CN106646344B - A kind of Wave arrival direction estimating method using relatively prime battle array - Google Patents

A kind of Wave arrival direction estimating method using relatively prime battle array Download PDF

Info

Publication number
CN106646344B
CN106646344B CN201611163847.7A CN201611163847A CN106646344B CN 106646344 B CN106646344 B CN 106646344B CN 201611163847 A CN201611163847 A CN 201611163847A CN 106646344 B CN106646344 B CN 106646344B
Authority
CN
China
Prior art keywords
array
vector
battle array
signal
relatively prime
Prior art date
Application number
CN201611163847.7A
Other languages
Chinese (zh)
Other versions
CN106646344A (en
Inventor
王海燕
贾天
贾天一
申晓红
花飞
荆海霞
刘兴晨
马志东
姚海洋
Original Assignee
西北工业大学
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 西北工业大学 filed Critical 西北工业大学
Priority to CN201611163847.7A priority Critical patent/CN106646344B/en
Publication of CN106646344A publication Critical patent/CN106646344A/en
Application granted granted Critical
Publication of CN106646344B publication Critical patent/CN106646344B/en

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

Abstract

The present invention provides a kind of Wave arrival direction estimating methods using relatively prime battle array, it is related to array signal processing field, based on the sparse reconstruct thought in compressed sensing, Wave arrival direction estimating method based on OMP algorithm is constructed for relatively prime battle array, overcome the problem of smooth MUSIC algorithm in airspace causes Virtual array to reduce, so that detection target numbers are more than MN, due to using this novel nested battle array of relatively prime battle array, so that Mutual coupling performance significantly improves the destination number of estimation, there is estimation performance well in the estimation of array multiple target, when equally using relatively prime battle array, so that the target numbers of estimation are more than NM target, under same array element quantity, further improve the estimation target numbers of relatively prime battle array, what can be simple and efficient is reconstructed sparse signal, finally the direction of arrival that provides rapidly and efficiently is estimated For meter as a result, in high s/n ratio, the estimated accuracy of the method for the present invention is better than the MUSIC algorithm of space smoothing.

Description

A kind of Wave arrival direction estimating method using relatively prime battle array

Technical field

The present invention relates to array signal processing field, especially a kind of Wave arrival direction estimating method.

Background technique

Mutual coupling has and is widely applied very much in radar, sonar and wireless communication, however traditional direction of arrival Estimation method can only solve the case where number of targets is less than array number, therefore how to detect more targets with a small amount of array element is one A good problem to study.In practice, general common N member concentrating rate, can only at most estimate the incoming wave of N-1 target Direction.In recent years, propose a kind of linear array of new geometric structure --- relatively prime battle array so that estimation target numbers far more than Array element number N, in fact, the target numbers for having the relatively prime battle array of N+M array element that can estimate can achieve O (MN), i.e., and MN The target numbers of same order size, since the ingenious distribution of relatively prime battle array element position can be formed after mathematical operation is handled The bigger virtual array in aperture.Specifically, the relatively prime battle array of N+2M-1 physics array element mentioned above can fictionalize aperture and be The concentrating rate of 2MN+1.

For relatively prime battle array, it has been proposed that a kind of smooth MUSIC algorithm in airspace based on relatively prime battle array, for using N+2M-1 The relatively prime battle array of a physics array element, MN target is at most estimated that using this method, although the target that this algorithm estimates Number has been much larger than the concentrating rate of equal number physics array element, but due to having used airspace smoothing method, leads Virtual array is caused to be reduced to MN+1, detection performance reduces.Recent years, by D.Donoho, E.Candes and scientist of Chinese origin T.Tao et al. proposes a kind of new acquisition of information guiding theory, i.e. compressed sensing, the theory once proposition, just information theory, The fields such as signal processing, pattern-recognition, wireless communication are paid high attention to, which thinks, for a sparse signal or It is that can be carried out after observing on a small quantity to the signal with the signal of rarefaction representation, it is then former using the reconstruct of specific restructing algorithm Beginning signal meets y=Ax, observing matrix in formula for the finite observation y of signal xM < N, if meeting certain item Part can reconstruct original sparse signal x from finite observation y.

Summary of the invention

For overcome the deficiencies in the prior art, the present invention is based on the sparse reconstruct thoughts in compressed sensing, for relatively prime battle array The Wave arrival direction estimating method based on OMP algorithm is constructed, overcoming the smooth MUSIC algorithm in airspace leads to asking for Virtual array reduction Topic so that detection target numbers be more than MN, and emulate proof this method have good estimated accuracy in high s/n ratio.

The technical solution adopted by the present invention to solve the technical problems specifically comprises the following steps:

Step 1: relatively prime battle array structure design

Relatively prime battle array, which is respectively that Md is nested with the concentrating rate of Nd by two spacing, to be formed, and wherein M and N is two matter each other Several constants, d indicate the half-wavelength of reception signal, and a concentrating rate has N number of array element, and array element spacing is Md, another Concentrating rate has 2M array element, and array element spacing is Nd, and first array element of two concentrating rates is overlapped, and shares N+2M-1 Array element, since M and N is prime number, so this linear array is referred to as relatively prime battle array, relatively prime battle array is a kind of heterogeneous line of special construction Array;

Step 2: estimating that relatively prime battle array receives the covariance matrix of signal

Relatively prime battle array receives the covariance matrix R of signalxxEstimated by formula (1):

In formula (1),Indicate that relatively prime battle array received signal vector, T indicate array number of snapshots;

Step 3: constructing the received vector y of virtual concentrating rate

Construct the element b that size is the i-th row jth column in (N+2M-1) × (N+2M-1) identity matrix B, Bij=i-j, mark Matrix B and covariance matrix RxxIt is straightened, obtains by column respectivelyMark vectorThe position that element value is-MN to MN is successively found according to sequence from small to large from mark vector b It sets, and is sequentially recorded lower 2MN+1 location information, then successively come out the element extraction of corresponding position from vector z, structure Produce the received vector of virtual concentrating rate

Step 4: angular regions subdivision grid to be detected, constructs array manifold matrix A

By angular regions discretization subdivision grid to be detected, grid vector θ=[θ is formed12,…,θD]T, wherein θkTable Showing discrete grid angle, k=1 ..., D, D indicates lattice number, and the lattice number D of subdivision is made to be greater than signal number, by Formula (2) and formula (3) construct array manifold matrix A:

A(θ12,…,θD)=[a (θ1),a(θ2),…a(θk),...,a(θD)] (2)

In formula (2), a (θk) be corresponding array array manifold vector, k=1,2 ..., D, in formula (3), j indicates imaginary number Unit, d indicate to receive the half-wavelength of signal, the wavelength of λ expression reception signal, θkIndicate discrete grid angle;

Step 5: estimating sparse signal using OMP algorithm

Array manifold matrix A (θ12,…,θD) and the received vector y of virtual concentrating rate meet following equation:

P is corresponding to subdivision grid θ=[θ in formula (4)12,…,θD]TSparse vector,Indicate noise vector, it is dilute Sparse vector estimated value can be obtained using the OMP algorithm solution in compressed sensing by dredging vector p

Step 6: obtaining signal direction of arrival by sparse vector

If sparse vector estimated valueIn i-th it is non-zero, then it represents that corresponding θiDirection has signal, otherwise indicates do not have Signal.

The present invention uses concentrating rate compared to traditional Mutual coupling, this novel due to using relatively prime battle array Nested battle array has so that Mutual coupling performance significantly improves the destination number of estimation in the estimation of array multiple target Estimation performance well;When equally using relatively prime battle array, this method is for the MUSIC algorithm of space smoothing, due to making With the OMP algorithm of compressed sensing, so that the target numbers of estimation are more than NM target, under same array element quantity, further Improve the estimation target numbers of relatively prime battle array;Due to having used OMP algorithm, what can be simple and efficient carries out weight to sparse signal Structure, finally rapidly and efficiently provide Mutual coupling result;In high s/n ratio, the estimated accuracy of the method for the present invention is better than sky Between smooth MUSIC algorithm, in general, performance of the invention is not less than the MUSIC algorithm of space smoothing.

Detailed description of the invention

Fig. 1 is the method flow diagram that the present invention carries out Mutual coupling.

Fig. 2 is the geometry of the relatively prime battle array of the present invention.

Fig. 3 is the Mutual coupling result of the present invention with space smoothing MUSIC algorithm.

Fig. 4 is Mutual coupling result of the present invention to 16 incoming wave signals.

Fig. 5 is the estimated accuracy comparing result of the present invention with the single goal of space smoothing MUSIC algorithm.

The smooth MUSIC algorithm of MUSIC representation space in Fig. 5, OMP indicate that the wave proposed by the present invention based on OMP algorithm reaches Direction determining method.

Specific embodiment

Present invention will be further explained below with reference to the attached drawings and examples.

Step 1: relatively prime battle array structure design

The constant M and N for choosing two prime numbers each other indicate the half-wavelength for receiving signal with constant d, and relatively prime battle array is by between two Away from respectively Md composition nested with the concentrating rate of Nd, one of even linear array has N number of array element, and array element spacing is Md, separately An outer linear array has 2M array element, and array element spacing is Nd, and two linear arrays, first array element is overlapped, a total of N+2M-1 battle array Member, since M and N is prime number, so this linear array is referred to as relatively prime battle array, the structure of relatively prime battle array is as shown in Figure 1, relatively prime battle array is one The unequally spaced linear array of kind special construction, therefore can further be derived on the model of linear array;

The linear array comprising N number of array element, element position l are considered firsti, it is assumed that K target bearing isAnd And K echo signal is set as narrow band signal, then array received signal indicates are as follows:

1≤t≤T, s in formulakIt (t) is k-th of incoming wave signal, n (t) is independent identically distributed white noise, It is array manifold vector, and

I-th of element represent k-th of signal phase change brought by the time delay of i-th of array element.

Step 2: estimating that relatively prime battle array receives the covariance matrix of signal

Relatively prime battle array receives the covariance matrix R of signalxxEstimated by formula (1):

In formula,Indicate that relatively prime battle array received signal vector, T indicate array number of snapshots;

If signal sk(t) obeying variance isIndependent Gaussian distribution, consider that each array element receives the second-order statistics of data Amount solves the covariance matrix R of array received signal x (t)xx

σ in formula2It is noise power, by matrix RxxIt is straightened by column, according to the operation of matrix Padé approximants (also referred to as direct product) Property obtains vector z by formula (7):

Wherein, Indicate the variance of k-th of transmitting signal,Expression formula is as follows:

Represent in addition to i-th of position be all as 1 remaining element 0 column vector, letter as is regarded vector z according to (8) formula The signal that source vector q is received after matrix Φ observation, matrixIt is middle that there are an array manifold matrixes, and And this matrix is the array manifold matrix of a linear array with more Virtual arrays.

Step 3: constructing the received vector y of virtual concentrating rate

Construct the element b that size is the i-th row jth column in (N+2M-1) × (N+2M-1) identity matrix B, Bij=i-j, mark Matrix B and covariance matrix RxxIt is straightened, obtains by column respectivelyMark vector Sequence from mark vector b from small to large successively finds the position that element value is-MN to MN, is sequentially recorded lower total 2MN+1 A location information, it is assumed that b=[- 3, -1,1,5,0] successively finds the position that element value is -1 to 1 to big sequence from b with small It sets, the corresponding position information recorded is exactly [2,5,3], then successively the element extraction of corresponding position come out from vector z, Construct the received vector of virtual concentrating rate

Step 4: angular regions subdivision grid to be detected, constructs array manifold matrix A

By angular regions discretization subdivision grid to be detected, grid vector θ=[θ is formed12,…,θD]T, wherein θk(k =1 ..., D) indicate discrete grid angle, D indicates lattice number, and the lattice number D of subdivision is made to be greater than signal number, Taking D is 10 times or more of signal number, constructs array manifold matrix A by following formula:

A(θ12,…,θD)=[a (θ1),a(θ2),…,a(θD)] (2)

In formula (2), a (θk) be corresponding array array manifold vector, k=1,2 ..., D, in formula (3), j indicates imaginary number Unit, d indicate to receive the half-wavelength of signal, the wavelength of λ expression reception signal, θkIndicate discrete grid angle;

Step 5: estimating sparse signal using OMP algorithm

Array manifold matrix A (θ12,…,θD) and the received vector y of virtual concentrating rate meet following equation:

P is corresponding to subdivision grid θ=[θ in formula (4)12,…,θD]TSparse vector,Indicate noise vector, it is dilute Sparse vector estimated value can be obtained using the OMP algorithm solution in compressed sensing by dredging vector p

Step 6: obtaining signal direction of arrival by sparse vector

If sparse vector estimated valueIn i-th it is non-zero, then it represents that θiDirection has signal, otherwise indicates no signal.

In the present invention, more specifically assume that this linear array is relatively prime battle array as shown in Figure 1, be respectively by two spacing Md composition nested with the concentrating rate of Nd, a shared N+2M-1 physics array element, one of even linear array have N number of array element, Another linear array has 2M array element, and first array element of two linear arrays is overlapped.Assuming that the covariance matrix R of signalxxIt is known that square Battle array size is (N+2M-1) × (N+2M-1), according to formula (4), by the property of direct product it is recognised that matrix Φ shares (N+2M- 1)2A row vector, it has been demonstrated that, if N, M are prime numbers, the subset of these row vectors corresponds to bigger equal in an aperture The array manifold matrix A of even linear array, element position is from-MNd to MNd, and array element spacing is d, at this moment virtual concentrating rate Aperture can achieve (2MN+1) d, below by matrixConstruct the array manifold matrix of concentrating rate

Corresponding row vector is selected from Φ, and these row vectors are ranked up to obtain A, so that array manifold matrix A In (n, k) a element beK=1,2 ..., K, n=-MN ..., 0 ..., MN ultimately form size For the matrix A of (2NM+1) × K, A is just the array manifold matrix of the concentrating rate of (2MN+1) d for an aperture at this time, For the vector z of formula (8) left end, according to the corresponding item of method choice that vector y is constructed described in step 3 and sorting obtain to Y is measured, y is the received vector corresponding to this virtual concentrating rate, then A, y meet

Vector in formulaIt is vector 1nRemove the vector that corresponding entry sorts, and NM+1 are 1, remaining is 0.

Mutual coupling is carried out to the sparse reconstruct thought of compressed sensing below to be illustrated:

First to angular regions discretization to be detected: θ=[θ12,…,θD]T, and D > > K, the i.e. grid number of subdivision Mesh is greater than signal number, constructs array manifold matrix A (θ12,…,θD):

A(θ12,…,θD)=[a (θ1),a(θ2),…,a(θD)] (2)

In formula It is the array manifold vector for the concentrating rate that an array number is 2NM+1.Therefore formula (4) can be rewritten as

Wherein, y is the received vector for the virtual concentrating rate being calculated according to the reception signal x (t) of relatively prime battle array, such as The arrival bearing of k-th of signal of fruit is θi, then i-th of position of vector pAssuming that the lucky position in the position of all targets In on the mesh point of subdivision, then being only non-zero on K arrival bearing, at this point, p is K sparse vector.For formula (9), If given y and A, solves equation by the sparse restructing algorithm in compressed sensing.

According to compressive sensing theory, need to acquire the most sparse solution for meeting formula (10)Therefore sparse reconstruction indicates Are as follows:

In formula | | | |0The number of vector nonzero term is indicated, in order to indicate to facilitate A=A (θ12,…,θD), it is assumed that noise Variances sigma2Be it is unknown, in order to solve formula (10), the present invention is solved using OMP algorithm, and the iterative step of OMP algorithm is as follows:

Input: array manifold matrix A, the received vector y of virtual concentrating rate, echo signal quantity K;

Output: the K sparse bayesian learning of pError vector E;

Initialization: surplus E0=y, reconstruction signal p0=0, indexed setThe number of iterations n=0;

Step 1: calculating surplus and construction array manifold matrix A (θ12,…,θD) each column between inner product Gn= ATEn-1, En-1It is the surplus of nth iteration, when n=1, En-1=E0

Step 2: finding out GnThe corresponding position of the element of middle maximum absolute value, i.e.,

Step 3: updating indexed set Γnn-1∪ { k } and atom set

Step 4: acquiring approximate solution using least square method

Step 5: updating surplus

Step 6: judging whether iteration meets stop condition, i.e., whether the number of iterations n is greater than K, when the number of iterations is greater than K then Meet stop condition, if meeting stop condition,E=En, outputOtherwise n ← n+1 is enabled, step is turned Rapid 1, whereinIt isI-th.

Method of the invention is further described below: implementing under the premise of the technical scheme of the present invention, provides Detailed embodiment and specific operating process.

Consider the relatively prime battle array comprising 10 physics array element, i.e., N=5, M=3 is taken to the relatively prime battle array in Fig. 2, specifically, the One layer of element position is in [0,3,6,9,12] d, and for second layer element position in [0,5,10,15,20,25] d, taking d is half-wave long value, Several narrow band signals are chosen, arrival bearing is evenly distributed in -60 degree to 60 degree of section, and frequency is taken as f=1000Hz, adopts Sample frequency fs=8192Hz, number of snapshots 500, signal-to-noise ratio 0dB.

The narrow band signal that 15 different directions are had chosen in Fig. 3 emulates the detection performance of two kinds of algorithms.In Fig. 4 left figure be by The result obtained according to space smoothing matrix MUSIC algorithm.Right figure illustrates the Mutual coupling based on OMP algorithm in Fig. 3.It is first It first spends to 60 degree with 0.5 degree from -60 for step-length grid division, i.e. then D=240 arrives (3) according to formula (1), calculate structural formula (4) equation equation finally executes OMP algorithm, obtains Mutual coupling result.Dotted line indicates original signal direction in Fig. 4, Solid line indicates estimated result.As can be seen that two kinds of algorithms can obtain good estimation effect, however space smoothing MUSIC is calculated Method can only at most estimate the arrival bearing of MN=15 signal.For 10 yuan of relatively prime battle arrays mentioned above, when there is 16 incoming waves When signal, space smoothing MUSIC algorithm will go wrong, and MUSIC algorithm can only at most estimate NM=15 signal, when 16 When a signal, method is without solution, if still solving according to the input of 15 signals, estimated result differs greatly with true value, at this time Mutual coupling based on OMP algorithm will show advantage.

When Fig. 4 gives 16 incoming wave signals, the result of Mutual coupling is carried out using OMP algorithm, it can be seen that OMP The Mutual coupling result of algorithm can provide accurate estimated result, and this result is that space smoothing MUSIC algorithm cannot Reach.

In the analogous diagram that Fig. 5 is provided, when the present invention considers single target estimation, the estimated accuracy of two kinds of algorithms is with letter It makes an uproar the situation of change of ratio.Different from the emulation of front, specifically in the range of -90 degree are to 90 degree, a signal side is randomly generated To the SNR ranges that horizontal axis indicates are using 2dB as step-length, from -20dB to 10dB;The longitudinal axis indicate error using estimation angle with The absolute value of real angle difference.Each point executes 5000 Monte Carlo Experiments, remaining condition is identical as front emulation.From figure Result in 5 can be seen that the Mutual coupling for single goal, in low signal-to-noise ratio, the estimation performance base of two kinds of algorithms This is the same, but in high s/n ratio, the estimated accuracy of OMP algorithm is better than the MUSIC algorithm of space smoothing.

Claims (1)

1. a kind of Wave arrival direction estimating method using relatively prime battle array, it is characterised in that include the following steps:
Step 1: relatively prime battle array structure design
Relatively prime battle array, which is respectively that Md is nested with the concentrating rate of Nd by two spacing, to be formed, and wherein M and N is two prime numbers each other Constant, d indicate the half-wavelength of reception signal, and a concentrating rate has N number of array element, and array element spacing is Md, another is uniformly Linear array has 2M array element, and array element spacing is Nd, and first array element of two concentrating rates is overlapped, and shares N+2M-1 battle array Member, since M and N is prime number, so this linear array is referred to as relatively prime battle array, relatively prime battle array is a kind of non-homogeneous alignment of special construction Battle array;
Step 2: estimating that relatively prime battle array receives the covariance matrix of signal
Relatively prime battle array receives the covariance matrix R of signalxxEstimated by formula (1):
In formula (1),Indicate that relatively prime battle array received signal vector, T indicate array number of snapshots;
Step 3: constructing the received vector y of virtual concentrating rate
Construct the element b that size is the i-th row jth column in (N+2M-1) × (N+2M-1) identity matrix B, Bij=i-j, identity matrix B and covariance matrix RxxIt is straightened, obtains by column respectivelyMark vector Element value is successively found according to sequence from small to large from mark vector b and is the position of-MN to MN, and is sequentially recorded down Then 2MN+1 location information successively comes out the element extraction of corresponding position from vector z, construct virtual uniform alignment The received vector of battle array
Step 4: angular regions subdivision grid to be detected, constructs array manifold matrix A
By angular regions discretization subdivision grid to be detected, grid vector θ=[θ is formed12,…,θD]T, wherein θkIndicate from Scattered grid angle, k=1 ..., D, D indicate lattice number, and the lattice number D of subdivision is made to be greater than signal number, by formula (2) and formula (3) constructs array manifold matrix A:
A(θ12,…,θD)=[a (θ1),a(θ2),…a(θk),...,a(θD)] (2)
In formula (2), a (θk) be corresponding array array manifold vector, k=1,2 ..., D, in formula (3), j indicates imaginary unit, d Indicate the half-wavelength of reception signal, λ indicates to receive the wavelength of signal, θkIndicate discrete grid angle;
Step 5: estimating sparse signal using OMP algorithm
Array manifold matrix A (θ12,…,θD) and the received vector y of virtual concentrating rate meet following equation:
P is corresponding to grid vector θ=[θ in formula (4)12,…,θD]TSparse vector,Indicate noise vector, it is sparse to Amount p can obtain sparse vector estimated value using the OMP algorithm solution in compressed sensing
Step 6: obtaining signal direction of arrival by sparse vector
If sparse vector estimated valueIn i-th it is non-zero, then it represents that corresponding θiDirection has signal, otherwise indicates not believe Number.
CN201611163847.7A 2016-12-16 2016-12-16 A kind of Wave arrival direction estimating method using relatively prime battle array CN106646344B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611163847.7A CN106646344B (en) 2016-12-16 2016-12-16 A kind of Wave arrival direction estimating method using relatively prime battle array

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611163847.7A CN106646344B (en) 2016-12-16 2016-12-16 A kind of Wave arrival direction estimating method using relatively prime battle array

Publications (2)

Publication Number Publication Date
CN106646344A CN106646344A (en) 2017-05-10
CN106646344B true CN106646344B (en) 2019-02-01

Family

ID=58822798

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611163847.7A CN106646344B (en) 2016-12-16 2016-12-16 A kind of Wave arrival direction estimating method using relatively prime battle array

Country Status (1)

Country Link
CN (1) CN106646344B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107315161B (en) * 2017-06-07 2019-06-21 西安电子科技大学 Compressed sensing based non-circular signal wave reaches bearing estimate method
CN107576931B (en) * 2017-07-18 2020-08-11 电子科技大学 Covariance low-dimensional iteration sparse reconstruction-based correlation/coherent signal direction-of-arrival estimation method
CN107329110B (en) * 2017-08-24 2019-08-30 浙江大学 Wave arrival direction estimating method based on thinned array Direct interpolation
CN108594164B (en) * 2017-11-30 2020-09-15 山东农业大学 Plane array DOA estimation method and equipment
CN108120967B (en) * 2017-11-30 2020-01-10 山东农业大学 Plane array DOA estimation method and equipment
CN109491009B (en) * 2018-11-06 2020-06-16 哈尔滨工程大学 Optical fiber combined array and grating lobe suppression method based on optical fiber combined array
CN110161453A (en) * 2019-05-23 2019-08-23 成都理工大学 Relatively prime array wide and narrow strip target DOA estimation algorithm based on multi input residual error network

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4965732A (en) * 1985-11-06 1990-10-23 The Board Of Trustees Of The Leland Stanford Junior University Methods and arrangements for signal reception and parameter estimation
CN103941220A (en) * 2014-04-25 2014-07-23 电子科技大学 Out-of-grid target direction-of-arrival estimation method based on sparse reconstruction
CN104749552A (en) * 2015-03-21 2015-07-01 西安电子科技大学 Estimation method of co-prime array DOA (Direction Of Arrival) angle based on sparse reconstruction
CN105388462A (en) * 2015-11-20 2016-03-09 成都理工大学 Virtual wave beam forming method based on coprime arrays
CN105824002A (en) * 2016-04-15 2016-08-03 西安电子科技大学 Direction-of-arrival estimation method based on nested subarray array

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4965732A (en) * 1985-11-06 1990-10-23 The Board Of Trustees Of The Leland Stanford Junior University Methods and arrangements for signal reception and parameter estimation
CN103941220A (en) * 2014-04-25 2014-07-23 电子科技大学 Out-of-grid target direction-of-arrival estimation method based on sparse reconstruction
CN104749552A (en) * 2015-03-21 2015-07-01 西安电子科技大学 Estimation method of co-prime array DOA (Direction Of Arrival) angle based on sparse reconstruction
CN105388462A (en) * 2015-11-20 2016-03-09 成都理工大学 Virtual wave beam forming method based on coprime arrays
CN105824002A (en) * 2016-04-15 2016-08-03 西安电子科技大学 Direction-of-arrival estimation method based on nested subarray array

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Direction of Arrival Estimation With Co-prime Arrays Via Compressed Sensing Methods;Tianyi Jia等;《 OCEANS 2016 - Shanghai》;20160609;1-5

Also Published As

Publication number Publication date
CN106646344A (en) 2017-05-10

Similar Documents

Publication Publication Date Title
Vasil’ev et al. Doubly stochastic models of images
Wen et al. Angle estimation and mutual coupling self-calibration for ULA-based bistatic MIMO radar
Tan et al. Sparse learning via iterative minimization with application to MIMO radar imaging
CN105917249B (en) Method and system for reconstructing the scene after wall
Liu et al. Sparsity-inducing direction finding for narrowband and wideband signals based on array covariance vectors
CN104730491B (en) A kind of virtual array DOA estimation method based on L-type battle array
CN103412305B (en) Scanning radar super-resolution imaging method
CN105403856B (en) Wave arrival direction estimating method based on nested type minimum redundant array
JP4990919B2 (en) Direction-of-arrival estimation method and system in scenarios with low intensity or small sample size
Ye et al. On the resiliency of MUSIC direction finding against antenna sensor coupling
CN103149561B (en) Microwave imaging method based on scenario block sparsity
CN103344940B (en) The DOA estimation method of low complex degree and system
US20120259590A1 (en) Method and apparatus for compressed sensing with joint sparsity
CN104749553B (en) Direction of arrival angle method of estimation based on rapid sparse Bayesian learning
CN103780522B (en) Non-orthogonal joint diagonalization instantaneous blind source separation method based on double iteration
CN103353596B (en) Wave beam space domain meter wave radar height measurement method based on compressed sensing
CN104237883B (en) Airborne radar space time self-adaptation processing method with sparse representation
CN105182293A (en) Method for estimating DOA and DOD of MIMO radar based on co-prime array
CN102213761B (en) Multi-target location method of bistatic common-address multi-input-multi-output radar
CN103616687B (en) The fitting of a polynomial ISAR envelope alignment method that piecewise linearity is estimated
US20110146371A1 (en) Method for calibrating a transducer array
CN103353595B (en) Meter wave radar height measurement method based on array interpolation compression perception
CN103983958A (en) MIMO radar continuous target angle estimation method based on multi-measurement-vector sparse representation
CN103245956B (en) A kind of GPS anti-multipath method based on robust ada-ptive beamformer algorithm
CN105259550B (en) MIMO radar two dimension angular method of estimation based on compressed sensing

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