CN102590797A - Efficient imaging processing method for onboard spotlight synthetic aperture radar (SAR) - Google Patents

Abstract

The invention provides an efficient imaging processing method suitable for onboard spotlight synthetic aperture radar (SAR), which mainly overcomes the defects that the conventional imaging method is long in data processing time and cannot meet the requirement for instantaneity in actual electronic warfare. The method comprises the following imaging processes of: performing de-chirp processing on signals by adopting a linear de-chirp technology, and determining range and azimuth direction positions of the target signals by adopting a Fourier transform technology. During the de-chirp processing, the vertical distance between an imaging regional center and an onboard air route is selected as the reference distance in numerous imaging algorithms, and the instantaneous slant distance between the imaging regional center and the radar is selected in the method, so that the range migration curve is similar to a straight line, range migration compensation is not required, the calculation quantity is greatly reduced, and real-time imaging of a target can be realized. The imaging processing method only requires two times of complex multiplication and two times of Fourier transform, and is a simple, quick and efficient imaging method.

Description

The efficient image processing method of airborne bunching type SAR

Technical field

The present invention relates to a kind of Radar Signal Processing technology, be specifically related to a kind of echo data disposal route that is applicable to the high-resolution radar of little scene imaging.

Background technology

In the radar imagery field, synthetic-aperture radar becomes the research focus at present, and according to imaging pattern commonly used, synthetic-aperture radar can be divided into strip-type, bunching type and scan-type.Than other imaging pattern, the resolution of bunching type SAR generally is applicable to little scene imaging than higher, particularly certain target is carried out to picture.Along with science and technology development, in the electronic warfare, need follow the tracks of and accurately image in modern times target, development has great importance for national defence therefore to study bunching type SAR imaging algorithm.

So far the imaging algorithm of bunching type SAR is existing a lot, can accomplish the accurately image to target, but algorithm is too complicated; The overlong time of data processing, in the electronic warfare of reality, real-time is determining the victory or defeat of fight to a great extent; For little scene imaging region, this paper considers the requirement based on real time imagery in the real system, has adopted the imaging algorithm that a kind of step is simple, operand is little; This method can realize the real time imagery to target, is the high-efficient treatment method of bunching type SAR.

Summary of the invention

The objective of the invention is in order to adapt to the needs of real-time in the hyundai electronics war, provide that a kind of step is simple, calculated amount is little, the image processing method of the bunching type SAR of easy realization.

The objective of the invention is to realize like this:

Whole radar to the course of work of target imaging is: radar emission electromagnetic wave signal at first, signal are received by radar receiver behind target reflection in communication process, then adopt the respective imaging algorithm to accomplish the imaging to target.For the imaging pattern of bunching type, multiple imaging algorithm is arranged, be respectively apart from range and Doppler, polar format algorithm; Wavenumber Domain Algorithms, range migration algorithm, frequency become the mark algorithm, and these algorithms satisfy the application demand in each field basically, but in electronic warfare; Real-time is the key factor of decision war victory; These algorithms are too complicated, and the overlong time of data processing can not satisfy the requirement of real-time.The algorithm that the present invention proposes only needs twice complex multiplication operation and twice Fourier transform can accomplish the imaging to target, is a kind of simple, fast and efficient formation method.

Data handling procedure is: at first signal is separated linear frequency modulation and handle; Make echoed signal the distance to the orientation upwards all become the simple signal component and; The frequency values size is relevant with the position of impact point, and the frequency values of definite signal can be confirmed the position of point target then.Then signal is carried out the Fourier transform that distance makes progress, this moment, the frequency of distance value can be determined, and can confirm that promptly point target is in position that distance makes progress.Introduced remaining video phase item owing to adopt to separate after linear frequency modulation is handled, it can make the different frequency component produce the delay of different distance time, is equivalent to and in apart from frequency domain, has introduced a phase place, therefore need get rid of.At last signal is carried out the Fourier transform that the orientation makes progress and to accomplish imaging process target.

Core technology of the present invention is to separate choosing of linear frequency modulation process middle distance parameter; The numerous imaging algorithm of tradition all is imaging scene center and the vertical range of carrying the machine line of flight doing the reference distance of choosing when frequency modulation removal is handled, and the distance parameter that the present invention chooses is the instantaneous oblique distance at radar and imaging region center.

Linear accent process is frequently separated in concrete realization: separate the completion before radar receiver carries out the A/D sampling of linear frequency modulation technology; It is with an echoed signal and a linear frequency modulation reference signal phase mixing with identical chirp rate, and the signal of doing frequency in time linear change becomes single-frequency signals.The echoed signal that distance makes progress; In fast time domain its be the simple signal component and; The frequency size of signal is corresponding with the distance and position of target; Be each frequency range and distance on the frequency domain upwards the impact point of certain position is corresponding, only need to signal carry out Fourier transform can confirm point target distance to the position.After handling through frequency modulation removal, the point target orientation to echoed signal also be simple signal, and signal frequency is relevant with the position of orientation of impact point, therefore need one time Fourier transform can confirm the orientation to the position.

Advantage of the present invention is: when satisfying imaging requirements, the imaging algorithm of numerous complicated is simplified, removed redundant operation, image-forming step is simplified, become the imaging algorithm that operand is less, elapsed time is short, can satisfy the requirement of real-time.

Description of drawings

Fig. 1 is the process flow diagram of bunching type SAR imaging algorithm of the present invention;

Fig. 2 is that the present invention carries out the orientation when handling to signal, and used reference distance concerns synoptic diagram, and its mid point O is the reference center of imaging region, and A is the imaging point of point target, and x is the coordinate that point target makes progress in the orientation, R ₀Be imaging center and the vertical range of carrying the machine course line, R _aBe the distance of reference center and radar, R (t _a) be the distance of point target and radar, v is for carrying the movement velocity of machine;

Fig. 3 is under simulated environment, the position of the point target of setting relation;

Fig. 4 is the one-dimensional range profile of point target on range direction;

Fig. 5 is the one dimension orientation picture that point target makes progress in the orientation;

Fig. 6 is the two-dimensional imaging figure as a result of point target.

Embodiment

With reference to Fig. 1, the present invention realizes that the imaging process of bunching type SAR is following:

[0016] step 1: the original echo to radar receives is separated the linear frequency modulation processing; Reference signal is the linear FM signal that and echoed signal have identical chirp rate; And reference distance is chosen for the imaging reference center and carries the instantaneous oblique distance between the machine course line; Separate linear frequency modulation handle signal on the range direction of back become the simple signal component with; Its frequency values size is one to one with the distance of target scattering point, therefore can confirm the position of each scattering point on range direction by the size of frequency values.Simultaneously, separate linear frequency modulation and handle and also to have removed the frequency modulation characteristic that signal makes progress in the orientation, make signal upwards also become in the orientation simple signal component and, the size of its frequency values is relevant with the position that point target makes progress in the orientation.

[0017] its detailed process is following: radar adopts the pattern of " step one stops " in motion process; Be that signal is regarded radar as static in communication process, describe the electromagnetic wave propagation time respectively and carry machine exomonental moment in motion process with fast time and slow time.For guarantee effect distance and high-resolution imaging, radar selects for use linear FM signal as the transmitted waveform of gathering the object scene data, and its expression formula is:

$S_t(\hat{t},t_m)=\mathrm{rect}\left(\frac{\hat{t}}{T_p}\right)\exp \left(j2\mathrm{\π}(f_{c}t+\mathrm{\γ}\hat{t}/2)\right)$

In the formula $\mathrm{Rect}\left(\frac{\hat{t}}{T_p}\right)=\left\{\begin{array}{cc}1& \left|\hat{t}\right|<T_p/2\\ 0& \mathrm{Else}\end{array}\right.,$ T _pBe pulse width, γ is γ=B/T for the linear frequency modulation rate _p, B is a signal bandwidth; f _cBe carrier frequency;

For the fast time, be defined as

M ∈ Z, wherein T is the signal repetition period; t _mFor the slow time, be defined as t _m=mT, m ∈ Z;

When resolution was very high, target can be several strong scattering points by equivalence, so the echo of radar reception is the echo sum of each scattering point, and its expression is:

$S_r(\hat{t},t_m)=\underset{i}{\mathrm{\&Sigma;}}{\mathrm{\&sigma;}}_i\mathrm{rect}\left(\frac{\hat{t}-{\mathrm{\&tau;}}_i}{T_p}\right)\exp \left(j2\mathrm{\&pi;}\right[f_c(t-{\mathrm{\&tau;}}_i)+\mathrm{\&gamma;}{(\hat{t}-{\mathrm{\&tau;}}_i)}^2/2\left]\right)$

σ in the formula _iIt is the complex reflection coefficient of i scattering point; τ _iBe i pairing echo time delay of scattering point and τ _i=2R _i/ c; R _iIt is the distance of i scattering point and radar; C is a velocity of wave.

Separate the linear frequency modulation technology and can remove the frequency modulation characteristic of signal, reduce the bandwidth of signal simultaneously, reduce rear end equipment the signals sampling rate.It is with an echoed signal and a linear frequency modulation reference signal phase mixing with identical chirp rate, and the signal of doing frequency in time linear change becomes single-frequency signals.

This reference signal is: $S_{\mathrm{Ref}}(\hat{t},t_m)=\mathrm{Exp}\left(j2\mathrm{\&pi;}\right[f_c(t-{\mathrm{\&tau;}}_m)+\mathrm{\&gamma;}{(\hat{t}-{\mathrm{\&tau;}}_m)}^2/2\left]\right)$

τ wherein _m=2R _m/ c, R _mBe the instantaneous distance of difference moment radar to the imaging region center.

Echoed signal can be expressed as after linear frequency modulation is handled through separating:

Wherein

R (t _m) be the distance between different slow Texas towers constantly and the impact point, R _mBe the distance of slow Texas tower constantly of difference and imaging scene center, first phase place be mutually distance to echoed signal, the orientation that second phase term is echo is to signal, the 3rd phase term is the remaining video phase place, this is the garbage that needs removal.

First phase term is the echoed signal that distance makes progress, and it is a simple signal in fast time domain, and the frequency size of signal is corresponding with the distance and position of target, and promptly each frequency range on the frequency domain is with corresponding apart from the impact point of the certain position that makes progress.Only need to signal carry out Fourier transform can confirm point target distance to the position.

R is supposed to signal in the orientation that second phase term is echo ₀Be the vertical range of Texas tower with imaging scene center line, point target is positioned at the center line of scene, and the orientation is x to coordinate, and v is the movement velocity of Texas tower, and its concrete position relation is as shown in Figure 2, and have this moment:

$R\left(t_m\right)-R_m=\sqrt{{\mathrm{<figure-callout\; id="0"\; label="R"\; filenames="HSA00000671220400012.png,HSA00000671220400022.png"\; state="\{\{state\}\}">R</figure-callout>}}_0^2+v^2{t}_a^2}-\sqrt{{\mathrm{<figure-callout\; id="0"\; label="R"\; filenames="HSA00000671220400012.png,HSA00000671220400022.png"\; state="\{\{state\}\}">R</figure-callout>}}_0^2+{({\mathrm{vt}}_a-x)}^2}\&ap;-\frac{2{\mathrm{vxt}}_a}{2{\mathrm{<figure-callout\; id="0"\; label="R"\; filenames="HSA00000671220400012.png,HSA00000671220400022.png"\; state="\{\{state\}\}">R</figure-callout>}}_0}+\frac{x^2}{2{\mathrm{<figure-callout\; id="0"\; label="R"\; filenames="HSA00000671220400012.png,HSA00000671220400022.png"\; state="\{\{state\}\}">R</figure-callout>}}_0}$

The orientation to the Doppler frequency of echoed signal is:

This shows, after handling through frequency modulation removal, the point target orientation to echoed signal be the simple signal component with, and the signal frequency size is relevant with the position of orientation of impact point, therefore only need one time Fourier transform can confirm the orientation to the position.

Step 2: on range direction, carrying out Fourier transform, the different frequency values of signal is separated in frequency domain, can confirms the position of point target on range direction like this through the signal of separating after linear frequency modulation is handled.

Step 3 after above-mentioned processing, is then removed the remaining video phase of signal.This remnants video phase is to separate linear frequency modulation to handle the intrinsic phase term in back; It makes the different frequency component produce the delay of different distance time, is equivalent to and in apart from frequency domain, has introduced a phase place, if do not remove; Can produce, reduce image quality at last apart from phase error

Step 4 is upwards carried out Fourier transform to the signal after handling in the orientation, the simple signal that the orientation is made progress is separated in frequency domain, can confirm the position that point target makes progress in the orientation like this.

In order to verify the validity of the inventive method, point target is carried out simulating, verifying at present.Simulation parameter, bandwidth: 150MHz, wavelength: 0.2m, carrier frequency: 1.5GHz, sampling rate: 512MHz, pulse repetition rate: 256Hz, pulsewidth: 1.33us, length of synthetic aperture: 200m, year motor speeds: 100m/s, operating distance: 3Km.

The position of point target relation is set to as shown in Figure 3 in the emulation; Simulation result is like Fig. 4,5, shown in 6, and by Figure 4 and 5 are visible as a result, the main lobe peak value that impact point is corresponding is higher; Be easy to distinguish the number and the position of impact point; Owing to there are two positions of scattering point on range direction identical, so Fig. 4 shown two main lobes, tallies with the actual situation.Visible by Fig. 5; Three scattering points are fixed to the position that makes progress with the orientation in distance exactly; With the position relationship consistency of three scattering points shown in Figure 3, and the focusing effect of image is better, shows that thus the inventive method can accomplish the accurately image to target; And be the imaging algorithm that a kind of step is simple, operand is little, can realize real time imagery target.

Claims (3)

1. the efficient image processing method of an airborne bunching type SAR is characterized in that:

Echoed signal at first will be passed through and separated the linear frequency modulation processing, can reduce the bandwidth of signal so effectively, reduces rear end A/D signals sampling rate, still is the important step that echoed signal is carried out process of pulse-compression simultaneously;

Signal after separating the processing of linear conditioning technology frequently only needs promptly can confirm the position of target on range direction apart to accomplishing the pulse compression process through fourier transform technique;

Signal is carried out the orientation to the confirming of position, need to the signal after handling adopt fourier transform technique can confirm target in the orientation to the position;

2. according to the efficient image processing method of the bunching type SAR in the claim 1, wherein step (1) is described separates linear accent process frequently, realizes as follows:

Separating linear accent method frequently is to fix with a time; Frequency, chirp rate and the identical linear FM signal signal as a reference that transmits; Do Frequency mixing processing with the echoed signal that receives; Handle the simple signal of the target echo at different distance place, back, if measure each frequency component of echoed signal can accomplish confirming of different distance place target corresponding to different frequency;

3. according to the linear frequency modulation method of separating in the claim 2, in its reference function each parameter to choose situation following:

As linear FM signal; Its frequency, chirp rate and transmit identically, wherein also having an important parameter is reference distance, how choosing reference distance is emphasis of the present invention; What general imaging algorithm was chosen is the vertical range of imaging reference center and radar; The reference distance that the present invention chooses is the instantaneous distance of radar and imaging reference center, and processing can be in line the range migration curve approximation like this, need not to carry out the range migration compensation; Through with the reference signal Frequency mixing processing after, signal the distance to the orientation upwards all become the simple signal component and.

Images