CN106093870A - The SAR GMTI clutter suppression method of hypersonic aircraft descending branch - Google Patents

Abstract

The invention discloses the SAR GMTI clutter suppression method of a kind of hypersonic aircraft descending branch, its thinking is: set up the radar motion geometric model of hypersonic aircraft descending branch, wherein radar comprises N number of antenna channels, P is radar any one moving target in the scene, and the instantaneous oblique distance between the n-th antenna channels and moving target P is expressed as R_n(t_a), obtain the echo-signal of N number of displaced phase center passage；Obtain the echo-signal of the N number of displaced phase center passage distance frequency domain orientation laggard line phase of time-domain representation form after Range compress to compensate, the distance frequency domain orientation time domain echo-signal of N number of displaced phase center passage after obtaining phase compensation, the echo-signal of the moving target P of N number of displaced phase center passage echo-signal of computed range Doppler domain, range-Dopler domain and the noise signal of range-Dopler domain the most successively, calculate the optimum weight coefficient vector of space-time adaptive clutter recognition, finally give the echo-signal of moving target P after clutter recognition.

Description

The SAR-GMTI clutter suppression method of hypersonic aircraft descending branch

Technical field

The invention belongs to Radar Technology field, particularly to the SAR-GMTI clutter of a kind of hypersonic aircraft descending branch Synthetic aperture radar-Ground moving target detection (SAR-GMTI) clutter of suppressing method, i.e. hypersonic aircraft descending branch Suppressing method, it is adaptable to the hypersonic aircraft descending branch suppression to SAR-GMTI clutter, thus complete dynamic mesh in SAR imaging Target testing goal.

Background technology

Synthetic aperture radar (SAR) imaging technique initially grows up in the 1950's, it is possible to provide simultaneously High two-dimensional resolution, including High Range Resolution and high azimuth resolution, and then can carry out imaging to radar target；And close Aperture radar-Ground moving target detection (SAR-GMTI) is become to combine Clutter Rejection Technique, it is possible at the shadow of strong surface feature clutter The detection to ground moving object is completed under sound, the most significant for the field such as battle reconnaissance, remote sensing.

Existing research is all based on spaceborne or airborne synthetic aperture radar/synthetic aperture radar-ground moving object Detection (SAR/SAR-GMTI), carried SAR flying height is low and flight speed slow, and detection range is less；Satellite-borne SAR flying height Height, need larger emitted energy detection ground target, due to practical devices launch power restriction be extremely difficult to satellite-borne SAR- Emitted energy requirement needed for GMTI, adds satellite orbit the most fixing, so it is easily spied out by enemy and implements Interference or strike.

In view of the disadvantages mentioned above of carried SAR and satellite-borne SAR carries out studying the synthesis hole that hypersonic aircraft (HSV) carries Footpath radar (HSV-borne SAR) imaging problem, hypersonic aircraft (HSV) can around the earth one week in 2 hours, and There is high speed and high maneuverability advantage simultaneously, be therefore difficult to be scouted and hit；It addition, hypersonic aircraft (HSV) leads to Often flying in the range of near-space, compared to space spaceborne radar, hypersonic aircraft (HSV) needs less energy just can Ground target detected；In addition, hypersonic aircraft (HSV) is also expected to for completing remote battle reconnaissance, fire Power controls and the weapon platform of precision strike task, and therefore hypersonic aircraft (HSV) has researching value greatly.So And, it is different from traditional platform, the radar motion state complex in hypersonic aircraft (HSV), is generally of spasmodic motion State, and time in jump phase the acceleration time generally the shortest, only account for the 1/6 or 1/9 of whole jump phase, so high ultrasonic The synthetic aperture radar (HSV-borne SAR) that speed aircraft (HSV) is carried only can be set in descending branch and work, and When entering into descending branch, the strongest ground clutter can be faced, and then to moving target detection and tracking, imaging and knowledge afterwards Do not bring difficulty justice.

Summary of the invention

The deficiency existed for vacancy and the investigative technique of above research field, it is an object of the invention to propose a kind of height The SAR-GMTI clutter suppression method of supersonic aircraft descending branch, the SAR-GMTI of this kind of hypersonic aircraft descending branch is miscellaneous Ripple suppressing method can preferably solve the synthetic aperture radar (HSV-borne SAR) that hypersonic aircraft (HSV) carries The strong clutter faced in the descending branch impact on SAR-GMTI.

For reaching object above, the present invention uses following scheme to be achieved.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch, comprises the following steps:

Step 1, sets up the radar motion geometric model of hypersonic aircraft descending branch, at described hypersonic aircraft In the radar motion geometric model of descending branch, radar comprises N number of antenna channels, and P is radar any one motion in the scene Target, and the instantaneous oblique distance between the n-th antenna channels and moving target P is expressed as R_n(t_a)；Wherein, n ∈ 1,2 ..., N}, N represent the antenna channels number that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises, and N is Odd number more than 1, t_aRepresent the orientation slow time；

Step 2, the N number of antenna channels comprised by radar is designated as the 1st passage successively to N channel, and using the 1st passage as Reference channel, by (N+1)/2 channel emission linear FM signal, uses N number of antenna channels to receive moving target P place simultaneously The echo-signal of scene, obtains the echo-signal s (t of N number of passage_r,t_a), s (t_r,t_a)={ s₁(t_r,t_a),…,s_n(t_r, t_a),…,s_N(t_r,t_a)}；s_n(t_r,t_a) represent the echo-signal of the n-th passage, t_rRepresent that distance is to fast time, t_aRepresent that orientation is slow Time；Then set the constant phase compensating factor relative to reference channel, and the echo-signal of N number of passage is carried out perseverance respectively Phase bit compensates, and obtains the echo-signal of N number of displaced phase center passage Represent the echo-signal of the n-th displaced phase center passage；

Step 3, obtains the echo-signal of N number of displaced phase center passageDistance frequency domain after Range compress- Orientation time-domain representation form Represent n-th The echo-signal of individual displaced phase center passageDistance frequency domain after Range compress-orientation time-domain representation form, and Echo-signal to N number of displaced phase center passageDistance frequency domain after Range compress-orientation time-domain representation formCarry out phase compensation, the distance frequency domain of N number of displaced phase center passage-orientation time domain echo letter after obtaining phase compensation Number Represent after phase compensation n-th etc. Distance frequency domain-orientation time domain the echo-signal of effect phase center passage, f_rRepresent frequency of distance, t_aRepresent orientation slow time, t_rTable Show that distance is to the fast time；

Step 4, according to the distance frequency domain-orientation time domain echo-signal of displaced phase center passage N number of after phase compensationIt is calculated N number of displaced phase center passage echo-signal s (t of range-Dopler domain successively_r,f_a), distance how general Strangle the echo-signal st (t of the moving target P in territory_r,f_a) and the noise signal sc (t of range-Dopler domain_r,f_a), then according to such as The weight vector w of lower equations space-time adaptive clutter recognition:

$\left\{\begin{array}{cc}\underset{w}{min}& w^{H}Rw\\ s.t.& w^{H}a\left(f_a\right)=1\end{array}\right.$

Meet the optimum weight coefficient vector w that weight vector is space-time adaptive clutter recognition of above-mentioned formula_opt, and then obtain Echo-signal s of moving target P after clutter recognition_P(t_r,f_a)；Wherein,The value of w, s.t. table when expression obtains minima Show that constraints, R represent N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a) in Doppler frequency The covariance matrix of the multichannel output correspondence of unit, a (f_a) represent the steering vector matrix of N number of displaced phase center passage, f_a Represent the Doppler frequency of moving target P.

The present invention compared with prior art, has the advantage that

First, the inventive method uses the synthetic aperture radar (HSV-borne that hypersonic aircraft (HSV) carries SAR), on the one hand, compare satellite-borne synthetic aperture radar, it is possible to reduce transmitting signal power and reach the moving target inspection of good ground equally Survey purpose；On the other hand, compared to airborne synthetic aperture radar, there is higher flying height, it is possible to observe bigger scope； In addition, hypersonic aircraft (HSV) also has high maneuverability, is difficult to be scouted and hit, more conducively battle reconnaissance；

Second, the inventive method can effectively suppress the strong clutter of hypersonic aircraft (HSV) descending branch and therefrom examine Measure moving target, tracking afterwards, imaging and identification etc. are all had and is of great significance.

Accompanying drawing explanation

With detailed description of the invention, the present invention is described in further detail below in conjunction with the accompanying drawings.

Fig. 1 is the SAR-GMTI clutter suppression method flow chart of a kind of hypersonic aircraft descending branch of the present invention；

Fig. 2 is the radar motion geometric model figure of hypersonic aircraft descending branch；Wherein, in three-dimensional system of coordinate XYZ, Radar platform with constant angle of descent γ (referring to the angle between radar speed direction and X-axis) diagonally under direction do perseverance Constant speed degree is the linear uniform motion of V, and radar is φ at the downwards angle of visibility of platform, and radar is at platform to moving target P Nearly oblique distance is R₀, P is radar any one moving target in the scene, and radar is V in the horizontal velocity component of platform_x, thunder The vertical velocity component reaching place platform is V_z；

Fig. 3 is the single channel amplitude result figure after Range compress after distance-Doppler territory carries out three phase compensation；Its Middle transverse axis represent distance to sampled point, the longitudinal axis represents Doppler sample point；

Fig. 4 is three and channel combined carries out the amplitude result figure after clutter recognition；Wherein, transverse axis represents that distance is to sampling Point, the longitudinal axis represents Doppler sample point.

Detailed description of the invention

With reference to Fig. 1, for the SAR-GMTI clutter suppression method flow process of a kind of hypersonic aircraft descending branch of the present invention Figure；The SAR-GMTI clutter suppression method of described hypersonic aircraft descending branch, comprises the following steps:

Step 1, sets up the radar motion geometric model of hypersonic aircraft descending branch, at described hypersonic aircraft In the radar motion geometric model of descending branch, radar comprises N number of antenna channels, and P is radar any one motion in the scene Target, and the instantaneous oblique distance between the n-th antenna channels and moving target P is expressed as R_n(t_a)；Wherein, n ∈ 1,2 ..., N}, N represent the antenna channels number that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises, t_aTable Show the orientation slow time.

Specifically, with reference to Fig. 2, for the radar motion geometric model figure of hypersonic aircraft descending branch；Set up high ultrasonic The radar motion geometric model of speed aircraft descending branch, as in figure 2 it is shown, in three-dimensional system of coordinate XYZ, X-axis in the horizontal direction, Y On the right side of axle points to, Z axis is away from direction, the earth's core, and radar is H in the vertical height of platform, and radar platform is by along flight path direction Linearly aligned three antenna channels composition, the distance between adjacent two antennas is 2d；A relevant treatment time (CPI) In, radar platform with constant angle of descent γ (referring to the angle between radar speed direction and X-axis) diagonally under direction Doing the linear uniform motion that constant speed is V, radar is φ at the downwards angle of visibility of platform, and radar is at platform to moving target P Nearest oblique distance be R₀, then radar is V in the horizontal velocity component of platform_x, radar at the vertical velocity component of platform is V_z；P is radar any one moving target in the scene, and 0 moment moving target P is at (x_p,y_p) place, and 0 moment moving target The horizontal velocity component of P is v_xp, the vertical velocity component of 0 moment moving target P is v_yp, radar is at platform to moving target P Nearest oblique distance be R₀, W_gObserve the strip width of scene for moving target P place, and assume that radar is to be operated in positive side-looking In the case of (i.e. angle of strabismus is 0 °)；After the radar motion geometric model of hypersonic aircraft descending branch is set up, by n-th day Instantaneous oblique distance between line passage and moving target P is expressed as R_n(t_a):

$\begin{array}{c}{R}_n\left(t_a\right)=\sqrt{{(V_x{t}_a+x_n-x_p-v_{xp}{t}_a)}^2+{(H-V_z{t}_a+z_n)}^2+{(y_p+v_{yp}{t}_a)}^2}\\ \≈r+(-V_z\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})(t_a-t_c)+\frac{{(V_x-v_{xp})}^2+{(v_{yp}\cos \mathrm{\φ}+V_z\sin \mathrm{\φ})}^2}{2r}{(t_a-t_c)}^2+\\ z_n\cos \mathrm{\φ}-(\frac{V_z{z}_n{\sin }^2\mathrm{\φ}}{r}+\frac{v_{yp}{z}_n\sin \mathrm{\φ}\cos \mathrm{\φ}}{r})(t_a-t_c)\end{array}$

Wherein, V_xRepresent the radar horizontal velocity component at platform, t_aRepresent orientation slow time, t_cRepresent that the n-th passage arrives The time that distance between moving target P is required when being nearest oblique distance, x_pRepresent the 0 moment moving target P coordinate in X-axis, v_xp Representing the horizontal velocity component of 0 moment moving target P, H represents the radar vertical height at platform, V_zRepresent that radar is flat The vertical velocity component of platform, y_pRepresent the 0 moment moving target P coordinate in Y-axis, v_ypRepresent the vertical speed of 0 moment moving target P Degree component,R represents radar distance between platform and moving target P, and φ represents that radar is at platform Downwards angle of visibility, n ∈ 1,2 ..., N}, N represent that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises Antenna channels number, x_nRepresent the horizontal range between the n-th antenna channels and initial point, z_nRepresent that the n-th antenna channels is with former Vertical distance between point.

Step 2, the N number of antenna channels comprised by radar is designated as the 1st passage successively to N channel, and using the 1st passage as Reference channel, by (N+1)/2 channel emission linear frequency modulation (LFM) signal, uses N number of antenna channels to receive ground motion simultaneously The echo-signal of target P place scene, obtains the echo-signal s (t of N number of passage_r,t_a), s (t_r,t_a)={ s₁(t_r,t_a),…,s_n (t_r,t_a),…,s_N(t_r,t_a), s_n(t_r,t_a) represent the echo-signal of the n-th passage, t_rRepresent that distance is to fast time, t_aExpression side The position slow time；Then set the constant phase compensating factor relative to reference channel, and the echo-signal of N number of passage is entered respectively Row constant phase compensates, and obtains the echo-signal of N number of displaced phase center passage Represent the echo-signal of the n-th displaced phase center passage.

Specifically, the N number of antenna channels comprised by radar time is designated as the 1st passage to N channel according to putting in order, and will 1st passage is as reference channel, by (N+1)/2 channel emission linear frequency modulation (LFM) signal, uses N number of antenna channels simultaneously Receive the echo-signal of moving target P place scene, obtain the echo-signal s (t of N number of passage_r,t_a), s (t_r,t_a)={ s₁(t_r, t_a),…,s_n(t_r,t_a),…,s_N(t_r,t_a), s_n(t_r,t_a) represent the echo-signal of the n-th passage, t_rRepresent distance to the fast time, t_aRepresent the orientation slow time.

Then set the constant phase compensating factor relative to reference channel, and the echo-signal of N number of passage is entered respectively Row constant phase compensates, and obtains the echo-signal of N number of displaced phase center passage Representing the echo-signal of the n-th displaced phase center passage, wherein the echo compensated signal to the n-th passage sets N-th relative to the constant phase compensating factor of reference channelThen by separate for each two transmitting Passage and receive passage and be equivalent to the passage of the n-th internal loopback, the most correspondingly received echo-signal is also converted into n-th etc. The echo-signal of effect phase center passage, the adjacent distance between displaced phase center is d, and then respectively obtains N number of equivalence The echo-signal of phase center passage Represent The echo-signal of the n-th displaced phase center passage,

Wherein, d_nRepresent the distance between displaced phase center and the displaced phase center of reference channel of the n-th passage, d_n=(n-1) d, n ∈ 1,2 ..., N}, N represent the radar in the radar motion geometric model of hypersonic aircraft descending branch The antenna channels number comprised, r represents radar distance between platform and moving target P, and γ represents that radar is at platform The angle of descent of motion, λ represents the echo-signal wavelength that each channel reception arrives, t_rRepresent that distance is to fast time, t_aRepresent that orientation is slow Time.

Step 3, obtains the echo-signal of N number of displaced phase center passageDistance frequency domain after Range compress- Orientation time-domain representation form Represent n-th The echo-signal of individual displaced phase center passageDistance frequency domain after Range compress-orientation time-domain representation form, and Echo-signal to N number of displaced phase center passageDistance frequency domain after Range compress-orientation time-domain representation formCarry out phase compensation, the distance frequency domain of N number of displaced phase center passage-orientation time domain echo letter after obtaining phase compensation Number Represent after phase compensation n-th etc. Distance frequency domain-orientation time domain the echo-signal of effect phase center passage, f_rRepresent frequency of distance, t_aRepresent orientation slow time, t_rTable Show that distance is to the fast time.

The concrete sub-step of step 3 is:

3a) set the conjugate function s of transmitting signal copy_r(t_r), s_r(t_r)=W_r(t_r)·exp[-jπμt_r ²], W_r(·) Expression distance is to rectangular pulse window function, and μ represents the frequency modulation rate launching linear frequency modulation (LFM) signal, t_rRepresent that distance is when fast Between, and by the echo-signal of N number of displaced phase center passageIt is arranged together in N₁×N₂Dimension matrix, obtains N number of N₁×N₂ Dimension matrix, N₁Represent that distance that the echo-signal of each displaced phase center passage comprises is to sampling number, N₂Represent each equivalence The orientation that the echo-signal of phase center passage comprises is to sampling number；Then to N number of N₁×N₂Every a line of dimension matrix is entered respectively Row fast Fourier transform (FFT) is multiplied by, after processing, the conjugate function s launching signal copy again_r(t_r), and then obtain N number of equivalence The echo-signal of phase center passageDistance frequency domain after Range compress-orientation time-domain representation form Represent the echo of the n-th displaced phase center passage SignalDistance frequency domain after Range compress-orientation time-domain representation form, f_rRepresent frequency of distance, t_aRepresent that orientation is slow Time.

Specifically, in the echo-signal of described N number of displaced phase center passageDistance frequency after Range compress Territory-orientation time-domain representation formIn,Represent the echo-signal of the n-th displaced phase center passageDistance frequency domain after Range compress-orientation time-domain representation form, it obtains process and is:

Wherein, n ∈ 1,2 ..., and N}, represent dot product, FFT represents that fast Fourier transform operates, f_rRepresent distance frequency Rate, t_aRepresent orientation slow time, t_rRepresent distance to the fast time,Represent the echo of the n-th displaced phase center passage Signal.

Represent the echo-signal of the n-th displaced phase center passageDistance frequency after Range compress Territory-orientation time-domain representation form, its expression formula is:

$\begin{array}{c}{\hat{s}}_n(f_r,t_a)=W_r\left(f_r\right)W_a(t_a-t_c)\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}(R_0+z_n\cos \mathrm{\φ}\left)\right)\\ \×\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}\frac{(-V_y\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})}{V_x-v_{xp}}(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\left)\right)\\ \×\exp (-j2\mathrm{\π}\frac{(f_c+f_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}\cos \mathrm{\φ}+V_y\sin \mathrm{\φ}\right)}^2)}{{\mathrm{cR}}_0{(V_x-v_{xp})}^2}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_x\right)+x_n\right)}^2)\\ \×\exp \left(j2\mathrm{\π}\frac{(f_c+f_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}\cos \mathrm{\φ}+V_y\sin \mathrm{\φ}\right)}^2)(-V_z\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})}{{\mathrm{cR}}_0^2{(V_x-v_{xp})}^3}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^3\right)\end{array}$

Wherein, f_cRepresent the carrier frequency of passage echo-signal, W_a() represents that orientation is to window function, W_r() represent distance to Rectangular pulse window function, f_rRepresent frequency of distance, t_aRepresent orientation slow time, t_cRepresent that the n-th passage is between moving target P Distance is time required during nearest oblique distance, R₀Represent the radar nearest oblique distance between platform and moving target P,H represents the radar vertical height at platform, and φ represents that radar is under platform Visual angle, x_nRepresent 0 moment the n-th passage coordinate in X-axis, z_nRepresent 0 moment the n-th passage coordinate on Z axis, V_xRepresent thunder Reach the horizontal velocity component of place platform, V_zRepresent the radar vertical velocity component at platform, y_pRepresent 0 moment moving target P Coordinate in Y-axis, v_ypRepresent the horizontal component of moving target P speed, v_xpRepresent that the horizontal velocity of 0 moment moving target P is divided Amount, andC represents the light velocity, c=3 × 10⁸(m/s)。

3b) the echo-signal to N number of displaced phase center passageDuring distance frequency domain-orientation after Range compress Domain representation formCarry out phase compensation, the distance frequency domain-orientation of N number of displaced phase center passage after obtaining phase compensation Time domain echo-signal Represent phase compensation The distance frequency domain of rear n-th displaced phase center passage-orientation time domain echo-signal, f_rRepresent frequency of distance, t_aRepresent that orientation is slow Time.

Specifically, the distance frequency domain of N number of displaced phase center passage-orientation time domain echo-signal after described phase compensationIn,After representing phase compensation, the distance frequency domain-orientation time domain echo of the n-th displaced phase center passage is believed Number, it obtains process and is:

Set the quadratic term phase compensation function h of the n-th displaced phase center passage the most respectively_n1With the n-th equivalence phase The cubic term phase compensation function h of position central passage_n2, its expression formula is respectively as follows:

$h_{n1}=\exp \left(j2\mathrm{\π}\frac{(f_c+f_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}cos\mathrm{\φ}+V_{z}sin\mathrm{\φ}\right)}^2)}{{\mathrm{cR}}_0{(V_x-v_{xp})}^2}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^2\right)$

$h_{n2}=\exp (-j2\mathrm{\π}\frac{(f_c+f_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}cos\mathrm{\φ}+V_{z}sin\mathrm{\φ}\right)}^2)(-V_{z}cos\mathrm{\φ}+v_{yp}sin\mathrm{\φ})}{{\mathrm{cR}}_0^2{(V_x-v_{xp})}^3}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^3)$

Then by the echo-signal of the n-th displaced phase center passageDistance frequency domain-side after Range compress Position time-domain representation formQuadratic term phase compensation function h with the n-th displaced phase center passage set_n1With The cubic term phase compensation function h of the n-th displaced phase center passage set_n2Carry out dot product successively,

I.e.And then it is calculated the distance frequency of the n-th displaced phase center passage after phase compensation Territory-orientation time domain echo-signalIts expression formula is:

$\begin{array}{c}{\hat{s}}_n(f_r,t_a)=W_r\left(f_r\right)W_a(t_a-t_c)\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}(R_0+z_n\cos \mathrm{\φ}\left)\right)\\ \×\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}\frac{(-V_z\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})}{V_x-v_{xp}}(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\left)\right)\end{array}$

Wherein, R₀Represent the radar nearest oblique distance between platform and moving target P, f_cRepresent passage echo-signal Carrier frequency, W_r() represents that distance is to rectangular pulse window function, W_a() represents that orientation is to window function, f_rRepresent frequency of distance, t_aTable Show orientation slow time, t_cRepresent the n-th passage to the distance between moving target P be nearest oblique distance time required time, c represents The light velocity, V_xRepresent the radar horizontal velocity component at platform, v_xpRepresent the horizontal velocity component of 0 moment moving target P, v_ypTable Showing the vertical velocity component of 0 moment moving target P, φ represents the radar downwards angle of visibility at platform, V_zRepresent that radar is at platform Vertical velocity component, z_nRepresent the vertical distance between the n-th passage and initial point, x_nRepresent the level between the n-th passage and initial point Distance, n ∈ 1,2 ..., N}, N represent what the radar in the radar motion geometric model of hypersonic aircraft descending branch comprised Antenna channels number；After can be seen that phase compensation, the distance frequency domain-orientation time domain echo of the n-th displaced phase center passage is believed NumberIt is the distance frequency domain-orientation time domain echo-signal of the n-th passage after three phase compensation, no longer contains secondary and three Secondary exponential term, this processes for follow-up clutter recognition and provides conveniently.

Step 4, according to the distance frequency domain-orientation time domain echo-signal of displaced phase center passage N number of after phase compensationIt is calculated N number of displaced phase center passage echo-signal s (t of range-Dopler domain successively_r,f_a), distance how general Strangle the echo-signal st (t of the moving target P in territory_r,f_a) and the noise signal sc (t of range-Dopler domain_r,f_a), then according to such as The weight vector w of lower equations space-time adaptive clutter recognition:

$\left\{\begin{array}{cc}\underset{w}{\min }& w^{H}Rw\\ s.t.& w^{H}a\left(f_a\right)=1\end{array}\right.$

Meet the optimum weight coefficient vector w that weight vector is space-time adaptive clutter recognition of above-mentioned formula_opt, and then obtain Echo-signal s of moving target P after clutter recognition_P(t_r,f_a)；Wherein,The value of w, s.t. table when expression obtains minima Show that constraints, R represent N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a) in Doppler frequency The covariance matrix of the multichannel output correspondence of unit, a (f_a) represent the steering vector matrix of N number of displaced phase center passage, f_a Represent the Doppler frequency of moving target P.

The sub-step of step 4 is:

4a) distance frequency domain-orientation time domain the echo-signal to displaced phase center passage N number of after phase compensation's Every a line carries out inverse fast fourier transform (IFFT) respectively, every string carries out fast Fourier transform more respectively simultaneously (FFT) N number of displaced phase center passage echo-signal s (t of range-Dopler domain, it is calculated_r,f_a),

s(t_r,f_a)={ s₁(t_r,f_a),…,s_n(t_r,f_a),…,s_N(t_r,f_a), s_n(t_r,f_a) represent range-Dopler domain The n-th displaced phase center passage echo-signal, its expression formula is:

$s_n(t_r,f_a)={\mathrm{FFT}}_{t_a}\{{\mathrm{IFFT}}_{t_r}\[{\tilde{s}}_n(f_r,t_a)\]\}$

Wherein, f_rRepresent frequency of distance, f_aRepresent the Doppler frequency of moving target P, t_rRepresent that distance is to fast time, n ∈ 1,2 ..., N}, N represent the antenna channels that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises Number, t_aRepresenting the orientation slow time, FFT represents that fast Fourier transform operates, and IFFT represents that inverse fast fourier transform operates,Distance frequency domain-orientation time domain the echo-signal of N number of displaced phase center passage after expression phase compensation.

4b) according to N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a), it is calculated respectively Echo-signal st (the t of the moving target P of range-Dopler domain_r,f_a) and the noise signal sc (t of range-Dopler domain_r,f_a), its Expression formula is respectively as follows:

$\begin{array}{c}st(t_r,f_a)=\sin c\[B_r(t_r-\frac{2}{c}(R_0+\frac{\mathrm{\λ}}{4K_{at}}{f}_a^2-\frac{\mathrm{\λ}}{4K_{at}}{f}_{dct}^2\left)\right)\]W_a\left(f_a\right)\×\\ \exp (-j\frac{4\mathrm{\π}}{\mathrm{\λ}}{R}_0)\exp \left(j\frac{\mathrm{\π}}{K_{at}}{\left(f_a+f_{dct}\right)}^2\right)\exp (-j2{\mathrm{\πf}}_a{t}_c)\end{array}$

$\begin{array}{c}sc(t_r,f_a)=\sin c\[B_r(t-\frac{2}{c}(R_0+\frac{\mathrm{\λ}}{4K_{ac}}{f}_a^2-\frac{\mathrm{\λ}}{4K_{ac}}{f}_{dcc}^2\left)\right)\]W_a\left(f_a\right)\×\\ \exp (-j\frac{4\mathrm{\π}}{\mathrm{\λ}}{R}_0)\exp \left(j\frac{\mathrm{\π}}{K_{ac}}{f}_a^2\right)\exp \left(j\frac{2{\mathrm{\πf}}_{dcc}}{K_{ac}}{f}_a\right)\exp \left(j\frac{{\mathrm{\πf}}_{dcc}^2}{K_{ac}}\right)\end{array}$

Wherein, B_rRepresent that the distance of linear frequency modulation (LFM) signal is to bandwidth, t_rRepresent that distance is to fast time, sinc [] table Show Sinc function, f_aRepresenting the Doppler frequency of moving target P, c represents the light velocity, t_cRepresent the n-th passage to moving target P it Between distance time required when being nearest oblique distance, R₀Represent the radar nearest oblique distance between platform and moving target P, λ Represent the echo-signal wavelength that each channel reception arrives, f_dccRepresent the doppler centroid of noise signal,W_a() represents that orientation is to window function, K_acRepresent the doppler frequency rate of noise signal,f_dctRepresent the doppler centroid of moving target P, K_atRepresent the doppler frequency rate of moving target P,V_xRepresent radar The horizontal velocity component of place platform, v_xpRepresent the horizontal velocity component of 0 moment moving target P, v_ypRepresent 0 moment motion mesh The vertical velocity component of mark P, φ represents the radar downwards angle of visibility at platform, V_zRepresent that radar divides at the vertical velocity of platform Amount.

N number of displaced phase center passage echo-signal s (t of described range-Dopler domain_r,f_a), for range-Dopler domain Echo-signal st (the t of moving target P_r,f_a) and the noise signal sc (t of range-Dopler domain_r,f_a) sum.

4c) based on maximum output signal-to-noise ratio criterion, and solve the power arrow of space-time adaptive clutter recognition according to equation below Amount w:

$\left\{\begin{array}{cc}\underset{w}{\min }& w^{H}Rw\\ s.t.& w^{H}a\left(f_a\right)=1\end{array}\right.$

Meet the optimum weight coefficient vector w that weight vector is space-time adaptive clutter recognition of above-mentioned formula_opt, its expression formula For:

$w_{opt}={\[w_1,...,w_n,...,w_N\]}^T=\frac{R^{-1}a\left(f_a\right)}{a^H\left(f_a\right)R^{-1}a\left(f_a\right)}$

Wherein, w_nRepresent the space-time adaptive clutter recognition optimum weight coefficient vector of the n-th displaced phase center passage, R Represent N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a) at the multichannel of Doppler frequency unit The covariance matrix that output is corresponding, and R=E{s (t_r,f_a)s^H(t_r,f_a), s (t_r,f_a) represent the N number of etc. of range-Dopler domain Effect phase center passage echo-signal, ()^-1The inverse matrix represented, []^HThe conjugate transpose represented, []^TRepresent Transposition, represent dot product, a (f_a) represent the steering vector matrix of N number of displaced phase center passage, n ∈ 1,2 ..., N}, N Represent antenna channels number f that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises_aRepresent fortune The Doppler frequency of moving-target P, t_rRepresent that distance is to fast time, ()^-1The inverse matrix represented, E{ } represent the behaviour that averages Make, Represent echo-signal and ginseng that the n-th displaced phase center channel reception arrives Examining the phase contrast between passage, its expression formula is:

f_aTable Showing the Doppler frequency of moving target P, γ represents the angle of descent that radar moves, V at platform_xRepresent the radar water at platform Flat velocity component, v_xpRepresent the horizontal velocity component of 0 moment moving target P, d_nRepresent the displaced phase center of the n-th passage with Distance between the displaced phase center of reference channel, v_ypRepresenting the vertical velocity component of 0 moment moving target P, φ represents thunder Reach the downwards angle of visibility of place platform, V_zRepresenting the radar vertical velocity component at platform, λ represents the echo that each channel reception arrives Signal wavelength.

4d) according to N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a) and space-time adaptive miscellaneous The optimum weight coefficient vector w of ripple suppression_opt, it is calculated echo-signal s of moving target P after clutter recognition_P(t_r,f_a)；Wherein, f_aRepresent the Doppler frequency of moving target P, t_rRepresent that distance is to the fast time.

Specifically, the passage echo-signal s (t of range-Dopler domain is extracted_r,f_a) same distance is to, identical doppler cells But the passage echo-signal of different antennae passage, and by it with this distance to the space-time adaptive clutter recognition of respective channel Excellent weight coefficient vector carries out dot product, thus completes space domain self-adapted clutter recognition, obtains the echo of moving target P after clutter recognition Signal s_P(t_r,f_a), its expression formula is:

s_P(t_r,f_a)=s (t_r,f_a)·w_opt,

Wherein, s (t_r,f_a) represent range-Dopler domain N number of displaced phase center passage echo-signal, f_aRepresent motion The Doppler frequency of target P, t_rRepresent that distance is to fast time, w_optRepresent that the optimum weight coefficient of space-time adaptive clutter recognition is vowed Amount；Obtain echo-signal s of moving target P after clutter recognition_P(t_r,f_aAfter), just can farthest extract moving target The echo-signal of P, and the noise signal amplitude beyond the echo-signal of moving target P is greatly reduced even 0, thus filter Noise signal, retains the echo-signal of moving target P.

The effect of the present invention can be described further by following emulation experiment:

(1) simulation parameter

The carrier frequency of passage echo-signal is 5.405GHz, launches a width of 50MHz of band of linear frequency modulation (LFM) signal, pulse Width is 10us, and pulse recurrence frequency is 3000Hz, and radar antenna port number is 3, and adjacency channel spacing is 1m, and radar exists Horizontal velocity component V of platform_xFor 3000m/s, radar is at the vertical velocity component V of platform_yFor 300m/s, radar is flat Vertical height H of platform is 30km, and the orientation of moving target P is 10m/s to speed, and the distance of moving target P is 20m/ to speed s。

(2) emulation content

Emulation content 1, uses the inventive method that radar return carries out Range compress, and in distance frequency domain-orientation time domain Use the range Doppler result figure after three phase compensation, with reference to Fig. 3, for carrying out in distance-Doppler territory after Range compress Single channel amplitude result figure after three phase compensation；Wherein transverse axis represents that distance represents Doppler sample to sampled point, the longitudinal axis Point；The Range compress result of moving target is that middle oblique line, and the compression image that remaining 4 oblique line is clutter.

Emulation content 2, the image result figure after using the inventive method that the above results carries out clutter recognition further, ginseng According to Fig. 4, it is three and channel combined carries out the amplitude result figure after clutter recognition；Wherein, transverse axis represents that distance is to sampled point, the longitudinal axis Represent Doppler sample point.

(3) analysis of simulation result

From figure 3, it can be seen that Moving Target Return and clutter occupy different Doppler spreads respectively, thus it is follow-up Clutter recognition processes and provides theoretical basis.

Figure 4, it is seen that after using the inventive method to carry out clutter recognition, clutter can be curbed well, only Retain the result of moving target.

In sum, emulation experiment demonstrates the correctness of the present invention, validity and reliability.

Obviously, those skilled in the art can carry out various change and the modification essence without deviating from the present invention to the present invention God and scope；So, if these amendments of the present invention and modification belong to the scope of the claims in the present invention and equivalent technologies thereof Within, then the present invention is also intended to comprise these change and modification.

Claims (10)

1. the SAR-GMTI clutter suppression method of a hypersonic aircraft descending branch, it is characterised in that comprise the following steps:

Step 1, sets up the radar motion geometric model of hypersonic aircraft descending branch, declines at described hypersonic aircraft In the radar motion geometric model of section, radar comprises N number of antenna channels, and P is radar any one moving target in the scene, And the instantaneous oblique distance between the n-th antenna channels and moving target P is expressed as R_n(t_a)；Wherein, n ∈ 1,2 ..., N}, N table Showing the antenna channels number that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises, N is more than 1 Odd number, t_aRepresent the orientation slow time；

Step 2, the N number of antenna channels comprised by radar is designated as the 1st passage successively to N channel, and using the 1st passage as reference Passage, by (N+1)/2 channel emission linear FM signal, uses N number of antenna channels to receive moving target P place scene simultaneously Echo-signal, obtain the echo-signal s (t of N number of passage_r,t_a), s (t_r,t_a)={ s₁(t_r,t_a),…,s_n(t_r,t_a),…,s_N (t_r,t_a), s_n(t_r,t_a) represent the echo-signal of the n-th passage, t_rRepresent that distance is to fast time, t_aRepresent the orientation slow time；

Then set the constant phase compensating factor relative to reference channel, and the echo-signal of N number of passage is carried out perseverance respectively Phase bit compensates, and obtains the echo-signal of N number of displaced phase center passage Represent the echo-signal of the n-th displaced phase center passage；

Step 3, obtains the echo-signal of N number of displaced phase center passageDistance frequency domain-orientation after Range compress Time-domain representation form Expression n-th etc. The echo-signal of effect phase center passageDistance frequency domain after Range compress-orientation time-domain representation form；

And the echo-signal to N number of displaced phase center passageDistance frequency domain after Range compress-orientation time domain table Show form s (f_r,t_a) carry out phase compensation, after obtaining phase compensation during the distance frequency domain-orientation of N number of displaced phase center passage Territory echo-signal After representing phase compensation The distance frequency domain of the n-th displaced phase center passage-orientation time domain echo-signal, f_rRepresent frequency of distance, t_aRepresent when orientation is slow Between, t_rRepresent that distance is to the fast time；

Step 4, according to the distance frequency domain-orientation time domain echo-signal of displaced phase center passage N number of after phase compensation It is calculated N number of displaced phase center passage echo-signal s (t of range-Dopler domain successively_r,f_a), the fortune of range-Dopler domain Echo-signal st (the t of moving-target P_r,f_a) and the noise signal sc (t of range-Dopler domain_r,f_a), then ask according to equation below The weight vector w of solution space-time adaptive clutter recognition:

$\left\{\begin{array}{cc}\underset{w}{min}& w^{H}Rw\\ s.t.& w^{H}a\left(f_a\right)=1\end{array}\right.$

Meet the optimum weight coefficient vector w that weight vector is space-time adaptive clutter recognition of above-mentioned formula_opt, and then obtain clutter Echo-signal s of moving target P after suppression_P(t_r,f_a)；Wherein,The value of w when expression obtains minima, s.t. represents about Bundle condition, R represents N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a) at Doppler frequency unit Covariance matrix corresponding to multichannel output, a (f_a) represent the steering vector matrix of N number of displaced phase center passage, f_aRepresent The Doppler frequency of moving target P.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 1, its feature It is, in step 1, the radar motion geometric model of described hypersonic aircraft descending branch, particularly as follows:

In three-dimensional system of coordinate XYZ, in the horizontal direction, on the right side of Y-axis sensing, Z axis is away from direction, the earth's core, and radar is at platform for X-axis Vertical height be H, and radar platform is made up of three antenna channels arranged along flight path dimension linear, adjacent two antennas it Between distance be 2d；Within a relevant treatment time, radar (refers to radar speed direction at platform with constant angle of descent γ And the angle between X-axis) diagonally under direction do the linear uniform motion that constant speed is V, radar regards at the lower of platform Angle is φ, and radar is R in the nearest oblique distance of platform to moving target P₀, then radar is V in the horizontal velocity component of platform_x, Radar is V at the vertical velocity component of platform_z；P is radar any one moving target in the scene, 0 moment motion mesh Mark P is at (x_p,y_p) place, and the horizontal velocity component of 0 moment moving target P is v_xp, the vertical speed of 0 moment moving target P is divided Amount is v_yp, radar is R in the nearest oblique distance of platform to moving target P₀, W_gBand for moving target P place observation scene Width, and in the case of radar is operated in positive side-looking.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 1, its feature It is, in step 1, described instantaneous oblique distance between n-th antenna channels and moving target P is expressed as R_n(t_a):

$\begin{array}{c}{R}_n\left(t_a\right)=\sqrt{{(V_x{t}_a+x_n-x_p-v_{xp}{t}_a)}^2+{(H-V_z{t}_a+z_n)}^2+{(y_p+v_{yp}{t}_a)}^2}\\ \≈r+(-V_z\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})(t_a-t_c)+\frac{{(V_x-v_{xp})}^2+{(v_{yp}\cos \mathrm{\φ}+V_z\sin \mathrm{\φ})}^2}{2r}{(t_a-t_c)}^2+\\ z_n\cos \mathrm{\φ}-(\frac{V_z{z}_n{\sin }^2\mathrm{\φ}}{r}+\frac{v_{yp}{z}_n\sin \mathrm{\φ}\cos \mathrm{\φ}}{r})(t_a-t_c)\end{array}$

Wherein, V_xRepresent the radar horizontal velocity component at platform, t_aRepresent orientation slow time, t_cRepresent that the n-th passage is to motion The time that distance between target P is required when being nearest oblique distance, x_pRepresent the 0 moment moving target P coordinate in X-axis, v_xpRepresent 0 The horizontal velocity component of moment moving target P, H represents the radar vertical height at platform, V_zRepresent that radar is at platform Vertical velocity component, y_pRepresent the 0 moment moving target P coordinate in Y-axis, v_ypRepresent that the vertical speed of 0 moment moving target P is divided Amount,R represents radar distance between platform and moving target P, and φ represents that radar regards at the lower of platform Angle, n ∈ 1,2 ..., N}, N represent the sky that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises Line channel number, x_nRepresent the horizontal range between the n-th antenna channels and initial point, z_nRepresent the n-th antenna channels and initial point it Between vertical distance.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 1, its feature It is, in step 2, describedRepresenting the echo-signal of the n-th displaced phase center passage, it obtains process and is: set Fixed n-th relative to the constant phase compensating factor of reference channelThen n-th it is calculated The echo-signal of displaced phase center passageIts expression formula is:

Wherein, d_nRepresent the distance between displaced phase center and the displaced phase center of reference channel of the n-th passage, n ∈ 1,2 ..., N}, N represent the antenna channels that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises Number, r represents radar distance between platform and moving target P, and γ represents the angle of descent that radar moves, λ at platform Represent the echo-signal wavelength that each channel reception arrives, t_rRepresent that distance is to fast time, t_aRepresent the orientation slow time.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 1, its feature Being, the sub-step of step 3 is:

3a) set the conjugate function s of transmitting signal copy_r(t_r), and by the echo-signal of N number of displaced phase center passageIt is arranged together in N₁×N₂Dimension matrix, obtains N number of N₁×N₂Dimension matrix, N₁Represent returning of each displaced phase center passage The distance that ripple signal packet contains is to sampling number, N₂Represent that orientation that the echo-signal of each displaced phase center passage comprises is to adopting Number of samples；Then to N number of N₁×N₂Every a line of dimension matrix is multiplied by transmitting signal after carrying out fast Fourier transform process respectively again The conjugate function s of copy_r(t_r), and then obtain the echo-signal of N number of displaced phase center passageAfter Range compress Distance frequency domain-orientation time-domain representation form Represent the echo-signal of the n-th displaced phase center passageDuring distance frequency domain-orientation after Range compress Domain representation form, t_rRepresent that distance is to fast time, f_rRepresent frequency of distance, t_aRepresent the orientation slow time；

3b) the echo-signal to N number of displaced phase center passageDistance frequency domain after Range compress-orientation time domain table Show formCarry out phase compensation, the distance frequency domain-orientation time domain of N number of displaced phase center passage after obtaining phase compensation Echo-signal Represent after phase compensation the The distance frequency domain of n displaced phase center passage-orientation time domain echo-signal, f_rRepresent frequency of distance, t_aRepresent when orientation is slow Between.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 5, its feature It is, the conjugate function s of described transmitting signal copy_r(t_r), its expression formula is: s_r(t_r)=W_r(t_r)·exp[-jπμt_r ²], t_r Represent that distance is to fast time, W_r() expression distance is to rectangular pulse window function, and μ represents the frequency modulation launching linear FM signal Rate；

Echo-signal at described N number of displaced phase center passageDistance frequency domain-orientation time domain after Range compress RepresentationIn,Represent the echo-signal of the n-th displaced phase center passageThrough Range compress After distance frequency domain-orientation time-domain representation form, it obtains process and is:

Wherein, n ∈ 1,2 ..., and N}, represent dot product, FFT represents that fast Fourier transform operates, f_rRepresent frequency of distance, t_a Represent orientation slow time, t_rRepresent distance to the fast time,Represent the echo-signal of the n-th displaced phase center passage.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 5, its feature It is, describedRepresent the echo-signal of the n-th displaced phase center passageDistance after Range compress Frequency domain-orientation time-domain representation form, its expression formula is:

$\begin{array}{c}{\hat{s}}_n(f_r,t_a)=W_r\left(f_r\right)W_a(t_a-t_c)\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}(R_0+z_n\cos \mathrm{\φ}\left)\right)\\ \×\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}\frac{(-V_y\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})}{V_x-v_{xp}}(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\left)\right)\\ \×\exp (-j2\mathrm{\π}\frac{(f_c+r_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}\cos \mathrm{\φ}+V_y\sin \mathrm{\φ}\right)}^2)}{{\mathrm{cR}}_0{(V_x-v_{xp})}^2}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^2)\\ \×\exp (-j2\mathrm{\π}\frac{(f_c+r_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}\cos \mathrm{\φ}+V_y\sin \mathrm{\φ}\right)}^2)(-V_z\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})}{{\mathrm{cR}}_0^2{(V_x-v_{xp})}^3}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^3)\end{array}$

Wherein, f_cRepresent the carrier frequency of passage echo-signal, W_a() represents that orientation is to window function, W_r() represents that distance is to rectangle Pulse window function, f_rRepresent frequency of distance, t_aRepresent orientation slow time, t_cRepresent that the n-th passage is to the distance between moving target P For time required during nearest oblique distance, R₀Represent the radar nearest oblique distance between platform and moving target P,H represents the radar vertical height at platform, and φ represents that radar is under platform Visual angle, x_nRepresent 0 moment the n-th passage coordinate in X-axis, z_nRepresent 0 moment the n-th passage coordinate on Z axis, V_xRepresent thunder Reach the horizontal velocity component of place platform, V_zRepresent the radar vertical velocity component at platform, y_pRepresent 0 moment moving target P Coordinate in Y-axis, v_ypRepresent the horizontal component of moving target P speed, v_xpRepresent that the horizontal velocity of 0 moment moving target P is divided Amount, andC represents the light velocity.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 5, its feature It is, the distance frequency domain of N number of displaced phase center passage-orientation time domain echo-signal after described phase compensationIn,Distance frequency domain-orientation time domain the echo-signal of the n-th displaced phase center passage after expression phase compensation, it obtains Process is:

Set the quadratic term phase compensation function h of the n-th displaced phase center passage the most respectively_n1With in the n-th equivalent phase The cubic term phase compensation function h of heart passage_n2, and according to the distance frequency domain of the n-th displaced phase center passage after Range compress- Orientation time domain echo-signalThen by the echo-signal of the n-th displaced phase center passageThrough distance pressure Distance frequency domain after contracting-orientation time-domain representation formQuadratic term with the n-th displaced phase center passage set Phase compensation function h_n1Cubic term phase compensation function h with the n-th displaced phase center passage set_n2Carry out a little successively Take advantage of, i.e.And then it is calculated the distance frequency domain-orientation of the n-th displaced phase center passage after phase compensation Time domain echo-signalIts expression formula is respectively as follows:

$h_{n1}=\exp \left(j2\mathrm{\π}\frac{(f_c+f_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}cos\mathrm{\φ}+V_{z}sin\mathrm{\φ}\right)}^2)}{{\mathrm{cR}}_0{(V_x-v_{xp})}^2}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^2\right)$

$h_{n2}=\exp (-j2\mathrm{\π}\frac{(f_c+f_r)({\left(V_x-v_{xp}\right)}^2+{\left(v_{yp}cos\mathrm{\φ}+V_{z}sin\mathrm{\φ}\right)}^2)(-V_{z}cos\mathrm{\φ}+v_{yp}sin\mathrm{\φ})}{{\mathrm{cR}}_0^2{(V_x-v_{xp})}^3}{\left(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\right)}^3)$

$\begin{array}{c}{\hat{s}}_n(f_r,t_a)=W_r\left(f_r\right)W_a(t_a-t_c)\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}(R_0+z_n\cos \mathrm{\φ}\left)\right)\\ \×\exp (-j4\mathrm{\π}\frac{f_c+f_r}{c}\frac{(-V_z\cos \mathrm{\φ}+v_{yp}\sin \mathrm{\φ})}{V_x-v_{xp}}(\left(V_x-v_{xp}\right)\left(t_a-t_c\right)+x_n\left)\right)\end{array}$

Wherein, R₀Represent the radar nearest oblique distance between platform and moving target P, f_cRepresent the carrier frequency of passage echo-signal, W_r() represents that distance is to rectangular pulse window function, W_a() represents that orientation is to window function, f_rRepresent frequency of distance, t_aExpression side The position slow time, t_cRepresent the n-th passage to the distance between moving target P be nearest oblique distance time required time, c represents light Speed, V_xRepresent the radar horizontal velocity component at platform, v_xpRepresent the horizontal velocity component of 0 moment moving target P, v_ypRepresent The vertical velocity component of 0 moment moving target P, φ represents the radar downwards angle of visibility at platform, V_zRepresent that radar is at platform Vertical velocity component, z_nRepresent the vertical distance between the n-th passage and initial point, x_nRepresent level between the n-th passage and initial point away from From, n ∈ 1,2 ..., N}, N represent the sky that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises Line channel number.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 1, its feature Being, the sub-step of step 4 is:

4a) distance frequency domain-orientation time domain the echo-signal to displaced phase center passage N number of after phase compensationEach Row carries out inverse fast fourier transform respectively, every string carries out fast Fourier transform (FFT) more respectively simultaneously, is calculated N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a), s (t_r,f_a)={ s₁(t_r,f_a),…,s_n(t_r, f_a),…,s_N(t_r,f_a), s_n(t_r,f_a) represent range-Dopler domain the n-th displaced phase center passage echo-signal；

4b) according to N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a), it is calculated distance respectively Echo-signal st (the t of the moving target P of Doppler domain_r,f_a) and the noise signal sc (t of range-Dopler domain_r,f_a)；

The weight vector w of space-time adaptive clutter recognition 4c) is solved according to equation below:

$\left\{\begin{array}{cc}\underset{w}{\min }& w^{H}Rw\\ s.t.& w^{H}a\left(f_a\right)=1\end{array}\right.$

Meet the optimum weight coefficient vector w that weight vector is space-time adaptive clutter recognition of above-mentioned formula_opt；

4d) according to N number of displaced phase center passage echo-signal s (t of range-Dopler domain_r,f_a) and space-time adaptive clutter press down The optimum weight coefficient vector w of system_opt, it is calculated echo-signal s of moving target P after clutter recognition_P(t_r,f_a)；Wherein, f_aTable Show the Doppler frequency of moving target P, t_rRepresent that distance is to the fast time.

The SAR-GMTI clutter suppression method of a kind of hypersonic aircraft descending branch the most as claimed in claim 9, its feature It is, described s_n(t_r,f_a) represent that how general the n-th displaced phase center passage echo-signal of range-Dopler domain, described distance be Strangle the echo-signal st (t of the moving target P in territory_r,f_a), the noise signal sc (t of described range-Dopler domain_r,f_a), described sky Time self-adapting clutter suppression optimum weight coefficient vector w_optWith echo-signal s of moving target P after described clutter recognition_P(t_r, f_a), its expression formula is respectively as follows:

$s_n(t_r,f_a)={\mathrm{FF}}_{t_a}\{{\mathrm{IFFT}}_{t_r}\[{\tilde{s}}_n(f_r,t_a)\]\}$

$\begin{array}{c}st(t_r,f_a)=\sin c\[B_r(t_r-\frac{2}{c}(R_0+\frac{\mathrm{\λ}}{4K_{at}}{f}_a^2-\frac{\mathrm{\λ}}{4K_{at}}{f}_{dct}^2\left)\right)\]W_a\left(f_a\right)\×\\ \exp (-j\frac{4\mathrm{\π}}{\mathrm{\λ}}{R}_0)\exp \left(j\frac{\mathrm{\π}}{K_{at}}{\left(f_a+f_{dct}\right)}^2\right)\exp (-j2{\mathrm{\πf}}_a{t}_c)\end{array}$

$\begin{array}{c}sc(t_r,f_a)=\sin c\[B_r(t-\frac{2}{c}(R_0+\frac{\mathrm{\λ}}{4K_{ac}}{f}_a^2-\frac{\mathrm{\λ}}{4K_{ac}}{f}_{dcc}^2\left)\right)\]W_a\left(f_a\right)\×\\ \exp (-j\frac{4\mathrm{\π}}{\mathrm{\λ}}{R}_0)\exp \left(j\frac{\mathrm{\π}}{K_{ac}}{f}_a^2\right)\exp \left(j\frac{2{\mathrm{\πf}}_{dcc}}{K_{ac}}{f}_a\right)\exp \left(j\frac{{\mathrm{\πf}}_{dcc}^2}{K_{ac}}\right)\end{array}$

$w_{opt}={\[w_1,...,w_n,...,w_N\]}^T=\frac{R^{-1}a\left(f_a\right)}{a^H\left(f_a\right)R^{-1}a\left(f_a\right)}$

s_P(t_r,f_a)=s (t_r,f_a)·w_opt

Wherein, f_rRepresent frequency of distance, f_aRepresent the Doppler frequency of moving target P, t_rRepresent distance to the fast time, n ∈ 1, 2 ..., N}, N represent the antenna channels that the radar in the radar motion geometric model of hypersonic aircraft descending branch comprises Number, t_aRepresenting the orientation slow time, FFT represents that fast Fourier transform operates, and IFFT represents that inverse fast fourier transform operates,Distance frequency domain-orientation time domain the echo-signal of N number of displaced phase center passage, B after expression phase compensation_rRepresent linear The distance of FM signal represents Sinc function to bandwidth, sinc [], and c represents the light velocity, t_cRepresent the n-th passage to moving target The time that distance between P is required when being nearest oblique distance, R₀Represent radar between platform and moving target P recently tiltedly Away from, λ represents the echo-signal wavelength that each channel reception arrives, f_dccRepresent the doppler centroid of noise signal,W_a() represents that orientation is to window function, K_acRepresent the doppler frequency rate of noise signal,f_dctRepresent the doppler centroid of moving target P, K_atRepresent the doppler frequency rate of moving target P,V_xRepresent radar The horizontal velocity component of place platform, v_xpRepresent the horizontal velocity component of 0 moment moving target P, v_ypRepresent 0 moment motion mesh The vertical velocity component of mark P, φ represents the radar downwards angle of visibility at platform, V_zRepresent that radar divides at the vertical velocity of platform Amount, w_nRepresenting the space-time adaptive clutter recognition optimum weight coefficient vector of the n-th displaced phase center passage, R represents that distance is many N number of displaced phase center passage echo-signal s (t in general Le territory_r,f_a) export at the multichannel of Doppler frequency unit corresponding Covariance matrix, ()^-1The inverse matrix represented, []^HThe conjugate transpose represented, []^TThe transposition represented, represents Dot product, a (f_a) represent the steering vector matrix of N number of displaced phase center passage, Represent the phase contrast between echo-signal and the reference channel that the n-th displaced phase center channel reception arrives, s (t_r,f_a) represent away from N number of displaced phase center passage echo-signal from Doppler domain.