CN107015225B - A kind of SAR platform elemental height error estimation based on self-focusing - Google Patents
A kind of SAR platform elemental height error estimation based on self-focusing Download PDFInfo
- Publication number
- CN107015225B CN107015225B CN201710173161.4A CN201710173161A CN107015225B CN 107015225 B CN107015225 B CN 107015225B CN 201710173161 A CN201710173161 A CN 201710173161A CN 107015225 B CN107015225 B CN 107015225B
- Authority
- CN
- China
- Prior art keywords
- denoted
- sar
- radar
- population
- genetic algorithm
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S13/00—Systems 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
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9004—SAR image acquisition techniques
- G01S13/9017—SAR image acquisition techniques with time domain processing of the SAR signals in azimuth
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S13/00—Systems 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
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S13/00—Systems 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
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9004—SAR image acquisition techniques
- G01S13/9019—Auto-focussing of the SAR signals
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/40—Means for monitoring or calibrating
- G01S7/4052—Means for monitoring or calibrating by simulation of echoes
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Signal Processing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention proposes a kind of SAR platform elemental height error estimation based on self-focusing, it carries out Range compress to SAR original echoed signals first；Initialize rough estimate parameter, it divides scene objects observation section and carries out rear orientation projection (BP) imaging, image sharpness value is calculated using BP imaging results, elemental height error rough estimate evaluation is then obtained according to a preliminary estimate to elemental height error with genetic algorithm；Essence estimation parameter is adjusted using SAR platform elemental height error rough estimate evaluation, observation scene object space is repartitioned, carries out the estimation of elemental height error essence, finally obtain SAR platform elemental height error essence estimated value.It is compared with the traditional method, the present invention has the characteristics that calculation amount is smaller, the speed of service is fast and higher to SAR elemental height error estimation accuracy, therefore is more suitable for large scene, large slanting view angle machine, high-precision SAR imaging.
Description
Technical field
This technology invention belongs to Radar Technology field, its in particular to synthetic aperture radar (SAR) technical field of imaging.
Background technique
Synthetic aperture radar (SAR) is a kind of high-resolution microwave imaging radar, with round-the-clock and all weather operations
Advantage, be widely used every field, such as mapping, guidance, environmental remote sensing and resource exploration.SAR is applied important
Premise and the main target of signal processing are to obtain high-resolution, high-precision microwave imagery by imaging algorithm.Its height obtained
It differentiates microwave imagery and has been widely used in numerous areas, such as generation Digital height model, observation colcanism and big flood situation,
Monitor land and sea traffic etc..
Fast dive SAR (HSD-SAR) has very high application value, it can be applied to the necks such as civil aircraft navigation
Domain.It can be used for improving navigation accuracy by carrying out target identification, positioning and scene matching, SAR image.HSD-SAR is usual
Work is in high angle of squint state.Due to the high speed of HSD-SAR and the characteristic of high angle of squint, frequency domain imaging algorithm hardly results in focusing
The good wide swath HSD-SAR image of effect, and rear orientation projection's (BP) algorithm is by carrying out accurate for each pixel
With available good HSD-SAR image.In BP imaging algorithm, the relative position between target and observation scene must quilt
Precise measurement.
Rear orientation projection (BP) algorithm is a kind of accurate SAR time domain imaging algorithms, it is original by synthetic aperture radar first
Data, to Range compress (pulse compression) is carried out, then pass through any pixel in the different slow time observation spaces of selection along distance
Data after Range compress in SAR data space compensate orientation doppler phase, and carry out coherent accumulation, final to obtain
The imaging algorithm of each pixel scattering coefficient.Due to accurately known antenna phase center (Antenna Phase Center,
APC under the premise of), BP algorithm can with effective compensation kinematic error, because due to be widely used, be detailed in " Shi Jun double-base SAR
[D] University of Electronic Science and Technology doctoral thesis .2009 " is studied with linear array SAR principle and imaging technique.
Self-focusing technology is estimated using SAR data itself and removes irreducible phase errors.Wherein, the essence of phase error estimation and phase error
Degree is dependent on specific image scene and the self-focusing method of use.Existing self-focusing method can be divided into two major classes: parameter
Modelling, including sub-aperture correlation method (MD), multiple sub-apertures correlation method (MAM) and phase difference method (PD)；Nonparametric model method, packet
Include Phase-gradient autofocus algorithem (PGA).
Self-focusing BP algorithm is a kind of autofocus algorithm based on spatial domain picture quality, also be can be regarded as a kind of for BP
The motion compensation process of algorithm, main process are to optimize orientation phase compensation error vector according to image quality index, when
When image quality index is optimal, SAR image focuses best.Current main self-focusing BP algorithm has based on minimum image entropy
Self-focusing BP algorithm (be detailed in " M.Liu, C.S.Li, X.H.Shi, A back-projection fast autofocus
algorithm based on minimum entropy for SAR imaging[C].3rd APSAR
Conference.2011:1-4 "), in conjunction with self-focusing and rapid bp high-precision imaging algorithm (be detailed in " L.Zhang, H.L.Li,
Z.J.Qiao,M.D.Xing,Z.Bao,Integrating autofocus techniques with fast factorized
back-projection for high-resolution spotlight SAR imaging[J].IEEE Geoscience
And Remote Sensing Letters.2013,10 (6): self-focusing BP algorithm 1394-1398 ") and based on image sharpness
(it is detailed in " J.N.Ash, An autofocus method for back projection imagery in synthetic
aperture radar[J].IEEE Geoscience and Remote Sensing Letters.2012,9(1):104-
108").Wherein the self-focusing BP algorithm imaging effect based on image sharpness is best.
Hereditary (Genetic Algorithm, GA) algorithm is one of four big Main Branches of evolutionary computation, it is also nearly ten
The main evolution algorithm developed rapidly in remaining year.It is fast together with evolution strategy, evolutional programming and Genetic Programming
Speed develops and gradually moves towards fusion, forms a kind of computational theory of novel Simulating Evolution.Genetic algorithm is that one kind is searched at random
Rope algorithm.But it is again simultaneously a kind of process by iteration optimizing, has adaptive feature.
GPS (Global Positioning System) positioning system can provide relatively accurate for many civil fields
Target and scene between relative position, but be used to replace using inertial navigation system in many fields are for example guided
GPS positioning system.However the systematic error as accumulated in inertial navigation system, measured obtained target position will deposit
In thousands of meters of offset.The error of these positions especially high speed error will will lead in image exist defocus, positional shift and
Geometric distortion.Based on existing these problems, there is presented herein at the beginning of a kind of platform based on self-focusing and hereditary (GA) algorithm
Beginning height error (IAE) estimation method.
Summary of the invention
Occur defocusing since SAR initial platform height error will will lead to SAR image, positional shift and geometric distortion etc. are asked
The imaging precision of SAR image is inscribed and then influences, the invention proposes a kind of, and the SAR platform elemental height error based on self-focusing is estimated
Meter method, it mainly uses hereditary (GA) algorithm and rear orientation projection (BP) algorithm, with genetic algorithm to the elemental height of SAR platform
Error carries out estimation selection elemental height error optimization solution, using optimal solution as initial platform height error estimated value, this side
Method can efficiently solve defocused present in the high speed SAR image due to caused by elemental height error, geometric distortion, position it is inclined
The problem of shifting.
In order to facilitate the description contents of the present invention, make following term definition first:
Define 1, synthetic aperture radar (SAR)
Synthetic aperture radar be to be fixed on radar radar on loading movement platform, in conjunction with motion platform movement with
Synthesizing linear array with reach movement to resolution ratio, recycle radar beam to realize to echo delay apart from one-dimensional image, from
And realize a kind of synthetic aperture radar technique of observed object two-dimensional imaging.
Define 2, synthetic aperture radar echo data Range compress
Standard synthetic aperture distance by radar compression method refers to using synthetic aperture radar transmission signal parameters, using matching
Filtering technique carries out signal focus imaging process to signal to the distance of synthetic aperture radar.It is detailed in document " radar imagery skill
Art ", guarantor polished, Xing Mengdao, Wang Tong, Electronic Industry Press, 2005.
Define 3, norm
If X is number fieldLinear Space, whereinIndicate complex field, if it meets following property: | | X | | >=0, and | | X
| |=0 only X=0；| | aX | |=| a | | | X | |, a is arbitrary constant；||X1+X2||≤||X1||+||X2| |, then claim | | X | |
For the norm (norm) of X spatially, wherein X1And X2For any two value of X spatially.It is discrete for defining the dimension of N × 1 in 1
Signal vector X=[x1,x2,…,xN]T, the LP norm expression formula of vector X isWherein xiFor vector X
I-th of element, ∑ | | indicate absolute value summation operation symbol, the L1 norm expression formula of vector X isTo
Amount X L2 norm expression formula beThe L0 norm expression formula of vector X isAnd xi≠0.In detail
See document " matrix theory ", Huang Ting, which wishes, etc. writes, and Higher Education Publishing House publishes.
Define 4, synthetic aperture radar rear orientation projection (BP) imaging algorithm
Rear orientation projection's imaging algorithm of synthetic aperture radar, abbreviation BP imaging algorithm.BP imaging algorithm is first with radar
The trace information of platform finds out the history at a distance from scene pixel point of radar platform, then by finding out number of echoes apart from history
The corresponding complex data in carries out coherent accumulation after carrying out phase compensation to echo data, to obtain answering for the pixel
Image value.It is detailed in " Shi Jun, double-base SAR and linear array SAR principle and imaging technique [D] University of Electronic Science and Technology doctoral thesis
.2009”。
Define 5, orientation, distance to
By radar platform move direction be called orientation, will be perpendicular to orientation direction be called distance to.
The 6, distance of synthetic aperture radar is defined to fast moment and orientation slow moment
The distance of synthetic aperture radar to the fast moment refer to a radar system work pulse repetition period in, distance
The time interval variable of different sampled points during to echo signal sample.Polarization sensitive synthetic aperture radar system is with certain time length
Repetition period transmitting and return pulse signal, orientation slow moment indicate one using the pulse repetition period as the discretization of step-length when
Between variable, wherein each pulse repetition period discrete-time variable value be an orientation slow moment.It is detailed in document " synthesis hole
Diameter radar imagery principle ", Pi Yiming writes, and prospects society, University of Electronic Science and Technology publishes.
Define 7, population
Basic genetic algorithmic generates several groups of individuals using random device, which is collectively referred to as initial population.It is losing
In propagation algorithm, a population also just contains practical problem in the space of the solution of certain generation, and the set of possible solution.Population
The genetic evolution search space of search solution is provided for genetic algorithm.
Define 8, genetic algorithm (Genetic Algorithm)
Genetic algorithm (GA) is a kind of learning method inspired by biological evolution, and inherently one kind, which does not depend on, specifically asks
The direct search method of topic, it is to generate subsequent hypothesis by making a variation and recombinating currently known preferably hypothesis.Heredity is calculated
Breeding, intersection and the gene mutation phenomenon occurred in method simulation natural selection and natural genetic process, is all protected in each iteration
One group of candidate solution is stayed, and chooses optimal individual from solution population by certain index, utilizes genetic operator (selection operator, friendship
Fork operator and mutation operator) these individuals are combined, the candidate solution group of a new generation is generated, this process is repeated, until meeting
Until certain convergence index, specific implementation procedure can refer to document: " MATLAB GAs Toolbox and application ", thunder hero etc.
It writes, publishing house, Xian Electronics Science and Technology University.
Define 9, fitness function
Fitness function, which refers to, is used to distinguish individual in population quality according to what the objective function in optimization problem determined
Standard.The target function value of required problem is used only in genetic algorithm, so that it may obtain the related search information of next step.To target
The use of functional value is realized by the fitness of evaluation individual.
Define 10, generation gap rate
In genetic algorithm, a new population is selected and is recombinated generation by the individual to old population, if newly
The number of individuals of population is less than the size of initial population, and the difference of new population and old Population Size is referred to as generation gap, and difference is big
It is small to be then known as generation gap rate.
Define 11 self-focusings
Self-focusing technology is estimated using SAR data itself and removes irreducible phase errors.Wherein, the essence of phase error estimation and phase error
Degree is dependent on specific image scene and the self-focusing method of use.Self-focusing method is divided into parameter model, nonparametric model
Method, optimized parameter search method.The standard that optimized parameter search method is measured whether image focuses by setting one, in a certain section
It is upper to carry out error coefficient search to obtain the estimation of phase error, and then realize the focusing of image.
It is provided by the invention it is a kind of be based on self-focusing SAR platform elemental height error estimation, it the following steps are included:
Step 1, initialization SAR system parameter:
Initializing SAR system parameter includes: platform speed vector, is denoted as V；Radar initial position vector is denoted as P (0)；Thunder
Up to operating center frequency, it is denoted as fc；Radar carrier frequency wavelength, is denoted as λ；The signal bandwidth of radar emission baseband signal, is denoted as Br；Thunder
Up to transmitting signal pulse width, it is denoted as Tr；The chirp rate of radar emission signal, is denoted as fdr；The sampling frequency of Radar Receiver System
Rate is denoted as fs；The pulse recurrence frequency of radar emission system, is denoted as PRF；The aerial spread speed of electromagnetic wave, is denoted as C；
Distance is denoted as t, t=1,2 ... to the fast moment, Nr, NrIt is total to the fast moment for distance；At the orientation slow moment, it is denoted as l, l=1,
2,…,Na, NaFor the slow moment sum of orientation；Above-mentioned parameter is SAR system standard parameter, wherein radar center frequency fc, thunder
Up to carrier frequency wavelength X, the signal bandwidth B of radar emission baseband signalr, radar emission signal pulse width Tr, radar emission signal tune
Frequency slope fdr, radar received wave door continues width To, the sample frequency f of Radar Receiver Systems, the pulse weight of radar emission system
Complex frequency PRF has determined in linear array SAR system design process；Platform speed vector V, radar initial position vector P (0),
Distance has determined in the design of SAR observation program to the fast moment t and slow moment l of orientation；According to SAR imaging system scheme and
Observation program, the initialization imaging system parameters that SAR imaging method needs are known；SAR primary echo signals matrix is S；
Step 2, the observation scene object space parameter for initializing SAR:
Initialize the observation scene object space parameter of SAR, comprising: constituted with radar beam irradiation field areas ground level
Observation scene object space Ω of the two-dimensional space as SAR；Observation scene object space Ω is evenly dividing into equal-sized
Cell, unit grid are denoted as d in the direction x, the direction y side length respectivelyx、dy, it is traditional that cell size is selected as linear array SAR system
The half of theoretical imaging resolution；The coordinate vector for observing m-th of cell in scene object space Ω, is denoted as Pm, m table
Show that m-th of cell in observation scene object space Ω, m=1,2 ..., M, M are the cell observed in scene object space Ω
Sum；The scattering coefficient opsition dependent sequence of all cells rearranges vector in observation scene object space Ω, is denoted as α, to
Amount α is made of the column of M row 1；The scattering coefficient of m-th of element, is denoted as α in scattering coefficient vector αm；Observe scene object space Ω
It is had determined in SAR imaging conceptual design；
Step 3 carries out Range compress to raw radar data:
Distance is carried out to SAR primary echo signals S using SAR gauged distance compression method to compress to pulse, obtains distance
Compressed echo data is denoted as E, and wherein S is that step 1 initializes obtained SAR primary echo signals matrix；
Step 4, platform elemental height error rough estimate:
Step 4.1, initialization rough estimate parameter:
Initializing platform elemental height estimation error parameter includes: Population in Genetic Algorithms individual amount, is denoted as N1；Heredity is calculated
Method generation gap rate, is denoted as Gp1；Genetic algorithm maximum number of iterations, is denoted as Mg1；The sample territory of platform elemental height error rough estimate,
It is denoted as [- H, H]；Imaging is carried out using BP algorithm to need observation scene object space Ω being evenly dividing into equal-sized net
Lattice, grid are Nx in lateral division unit number scale1, gap size is denoted as Δ x1=10dx, it is in longitudinal division unit number scale
Ny1, gap size is denoted as Δ y1=10dy, observed object space is divided into Nx1Row Ny1The two-dimensional grid of column, wherein Ω is step
The observation scene object spaces of rapid 2 definition, wherein dxThe cell defined for step 2 is in the side length in the direction x, wherein dyFor step 2
The cell of definition is in the direction y side length；
Step 4.2 is imaged using BP algorithm, and calculates image sharpness value:
According to the platform speed vector V initialized in step 1, radar initial position vector P (0) and radar emission system
Pulse recurrence frequency PRF, using formula Pc(l)=P (0)+Vl/PRF, l=1,2 ..., Na, radar is calculated at first
The position vector at orientation slow moment, as the measurement antenna phase center of radar, are denoted as Pc, Pc=[Pc(1),Pc(2),…,Pc
(Na)]；
Utilize Nx1、Ny1、dx、dy, according to formula Pai1=(i-Nx1/2)*Δx1、Paj1=(j-Ny1/2)*Δy1, calculate
I-th of the direction x of object space after to division, j-th of the direction y mesh point position (Pai1,Paj1), in order by grid
The position vector of point, which is arranged successively, forms a vector, the grid point locations vector after as repartitioning object space, as
(Pax1,Pay1), wherein Nx1The grid defined for step 4.1 is in lateral division unit number, wherein Ny1It is defined for step 4.1
The division unit number of grid longitudinal direction, wherein Δ x1The cell defined for step 4.1 is in the side length in the direction x, wherein Δ y1For step
The cell of 4.1 definition is in the direction y side length；
Utilize the antenna phase center P of measurementc, grid point locations (Pax1,Pay1) and number of echoes after Range compress
According to E, it is imaged with traditional synthetic aperture radar rear orientation projection-BP algorithm, obtains SAR image data, be denoted as B1, B1For Nx1
Row Ny1The two-dimensional complex number matrix of column, wherein E is the echo data after the initial SAR echo signal Range compress that step 3 obtains；
Using formulaThe acutance value function of SAR image is calculated, wherein | |4Indicate multiple to one
4 powers after number modulus；
Step 4.3 carries out rough estimate to platform elemental height error using genetic algorithm:
Step 4.3.1: according to N1D is denoted as using traditional genetic algorithm random initializtion population with [- H, H]0, wherein
N1Obtained population at individual number is initialized for step 4.1, wherein [- H, H] is that the platform that step 4.1 initialization obtains is initially high
Spend the sample territory of error rough estimate；
Step 4.3.2: initial time genetic algorithm the number of iterations is denoted as gen1；
Step 4.3.3: according to formula J1=-f1Population at individual fitness function in genetic algorithm is defined, is planted in genetic algorithm
Group's individual adaptation degree function is denoted as J1, wherein f1It is the SAR image sharpness value that step 4.2 obtains；
Step 4.3.4: according to J1With Gp1, using traditional genetic algorithm selection operator to D0Selection operation is carried out, is obtained
More excellent population D1, wherein J1For the population at individual fitness function that step 4.3.3 is obtained, wherein Gp1It is initialized for step 4.1
The genetic algorithm generation gap rate function arrived, wherein D0The initialization population initialized for step 4.3.1；
Step 4.3.5: using the crossover operator in traditional genetic algorithm to D1Operation of reporting to the leadship after accomplishing a task is carried out, then again to reporting to the leadship after accomplishing a task
The population obtained after the completion of operation carries out the mutation operation of traditional genetic algorithm, obtains new population D2, wherein D1For step
4.3.4 the more excellent population obtained；
Step 4.3.6: termination condition judgement, if gen1Meet gen1< Mg1, then repeat step 4.3.4~step
And gen 4.3.51=gen1+1；Work as gen1=Mg1When, step 4.3.7 is gone to, wherein gen1The initialization defined for step 4.3.2
Genetic algorithm the number of iterations, wherein Mg1Obtained maximum number of iterations is initialized for step 4.1,；
Step 4.3.7: after terminating iteration, optimal estimation individual, as platform elemental height error rough estimate evaluation, meter are obtained
For V1；Step 5, platform elemental height error high-precision are estimated:
Step 5.1, initialization high-precision estimation parameter:
Population in Genetic Algorithms individual amount, is denoted as N2；Genetic algorithm generation gap rate, is denoted as Gp2；Maximum number of iterations is denoted as
Mg2；The sample territory that the high-precision estimation of platform elemental height error is adjusted according to platform elemental height error rough estimate evaluation, is denoted as
[V1-h,V1+ h], wherein V1The platform elemental height error rough estimate evaluation estimated for step 4.3.7；It is carried out using BP algorithm
Imaging needs observation scene object space Ω being evenly dividing into equal-sized grid, and grid is in lateral division unit number scale
For Nx2Gap size is denoted as Δ x2=2dx, it is Ny in longitudinal division unit number scale2, gap size is denoted as Δ y2=2dy, in this way
Observed object space is just divided into Nx2Row Ny2The two-dimensional grid of column is imaged for next BP, and wherein Ω is fixed for step 2
The observation scene object space of justice；
Step 5.2 is imaged using BP algorithm, and calculates image sharpness value:
According to formula Pai2=(i-Nx2/2)*Δx2、Paj2=(j-Ny2/2)*Δy2, the mesh after repartitioning is calculated
Mark i-th of the direction x in space, the position (Pa of j-th of the direction y mesh pointi2,Paj2), in order by the position vector of mesh point
It is arranged successively one vector of composition, the grid point locations vector after as repartitioning object space, as (Pax2,Pay2),
Middle Nx2The grid defined for step 5.1 is in lateral division unit number, wherein Ny2For stroke for the grid longitudinal direction that step 5.1 defines
Sub-unit number, wherein Δ x2The cell defined for step 5.1 is in the side length in the direction x, wherein Δ y2The list defined for step 5.1
First lattice are in the direction y side length；
Utilize the antenna phase center P of measurementc, grid point locations (Pax2,Pay2) and Range compress after echo data E,
It is imaged with traditional synthetic aperture radar rear orientation projection-BP algorithm, obtains SAR image data, be denoted as B2, B2For Nx2Row
Ny2The two-dimensional complex number matrix of column, wherein E is the echo data after the initial SAR echo signal Range compress that step 3 obtains, PcFor
The antenna phase center position that step 4.2 obtains；
Using formulaThe acutance value function of SAR image is calculated, wherein | |4Indicate multiple to one
4 powers after number modulus；
Step 5.3 carries out rough estimate to platform elemental height error using genetic algorithm:
Step 5.3.1: according to N2With [V1-h,V1+ h], using traditional genetic algorithm random initializtion population, it is denoted as G0,
Wherein N2Obtained population at individual number is initialized for step 5.1, wherein [V1-h,V1+ h] it is put down for what step 5.1 initialization obtained
The sample territory of platform elemental height error rough estimate；
Step 5.3.2: initial time genetic algorithm the number of iterations is denoted as gen2；
Step 5.3.3: according to formula J2=-f2Population at individual fitness function in genetic algorithm is defined, is planted in genetic algorithm
Group's individual adaptation degree function is denoted as J2, wherein f2The SAR image sharpness value that step 5.2 obtains；
Step 5.3.4: according to J2With Gp2, using the selection operator in traditional genetic algorithm, to G0Selection operation is carried out to obtain
To more excellent population G1, wherein J2For the population at individual fitness function that step 5.3.3 is obtained, wherein Gp2For step 5.1 initialization
Obtained genetic algorithm generation gap rate function, wherein G0The initialization population initialized for step 5.3.1；
Step 5.3.5: using the crossover operator in traditional genetic algorithm to G1Report to the leadship after accomplishing a task after operation, operated to reporting to the leadship after accomplishing a task
The population obtained after carries out mutation operation in traditional genetic algorithm, obtains new population G2Wherein G1It is obtained for step 5.3.4
More excellent population；
Step 5.3.6: termination condition judgement, if gen2Meet gen2< Mg2, then repeat step 5.3.4~step
And gen 5.3.52=gen2+1；Work as gen2=Mg2When, step 5.3.7 is gone to, wherein gen2The initialization defined for step 5.3.2
Genetic algorithm the number of iterations, wherein Mg2Obtained maximum number of iterations is initialized for step 5.1；
Step 5.3.7: after terminating iteration, optimal estimation individual, as platform elemental height error rough estimate evaluation, meter are obtained
For V2；
So far, we have obtained the final estimated value of platform elemental height error.
The main thought of the method for the present invention is: after carrying out Range compress to original echoed signals, being drawn using rough estimate parameter
Divide after observing scene object space compressed signal of adjusting the distance to carry out SAR rear orientation projection-BP imaging, passes through and calculate image sharpness
Fitness function needed for obtaining genetic algorithm；Then elemental height error is obtained initially according to a preliminary estimate with genetic algorithm
Height error rough estimate evaluation；Essence estimation parameter is adjusted using SAR platform elemental height error rough estimate evaluation, repartitions observation field
Scape object space carries out the estimation of elemental height error essence, finally obtains SAR platform elemental height error essence estimated value.
The advantage of the invention is that estimating using autofocus algorithm SAR platform elemental height error, pass through calculating
Elemental height error is estimated using genetic algorithm after image sharpness, available SAR initial rapid error essence estimated value,
Calculation amount of the present invention is smaller, and the speed of service is fast, and higher to SAR elemental height error estimation accuracy, therefore is more suitable for big
Scene, large slanting view angle machine, high-precision SAR imaging.
Detailed description of the invention
Fig. 1 is flow chart of the present invention；
Specific embodiment
The method that the present invention mainly uses Computer Simulation is verified, and all steps, conclusion are all in MATLAB-R2012b
Upper verifying is correct.Specific implementation step is as follows:
Step 1, initialization SAR system parameter:
Initializing SAR system parameter includes: platform speed vector, is denoted as V=[740；1100；-1500]m/s；At the beginning of radar
Beginning position vector is denoted as P (0)=[- 7001.8；41000；57564]m；Radar operating center frequency, is denoted as fc=60 ×
109Hz；Radar carrier frequency wavelength, is denoted as λ=0.005m；The signal bandwidth of radar emission baseband signal, is denoted as Br=100MHz；Thunder
Up to transmitting signal pulse width, it is denoted as Tr=10 μ s；The chirp rate of radar emission signal, is denoted as fdr=100 × 1011Hz/s；
The sample frequency of Radar Receiver System, is denoted as fs=120MHz；The pulse recurrence frequency of radar emission system, is denoted as PRF=
5848Hz；The aerial spread speed of electromagnetic wave, is denoted as C=3 × 108m/s；Distance is denoted as t, t=1 to the fast moment,
2,…,Nr, Nr=4096 is total to the fast moment for distance；At the orientation slow moment, it is denoted as l, l=1,2 ..., Na, Na=2924 are
The slow moment sum of orientation；Above-mentioned parameter is SAR system standard parameter, wherein radar center frequency fc, radar carrier frequency wavelength
λ, the signal bandwidth B of radar emission baseband signalr, radar emission signal pulse width Tr, radar emission signal chirp rate fdr,
Radar received wave door continues width To, the sample frequency f of Radar Receiver Systems, the pulse recurrence frequency PRF of radar emission system
It is had determined in linear array SAR system design process；Platform speed vector V, radar initial position vector P (0), distance to it is fast when
The t and slow moment l of orientation is carved to have determined in the design of SAR observation program；According to SAR imaging system scheme and observation program,
The initialization imaging system parameters that SAR imaging method needs are known；SAR primary echo signals matrix is S；
Step 2, the observation scene object space parameter for initializing SAR:
Initialize the observation scene object space parameter of SAR, comprising: constituted with radar beam irradiation field areas ground level
Observation scene object space Ω of the two-dimentional rectangular co-ordinate as SAR；Initialization observation scene object space Ω size be 50 ×
The centre coordinate position of 50 × 1 pixels, observation scene object space Ω is located at [0,0,0], and observation scene object space Ω is equal
Even to be divided into equal-sized cell, unit grid is denoted as d in transverse direction, longitudinal side length respectivelyx=0.5m, dy=0.5m, dz=
0.5m, it is M=20000 that observation scene object space cell sum M, which is calculated,；It observes in scene object space Ω m-th
The coordinate vector of cell, is denoted as Pm, m indicate observation scene object space Ω in m-th of cell, m=1,2 ..., 20000M
For the cell sum in observation scene object space Ω；The scattering coefficient of all cells is pressed in observation scene object space Ω
Sequence of positions rearranges vector, is denoted as α, and vector α is made of 20000 rows 1 column；M-th element dissipates in scattering coefficient vector α
Coefficient is penetrated, α is denoted asm；Observation scene object space Ω has determined in SAR imaging conceptual design；
Step 3 carries out Range compress to raw radar data:
Distance is carried out to SAR primary echo signals S using SAR gauged distance compression method to compress to pulse, obtains distance
Compressed echo data is denoted as E, and wherein S is that step 1 initializes obtained SAR primary echo signals matrix；
Step 4, platform elemental height error rough estimate:
Step 4.1, initialization rough estimate parameter:
Initializing platform elemental height estimation error parameter includes: Population in Genetic Algorithms individual amount, is denoted as N1=5；It loses
Propagation algorithm generation gap rate, is denoted as Gp1=1；Genetic algorithm maximum number of iterations, is denoted as Mg1=5；Platform elemental height error rough estimate
Sample territory, be denoted as [- 3000,3000]；Imaging is carried out using BP algorithm to need for observation scene object space Ω to be evenly dividing
At equal-sized grid, grid is Nx in lateral division unit number scale1=10, gap size is denoted as Δ x1=10dx=5,
It is Ny in longitudinal division unit number scale1=10, gap size is denoted as Δ y1=10dy=5, thus observed object space is drawn
It is divided into the two-dimensional grid of 10 rows 10 column, is imaged for next standard BP, wherein Ω is the observation scene objects that step 2 defines
Space, wherein dx=0.5 cell defined for step 2 is in the side length in the direction x, wherein dy=0.5 unit defined for step 2
Lattice are in the direction y side length；
Step 4.2 is imaged using BP algorithm, and calculates image sharpness value:
According to the platform speed vector V initialized in step 1, radar initial position vector P (0) and radar emission system
Pulse recurrence frequency PRF, using formula Pc(l)=P (0)+Vl/PRF, l=1,2 ..., 2924, radar is calculated in l
The position vector at a orientation slow moment, as the measurement antenna phase center of radar, are denoted as Pc, Pc=[Pc(1),Pc
(2),…,Pc(2924)] Nx, is utilized1、Ny1、dx、dy, according to formula Pai1=(i-Nx1/2)*Δx1、Paj1=(j-Ny1/2)*
Δy1, i-th of the direction x of the object space after dividing, the position (Pa of j-th of the direction y mesh point is calculatedi1,Paj1), it presses
The position vector of mesh point is stored the grid point locations vector at a vector, after as repartitioning object space by sequence,
As (Pax1,Pay1), wherein Nx1=10 be the grid of step 4.1 definition in lateral division unit number, Ny1=10 be step
The division unit number of the grid longitudinal directions of 4.1 definition, wherein Δ x1=5 be side length of the cell in the direction x of step 4.1 definition,
Wherein Δ y1=5 be the cell of step 4.1 definition in the direction y side length；Utilize the antenna phase center P of measurementc, grid point
Set (Pax1,Pay1) and echo data E after Range compress, it is imaged with BP algorithm, obtains SAR image data, be denoted as
B1, B1For Nx1Row Ny1The two-dimensional complex number matrix of column, after wherein E is the obtained initial SAR echo signal Range compress of step 3
Echo data is denoted as B1, B1For the two-dimensional complex number matrix of 10 rows 10 column, wherein E is that step 3 initializes obtained initial SAR echo
The compressed echo data of signal distance；Using formulaThe sharpness value of SAR image is calculated, wherein |
|4It indicates to 4 powers after a plural modulus；
Step 4.3 carries out rough estimate to platform elemental height error using genetic algorithm:
Step 4.3.1: according to N1It is denoted as with [- 3000,3000] using traditional genetic algorithm random initializtion population
D0, wherein N1=5 initialize obtained population at individual number for step 4.1, wherein [- 3000,3000] are step 4.1 initialization
The sample territory of obtained platform elemental height error rough estimate；
Step 4.3.2: initial time genetic algorithm the number of iterations is denoted as gen1=0；
Step 4.3.3: according to formula J1=-f1Population at individual fitness function in genetic algorithm is defined, is planted in genetic algorithm
Group's individual adaptation degree function is denoted as J1, wherein f1The SAR image sharpness value that step 4.2 obtains；
Step 4.3.4: according to J1With Gp1, using traditional genetic algorithm selection operator to D0Selection operation is carried out to obtain more
Excellent population D1, wherein J1For the population at individual fitness function that step 4.3.3 is obtained, wherein Gp1=1 initializes for step 4.1
The genetic algorithm generation gap rate function arrived, wherein D0The initialization population initialized for step 4.3.1；
Step 4.3.5: using the crossover operator in traditional genetic algorithm to D1Report to the leadship after accomplishing a task after operation, to operation of reporting to the leadship after accomplishing a task
The mutation operation that the population obtained after the completion carries out in traditional genetic algorithm obtains new population D2Wherein D1For step 4.3.4
Obtained more excellent population；
Step 4.3.6: termination condition judgement, if gen1Meet gen1< Mg1, then repeat step 4.3.4~step
And gen 4.3.51=gen1+1；Work as gen1=Mg1When, step 4.3.7 is gone to, wherein gen1=0 for step 4.3.2 define just
Beginning time genetic algorithm the number of iterations, wherein Mg1=5 initialize obtained maximum number of iterations for step 4.1；
Step 4.3.7: after terminating iteration, searching out optimal estimation individual, as platform elemental height error rough estimate evaluation,
It is calculated as V1；
Step 5, platform elemental height error high-precision are estimated:
Step 5.1, initialization high-precision estimation parameter:
Population in Genetic Algorithms individual amount, is denoted as N2=5；Genetic algorithm generation gap rate, is denoted as Gp2=1；Maximum number of iterations,
It is denoted as Mg2=5；Utilize the sample of platform elemental height error rough estimate evaluation adjustment platform elemental height error high-precision estimation
Domain is denoted as [V1-500,V1+ 500], wherein V1The platform elemental height error rough estimate evaluation estimated for step 4.3.7；It will
Observation scene object space Ω is evenly dividing into equal-sized grid, and grid is Nx in lateral division unit number scale2=50,
Gap size is denoted as Δ x2=2dx=1, it is Ny in longitudinal division unit number scale2=50, gap size is denoted as Δ y2=2dy=
1, observed object space is thus divided into the two-dimensional grid that 50 rows 50 arrange, is imaged for next standard rear orientation projection,
Wherein Ω is the observation scene object space that step 2 defines；
Step 5.2 is imaged using BP algorithm, and calculates image sharpness value:
According to formula Pai2=(i-Nx2/2)*Δx2、Paj2=(j-Ny2/2)*Δy2, the mesh after repartitioning is calculated
Mark i-th of the direction x in space, the position (Pa of j-th of the direction y mesh pointi2,Paj2), in order by the position vector of mesh point
Store into a vector, the grid point locations vector after as repartitioning object space, as (Pax2,Pay2), wherein Nx2=
50 grids defined for step 5.1 are in lateral division unit number, wherein Ny2=50 grid longitudinal directions defined for step 5.1
Division unit number, wherein Δ x2=1 cell defined for step 5.1 is in the side length in the direction x, wherein Δ y2=1 is step 5.1
The cell of definition is in the direction y side length；Utilize the antenna phase center P of measurementc, grid point locations (Pax2,Pay2) and apart from pressure
Echo data E after contracting, is imaged with BP algorithm, is obtained SAR image data, is denoted as B2, B2For the two-dimensional complex number of 50 rows 50 column
Matrix, wherein E is the echo data after the initial SAR echo signal Range compress that step 3 initialization obtains, PcFor step 4.2
Initialize obtained antenna phase center position；Using formulaThe acutance value function of SAR image is calculated,
Wherein | |4It indicates to 4 powers after a plural modulus；
Step 5.3 carries out rough estimate to platform elemental height error using genetic algorithm:
Step 5.3.1: according to N2With [V1-500,V1+ 500], using traditional genetic algorithm random initializtion population, note
For G0, wherein N2=5 initialize obtained population at individual number for step 5.1, wherein [V1-500,V1+ 500] at the beginning of step 5.1
The sample territory for the platform elemental height error essence estimation that beginningization obtains；
Step 5.3.2: initial time genetic algorithm the number of iterations is denoted as gen2=0；
Step 5.3.3: according to formula J2=-f2Population at individual fitness function in genetic algorithm is defined, is planted in genetic algorithm
Group's individual adaptation degree function is denoted as J2, wherein f2The SAR image sharpness value that step 5.2 obtains；
Step 5.3.4: according to J2With Gp2=1, using the selection operator of traditional genetic algorithm, to G0Carry out selection operation
Obtain more excellent population G1, wherein J2For the population at individual fitness function that step 5.3.3 is obtained, wherein Gp2It is initial for step 5.1
Change obtained genetic algorithm generation gap rate function, wherein G0The initialization population initialized for step 5.3.1；
Step 5.3.5: using the crossover operator of traditional genetic algorithm to G1Report to the leadship after accomplishing a task after operation, operated to reporting to the leadship after accomplishing a task
The population obtained after carries out mutation operation and obtains new population G2, wherein G1The more excellent population obtained for step 5.3.4；
Step 5.3.6: termination condition judgement, if gen2Meet gen2< Mg2=5, then repeat step 5.3.4~step
Rapid 5.3.5 and gen2=gen2+1；Work as gen2When=5, step 5.3.7 is gone to, wherein gen2=0 for step 5.3.2 define just
Beginning time genetic algorithm the number of iterations, wherein Mg2=5 initialize obtained maximum number of iterations for step 5.1；
Step 5.3.7: after terminating iteration, searching out optimal estimation individual, as platform elemental height error rough estimate evaluation,
It is calculated as V2；
So far, we have obtained the final estimated value of platform elemental height error, and entire method terminates.
It is proved by computer artificial result, the present invention calculates image sharpness function by the method for self-focusing, and utilizes
Genetic algorithm tests SAR elemental height estimation error, realizes the essence estimation to SAR platform elemental height error, the present invention can
With quick and high-precision estimation SAR platform elemental height error.
Claims (1)
1. one kind be based on self-focusing SAR platform elemental height error estimation, it is characterized in that it the following steps are included:
Step 1, initialization SAR system parameter:
Initializing SAR system parameter includes: platform speed vector, is denoted as V；Radar initial position vector is denoted as P (0)；Radar work
Make centre frequency, is denoted as fc；Radar carrier frequency wavelength, is denoted as λ；The signal bandwidth of radar emission baseband signal, is denoted as Br；Radar hair
Signal pulse width is penetrated, T is denoted asr；The chirp rate of radar emission signal, is denoted as fdr；The sample frequency of Radar Receiver System, note
It is fs；The pulse recurrence frequency of radar emission system, is denoted as PRF；The aerial spread speed of electromagnetic wave, is denoted as C；Distance to
At the fast moment, it is denoted as t, t=1,2 ..., Nr, NrIt is total to the fast moment for distance；At the orientation slow moment, it is denoted as l, l=1,2 ...,
Na, NaFor the slow moment sum of orientation；Above-mentioned parameter is SAR system standard parameter, wherein radar center frequency fc, radar load
Frequency wavelength X, the signal bandwidth B of radar emission baseband signalr, radar emission signal pulse width Tr, radar emission signal frequency modulation is oblique
Rate fdr, radar received wave door continues width To, the sample frequency f of Radar Receiver Systems, the pulse repetition frequency of radar emission system
Rate PRF has determined in linear array SAR system design process；Platform speed vector V, radar initial position vector P (0), distance
It is had determined in the design of SAR observation program to the fast moment t and slow moment l of orientation；According to SAR imaging system scheme and observation
Scheme, the initialization imaging system parameters that SAR imaging method needs are known；SAR primary echo signals matrix is S；
Step 2, the observation scene object space parameter for initializing SAR:
Initialize the observation scene object space parameter of SAR, comprising: the two dimension constituted with radar beam irradiation field areas ground level
Observation scene object space Ω of the space as SAR；Observation scene object space Ω is evenly dividing into equal-sized unit
Lattice, unit grid are denoted as d in the direction x, the direction y side length respectivelyx、dy, cell size is selected as linear array SAR system traditional theory
The half of imaging resolution；The coordinate vector for observing m-th of cell in scene object space Ω, is denoted as Pm, m expression sight
M-th of cell in the Ω of scene objects space, m=1,2 ..., M are surveyed, M is that the cell in observation scene object space Ω is total
Number；The scattering coefficient opsition dependent sequence of all cells rearranges vector in observation scene object space Ω, is denoted as α, vector α
It is made of the column of M row 1；The scattering coefficient of m-th of element, is denoted as α in scattering coefficient vector αm；Observation scene object space Ω exists
It is had determined in SAR imaging conceptual design；
Step 3 carries out Range compress to raw radar data:
Distance is carried out to SAR primary echo signals S using SAR gauged distance compression method to compress to pulse, obtains Range compress
Echo data afterwards is denoted as E, and wherein S is that step 1 initializes obtained SAR primary echo signals matrix；
Step 4, platform elemental height error rough estimate:
Step 4.1, initialization rough estimate parameter:
Initializing platform elemental height estimation error parameter includes: Population in Genetic Algorithms individual amount, is denoted as N1；Genetic algorithm generation
Ditch rate, is denoted as Gp1；Genetic algorithm maximum number of iterations, is denoted as Mg1；The sample territory of platform elemental height error rough estimate, is denoted as
[-H,H]；Imaging is carried out using BP algorithm to need observation scene object space Ω being evenly dividing into equal-sized grid, net
Lattice are Nx in lateral division unit number scale1, gap size is denoted as Δ x1=10dx, it is Ny in longitudinal division unit number scale1,
Gap size is denoted as Δ y1=10dy, observed object space is divided into Nx1Row Ny1The two-dimensional grid of column, wherein Ω is step 2
The observation scene object space of definition, wherein dxThe cell defined for step 2 is in the side length in the direction x, wherein dyIt is fixed for step 2
The cell of justice is in the direction y side length；
Step 4.2 is imaged using BP algorithm, and calculates image sharpness value:
According to the platform speed vector V initialized in step 1, the pulse of radar initial position vector P (0) and radar emission system
Repetition rate PRF, using formula Pc(l)=P (0)+Vl/PRF, l=1,2 ..., Na, radar is calculated in first of orientation
To the position vector at slow moment, as the measurement antenna phase center of radar, it is denoted as Pc, Pc=[Pc(1),Pc(2),…,Pc
(Na)]；
Utilize Nx1、Ny1、dx、dy, according to formula Pai1=(i-Nx1/2)*Δx1、Paj1=(j-Ny1/2)*Δy1, it is calculated and draws
I-th of the direction x of object space after point, j-th of the direction y mesh point position (Pai1,Paj1), in order by mesh point
Position vector is arranged successively one vector of composition, the grid point locations vector after as repartitioning object space, as (Pax1,
Pay1), wherein Nx1The grid defined for step 4.1 is in lateral division unit number, wherein Ny1The grid defined for step 4.1
Longitudinal division unit number, wherein Δ x1The cell defined for step 4.1 is in the side length in the direction x, wherein Δ y1For step 4.1
The cell of definition is in the direction y side length；
Utilize the antenna phase center P of measurementc, grid point locations (Pax1,Pay1) and echo data E after Range compress,
It is imaged with traditional synthetic aperture radar rear orientation projection-BP algorithm, obtains SAR image data, be denoted as B1, B1For Nx1Row
Ny1The two-dimensional complex number matrix of column, wherein E is the echo data after the initial SAR echo signal Range compress that step 3 obtains；
Using formulaThe acutance value function of SAR image is calculated, wherein | |4Expression takes a plural number
4 powers after mould；
Step 4.3 carries out rough estimate to platform elemental height error using genetic algorithm:
Step 4.3.1: according to N1D is denoted as using traditional genetic algorithm random initializtion population with [- H, H]0, wherein N1For
Step 4.1 initializes obtained population at individual number, wherein [- H, H] is that the platform elemental height that step 4.1 initialization obtains misses
The sample territory of poor rough estimate；
Step 4.3.2: initial time genetic algorithm the number of iterations is denoted as gen1；
Step 4.3.3: according to formula J1=-f1Population at individual fitness function in genetic algorithm is defined, population in genetic algorithm
Body fitness function is denoted as J1, wherein f1It is the SAR image sharpness value that step 4.2 obtains；
Step 4.3.4: according to J1With Gp1, using traditional genetic algorithm selection operator to D0Selection operation is carried out, is obtained more excellent
Population D1, wherein J1For the population at individual fitness function that step 4.3.3 is obtained, wherein Gp1It is obtained for step 4.1 initialization
Genetic algorithm generation gap rate function, wherein D0The initialization population initialized for step 4.3.1；
Step 4.3.5: using the crossover operator in traditional genetic algorithm to D1Operation of reporting to the leadship after accomplishing a task is carried out, then again to operation of reporting to the leadship after accomplishing a task
The population obtained after the completion carries out the mutation operation of traditional genetic algorithm, obtains new population D2, wherein D1For step 4.3.4
Obtained more excellent population；
Step 4.3.6: termination condition judgement, if gen1Meet gen1< Mg1, then repeat step 4.3.4~step 4.3.5
And gen1=gen1+1；Work as gen1=Mg1When, step 4.3.7 is gone to, wherein gen1The initialization heredity defined for step 4.3.2
Algorithm iteration number, wherein Mg1Obtained maximum number of iterations is initialized for step 4.1,；
Step 4.3.7: after terminating iteration, optimal estimation individual, as platform elemental height error rough estimate evaluation is obtained, V is calculated as1；
Step 5, platform elemental height error high-precision are estimated:
Step 5.1, initialization high-precision estimation parameter:
Population in Genetic Algorithms individual amount, is denoted as N2；Genetic algorithm generation gap rate, is denoted as Gp2；Maximum number of iterations is denoted as Mg2；Root
According to the sample territory of platform elemental height error rough estimate evaluation adjustment platform elemental height error high-precision estimation, it is denoted as [V1-h,V1+
H], wherein V1The platform elemental height error rough estimate evaluation estimated for step 4.3.7；Imaging needs are carried out using BP algorithm
Observation scene object space Ω is evenly dividing into equal-sized grid, grid is Nx in lateral division unit number scale2Between
Δ x is denoted as every size2=2dx, it is Ny in longitudinal division unit number scale2, gap size is denoted as Δ y2=2dy, will thus see
It surveys object space and is divided into Nx2Row Ny2The two-dimensional grid of column is imaged for next BP, and wherein Ω is the sight that step 2 defines
Survey scene objects space；
Step 5.2 is imaged using BP algorithm, and calculates image sharpness value:
According to formula Pai2=(i-Nx2/2)*Δx2、Paj2=(j-Ny2/2)*Δy2, the target empty after repartitioning is calculated
Between i-th of the direction x, j-th of the direction y mesh point position (Pai2,Paj2), in order successively by the position vector of mesh point
Rearrange a vector, the grid point locations vector after as repartitioning object space, as (Pax2,Pay2), wherein Nx2
The grid defined for step 5.1 is in lateral division unit number, wherein Ny2For the division list for the grid longitudinal direction that step 5.1 defines
First number, wherein Δ x2The cell defined for step 5.1 is in the side length in the direction x, wherein Δ y2The cell defined for step 5.1
In the direction y side length；
Utilize the antenna phase center P of measurementc, grid point locations (Pax2,Pay2) and Range compress after echo data E, with biography
Synthetic aperture radar rear orientation projection-BP algorithm of system is imaged, and is obtained SAR image data, is denoted as B2, B2For Nx2Row Ny2Column
Two-dimensional complex number matrix, wherein E is the echo data after the obtained initial SAR echo signal Range compress of step 3, PcFor step
4.2 obtained antenna phase center positions；
Using formulaThe acutance value function of SAR image is calculated, wherein | |4Expression takes a plural number
4 powers after mould；
Step 5.3 carries out rough estimate to platform elemental height error using genetic algorithm:
Step 5.3.1: according to N2With [V1-h,V1+ h], using traditional genetic algorithm random initializtion population, it is denoted as G0, wherein
N2Obtained population at individual number is initialized for step 5.1, wherein [V1-h,V1+ h] it is at the beginning of step 5.1 initializes obtained platform
The sample territory of beginning height error rough estimate；
Step 5.3.2: initial time genetic algorithm the number of iterations is denoted as gen2；
Step 5.3.3: according to formula J2=-f2Population at individual fitness function in genetic algorithm is defined, population in genetic algorithm
Body fitness function is denoted as J2, wherein f2The SAR image sharpness value that step 5.2 obtains；
Step 5.3.4: according to J2With Gp2, using the selection operator in traditional genetic algorithm, to G0Selection operation is carried out to obtain more
Excellent population G1, wherein J2For the population at individual fitness function that step 5.3.3 is obtained, wherein Gp2It is obtained for step 5.1 initialization
Genetic algorithm generation gap rate function, wherein G0The initialization population initialized for step 5.3.1；
Step 5.3.5: using the crossover operator in traditional genetic algorithm to G1Report to the leadship after accomplishing a task after operation, after the completion of operation of reporting to the leadship after accomplishing a task
Obtained population carries out mutation operation in traditional genetic algorithm, obtains new population G2Wherein G1It is obtained for step 5.3.4 more excellent
Population；
Step 5.3.6: termination condition judgement, if gen2Meet gen2< Mg2, then repeat step 5.3.4~step 5.3.5
And gen2=gen2+1；Work as gen2=Mg2When, step 5.3.7 is gone to, wherein gen2The initialization heredity defined for step 5.3.2
Algorithm iteration number, wherein Mg2Obtained maximum number of iterations is initialized for step 5.1；
Step 5.3.7: after terminating iteration, optimal estimation individual, as platform elemental height error rough estimate evaluation is obtained, V is calculated as2。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710173161.4A CN107015225B (en) | 2017-03-22 | 2017-03-22 | A kind of SAR platform elemental height error estimation based on self-focusing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710173161.4A CN107015225B (en) | 2017-03-22 | 2017-03-22 | A kind of SAR platform elemental height error estimation based on self-focusing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107015225A CN107015225A (en) | 2017-08-04 |
| CN107015225B true CN107015225B (en) | 2019-07-19 |
Family
ID=59440725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710173161.4A Expired - Fee Related CN107015225B (en) | 2017-03-22 | 2017-03-22 | A kind of SAR platform elemental height error estimation based on self-focusing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107015225B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108614249B (en) * | 2018-04-12 | 2021-06-11 | 北京航空航天大学 | Phase error estimation method, device, compensation method and system |
| CN109752698A (en) * | 2018-12-12 | 2019-05-14 | 北京无线电测量研究所 | A kind of inertial navigation method for estimating error of airborne synthetic aperture radar |
| CN110109107B (en) * | 2019-04-24 | 2022-05-31 | 电子科技大学 | Motion error compensation method of synthetic aperture radar frequency domain BP algorithm |
| CN110221263B (en) * | 2019-07-03 | 2021-12-14 | 北京电子工程总体研究所 | Error estimation method and system for multi-sensor system |
| CN112115642B (en) * | 2020-09-14 | 2023-05-02 | 四川航天燎原科技有限公司 | SAR imaging parameter optimization design method for high maneuvering platform |
| CN113255887A (en) * | 2021-05-25 | 2021-08-13 | 上海机电工程研究所 | Radar error compensation method and system based on genetic algorithm optimization BP neural network |
| CN113484862B (en) * | 2021-08-04 | 2023-10-17 | 电子科技大学 | Self-adaptive high-resolution wide-amplitude SAR clear reconstruction imaging method |
| CN115856888B (en) * | 2022-12-07 | 2024-04-19 | 北京理工大学 | Radiation source positioning method based on back projection |
| CN117289272A (en) * | 2023-09-15 | 2023-12-26 | 西安电子科技大学 | Regularized synthetic aperture radar imaging method based on non-convex sparse optimization |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7301495B2 (en) * | 2006-01-11 | 2007-11-27 | Raytheon Company | Interrupt SAR implementation for range migration (RMA) processing |
| CN102426353A (en) * | 2011-08-23 | 2012-04-25 | 北京航空航天大学 | Method for offline acquisition of airborne InSAR (Interferometric Synthetic Aperture Radar) motion error by utilizing high-precision POS (Position and Orientation System) |
| CN104316923A (en) * | 2014-10-14 | 2015-01-28 | 南京航空航天大学 | Self-focusing method aiming at synthetic aperture radar (Back Projection) imaging |
| CN104730520A (en) * | 2015-03-27 | 2015-06-24 | 电子科技大学 | Circumference SAR back projection self-focusing method based on subaperture synthesis |
| CN104833973A (en) * | 2015-05-08 | 2015-08-12 | 电子科技大学 | Linear array SAR backward projection self-focusing imaging method based on positive semi-definite programming |
| CN105929399A (en) * | 2016-04-25 | 2016-09-07 | 电子科技大学 | Interference SAR data imaging and elevation estimation method |
-
2017
- 2017-03-22 CN CN201710173161.4A patent/CN107015225B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7301495B2 (en) * | 2006-01-11 | 2007-11-27 | Raytheon Company | Interrupt SAR implementation for range migration (RMA) processing |
| CN102426353A (en) * | 2011-08-23 | 2012-04-25 | 北京航空航天大学 | Method for offline acquisition of airborne InSAR (Interferometric Synthetic Aperture Radar) motion error by utilizing high-precision POS (Position and Orientation System) |
| CN104316923A (en) * | 2014-10-14 | 2015-01-28 | 南京航空航天大学 | Self-focusing method aiming at synthetic aperture radar (Back Projection) imaging |
| CN104730520A (en) * | 2015-03-27 | 2015-06-24 | 电子科技大学 | Circumference SAR back projection self-focusing method based on subaperture synthesis |
| CN104833973A (en) * | 2015-05-08 | 2015-08-12 | 电子科技大学 | Linear array SAR backward projection self-focusing imaging method based on positive semi-definite programming |
| CN105929399A (en) * | 2016-04-25 | 2016-09-07 | 电子科技大学 | Interference SAR data imaging and elevation estimation method |
Non-Patent Citations (3)
| Title |
|---|
| An Autofocus Method for Backprojection Imagery in Synthetic Aperture Radar;Joshua N. Ash;《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》;20120101;第104-108页 * |
| 一种基于混合模型的合成孔径雷达自聚焦算法;杨洋，王岩飞;《中国科学院大学学报》;20160930;第656-663页 * |
| 基于图像强度最优的SAR高精度运动补偿方法;胡克彬 等;《雷达学报》;20150228;第60-69页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107015225A (en) | 2017-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107015225B (en) | A kind of SAR platform elemental height error estimation based on self-focusing | |
| CN107037429B (en) | Linear array SAR three-dimensional imaging method based on threshold gradient tracking algorithm | |
| CN104833973B (en) | Linear array SAR backward projection self-focusing imaging method based on positive semi-definite programming | |
| CN104730520B (en) | Circumference SAR back projection self-focusing method based on subaperture synthesis | |
| Buckley et al. | Terrestrial laser scanning in geology: data acquisition, processing and accuracy considerations | |
| CN103713288B (en) | Sparse Bayesian reconstruct linear array SAR formation method is minimized based on iteration | |
| Capraro et al. | Implementing digital terrain data in knowledge-aided space-time adaptive processing | |
| CN107238824B (en) | Satellite-borne SAR image geometric accurate correction method based on priori dem data | |
| CN107748362A (en) | A kind of quick autohemagglutination focusing imaging methods of linear array SAR based on maximum sharpness | |
| CN103698763A (en) | Hard threshold OMP (orthogonal matching pursuit)-based linear array SAR (synthetic aperture radar) sparse imaging method | |
| CN105388476B (en) | A kind of chromatography SAR imaging methods based on joint sparse model | |
| CN103913741A (en) | Synthetic aperture radar efficient autofocus BP method | |
| JP2017096918A (en) | Method for generating image of area of interest using radar system | |
| Denbina et al. | Forest height estimation using multibaseline PolInSAR and sparse lidar data fusion | |
| CN105699969A (en) | A maximum posterior estimated angle super-resolution imaging method based on generalized Gaussian constraints | |
| CN106918810B (en) | A kind of microwave relevance imaging method when the amplitude phase error there are array element | |
| CN114003981A (en) | Electromagnetic spectrum visual analysis method based on space-time integrated digital earth | |
| CN106054190B (en) | Bistatic Forward-looking SAR frequency domain imaging method based on frequency spectrum optimization modeling | |
| WO2013154645A2 (en) | Systems and methods for image sharpening | |
| CN106908790B (en) | A kind of optimal estimating method of SAR radar speed | |
| Kröhnert et al. | Watching grass grow-a pilot study on the suitability of photogrammetric techniques for quantifying change in aboveground biomass in grassland experiments | |
| CN106772368B (en) | The super-resolution three-D imaging method of more random frequency radar arrays | |
| CN103728617B (en) | Double-base synthetic aperture radar time domain fast imaging method | |
| Grossmann et al. | Digital twinning in the ocean-chanllenges in multimodal sensing and multiscale fusion based on faithful visual models | |
| CN108427111A (en) | A kind of radar range finding method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for 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 |
Granted publication date: 20190719 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |