CN103064072B - Background signal extraction method in radar scattering cross section measurement - Google Patents

Abstract

The invention provides a background signal extraction method in radar scattering cross section measurement. The method includes measuring echo signals used for obtaining radar transmission signals reflected by a dihedral angle reflector, isolating the echo signals into in-phase channel signals and orthogonal channel signals, obtaining first complex signals through reconstruction of the in-phase channel signals, obtaining second complex signals through the orthogonal channel signals, respectively obtaining real part signals and imaginary part signals of the first complex signals and the second complex signals by means of a curve fitting algorithm, obtaining the real part signals of background signals according to the real part signals and the imaginary part signals of the first complex signals, and obtaining the imaginary part signals of background signals according to the real part signals and the imaginary part signals of the second complex signals. Due to the fact that the background signals are extracted through the curve fitting algorithm in high precision, the background signal extraction method in the radar scattering cross section measurement solves the problem that robustness is poor in practical engineering application.

Description

Background signal extracting method in RCS measurement

Technical field

The present invention relates to Radar Signal Processing Technology, (Radar CrossSection is called for short: the background signal extracting method in RCS) measuring to particularly relate to a kind of RCS.

Background technology

In RCS, (Radar Cross Section is called for short: in RCS) measuring, the extraction of background signal and the counteracting of background signal are its important steps.Wherein, the precision that background signal extracts directly affects the effect that background signal is offset, so the precision that background signal extracts is particularly important.

Prior art proposes a kind of Dihedral Corner Reflectors wheel measuring to extract the method for polarization background signal, and polarized signal is expressed as by it:

M _pq(θ)=c _pqcos2θ+s _pqsin2θ+B _i+N _i （11）

Wherein, M is the echoed signal of the radar emission signal measuring the Dihedral Corner Reflectors reflection obtained, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives.B is background signal, and N is noise signal, i=1,2,3,4, c _pqand s _pqby the wavelength of wide and high, the radar emission signal of Dihedral Corner Reflectors, and normalized cross polar component ε _h, ε _vdetermine, be the relevant physical quantity such as same Dihedral Corner Reflectors physical dimension, measuring system parameter, without the need to obtaining in background extracting, different polarization modes has different computing method.θ is the angle that the broken line of dihedral angle turns clockwise in yOz plane.

Again mathematical expectation is asked to formula (11):

E{M _pq(θ)}=E{c _pqcos2θ}+E{s _pqsin2θ}+E{B _i}+E{N _i} （12）

If to Dihedral Corner Reflectors by anglec of rotation uniform sampling and the angle making it rotate is 2n π (n>=1), then due to E{c _pqcos2 θ }=0, E{s _pqsin2 θ }=0, E{N _i}=0, therefore

E{M _pq(θ)}=E{B _i} （13）

Therefore, if total measurement hits is K, then have

$B_i\≈\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{M}_{\mathrm{pq}}\left({\mathrm{\θ}}_k\right)---\left(14\right)$

Therefore, averaging by measuring sample summation to K, can background signal be extracted.

In prior art, because formula (11) is based upon measurement data not exist on the hypothesis basis of any undesired signal, there is relatively large deviation in result and the ideal value of gained, the precision of the background signal namely extracted is low, so the poor robustness of this algorithm.

Summary of the invention

The invention provides the background signal extracting method in a kind of RCS measurement, for extracting the background signal of degree of precision, solving the problem of the poor robustness in practical engineering application.

The invention provides the background signal extracting method in a kind of RCS measurement, comprising:

Measure the echoed signal of the radar emission signal obtaining Dihedral Corner Reflectors reflection;

Described echoed signal is separated into homophase channel signal and orthogonal channel signal;

Adopt described homophase channel signal to reconstruct acquisition first complex signal, adopt described orthogonal channel signal reconstruction to obtain the second complex signal;

Curve fitting algorithm is adopted to obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively;

According to solid part signal and the imaginary signals of the background signal of described first complex signal, obtain the solid part signal of the background signal of described echoed signal, according to solid part signal and the imaginary signals of the background signal of described second complex signal, obtain the imaginary signals of the background signal of described echoed signal.

Further, described described echoed signal is separated into homophase channel signal and orthogonal channel signal, comprises:

Formula (1) is adopted to obtain described homophase channel signal M _ipq(θ), this signal can be expressed as:

M _Ipq(θ)=c _Ipqcos2θ+s _Ipqsin2θ+B _Ii+N _Ii （1）

Formula (2) is adopted to obtain described orthogonal channel signal M _qpq(θ), this signal can be expressed as:

M _Qpq(θ)=c _Qpqcos2θ+s _Qpqsin2θ+B _Qi+N _Qi； （2）

Wherein, I is homophase passage, and Q is orthogonal channel, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives.C _ipq, s _ipq, c _qpqand s _ipqfor the physical quantity that same Dihedral Corner Reflectors physical dimension, measuring system parameter etc. are relevant, without the need to obtaining in background extracting.B _ifor the background signal of homophase channel signal, B _qfor the background signal of orthogonal channel signal, N _ifor the noise of homophase channel signal, N _qfor the noise of orthogonal channel signal, i=1,2,3,4.θ is the angle that the broken line of dihedral angle turns clockwise in yOz plane.

Further, described employing described homophase channel signal reconstruct acquisition first complex signal, comprising:

By described homophase channel signal M _ipq(θ) in, the signal of θ <n π+3 π/4 is as the solid part signal of described first complex signal, by described homophase channel signal M _ipq(θ) in, the signal of θ>=n π+3 π/4 is as the imaginary signals of described first complex signal;

The described orthogonal channel signal reconstruction of described employing obtains the second complex signal, comprising:

By described orthogonal channel signal M _qpq(θ) in, the signal of θ <n π+3 π/4 is as the solid part signal of described second complex signal, by described orthogonal channel signal M _qpq(θ) in, the signal of θ>=n π+3 π/4 is as the imaginary signals of described second complex signal.

Further, described by described homophase channel signal M _ipq(θ) in, the signal of θ <n π+3 π/4 is as the solid part signal of described first complex signal, by described homophase channel signal M _ipq(θ) in, the signal of θ>=n π+3 π/4 is as the imaginary signals of described first complex signal, comprising:

Formula (3) reconstruct is adopted to obtain described first complex signal:

${{M_I}^{\&prime;}}_{\mathrm{pq}}\left(\mathrm{\&theta;}\right)=c_{\mathrm{Ipq}}{e}^{j2\mathrm{\&theta;}}+{s_{\mathrm{Ipq}}e}^{j(\frac{\mathrm{\&pi;}}{2}-2\mathrm{\&theta;})}+\sqrt{2}(B_{\mathrm{Ii}}+N_{\mathrm{Ii}})e^{j\frac{\mathrm{\&pi;}}{4}}---\left(3\right)$

Wherein, I is homophase passage, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives.B _ifor the background signal of homophase channel signal, N _ifor the noise of homophase channel signal, i=1,2,3,4;

Described by described orthogonal channel signal M _qpq(θ) in, the signal of θ <n π+3 π/4 is as the solid part signal of described second complex signal, by described orthogonal channel signal M _qpq(θ) in, the signal of θ>=n π+3 π/4 is as the imaginary signals of described second complex signal, comprising:

Formula (4) reconstruct is adopted to obtain described second complex signal:

${{M_Q}^{\&prime;}}_{\mathrm{pq}}\left(\mathrm{\&theta;}\right)=c_{\mathrm{Qpq}}{e}^{j2\mathrm{\&theta;}}+{s_{\mathrm{Qpq}}e}^{j(\frac{\mathrm{\&pi;}}{2}-2\mathrm{\&theta;})}+\sqrt{2}(B_{\mathrm{Qi}}+N_{\mathrm{Qi}})e^{j\frac{\mathrm{\&pi;}}{4}}---\left(4\right)$

Wherein, Q is orthogonal channel, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives.B _qfor the background signal of orthogonal channel signal, N _qfor the noise of orthogonal channel signal, i=1,2,3,4.

Further, described employing curve fitting algorithm obtains solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively, comprising:

Formula (5) and formula (6) is adopted to obtain fitting circle parameter A, B, C, D:

$\min \underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{({\mathrm{Ax}}_i^2+{\mathrm{Ay}}_i^2+{\mathrm{Bx}}_i+{\mathrm{Cy}}_i+D)}^2---\left(5\right)$

Constraint condition: $8\stackrel{\&OverBar;}{z}{A}^2+8\stackrel{\&OverBar;}{x}\mathrm{AB}+8\stackrel{\&OverBar;}{y}\mathrm{AC}+4\mathrm{AD}+B^2+C^2=1---\left(6\right)$ Wherein, A ≠ 0, z _i=x _i ²+ y _i ², for x _i, y _i, z _iaverage, namely (x, y) represents [(M ' _ipq) _i, (M ' _ipq) _q] or [(M ' _qpq) _i, (M ' _qpq) _q], N is sampling number;

Adopt formula (7), formula (8), formula (9), formula (10), obtain solid part signal and the imaginary signals of the first complex signal and the second complex signal:

$b_I=-\frac{B}{2A}---\left(7\right)$

$b_Q=-\frac{C}{2A}---\left(8\right)$

${b_I}^{\&prime;}=-\frac{B^{\&prime;}}{2A^{\&prime;}}---\left(9\right)$

${b_Q}^{\&prime;}=-\frac{C^{\&prime;}}{2A^{\&prime;}}---\left(10\right)$

Wherein, b _i, b _qbe respectively solid part signal and the imaginary signals of the background signal of the first complex signal, A, B, C for will [(M ' _ipq) _i, (M ' _ipq) _q] substitute into formula (5), (6) gained fitting circle parameter; b _i', b _q' be respectively solid part signal and the imaginary signals of the background signal of the second complex signal, A', B', C' for will [(M ' _qpq) _i, (M ' _qpq) _q] substitute into formula (5), (6) gained fitting circle parameter.

Further, before the solid part signal that described employing curve fitting algorithm obtains the background signal of the first complex signal and the second complex signal respectively and imaginary signals, also comprise:

Adopt the undesired signal in least square median method described first complex signal of rejecting and the second complex signal.

Further, the solid part signal of the described background signal according to described first complex signal and imaginary signals, obtain the solid part signal of the background signal of described echoed signal, according to solid part signal and the imaginary signals of the background signal of described second complex signal, obtain the imaginary signals of described echoed signal background signal, comprising:

Calculate and obtain described b _iand b _qaverage, this average is the solid part signal of described background signal;

Calculate and obtain described b _i' and b _q' average, this average is the imaginary signals of described background signal.

Background signal extracting method in RCS measurement provided by the invention, by being separated the echoed signal of radar emission signal and reconstructing acquisition first complex signal and the second complex signal, curve fitting algorithm is adopted to obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively again, solid part signal and the imaginary signals of the background signal of echoed signal is obtained, to realize the extraction of the background signal of echoed signal according to the solid part signal of the background signal of the first complex signal and the second complex signal and imaginary signals.Apply method provided by the invention, adopt the complex signal of curve fitting algorithm process reconstruct to extract the background signal of degree of precision, solve the problem of the poor robustness in practical engineering application.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the background signal extracting method embodiment one in RCS measurement of the present invention;

The schematic diagram of Dihedral Corner Reflectors wheel measuring of Fig. 2 for adopting in embodiment of the method shown in Fig. 1;

Fig. 3 is the process flow diagram of the background signal extracting method embodiment two in RCS measurement of the present invention;

The M of Fig. 4 for emulating in embodiment of the method shown in Fig. 3 _hHthe track distribution plan of passage echoed signal;

The first complex signal track distribution plan on I/Q plane of Fig. 5 a for reconstructing in embodiment of the method shown in Fig. 3;

The second complex signal track distribution plan on I/Q plane of Fig. 5 b for reconstructing in embodiment of the method shown in Fig. 3;

Fig. 6 is averaging the probability of error distribution plan of method for the LMS-Hyper algorithm that adopts in embodiment of the method shown in Fig. 3 and prior art.

Embodiment

Fig. 1 is the process flow diagram of the background signal extracting method embodiment one in RCS measurement of the present invention, and as shown in Figure 1, the method comprises:

Step 101: the echoed signal measuring the radar emission signal obtaining Dihedral Corner Reflectors reflection.

In the situation of single station, radar emission and receiving antenna are in same position, and the echoed signal that receiving antenna receives can be expressed as formula (15):

M＝R(S+B+N)T＝RST+RBT+RNT （15）

Wherein M, R, S, B, N, T is the echoed signal of receiving antenna reception, the receiving matrix of radar system, target polarization matrix, background matrix, noise matrix and systems radiate matrix respectively.

Receiving matrix R can be written as formula (16)

$R=\left(\begin{array}{cc}{R}_{\mathrm{HH}}& R_{\mathrm{HV}}\\ R_{\mathrm{VH}}& R_{\mathrm{VV}}\end{array}\right)---\left(16\right)$

In the matrix element of formula (16), bottom right mark represents the polarised direction transmitted, and pre-sub represents the polarised direction of echoed signal, and H represents horizontal polarization directions, and V represents vertical polarization directions.To a reciprocity system, emission matrix T can be obtained by the transposition of receiving matrix R, namely has T=R ^t.

By formula (17) to R normalization,

$R=R_n\mathrm{\&epsiv;}=\left(\begin{array}{cc}{R}_{\mathrm{HH}}& 0\\ 0& R_{\mathrm{VV}}\end{array}\right)\left(\begin{array}{cc}1& {\mathrm{\&epsiv;}}_H\\ {\mathrm{\&epsiv;}}_V& 1\end{array}\right)---\left(17\right)$

Wherein, ε _hand ε _vfor normalized cross polar component, illustrate cross polarization error during target illumination.

According to above each formula, formula (15) can be written as:

M=R _nε(S+B+N)ε ^TR _n （18）

Formula (18) is expanded into matrix form, namely

$\left[\begin{array}{cc}{M}_{\mathrm{HH}}& M_{\mathrm{HV}}\\ M_{\mathrm{VH}}& M_{\mathrm{VV}}\end{array}\right]=\left[\begin{array}{cc}{R}_{\mathrm{HH}}& 0\\ 0& R_{\mathrm{VV}}\end{array}\right]\left[\begin{array}{cc}1& {\mathrm{\&epsiv;}}_H\\ {\mathrm{\&epsiv;}}_V& 1\end{array}\right]\left(\begin{array}{c}\left[\begin{array}{cc}{S}_{\mathrm{HH}}& S_{\mathrm{HV}}\\ S_{\mathrm{VH}}& S_{\mathrm{VV}}\end{array}\right]+\left[\begin{array}{cc}{B}_{\mathrm{HH}}& B_{\mathrm{HV}}\\ B_{\mathrm{VH}}& B_{\mathrm{VV}}\end{array}\right]+\left[\begin{array}{cc}{N}_{\mathrm{HH}}& N_{\mathrm{HV}}\\ N_{\mathrm{VH}}& N_{\mathrm{VV}}\end{array}\right]\end{array}\right)---\left(19\right)$

$\&CenterDot;\left[\begin{array}{cc}1& {\mathrm{\&epsiv;}}_V\\ {\mathrm{\&epsiv;}}_H& 1\end{array}\right]\left[\begin{array}{cc}{R}_{\mathrm{HH}}& 0\\ 0& R_{\mathrm{VV}}\end{array}\right]$

After merger is carried out to formula (19) be:

$\left[\begin{array}{cc}{M}_{\mathrm{HH}}& M_{\mathrm{HV}}\\ M_{\mathrm{VH}}& M_{\mathrm{VV}}\end{array}\right]=\left[\begin{array}{cc}{R}_{\mathrm{HH}}^2(S_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H^2{S}_{\mathrm{VV}})& R_{\mathrm{HH}}{R}_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V{S}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{S}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{VV}})\\ R_{\mathrm{HH}}{R}_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V{S}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{S}_{\mathrm{HV}}+S_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{VV}})& R_{\mathrm{VV}}^2({\mathrm{\&epsiv;}}_V^2{S}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_V{S}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_V{S}_{\mathrm{VH}}+S_{\mathrm{VV}})\end{array}\right]$

（20）

$+\left[\begin{array}{cc}{R}_{\mathrm{HH}}^2(B_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{B}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{B}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H^2{B}_{\mathrm{VV}})& R_{\mathrm{HH}}{R}_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V{B}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{B}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{B}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{B}_{\mathrm{VV}})\\ R_{\mathrm{HH}}{R}_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V{B}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{B}_{\mathrm{HV}}+B_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{B}_{\mathrm{VV}})& R_{\mathrm{VV}}^2({\mathrm{\&epsiv;}}_V^2{B}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_V{B}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_V{B}_{\mathrm{VH}}+B_{\mathrm{VV}})\end{array}\right]$

$+\left[\begin{array}{cc}{R}_{\mathrm{HH}}^2(N_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{N}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{N}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H^2{N}_{\mathrm{VV}})& R_{\mathrm{HH}}{R}_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V{N}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{N}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{N}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{N}_{\mathrm{VV}})\\ R_{\mathrm{HH}}{R}_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V{N}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{N}_{\mathrm{HV}}+N_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{N}_{\mathrm{VV}})& R_{\mathrm{VV}}^2({\mathrm{\&epsiv;}}_V^2{N}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_V{N}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_V{N}_{\mathrm{VH}}+N_{\mathrm{VV}})\end{array}\right]$

Referred to as formula (21):

$\left[\begin{array}{cc}{M}_{\mathrm{HH}}& M_{\mathrm{HV}}\\ M_{\mathrm{VH}}& M_{\mathrm{VV}}\end{array}\right]=\left[\begin{array}{cc}{A}_{\mathrm{HH}}(S_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H^2{S}_{\mathrm{VV}})& A_{\mathrm{HV}}({\mathrm{\&epsiv;}}_V{S}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{S}_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{VV}})\\ A_{\mathrm{VH}}({\mathrm{\&epsiv;}}_V{S}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_H{\mathrm{\&epsiv;}}_V{S}_{\mathrm{HV}}+S_{\mathrm{VH}}+{\mathrm{\&epsiv;}}_H{S}_{\mathrm{VV}})& A_{\mathrm{VV}}({\mathrm{\&epsiv;}}_V^2{S}_{\mathrm{HH}}+{\mathrm{\&epsiv;}}_V{S}_{\mathrm{HV}}+{\mathrm{\&epsiv;}}_V{S}_{\mathrm{VH}}+S_{\mathrm{VV}})\end{array}\right]---\left(21\right)$

$+\left[\begin{array}{cc}{A}_{\mathrm{HH}}{B}_1& A_{\mathrm{HV}}{B}_2\\ A_{\mathrm{VH}}{B}_3& A_{\mathrm{VV}}{B}_4\end{array}\right]+\left[\begin{array}{cc}{A}_{\mathrm{HH}}{N}_1& A_{\mathrm{HV}}{N}_2\\ A_{\mathrm{VH}}{N}_3& A_{\mathrm{VV}}{N}_4\end{array}\right]$

Wherein, R _hHt _hH=R _hH ²=A _hH, R _vVt _vV=R _vV ²=A _vV, R _hHt _vV=R _hHr _vV=A _hV, R _vVt _hH=R _vVr _hH=A _vH; B _ifor background signal, N _ifor noise, i=1,2,3,4.

The schematic diagram of Dihedral Corner Reflectors wheel measuring of Fig. 2 for adopting in embodiment of the method shown in Fig. 1, as shown in Figure 2, fixing x-axis, the broken line of the dihedral angle scattering matrix behind θ angle that turns clockwise in yOz plane is:

$S=\left[\begin{array}{cc}{S}_{\mathrm{HH}}& S_{\mathrm{HV}}\\ S_{\mathrm{VH}}& S_{\mathrm{VV}}\end{array}\right]=k_D\left[\begin{array}{cc}-\cos 2\mathrm{\&theta;}& \sin 2\mathrm{\&theta;}\\ \sin 2\mathrm{\&theta;}& \cos 2\mathrm{\&theta;}\end{array}\right]---\left(22\right)$

Wherein, under high frequency condition, a, b are the wide and high of dihedral angle respectively, and λ is radar wavelength.

Formula is substituted into formula (22), can obtain

$M_{\mathrm{HH}}=A_{\mathrm{HH}}[k_D\&CenterDot;(-1+{\mathrm{\&epsiv;}}_H^2)\cos 2\mathrm{\&theta;}+k_D\&CenterDot;{2\mathrm{\&epsiv;}}_H\sin 2\mathrm{\&theta;}+B_1+N_1]---\left(23\right)$

$M_{\mathrm{VV}}=A_{\mathrm{VV}}[k_D\&CenterDot;(1+{\mathrm{\&epsiv;}}_V^2)\cos 2\mathrm{\&theta;}+k_D\&CenterDot;{2\mathrm{\&epsiv;}}_V\sin 2\mathrm{\&theta;}+B_2+N_2]---\left(24\right)$

M _HV=A _HV[k _D·(ε _H-ε _V)cos2θ+k _D·(1+ε _Hε _V)sin2θ+B ₃+N ₃] （25）

M _VH=A _HV[k _D·(ε _H-ε _V)cos2θ+k _D·(1+ε _Hε _V)sin2θ+B ₄+N ₄] （26）

To all polarization combination, echoed signal is expressed as the function of rotation angle θ, can write out general formula is formula (11):

M _pq(θ)=c _pqcos2θ+s _pqsin2θ+B _i+N _i （11）

Wherein, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives.B is background signal, and N is noise signal, i=1,2,3,4, c _pqand s _pqby the wavelength of wide and high, the radar emission signal of Dihedral Corner Reflectors, and normalized cross polar component ε _h, ε _vdetermine, be the relevant physical quantity such as same Dihedral Corner Reflectors physical dimension, measuring system parameter, without the need to obtaining in background extracting, different polarization modes has different computing method.θ is the angle that the broken line of dihedral angle turns clockwise in yOz plane.

The echoed signal M of the radar emission signal of the Dihedral Corner Reflectors reflection obtained is measured by formula (11) _pq(θ).

Step 102: echoed signal is separated into homophase channel signal and orthogonal channel signal.

Echoed signal M _pq(θ) I, Q two paths is had.Wherein, I passage is homophase passage, and Q passage is orthogonal channel.

For example, can according to formula (11) by echoed signal M _pq(θ) signal of I, Q two paths is separated into:

M _Ipq(θ)＝c _Ipqcos2θ+s _Ipqsin2θ+B _Ii+N _Ii （1）

M _Qpq(θ)=c _Qpqcos2θ+s _Qpqsin2θ+B _Qi+N _Qi （2）

Formula (1) can be adopted to obtain echoed signal M _pq(θ) homophase channel signal M _ipq(θ); Formula (2) can be adopted to obtain echoed signal M _pq(θ) orthogonal channel signal M _qpq(θ).

Step 103: adopt homophase channel signal reconstruct acquisition first complex signal, adopt orthogonal channel signal reconstruction to obtain the second complex signal.

Specifically, for the measurement data by anglec of rotation uniform sampling, according to anglec of rotation θ _k, k=1,2 ... the order of K, if hypothesis θ ₁=0, θ _k=2n π, then because according to formula (11), measurement data M _pq(θ) θ in _kthe signal of <n π+3 π/4 with from θ _kbetween the signal of>=n π+3 π/4, its phase place just in time differs pi/2, therefore by homophase channel signal M _ipq(θ) θ in _kthe signal of <n π+3 π/4 as the solid part signal of the first complex signal, by homophase channel signal M _ipq(θ) θ in _kthe signal of>=n π+3 π/4, as the imaginary signals of the first complex signal, makes the track of the first complex signal become a circle or one section of circular arc; By orthogonal channel signal M _qpq(θ) θ in _kthe signal of <n π+3 π/4 as the solid part signal of described second complex signal, by orthogonal channel signal M _qpq(θ) θ in _kthe signal of>=n π+3 π/4, as the imaginary signals of described second complex signal, makes the track of the second complex signal become a circle or one section of circular arc.

By the way, formula (27) i.e. restructural is adopted to obtain the first complex signal:

M′ _Ipq(θ)=c _Ipqcos2θ+s _Ipqsin2θ+B _Ii+N _Ii+j(c _Ipqsin2θ+s _Ipqcos2θ+B _Ii+N _Ii)（27）

After formula (27) is simplified be:

${{M_I}^{\&prime;}}_{\mathrm{pq}}\left(\mathrm{\&theta;}\right)=c_{\mathrm{Ipq}}{e}^{j2\mathrm{\&theta;}}+{s_{\mathrm{Ipq}}e}^{j(\frac{\mathrm{\&pi;}}{2}-2\mathrm{\&theta;})}+\sqrt{2}(B_{\mathrm{Ii}}+N_{\mathrm{Ii}})e^{j\frac{\mathrm{\&pi;}}{4}}---\left(3\right)$

Therefore, the present embodiment can adopt formula (3) to reconstruct acquisition first complex signal.

In like manner, similar formula (4) reconstruct acquisition second complex signal can be adopted:

${{M_Q}^{\&prime;}}_{\mathrm{pq}}\left(\mathrm{\&theta;}\right)=c_{\mathrm{Qpq}}{e}^{j2\mathrm{\&theta;}}+{s_{\mathrm{Qpq}}e}^{j(\frac{\mathrm{\&pi;}}{2}-2\mathrm{\&theta;})}+\sqrt{2}(B_{\mathrm{Qi}}+N_{\mathrm{Qi}})e^{j\frac{\mathrm{\&pi;}}{4}}---\left(4\right)$

According to formula (3) and (4) known, the right Section 3 can not be approximately complex constant with rotation angle θ change, therefore the first complex signal reconstructed and the track of the second complex signal in I/Q plane are primarily of front two formations, although all have two e index items in the first complex signal of reconstruct and the second complex signal, due to parameter ε in actual polarization radar measurement _hand ε _vthe polarization isolation parameter of representative system, its isolation is generally all better than 20dB, is also ε _hand ε _vall be less than 0.1.Therefore, according to formula (23) ~ (26), for each polarization combination, c _pqand s _pqin only have one-component to occupy an leading position, another component is Comparatively speaking very little, and the first complex signal reconstructed thus and the second complex signal track are on a complex plane essentially a circle.Therefore, the real part of the first complex signal of reconstruct and the background signal of the second complex signal and the amplitude of imaginary part should be equal.Due to the impact of the reasons such as noise in reality, both amplitude slightly difference.

Step 104: adopt curve fitting algorithm to obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively;

Preferably, the multiprecision arithmetic (Hyper) that the present embodiment adopts Chernov etc. to propose obtains solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively.

The fitting criterion of Hyper algorithm is:

$\min \underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{({\mathrm{Ax}}_i^2+{\mathrm{Ay}}_i^2+{\mathrm{Bx}}_i+{\mathrm{Cy}}_i+D)}^2---\left(5\right)$

Constraint condition is:

$8\stackrel{\&OverBar;}{z}{A}^2+8\stackrel{\&OverBar;}{x}\mathrm{AB}+8\stackrel{\&OverBar;}{y}\mathrm{AC}+4\mathrm{AD}+B^2+C^2=1---\left(6\right)$

Wherein, A, B, C, D are fitting circle parameter, A ≠ 0.Z _i=x _i ²+ y _i ², for x _i, y _i, z _iaverage, namely $\stackrel{\&OverBar;}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{x}_i,$ $\stackrel{\&OverBar;}{y}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{y}_i,$ $\stackrel{\&OverBar;}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{z}_i=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}(x_i^2+y_i^2),$ N is sampling number.

First complex signal and the second complex signal are separated into respectively I/Q two paths complex signal [(M ' _ipq) _i, (M ' _ipq) _q] and [(M ' _qpq) _i, (M ' _qpq) _q].For stating conveniently, represent with (x, y) [(M ' _ipq) _i, (M ' _ipq) _q] or [(M ' _qpq) _i, (M ' _qpq) _q].

(x, y) is substituted into formula (5) and formula (6), Optimization Solution, draw [(M ' _ipq) _i, (M ' _ipq) _q] or [(M ' _qpq) _i, (M ' _qpq) _q] after respectively corresponding A, B, C, D fitting circle parameter, by adopting formula (7), formula (8), formula (9), formula (10), obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal:

$b_I=-\frac{B}{2A}---\left(7\right)$

$b_Q=-\frac{C}{2A}---\left(8\right)$

${b_I}^{\&prime;}=-\frac{B^{\&prime;}}{2A^{\&prime;}}---\left(9\right)$

${b_Q}^{\&prime;}=-\frac{C^{\&prime;}}{2A^{\&prime;}}---\left(10\right)$

Wherein, b _i, b _qbe respectively solid part signal and the imaginary signals of the background signal of the first complex signal, A, B, C for will [(M ' _ipq) _i, (M ' _ipq) _q] substitute into formula (5), (6) gained fitting circle parameter; b _i', b _q' be respectively solid part signal and the imaginary signals of the background signal of the second complex signal, A', B', C' for will [(M ' _qpq) _i, (M ' _qpq) _q] substitute into formula (5), (6) gained fitting circle parameter.

Step 105: according to solid part signal and the imaginary signals of the background signal of the first complex signal, obtain the solid part signal of the background signal of echoed signal, according to solid part signal and the imaginary signals of the background signal of the second complex signal, obtain the imaginary signals of the background signal of echoed signal.

To the solid part signal b of the background signal of the first complex signal _iwith imaginary signals b _qsummation is averaged, and this average is echoed signal M _pq(θ) solid part signal of background signal;

To the solid part signal b of the background signal of the second complex signal _i' and imaginary signals b _q' sue for peace and average, this average is echoed signal M _pq(θ) imaginary signals of background signal.

Background signal extracting method in the RCS measurement that the present embodiment provides, by being separated the echoed signal of radar emission signal and reconstructing acquisition first complex signal and the second complex signal, curve fitting algorithm is adopted to obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively again, solid part signal and the imaginary signals of the background signal of echoed signal is obtained, to realize the extraction of the background signal of echoed signal according to the solid part signal of the background signal of the first complex signal and the second complex signal and imaginary signals.The method that application the present embodiment provides, adopts the complex signal of curve fitting algorithm process reconstruct to extract the background signal of degree of precision, solves the problem of the poor robustness in practical engineering application.

Fig. 3 is the process flow diagram of the background signal extracting method embodiment two in RCS measurement of the present invention, and as shown in Figure 3, the method comprises:

Step 301: the echoed signal measuring the radar emission signal obtaining Dihedral Corner Reflectors reflection;

Can see the step 101 in embodiment one to the description of step 301.

Step 302: echoed signal is separated into homophase channel signal and orthogonal channel signal.

Can see the step 102 in embodiment one to the description of step 302.

Step 303: adopt homophase channel signal reconstruct acquisition first complex signal, adopt orthogonal channel signal reconstruction to obtain the second complex signal.

Can see the step 103 in embodiment one to the description of step 303.

Step 304: employing least square median method rejects the undesired signal in the first complex signal and the second complex signal.

The present embodiment before the solid part signal extracting background signal and imaginary signals, can first carry out the rejecting of undesired signal.

Specifically, the first complex signal and the second complex signal can be separated into the complex signal [(M' of I/Q two paths by the present embodiment _ipq) _i, (M ' _ipq) _q] and [(M ' _qpq) _i, (M ' _qpq) _q].For stating conveniently, represent [(M' with (x, y) _ipq) _i, (M' _ipq) _q] or [(M ' _qpq) _i, (M ' _qpq) _q].

Being expressed as of round parametric equation:

$\begin{array}{cc}\tilde{A}{x}^2+\tilde{A}{y}^2+\tilde{B}x+\tilde{C}y+\tilde{D}=0& \tilde{A}\&NotEqual;0\end{array}---\left(28\right)$

Wherein, for the required fitting circle parameter obtained.

By formula (28), the fitting criterion of least square median method (LMS) is:

$\underset{\tilde{A},\tilde{B},\tilde{C}}{\min }\mathrm{median}\left\{d_i^2\right\}=\underset{\tilde{A},\tilde{B},\tilde{C}}{\min }\mathrm{median}\{{(\tilde{A}{x}_i^2+\tilde{A}{y}_i^t+\tilde{B}{x}_i+\tilde{C}{y}_i+1)}^2,i=1,.....,N\}---\left(29\right)$

Wherein, (x _i, y _i) represent i-th Received signal strength measured and obtain, d _ibe the residual error of i-th Received signal strength, it is conveniently solve that fitting circle parameter D is set to 1 at this.Compared to least square method (LSS), the summation in LSS becomes by minimum intermediate value square (LMS) asks intermediate value.

(x, y) is substituted into formula (29) and obtain fitting circle parameter then absolute residuals formula (30) is used:

${\mathrm{rd}}_i=\frac{|\tilde{A}{x}_i^2+\tilde{A}{y}_i^2+\tilde{B}{x}_i+\tilde{C}{y}_i+1|}{\widehat{\mathrm{\&sigma;}}}---\left(30\right)$

Pass judgment on data, retain the point of absolute residuals within threshold value, reject the point of absolute residuals beyond threshold value, wherein threshold value needs setting in advance, is generally between 2 ~ 3.In formula that the data standard difference constructed for LMS matching is estimated.Can find out that LMS method only can reject undesired signal from raw data by analyzing above, and other signal can not be changed.

Step 305: adopt curve fitting algorithm to obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively.

Can see the step 104 in embodiment one to the description of step 305.Wherein, this first complex signal and the second complex signal are reject the first complex signal after undesired signal and the second complex signal.

Step 306: according to solid part signal and the imaginary signals of the background signal of the first complex signal, obtain the solid part signal of the background signal of echoed signal, according to solid part signal and the imaginary signals of the background signal of the second complex signal, obtain the imaginary signals of the background signal of echoed signal.

Can see the step 105 in embodiment one to the description of step 306.

Below that minimum intermediate value square-high precision (LMS-Hyper) algorithm of application the present embodiment and prior art are averaging method and carry out a simple comparison.

Because the rule of each POLARIZATION CHANNEL is substantially consistent, exemplarily, select HH polarization as research object here.Fundamental simulation parameter is as follows: radar is 100m to the distance of Dihedral Corner Reflectors, Dihedral Corner Reflectors is 0.299m to the distance of support, radar wavelength is 0.03m, the wide of metal Dihedral Corner Reflectors is 0.15m, long is 0.3m, signal to noise ratio SCR is 20dB, and signal to noise ratio snr is 40dB, normalized cross polar component ε _hand ε _vbe respectively 0.12,0.13, Dihedral Corner Reflectors turns over 360 °, and sampling interval is 0.2 °.A _pqbe a complex constant, its amplitude is determined by factors such as antenna gains, and its phase place is determined by the Distance geometry phase delay of radar and target.In order to study conveniently, by A _pqamplitude is set to 1, A _pqphase place is determined by many factors, and discuss for simplifying and under the prerequisite not affecting emulation, phase place can be established to be determined by radar range-to-go, the phase place of background signal will increase the distance of a target to support in addition.

According to above simulation parameter and formula (23), to M _hHchannel signal emulates.The M of Fig. 4 for emulating in embodiment of the method shown in Fig. 3 _hHthe track distribution plan of passage echoed signal, as shown in Figure 4, the exceptional data point wherein in hypothesis measurement accounts for 5%, and target and background signal is all without fluctuating.Can find out, the track of measurement data in I/Q plane is straight line.The first complex signal and the second complex signal is reconstructed according to formula (3) and formula (4), the first complex signal track distribution plan on I/Q plane of Fig. 5 a for reconstructing in embodiment of the method shown in Fig. 3, the second complex signal track distribution plan on I/Q plane of Fig. 5 b for reconstructing in embodiment of the method shown in Fig. 3, as shown in figure 5 a and 5b, first complex signal and the track of the second complex signal of reconstruct are all circles, and this is that the application of LMS-Hyper algorithm is laid a good foundation.

Next simply compare the background signal extractability of the method for being averaging and LMS-Hyper, can whether main these two kinds of algorithms of research have robustness, and accurately extract background signal when background signal and echo signal have fluctuating.

For weighing the precision that background signal extracts, definition background signal extracts error:

$\mathrm{\&Delta;}{e}_{\mathrm{BI}}=\left|{10\log }_{10}\frac{{{\hat{b}}_I}^2}{{b_I}^2}\right|---\left(31\right)$

$\mathrm{\&Delta;}{e}_{\mathrm{BQ}}=\left|{10\log }_{10}\frac{{{\hat{b}}_Q}^2}{{b_Q}^2}\right|---\left(32\right)$

Wherein, the LMS-Hyper method provided by being averaging method or the present embodiment obtains, (b _i, b _q) represent true value.

Draw Δ e _bI, Δ e _bQprobability distribution graph.Fig. 6 is averaging the probability of error distribution plan of method for the LMS-Hyper algorithm that adopts in embodiment of the method shown in Fig. 3 and prior art.Wherein, the longitudinal axis represents that background signal extracts error delta e _bI, Δ e _bQall be less than the probability of 0.5dB.

According to formula (31) and formula (32) known, if the value recorded and true value more close, Δ e _bI, Δ e _bQlower, as Δ e _bI, Δ e _bQall be less than 0.5dB, just think that precision is very high.Because each result is all not quite alike, in order to without loss of generality, so carry out several test to see its average case.If in test for several times, if there is the Δ e of m time _bI, Δ e _bQall be less than 0.5dB, that Δ e _bI<0.5 and Δ e _bQthe probability of <0.5 is just m/150, if probability very high (such as probability is 1) illustrates that the precision of algorithm is very high.When probability reduces, illustrate that precision reduces.

As can be seen here, background signal extracting method in the RCS measurement that the present embodiment provides, by being separated the echoed signal of radar emission signal and reconstructing acquisition first complex signal and the second complex signal, least square median method is adopted the undesired signal in the first complex signal and the second complex signal to be rejected, curve fitting algorithm is adopted to obtain solid part signal and the imaginary signals of the background signal of the first complex signal after rejecting undesired signal and the second complex signal respectively again, solid part signal and the imaginary signals of the background signal of echoed signal is obtained according to the solid part signal of background signal and imaginary signals of rejecting the first complex signal after undesired signal and the second complex signal, the extraction of the background signal of echoed signal is realized with this.The method that application the present embodiment provides, when there is undesired signal, adopting LMS-Hyper algorithm undesired signal can be rejected, and extracting background signal with higher precision, solving the problem of the poor robustness in practical engineering application.

Last it is noted that above each embodiment is only in order to illustrate technical scheme of the present invention, be not intended to limit; Although with reference to foregoing embodiments to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein some or all of technical characteristic; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.

Claims (3)

1. the background signal extracting method in RCS measurement, is characterized in that, comprising:

Measure the echoed signal of the radar emission signal obtaining Dihedral Corner Reflectors reflection;

Described echoed signal is separated into homophase channel signal and orthogonal channel signal;

Adopt described homophase channel signal to reconstruct acquisition first complex signal, adopt described orthogonal channel signal reconstruction to obtain the second complex signal;

Curve fitting algorithm is adopted to obtain solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively;

According to solid part signal and the imaginary signals of the background signal of described first complex signal, obtain the solid part signal of the background signal of described echoed signal, according to solid part signal and the imaginary signals of the background signal of described second complex signal, obtain the imaginary signals of the background signal of described echoed signal;

Described described echoed signal is separated into homophase channel signal and orthogonal channel signal, comprises:

Formula (1) is adopted to obtain described homophase channel signal this signal can be expressed as:

$M_{I_{\mathrm{pq}}}\left(\mathrm{\&theta;}\right)=c_{I_{\mathrm{pq}}}\cos 2\mathrm{\&theta;}+s_{I_{\mathrm{pq}}}\sin 2\mathrm{\&theta;}+B_{I_i}+N_{I_i}---\left(1\right)$

Formula (2) is adopted to obtain described orthogonal channel signal , this signal can be expressed as:

$M_{Q_{\mathrm{pq}}}\left(\mathrm{\&theta;}\right)=c_{Q_{\mathrm{pq}}}\cos 2\mathrm{\&theta;}+s_{Q_{\mathrm{pq}}}\sin 2\mathrm{\&theta;}+B_{Q_i}+N_{Q_i};---\left(2\right)$

Wherein, I is homophase passage, and Q is orthogonal channel, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives, with by the wavelength of wide and high, the radar emission signal of Dihedral Corner Reflectors, and normalized cross polar component ε _h, ε _vdetermine, without the need to obtaining in background extracting, B _ifor the background signal of homophase channel signal, B _qfor the background signal of orthogonal channel signal, N _ifor the noise of homophase channel signal, N _qfor the noise of orthogonal channel signal, i=1,2,3,4, θ is the angle that the broken line of dihedral angle turns clockwise in yOz plane;

Described employing described homophase channel signal reconstruct acquisition first complex signal, comprising:

By described homophase channel signal the signal of middle θ < n π+3 π/4 as the solid part signal of described first complex signal, by described homophase channel signal the signal of middle θ>=n π+3 π/4 is as the imaginary signals of described first complex signal;

The described orthogonal channel signal reconstruction of described employing obtains the second complex signal, comprising:

By described orthogonal channel signal the signal of middle θ < n π+3 π/4 as the solid part signal of described second complex signal, by described orthogonal channel signal the signal of middle θ>=n π+3 π/4 is as the imaginary signals of described second complex signal;

Described by described homophase channel signal the signal of middle θ < n π+3 π/4 as the solid part signal of described first complex signal, by described homophase channel signal the signal of middle θ>=n π+3 π/4, as the imaginary signals of described first complex signal, comprising:

Formula (3) reconstruct is adopted to obtain described first complex signal:

$M_{I^{{\&prime;}_{\mathrm{pq}}}}\left(\mathrm{\&theta;}\right)=c_{I_{\mathrm{pq}}}{e}^{j2\mathrm{\&theta;}}+s_{I_{\mathrm{pq}}}{e}^{j(\frac{\mathrm{\&pi;}}{2}-2\mathrm{\&theta;})}+\sqrt{2}(B_{I_i}+N_{I_i})e^{j\frac{\mathrm{\&pi;}}{4}}---\left(3\right)$

Wherein, I is homophase passage, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives, B _ifor the background signal of homophase channel signal, N _ifor the noise of homophase channel signal, i=1,2,3,4;

Described by described orthogonal channel signal the signal of middle θ <n π+3 π/4 as the solid part signal of described second complex signal, by described orthogonal channel signal the signal of middle θ>=n π+3 π/4, as the imaginary signals of described second complex signal, comprising:

Formula (4) reconstruct is adopted to obtain described second complex signal:

$M_{Q^{{\&prime;}_{\mathrm{pq}}}}\left(\mathrm{\&theta;}\right)=c_{Q_{\mathrm{pq}}}{e}^{j2\mathrm{\&theta;}}+s_{Q_{\mathrm{pq}}}{e}^{j(\frac{\mathrm{\&pi;}}{2}-2\mathrm{\&theta;})}+\sqrt{2}(B_{Q_i}+N_{Q_i})e^{j\frac{\mathrm{\&pi;}}{4}}---\left(4\right)$

Wherein, Q is orthogonal channel, p and q refers to horizontal polarization directions or vertical polarization directions respectively, and q represents the polarised direction of radar emission signal, and p represents the polarised direction of the echoed signal that receiving antenna receives, B _qfor the background signal of orthogonal channel signal, N _qfor the noise of orthogonal channel signal, i=1,2,3,4;

Described employing curve fitting algorithm obtains solid part signal and the imaginary signals of the background signal of the first complex signal and the second complex signal respectively, comprising:

Formula (5) and formula (6) is adopted to obtain fitting circle parameter A, B, C, D:

$\min \underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{(A{x_i}^2+A{y_i}^2+B{x}_i+C{y}_i+D)}^2---\left(5\right)$

Constraint condition: $8\stackrel{\&OverBar;}{z}{A}^2+8\stackrel{\&OverBar;}{x}\mathrm{AB}+8\stackrel{\&OverBar;}{y}\mathrm{AC}+4\mathrm{AD}+B^2+C^2=1---\left(6\right)$

Wherein, A ≠ 0, z _i=x _i ²+ y _i ², for x _i, y _i, z _iaverage, namely $\stackrel{\&OverBar;}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{x}_i,\stackrel{\&OverBar;}{y}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{y}_i,$ $\stackrel{\&OverBar;}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{z}_i=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}({x_i}^2+{y_i}^2),$ (x, y) represents [(M _i' _pq) _i, (M _i' _pq) _q] or [(M _q' _pq) _i, (M _q' _pq) _q], N is sampling number;

Adopt formula (7), formula (8), formula (9), formula (10), obtain solid part signal and the imaginary signals of the first complex signal and the second complex signal:

$b_I=-\frac{B}{2A}---\left(7\right)$

$b_Q=-\frac{C}{2A}---\left(8\right)$

${b_I}^{\&prime;}=-\frac{B\&prime;}{2A\&prime;}---\left(9\right)$

${b_Q}^{\&prime;}=-\frac{C\&prime;}{2A\&prime;}---\left(10\right)$

Wherein, b _i, b _qbe respectively solid part signal and the imaginary signals of the background signal of the first complex signal, A, B, C are by [(M _i' _pq) _i, (M _i' _pq) _q] substitute into formula (5), (6) gained fitting circle parameter; b _i', b _q' being respectively solid part signal and the imaginary signals of the background signal of the second complex signal, A', B', C' are for by [(M _q' _pq) _i, (M _q' _pq) _q] substitute into formula (5), (6) gained fitting circle parameter.

2. method according to claim 1, is characterized in that, before the solid part signal that described employing curve fitting algorithm obtains the background signal of the first complex signal and the second complex signal respectively and imaginary signals, also comprises:

Adopt the undesired signal in least square median method described first complex signal of rejecting and the second complex signal.

3. method according to claim 1, it is characterized in that, the solid part signal of the described background signal according to described first complex signal and imaginary signals, obtain the solid part signal of the background signal of described echoed signal, according to solid part signal and the imaginary signals of the background signal of described second complex signal, obtain the imaginary signals of described echoed signal background signal, comprising:

Calculate and obtain described b _iand b _qaverage, this average is the solid part signal of described background signal;

Calculate and obtain described b _i' and b _q' average, this average is the imaginary signals of described background signal.