.

Accompanying drawing explanation

Fig. 1 is the processing flow chart of the inventive method;

Fig. 2 is flight path normal plane site error geometric relationship;

Fig. 3 is the point target contour map adopting the airborne SAR emulated data of conventional motion compensation algorithm process to obtain, and envelope cancellation is completed by desirable interpolation operation;

Fig. 4 is the point target contour map adopting the movement compensating algorithm process carried SAR emulated data based on distance subband to obtain, and envelope cancellation is completed by the fast algorithm without the need to interpolation, and conventional phase compensates rear;

Fig. 5 is the point target contour map adopting movement compensating algorithm process carried SAR emulated data of the present invention to obtain, and envelope cancellation is completed by the fast algorithm without the need to interpolation, improves phase compensation front;

The imaging results figure that Fig. 6 to 9 adopts the inventive method process to obtain for other range unit point targets in scene;

Figure 10 is the imaging results figure that carried SAR measured data adopts the inventive method process and obtains.

Embodiment

Below in conjunction with accompanying drawing, a kind of high resolution wide swath carried SAR real-time motion compensation algorithm that the present invention proposes is described in detail.

Invention describes a kind of high resolving power movement compensating algorithm of improvement, its basic thought improves the phase compensation section in traditional algorithm, the processing sequence of adjustment envelope cancellation and phase compensation, improves the treatment effeciency of algorithm under the prerequisite ensureing compensation effect.Adopt the Phase Compensation Algorithm improved, reduce the impact of phase error correction effect by other factors; Envelope cancellation uses the approximate fast algorithm without the need to interpolation, improves operation efficiency.This section makes checking and analysis by utilizing SAR simulative data and flight-test data to the algorithm proposed.Simulative data and flight-test data radar parameter is identical, is the running parameter of certain X-band high resolving power air-borne test SAR.Radar signal bandwidth 1.3GHz, azimuth resolution 0.1m, imaging scene center oblique distance 8271.7m, mapping band oblique distance scope is about 8km.Course error is generated by the inertial guidance data of reality, and error span mean value reaches 27m.

High resolution wide swath carried SAR real-time motion compensation algorithm of the present invention, as shown in Figure 1, specific implementation step is its basic implementation process:

First carry out Range compress, the echo data after compression of adjusting the distance divides distance subband.

Distance carries out matched filtering to transforming to distance frequency domain, then switches back to time domain and complete Range compress.Range compress reference function is

$H_{\mathrm{rc}}=\exp \left\{j\mathrm{\π}\frac{f_r^2}{k_r}\right\}---\left(1\right)$

Because method of the present invention operates based on distance subband, therefore first carry out Range compress, corresponding imaging processing model choice Omega-K algorithm.

Distance subband number is

N _subswath＝(N _r-N _{r_sub})/(N _{r_sub}-N _{r_overlap})+1 (2)

Should be noted during sub-band division, owing to being divide subband before RCM corrects, subband width should be not less than the twice of echo ultimate range migration amount, and between adjacent sub-bands, Multiplexing Unit number should be not less than ultimate range migration amount.

Calculate echo ultimate range migration amount according to radar parameter and be about 1813 range units, therefore distance subband width chooses 4096 range units, multiplexing 2048 range units between adjacent sub-bands.

Next carry out motion compensation, the phase compensation and the fast envelope that comprise improvement compensate.Site error Δ y and Δ z can be calculated by the inertial guidance data of carrier aircraft.First carry out phase compensation, calculate wave beam direction of visual lines kinematic error according to formula (1), complete the phase compensation of improvement in two-dimensional time-domain according to formula (6).Distance, to transforming to frequency domain, is multiplied by the linear phase in formula (7), and complete the approximate envelope cancellation without the need to interpolation, so far kinematic error is corrected.It is a kind of approximate processing that fast envelope without the need to interpolation compensates, and replaces the distance space-variant error of each range unit in subband with the empty constant envelope error of the distance that subband center is corresponding.Signal is multiplied by linear phase to complete envelope cancellation in distance frequency domain, orientation time domain.

Subsequent treatment is according to the flow process of Omega-K algorithm, and signal transforms to two-dimensional frequency, unanimously compresses, and Stolt interpolation completes remaining RCM correction, remaining SRC and remaining Azimuth Compression, finally transforms to the image that two-dimensional time-domain obtains distance subband.

In Omega-K algorithm process, i-th consistent reference compression function corresponding to distance subband is:

$H_{\mathrm{rfm},i}=\exp \{j\frac{4\mathrm{\π}{r}_{c,i}}{c}\sqrt{{(f_c+f_r)}^2-\frac{c^2{f}_a^2}{4V_a^2}}-j\frac{4\mathrm{\π}{r}_{c,i}}{c}{f}_r\}---\left(8\right)$

During measured data process, the precision of usual inertial guidance data cannot meet resolution requirement, and therefore, the motion compensation only relied on based on inertial guidance data generally cannot obtain desirable image.Method of the present invention obtains focusing on good distance sub-band images by Autofocus processing further.All distance subband datas are disposed, and splice each sub-band images, obtain the imaging results of whole mapping band.

In order to the validity of the inventive method is described, uses three kinds of movement compensating algorithms to process respectively to same emulated data, obtain the design sketch as shown in Fig. 3 to 5, wherein horizontal ordinate be orientation to sampled point, ordinate is that distance is to sampled point.Fig. 3 adopts traditional movement compensating algorithm, and envelope cancellation is accurately completed by desirable interpolation operation; Fig. 4 adopts the existing movement compensating algorithm based on distance subband, first completes the approximate envelope cancellation without the need to interpolation, then carries out traditional phase compensation; Fig. 5, for adopting movement compensating algorithm of the present invention, first completes the phase compensation of improvement, then carries out the approximate envelope cancellation without the need to interpolation.Contrast can find, adopts the inventive method can obtain the imaging results suitable with traditional ideal movements backoff algorithm.Fig. 6 to 9 is the imaging results of other range unit point targets in scene, demonstrates correctness and the validity of the inventive method further.

Figure 10 is the imaging results that carried SAR measured data adopts method process of the present invention and obtains, horizontal direction be orientation to, vertical direction be distance to, image from top to bottom corresponding low coverage to long distance.As can be seen from result, method of the present invention can obtain and focus on good SAR image, and the distance of image is to also without significantly splicing vestige, and algorithm performance meets imaging requirements.

In sum, above embodiment only in order to technical scheme of the present invention to be described, is not intended to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., it all should be encompassed in the middle of right of the present invention.

Claims (6)

1. a high resolution wide swath carried SAR real-time motion compensation algorithm, is characterized in that, comprise the following steps:

1) Range compress is carried out to original echoed signals;

2) echoed signal after compressing of adjusting the distance divides distance subband;

3) carry out the phase compensation improved, in i-th distance subband, wave beam direction of visual lines kinematic error computing formula is:

i·(N _{r_sub}-N _{r_overlap})≤l≤i·(N _{r_sub}-N _{r_overlap})+N _{r_sub}-1 (2)

In formula, H is the carrier aircraft height of Desired Track, and Δ y is the site error in path in elevation direction, and Δ z is the site error of short transverse, N _{r_sub}for the range unit number of subband, N _{r_overlap}for range unit number multiplexing between adjacent sub-bands, l represents the index of range unit, Δ r _irepresent wave beam direction of visual lines kinematic error in i-th distance subband, r _brepresent the oblique distance that in mapping band, each range unit is corresponding

In formula, r ₀represent the oblique distance that first range unit is corresponding, f _sfor distance is to the sampling rate of signal, c is the light velocity, N _rfor total range unit number;

4) fast envelope carried out without the need to interpolation compensates;

5) Omega-K algorithm process, transforms to the image that two-dimensional time-domain obtains distance subband;

6) to step 5) the distance sub-band images that obtains carries out Autofocus processing;

7) step 3 is repeated) ~ step 6), until all distance subband datas are disposed;

8) splice each sub-band images, obtain the imaging results of whole mapping band.

2. high resolution wide swath carried SAR real-time motion compensation algorithm according to claim 1, is characterized in that, described step 1) Range compress reference function is:

In formula, f _rfor distance is to frequency, k _rfor the chirp rate that transmits.

3. high resolution wide swath carried SAR real-time motion compensation algorithm according to claim 1, is characterized in that, described step 2) distance subband number be:

N _subswath＝(N _r-N _{r_sub})/(N _{r_sub}-N _{r_overlap})+1 (5)。

4. high resolution wide swath carried SAR real-time motion compensation algorithm according to claim 1, is characterized in that, described step 3) in when carrying out the phase compensation improved, the phase compensation reference function that i-th distance subband is corresponding is:

。

5. high resolution wide swath carried SAR real-time motion compensation algorithm according to claim 1, is characterized in that, described step 4) in when carrying out compensating without the need to the fast envelope of interpolation, i-th envelope cancellation reference function corresponding to distance subband is:

In formula, Δ r _c,irepresent the kinematic error that i-th distance subband center is corresponding.

6. high resolution wide swath carried SAR real-time motion compensation algorithm according to claim 1, is characterized in that, described step 5) in when carrying out Omega-K algorithm process, the consistent reference compression function that i-th distance subband is corresponding is:

In formula, f _cfor carrier frequency, f _afor orientation is to frequency, V _afor carrier aircraft speed, r _c,irepresent the shortest oblique distance of i-th distance subband center.