CN102955152A - Synthetic aperture radar (SAR) signal simulation method for sea waves - Google Patents

Abstract

The invention discloses a synthetic aperture radar (SAR) signal simulation method for sea waves. The SAR signal simulation method comprises the following steps of: S1, modeling the sea waves by adopting a sea wave spectrum, and constructing a three-dimensional sea surface comprising a plurality of surface elements; S2, calculating a scattering field of each surface element, then converting a motion effect of each surface element along the distance direction into excursion along the orientation direction according to a speed beaming model, and generating a sea surface scattering graph according to the scattering fields and the excursion in the orientation direction; and S3, simulating an SAR signal of the sea surface scattering graph for overall motion to obtain an analog SAR signal, and processing the analog SAR signal by an SAR imaging algorithm to obtain the analog SAR image. By the SAR signal simulation method for the sea waves, a Doppler effect caused by sea surface motion is considered, so that SAR signal simulation of the sea waves can be realized under the condition of strong storm, and comprehensive analog data can be supplied to design of an SAR system.

Description

A kind of SAR signal imitation method of wave

Technical field

The present invention relates to the ocean application of synthetic-aperture radar, particularly a kind of synthetic-aperture radar signal imitation method of wave.

Background technology

SAR (being synthetic-aperture radar) is to improve its spatial resolution by the signal that obtains in certain integral time, and for the land static scene, the relative position relation between atural object and the sensor is by the satellite orbit unilateral decision; For Ocean Scenes, the dynamic fluctuating on sea has caused extra motion, has influence on its imaging effect in imaging process.Therefore, the SAR signal imitation of carrying out wave can improve the understanding to whole imaging process, and has great significance for sensor design, parameter selection etc.

SAR signal imitation is based upon on the scene scattering model basis, in SAR signal imitation process, needs to consider aspect, obtains simulated data by the relation that makes up scene scattering model and aspect.The SAR signal imitation method of the wave of prior art has only been considered the motion of wave integral body, the model that adopts in SAR signal imitation process is distribution table surface model (DS), and do not consider the sea caused Doppler effect that moves, therefore the result who the SAR signal of wave is simulated is not accurate enough, consider that the move main models of caused Doppler effect of sea is speed pack model (VB), it is according to being the SAR image-forming principle of moving object.The SAR image modeling of wave mainly adopts M4S and two modeling tools of RIM in the prior art, the groundwork of these two modeling tools is impacts of analyzing stormy waves stream interaction partners sea surface roughness (being surface scattering), and do not consider that the sea moves caused Doppler effect to the impact of surface scattering, and the filter effect of the imaging of SAR system is not related to.

In sum, although the SAR signal imitation method of wave and sea SAR modeling tool can be set up backward scattered model in the prior art on the basis of ocean wave spectrum and scattering model, realize the SAR signal imitation of wave in the less stormy waves situation, but the problem owing to the caused Doppler effect that moves on existence shortage consideration sea aspect the SAR signal imitation, cause to realize the SAR signal imitation than wave in the high sea situation, therefore can't provide for the design of SAR system comprehensive simulated data.

Summary of the invention

In view of this, the synthetic-aperture radar signal imitation method that the purpose of this invention is to provide a kind of wave, with the SAR signal imitation method that solves prior art because the problem than the SAR signal imitation of wave in the high sea situation of can't realizing that lacks that the consideration sea moves that caused Doppler effect causes.

To achieve these goals, the invention provides a kind of SAR signal imitation method of wave, may further comprise the steps:

S1: adopt ocean wave spectrum that wave is carried out modeling, structure comprises the surface, three-dimensional sea of a plurality of bins;

S2: calculate the scattered field of each described bin, then according to speed pack model with described bin along distance to exercise effect be converted into the orientation to skew, and according to described scattered field and described orientation to skew generate surface scattering figure;

S3: the SAR signal imitation of described surface scattering figure being carried out mass motion obtains simulating the SAR signal, and adopts the SAR imaging algorithm that described simulation SAR signal is processed to obtain simulating the SAR image.

As preferably, among the described step S1, described ocean wave spectrum is expressed as long-pending form:

W(K)＝S(K)P(θ，K)

Wherein, W (K) is ocean wave spectrum;

S (K) is the one dimension spectrum, and at Long wavelength region, S (K) adopts Classical Spectrum; In short wavelength regions, S (K) is divided into gravity capillary wave S _Gc(K) and capillary wave S _c(K), wave number corresponding to difference;

P (θ, K) is spread function;

K is the wave wave vector, and mould K=2 π/Λ of K is the wave wave number, and Λ is the wave wavelength;

θ defines by following formula:

$\left\{\begin{array}{c}{K}_x=K\cos \mathrm{\θ}\\ K_y=K\sin \mathrm{\θ}\end{array}\right.$

Wherein, K _xThat wave wave vector K is at the component of x direction, K _yThat wave wave vector K is at the component of y direction.

As preferred further, described Classical Spectrum is the JONSWAP spectrum.

As preferably, among the described step S1, adopt the surface, the described three-dimensional sea of ocean surface Construction of A Model of two yardsticks, be specially: the profile of the corresponding large scale of long wave, by less than SAR resolution elements and approximate much larger than the plane bin of electromagnetic wavelength, generate at random sea; The random rough of the corresponding small scale of shortwave is approximate by the waveform of gravity capillary wave or capillary wave; The waveform of described gravity capillary wave or capillary wave is superimposed upon and forms extra large surface model on the bin of described plane.

As preferred further, described long wave adopts the section of surging to generate in sometime stochastic process, and parameter comprises significant wave height, peak period, the main direction of ocean wave spectrum and ocean wave spectrum type; The method that generates described at random sea from the long wave spectrum is to adopt Noise Filter, and be specially: at first generate the random complex matrix N, its real part and imaginary part are evenly distributed between 0 to 1; Then the correspondence long wave spectrum W that multiplies each other _L(K) root and matrix N obtain intermediate variable L _T=0:

$L_{t=0}=\sqrt{W_L\left(K\right)}N$

After following formula carries out Fourier transform, sea level height ζ _T=0(r) be expressed as:

ζ _t＝0(r)＝real(aF ^-1[L _t＝0](r))

Wherein, a is normalized parameter, and r is the position of wave in reference field, comprises the positional information on x and the y both direction, and real is got in real () expression.

As preferably, among the described step S2, the scattered field of each described bin adopts following formula to calculate:

$\left|\begin{array}{c}{E}_H^s\\ E_V^s\end{array}\right|=\frac{\mathrm{<figure-callout\; id="4"\; label="jk"\; filenames="HDA0000107306650000041.png,HDA0000107306650000063.png"\; state="\{\{state\}\}">jk</figure-callout>}}{4\mathrm{\&pi;R}}\exp (-\mathrm{jkR})\mathrm{\&chi;}\left|\begin{array}{c}{E}_H^i\\ E_V^i\end{array}\right|$

Wherein,

With

The incident field,

With

Be scattered field, R is that the bin central point is to the distance between the line of flight; K=|k|=2 π/λ is the electromagnetism wave number, and k is that incident electromagnetic wave vows that λ is electromagnetic wavelength; Subscript H and V represent respectively horizontal and vertical polarization part; Subscript i and s represent respectively incident field and scattered field; Matrix χ represents backscattering coefficient, and j is imaginary unit.

As preferred further, described matrix χ is provided by following formula:

χ＝FD(θ _i，δ _x，δ _y)

Wherein, θ _iThe radar incident angle, δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y;

Matrix F represents the radar back scattering because bin direction and electromagnetic parameter rely on the polarization effect that causes;

Scalar D is expressed as:

D(θ _i，δ _x，δ _y)＝∫ _Aexp(j2k·ρ)dA

Wherein, k is that incident electromagnetic wave is vowed, A is the bin area that bin projects to (x, y) plane, and ρ be from the bin central point to bin on the radial vector of any point;

$\mathrm{\&rho;}=(x-x_0)\hat{x}+(y-y_0)\hat{y}+(z(x,y)-z_0)\hat{z}$

Wherein, (x ₀, y ₀, z ₀) be the centre coordinate of bin,

Be respectively x, y, the unit vector of z direction.

Z (x, y)=z ₀+ δ _x(x-x ₀)+δ _y(y-y ₀)+z _s(x, y), wherein, δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y; z _s(x, y) is the stochastic process that meets the shortwave spectrum.

As preferably, among the described step S2, according to speed pack model with described bin along distance to exercise effect be converted into the orientation to skew be specially:

Scattering surface unit according to the distance to speed component carry out the orientation to adjustment:

$\mathrm{\&Delta;x}=-\frac{R}{V}v$

Wherein, R is the distance that sensor arrives described scattering bin, and V is sensor speed;

According to the movement velocity of sea bin, the orientation is that the scattering bin of x will be mapped to coordinate

$x^{\&prime;}=x-\frac{R}{V}v(r,t)$

And the scattering bin along the speed of Electromagnetic Wave Propagation direction is:

$v(r,t)=\underset{K}{\mathrm{\&Sigma;}}\mathrm{\&omega;}\sqrt{W_L\left(K\right)}l({\mathrm{\&theta;}}_i,{\mathrm{\&theta;}}_0)\cos \left({\mathrm{\&theta;}}_i\right)\sin (K\&CenterDot;r-\mathrm{\&omega;t}+\mathrm{\&delta;})$

Wherein, ω is the wave frequency, W _L(K) be the long wave spectrum, K is the wave wave vector;

δ=tan ^-1(tan θ _iSin θ ₀), θ _iThe radar incident angle, θ ₀That heading and wave wave vector K are at the angle of extra large surface direction; R represents wave position in the horizontal direction, and it comprises the positional information on x and the y both direction.

Compared with prior art, the present invention has following beneficial effect: the SAR signal imitation method of wave provided by the invention is owing to considered the sea caused Doppler effect that moves, make it possible to realize the SAR signal imitation than wave in the high sea situation, comprehensive simulated data can be provided for the design of SAR system.

Description of drawings

Fig. 1 is the schematic flow sheet of the SAR signal imitation method of wave of the present invention.

Fig. 2 is the SAR imaging coordinate system.

The sea level height figure of Fig. 3 for obtaining in the experiment one.

Fig. 4 for the sea distance of output in the experiment one to bin orientation that speed causes to deflection graph.

The surface scattering distribution plan of Fig. 5 for obtaining in the experiment one.

The analog echo signal figure of Fig. 6 for obtaining in the experiment one.

The simulation SAR image of Fig. 7 for obtaining in the experiment one.

Fig. 8 is the probability function density map of the simulation SAR image of acquisition in the experiment one.

Fig. 9 is the image spectrum of the simulation SAR image of acquisition in the experiment one.

Figure 10 is the ocean wave spectrum of input in the experiment one.

Figure 11 is the ocean wave spectrum comparison diagram of input in the experiment two.

The sea level height figure comparison diagram of Figure 12 for obtaining in the experiment two.

The simulation SAR image comparison figure of Figure 13 for obtaining in the experiment two.

Figure 14 is the image spectrum comparison diagram of the simulation SAR image of acquisition in the experiment two.

Embodiment

Below in conjunction with accompanying drawing specific embodiments of the invention are elaborated.

As shown in Figure 1, the SAR signal imitation method of wave of the present invention comprises following three steps:

S1: adopt ocean wave spectrum that wave is carried out modeling, structure comprises the surface, three-dimensional sea of a plurality of bins; Wherein,

1, ocean wave spectrum

The description of wave mainly adopts ocean wave spectrum to carry out, and ocean wave spectrum W (K) is typically expressed as long-pending form:

W(K)＝S(K)P(θ，K) (1)

Wherein, S (K) is the one dimension spectrum, and at Long wavelength region, S (K) can adopt various Classical Spectrum, for example the JONSWAP spectrum; In short wavelength regions, S (K) is divided into gravity capillary wave S _Gc(K) and capillary wave S _c(K), wave number corresponding to difference;

P (θ, K) is spread function;

K is the wave wave vector, and mould K=2 π/Λ of K is the wave wave number, and Λ is the wave wavelength;

θ defines by following formula:

$\left\{\begin{array}{c}{K}_x=K\cos \mathrm{\&theta;}\\ K_y=K\sin \mathrm{\&theta;}\end{array}\right.---\left(1\right)$

Wherein, K _xThat wave wave vector K is at the component of x direction, K _yThat wave wave vector K is at the component of y direction.

2, sea level height modeling

The height on sea is described as with the variation of space and time:

$\mathrm{\&zeta;}(r,t)=\underset{K}{\mathrm{\&Sigma;}}\sqrt{W_L\left(K\right)}\cos (K\&CenterDot;r-\mathrm{\&omega;t})---\left(2\right)$

Wherein, W _L(K) be the long wave spectrum, K is the wave wave vector; R represents wave position in the horizontal direction, and it comprises the positional information on x and the y both direction; ω is the wave frequency, and t is the time.

The ocean surface Construction of A Model of two yardsticks is adopted on surface, three-dimensional sea, is specially: the profile of the corresponding large scale of long wave, and by less than SAR resolution elements, but approximate much larger than the plane bin of electromagnetic wavelength, generate at random sea; The random rough of the corresponding small scale of shortwave, approximate by suitable shortwave spectrum (such as the waveform of gravity capillary wave or capillary wave), the waveform of these gravity capillary waves or capillary wave is superimposed upon and forms extra large surface model on the bin of described plane.It is very little that the Main Basis that adopts surface, the three-dimensional sea of said method structure is that the resolution of typical SAR system is compared with the long wave of wave, but compare very large with the shortwave of wave.

Long wave adopts the section of surging to generate in sometime stochastic process, and parameter comprises significant wave height, peak period, the main direction of ocean wave spectrum and ocean wave spectrum type etc.A common way that generates described at random sea from the long wave spectrum is to adopt Noise Filter, and its method is specially: at first generate the random complex matrix N, its real part and imaginary part are evenly distributed between 0 to 1, the in the same size on the size of matrix N and surface, three-dimensional sea; Then the correspondence long wave spectrum W that multiplies each other _L(K) root and matrix N obtain intermediate variable L _T=0:

$L_{t=0}=\sqrt{W_L\left(K\right)}N---\left(3\right)$

After formula (4) was carried out Fourier transform, the bin height on surface, three-dimensional sea can be expressed as:

ζ _t＝0(r)＝real(aF ^-1[L _t＝0](r)) (4)

Wherein, a is normalized parameter, and its value is relevant with Fourier Transform Algorithm; R represents wave position in the horizontal direction, and it comprises the positional information on x and the y both direction; Real is got in real () expression.

3, wave velocity modeling

The bin that consists of surface, three-dimensional sea along distance to speed be:

$v(r,t)=\underset{K}{\mathrm{\&Sigma;}}\mathrm{\&omega;}\sqrt{W_L\left(K\right)}l({\mathrm{\&theta;}}_i,{\mathrm{\&theta;}}_0)\cos \left({\mathrm{\&theta;}}_i\right)\sin (K\&CenterDot;r-\mathrm{\&omega;t}+\mathrm{\&delta;})---\left(6\right)$

Wherein, ω is the wave frequency, W _L(K) be the long wave spectrum, K is the wave wave vector; δ=tan ^-1(tan θ _iSin θ ₀), θ _iThe radar incident angle, θ ₀That heading and wave wave vector K are at the angle of extra large surface direction; R represents wave position in the horizontal direction, and it comprises the positional information on x and the y both direction.

S2: calculate to consist of the scattered field of each bin on surface, three-dimensional sea, then according to speed pack model with described bin along distance to exercise effect be converted into the orientation to skew, and according to described scattered field and described orientation to skew generate surface scattering figure;

Long wave spectrum W is being adopted in the impact that the scattering of bin is calculated the locality that needs the whole bin of consideration and depended on the shortwave spectrum of bin _L(K) in the sea profile that makes up, whole sea is by (being surface level) evenly distributes on the xy plane, and has the point of certain altitude to represent in the z direction.

Single bin scope adopts following formula to represent:

z(x，y)＝z ₀+δ _x(x-x ₀)+δ _y(y-y ₀)+z _s(x，y) (7)

Wherein, (x ₀, y ₀, z ₀) be the centre coordinate of bin; δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y; z _s(x, y) is the stochastic process that meets the shortwave spectrum.

1, the Bragg diffraction of oblique incidence

In order to estimate backscattering coefficient figure, we must consider the interaction that incident electromagnetic wave is surperficial with the sea, and for little incident angle, Specular reflection must be considered; For medium incident angle, dominant mechanism is the Bragg reflection that expands to oblique incidence; For large incident angle, significantly effect is for blocking and the diffraction phenomenon; But the minimum and great incident angle under the limiting case does not have actual value for the SAR system on sea.

The scattered field of each bin can pass through Kirchhoff approximation (KA, kirchhoff) approximate treatment, and final scattered field and the relation of incident field can be expressed as:

$\left|\begin{array}{c}{E}_H^s\\ E_V^s\end{array}\right|=\frac{\mathrm{<figure-callout\; id="4"\; label="jk"\; filenames="HDA0000107306650000041.png,HDA0000107306650000063.png"\; state="\{\{state\}\}">jk</figure-callout>}}{4\mathrm{\&pi;R}}\exp (-\mathrm{jkR})\mathrm{\&chi;}\left|\begin{array}{c}{E}_H^i\\ E_V^i\end{array}\right|---\left(8\right)$

Wherein,

With

The incident field,

With

Be scattered field, R is that the bin central point is to the distance between the line of flight; K=|k|=2 π/λ is the electromagnetism wave number, and k is that incident electromagnetic wave vows that λ is electromagnetic wavelength; Subscript H and V represent respectively horizontal and vertical polarization part; Subscript i and s represent respectively incident field and scattered field; J is imaginary unit; Matrix χ represents backscattering coefficient, is provided by following formula:

χ＝FD(θ _i，δ _x，δ _y) (9)

Wherein, θ _iThe radar incident angle, δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y;

Matrix F has represented the radar back scattering because bin direction (being the mean surface of bin) and electromagnetic parameter rely on the polarization effect that (being specific inductive capacity) causes;

Scalar D is expressed as:

D(θ _i，δ _x，δ _y)＝∫ _Aexp(j2k·ρ)dA (10)

It has represented shortwave spectrum on extra large surface for the contribution of scattered field, and k is that incident electromagnetic wave vows that A is the bin area that bin projects to (x, y) plane in following formula, and ρ is the radial vector of point general on from the bin central point to bin;

$\mathrm{\&rho;}=(x-x_0)\hat{x}+(y-y_0)\hat{y}+(z(x,y)-z_0)\hat{z}---\left(11\right)$

Wherein, (x ₀, y ₀, z ₀) be the centre coordinate of bin, formula (7) is seen in z (x, y) definition,

Be respectively x, y, the unit vector of z direction.

Calculate D (θ _i, δ _x, δ _y), mainly based on back scattering mainly by determining with the corresponding small wave spectrum composition of electromagnetic field wavelength.Because the height of wave is much smaller than its wavelength, so with exp (j2kz _sCos θ _i) carry out the progression decomposition afterwards and keep the first rank, can obtain:

$D({\mathrm{\&theta;}}_i,{\mathrm{\&delta;}}_x,{\mathrm{\&delta;}}_y)=\mathrm{<figure-callout\; id="4"\; label="j"\; filenames="HDA0000107306650000041.png,HDA0000107306650000063.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{4\mathrm{\&pi;}}{\mathrm{\&lambda;}}{Z}_s({\stackrel{\&OverBar;}{K}}_x,{\stackrel{\&OverBar;}{K}}_y)\cos {\mathrm{\&theta;}}_i---\left(12\right)$

Wherein,

$\left\{\begin{array}{c}{\stackrel{\&OverBar;}{K}}_x=\frac{4\mathrm{\&pi;}}{\mathrm{\&lambda;}}{\mathrm{\&delta;}}_x\cos {\mathrm{\&theta;}}_i\\ {\stackrel{\&OverBar;}{K}}_y=\frac{4\mathrm{\&pi;}}{\mathrm{\&lambda;}}(\sin {\mathrm{\&theta;}}_i+{\mathrm{\&delta;}}_y\cos {\mathrm{\&theta;}}_i)\end{array}\right.---\left(13\right)$

Wherein, λ is electromagnetic wavelength, δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y, θ _iIt is the radar incident angle;

And Z _s() is the fourth contact Gaussian random variable, and it all just is:

$<{\left|Z_s({\stackrel{\&OverBar;}{K}}_x,{\stackrel{\&OverBar;}{K}}_y)\right|}^2>=A\&CenterDot;W({\stackrel{\&OverBar;}{K}}_x,{\stackrel{\&OverBar;}{K}}_y)---\left(14\right)$

Wherein,＜expression statistical average, and W () basis

Value determine it is short gravity wave, gravity capillary wave or capillary wave wave spectrum, A is the area that bin projects to surface level.

2, speed pack

After scattering is calculated, in order to determine the position of this bin on scatter diagram, introduced speed pack model.

According to basic synthetic aperture principle, synthetic-aperture radar positions all targets according to Doppler frequency corresponding to position angle.When target along distance when moving, produce difference with reference frequency.Therefore, can produce the orientation to the skew of image and corresponding location mistake.

Scattering surface unit according to the distance to speed component v (r, t) carry out the orientation to adjustment:

$\mathrm{\&Delta;x}=-\frac{R}{V}v(r,t)---\left(15\right)$

Wherein, R is the distance that sensor arrives the scattering bin, and V is sensor speed.

According to the movement velocity v (r, t) of sea bin, the orientation is that the scattering bin of x will be mapped to coordinate

$x^{\&prime;}=x-\frac{R}{V}v(r,t)---\left(16\right)$

Wherein, the movement velocity v (r, t) of sea bin sees formula (6).

According to dispersion relation:

ω ²＝g|K| (17)

Wherein, g is acceleration of gravity;

Formula (16) is transformed to:

$x^{\&prime;}=x-\underset{K}{\mathrm{\&Sigma;}}\mathrm{\&Lambda;\&alpha;}\sin (K\&CenterDot;r-\mathrm{\&omega;t}+\mathrm{\&delta;})---\left(18\right)$

Wherein, the phase bit position of intrinsic (constant) is ignored.

$\mathrm{\&alpha;}=\sqrt{2\mathrm{\&pi;g}}\frac{R\cos {\mathrm{\&theta;}}_i}{V}\frac{\sqrt{W\left(K\right)}}{{\mathrm{\&Lambda;}}^{3/2}}l({\mathrm{\&theta;}}_i,{\mathrm{\&theta;}}_0)---\left(19\right)$

S3, the SAR signal imitation that described surface scattering figure is carried out mass motion obtain simulating the SAR signal, and adopt the SAR imaging algorithm that described simulation SAR signal is processed to obtain simulating the SAR image.

After obtaining the scatter diagram of each bin, the SAR signal imitation that it can be carried out mass motion obtains simulating the SAR signal, then adopts the SAR imaging algorithm that described simulation SAR signal is processed and obtains simulating the SAR image.Then adopt novel SAR treatment technology test focusing technology that simulation SAR image is carried out the aftertreatments such as feature extraction.This is the difference of SAR signal simulator and simple SAR image simulation, and is particularly useful for sea condition.

As shown in Figure 2, the SAR system is at height H=Rcos θ _i, fly at a constant speed along the x direction.Surface scattering according to bin with and upper shortwave spectrum and obtain the back scattering field according to the speed bunching effect.The position of x ' expression sensor.

The SAR signal indication of considering the sea mass motion is:

$s(x^{\&prime;},r^{\&prime;})=\frac{1}{{\left(2\mathrm{\&pi;}\right)}^2}\exp \left\{j\right[{\mathrm{\&omega;}}_0-{\mathrm{\&omega;}}_{\mathrm{\&xi;}}{\mathrm{\&xi;}}_0-{\mathrm{\&omega;}}_{\mathrm{\&eta;}}{\mathrm{\&eta;}}_0\left]\frac{x^{\&prime;}}{V}\right\}\&CenterDot;s_0[(1+\frac{{\mathrm{\&omega;}}_{\mathrm{\&xi;}}}{V})x^{\&prime;},r^{\&prime;}+\frac{{\mathrm{\&omega;}}_{\mathrm{\&eta;}}}{V}{x}^{\&prime;}]---\left(20\right)$

Wherein:

s ₀(x′，r′)＝∫∫dxdrγ(x，r，x′＝0)g(x′-x，r′-r；r) (21)

ω=ω (ξ, η) is the dispersion relation of long wave; ξ=K _x, η=K _y/ sin θ _i, θ wherein _iIt is the radar incident angle; ${\mathrm{\&omega;}}_{\mathrm{\&xi;}}=\&PartialD;\mathrm{\&omega;}/\&PartialD;\mathrm{\&xi;}{|}_{\mathrm{\&xi;}={\mathrm{\&xi;}}_0,\mathrm{\&eta;}={\mathrm{\&eta;}}_0},$ ${\mathrm{\&omega;}}_{\mathrm{\&eta;}}=\&PartialD;\mathrm{\&omega;}/\&PartialD;\mathrm{\&eta;}{|}_{\mathrm{\&xi;}={\mathrm{\&xi;}}_0,\mathrm{\&eta;}={\mathrm{\&eta;}}_0},$ Wherein, ξ ₀And η ₀Corresponding to dominant wavelength in the long wave spectrum; And ω ₀It is corresponding angular frequency.Scatter diagram after the process speed pack of γ (x, r, x '=0) expression position x '=0 is adjusted; G (x '-x, r '-r; R) be the transport function of static scene; SAR signal s before the interpolation ₀() at first calculates according to the method for Fourier transform, and then final s () obtains by interpolation.

After obtaining simulation SAR signal, can adopt general SAR formation method to process, obtain simulation SAR image.

Simulate effect for the SAR signal imitation method of analyzing wave provided by the invention has adopted the method for statistical distribution and image analysis of spectrum that the SAR signal imitation method of wave of the present invention is tested, and image setting is 1024 pixels.

Major parameter sees Table 1:

Table 1

Parametric variable	Parameter	Value	Unit
				spectype	The frequency spectrum kind	Jonswap
windspeed	Wind speed	8	m/s
				H m0	Significant wave height	1	m

[0144]

T p	Peak period	10	s
				θ 0	The frequency spectrum angle	45	°
s	The spread function coefficient	150
				Nx	The orientation is to the bin number	2000
Ny	Distance is to the bin number	2000
				dx	The orientation is to sizing grid	1.4	m
dy	Distance is to sizing grid	1.9	m
				θ i	Incident angle	30	°
Polarization	Polarization mode	HH
				λ	Electromagnetic wavelength	0.03	m
maxaz	The scatter diagram orientation is to pixel count	1024
				maxrg	The scatter diagram distance is to pixel count	1024
ColumnSpacing	Column pitch (orientation to)	4.2	m
				RowSpacing	Line space (distance to)	7.9	m
e	The specific inductive capacity of seawater	80+4i
				τ	Burst length	37.1E-6	sec
f c	Sample frequency	18.96E6	Hz
				Δf	Pulse bandwidth	15.55E6	Hz
c	The light velocity	3.E8	m/sec
				Daz	Antenna size	12	m
V	Flying speed	7100	m/sec
				PRF	Pulse repetition rate	1678	Hz
R 0	Sensor distance	6E5	m
				RangeResolution	Range resolution	9.6	m

[0145]

AzimuthResolution

Azimuth resolution

6.0

m

In order to simulate the high impact on imaging of different waves, the below is high to the 1m wave, and 3m wave height and the high situation of 8m wave are simulated, and at first carry out the description of whole simulation process as example take 1m wave height.

Experiment one:

The below analyzes the result in some simulation processes, and verifies whole simulation process.At first clear and definite input function has represented parameter and the value used in the simulation process in table, and the implication of expression.

The ocean wave spectrum of input is the Jonswap spectrum, and significant wave height is 1m, and peak period is 10s, and the angle of spectrum peak and radar heading is 45 °, and spread function uses the cos-2s type, and the s parameter selects 150; Obtain one 2000 * 2000 sea level height data, resolution be the orientation to sizing grid be 1.4m, distance to sizing grid be 1.9m.The sea level height data of obtaining are seen Fig. 3.

And exported for each bin along distance to the bin orientation that causes of speed to skew, such as Fig. 4.

The shortwave spectrum adopts formula (12) and (13), and parameter is 10 meters high wind speeds, and value is 8m/s.Consider extra large superficial velocity to the impact of scattering bin according to the speed bunching effect, the scatter diagram that obtains the sea is seen Fig. 5.

The simulation SAR signal indication that obtains is Fig. 6.

Carry out imaging acquisition simulation SAR image and see Fig. 7.

The below analyzes the probability density distribution situation of simulation SAR image.Adopted rayleigh distributed, Lognormal to distribute, Weibull distributes and Gamma distributes carries out parameter fitting.See Fig. 8.KS represents the KS test coefficient, and all the other are the parameter of each distribution.

Fig. 9 is seen in spectrum calculating to analog image.

Corresponding sea level height spectrum is seen Figure 10 with it.

As can be seen from the results, in the smaller situation of the significant wave height of wave, the pass between ocean wave spectrum and the image spectrum is linear relationship.

Experiment two:

After the simulation in the situation of carrying out the 1m significant wave height, the below is to 3m and than simulating in the high 8m situation of billow.Only provide final scatter diagram, simulation SAR image and ocean wave spectrum to compare.

At first be input message, the ocean wave spectrum of input is the Jonswap spectrum, sees Figure 11.

The wave height map of simulation is seen Figure 12.

Simulate the simulation SAR image of acquisition and see Figure 13.

The image spectrum of analyzing is separately seen Figure 14.

On the proof of analog result of different significant wave heights the impact of speed bunching effect on sea SAR imaging.The result shows that large unrestrained height has caused the nonlinear effect of SAR imaging.

The invention provides a kind of SAR signal imitation method of the wave based on the speed bunching effect, input ocean wave spectrum information, sensor information, carry out three-dimensional extra large surface simulation, consider that Bragg diffraction and speed bunching effect are to the adjustment of bin, adopt the SAR signal imitation of mass motion, and carry out imaging acquisition Ocean surface simulation SAR image.Significant wave height to 1m, 3m and 8m is analyzed, and has obtained the analog image under the different situations, has verified the impact of speed bunching effect on final imaging results.

The SAR signal imitation method of wave provided by the invention has been owing to considered the sea caused Doppler effect that moves, and makes it possible to realize the SAR signal imitation than wave in the high sea situation, and comprehensive simulated data can be provided for the design of SAR system.

Above embodiment is exemplary embodiment of the present invention only, is not used in restriction the present invention, and protection scope of the present invention is defined by the claims.Those skilled in the art can make various modifications or be equal to replacement the present invention in essence of the present invention and protection domain, this modification or be equal to replacement and also should be considered as dropping in protection scope of the present invention.

Claims (8)

1. the SAR signal imitation method of a wave is characterized in that, may further comprise the steps:

S1: adopt ocean wave spectrum that wave is carried out modeling, structure comprises the surface, three-dimensional sea of a plurality of bins;

S2: calculate the scattered field of each described bin, then according to speed pack model with described bin along distance to exercise effect be converted into the orientation to skew, and according to described scattered field and described orientation to skew generate surface scattering figure;

S3: the SAR signal imitation of described surface scattering figure being carried out mass motion obtains simulating the SAR signal, and adopts the SAR imaging algorithm that described simulation SAR signal is processed to obtain simulating the SAR image.

2. the SAR signal imitation method of wave according to claim 1 is characterized in that, among the described step S1, described ocean wave spectrum is expressed as long-pending form:

W(K)＝S(K)P(θ，K)

Wherein, W (K) is ocean wave spectrum;

S (K) is the one dimension spectrum, and at Long wavelength region, S (K) adopts Classical Spectrum; In short wavelength regions, S (K) is divided into gravity capillary wave S _Gc(K) and capillary wave S _c(K), wave number corresponding to difference;

P (θ, K) is spread function;

K is the wave wave vector, and mould K=2 π/Λ of K is the wave wave number, and Λ is the wave wavelength;

θ defines by following formula:

$\left\{\begin{array}{c}{K}_x=K\cos \mathrm{\&theta;}\\ K_y=K\sin \mathrm{\&theta;}\end{array}\right.$

Wherein, K _xThat wave wave vector K is at the component of x direction, K _yThat wave wave vector K is at the component of y direction.

3. the SAR signal imitation method of wave according to claim 2 is characterized in that, described Classical Spectrum is the JONSWAP spectrum.

4. the SAR signal imitation method of wave according to claim 1, it is characterized in that, among the described step S1, adopt the surface, the described three-dimensional sea of ocean surface Construction of A Model of two yardsticks, be specially: the profile of the corresponding large scale of long wave, by less than SAR resolution elements and approximate much larger than the plane bin of electromagnetic wavelength, generate at random sea; The random rough of the corresponding small scale of shortwave is approximate by the waveform of gravity capillary wave or capillary wave; The waveform of described gravity capillary wave or capillary wave is superimposed upon and forms extra large surface model on the bin of described plane.

5. the SAR signal imitation method of wave according to claim 4 is characterized in that, described long wave adopts the section of surging to generate in sometime stochastic process, and parameter comprises significant wave height, peak period, the main direction of ocean wave spectrum and ocean wave spectrum type; The method that generates described at random sea from the long wave spectrum is to adopt Noise Filter, and be specially: at first generate the random complex matrix N, its real part and imaginary part are evenly distributed between 0 to 1; Then the correspondence long wave spectrum W that multiplies each other _L(K) root and matrix N obtain intermediate variable L _T=0:

$L_{t=0}=\sqrt{W_L\left(K\right)}N$

After following formula carries out Fourier transform, sea level height ζ _T=0(r) be expressed as:

ζ _t＝0(r)＝real(aF ^-1[L _t＝0](r))

Wherein, a is normalized parameter, and r is the position of wave in reference field, comprises the positional information on x and the y both direction, and real is got in real () expression.

6. the SAR signal imitation method of wave according to claim 1 is characterized in that, among the described step S2, the scattered field of each described bin adopts following formula to calculate:

$\left|\begin{array}{c}{E}_H^s\\ E_V^s\end{array}\right|=\frac{\mathrm{jk}}{4\mathrm{\&pi;R}}\exp (-\mathrm{jkR})\mathrm{\&chi;}\left|\begin{array}{c}{E}_H^i\\ E_V^i\end{array}\right|$

Wherein,

With

The incident field,

With

Be scattered field, R is that the bin central point is to the distance between the line of flight; K=|k|=2 π/λ is the electromagnetism wave number, and k is that incident electromagnetic wave vows that λ is electromagnetic wavelength; Subscript H and V represent respectively horizontal and vertical polarization part; Subscript i and s represent respectively incident field and scattered field; Matrix χ represents backscattering coefficient, and j is imaginary unit.

7. the SAR signal imitation method of wave according to claim 6 is characterized in that, described matrix χ is provided by following formula:

χ＝FD(θ _i，δ _x，δ _y)

Wherein, θ _iThe radar incident angle, δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y;

Matrix F represents the radar back scattering because bin direction and electromagnetic parameter rely on the polarization effect that causes;

Scalar D is expressed as:

D(θ _i，δ _x，δ _y)＝∫ _Aexp(j2k·ρ)dA

Wherein, k is that incident electromagnetic wave is vowed, A is the bin area that bin projects to (x, y) plane, and ρ be from the bin central point to bin on the radial vector of any point;

$\mathrm{\&rho;}=(x-x_0)\hat{x}+(y-y_0)\hat{y}+(z(x,y)-z_0)\hat{z}$

Wherein, (x ₀, y ₀, z ₀) be the centre coordinate of bin,

Be respectively x, y, the unit vector of z direction.

Z (x, y)=z ₀+ δ _x(x-x ₀)+δ _y(y-y ₀)+z _s(x, y), wherein, δ _xAnd δ _yBe bin along the orientation bin slope to x and distance direction y; z _s(x, y) is the stochastic process that meets the shortwave spectrum.

8. the SAR signal imitation method of wave according to claim 1 is characterized in that, among the described step S2, according to speed pack model with described bin along the distance to exercise effect be converted into the orientation to skew be specially:

Scattering surface unit according to the distance to speed component carry out the orientation to adjustment:

$\mathrm{\&Delta;x}=-\frac{R}{V}v$

Wherein, R is the distance that sensor arrives described scattering bin, and V is sensor speed;

According to the movement velocity of sea bin, the orientation is that the scattering bin of x will be mapped to coordinate

$x^{\&prime;}=x-\frac{R}{V}v(r,t)$

And the scattering bin along the speed of Electromagnetic Wave Propagation direction is:

$v(r,t)=\underset{K}{\mathrm{\&Sigma;}}\mathrm{\&omega;}\sqrt{W_L\left(K\right)}l({\mathrm{\&theta;}}_i,{\mathrm{\&theta;}}_0)\cos \left({\mathrm{\&theta;}}_i\right)\sin (K\&CenterDot;r-\mathrm{\&omega;t}+\mathrm{\&delta;})$

Wherein, ω is the wave frequency, W _L(K) be the long wave spectrum, K is the wave wave vector;

δ=tan ^-1(tan θ _iSin θ ₀), θ _iThe radar incident angle, θ ₀That heading and wave wave vector K are at the angle of extra large surface direction; R represents wave position in the horizontal direction, and it comprises the positional information on x and the y both direction.

Images