CN102680974A

CN102680974A - Signal processing method of satellite-bone sliding spotlight synthetic aperture radar

Info

Publication number: CN102680974A
Application number: CN2012101688202A
Authority: CN
Inventors: 李财品; 谭小敏
Original assignee: Xian Institute of Space Radio Technology
Current assignee: Xian Institute of Space Radio Technology
Priority date: 2012-05-25
Filing date: 2012-05-25
Publication date: 2012-09-19
Anticipated expiration: 2032-05-25
Also published as: CN102680974B

Abstract

The invention discloses a signal processing method of a satellite-bone sliding spotlight synthetic aperture radar. The method includes: performing molecule aperture processing on original echo data of the sliding spotlight synthetic aperture radar (SAR); performing range compression and range migration correction through a chirp scaling (CS) algorithm; introducing squint angle compensation in the CS algorithm; compensating scene space-variant performance; adjusting range migration amount of all sub-apertures to be identical with Doppler parameters of full apertures in CS scalar variable factors as the standard; performing azimuth compression and quadratic term compensation in the frequency domain, and performing frequency modulation processing in the time domain; splicing data of all the sub-apertures and restoring resolution ratio of the full apertures; and enabling signals to have phase preservation performance through spectrum analyzer (SPECAN) processing. By adopting the signal processing method, imaging quality can be improved under the condition of large scene squinting.

Description

Signal processing method of satellite-borne sliding bunching synthetic aperture radar

Technical Field

The invention relates to a method for processing synthetic aperture radar signals, in particular to a signal processing method under a sliding bunching mode large-scene squint condition.

Background

Synthetic Aperture Radar (SAR) is a typical radar system for ground reconnaissance imaging, is mainly used for military reconnaissance and disaster monitoring, and is one of important microwave remote sensing instruments which can work all day long and all weather at present. At present, the common operation modes of the synthetic aperture radar mainly include a stripe mode, a scanning mode, a beam-bunching mode, a sliding beam-bunching mode, a TOPS mode, and so on. In the several SAR imaging modes, the strip mode can realize wide azimuth swath, but high resolution is difficult to realize; although a swath wide in the upward direction and the azimuthal direction can be obtained simultaneously in the scanning mode, the resolution is lower than that in the same case at the expense of the resolution; although the beam-bunching mode can obtain high resolution, the mapping zone in the azimuth direction is very small, and the size of the footprint of the beam is only one; the sliding bunching mode can break through the limitation of the bunching mode, and not only can high resolution be realized, but also a wide azimuth mapping band can be realized. The processing algorithm of the sliding spotlight mode mainly comprises a subaperture method and an azimuth preprocessing-based method. A signal processing block diagram in which the sub-aperture processes the sliding beamforming mode is shown in fig. 1. A block diagram of the sliding bunching process based on the azimuth preprocessing is shown in fig. 2. The existing signal processing method of the sliding bunching synthetic aperture radar can be summarized as follows: acquiring echo signals, reducing PRF by adopting a sub-aperture or azimuth preprocessing method, avoiding azimuth blurring of imaging, and then imaging by adopting conventional algorithms, such as CS algorithm, RD algorithm and RMA algorithm.

During the actual flight of a satellite or an airplane, due to control or other principles, the doppler frequency shift of a target pointed by the phase center of the antenna may not be zero, and the phase center of the antenna has a certain squint angle. The existence of the oblique angle causes the loss of signal-to-noise ratio during radar imaging, the reduction of azimuth ambiguity and image deviation. In addition, the range of beam irradiation of the satellite-borne synthetic aperture radar is large, and especially in high-resolution imaging, the space variability of a scene cannot be ignored. The existing sliding spotlight SAR imaging algorithm does not consider the situation under an oblique angle, does not consider the scene space-variant compensation under a large scene, does not carry out the consistency compensation on the phase of each sub-aperture (the image quality is influenced when a plurality of sub-apertures are spliced finally), and influences the imaging quality in the three aspects. Therefore, it is necessary to develop a sliding spotlight SAR imaging algorithm capable of improving image quality under large scene squint conditions.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: aiming at the defects of the prior art, the signal processing method of the satellite-borne sliding bunching synthetic aperture radar is provided, and the imaging quality can be improved under the condition of large scene squint.

The invention comprises the following technical scheme:

a signal processing method of a satellite-borne sliding spotlight synthetic aperture radar is characterized by comprising the following steps:

(1) sub-aperture division is carried out on the collected original data of the sliding bunching synthetic aperture radar;

(2) and respectively processing each divided sub-aperture data as follows:

performing direction FFT; multiplying the data subjected to the direction Fourier transform by a scaling factor; performing range-to-FFT conversion, and then performing range compression and range migration correction; performing distance inverse FFT; performing azimuth compression and compensating the residual phase after inverse FFT conversion; performing quadratic term compensation after azimuth compression; then performing the direction inverse FFT; after the azimuth direction inverse FFT conversion, frequency modulation removing processing is carried out;

(3) synthesizing the processed sub-aperture data to synthesize a full aperture data;

(4) performing azimuth FFT on the full-aperture data;

(5) carrying out residual frequency modulation compensation;

(6) performing azimuth inverse FFT;

(7) the Specan compensation is performed so that the signal has phase-holding properties.

The formula of the scaling factor in the step (2) is as follows:

wherein,

Δα(f_a)＝α(f_a)-α(f_dc)，

wherein f is_dcDoppler center frequency for full aperture, t satellite time of flight, f_aAzimuth frequency, c is speed of light; r₀Is the center slope of the scene, k (f)_a(ii) a R) is equivalent tuning frequency; r (f)_a，R₀) The instantaneous distance from the scene center to the satellite; v is the speed of flight of the satellite, theta_refIs the oblique angle of view at the center of the scene, theta_cAt the oblique viewing angle at the center of the full aperture, λ is the wavelength.

k_rFrequency adjustment for linear signals, where R ═ R₀+c/2/f_s[-nrn/2：nrn/2-1]，R₀Is the center slope of the scene, where f_sFor signal sampling rate, nrn is the number of distance-wise sampling points.

In the step (2), an azimuth compression function for azimuth compression and compensation of the residual phase is as follows:

wherein

R(f_a(ii) a r) is the slant distance of the satellite to the ground target point.

Compared with the prior art, the invention has the following advantages:

the method provided by the invention is used for imaging in a sliding bunching working mode under large-scene squint. The method not only considers the compensation of the squint angle and the phase consistency compensation among the sub apertures, but also considers the space-variant property of the range migration under the condition of a large scene.

Drawings

FIG. 1 is a block diagram of a conventional sub-aperture sliding beamforming process;

FIG. 2 is a block diagram of a prior art sliding bunching process based on azimuth preprocessing;

FIG. 3 is a flow chart of a processing method of the present invention;

FIG. 4 is a sub-aperture 1 imaging;

FIG. 5 is a sub-aperture 1-point target spectrum;

FIG. 6 is sub-aperture 2 imaging;

FIG. 7 is a sub-aperture 2-point target spectrum;

FIG. 8 is a sliding spotlight SAR full aperture imaging;

fig. 9 is a full aperture imaging point target spectrum.

Detailed Description

The invention will now be further described with reference to the accompanying drawings. As shown in fig. 3, the processing method of the present invention includes the steps of:

(1) subaperture partitioning

The method comprises the steps of firstly, carrying out subaperture division on original sliding spotlight SAR data, wherein the length of the subaperture is larger than the instantaneous bandwidth of a target point of the sliding spotlight SAR. Can be according to the formula

Determining the number of sub-aperture divisions PRF being the pulse repetition frequency, B_aFor instantaneous bandwidth of the signal, k_rotThe chirp rate of the satellite to the center of rotation. By dividing the sub-apertures, the azimuth ambiguity of the sliding spotlight synthetic aperture radar is effectively removed, the repetition frequency of the system is reduced, the selection of the system on the pulse repetition frequency is consistent with that of the system in a stripe mode, and the data volume of the synthetic aperture radar data transmission is greatly reduced. On the other hand, the range migration amount is reduced through the division of the sub-apertures, the coupling amount of the range direction and the azimuth direction is reduced, and the decoupling processing in the imaging algorithm is facilitated.

(2) Respectively carrying out azimuth FFT (fast Fourier transform) on the divided sub-aperture data

After the azimuth Fourier transform, the method is expressed as follows:

in the formula,

is an azimuth window function, X is the azimuth starting position of the ground target point, the length of the footprint of the L wave beam, v_aFor the beam footprint at ground speed, t is the satellite time of flight, rect ((t-2R (f)_a(ii) a R)/c)/T) is a range window function, T is a range echo delay period, R (f)_a(ii) a r) is the slant distance from the satellite to the ground target point, r is the closest distance from the platform to the ground, theta is the slant angle, lambda is the wavelength, V is the satellite velocity, k (f)_a(ii) a R) is the equivalent tuning frequency, f_aThe azimuth frequency, c is the speed of light.

(3) Multiplying the data after the direction Fourier transform by a scaling factor

The CS factor at this time is:

Δα(f_a)＝α(f_a)-α(f_dc)

R(f_a，R₀) Instantaneous distance from the center of the scene to the satellite, f_dcDoppler center frequency, theta, for a synthetic aperture divided by a front full aperture_cAt an oblique viewing angle at the center of the full aperture, theta_refIs the squint angle at the center of the scene. Here, it should be noted that: for consistency of final sub-aperture splicing, the CS factor for changing the frequency modulation scale needs to be normalized, that is, the CS factor value of each sub-aperture is made consistent through normalization processing.

The null-variant of a large scene is considered in the equivalent frequency modulation rate,

(4) Then, distance-to-FFT conversion is carried out

(5) Then, distance compression and distance migration correction are carried out

After the operation of the previous step, distance compression and distance migration correction are needed, and the phase function is

f_rIs the range frequency.

(6) The distance is transformed to the inverse FFT,

(7) and performing azimuth compression after transformation and compensating the residual phase multiplied by the following azimuth compression function:

wherein,

(8) quadratic term compensation after azimuth compression

Multiplication by a quadratic compensation term function H4:

wherein

Wherein

r_{scl} (r) = \frac{r_{scl 0}}{r_{rot 0}} r_{rot} (r),

r_{rot} (r) = \frac{r_{rot 0} - r}{1 - r_{scl 0} / r_{rot 0}},

r_scl(r) is a scene slope function, r_rot(r) is a function of the rotational slope distance, r_rot0Distance of the platform from the point of rotation, r_scl0The selection of (1) is related to the image azimuth interval, and generally takes the value as the distance from the platform to the ground target point.

(9) Azimuth inverse FFT transform

(10) After conversion, the frequency is removed

After the direction inverse FFT, in order to further reduce the processing bandwidth of the direction, a frequency modulation removing process H is adopted₅＝exp(-jπk_rot(r)(t_a-t_mid) Therein), wherein

t_aIs azimuth time, t_midThe scene center time.

(11) And (4) processing the data of each sub-aperture according to the steps (2) to (10), and performing sub-aperture synthesis according to the time sequence to synthesize data of a full aperture.

(12) Performing azimuth FFT processing on the processed full aperture data

(13) Then, residual frequency modulation compensation is carried out

Multiplying the synthesized full aperture data by H₆：

Wherein k is_eff(r)＝k_scl(r)-k_rot(r)

(14) Performing an azimuth inverse FFT transform

(15) Finally, performing spectrum analysis (Specan) compensation

In order to make the final imaging result phase-preserving, the following phase function needs to be multiplied:

wherein,

the steps consider the processing under the oblique angle, and the compensation of the oblique angle is introduced into the CS algorithm. Such as a factor

Azimuthal compression function

Wherein

Etc. introduce compensation for squint angles.

Performing point target simulation on the large-scene squint SAR sliding bunching algorithm, wherein the selected simulation parameters are as follows:

center frequency: 9.6GHz

Pulse repetition frequency: (ii) a 3800Hz

Signal bandwidth: 600MHz

Azimuth resolution: 1 m

The slant distance of the rotating center: 1234 km

Scene center slope distance: 617 km

The oblique angle of the corresponding antenna at the center of the scene is as follows: 3 degree

Antenna aperture: 4 m

Scene size: 15 kilometers (distance direction) × 20 kilometers (azimuth direction)

Point target placement in scene center

As can be seen from FIGS. 5 and 7, the sub-aperture data after being processed by the method achieves good focusing, the side ratio of the peak value is about-13 dB, and the integral side lobe ratio is about-9.5 dB. Since the final step of the imaging algorithm uses de-modulation, the azimuthal pixel spacing is now

Wherein, N is the number of azimuth processing points, and according to the method, the azimuth pixel interval is 0.3662 after the two sub-apertures are divided, and the pixel interval is 0.1881 after the full aperture is synthesized. In fig. 4 and 6, the separation at 3dB widths is about 6 and 5 or so, the sub-aperture imaging resolution is about 2 meters, while fig. 8 is improved to 1 meter after full aperture processing. Therefore, the full-aperture processing under the large-scene squint not only improves the resolution ratio, but also obtains good focusing effect.

The invention is not described in detail and is within the knowledge of a person skilled in the art.

Claims

1. A signal processing method of a satellite-borne sliding spotlight synthetic aperture radar is characterized by comprising the following steps:

(2) and respectively processing each divided sub-aperture data as follows:

(4) performing azimuth FFT on the full-aperture data;

(5) carrying out residual frequency modulation compensation;

(6) performing azimuth inverse FFT;

2. The method of claim 1, wherein: the formula of the scaling factor in the step (2) is as follows:

wherein,

Δα(f_a)＝α(f_a)-α(f_dc)，

3. The method of claim 2, wherein:

4. The method of claim 3, wherein: in the step (2), an azimuth compression function for azimuth compression and compensation of the residual phase is as follows:

wherein