CN104777478A - Target searching and acquisition method of phased array radar - Google Patents

Abstract

The invention discloses a target searching and acquisition method of phased array radar, and belongs to the field of phased array tracking radar signal processing. The target searching and acquisition method comprises a searching process and an acquisition process, wherein the searching process is that a radar signal is searched according to two beam position arrangement states through beam guidance information, if the energy for receiving the signal is greater than a threshold, the signal is determined to meet the requirement of shifting to the acquisition process; in the acquisition process, compared with an ordinary phased array radar acquisition method, the detection is performed repeatedly to reduce the false alarm probability, the processes of angle measurement of single pulse and difference beam, angle measurement of alpha-beta filter closed-loop, and tracking filter are additionally provided, an angle cosine residual in an innovation process in the tracking filter processing procedure is used as a parameter for determining whether the acquisition is successful, so that the acquisition process has a process of convergence, the radar signal acquired when an initial beam points to a target position in a large error is more stable and reliable, and the whole automatic phased array tracking system is more steady.

Description

A kind of phased-array radar code acquisition order calibration method

Technical field

The present invention relates to tracking radar with phase array signal transacting field, particularly the phased array minitrack method of searching for/catching, is transitioned into the tenacious tracking stage with this.

Background technology

In recent years, the development of mobile communication, the requirement of communication system to communication distance and scope is more and more higher.Mobile satellite communication system becomes a kind of excellent means of mobile communication, can be used for automobile, train, naval vessel, aircraft, on the various mobile vehicle such as guided missile, because of its wide coverage, the advantages such as insensitive, message capacity is large are required to region, be widely used in multiple fields of military and civilian.The key equipment of mobile satellite communication system is antenna stabilization tracker.

Active phase array antenna has the ability of wave beam rapid scanning, adopts Electric phase-shift mode to control beam direction, can overcome the conversion of mechanical scanning antennas beam position inertia and thus to the restriction of system performance.Avoid the movement velocity of the mechanical servo of traditional mechanical radar antenna comparatively slow, do not catch up with attitude and the geographic position change of carrier, make the sensing of antenna in the dynamic case depart from satellite, cause communication quality to decline or cause the situation of communication disruption.

Tracking radar refers to that those can the following the tracks of target of Automatic continuous, constantly carry out accurately measuring to target component and the radar of target coordinate position parameter can be exported, Continuous Closed ring type is followed the tracks of automatically, principal feature that the high-acruracy survey of coordinates of targets parameter and high data output rate are target following.Monopulse radar tracking accuracy is high, and realize simple, number all selects monopulse systems with typical high precision tracking radar.

Satellite communication phased array antenna autotracking system, mainly comprises the part such as receiving antenna array, transmitting antenna array, RF front-end module, autotracking baseband processing unit, wave beam guiding control module, external clock and power supply.Auto-tracking receive system is by becoming 4 submatrixs by antenna array partition.Signal syntheses four submatrixs of each antenna element, carry out 4 road A/D and convert, and 4 way battle array signals become intermediate-freuqncy signal through analog down and after ADC sampling, first complete the works for the treatment of such as complex signal generation, channel correcting by pretreatment module.Then enter code acquisition state, the guidance information provided according to carrier platform completes the code acquisition of satellite-signal.After acquisition success, system enters autotracking state, by measurement of angle and tracking filter etc. process make antenna beam auto-alignment satellite, realize receiving array wave beam to satellite from motion tracking.Target being detected in real time whether with losing in tracking filter process, being judged to then again to proceed to code acquisition state after losing.

There is certain angular error unavoidably in the satellite position forecast information obtained due to motion platform equipment and realistic objective, need radar to carry out code acquisition process within the specific limits, the angular error reduced between target location meets the track loop tracking mode condition of antenna.For code acquisition process, general code acquisition is exactly by Threshold detection and judgement, a kind of typical technology arranges suitable thresholding, make threshold voltage enough high, thresholding is exceeded to prevent most of noise spike, low being enough to allows weak signal pass through again, what just have has done twice or has repeatedly adjudicated confirmation, and the present invention by means of closed loop tracking filter method, measurement of angle and tracking filter are used, acquisition procedure is made to have a process restrained, more stable, whole algorithm is simple and easy stable, there is higher search speed, the autotracking system of mobile platform at a high speed can be applied to, ensure the accuracy of autotracking.

Tracking filtering technique is the core of tracker.Further investigation is also carried out to this technology both at home and abroad.Kalman (Kalman) describes stochastic linear system with state equation, proposes Kalman filtering algorithm.This algorithm is applicable to steadily and nonstationary random process simultaneously.Kalman filtering is optimum method of estimation in linear system, but Kalman filtering needs Real-time solution gain matrix, relate to more complicated matrix inversion process, for the Project Realization of algorithm adds difficulty, for nonlinear system, based on the thought of nonlinear equation being carried out linear-apporximation, propose the nonlinear filtering algorithm such as EKF.But this kind of algorithm is easily dispersed, and engineer applied is restricted.Unscented kalman filter and the newer nonlinear filtering algorithm of particle filter two kinds, but these methods are still in theoretical research stage at present, engineering are difficult to realize.Engineering adopts alpha-beta filtering algorithm usually, and this algorithm is a kind of filtering method of constant gain amplifier, this method avoid gain matrix Solve problems complicated in Kalman filtering, is easy to realize, thus in engineering, obtains more application.

Summary of the invention

The invention provides a kind of based on the phased-array radar code acquisition method of class monopulse with difference measuring angle by comparing amplitude, the method is compared with traditional code acquisition method, general code acquisition is by Threshold detection and judgement, what just have has done twice or has repeatedly adjudicated confirmation, and this method by means of angleonly tracking filtering, after search procedure is by Threshold detection to signal, carrying out class monopulse and difference measuring angle by comparing amplitude and the Closed loop track process of alpha-beta filtering makes acquisition procedure have a process restrained, and makes to catch more reliable and more stable.

The present invention specifically adopts following technical scheme:

A kind of phased-array radar code acquisition order calibration method, its flow process as shown in Figure 1, specifically comprises the following steps:

Step 1. phased-array radar Subarray partition;

Phased Array Radar Antenna array partition is become four submatrixs, be designated as submatrix 1, submatrix 2, submatrix 3 and submatrix 4 respectively;

Step 2. sets energy measuring thresholding E _t;

Note n ₁(n), n ₂(n), n ₃(n), n ₄n () is respectively when described four submatrixs all do not receive the signal of target acquisition and outputs signal accordingly, then now composite signal is designated as n _Σ=n ₁(n)+n ₂(n)+n ₃(n)+n ₄(n); Obtain noise gross energy thus:

$E_{\mathrm{\Σ}}=\underset{n=1}{\overset{N-1}{\mathrm{\Σ}}}{n_{\mathrm{\Σ}}}^2\left(n\right)$

Solve average and the variance of described noise gross energy σ _Σ, obtain energy measuring thresholding wherein k ₁value 1.0 to 1.2, k ₂value 4 to 5;

Step 3. target search procedure; After the guidance information provided from carrier platform obtains the initial position of target, point to this initial position by the beam. position controlling phased-array radar, and carry out beam search centered by this initial position; Beam search specifically carries out in the following ways:

The ripple position for beam search that step 3-1. builds two kinds of different range arranges, and is designated as arrangement 1, arrangement 2 respectively;

Described arrangement 1 as shown in Figure 4, comprises 5 ripple positions: middle cardiac wave position 1, lays respectively at the upper left side of middle cardiac wave position 1, upper right side, lower left and bottom-right ripple position 2,3,4 and ripple position 5; Described middle cardiac wave position 1 and ripple position 2,3,4,5 are overlapped mutually, the center of described arrangement 1 medium wave position 2,3,4,5 all with center half beam angle apart of ripple position 1;

Described arrangement 2 as shown in Figure 5, comprise nine ripple positions of the arrangement in nine grids: ripple position 1 is positioned at nine grids center, described nine grids first row is respectively ripple position 2,3 and ripple position 4 from left to right, the left and right of ripple position 1 is respectively ripple position 5, ripple position 6, and nine grids the 3rd row is respectively ripple position 7,8,9 from left to right; In arrangement 2 between left and right or neighbouring Liang Gebowei center all at a distance of a beam angle;

Step 3-2. ripple position 1 centered by the initial position of target builds the beam search arrangement mode shown in arrangement 1, the beam position of phased-array radar is adjusted to the initial position of target, calculates the energy of now four submatrix composite signals;

The energy of described composite signal is specifically obtained by following methods:

Be with reference to submatrix with submatrix 1, respectively Wave beam forming done to each submatrix, outputed signal y accordingly ₁(n), y ₂(n), y ₃(n) and y ₄n described four output signals are synthesized to obtain Y by () _Σ(n)=y ₁(n)+y ₂(n)+y ₃(n)+y ₄n (), solves the energy of four submatrix composite signals thus:

n represents the total sampling number in time domain;

If described energy is not less than described energy measuring thresholding E _t, then target search completes, and remembers that the beam position of now phased-array radar is for catching with beam position, performs step 4; Otherwise, the beam position of phased-array radar is adjusted to successively ripple position 2,3,4,5 until the energy of four submatrix composite signals is not less than described energy measuring thresholding E _tin time, stops; If when the beam position of phased-array radar is adjusted to ripple position 5, the energy of four submatrix composite signals is still less than described energy measuring thresholding E _t, perform step 3-3;

Step 3-3. ripple position 1 centered by the initial position of target builds the beam search arrangement mode shown in arrangement 2, the beam position of phased-array radar is adjusted to the initial position of target, calculates the energy of now four submatrix composite signals;

If described energy is not less than described energy measuring thresholding E _t, then target search completes, and remembers that the beam position of now phased-array radar is for catching with beam position, performs step 4; Otherwise, the beam position of phased-array radar is adjusted to successively ripple position 2,3,4,5,6,7,8,9 until the energy of four submatrix composite signals is not less than described energy measuring thresholding E _tin time, stops; If when the beam position of phased-array radar is adjusted to ripple position 9, the energy of four submatrix composite signals is still less than described energy measuring thresholding E _t, then carrier platform reinitializes and obtains up-to-date guidance information and the up-to-date initial position providing target to phased-array radar, and again performs step 3-1 to step 3-3;

Step 4. target acquistion process;

Step 4-1. catching with the initial value of beam position as class Monopulse estimation using step 3 gained, utilizes He er bu tong measuring angle by comparing amplitude method to solve the position angle of place's echo signal and the angle of pitch

The beam permutation that described and poor measuring angle by comparing amplitude method adopts as shown in Figure 6, comprises 5 ripple positions: middle cardiac wave position 5, to lay respectively at directly over middle cardiac wave position 5, directly over, the ripple position 1,2,3 of left and right and ripple position 4; Described middle cardiac wave position 5 and ripple position 1,2,3,4 are overlapped mutually, the center of described arrangement medium wave position 1,2,3,4 all with center half beam angle apart of ripple position 5;

Step 4-2. is taken measurement of an angle according to monopulse He er bu tong measuring angle by comparing amplitude try to achieve the direction cosine under corresponding angle:

With target in the direction cosine in three directions and the speed of change thereof for state vector:

${\left[\begin{array}{cccccc}{x}_c& {\stackrel{.}{x}}_c& y_c& {\stackrel{.}{y}}_c& z_c& {\stackrel{.}{z}}_c\end{array}\right]}^T$

Wherein for the speed of direction cosine change, application alpha/beta algorithm tracking filter obtains the initial angle of incident angle estimated value as angle measurement next time;

Step 4-3. is in filtering, and by various factors, may have in observation data between partial data and actual value and there is very big error, these data are called as outlier, and outlier can produce larger impact to tracking performance, needs to reject; When carrying out alpha-beta filtering, innovation process modulus value and a certain thresholding W are adjudicated, if meet:

$\left|\mathrm{\&alpha;}\left(n\right)\right|=|z\left(n\right)-\hat{z}\left(n\right|n-1)|<W$

Observation data is effective, utilizes alpha-beta filtering to carry out state estimation, and utilizes inverse trigonometric function to solve position angle after filtering and the angle of pitch initial value as class Monopulse estimation next time; Otherwise, Current observation value is judged to outlier, performs step 4-4;

Wherein, threshold value is chosen as W and elects 3 times of respective vectors formed of three direction observation noise criteria differences as;

If step 4-4. Current observation value is judged to outlier, then utilize the state vector estimated value in last cycle to do linear extrapolation when state estimation, be judged to be that tracking target is lost when continuous 3 observation moment all detect outlier, now should reinitialize alpha-beta tracking filter and re-execute whole search described in step 3 to step 4 and acquisition procedure;

By angle cosine residual error in step 4-5. filtering | α _t| with | β _t| adjudicate with thresholding, threshold value gets 1/6 to 1/3 beam angle usually, when | α _t| with | β _t| then think that acquisition procedure is restrained when being all less than described threshold value, target acquistion completes and can proceed to tracing process, otherwise, by repeated execution of steps 4-1 to step 4-4 until acquisition procedure convergence, complete target acquistion.

The invention has the beneficial effects as follows: classic method is carried out twice or the process of repeated detection after search procedure is by Threshold detection to signal, although can detection probability be improved but still cannot ensure that the beam position entering autotracking process meets autotracking conditional request at every turn, if beam position information has misunderstanding to make autotracking resultant error large or can not use, this method is after search procedure is by Threshold detection to signal, Closed loop track process has been carried out in acquisition procedure, and now 2 ~ 5 times of the normally normal tracking mode of the sampling rate of tracking filter can make acquisition procedure Fast Convergent, enter autotracking process beam position information and meet its tracking condition, guarantee tracking results is accurate, improve autotracking precision.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of phased-array radar target searching and acquisition method provided by the invention;

Fig. 2 is the coordinate distribution of Phased Array Radar Antenna;

Fig. 3 is adjacent two wave beam schematic diagram;

Fig. 4 is the arranged distribution 1 of search ripple position;

Fig. 5 is the arranged distribution 2 of search ripple position;

Fig. 6 is and differs from measuring angle by comparing amplitude beam permutation.

Embodiment

Code acquisition method of the present invention comprises search procedure and acquisition procedure, and first system carries out target location search according to the wave beam guidance information of platform, enters acquisition procedure after searching useful signal.Wherein, search procedure, by carrying out beam scanning to two kinds of Beam Position Arrangement states, carries out energy measuring to each ripple position, to obtain effective detection signal; Acquisition procedure, after search condition detects target, array beams is to detecting that the direction of signal repeats to receive and detect, after getting rid of false alarm condition, carry out the class monopulse of closed loop and difference measuring angle by comparing amplitude and alpha-beta tracking filter, make acquisition procedure Fast Convergent, enter autotracking state.

The present invention specifically adopts following technical scheme:

A kind of phased-array radar code acquisition order calibration method, its flow process as shown in Figure 1, specifically comprises the following steps:

Step 1. phased-array radar divides;

Phased-array radar is divided into four submatrixs, is designated as submatrix 1, submatrix 2, submatrix 3 and submatrix 4 respectively;

Step 2. sets energy measuring thresholding E _t;

Note n ₁(n), n ₂(n), n ₃(n), n ₄n () is respectively when described four submatrixs all do not receive the signal of target acquisition and outputs signal accordingly, then now composite signal is designated as n _Σ=n ₁(n)+n ₂(n)+n ₃(n)+n ₄(n); Obtain noise gross energy thus:

$E_{\mathrm{\&Sigma;}}=\underset{n=1}{\overset{N-1}{\mathrm{\&Sigma;}}}{n_{\mathrm{\&Sigma;}}}^2\left(n\right)---\left(1\right)$

Solve average and the variance of described noise gross energy σ _Σ, obtain energy measuring thresholding wherein k ₁∈ [1.0,1.2], k ₂∈ [4,5];

Step 3. target search procedure; After the guidance information provided from carrier platform obtains the initial position of target, the beam. position controlling phased-array radar points to this initial position, and carries out beam search centered by this initial position; Beam search specifically carries out in the following ways:

The ripple position for beam search that step 3-1. builds two kinds of different range arranges, and is designated as arrangement 1, arrangement 2 respectively;

Described arrangement 1 as shown in Figure 4, comprises 5 ripple positions: middle cardiac wave position 1, lays respectively at the upper left side of middle cardiac wave position 1, upper right side, lower left and bottom-right ripple position 2,3,4 and ripple position 5; Described middle cardiac wave position 1 and ripple position 2,3,4,5 are overlapped mutually, the center of described arrangement 1 medium wave position 2,3,4,5 all with center half beam angle apart of ripple position 1;

Described arrangement 2 as shown in Figure 5, comprise nine ripple positions of the arrangement in nine grids: ripple position 1 is positioned at nine grids center, described nine grids first row is respectively ripple position 2,3 and ripple position 4 from left to right, the left and right of ripple position 1 is respectively ripple position 5, ripple position 6, and nine grids the 3rd row is respectively ripple position 7,8,9 from left to right; In arrangement 2 between left and right or neighbouring Liang Gebowei center all at a distance of a beam angle;

Step 3-2. ripple position 1 centered by the initial position of target builds the beam search arrangement mode shown in arrangement 1, the beam position of phased-array radar is adjusted to the initial position of target, calculates the energy of now four submatrix composite signals;

The energy of described composite signal is specifically obtained by following methods:

Be with reference to submatrix with submatrix 1, respectively Wave beam forming done to each submatrix, outputed signal y accordingly ₁(n), y ₂(n), y ₃(n) and y ₄n described four output signals are synthesized to obtain Y by () _Σ(n)=y ₁(n)+y ₂(n)+y ₃(n)+y ₄n (), solves the energy of four submatrix composite signals thus:

wherein N is the total sampling number in time domain; (2)

If described energy F _Σbe not less than described energy measuring thresholding E _t, then target search completes, and remembers that the beam position of now phased-array radar is for catching with beam position, performs step 4; Otherwise, the beam position of phased-array radar is adjusted to successively ripple position 2,3,4,5 until the energy F of four submatrix composite signals _Σbe not less than described energy measuring thresholding E _tin time, stops; If when the beam position of phased-array radar is adjusted to ripple position 5, the energy F of four submatrix composite signals _Σstill be less than described energy measuring thresholding E _t, perform step 3-3;

Step 3-3. ripple position 1 centered by the initial position of target builds the beam search arrangement mode shown in arrangement 2, the beam position of phased-array radar is adjusted to the initial position of target, calculates the energy of now four submatrix composite signals;

If described energy F _Σbe not less than described energy measuring thresholding E _t, then target search completes, and remembers that the beam position of now phased-array radar is for catching with beam position, performs step 4; Otherwise, the beam position of phased-array radar is adjusted to successively ripple position 2,3,4,5,6,7,8,9 until the energy F of four submatrix composite signals _Σbe not less than described energy measuring thresholding E _tin time, stops; If when the beam position of phased-array radar is adjusted to ripple position 9, the energy F of four submatrix composite signals _Σstill be less than described energy measuring thresholding E _t, then carrier platform reinitializes and obtains up-to-date guidance information and the up-to-date initial position providing target to phased-array radar, and again performs step 3-1 to step 3-3;

Step 4. target acquistion process;

Step 4-1. catching with the initial value of beam position as class Monopulse estimation using step 3 gained, utilizes He er bu tong measuring angle by comparing amplitude method to solve the position angle of place's echo signal and the angle of pitch concrete method for solving is as follows:

Make radar array place plane be x-o-y plane, in acquisition procedure, the incoming signal initial directional of phased-array radar and the angle of x-axis and y-axis are respectively α _xand α _y; Incoming signal and z-axis angle are pitching angle theta, and its span is 0 ° ~ 90 °, and incoming signal is position angle in the projection of x-o-y plane and x-axis angle its span is 0 ° ~ 360 °, (α _x, α _y) with there is following relation:

The signal that 4 antenna submatrixs export after phase shift is designated as y respectively ₁(n), y ₂(n), y ₃(n) and y ₄n (), with the Received signal strength y of first passage ₁n () carries out relevant treatment respectively with four signals respectively, just can obtain the related coefficient between sub antenna battle array 1 and its complementary submatrix: R _1,1, R _1,2, R _1,3and R _{isosorbide-5-Nitrae}; The angle information in incident source is included in these four related coefficients, concrete:

$R_{1,i}=\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}{y}_1*\left(n\right)y_i\left(n\right),i=\mathrm{1,2,3,4}$

Total sampling number wherein in N time domain, four related coefficients are regarded as the complex magnitude that four submatrixs export, can obtain and signal:

F _Σi＝R _1,1+R _1,2+R _1,3+R _1,4(5)

Centered by wave beam 5, under angle cosine coordinate system, wave beam 5 is oriented to (α ₅, β ₅)=(α ₀, β ₀); Along half beam angle in this direction of skew each about α coordinate direction, order forms wave beam 1 and wave beam 2, two beam position is respectively:

$\left\{\begin{array}{c}({\mathrm{\&alpha;}}_1,{\mathrm{\&beta;}}_1)=({\mathrm{\&alpha;}}_0-{\mathrm{\&alpha;}}_{3\mathrm{dB}}/2,{\mathrm{\&beta;}}_0)\\ ({\mathrm{\&alpha;}}_2,{\mathrm{\&beta;}}_2)=({\mathrm{\&alpha;}}_0+{\mathrm{\&alpha;}}_{3\mathrm{dB}}/2,{\mathrm{\&beta;}}_0)\end{array}\right.---\left(6\right)$

Similarly, form wave beam 3 and wave beam 4 successively along β coordinate direction, beam position is respectively

$\left\{\begin{array}{c}({\mathrm{\&alpha;}}_3,{\mathrm{\&beta;}}_3)=({\mathrm{\&alpha;}}_0,{\mathrm{\&beta;}}_0-{\mathrm{\&beta;}}_{3\mathrm{dB}}/2)\\ ({\mathrm{\&alpha;}}_4,{\mathrm{\&beta;}}_4)=({\mathrm{\&alpha;}}_0,{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_{3\mathrm{dB}}/2)\end{array}\right.---\left(7\right)$

In above two formulas, α _3dBwith β _3dBrepresent the half-power beam width of α coordinate direction and β coordinate direction under angle cosine coordinate system respectively.With incoming signal direction the closer to the signal that receives of wave beam stronger, can judge that echo signal such as to depart from the direction of signal shaft accordingly.

Received signal strength synthesis is carried out to 5 beam positional, can obtain 5 orientation with wave beam Received signal strength energy F _{Σ 1}~ F _{Σ 5}.Wherein F _{Σ i}wherein try to achieve according to formula (28), following formula asks for the error voltage in α and β direction:

$u_{\mathrm{\&alpha;}}=\frac{\left|F_{\mathrm{\&Sigma;}1}\right|-\left|F_{\mathrm{\&Sigma;}2}\right|}{\left|F_{\mathrm{\&Sigma;}5}\right|},u_{\mathrm{\&beta;}}=\frac{\left|F_{\mathrm{\&Sigma;}4}\right|-\left|F_{\mathrm{\&Sigma;}3}\right|}{\left|F_{\mathrm{\&Sigma;}5}\right|}---\left(8\right)$

Remember respectively for target direction departs from the angle cosine value of central beam sensing, wherein with for target signal direction angle cosine, within the specific limits angular error signal u _αwith α _t, u _βwith β _tbe approximated to linear relationship, namely

$\left\{\begin{array}{c}{u}_{\mathrm{\&alpha;}}\&ap;k_{\mathrm{\&alpha;}}{\mathrm{\&alpha;}}_t\\ u_{\mathrm{\&beta;}}\&ap;k_{\mathrm{\&beta;}}{\mathrm{\&beta;}}_t\end{array}\right.---\left(9\right)$

Wherein slope k _αand k _βmatching can obtain by experiment, in practical application, solve k _αtime hypothesis β direction be oriented to target direction, at α ₀corresponding u in neighbouring left and right deviation beam angle scope _αwith α _tvalue and use beeline approaching u _α-α _trelation curve, the slope being obtained straight line by measured data matching solves k _α, the angular error signal in β direction and the relation curve slope k of angle cosine deviation _βwith the k in α direction _αmethod for solving identical, and then solve the angle cosine of sense

$\left\{\begin{array}{c}\widehat{\mathrm{\&alpha;}}=u_{\mathrm{\&alpha;}}/k_{\mathrm{\&alpha;}}+{\mathrm{\&alpha;}}_0\\ \widehat{\mathrm{\&beta;}}=u_{\mathrm{\&beta;}}/k_{\mathrm{\&beta;}}+{\mathrm{\&beta;}}_0\end{array}\right.---\left(10\right)$

Finally solve position angle and the angle of pitch of echo signal:

At α _t≤ α _3dB/ 2, β _t≤ β _3dBwhen/2, u _αwith α _t, u _βwith β _tbe approximated to linear relationship, along with at α _t> α _3dB/ 2, β _t> β _3dBwhen/2, along with α _t, β _tincrease u _αwith α _t, u _βwith β _tpresent nonlinear relationship, to position angle and the angle of pitch that nonlinear situation is carried out now adopting the processing mode of piecewise fitting or amplitude limiting processing to carry out matching is solved echo signal.

Step 4-2. is taken measurement of an angle according to monopulse He er bu tong measuring angle by comparing amplitude the direction cosine of trying to achieve under corresponding angle are:

With target in the direction cosine in three directions and the speed of change thereof for state vector:

${\left[\begin{array}{cccccc}{x}_c& {\stackrel{.}{x}}_c& y_c& {\stackrel{.}{y}}_c& z_c& {\stackrel{.}{z}}_c\end{array}\right]}^T---\left(13\right)$

Wherein for the speed of direction cosine change, application alpha/beta algorithm tracking filter obtains the initial angle of incident angle estimated value as angle measurement next time;

α-βfilter is a kind of the constant gain filters for uniform motion object module, and gain matrix wherein has following form:

$K\left(n\right)=\left[\begin{array}{ccc}{\mathrm{\&alpha;}}_{x_c}& 0& 0\\ {\mathrm{\&beta;}}_{x_c}/T& 0& 0\\ 0& {\mathrm{\&alpha;}}_{y_c}& 0\\ 0& {\mathrm{\&beta;}}_{y_c}/T& 0\\ 0& 0& {\mathrm{\&alpha;}}_{z_c}\\ 0& 0& {\mathrm{\&beta;}}_{z_c}/T\end{array}\right]---\left(14\right)$

Wherein with be respectively the position of dbjective state and the normal filter gain of speed component, T is the sampling period.First define three yaw maneuver indexs for:

$\left[\begin{array}{ccc}{\mathrm{\&lambda;}}_{x_c}& {\mathrm{\&lambda;}}_{y_c}& {\mathrm{\&lambda;}}_{z_c}\end{array}\right]=\left[\begin{array}{ccc}\frac{T^2\&CenterDot;{\mathrm{\&sigma;}}_1}{{\mathrm{\&sigma;}}_{x_c}}& \frac{T^2\&CenterDot;{\mathrm{\&sigma;}}_1}{{\mathrm{\&sigma;}}_{y_c}}& \frac{T^2\&CenterDot;{\mathrm{\&sigma;}}_1}{{\mathrm{\&sigma;}}_{z_c}}\end{array}\right]---\left(15\right)$

Wherein σ ₁for process noise and it is the standard deviation of three direction observation noises.With with for example expression formula as shown in the formula, with method for solving is identical

${\mathrm{\&alpha;}}_{x_c}=-\frac{{{\mathrm{\&lambda;}}_{x_c}}^2+8\&CenterDot;{\mathrm{\&lambda;}}_{x_c}-(4+{\mathrm{\&lambda;}}_{x_c})\&CenterDot;\sqrt{{{\mathrm{\&lambda;}}_{x_c}}^2+8\&CenterDot;{\mathrm{\&lambda;}}_{x_c}}}{8}---\left(16\right)$

${\mathrm{\&beta;}}_{x_c}={{\mathrm{\&alpha;}}_{x_c}}^2/(2-{\mathrm{\&alpha;}}_{x_c})---\left(17\right)$

State-transition matrix is

$F(n-1)=\left[\begin{array}{cccccc}1& T& 0& 0& 0& 0\\ 0& 1& 0& 0& 0& 0\\ 0& 0& 1& T& 0& 0\\ 0& 0& 0& 1& 0& 0\\ 0& 0& 0& 0& 1& T\\ 0& 0& 0& 0& 0& 1\end{array}\right]---\left(18\right)$

Observing matrix is

$H\left(n\right)=\left[\begin{array}{cccccc}1& 0& 0& 0& 0& 0\\ 0& 0& 1& 0& 0& 0\\ 0& 0& 0& 0& 1& 0\end{array}\right]---\left(19\right)$

Be filtering concrete steps below, the target state equation of discretize is:

$\hat{x}(n-1)={\left[\begin{array}{cccccc}{x}_c& {\stackrel{.}{x}}_c& y_c& {\stackrel{.}{y}}_c& z_c& {\stackrel{.}{z}}_c\end{array}\right]}^T---\left(20\right)$

Observation equation is z (n)=[x _c, y _c, z _c], filter step is specific as follows:

State one-step prediction:

$\hat{x}\left(n\right|n-1)=F(n-1)\hat{x}(n-1)---\left(21\right)$

Innovation process is calculated by observation signal z (n):

$\hat{z}\left(n\right|n-1)=H\left(n\right)\hat{x}\left(n\right|n-1)---\left(22\right)$

$\mathrm{\&alpha;}\left(n\right)=z\left(n\right)-\hat{z}\left(n\right|n-1)---\left(23\right)$

For alpha-beta filtering algorithm, utilize K (n), carry out state estimation:

$\hat{x}\left(n\right|n)=\hat{x}\left(n\right|n-1)+K\left(n\right)[\hat{z}\left(n\right)-\hat{z}\left(n\right|n-1)]---\left(24\right)$

After obtaining state estimation, extract in vector first, the 3rd, the value of the 5th parameter, namely extracts the angle cosine after filtering and solved by inverse trigonometric function relation as the input value of angle measurement next time, carry out the angle measurement of closed loop and the process of tracking.

Step 4-3. is in filtering, and by various factors, may have in observation data between partial data and actual value and there is very big error, these data are called as outlier; Outlier can produce larger impact to tracking performance, needs to reject; When carrying out alpha-beta filtering, its modulus value and a certain thresholding W should be adjudicated when calculating innovation process, if meet:

$\left|\mathrm{\&alpha;}\left(n\right)\right|=|z\left(n\right)-\hat{z}\left(n\right|n-1)|<W---\left(25\right)$

Then think that observation data is effective, otherwise be judged to outlier; Usual threshold value is chosen as W and elects 3 times of respective vectors formed of three direction observation noise criteria differences as;

If step 4-4. Current observation value is judged to outlier, then utilize the state vector estimated value in last cycle to do linear extrapolation when state estimation, be judged to be that tracking target is lost when continuous 3 observation moment all detect outlier, now should reinitialize alpha-beta tracking filter and re-execute whole search described in step 3 to step 4 and acquisition procedure;

By angle cosine residual error in step 4-5. filtering | α _t| with | β _t| adjudicate with thresholding, threshold value gets 1/6 to 1/3 beam angle usually, when | α _t| with | β _t| then think that acquisition procedure is restrained when being all less than described threshold value, target acquistion completes and can proceed to tracing process, otherwise, by repeated execution of steps 4-1 to step 4-4 until acquisition procedure convergence, complete target acquistion.

Claims (3)

1. a phased-array radar code acquisition order calibration method, specifically comprises the following steps:

Step 1. phased-array radar Subarray partition;

Phased Array Radar Antenna is divided into four submatrixs, is designated as submatrix 1, submatrix 2, submatrix 3 and submatrix 4 respectively;

Step 2. sets energy measuring thresholding E _t;

Note n ₁(n), n ₂(n), n ₃(n), n ₄n () is respectively when described four submatrixs all do not receive the signal of target acquisition and outputs signal accordingly, then now composite signal is designated as n _Σ=n ₁(n)+n ₂(n)+n ₃(n)+n ₄n (), obtains noise gross energy thus:

$E_{\mathrm{\&Sigma;}}=\underset{n=1}{\overset{N-1}{\mathrm{\&Sigma;}}}{n_{\mathrm{\&Sigma;}}}^2\left(n\right)$

Solve average and the variance of described noise gross energy σ _Σ, obtain energy measuring thresholding wherein k ₁∈ [1.0,1.2], k ₂∈ [4,5];

Step 3. target search procedure; After the guidance information provided from carrier platform obtains the initial position of target, the beam. position controlling Phased Array Radar Antenna points to this initial position, and carries out beam search centered by this initial position; Beam search specifically carries out in the following ways:

The ripple position for beam search that step 3-1. builds two kinds of different range arranges, and is designated as arrangement 1, arrangement 2 respectively;

Described arrangement 1 comprises 5 ripple positions: middle cardiac wave position 1, lays respectively at the upper left side of middle cardiac wave position 1, upper right side, lower left and bottom-right ripple position 2,3,4 and ripple position 5; Described middle cardiac wave position 1 and ripple position 2,3,4,5 are overlapped mutually, the center of described arrangement 1 medium wave position 2,3,4,5 all with center half beam angle apart of ripple position 1;

Described arrangement 2 comprises nine ripple positions of the arrangement in nine grids: ripple position 1 is positioned at nine grids center, described nine grids first row is respectively ripple position 2,3 and ripple position 4 from left to right, the left and right of ripple position 1 is respectively ripple position 5, ripple position 6, and nine grids the 3rd row is respectively ripple position 7,8,9 from left to right; In arrangement 2 between left and right or neighbouring Liang Gebowei center all at a distance of a beam angle;

Step 3-2. ripple position 1 centered by the initial position of target builds the beam search arrangement mode shown in arrangement 1, the beam position of phased-array radar is adjusted to the initial position of target, calculates the energy F of now four submatrix composite signals _Σ;

The energy of described composite signal is specifically obtained by following methods:

Be with reference to submatrix with submatrix 1, respectively Wave beam forming done to each submatrix, outputed signal y accordingly ₁(n), y ₂(n), y ₃(n) and y ₄n described four output signals are synthesized to obtain Y by () _Σ(n)=y ₁(n)+y ₂(n)+y ₃(n)+y ₄n (), solves the energy of four submatrix composite signals thus:

wherein N is the total sampling number in time domain;

If described energy F _Σbe not less than described energy measuring thresholding E _t, then target search completes, and remembers that the beam position of now phased-array radar is for catching with beam position, performs step 4; Otherwise, the beam position of phased-array radar is adjusted to successively ripple position 2,3,4,5 until the energy F of four submatrix composite signals _Σbe not less than described energy measuring thresholding E _tin time, stops; If when the beam position of phased-array radar is adjusted to ripple position 5, the energy F of four submatrix composite signals _Σstill be less than described energy measuring thresholding E _t, perform step 3-3;

Step 3-3. ripple position 1 centered by the initial position of target builds the beam search arrangement mode shown in arrangement 2, the beam position of phased-array radar is adjusted to the initial position of target, calculates the energy of now four submatrix composite signals;

If described energy F _Σbe not less than described energy measuring thresholding E _t, then target search completes, and remembers that the beam position of now phased-array radar is for catching with beam position, performs step 4; Otherwise, the beam position of phased-array radar is adjusted to successively ripple position 2,3,4,5,6,7,8,9 until the energy F of four submatrix composite signals _Σbe not less than described energy measuring thresholding E _tin time, stops; If when the beam position of phased-array radar is adjusted to ripple position 9, the energy F of four submatrix composite signals _Σstill be less than described energy measuring thresholding E _t, then carrier platform reinitializes and obtains up-to-date guidance information and the up-to-date initial position providing target to phased-array radar, and again performs step 3-1 to step 3-3;

Step 4. target acquistion process;

Step 4-1. catching with the initial value of beam position as class Monopulse estimation using step 3 gained, utilizes He er bu tong measuring angle by comparing amplitude method to solve the position angle of place's echo signal and the angle of pitch

Step 4-2. is taken measurement of an angle according to He er bu tong measuring angle by comparing amplitude the direction cosine of trying to achieve under corresponding angle are:

With target in the direction cosine in three directions and the speed of change thereof for state vector:

${\left[\begin{array}{cccccc}{x}_c& {\stackrel{\&CenterDot;}{x}}_c& y_c& {\stackrel{\&CenterDot;}{y}}_c& z_c& {\stackrel{\&CenterDot;}{z}}_c\end{array}\right]}^T$

Wherein for the speed of direction cosine change, application alpha/beta algorithm tracking filter obtains the initial angle of incident angle estimated value as angle measurement next time;

Its modulus value and thresholding W, when carrying out alpha-beta filtering, should be adjudicated when calculating innovation process α (n) by step 4-3., if meet:

wherein z (n) is the observation signal in alpha-beta filtering;

Then think that observation data is effective, and carry out state estimation, solve the initial angle of angle measurement next time; Otherwise innovation process α (n) is judged to outlier; Threshold value is elected W as and is elected 3 times of respective vectors formed of three direction observation noise criteria differences as;

If step 4-4. Current observation value is judged to outlier, then utilize the state vector estimated value in last cycle to do linear extrapolation when state estimation, be judged to be that tracking target is lost when continuous 3 observation moment all detect outlier, now should reinitialize alpha-beta tracking filter and re-execute whole search described in step 3 to step 4 and acquisition procedure;

By angle cosine residual error in step 4-5. filtering | α _t| with | β _t| adjudicate with thresholding, threshold value gets 1/6 to 1/3 beam angle usually, when | α _t| with | β _t| then think that acquisition procedure is restrained when being all less than described threshold value, target acquistion completes and can proceed to tracing process, otherwise, by repeated execution of steps 4-1 to step 4-4 until acquisition procedure convergence, complete target acquistion.

2. phased-array radar code acquisition order calibration method according to claim 1, is characterized in that, the position angle described in step 4 and the angle of pitch concrete method for solving as follows:

Wherein,

$\left\{\begin{array}{c}\widehat{\mathrm{\&alpha;}}=u_{\mathrm{\&alpha;}}/k_{\mathrm{\&alpha;}}+{\mathrm{\&alpha;}}_0\\ \widehat{\mathrm{\&beta;}}=u_{\mathrm{\&beta;}}/k_{\mathrm{\&beta;}}+{\mathrm{\&beta;}}_0\end{array}\right.$

The beam permutation that described and poor measuring angle by comparing amplitude method adopts comprises 5 ripple positions: middle cardiac wave position 5, to lay respectively at directly over middle cardiac wave position 5, directly over, the ripple position 1,2,3 of left and right and ripple position 4; Described middle cardiac wave position 5 and ripple position 1,2,3,4 are overlapped mutually, the center of described arrangement medium wave position 1,2,3,4 all with center half beam angle apart of ripple position 5;

Received signal strength synthesis is carried out to 5 beam positional, can obtain 5 orientation with wave beam Received signal strength energy F _{Σ 1}~ F _{Σ 5}, described energy can according to formula F _Σmethod for solving try to achieve, ask for error voltage:

$u_{\mathrm{\&alpha;}}=\frac{\left|F_{\mathrm{\&Sigma;}1}\right|-\left|F_{\mathrm{\&Sigma;}2}\right|}{\left|F_{\mathrm{\&Sigma;}5}\right|},u_{\mathrm{\&beta;}}=\frac{\left|F_{\mathrm{\&Sigma;}4}\right|-\left|F_{\mathrm{\&Sigma;}3}\right|}{\left|F_{\mathrm{\&Sigma;}5}\right|}$

Make phased-array antenna array place plane be x-o-y plane, in acquisition procedure, the incoming signal initial directional of phased-array radar and the angle of x-axis and y-axis are respectively α _xand α _y; Incoming signal and z-axis angle are pitching angle theta, and its span is 0 ° ~ 90 °, and incoming signal is position angle in the projection of x-o-y plane and x-axis angle its span is 0 ° ~ 360 °, (α _x, α _y) with there is following relation:

Wherein, slope k _αand k _βcan matching obtain by experiment: solve k _αtime hypothesis β direction be oriented to target direction, at α ₀corresponding u in neighbouring left and right deviation beam angle scope _αwith α _tvalue and use beeline approaching u _α-α _trelation curve, the slope being obtained straight line by measured data matching solves k _α; The angular error signal in β direction and the relation curve slope k of angle cosine deviation _βidentical with above-mentioned method for solving.

3. phased-array radar code acquisition order calibration method according to claim 1, it is characterized in that, described alpha-beta filtering method is specific as follows:

The target state equation of discretize is:

$\hat{x}(n-1)={\left[\begin{array}{cccccc}{x}_c& {\stackrel{\&CenterDot;}{x}}_c& y_c& {\stackrel{\&CenterDot;}{y}}_c& z_c& {\stackrel{\&CenterDot;}{z}}_c\end{array}\right]}^T$

Observation equation is z (n)=[x _c, y _c, z _c], filtering is specific as follows:

State one-step prediction:

$\hat{x}\left(n\right|n-1)=F(n-1)\hat{x}(n-1)$

Innovation process is calculated by observation signal z (n):

$\begin{array}{c}\hat{z}\left(n\right|n-1)=h\left(n\right)\hat{x}\left(n\right|n-1)\\ \mathrm{\&alpha;}\left(n\right)=z\left(n\right)-\hat{z}\left(n\right|n-1)\end{array}$

α-βfilter is a kind of the constant gain filters for uniform motion object module, and its gain matrix is K (n), utilizes K (n), carries out state estimation:

$\hat{x}\left(n\right|n)=\hat{x}\left(n\right|n-1)+K\left(n\right)[\hat{z}\left(n\right)-\hat{z}\left(n\right|n-1)]$

Extract the value of the first, the 3rd, the 5th parameter in vector, namely extracts the angle cosine after filtering and solved by inverse trigonometric function relation as the input value of angle measurement next time, carry out the angle measurement of closed loop and the process of tracking.