CN103973347A - Closed loop zero-setting shape-preserving method of satellite communication antenna - Google Patents

Abstract

The invention discloses a closed loop zero-setting shape-preserving method of a satellite communication antenna, and aims at mainly solving the problems that the zero-setting time is long and the communication range loss is large in the prior art. The method comprises the following steps: 1, processing signals received through the antenna in two ways, wherein one way of signals is subjected to beam formation to change into combined signals, and the other way of signals is subjected to down-conversion to change into middle-frequency signals; 2, selecting a middle-frequency signal of a branch channel to be correlated with the combined signals after the down-conversion, and outputting three ways of data, namely, branch self-correlation, branch and combined way inter-correlation and combined way self-correlation; 3, sequentially calculating the self-adaptation step length and the weights of the branches according to the three ways of data; 4, sequentially accomplishing the traverse of all the branch channels, and performing the beam formation on the obtained weights of all branches and the signals received through the antenna; and 5, judging whether the zero-setting formation process needs to be repeated according to output combined way self-correlation data of the beam formation. By adopting the method, the zero-setting time is short, the communication range loss is small, and the method can be applied to the field of practical satellite communication.

Description

The closed loop zeroing conformal method of satellite communication antena

Technical field

The invention belongs to communication technical field, a kind of zeroing of the closed loop for satellite communication antena conformal method, can be used for disturbing and returning to zero and communication zone conformal in satellite communication specifically.

Background technology

In traffic guidance, satellite communication system has important effect, and the wave beam that it can be realized flexibly, optimize in certain regional extent covers, and ensures that the link performance of all fixing or mobile terminals in this geographic area is in optimum state.But because the relative ground, position of communication satellite is fixed, therefore, it is easy to be subject to ground, aircraft or low orbit satellite interference effect, is especially easily subject to the artificial electromagnetic interference of having a mind to from ground.In addition, along with the electromagnetic environment of communication satellite self becomes increasingly complex, the signal transmission of communication satellite is easily subject to interim or reversible destruction, thereby causes the quality of communication sharply to worsen.Therefore, to communication satellite carry out that adaptive nulling suppresses to disturb and wave beam conformal to come optimize communicate overlay area be a requisite key technology that ensures satellite proper communication, corresponding, the adaptive nulling antenna returning to zero for satellite communication produces thereupon.

Adaptive nulling antenna, is in fact made up of aerial array and a multi-channel adaptive processor of one group of space configuration.It utilizes the variation of array aerial direction figure, for the environment self-adaption changing adjust the wave beam position at zero point, make it to aim at interference signal come to, and can realize offseting of interference signal by reducing the sidelobe level of antenna beam, ensure that the output of main beam is all the time in optimum state simultaneously.

Existing adaptive nulling antenna, can roughly be divided into closed loop system and the large class of open loop system two.Wherein:

Open loop system is to disturb in two steps inhibition: first estimate interference source come to, then disturb zeroing.The method of estimation at interference source direction of arrival angle is a lot, to single interference, can adopt the methods such as pulse angle measurement, interferometer; To multiple interference, can adopt multiple signal classification method MUSIC, the ultra-resolution method such as Parameter Estimation Method ESPRIT and maximum likelihood method ML of ESPRIT.Use various zeroing technology for the interference angle estimating, as adaptive antenna sidelobe cancellation technology, beam nulling technology, linear restriction minimum variance LCMV beam-forming technology etc., can suppress in theory strong jamming, and narrower at the zeroing recess of interference position in the accurate situation of interference angle estimation, but owing to there being various errors in reality, disturb angle estimation to have certain deviation, can have a strong impact on adjusting zero effect, and system realize to hardware device require highly, hardware complexity is very big.

Closed loop system, is to adopt the adaptive iteration algorithms such as power inversion, lowest mean square LMS, recursive least-squares RLS to disturb zeroing, do not need interference source to etc. priori, realization simply, to hardware device require low.But while using closed loop system to disturb zeroing, due to do not need interference source to etc. priori, therefore wider at the zeroing recess of interference position formation, can greatly reduce the coverage of communication zone, affect communication quality.

Summary of the invention

The object of the invention is to overcome the deficiency of prior art, be the contradiction of multi-beam zeroing performance and communication zone coverage rate in closed loop system, a kind of zeroing of the closed loop for satellite communication antena conformal method is proposed, to reach to interference signal fast zero-setting and to main beam pattern conformal, on the basis that ensures to disturb the zeroing degree of depth, coordinate the convergence rate of zeroing and the coverage rate of communication zone, improve communication quality.

Technical scheme of the present invention is: consider feasibility and the complexity of system, use closed loop system to realize interference zeroing and the communication zone conformal of satellite communication.Implementation step is as follows:

(1) the radiofrequency signal X of N antenna reception a being arrived bypass passage _e(t) divide two-way processing: the weights W that a road and Beam-former are pre-stored _smultiply each other and obtain beamformer output signal Y _e(t); Separately lead up to analog down, make the radiofrequency signal X of N bypass passage _e(t) become the analog if signal X (t) of N bypass passage;

(2) utilize beam selection device to select the analog if signal X of i bypass passage _i(t), to the branch road analog if signal X selecting _i(t) to the beamformer output signal y (t) after down-conversion carry out relevant, output three circuit-switched data: branch road autocorrelation is according to R _xx(i, n), branch road Yu He road cross-correlation data γ _i(n), close road autocorrelation according to R _yy(i, n), i=1,2 ..., N, N is total bypass passage number, n=1,2 ..., M, M is the total degree that whole zeroing conformal process is carried out;

(3) by branch road autocorrelation according to R _xx(i, n) and the zeroing thresholding u setting in advance compare: if R _xx(i, n) < u, shows that i bypass passage do not need zeroing, returns to step (2), if R _xx(i, n)>=u, execution step (4);

(4) adaptive step of calculating branch road: to beamformer output signal Y _e(t) carry out successively down-conversion and analog-to-digital conversion, obtain numeral and close circuit-switched data y (n); The autocorrelation that numeral is closed to circuit-switched data y (n) and branch road is according to R _xx(i, n) combines, and obtains the adaptive step μ of branch road _i(n);

(5) utilize following formula to calculate the weight w of branch road _i(n+1);

$w_i(n+1)=\left\{\begin{array}{cc}{W}_s\left(i\right)-{\mathrm{\&mu;}}_i\left(n\right)({\mathrm{\&gamma;}}_i\left(n\right)+{\mathrm{\&beta;}}_i\left(n\right))& n=1\\ w_i\left(n\right)+\frac{\left|y\left(n\right)y(n-1)\right|}{\mathrm{\&alpha;}{\mathrm{\&lambda;}}_{\max }}& n>1\end{array}\right.$

Wherein, i=1,2 ..., N, N is total bypass passage number, n=1,2 ..., M, M is the total degree that whole zeroing conformal process is carried out; w _i(n+1) be the weights of i branch road of the n+1 time zeroing, w _i(n) be the weights of i branch road of the n time zeroing, w _i(n-1) be the weights of i branch road of the n-1 time zeroing, W _s=[W _s(1) ..., W _s(i) ..., W _s(N)] represent the pre-stored weights of Beam-former, W _s(i) be the pre-stored weights of i bypass passage, y (n) is that the numeral of the n time zeroing is closed circuit-switched data, and y (n-1) is that the numeral of the n-1 time zeroing is closed circuit-switched data, γ _i(n) be the cross-correlation data on branch road Yu He road, μ _i(n) be the adaptive step of branch road, α is a variable, and the span of α is 0 < α < 0.5, λ _maxthe autocorrelation matrix R of analog if signal X (t) _n=E[X (t) X ^*(t) eigenvalue of maximum], E[] expression average, () ^*represent conjugation, β _i(n) be zeroing conformal parameter.

(6) repeating step (2)～(5), complete after the traversal of N bypass passage successively, obtain the weight w of N branch road _i(n), use this weight w _i(n) be multiplied by the radiofrequency signal X of antenna reception _e(t), obtain new beamformer output signal Y _e(t), i=1,2 ..., N, N is total bypass passage number;

(7) to new beamformer output signal Y _e(t) carry out down-conversion, obtain down-conversion signal y _p(t), by this down-conversion signal y _p(t) carry out auto-correlation, obtain Xin He road autocorrelation according to R _p(i, n), Jiang Xinhe road autocorrelation is according to R _pthe judgement threshold value v that (i, n) and whole antenna zero-setting system are given compares, if R _p(i, n)>=v, illustrates that zeroing conformal is unsuccessful, repeating step (2)～(6); If R _p(i, n) < v, explanation zeroing conformal success, finishes zeroing conformal process.

The present invention compared with prior art has the following advantages:

1) the present invention is because the autocorrelation certificate and the numeral that have added antenna branch passage in the calculating of branch road adaptive step are closed circuit-switched data, the step-length of branch road can be changed along with antenna environment self-adaption of living in, improve the convergence rate of zeroing, shorten the zeroing time, improved real-time performance and the tracking performance of zero-setting system;

2) the present invention adds zeroing conformal parameter in branch road weights are calculated, and has realized the conformal to main beam;

3) the present invention is by the adaptivity of comprehensive branch road step-length and the zeroing conformality of branch road weights, coordinate convergence rate and this pair of contradictory relation of communication zone coverage rate of zeroing, under the condition that ensures good zeroing convergence rate and the zeroing degree of depth, improved communication zone coverage rate.

Brief description of the drawings

Fig. 1 is the block diagram of realizing of the present invention;

Fig. 2 is the cloth system of battle formations of the main beam of antenna in the present invention;

Fig. 3 is the front main beam of zeroing overlay area in the present invention;

Fig. 4 is main beam contour map after the single interference zeroing in emulation of the present invention;

Fig. 5 is that two in emulation of the present invention disturb the rear main beam contour map of zeroing;

Fig. 6 is main beam contour map after three near-by interference zeroings in emulation of the present invention;

Fig. 7 is the rear main beam contour maps of zeroing that disturb far away of three in emulation of the present invention;

Fig. 8 is three the rear main beam contour maps of zeroing that disturb far away that are positioned at beam center in emulation of the present invention;

Fig. 9 is the single interference iteration convergence curve chart in emulation of the present invention;

Figure 10 is that two in emulation of the present invention disturb iteration convergence curve chart;

Figure 11 is that three in emulation of the present invention disturb iteration convergence curve chart.

Embodiment

With reference to Fig. 2, the main beam of antenna of the present invention is structured the formation and is made up of 7 wave beams, respectively as shown in Figure 3, the interference that needs zeroing is the interference occurring in communication coverage area in main beam overlay area before antenna zeroing, the interference beyond communication zone on satellite proper communication almost without affecting; For the measurement of communication capacity after zeroing, major concern be the variation of communication coverage area shape, be substantially not considered for the variation of shape beyond communication coverage area.

With reference to Fig. 1, the present invention is based on the antenna main beam shown in Fig. 2 and structure the formation, the method for the conformal of returning to zero, carry out as follows:

Step 1: antenna reception radiofrequency signal is processed.

The radiofrequency signal X of 7 bypass passages that antenna reception is arrived _e(t) be divided into two-way processing: the weights W that a road and Beam-former are pre-stored _smultiply each other, obtain beamformer output signal Y _e(t); Separately lead up to analog down, make the radiofrequency signal X of 7 bypass passages _e(t) become the analog if signal X (t) of 7 bypass passages.

Step 2: obtain three tunnel related datas.

Utilize beam selection device to select the analog if signal X of i bypass passage _i(t), by the branch road analog if signal X to selecting _i(t) carry out relevantly to the beamformer output signal y (t) after down-conversion, export following three circuit-switched data:

Branch road autocorrelation certificate: $R_{\mathrm{xx}}(i,n)=E\left[X_i\left(t\right)X_i^{*}\left(t\right)\right],$

Branch road Yu He road cross-correlation data: γ _i(n)=E[X _i(t) y ^*(t)],

Close road autocorrelation certificate: R _yy(i, n)=E[y (t) y ^*(t)],

Wherein, E[] expression average, () ^*represent conjugation, i=1,2 ..., 7, n=1,2 ..., M, M is the total degree that whole zeroing conformal process is carried out.

Step 3: judge whether bypass passage needs zeroing.

By branch road autocorrelation according to R _xx(i, n) and the branch road zeroing thresholding u that zero-setting system sets in advance compare: if R _xx(i, n) < u, shows that i bypass passage do not need zeroing, returns to step 2, if R _xx(i, n)>=u, shows that i bypass passage needs zeroing, performs step 4.

Step 4: the adaptive step that calculates branch road.

To beamformer output signal Y _e(t) carry out successively down-conversion and analog-to-digital conversion, obtain numeral and close circuit-switched data y (n); The autocorrelation that numeral is closed to circuit-switched data y (n) and branch road is according to R _xx(i, n) is calculated as follows the adaptive step μ of branch road _i(n):

${\mathrm{\&mu;}}_i\left(n\right)=\frac{1}{\mathrm{\&epsiv;}+R_{\mathrm{xx}}(i,n)}\&times;(\frac{1}{1+e^{-\mathrm{\&alpha;}{\left|y\left(n\right)\right|}^3}}-\mathrm{\&beta;}),$

Wherein, ε is step-length control coefrficient, and the span of ε is ε > 0; R _xxthe autocorrelation certificate of i branch road when (i, n) is the n time zeroing, R _xx(i, n) has certain adaptivity, can control the adaptive step μ of branch road _i(n) span, affects convergence rate and the stability of whole zeroing process; Y (n) is that the numeral of the n time zeroing is closed circuit-switched data, α is the control coefrficient that numeral is closed circuit-switched data y (n), α controls convergence and the convergence rate of iteration zeroing process, the span of α is α > 0, β is the adjustment coefficient that numeral is closed circuit-switched data y (n), and span is 0 < β < 1.

Step 5: the weight w of calculating branch road _i(n+1).

5a) cost function ε is optimized in definition ²for:

${\mathrm{\&epsiv;}}^2=w{\left(n\right)}^H{R}_{n}w\left(n\right)+\frac{k}{M}\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{({\left|P_w(m,w)\right|}^p-{\left|P_{w_q}\left(m\right)\right|}^2)}^2$

Wherein, Part I w (n) ^hr _nw (n) is the anti-interference part of zeroing, w (n) ^hr _nmore ditty zero interference free performance is better for the value of w (n); Part II for zeroing conformal part, utilize 2 norms of adaptive direction figure and static directional diagram, directional diagram amplitude difference approaches to make both direction figure consistent as far as possible in communication coverage area, thereby saves static directional diagram from damage;

In formula, w (n)=[w ₁(n) ..., w _i(n) ..., w _n(n)] be the set of branch road weights, W _s=[W _s(1) ..., W _s(i) ..., W _s(7)] represent Beam-former pre-stored weights, these weights are as the primary iteration weights of Beam-former, m=1, and 2 ..., M, M is the discrete point number that communication coverage area is divided, P _w(m, w)=w ^h(n) a (m) is adaptive direction figure, for static directional diagram, a (m)=[a ₁(m) ..., a _i(m) ..., a ₇(m)] be antenna direction diagram data, p is variable, different application background corresponding to p value, and k is zeroing parameter, k > 0;

5b) calculation cost function ε ²gradient ▽ _wε ²:

$\begin{array}{c}{\&dtri;}_w{\mathrm{\&epsiv;}}^2=2(R_{n}w\left(n\right)+\frac{\mathrm{kp}}{M}\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}({\left|P_w(m,w)\right|}^p-{\left|P_{w_q}\left(m\right)\right|}^p)\frac{P_w(m,w)}{P_w{(m,w)}^{2-p}}{a}^{*}\left(m\right))\\ =2\left(E\right[X\left(t\right)X^{*}\left(t\right)w\left(n\right)]+{\mathrm{\&beta;}}_i\left(n\right))=2\left(E\right[X\left(t\right)y^{*}\left(t\right)]+{\mathrm{\&beta;}}_i\left(n\right))\\ =2({\mathrm{\&gamma;}}_i\left(n\right)+{\mathrm{\&beta;}}_i\left(n\right))\end{array},$

In formula, ${\mathrm{\&beta;}}_i\left(n\right)=\frac{\mathrm{kp}}{M}\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}({\left|P_w(m,w)\right|}^p-{\left|P_{w_q}\left(m\right)\right|}^p)\frac{P_w(m,w)}{P_w{(m,w)}^{2-p}}{a_i}^{*}\left(m\right)$ For zeroing conformal parameter;

5c) use steepest descent method, obtain the following formula of the weights that calculate branch road:

$w_i(n+1)=\left\{\begin{array}{cc}{W}_s\left(i\right)-\frac{1}{2}{\mathrm{\&mu;}}_i\left(n\right){\&dtri;}_w{\mathrm{\&epsiv;}}^2& n=1\\ w_i\left(n\right)+\frac{\left|y\left(n\right)y(n-1)\right|}{\mathrm{\&alpha;}{\mathrm{\&lambda;}}_{\max }}(w_i\left(n\right)-w_i(n-1))-\frac{1}{2}{\mathrm{\&mu;}}_i\left(n\right){\&dtri;}_w{\mathrm{\&epsiv;}}^2& n>1\end{array}\right.,$

Wherein w _i(n+1) be the weights of i branch road of the n+1 time zeroing, w _i(n) be the weights of i branch road of the n time zeroing, w _i(n-1) be the weights of i branch road of the n-1 time zeroing, W _s(i) be the pre-stored weights of i bypass passage, μ _i(n) be the adaptive step of branch road, y (n) is that the numeral of the n time iteration zeroing is closed circuit-switched data, y (n-1) is that the numeral of the n-1 time iteration zeroing is closed circuit-switched data, α is a variable, the span of α is 0 < α < 0.5, λ _maxthe autocorrelation matrix R of analog if signal _n=E[X (t) X ^*(t) eigenvalue of maximum].

By cost function gradient ▽ _wε ²substitution weights formula, obtains the weight w of branch road _i(n+1):

$w_i(n+1)=\left\{\begin{array}{cc}{W}_s\left(i\right)-{\mathrm{\&mu;}}_i\left(n\right)({\mathrm{\&gamma;}}_i\left(n\right)+{\mathrm{\&beta;}}_i\left(n\right))& n=1\\ w_i\left(n\right)+\frac{\left|y\left(n\right)y(n-1)\right|}{\mathrm{\&alpha;}{\mathrm{\&lambda;}}_{\max }}(w_i\left(n\right)-w_i(n-1)){\mathrm{\&mu;}}_i\left(n\right)({\mathrm{\&gamma;}}_i\left(n\right)+{\mathrm{\&beta;}}_i\left(n\right))& n>1\end{array}\right..$

Step 6: obtain new beamformer output signal.

Repeating step 2～5, completes after the traversal of 7 bypass passages successively, obtains the weight w of 7 branch roads _i(n), i=1,2 ..., 7; Use this weight w _i(n) be multiplied by the radiofrequency signal X of antenna reception _e(t), obtain new beamformer output signal Y _e(t).

Step 7: judge that whether zeroing conformal is successful.

7a) to new beamformer output signal Y _e(t) carry out down-conversion, obtain down-conversion signal y _p(t), by this down-conversion signal y _p(t) carry out auto-correlation, obtain Xin He road autocorrelation certificate

7b) Jiang Xinhe road autocorrelation is according to R _pthe judgement threshold value v that (i, n) and whole antenna zero-setting system are given compares: if R _p(i, n)>=v, illustrates that zeroing conformal is unsuccessful, disturbs and still can affect communication zone, repeating step 2～6; If R _p(i, n) < v, explanation zeroing conformal success, finishes zeroing conformal process.

Effect of the present invention can further illustrate by following emulation:

1. simulation parameter: the shaped aerial of 7 wave beams, the dry of each interference made an uproar than being 30dB.

2. emulation content and result:

Emulation 1, adopts the present invention to return to zero to the interference of different numbers, and result is as Fig. 4-Fig. 8.Wherein:

Fig. 4 is that the present invention is to the rear main beam contour map of single interference zeroing.As can be seen from Figure 4, while only there is single interference, occur a significant depressions in interference position, and the recessed position zeroing degree of depth is more than 30dB, outside disturbance null region, communication zone shape is good.

Fig. 5 is that the present invention disturbs the rear main beam contour map of zeroing to two.As can be seen from Figure 5, while existing two to disturb, occur two significant depressions in interference position, and the recessed position zeroing degree of depth is more than 30dB, outside disturbance null region, communication zone shape is good.

Fig. 6 is that the present invention is to three rear main beam contour maps of near-by interference zeroing.As can be seen from Figure 6, while there is three near-by interference, after zeroing, the zeroing degree of depth of interference position is more than 30dB, and zeroing has no significant effect communication zone.

Fig. 7 is that the present invention is to three rear main beam contour maps of zeroing that disturb far away, as can be seen from Figure 7, while existing three far to disturb, after zeroing, the zeroing degree of depth of interference position is more than 30dB, due to the locus apart from each other disturbing, three interference recess separately is all similar to a band, falls into bands and all appears at edge due to three zero, minimum on the impact of core communication zone.

Fig. 8 is that the present invention is to being positioned at three rear main beam contour maps of zeroing that disturb far away of beam center.As can be seen from Figure 8, exist when being positioned at three of beam center and far disturbing, after zeroing obviously there are three depressions in interference position, and the recessed position zeroing degree of depth is more than 30dB, and after zeroing, communication zone shape is good.

Emulation 2, adopts the present invention to carry out iteration zeroing to the interference of different numbers, and the convergence result of zeroing is as Fig. 9-Figure 11.Wherein:

Fig. 9 is that the present invention is to single interference iteration convergence curve chart.As can be seen from Figure 9, while only having single interference, iterative process just can restrain in 5 times, fast convergence rate.

Figure 10 is that the present invention disturbs iteration convergence curve chart to two.As can be seen from Figure 10, while existing two to disturb, iterative process can convergence in 20 times, fast convergence rate.

Figure 11 is that the present invention disturbs iteration convergence curve chart to three.As can be seen from Figure 11, while existing three to disturb, iterative process can restrain in 35 times, fast convergence rate.

Emulation 3, adopts existing single channel conformal closed-loop policy and the present invention to estimate the interference of the different numbers coverage of communication after conformal that returns to zero, and estimated result is as shown in table 1:

Table 1 algorithm of the present invention and the coverage rate of existing single channel conformal closed-loop policy under disturbance number

As can be seen from Table 1, in the situation that ensureing to disturb number and interference space position all identical, adopt the coverage of communication of algorithm of the present invention to be significantly improved than the coverage of communication that adopts existing single channel conformal closed-loop policy, to the conformal better effects if of communication zone.

Claims (6)

1. for a closed loop zeroing conformal method for satellite communication antena, comprise the steps:

(1) the radiofrequency signal X of N antenna reception a being arrived bypass passage _e(t) divide two-way processing: the weights W that a road and Beam-former are pre-stored _smultiply each other and obtain beamformer output signal Y _e(t); Separately lead up to analog down, make the radiofrequency signal X of N bypass passage _e(t) become the analog if signal X (t) of N bypass passage;

(2) utilize beam selection device to select the analog if signal X of i bypass passage _i(t), to the branch road analog if signal X selecting _i(t) to the beamformer output signal y (t) after down-conversion carry out relevant, output three circuit-switched data: branch road autocorrelation is according to R _xx(i, n), branch road Yu He road cross-correlation data γ _i(n), close road autocorrelation according to R _yy(i, n), i=1,2 ..., N, N is total bypass passage number, n=1,2 ..., M, M is the total degree that whole zeroing conformal process is carried out;

(3) by branch road autocorrelation according to R _xx(i, n) and the zeroing thresholding u setting in advance compare: if R _xx(i, n) < u, shows that i bypass passage do not need zeroing, returns to step (2), if R _xx(i, n)>=u, execution step (4);

(4) adaptive step of calculating branch road: to beamformer output signal Y _e(t) carry out successively down-conversion and analog-to-digital conversion, obtain numeral and close circuit-switched data y (n); The autocorrelation that numeral is closed to circuit-switched data y (n) and branch road is according to R _xx(i, n) combines, and obtains the adaptive step μ of branch road _i(n);

(5) utilize following formula to calculate the weight w of branch road _i(n+1);

$w_i(n+1)=\left\{\begin{array}{cc}{W}_s\left(i\right)-{\mathrm{\&mu;}}_i\left(n\right)({\mathrm{\&gamma;}}_i\left(n\right)+{\mathrm{\&beta;}}_i\left(n\right))& n=1\\ w_i\left(n\right)+\frac{\left|y\left(n\right)y(n-1)\right|}{\mathrm{\&alpha;}{\mathrm{\&lambda;}}_{\max }}& n>1\end{array}\right.$

Wherein, i=1,2 ..., N, N is total bypass passage number, n=1,2 ..., M, M is the total degree that whole zeroing conformal process is carried out; w _i(n+1) be the weights of i branch road of the n+1 time zeroing, w _i(n) be the weights of i branch road of the n time zeroing, w _i(n-1) be the weights of i branch road of the n-1 time zeroing, W _s=[W _s(1) ..., W _s(i) ..., W _s(N)] represent the pre-stored weights of Beam-former, W _s(i) be the pre-stored weights of i bypass passage, y (n) is that the numeral of the n time zeroing is closed circuit-switched data, and y (n-1) is that the numeral of the n-1 time zeroing is closed circuit-switched data, γ _i(n) be the cross-correlation data on branch road Yu He road, μ _i(n) be the adaptive step of branch road, α is a variable, and the span of α is 0 < α < 0.5, λ _maxthe autocorrelation matrix R of analog if signal X (t) _n=E[X (t) X ^*(t) eigenvalue of maximum], E[] represent average, () * represents conjugation, β _i(n) be zeroing conformal parameter.

(6) repeating step (2)～(5), complete after the traversal of N bypass passage successively, obtain the weight w of N branch road _i(n), use this weight w _i(n) be multiplied by the radiofrequency signal X of antenna reception _e(t), obtain new beamformer output signal Y _e(t), i=1,2 ..., N, N is total bypass passage number;

(7) to new beamformer output signal Y _e(t) carry out down-conversion, obtain down-conversion signal y _p(t), by this down-conversion signal y _p(t) carry out auto-correlation, obtain Xin He road autocorrelation according to R _p(i, n), Jiang Xinhe road autocorrelation is according to R _pthe judgement threshold value v that (i, n) and whole antenna zero-setting system are given compares, if R _p(i, n)>=v, illustrates that zeroing conformal is unsuccessful, repeating step (2)～(6); If R _p(i, n) < v, explanation zeroing conformal success, finishes zeroing conformal process.

2. the closed loop for satellite communication antena according to claim 1 zeroing conformal method, is characterized in that the adaptive step μ of branch road in described step (4) _i(n), be calculated as follows:

${\mathrm{\&mu;}}_i\left(n\right)=\frac{1}{\mathrm{\&epsiv;}+R_{\mathrm{xx}}(i,n)}\&times;(\frac{1}{1+e^{-\mathrm{\&alpha;}{\left|y\left(n\right)\right|}^3}}-\mathrm{\&beta;}),$

Wherein, ε is step-length control coefrficient, and the span of ε is ε > 0; R _xx(i, n) the autocorrelation certificate of i branch road while being the n time zeroing, y (n) is that the numeral of the n time zeroing is closed circuit-switched data, α is the control coefrficient that numeral is closed circuit-switched data y (n), the span of α is α > 0, β is the adjustment coefficient that numeral is closed circuit-switched data y (n), and span is 0 < β < 1.

3. the zeroing of the closed loop for satellite communication antena conformal method according to claim 1, is characterized in that the zeroing conformal parameter β in described step (5) _i(n), calculate by following formula:

${\mathrm{\&beta;}}_i\left(n\right)=\frac{\mathrm{kp}}{M}\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}({\left|P_w(m,w)\right|}^p-{\left|P_{w_q}\left(m\right)\right|}^p)\frac{P_w(m,w)}{P_w{(m,w)}^{2-p}}{a_i}^{*}\left(m\right)$

Wherein, m=1,2 ..., M, M is the discrete point number that communication coverage area is divided, P _w(m, w)=w ^h(n) a (m) is adaptive direction figure, w (n)=[w ₁(n) ..., w _i(n) ..., w _n(n)] be the set of branch road weights, () ^hrepresent conjugate transpose, for static directional diagram, Ws=[W _s(1) ..., W _s(i) ..., W _s(N)] represent the pre-stored weights of Beam-former, a (m)=[a ₁(m) ..., a _i(m) ..., a _n(m)] be antenna direction diagram data, p is variable, different application background corresponding to p value, and k is zeroing parameter, k > 0.

4. the closed loop for satellite communication antena according to claim 1 zeroing conformal method, wherein step (2) described to the branch road analog if signal X selecting _i(t) to the beamformer output signal y (t) after down-conversion carry out relevant, output three circuit-switched data, undertaken by following formula:

$R_{\mathrm{xx}}(i,n)=E\left[X_i\left(t\right)X_i^{*}\left(t\right)\right]$

γ _i(n)＝E[X _i(t)y ^*(t)]

R _yy(i,n)＝E[y(t)y ^*(t)]

Wherein, R _xx(i, n) is branch road autocorrelation certificate, γ _i(n) be branch road Yu He road cross-correlation data, R _yy(i, n) Wei He road autocorrelation certificate, E[] expression average, () ^*represent conjugation.

5. the zeroing of the closed loop for satellite communication antena conformal method according to claim 1, the autocorrelation that numeral is closed to circuit-switched data y (n) and branch road that wherein step (4) is described is according to R _xx(i, n) combines, and obtains the adaptive step μ of branch road _i(n) be, that numeral is closed to circuit-switched data y (n) and branch road autocorrelation according to R _xx(i, n) substitution branch road adaptive step μ _i(n) computing formula, thus the adaptive step μ of branch road obtained _i(n).

6. the closed loop for satellite communication antena according to claim 1 zeroing conformal method, wherein step (7) described by down-conversion signal y _p(t) carry out auto-correlation, be calculated as follows:

$R_p(i,n)=E\left[y_p\left(t\right)y_p^{*}\left(t\right)\right]$

Wherein, R _p(i, n) represents new several autocorrelation certificates, the E[of closing] represent average, () ^*represent conjugation.