CN101477195B - Method for acquiring radar cross section - Google Patents

Abstract

The present invention relates to a kind of radar remote sensing technology, especially a kind of method for obtaining radar cross section. The described method includes: obtaining every ray from the path parameter for being incident on reflection after infinite ray invests target face element; According to the every ray got from the path parameter for being incident on reflection, the mirror field of every ray projection surface is obtained; The mirror field of acquisition is passed through into face element integral formula:

Obtain the scattered field of every ray projection surface; The scattered field of every ray projection surface of acquisition is overlapped; It is an advantage of the invention that not only speed is fast, accuracy is higher, but also is conducive to the higher-frequency radar sectional area simulation of multi-angle multi-frequency multipolarization, to realize the simulation to higher-frequency radar image.

Description

Obtain the method for radar cross section

Technical field

The present invention relates to a kind of radar remote sensing technology, particularly a kind of method that obtains radar cross section.

Background technology

In the radar application field, the scattering of calculation of complex target object has important effect to the radar cross section that obtains target object.There is the accurate predictor method of a large amount of radar cross sections in the prior art, as MOM, FEM, FDTD, though very high by the radar cross section accuracy of these methods acquisitions, be only applicable to finding the solution of low frequency range and resonance region.And at high frequency region, on object and the electromagnetic wave part interaction very strong, so prior art mainly adopts ray tracking method approximate treatment complex target radar cross section.So-called ray tracking method is meant by the path of following the tracks of ray judges visible face, when the nearest target bin that search and ray intersect, comes the acceleration search process by constructing certain bin storage organization.The main advantage of ray tracking method is to calculate flexibly, can calculate repeatedly reflection, but that its major defect is a speed is slow, and the result of the radar cross section of acquisition is also inaccurate.

Summary of the invention

The purpose of this invention is to provide and a kind ofly also can accurately obtain the method for radar cross section at high band, this method not only speed is fast, and the result of the radar cross section that obtains also relatively accurately.

In order to realize the foregoing invention purpose, the embodiment of the invention provides a kind of method that obtains radar cross section, and this method comprises:

After infinite ray is invested the target bin, obtain every ray from inciding the path parameter of reflection;

From inciding the path parameter of reflection, obtain the mirror field of every ray projection surface according to every the ray that gets access to;

The mirror field that obtains is passed through the bin integral formula: $S={\∫}_{S_1}\exp \[\mathrm{jkr}(\hat{i}-\hat{s})\]\mathrm{dS}$ Obtain the scattered field of every ray projection surface;

The scattered field of every ray projection surface obtaining is superposeed;

Described integral formula: $S={\∫}_{S_1}\exp \[\mathrm{jkr}(\hat{i}-\hat{s})\]\mathrm{dS}$ In

Be the incident direction vector,

Be the scattering direction vector, r is a position vector.

Effect intentionally of the present invention: not only speed is fast, accuracy is higher, and helps the higher-frequency radar sectional area simulation of multi-angle multi-frequency multipolarization, thereby realizes the simulation to the higher-frequency radar image.

Description of drawings

Fig. 1 is ray reflecting effect figure;

Fig. 2 obtains the process flow diagram of the method for radar cross section for the present invention;

Fig. 3 is for obtaining the method flow diagram of raypath by the BSP tree; Fig. 4 is the coordinate synoptic diagram;

Fig. 5 is the corner reflector point yupin effect figure of the embodiment of the invention.

Embodiment

The present invention is described in further detail below in conjunction with accompanying drawing.

As shown in Figure 2, this method comprises the steps:

Step 1, after infinite ray is invested the target bin, obtain every ray from inciding the path parameter of reflection;

, earlier a large amount of infinite rays is invested target from sensor herein, thereby make ray evenly distribute and cover whole target as Fig. 1.From the incidence point of ray, at first the bin on search and its crossing nearest target object is tried to achieve intersection point, obtains reflected ray according to reflection theorem then, follows the tracks of so repeatedly, leaves target object up to the final reflection of ray.

In the process of finding the solution nearest joining, the method that has adopted hierarchical data to divide the space, i.e. y-bend spatial division tree BSP (Binary Space Partitioning Tree).For cutting apart of plane, to use and axially cut out the plane, segmentation plane is parallel to three coordinate axis respectively.In the BSP structure, a node comprises one and cuts out face, and two subtrees, and these two subtrees comprise all bin data that are positioned at the face of cutting out one side respectively.And in the process that makes up the BSP tree, need all models be placed on respectively in the subtree of the left and right sides according to reduction face.In the middle of divisional plane all is arranged at every turn.The condition setting that stops to iterate be the bin number of single subtree less than designated value or subtree level greater than designated value.

Concrete method is as follows:

At first calculate the little bounding box of each triangle bin, just can surround the cubical a pair of summit of the minimum of each triangle bin, be called the angle point of bounding box; According to the angle point of these little bounding boxs, calculate total big bounding box of all bins again; At the center of the x axle of big bounding box, set up divisional plane, the bin about divisional plane is saved in left and right sides subtree respectively; At the center of the y axle of big bounding box, set up divisional plane, the bin about divisional plane is saved in left and right sides subtree respectively; At the center of the z of big bounding box axle, set up divisional plane, the bin about divisional plane is saved in left and right sides subtree respectively;

Finding the solution of ray and the nearest joining of target.Structure by the BSP tree quickens this search procedure.To the node of BSP tree, the near subtree of selected distance ray eye point is operated, and until leaf node, obtains the bin that leaf node contains and the intersection point of ray.

Concrete algorithm flow as shown in Figure 3.

Wherein calculate sp ₃, sp ₁, sp ₂, K _N, K _FDefinition and be calculated as follows:

At first according to initial point s and direction

Definition ray ray.The present node of BSP tree is K, and its left subtree is K _L, its right subtree is K _R, bounding box is box, divisional plane is C.Calculate two intersection point p of ray ray and big bounding box box ₁, p ₂And the intersection point p of ray ray and divisional plane C ₃Try to achieve apart from sp ₃, sp ₁, sp ₂(sp ₁＜sp ₂).Define nearly subtree K _NBe the subtree at ray initial point s place, another subtree is subtree K far away _F

Three kinds of situations are defined as follows:

Situation 1:sp ₃＜0 or sp ₃＞sp ₂

Situation 2:0＜sp ₃＜sp ₁

Situation 3: except situation 1,2.

1. calculate two intersection point p of ray ray and big bounding box box ₁, p ₂And the intersection point p of ray ray and divisional plane C ₃Try to achieve apart from sp ₃, sp ₁, sp ₂(sp ₁＜sp ₂).Define nearly subtree K _NBe the subtree at ray initial point s place, another subtree is subtree K far away _F

2. when satisfying situation 1sp ₃＜0 or sp ₃＞sp ₂The time, only to nearly subtree K _NContinue step 1; Satisfy situation 2 as 0＜sp ₃＜sp ₁The time, only to subtree K far away _FContinue step 1.Be arrival leaf node K by condition _DAll the other are the third situation, earlier to nearly subtree K _N, continue step 1; With subtree K far away _FBe pressed into stack Z.

3. behind the arrival leaf node, ask ray ray and leaf node K _DThe joining of all bins that comprise if intersection point is arranged, is chosen nearest joining, returns joining.If no intersection point, and current subtree is nearly subtree K _N, higher level's subtree satisfies situation three, then continues subtree K far away _FCarry out step 1.Otherwise, from stack Z, take out subtree as root node, continue the 1-3 step.If stack Z is empty, then return no intersection point.

According to above-mentioned flow process, if after finding joining, the reflection spot of trying to achieve next time from this joining along reflection direction leaves target up to order of reflection above designated value or ray.According to above flow process, calculate all rays follow to the path of leaving target, and note.

Two uses that parameter is come the standard ray have been used in actual applications: radiographic density and maximum reflection number of times.Ray number in the radiographic density, the unit's of being meant wavelength.Radiographic density is below a times of single beam wavelength preferably, and radiographic density little then computing time of short precision is low.Because the employing ray trace is found the solution electromagnetic scattering and the GRECO method all needs the discrete dough sheet that turns to less pixel character of bin.Need to consider the size of dough sheet under discrete case, for the purpose of accurately, radiographic density is preferably below a times of a beam wavelength, is 1/10th of a beam wavelength such as radiographic density.The ray minor shifts will overlap with the ray on next door like this, and the contribution difference of adjacent ray is little, all can be very accurate to the calculating of fluctuating bin.The maximum reflection number of times is meant the order of reflection of the maximum of following the tracks of ray.It is the ray trace stage, stops one of condition of following the tracks of.Arrive the maximum reflection number and ray stops following the tracks of when not leaving target at order of reflection, can raise the efficiency though reduced precision like this.

Every ray that step 2, basis get access to obtains the mirror field of every ray projection surface from inciding the path parameter of reflection;

, obtain ray after the reflection paths parameter of target surface herein, need further try to achieve the mirror field of ray projection surface, and the method for approximation in geometric optics is mainly used in the acquisition of mirror field, promptly be similar to and think ray, and intensity is constant along rectilinear propagation.Because the initial value of field intensity is the component of electric field in polarised direction, so after ray runs into target, according to the difference of targeted surface material, adopt different specific inductive capacity to calculate mirror field, reflection coefficient adopts the Fresnel coefficient.

Concrete grammar is:

From the path parameter that incides reflection the incident field is set according to every the ray that gets access to.The field intensity direction is decomposed into TM, TE ripple, calculates TM, TE wave reflection coefficient respectively, stack obtains mirror field polarization coefficient Γ _nAgain with the mirror field of previous step as the incident field, calculate the mirror field of this step once more.Thereby repeated calculation is left target up to reaching maximum reflection number of times or ray, gets reflection polarization coefficient Γ to the end _N, Γ _NProject to and receive polarised direction E _r, Γ _NE _rCalculate the light path of every ray at last.The distance, delta r of ray initial point to reflection spot calculated in each reflection _nAll distances and last reflection spot are obtained a light path that ray is total to reception place apart from addition $r=\underset{n=1}{\overset{\max }{\mathrm{\Σ}}}\mathrm{\Δ}{r}_n+\mathrm{\Δ}{r}_r,$ Wherein n is an order of reflection, and max is maximum order of reflection.

Step 3, with the mirror field that obtains by the bin integral formula: $S={\∫}_{S_1}\exp [jkr(\hat{i}-\hat{s})]\mathrm{dS}$ Obtain the scattered field of every ray projection surface;

Herein, be that four rays are chosen at the center in incident ray on every side with the incident ray, follow the tracks of this four rays, primary event to the last gets four eye points to the end.According to these four outgoing position calculation shape functions.Shape function specifically is defined as:

$S={\∫}_{S_1}\exp [jkr(\hat{i}-\hat{s})]\mathrm{dS}$

Wherein

The incident direction vector, Be the scattering direction vector, r is a position vector.The scope of integration is four quadrilaterals that ray surrounds.Be converted to according to rectangle is approximate:

$S=\mathrm{LW}\frac{\sin \[\frac{1}{2}\mathrm{kL}\·(\hat{i}-\hat{s})\]}{\frac{1}{2}\mathrm{kL}\·(\hat{i}-\hat{s})}\·\frac{\sin \[\frac{1}{2}\mathrm{kW}\·(\hat{i}-\hat{s})\]}{\frac{1}{2}\mathrm{kW}\·(\hat{i}-\hat{s})}$

The scattered field of step 4, every ray projection surface that will obtain superposes.

Herein, each bar ray all carries out field intensity, light path, bin integral and calculating, and the contribution of scatters of every ray just can be found the solution by these three parameters.Total scattered field is the stack of the contribution of each bar ray.

The scattering of each ray of superposeing is asked for total electric field contribution and is specially:

$E_{\mathrm{all}}=\underset{\mathrm{allrays}}{\mathrm{\Σ}}{\mathrm{\Γ}}_N\·E_r\·S\·e^{\mathrm{jkr}}$

Γ wherein _NBe the reflection polarization coefficient, E _rBe the polarised direction that receives, S is a shape function, and r is total light path.

Shown in following formula, scattered field is projected to polarised direction, and add corresponding coefficient, just obtain radar cross section.

Embodiment

Present embodiment is example with the corner reflector, because corner reflector is a kind of more special scatterer, because three scatterings, it all has bigger contribution of scatters in very big angular range.Because this specific character, corner reflector is commonly used for the calibration object in radar remote sensing.Therefore this paper has also selected corner reflector, and by calculating its contribution of scatters, physical optics and geometrical optics approach are used in checking, to the validity of the scattering mechanism modeling of reflection repeatedly.

The corner reflector length of side in the present embodiment is 13 times of beam wavelength, i.e. 13 λ; Another length of side is 5 times of beam wavelength, i.e. 5 λ; Calculate in a series of angles of vertical and horizontal direction respectively.Final experimental result and measurement result contrast, maximum error illustrates that analog result is effectively reliable within 2dB.The RCS that 90 different angles that the orientation is made progress carried out for 1 to 4 step calculates, and obtains the result and True Data compares.

System coordinate system is based upon on the coordinate system of object module as shown in Figure 4, and Target Setting is at initial point O.The P point is a sensor, and the incident direction of wave travel is

Horizontal polarization directions is $\stackrel{\→}{H}=\hat{z}\×\stackrel{\→}{\mathrm{PO}},$ Vertical polarization directions is $\stackrel{\→}{V}=\stackrel{\→}{\mathrm{PO}}\×\stackrel{\→}{H}.$ The direction parameter of imaging is specially:

Be called azimuth angle theta in the projection on xy plane and the angle of x axle,

With the angle on xy plane be the angle of pitch

Step 1 adopts the BSP tree to organize on model.Determine the position distribution of space parallel ray initiating terminal according to position angle and the angle of pitch, directions of rays is definite object then.Adopting a ray of the radiographic density unit of equaling beam wavelength, is 1.7cm for the wavelength of the frequency correspondence of 18GHz, every 1.7cm a ray emission is arranged like this.Set up a plane in the position of distance object 800m, every ray interval 1.7cm evenly distributes on vertical direction and horizontal direction.Ray can surround whole target.Carry out finding the solution of closest approach at every ray, adopt the BSP tree to quicken.Starting point and three parameters of direction according to the root node in the BSP tree, incident ray are carried out finding the solution of closest approach as the condition of input, again according to ray, bounding box and three parametric solution joinings of divisional plane distance to the ray starting point.Direction after the reflection is carried out secondary tracking again as reflected ray.So repeatedly, find the solution repeatedly reflection, order of reflection is the parameter as ray trace to the maximum 5 times.Obtain the path of single ray like this, adopt several sections rays to represent, and comprise normal direction, the angle point information of reflecting surface.All ray trace are obtained the set of paths of one group of ray.These information are delivered to step 2 and carry out electromagnetism calculating.

Step 2 is carried out electromagnetism Scattering Calculation according to the routing information of top ray, the polarised direction that adopts the HH polarization setting to receive and launch.The incident electric field is set to 1.From step 1 path, obtain the information of first joining, comprised the angle point information of crossing bin, joining position and material information.Calculate the normal vector of bin by angle point

, the reflected ray direction $\stackrel{\→}{r}=\stackrel{\→}{i}+2\stackrel{\→}{n}.$ Light path is calculated as the joining position and arrives the distance of ray initial point and the distance that joining arrives acceptance point.Incident wave can be decomposed into two groups of independently electromagnetic waves, and wherein electric vector is called the TM ripple in the plane of incidence, and electric vector is called the TE ripple perpendicular to the plane of incidence.The reflecting part is decomposed into TM, TE ripple with the field intensity direction, calculates TM, TE wave reflection coefficient respectively, all equals 1 as perfect conductor, and stack obtains mirror field polarization coefficient Γ _nFor single ray, comprised 5 secondary reflections, need find the solution the field intensity of reflection and find the solution again the field intensity of secondary reflection as next incident field intensity for each reflection.

Step 3 is to find the solution the step of bin integration, has adopted rectangle approximate, thinks that the represented beam of wall scroll ray all passes through the reflection on same surface.Algebraic expression according to bin integral representation: $S=\mathrm{LW}\frac{\sin \[\frac{1}{2}\mathrm{kL}\·(\hat{i}-\hat{s})\]}{\frac{1}{2}\mathrm{kL}\·(\hat{i}-\hat{s})}\·\frac{\sin \[\frac{1}{2}\mathrm{kW}\·(\hat{i}-\hat{s})\]}{\frac{1}{2}\mathrm{kW}\·(\hat{i}-\hat{s})}.$ Parameter in the formula is tried to achieve according to four positions that surround ray, these four are surrounded ray and obtain for the initial point from central ray respectively moves 1.7/2cm up and down, find the solution repeatedly up to this secondary reflector, L and W are respectively two limits that these four rays are formed rectangle in the formula.

Step 4 is utilized formula according to step 2, step 3:

$E_{\mathrm{all}}=\underset{\mathrm{allrays}}{\mathrm{\Σ}}{\mathrm{\Γ}}_N\·E_r\·S\·e^{\mathrm{jkr}}$

Obtain the parametric solution radar cross section.All scattering stacks of all rays obtain total scattered field.

By changing the radar cross section that the position angle or the angle of pitch obtain different azimuth, effect finds out that significantly improved algorithm can obtain good radar cross section and calculate effect as shown in Figure 5 at last.

Claims (3)

1. a method that obtains radar cross section is characterized in that, comprising:

Infinite ray is invested target, and obtain every ray from inciding the path parameter that leaves described target;

According to the described path parameter of every the ray that gets access to, obtain the mirror field of every ray at projection surface;

With the described mirror field that obtains according to the first bin integral formula:

Carry out integration to obtain the scattered field of every ray at projection surface, wherein, Be the incident direction vector,

Be the scattering direction vector, r is a position vector;

Every the ray that obtains is superposeed at the scattered field of projection surface, and described stack is concrete adopts following formula to carry out:

$E_{\mathrm{all}}=\underset{\mathrm{all\; rays}}{\mathrm{\Σ}}{\mathrm{\Γ}}_N\·E_r\·S\·e^{\mathrm{jkr}}$

Γ wherein _NBe the reflection polarization coefficient, E _rBe the polarised direction that receives, S is a shape function, and r is total light path; Obtain the radar cross section of described target then according to the result of stack.

2. the method for acquisition radar cross section according to claim 1 is characterized in that, the radiographic density of described ray is below a times of beam wavelength, and described radiographic density is the ray number in the unit wavelength.

3. the method for acquisition radar cross section according to claim 1, it is characterized in that, when the scope of described integration be four rays surround quadrilateral the time, according to following step of carrying out described integration by the second converted bin integral formula of the described first bin integral formula:

Wherein, L and W are the vector along plane width and length direction.

Images