CN104698431A - Method for estimating fussy component space angle and ambiguity-resolving multi-channel SAR (segmentation and resassembly sublayer) orientation - Google Patents

Method for estimating fussy component space angle and ambiguity-resolving multi-channel SAR (segmentation and resassembly sublayer) orientation Download PDF

Info

Publication number
CN104698431A
CN104698431A CN201510114770.3A CN201510114770A CN104698431A CN 104698431 A CN104698431 A CN 104698431A CN 201510114770 A CN201510114770 A CN 201510114770A CN 104698431 A CN104698431 A CN 104698431A
Authority
CN
China
Prior art keywords
ambiguity
doppler
component
signal
overbar
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201510114770.3A
Other languages
Chinese (zh)
Other versions
CN104698431B (en
Inventor
沈明威
杨柳
于佳
胡佩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hohai University HHU
Original Assignee
Hohai University HHU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hohai University HHU filed Critical Hohai University HHU
Priority to CN201510114770.3A priority Critical patent/CN104698431B/en
Publication of CN104698431A publication Critical patent/CN104698431A/en
Application granted granted Critical
Publication of CN104698431B publication Critical patent/CN104698431B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S3/00Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
    • G01S3/02Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using radio waves
    • G01S3/14Systems for determining direction or deviation from predetermined direction
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88Radar or analogous systems specially adapted for specific applications
    • G01S13/89Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/9004SAR image acquisition techniques
    • G01S13/9011SAR image acquisition techniques with frequency domain processing of the SAR signals in azimuth

Abstract

The invention discloses a method for estimating a fussy component space angle and ambiguity-resolving multi-channel SAR (segmentation and resassembly sublayer) orientation, which solves a problem that ADBF (adaptive digital beam forming) airspace covariance matrix estimation signals under system errors are balanced out and target guide vectors are mismatched when a conventional method solves azimuth ambiguity. The method comprises the following steps: firstly, constructing a multichannel SAR echo signal mode; then, performing fast Fourier transform (FFT) on an echo signal tracing distance unit subjected to distance dimensional pulse compression; then, utilizing NSIT to accurately estimate a space angle of each doppler ambiguity component in real time, assisting designing a space filter based on spatial information of each ambiguity component, and extracting each doppler ambiguity component; splicing to obtain a complete non-ambiguity signal to achieve an azimuth ambiguity dissolving purpose. The method effectively solves a problem that the signals are balanced out as each secondary ambiguity component exists in a sample of ADBF covariance matrix estimation; moreover, the inhibition to the ambiguity component is greatly improved.

Description

The estimation of obscuring component Space Angle and hyperchannel SAR orientation ambiguity solution method
Technical field
The invention belongs to airborne hyperchannel SAR wide-scene survey field, be specifically related to a kind of method of estimation and hyperchannel SAR orientation ambiguity solution method of obscuring component Space Angle.
Background technology
In airborne synthetic aperture radar imaging, for solving the contradiction between wide swath and orientation high resolving power, Chinese scholars has mainly done a lot of research work from two aspects, is SAR ambiguity solution treatment technology on the one hand, is on the other hand to research and develop newer SAR system; Usually use lower pulse repetition rate (PRF) to ensure coverage rate in wider scope in usual SAR system, but bearing signal is fuzzy, therefore ambiguity solution is that SAR imaging is necessary.In addition, the principal character of new SAR system is the combination of multiple sending/receiving passage and suitable Digital Signal Processing.
By the corresponding relation of analytic signal Doppler frequency and locus, propose to adopt space domain self-adapted Wave beam forming (ADBF) technology to suppress doppler ambiguity, it can adaptively to the zero setting of doppler ambiguity component.
But traditional ADBF technology exists each obscuring component in the sample of covariance matrix, can produce signal cancellation problem simultaneously, thus the secondary lobe of sef-adapting filter is caused to be raised.Particularly when carrier aircraft speed exists error, between each obscuring component real space angle and theoretical value, there is deviation, cause follow-up target guiding vector severe mismatch.This all makes follow-up SAR azimuth focus hydraulic performance decline.
Summary of the invention
Technical matters to be solved by this invention is: the estimation and the hyperchannel SAR orientation ambiguity solution method that provide a kind of obscuring component Space Angle, the problem of ADBF spatial domain covariance matrix signal cancellation and target guiding vector mismatch under systematic error when solving classic method solution azimuth ambiguity.
The present invention, for solving the problems of the technologies described above, adopts following technical scheme:
The method of estimation of obscuring component Space Angle,
First, the signal received radar carries out distance dimension pulse compression and direction dimension Fourier transform obtains radar each channel signal range Doppler figure; Then choose K+1 array element and form M-K submatrix, wherein, K is doppler ambiguity number of components, and M is spatial domain array number, first submatrix be 1 ..., and K+1}, second submatrix be 2 ... K+2}, the method comprises the steps:
Step 1, the mean value of covariance matrix estimated according to each submatrix of following formulae discovery
R ^ sa = 1 M sa Σ p = 1 M sa R ^ p , Wherein, R ^ p = 1 2 L + 1 Σ l = 1 2 L + 1 X p ( i + l - ( L + 1 ) ) X p H ( i + l - ( L + 1 ) ) , p = 1 , . . . , M sa
Wherein, L is the sample number of Received signal strength, and l is the natural number being less than or equal to 2L+1, and i is the sequence number of the range unit in p submatrix, and p is natural number, hfor conjugate transpose, M sa=M-K, X p(i)=[S p(i, r) ..., S p+K(i, r)] tfor the vector form of i-th range unit r doppler cells signal that each submatrix receives;
Step 2, preset the Space Angle initial value sequence θ of K obscuring component q, according to ε q=-(2 π/λ) dsin (θ q) obtain vector sequence ε q, wherein q=1 ..., K, makes vector ε=[ε 1..., ε k], preset the signal subspace W that obscuring component is formed 1=1, convergence threshold value χ=0.5, λ is the wavelength of radar emission signal, and d is array element distance;
Step 3, according to following formulae discovery secondary cost function:
C = W K + 1 H R ^ sa W K + 1
W in formula k+1the signal subspace that K obscuring component is formed, and W k+1with meet following relation:
W K + 1 = W ‾ K - e jϵ ( K ) P W ‾ K
Wherein W ‾ K = W K 0 , P = 0 U K × ( K + 1 ) , In P, zero row vector is 1 × (K+1), U k × (K+1)for the capable K+1 column matrix of K, make U k × (K+1)in matrix, a ranks number identical element is 1, and remaining element is 0;
Step 4, perform step 3 according to the initial value in step 2, calculate the initial function value C of secondary cost function 0;
The Space Angle of step 5, a calculating kth obscuring component, detailed process is as follows:
Step 5.1, the element in vector sequence ε to be resequenced, obtain ε=[ε 1..., ε k-1, ε k+1..., ε k, ε k], according to ε and W after adjustment 1utilize formula W K + 1 = W ‾ K - e jϵ ( K ) P W ‾ K Calculate again by formula e j ϵ ^ k = W ‾ K H R ^ sa W ‾ K W ‾ K H R ^ sa P W ‾ K Obtain estimate will value replaces ε in ε kvalue;
Step 5.2, repeats step 5.1, estimates all ε values, obtains the ε vector sequence after the renewal of whole element;
ε vector sequence repeated execution of steps 3 after the renewal that step 6, applying step 5 obtain obtains the cost function value C after upgrading 1;
Step 7, judge threshold value whether be greater than initial threshold χ, if make C 0=C 1, then perform step 5 and step 6, otherwise perform step 8;
Step 8, Space Angle according to a following formulae discovery kth obscuring component:
ε k=-(2π/λ)dsin(θ k);
Step 9: repeat step 8, obtain the Space Angle of all obscuring component.
Described acquisition radar each channel signal range Doppler figure adopts with the following method:
Steps A, according to following formula to radar m channel receiving signal carry out distance dimension pulse compression:
S m ( τ , t ) = { IFFT f r { FFT τ [ S m ( τ , t ) ] · FFT τ [ H r ( τ ) ] } } Nr × Na
Wherein, t be orientation to the time, τ be distance to the time, Nr be distance dimension sampling number, Na is orientation synthetic aperture umber of pulse, FFT τand IFFT frrepresent about distance to the Fourier transform of τ and the inverse Fourier transform about distance frequency domain respectively,
k rfor the chirp rate of transponder pulse signal;
Step B, according to following formula by the signal after the pulse compression of distance dimension by the Fast Fourier Transform (FFT) of range unit travel direction dimension, obtain range Doppler figure:
S m(τ,f)=FFT t[S m(τ,t)]
Wherein FFT trepresent about the Fourier transform of orientation to time t.
A kind of hyperchannel SAR orientation ambiguity solution method, comprises the steps:
Step a, obtain the Space Angle θ of K doppler ambiguity component of each doppler cells 1, θ 2... θ k;
Step b, normalization spatial domain steering vector according to a following formulae discovery kth obscuring component:
A k ( θ k ) = [ exp ( j 2 π λ · 1 · d sin ( θ k ) ) , . . . , exp ( j 2 π λ · M · d sin ( θ k ) ) ] T / M , Wherein, k=1 ..., K;
Build the covariance matrix R of a kth doppler ambiguity component k, and calculate corresponding ADBF weights; Its ADBF weights meet following formula:
s . t . W k H A k = 1 min w k W k H R ‾ k W k
Covariance matrix meet:
R ‾ k = Σ g = 1 K A g · A g H + σ 0 I , G=1 ..., K and g ≠ k,
Wherein, σ 0for the noise power loaded, I is the unit matrix of M × M;
Step c, ADBF weights according to a following formulae discovery kth doppler ambiguity component:
W k = R ‾ k - 1 A k A k H R ‾ k - 1 A k
Steps d, write i-th range unit r doppler cells signal that M array element receives as vector form and be:
Y=[S 1(i,r),...,S M(i,r)] T
Step e, obtain a kth doppler ambiguity component signal P according to following formula r; k:
P r ; k = W k H Y
Step f, to extracting a kth doppler ambiguity component signal respectively by doppler cells in i-th range unit and being designated as P (k), P (k)=[P 1; k..., P na; k], then corresponding according to different Doppler frequency obscuring component signal is arranged in order, and obtains i-th range unit full bandwidth doppler frequency data S (i, f after ambiguity solution a), that is:
S(i,f a)=[P(1),...,P(K)]
Step g, by range unit repeat step b ~ step f obtain data S (τ, f after Doppler ambiguity-resolution a), and orientation synthetic aperture processing is carried out to it, obtain SAR image.
Described step a application rights requires that the method for 1 obtains the Space Angle θ of K doppler ambiguity component of each doppler cells 1, θ 2... θ k.
In described step b, σ 0=10 -4.
Compared with prior art, the present invention has following beneficial effect:
1, when carrier aircraft speed does not have error, compare traditional ADBF method, the suppression of context of methods obscuring component improves 7dB, efficiently solves in the sample because of ADBF covariance matrix the signal cancellation problem that there is each obscuring component simultaneously and bring;
2, when carrier aircraft speed has error, due to the Space Angle of real-time ambiguous estimation component, make steering vector not have mismatch and overcome covariance matrix signal cancellation problem, comparing traditional ADBF method, the suppression of context of methods obscuring component improves about about 10dB.
Accompanying drawing explanation
Fig. 1 is positive side-looking stripmap SAR geometric relationship schematic diagram.
Fig. 2 (a) is ground echo Doppler frequency without doppler ambiguity and position angle graph of a relation.
Fig. 2 (b) is for having ground echo Doppler frequency and the position angle graph of a relation of doppler ambiguity.
Fig. 3 is the Doppler ambiguity-resolution design of filter process flow diagram based on NISE.
Fig. 4 (a) schemes without being compressed into picture under hazy condition for carrier aircraft speed is error free.
Fig. 4 (b) is for the error free orientation without point target under hazy condition of carrier aircraft speed is to sectional view.
Fig. 5 (a) has image under hazy condition for carrier aircraft speed is error free.
Fig. 5 (b) has the orientation of point target under hazy condition to sectional view for carrier aircraft speed is error free.
Fig. 6 (a) is the error free actual two-dimensional spectrum relation without ambiguity ground echo of carrier aircraft speed.
Fig. 6 (b) is the error free desirable two-dimensional spectrum relation without ambiguity ground echo of carrier aircraft speed.
Fig. 6 (c) has the actual two-dimensional spectrum relation of ambiguity ground echo for carrier aircraft speed is error free.
Fig. 6 (d) has the desirable two-dimensional spectrum relation of ambiguity ground echo for carrier aircraft speed is error free.
Fig. 7 (a) extracts each obscuring component and spliced result in the error free situation of carrier aircraft speed through filtering for the inventive method.
Fig. 7 (b) is the error free traditional ADBF disposal route result of carrier aircraft speed.
Fig. 8 (a) has error without ambiguity point target orientation to sectional view for carrier aircraft speed.
Fig. 8 (b) has error to have ambiguity point target orientation to sectional view for carrier aircraft speed.
Fig. 9 (a) has error without the actual two-dimensional spectrum relation of ambiguity ground echo for carrier aircraft speed.
Fig. 9 (b) has error without the desirable two-dimensional spectrum relation of ambiguity ground echo for carrier aircraft speed.
Fig. 9 (c) has error to have the actual two-dimensional spectrum relation of ambiguity ground echo for carrier aircraft speed.
Fig. 9 (d) has error to have the desirable two-dimensional spectrum relation of ambiguity ground echo for carrier aircraft speed.
Figure 10 (a) has error condition to extract each obscuring component and spliced result through filtering in carrier aircraft speed for the inventive method.
The ADBF disposal route result that Figure 10 (b) has error traditional for carrier aircraft speed.
Embodiment
Below in conjunction with accompanying drawing, technical scheme of the present invention is described in detail:
Airborne radar geometric configuration as shown in Figure 1, carrier aircraft is flown along X-axis with speed V, and carrier aircraft flying height H, a M even linear array is along course lineal layout, regulation is transmitting-receiving array element (being No. 1, array element in Fig. 1) along navigation direction high order end array element, and all the other are for receiving array element.Suppose that aircraft is in true origin overhead when t=0, certain one time t, the position x=V*t of aircraft, then target to the oblique distance of m array element is:
R m 2(t)=R 0 2+(x+(m-1)d) 2m=1,...,N
=R 0 2+(V·t+(m-1)d) 2
Wherein R 0 2=H 2+ y 0 2, H represents aircraft altitude, and d represents array element distance.
Obtain through abbreviation, the echo model of each passage is:
X m ( τ , t ) = exp [ jπ K r τ 2 ] · exp [ - j 2 π λ ( R 1 ( t ) + R m ( t ) ) ] , m = 1 , . . . , N
K in formula rrefer to the frequency modulation rate of transponder pulse signal, τ represents that distance is to the time, and λ is wavelength.
The ground echo Doppler frequency f of carried SAR dand the relation between azimuth angle theta is as shown in Fig. 2 (a), that is:
f d = 2 V λ sin ( θ )
When echoed signal doppler bandwidth is higher than radar pulse repetition frequency (PRF), now azimuth spectrum produces fuzzy.Fig. 2 (b) give azimuth spectrum 3 times fuzzy time space-time-frequency relation.Adopt Adaptive beamformer (ADBF) technology to extract each doppler ambiguity component signal successively by range unit in conventional methods where, and then realize azimuth spectrum reconstruction.But owing to there is each obscuring component in the sample of ADBF covariance matrix simultaneously, ADBF signal cancellation can be caused to lose; In addition when carrier aircraft speed exists error, between each obscuring component real space angle and calculated value, there is deviation, cause follow-up ADBF target guiding vector mismatch.Above-mentioned factor all will cause ADBF to decline to obscuring component rejection, cause the lifting of synthesis frequency spectrum azimuth focus minor level.Given this, this patent proposes the hyperchannel SAR orientation ambiguity solution method based on obscuring component DOA estimation.Comprise: the pulse compression of echoed signal distance dimension, obscuring component DOA estimation, Doppler ambiguity-resolution design of filter.Fig. 3 gives the part signal treatment scheme of this algorithm.
1. distance dimension pulse compression
Assuming that radar m channel receiving signal is S m(τ, t), and S m(τ, t) size is Nr × Na, wherein t be orientation to the time, remember t=0 with impact point to flight track (orientation to) perpendicular line and flight path point of intersection; τ represents that distance is to the time, is the uniform sampling point in the pulse width transmitted; Nr is distance dimension sampling number, and Na is orientation synthetic aperture umber of pulse, existing to S m(τ, t) carries out the pulse compression of distance dimension, that is:
S m ( τ , t ) = { IFFT f r { FFT τ [ S m ( τ , t ) ] · FFT τ [ H r ( τ ) ] } } Nr × Na - - - ( 1 )
FFT in formula τwith represent about distance to the Fourier transform of τ and the inverse Fourier transform about distance frequency domain respectively, for distance is to the matched filter function of pulse pressure, K rfor the chirp rate of transponder pulse signal;
2. obscuring component DOA estimates
Due under wide-scene mapping condition, the orientation doppler bandwidth of echoed signal is much larger than radar pulse repetition frequency (PRF), therefore there is multiple obscuring component in each doppler cells, we adopt noise subspace iterative estimation technique (NSIT) to estimate the space angle of different obscuring component in each doppler cells one by one; Assuming that doppler ambiguity number of components is K, M is spatial domain array number, meet M>K; First each channel signal after Range compress is tieed up Fast Fourier Transform (FFT) (FFT) by range unit travel direction, obtain range Doppler figure, that is:
S m(τ,f)=FFT t[S m(τ,t)] (2)
Wherein FFT trepresent about the Fourier transform of orientation to time t;
To estimate the Space Angle of obscuring component in i-th range unit r doppler cells, NSIT algorithm picks M-K the sub-aperture that wherein K+1 array element is formed estimates the Space Angle of K different obscuring component; Array number due to sub-aperture is K+1, will 1 ..., K+1} array element forms the 1st submatrix, 2 ... K+2} array element forms the 2nd submatrix, by that analogy, the submatrix of composition M altogether like this sa=M-K; I-th range unit r doppler cells signal that each submatrix receives is write as vector form and is:
X p(i)=[S p(i,r),...,S p+K(i,r)] T,p=1,...,M sa(3)
S in formula p(i, r) is p passage i-th range unit r doppler cells information, subscript tfor transpose operator;
NSIT estimates the concrete steps of each obscuring component DOA:
Step 1: the mean value detecing the covariance matrix of estimation according to following formulae discovery separately wherein, R ^ p = 1 2 L + 1 Σ l = 1 2 L + 1 X p ( i + l - ( L + 1 ) ) X p H ( i + l - ( L + 1 ) ) , p = 1 , . . . , M sa - - - ( 4 )
In formula, subscript H represents that conjugate transpose operation accords with;
Step 2: initialization of variable: the DOA initial value θ of setting K obscuring component q, according to ε q=-(2 π/λ) dsin (θ q) obtain vector sequence ε p, wherein q=1,2 ..., K, λ are wavelength, and d is array element distance, make vector ε=[ε 1..., ε k], ε (q) is the value of q element in vector ε, i.e. ε (q)=ε q; ; The signal subspace W that setting obscuring component is formed 1=1; Convergence threshold value χ=0.5;
Step 3: calculate secondary cost function: first defining secondary cost function is:
C = W K + 1 H R ^ sa W K + 1 - - - ( 5 )
W in formula k+1the signal subspace that K obscuring component is formed, and W k+1with meet following relation:
W K + 1 = W ‾ K - e jϵ ( K ) P W ‾ K - - - ( 6 )
Wherein W ‾ K = W K 0 , P = 0 U K × ( K + 1 ) , In P, zero row vector is 1 × (K+1); U k × (K+1)for the capable K+1 column matrix of K, make U k × (K+1)in matrix, a ranks number identical element is 1, and remaining element is 0; Step 4: according to initial value ε and W in step 2 1, can recurrence calculation W one by one according to formula (6) k+1, then calculate secondary cost function C according to formula (5), and perform C 0=C;
Step 5, calculate the Space Angle of a kth obscuring component, detailed process is as follows: step 5.1, resequenced by the element in vector sequence ε, obtains ε=[ε 1..., ε k-1, ε k+1..., ε k, ε k], according to ε and W after adjustment 1utilize formula W K + 1 = W ‾ K - e jϵ ( K ) P W ‾ K Calculate again by formula e j ϵ ^ k = W ‾ K H R ^ sa W ‾ K W ‾ K H R ^ sa P W ‾ K Obtain estimate will value replaces ε in ε kvalue;
Step 5.2, repetition step 5.1, estimate all ε values, obtains the ε vector sequence after the renewal of whole element;
Step 6: the ε vector sequence repeated execution of steps 3 after the renewal that applying step 5 obtains obtains the cost function value C after upgrading 1;
Step 7, judge threshold value whether be greater than initial threshold χ, if make C 0=C 1, then perform step 5 and step 6, otherwise perform step 8; Step 8, Space Angle according to a following formulae discovery kth obscuring component: according to the vector ε=[ε of final updating 1..., ε k], to each ε (k), k=1 ..., K, by formula the DOA value of K obscuring component can be obtained respectively;
Here is NSIT algorithm flow chart:
The DOA Calculation:
By the vector ε of final updating, computing formula
ε (k), k=1 ... K, can obtain the DOA value of K obscuring component respectively;
3. Doppler ambiguity-resolution design of filter
NSIT alternative manner is utilized to obtain the DOA value of K obscuring component of each doppler cells; Now for the i-th range unit r doppler cells above, assuming that the space angle of the K an estimated obscuring component is respectively θ 1, θ 2... θ k, then the normalization spatial domain steering vector of each obscuring component is:
A k ( θ k ) = [ exp ( j 2 π λ · 1 · d sin ( θ k ) ) , . . . , exp ( j 2 π λ · M · d sin ( θ k ) ) ] T / M , k = 1 , . . . , K - - - ( 7 )
Because each range unit exists each obscuring component simultaneously, directly utilize and receive data estimation covariance matrix, and calculate ADBF weights and will cause signal cancellation; Therefore we build corresponding covariance matrix R respectively based on the doppler ambiguity component extracted, and calculate corresponding ADBF weights; Assuming that extract a kth doppler ambiguity component, its spatial domain steering vector is A k, its ADBF weights meet:
s . t . W k H A k = 1 min w k W k H R ‾ k W k
Covariance matrix meet:
R ‾ k = Σ g = 1 K A g · A g H + σ 0 I , G=1 ..., K and g ≠ k (8)
σ in formula 0for the noise power loaded, σ can be set 0=10 -4, I is the unit matrix of M × M; The ADBF weights then extracting a kth doppler ambiguity component are:
W k = R ‾ k - 1 A k A k H R ‾ k - 1 A k - - - ( 9 )
Being write i-th of full array received range unit r doppler cells signal as vector form is:
Y=[S 1(i,r),...,S M(i,r)] T(10)
Now utilize W kin i-th range unit r doppler cells, extract a kth doppler ambiguity component signal, use P r; krepresent, that is:
P r ; k = W k H Y - - - ( 11 )
To extracting a kth doppler ambiguity component signal respectively by doppler cells in i-th range unit and being designated as P (k), P (k)=[P 1; k..., P na; k], then corresponding according to different Doppler frequency obscuring component signal is arranged in order, and namely obtains i-th range unit full bandwidth doppler frequency data S (i, f after ambiguity solution a), that is:
S(i,f a)=[P(1),...,P(K)] (12)
Repeat said process by range unit and can obtain data S (τ, f after Doppler ambiguity-resolution a), carrying out orientation synthetic aperture processing to it can obtain High Resolution SAR Images.
Below by Computer Simulation checking this patent correctness.Airborne positive side-looking array synthetic-aperture radar imagery simulation parameter is as shown in table 1, array number M=4, obscuring component number K=3, and the scattering P that counts gets 51, and it is in a row that point target is distributed in longitudinal shape, and each point is spaced apart 3m.In following emulation experiment, with the 512nd range unit for reference unit.
Table 1 simulation of Radar System parameter
Parameter name Parameter values
Pulse repetition rate 630Hz
Transmitted signal bandwidth 60MHz
Distance is to sampling rate 75MHz
Array element distance 0.3m
Carrier aircraft speed 150m/s
Length of synthetic aperture 360m
Carrier aircraft flying height 4472m
Wavelength 0.03m
(1) carrier aircraft speed error free situation processing procedure is as follows
First, the dimension two dimensional compaction process of distance peacekeeping direction is done to received signal.As Fig. 4 (a), Fig. 4 (b) indicate that without result figure, Fig. 4 (a) during azimuth ambiguity be impact point after distance peacekeeping azimuth dimension pulse pressure, the orientation that Fig. 4 (b) is point target to sectional view.When Fig. 5 (a), Fig. 5 (b) indicate azimuth ambiguity result figure, Fig. 5 (a) be depicted as distance to orientation to the impact point after pulse pressure, the orientation that Fig. 5 (b) is point target is to figure.
Secondly, utilize NSIT to estimate azimuth ambiguity and the DOA without obscuring component during azimuth ambiguity, draw the theoretical value by formulae discovery simultaneously.Result such as Fig. 6 (a) is the actual value without the corresponding Space Angle relation of Doppler frequency during azimuth ambiguity, Fig. 6 (b) is without the corresponding Space Angle relation of Doppler frequency during azimuth ambiguity ideal value, Fig. 6 (c) represents the actual value of Doppler frequency corresponding Space Angle relation when there is azimuth ambiguity, Fig. 6 (d) represents the ideal value of Doppler frequency corresponding Space Angle relation when there is azimuth ambiguity, from Fig. 6 (a), Fig. 6 (b) can find out without actual value during azimuth ambiguity and ideal value about the same, error is less.Doppler frequency correspondence Space Angle relation actual and desirable when Fig. 6 (c), Fig. 6 (d) indicate azimuth ambiguity, known in without velocity error situation actual value and theoretical value close, but actual curve has certain roughness, and ideal value is a proper linear line, therefore there is certain error.It should be noted that at the junction of three sections of obscuring component generation frequency spectrum hopping phenomenon, is because the finite-length effect of signal causes.
Finally, utilize the Space Angle actual value Computer Aided Design adaptive spatial filter estimated, extract each obscuring component and be spliced into complete Doppler signal, as shown in Fig. 7 (a), Fig. 7 (b).Fig. 7 (a) is context of methods result, Fig. 7 (b) is the spatial domain covariance matrix that neighbor distance cell signal estimates each blurred signal, suppress each blurred signal respectively according to theoretical spatial domain steering vector again, and then carry out frequency spectrum splicing.By contrast, context of methods filtered blurry component suppresses to improve 7dB, thus overcomes ADBF covariance matrix signal cancellation and target guiding vector mismatch problems under systematic error.
(2) carrier aircraft speed has error condition processing procedure as follows
There is certain deviation according to actual conditions carrier aircrafts speed, setting carrier aircraft speed V existence ± 5m/s error, be (V+5) m/s with velocity deviation during emulation, other simulated conditions are constant.
First, the orientation of point target when drawing without azimuth ambiguity and have azimuth ambiguity is to sectional view, and result is as shown in Fig. 8 (a), Fig. 8 (b).
Secondly azimuth ambiguity is estimated according to NSIT and without real space angle during azimuth ambiguity, to provide when there is velocity deviation their theoretical space angle simultaneously, as the actual value that Fig. 9 (a) is without the corresponding Space Angle relation of Doppler frequency during azimuth ambiguity, Fig. 9 (b) is without the corresponding Space Angle relation of Doppler frequency during azimuth ambiguity ideal value, the actual value of Doppler frequency corresponding Space Angle relation when Fig. 9 (c) indicates azimuth ambiguity, Fig. 9 (d) represents the ideal value of Doppler frequency corresponding Space Angle relation when there is azimuth ambiguity, from Fig. 9 (a), Fig. 9 (b) can find out because velocity deviation reason having certain deviation without orientation to Space Angle desirable time fuzzy compared with actual value, large with compared with velocity error.When Fig. 9 (c), Fig. 9 (d) represent that carrier aircraft speed exists error, the DOA actual value of obscuring component and ideal value contrast, there are some fluctuations from the known actual value of figure, a not linear line, and ideal value is the linear line of stricti jurise.Therefore because carrier aircraft speed exists deviating cause, if recycling ideal value process data will cause target guiding vector severe mismatch, can not effective filtering obscuring component.
Finally utilize and obtain real space angle Computer Aided Design adaptive spatial filter, extract each obscuring component and be spliced into complete Doppler signal, as Suo Shi Figure 10 (a), Figure 10 (b), the orientation of point target is to sectional view.Figure 10 (a) is context of methods result, Figure 10 (b) is the spatial domain covariance matrix method process utilizing neighbor distance cell signal to estimate each blurred signal, suppress each blurred signal respectively according to desirable spatial domain steering vector again, and then carry out frequency spectrum splicing.By contrast, context of methods filtered blurry component suppresses totally to improve about 10dB, efficiently solves the problem of covariance matrix signal cancellation problem and target guiding vector mismatch.
The present invention is directed to the mapping of airborne hyperchannel SAR wide-scene and there is doppler ambiguity problem, propose a kind of hyperchannel SAR orientation ambiguity solution method estimated based on obscuring component DOA.The DOA utilizing NSIT accurately to estimate each doppler ambiguity component is in real time proposed in literary composition, and then based on the spatial information (si) Computer Aided Design spatial filter of each obscuring component, and extract each doppler ambiguity component successively.Emulation experiment shows, context of methods efficiently solves conventional ADBF covariance matrix signal cancellation and target guiding vector mismatch problems under systematic error.When carrier aircraft speed does not have error, compare traditional ADBF method, the suppression of context of methods obscuring component improves 7dB; When carrier aircraft speed has error, due to the Space Angle of real-time ambiguous estimation component, make steering vector not have mismatch and overcome covariance matrix signal cancellation problem, comparing traditional ADBF method, the suppression of context of methods obscuring component improves about about 10dB.If when last orientation increases to umber of pulse accumulation, this algorithm advantage is more outstanding, and algorithm operation efficiency is high, is easy to engineering construction.

Claims (5)

1. the method for estimation of obscuring component Space Angle, is characterized in that:
First, the signal received radar carries out distance dimension pulse compression and direction dimension Fourier transform obtains radar each channel signal range Doppler figure; Then choose K+1 array element and form M-K submatrix, wherein, K is doppler ambiguity number of components, and M is spatial domain array number, first submatrix be 1 ..., K+1}, second submatrix be 2 ... K+2}, the method comprises the steps:
Step 1, the mean value of covariance matrix estimated according to each submatrix of following formulae discovery
R ^ sa = 1 M sa Σ p = 1 M sa R ^ p , Wherein, R ^ p = 1 2 L + 1 Σ l = 1 2 L + 1 X p ( i + l - ( L + 1 ) ) X p H ( i + l - ( L + 1 ) ) , p = 1 , . . . , M sa
Wherein, L is the sample number of Received signal strength, and l is the natural number being less than or equal to 2L+1, and i is the sequence number of the range unit in p submatrix, and p is natural number, and H is conjugate transpose, M sa=M-K, X p(i)=[S p(i, r) ..., S p+K(i, r)] tfor the vector form of i-th range unit r doppler cells signal that each submatrix receives;
Step 2, preset the Space Angle initial value sequence θ of K obscuring component q, according to ε q=-(2 π/λ) d sin (θ q) obtain vector sequence ε q, wherein q=1 ..., K, makes vector ε=[ε 1..., ε k], preset the signal subspace W that obscuring component is formed 1=1, convergence threshold value χ=0.5, λ is the wavelength of radar emission signal, and d is array element distance;
Step 3, according to following formulae discovery secondary cost function:
C = W K + 1 H R ^ sa W K + 1
W in formula k+1the signal subspace that K obscuring component is formed, and W k+1with meet following relation:
W K + 1 = W ‾ K - e jϵ ( K ) P W ‾ K
Wherein W ‾ K = W K 0 , P = 0 U K × ( K + 1 ) , In P, zero row vector is 1 × (K+1), U k × (K+1)for the capable K+1 column matrix of K, make U k × (K+1)in matrix, a ranks number identical element is 1, and remaining element is 0;
Step 4, perform step 3 according to the initial value in step 2, calculate the initial function value C of secondary cost function 0;
The Space Angle of step 5, a calculating kth obscuring component, detailed process is as follows:
Step 5.1, the element in vector sequence ε to be resequenced, obtain ε=[ε 1..., ε k-1, ε k+1..., ε k, ε k], according to ε and W after adjustment 1utilize formula W K + 1 = W ‾ K - e jϵ ( K ) P W ‾ K Calculate again by formula e j ϵ ^ k = W ‾ K H R ^ sa W ‾ K W ‾ K H R ^ sa P W ‾ K Obtain estimate will value replaces ε in ε kvalue;
Step 5.2, repeats step 5.1, estimates all ε values, obtains the ε vector sequence after the renewal of whole element;
ε vector sequence repeated execution of steps 3 after the renewal that step 6, applying step 5 obtain obtains the cost function value C after upgrading 1;
Step 7, judge threshold value whether be greater than initial threshold χ, if make C 0=C 1, then perform step 5 and step 6, otherwise perform step 8;
Step 8, Space Angle according to a following formulae discovery kth obscuring component:
ε k=-(2π/λ)d sin(θk);
Step 9: repeat step 8, obtain the Space Angle of all obscuring component.
2. the method for estimation of obscuring component Space Angle according to claim 1, is characterized in that: described acquisition radar each channel signal range Doppler figure adopts with the following method:
Steps A, according to following formula to radar m channel receiving signal carry out distance dimension pulse compression:
S m ( τ , t ) = { IFFFT f r { FFT τ [ S m ( τ , t ) ] · FFT τ [ H r ( τ ) ] } } Nr × Na
Wherein, t be orientation to the time, τ be distance to the time, Nr be distance dimension sampling number, Na is orientation synthetic aperture umber of pulse, FFT τwith represent about distance to the Fourier transform of τ and the inverse Fourier transform about distance frequency domain respectively,
H r ( τ ) exp ( - jπ τ 2 K r ) , K rfor the chirp rate of transponder pulse signal;
Step B, according to following formula by the signal after the pulse compression of distance dimension by the Fast Fourier Transform (FFT) of range unit travel direction dimension, obtain range Doppler figure:
S m(τ,f)=FFT t[S m(τ,t)]
Wherein FFT trepresent about the Fourier transform of orientation to time t.
3., based on the hyperchannel SAR orientation ambiguity solution method of the method for estimation of obscuring component Space Angle described in claim 1, it is characterized in that: comprise the steps:
Step a, obtain the Space Angle θ of K doppler ambiguity component of each doppler cells 1, θ 2... θ k;
Step b, normalization spatial domain steering vector according to a following formulae discovery kth obscuring component:
A k ( θ k ) = [ exp ( j 2 π λ · 1 · d sin ( θ k ) , . . . , exp ( j 2 π λ · M · d sin ( θ k ) ) ] T / M , Wherein, k=1 ..., K;
Build the covariance matrix of a kth doppler ambiguity component and calculate corresponding ADBF weights; Its ADBF weights meet following formula:
s . t . W k H A k = 1 min w k W k H R ‾ k W k
Covariance matrix meet:
R ‾ k = Σ g = 1 K A g · A g H + σ 0 I , G=1 ..., K and g ≠ k,
Wherein, σ 0for the noise power loaded, I is the unit matrix of M × M;
Step c, ADBF weights according to a following formulae discovery kth doppler ambiguity component:
W k = R ‾ k - 1 A k A k H R ‾ k - 1 A k
Steps d, write i-th range unit r doppler cells signal that M array element receives as vector form and be:
Y=[S 1(i,r),…,S M(i,r)] T
Step e, obtain a kth doppler ambiguity component signal P according to following formula r; k:
P r;k=W k HY
Step f, to extracting a kth doppler ambiguity component signal respectively by doppler cells in i-th range unit and being designated as P (k), P (k)=[P 1; k..., P na; k], then corresponding according to different Doppler frequency obscuring component signal is arranged in order, and obtains i-th range unit full bandwidth doppler frequency data S (i, f after ambiguity solution a), that is:
S(i,f a)=[P(1),…,P(K)]
Step g, by range unit repeat step b ~ step f obtain data S (τ, f after Doppler ambiguity-resolution a), and orientation synthetic aperture processing is carried out to it, obtain SAR image.
4. the hyperchannel SAR orientation ambiguity solution method of the method for estimation of obscuring component Space Angle according to claim 3, is characterized in that:
Described step a application rights requires that the method for 1 obtains the Space Angle θ of K doppler ambiguity component of each doppler cells 1, θ 2... θ k.
5. the hyperchannel SAR orientation ambiguity solution method of the method for estimation of obscuring component Space Angle according to claim 3, is characterized in that: in described step b, σ 0=10 -4.
CN201510114770.3A 2015-03-17 2015-03-17 Based on the multichannel SAR orientation ambiguity solution method that obscuring component DOA estimates Expired - Fee Related CN104698431B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510114770.3A CN104698431B (en) 2015-03-17 2015-03-17 Based on the multichannel SAR orientation ambiguity solution method that obscuring component DOA estimates

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510114770.3A CN104698431B (en) 2015-03-17 2015-03-17 Based on the multichannel SAR orientation ambiguity solution method that obscuring component DOA estimates

Publications (2)

Publication Number Publication Date
CN104698431A true CN104698431A (en) 2015-06-10
CN104698431B CN104698431B (en) 2017-06-30

Family

ID=53345754

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510114770.3A Expired - Fee Related CN104698431B (en) 2015-03-17 2015-03-17 Based on the multichannel SAR orientation ambiguity solution method that obscuring component DOA estimates

Country Status (1)

Country Link
CN (1) CN104698431B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106443672A (en) * 2016-08-30 2017-02-22 西安电子科技大学 Azimuth multichannel SAR signal adaptive reconstruction method
CN108399607A (en) * 2017-11-21 2018-08-14 北京航空航天大学 Image processing method, device, equipment and computer readable storage medium

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
ANDREA MONTI GUARNIERI: "Adaptive Removal of Azimuth Ambiguities in SAR Images", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》 *
MICHELANGELO VILLANO 等: "Spectral-Based Estimation of the Local Azimuth Ambiguity-to-Signal Ratio in SAR Images", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》 *
杨磊 等: "多发多收Scan模式实现高分辨宽测绘带SAR成像", 《系统工程与电子技术》 *
陈倩 等: "基于自适应滤波的DPC-MAB SAR方位向信号重建", 《电子与信息学报》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106443672A (en) * 2016-08-30 2017-02-22 西安电子科技大学 Azimuth multichannel SAR signal adaptive reconstruction method
CN106443672B (en) * 2016-08-30 2019-03-29 西安电子科技大学 A kind of orientation multichannel SAR signal adaptive reconstructing method
CN108399607A (en) * 2017-11-21 2018-08-14 北京航空航天大学 Image processing method, device, equipment and computer readable storage medium
CN108399607B (en) * 2017-11-21 2021-08-17 北京航空航天大学 Method, device and equipment for inhibiting image orientation blur and computer readable storage medium

Also Published As

Publication number Publication date
CN104698431B (en) 2017-06-30

Similar Documents

Publication Publication Date Title
CN103869311B (en) Real beam scanning radar super-resolution imaging method
CN104950305B (en) A kind of real beam scanning radar angle super-resolution imaging method based on sparse constraint
CN103176168B (en) A kind of airborne non-working side battle array radar short range clutter cancellation method
CN102967858B (en) Radar foresight super-resolution imaging method
CN104035095B (en) Based on the low level wind shear velocity estimation method of optimal processor during sky
CN106291543A (en) A kind of motion platform scanning radar super-resolution imaging method
CN102323583B (en) Super-resolution linear array three-dimensional synthetic aperture radar imaging method
CN103454635B (en) The squint SAR formation method of the section of flying is put down based on hypersonic aircraft
CN103207380B (en) Broadband target direction finding method based on two-dimensional frequency domain sparse constraint
CN104345300B (en) The airborne non-working side battle array radar STAP method of clutter space-time spectrum linear compensation
CN103116162B (en) High-resolution sonar location method based on sparsity of objective space
CN103207387B (en) Method for quickly simulating airborne phased array pulse Doppler (PD) radar clutter
CN105137430A (en) Forward-looking array SAR echo sparse acquisition and three-dimensional imaging method
CN103383449A (en) ESPRIT algorithm based short-range clutter suppression method for airborne radar
CN105137424A (en) Real-beam scanning radar angular super-resolution method under clutter background
CN109613532B (en) Airborne radar real-time Doppler beam sharpening super-resolution imaging method
CN104166129A (en) Real beam radar iteration minimum mean square error angle super-resolution method
CN105301589A (en) High-resolution wide-swath SAR (synthetic aperture radar) ground motion object imaging method
CN103630886B (en) Based on ISAR goal decomposition and the reconstructing method of attribute scattering center model
CN104698431A (en) Method for estimating fussy component space angle and ambiguity-resolving multi-channel SAR (segmentation and resassembly sublayer) orientation
CN104698448A (en) Conformal array robust angle estimation method based on manifold separation under movement platform
CN105445704A (en) Radar moving object inhibition method in SAR image
Yang et al. Doppler beam sharpening imaging based on fast iterative adaptive approach
CN103630899A (en) Method for high-resolution radar compressed sensing imaging of moving object on ground
CN104330779A (en) Airborne synthetic aperture radar kinematic error compensating method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20170630

Termination date: 20200317