CN105891814A - Range-only radar networking single-target clustering positioning method - Google Patents

Range-only radar networking single-target clustering positioning method Download PDF

Info

Publication number
CN105891814A
CN105891814A CN201610031441.7A CN201610031441A CN105891814A CN 105891814 A CN105891814 A CN 105891814A CN 201610031441 A CN201610031441 A CN 201610031441A CN 105891814 A CN105891814 A CN 105891814A
Authority
CN
China
Prior art keywords
step
amp
target
radar
outlier
Prior art date
Application number
CN201610031441.7A
Other languages
Chinese (zh)
Inventor
王本才
李发均
边保平
卫国爱
张祖峰
Original Assignee
中国人民解放军空军预警学院黄陂士官学校
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 中国人民解放军空军预警学院黄陂士官学校 filed Critical 中国人民解放军空军预警学院黄陂士官学校
Priority to CN201610031441.7A priority Critical patent/CN105891814A/en
Publication of CN105891814A publication Critical patent/CN105891814A/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
    • G01S13/00Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/06Systems determining position data of a target
    • 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
    • G01S7/00Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section

Abstract

The invention discloses a range-only radar networking single-target clustering positioning method and solves the problem that a conventional outlier identification method is not suitable for dynamic situations and is poor in real-time performance in the radar networking single-target positioning process. The method comprises the following steps: 1) carrying out combination on networking radars to obtain intersection point coordinates; 2) determining intensity degree at the places where intersection points locate according to the distance sum of the intersection points; and 3) comparing the intensity degree of the intersection points with a clustering detection threshold value, and finishing outlier point identification and target positioning. Compared with a conventional method, the method is easy to achieve engineering realization and is high in real-time performance and accuracy, and can improve target positioning precision effectively.

Description

Pure distance radar network composite single goal clusters localization method

Technical field

The invention belongs to radar network composite field of locating technology, it is adaptable to land-based radar networking, fleet radar network composite etc. are to sky The occasion of middle single goal location;In the occasion that positioning accuracy request is relatively low, this method can be directly applied to obtain target location;? The occasion that positioning accuracy request is higher, can particularly relate to a kind of pure distance radar group using the result of this method as estimating initial value Net single goal cluster localization method.

Background technology

During single goal is positioned by radar network composite, due to various objective factors or the impact of subjective operation, each thunder Reach in the target measurement of acquisition and be usually present outlier.Although the negligible amounts of outlier, but seriously reduce the positioning precision of target. Traditional outlier recognition methods such as " 3 σ " criterion, " Ge Luobusi " criterion etc., mainly to measure take afterwards batch at Reason form.Specifically, after obtaining all of measurement, suppose that it obeys a certain distribution, then according to above-mentioned criterion structure Statistic of test and then identification outlier.But practical experience shows, the method is appropriate only under quiescent conditions entering a certain fixed amount Row outlier identification, the most largely effective for current intelligences such as air mobile targets;The method takes batch processing afterwards simultaneously Mode, is unfavorable for the real-time positioning of target.If location algorithm can identify outlier in real time, accurately, it is possible not only to mesh Mark quickly positions and follows the tracks of, and can be effectively improved the positioning precision of target simultaneously.

Summary of the invention

The purpose of the present invention is to propose to a kind of pure distance radar network composite single goal and cluster localization method, solve in conventional location algorithm Outlier discrimination is low in the dynamic case, the problem of poor real.

For realizing the purpose of the present invention, the pure distance radar network composite single goal cluster localization method that the present invention proposes includes following step Rapid:

Step 1: by a certain moment N (N > 3) portion radar receiver Ri=(xi,yi)T(i=1,2 ..., N) record target range letter Breath diSend into Radar Signal Processing computer;

Step 2: initialize

npFor number of intersections, initialize np=1;

nmaxFor number of intersections maximum,

χ0For cluster detection threshold, take χ0∈(0,1);

ngFor the number of intersections that dense degree is high, initialize ng=0;

Step 3: calculate intersecting point coordinate

(1) three radar R are arbitrarily selectedm=(xm,ym)T、Rn=(xn,yn)T、Rk=(xk,yk)T

(2) intersecting point coordinate about target is recordedFor:

x ^ n p = 1 2 L 1 [ ( y m - y k ) L 2 - ( y m - y n ) L 3 ] - - - ( 1 )

y ^ n p = 1 2 L 1 [ - ( x m - x k ) L 2 + ( x m - x n ) L 3 ] - - - ( 2 )

Wherein, L1=-xmyk-xnym+xnyk+ymxk+ynxm-ynxk

L 2 = d n 2 - d m 2 + x m 2 - x n 2 + y m 2 - y n 2 ;

L 3 = d k 2 - d m 2 + x m 2 - x k 2 + y m 2 - y k 2 ;

(3)np=np+1;

(4) during circulation performs this step (1)~(3), until np=nmax+ 1 terminates;

(5)np=1;

Step 4: calculate intersection point distance and

(1) intersection pointTo the distance of other all intersection points be:

D n p = Σ j = 1 C N 3 - 1 ( x ^ n p - x ^ j ) 2 + ( y ^ n p - y ^ j ) 2 , ( j ≠ n p ) - - - ( 3 )

(2)np=np+1;

(3) during circulation performs this step (1)~(2), until np=nmax+ 1 terminates;

(4)np=1;

Step 5: outlier detection based on clustering method

(1) minimum range and D are initializedmin=Dnp

(2) D is comparedminSize, and to DminAssignment:

D m i n = m i n ( D m i n , D n p ) - - - ( 4 )

(3)np=np+1;

(4) during circulation performs this step (2)~(3), until np=nmax+ 1 terminates;

(5)np=1;

(6) intersection pointThe intersection point dense degree following formula of present position calculates

γ n p = D m i n D n p - - - ( 5 )

(7) if γnp< χ0, willIt is considered as outlier, and noteOn the contrary, ng=ng+1;

(8)np=np+1;

(9) during circulation performs this step (6)~(8), until np=nmax+ 1 terminates;

Step 6: estimate target location

Target state estimator positionCalculate with following formula

x ~ e = 1 n g Σ i = 1 C N 3 x ^ i y ~ e = 1 n g Σ i = 1 C N 3 y ^ i - - - ( 6 )

Thus complete the location to target.

Comparing with background technology, beneficial effects of the present invention illustrates: the change of (1) outlier recognition methods;Traditional outlier is known Other method is directly to be identified measurement, the intersection point that first measurement conversion is target by the method that the present invention uses data to change, Outlier identification is carried out according to the dense degree between intersection point;(2) raising of outlier recognition efficiency;Traditional outlier recognition methods Taking the form of batch processing, the method that the present invention proposes can carry out outlier identification after each radar obtains the single measurement of target, Real-time is improved;(3) extension of outlier identification range;For the outlier of traditional method None-identified, the present invention is permissible How many self-adaptative adjustment according to number of sensors cluster detection threshold, and then science determines the scope of outlier;(4) target is fixed The improvement of position precision;Traditional outlier recognition methods causes the positioning precision of algorithm not owing to being not suitable for the factors such as dynamic case Ideal, and the cluster localization method that the present invention proposes can be effectively improved the positioning precision of target.

Accompanying drawing explanation

Accompanying drawing 1 is the flow chart of steps of the pure distance radar network composite single goal cluster localization method of the present invention;

Accompanying drawing 2 is that the inventive method is to the positioning result figure after target single measurement;

Accompanying drawing 3 is the inventive method and traditional positioning result comparison diagram based on 3 σ criterions after outlier identification;

Accompanying drawing 4 is the inventive method and traditional positioning precision comparison diagram based on 3 σ criterions after outlier identification.

Detailed description of the invention

Below in conjunction with the accompanying drawings the pure distance radar network composite single goal cluster localization method of the present invention is described in detail.

Implementation condition: assume that quantity N=5 of radar network, each radar site are respectively as follows: R1=(-20,30)T、R2=(0,0)T、 R3=(15,10)T、R4=(30,0)T、R5=(60,15)T(unit: km);Each radar ranging accuracy is identical, and sets its value For σd=0.1km;Target current location is set to T=(40,60)Tkm;Take cluster detection threshold χ0=0.6.

According to above-mentioned condition, the concrete steps of the present invention are as shown in Figure 1.Specific as follows:

Step 1: by a certain moment N (N > 3) portion radar receiver Ri=(xi,yi)T(i=1,2 ..., N) record target range letter Breath diSend into Radar Signal Processing computer;

Step 2: initialize

npFor number of intersections, initialize np=1;

nmaxFor number of intersections maximum,

χ0For cluster detection threshold, take χ0∈(0,1);

ngFor the number of intersections that dense degree is high, initialize ng=0;

Step 3: calculate intersecting point coordinate

(1) three radar R are arbitrarily selectedm=(xm,ym)T、Rn=(xn,yn)T、Rk=(xk,yk)T

(2) intersecting point coordinate about target is recorded

The distance measuring of above-mentioned three radars is respectively dm、dnAnd dk, available equation below group:

( x m - x ^ n p ) 2 + ( y m - y ^ n p ) 2 = d m 2 ( x n - x ^ n p ) 2 + ( y n - y ^ n p ) 2 = d n 2 ( x k - x ^ n p ) 2 + ( y k - y ^ n p ) 2 = d k 2 - - - ( 1 )

Above formula is organized into following matrix form:

X ^ n p = A - 1 · B - - - ( 2 )

Wherein A = x m - x n y m - y n x m - x k y m - y k ;

B = 1 2 · d n 2 - d m 2 + x m 2 - x n 2 + y m 2 - y n 2 d k 2 - d m 2 + x m 2 - x k 2 + y m 2 - y k 2 ;

Solve:

x ^ n p = 1 2 L 1 [ ( y m - y k ) L 2 - ( y m - y n ) L 3 ] - - - ( 3 )

y ^ n p = 1 2 L 1 [ - ( x m - x k ) L 2 + ( x m - x n ) L 3 ] - - - ( 4 )

Wherein, L1=-xmyk-xnym+xnyk+ymxk+ynxm-ynxk

L 2 = d n 2 - d m 2 + x m 2 - x n 2 + y m 2 - y n 2 ;

L 3 = d k 2 - d m 2 + x m 2 - x k 2 + y m 2 - y k 2 ;

(3)np=np+1;

(4) during circulation performs this step (1)~(3), until np=nmax+ 1 terminates;

(5)np=1;

Step 4: calculate intersection point distance and

(1) intersection pointTo the distance of other all intersection points be:

D n p = Σ j = 1 C N 3 - 1 ( x ^ n p - x ^ j ) 2 + ( y ^ n p - y ^ j ) 2 , ( j ≠ n p ) - - - ( 5 )

(2)np=np+1;

(3) during circulation performs this step (1)~(2), until np=nmax+ 1 terminates;

(4)np=1;

Step 5: outlier detection based on clustering method

(1) initialize minimum range and

(2) D is comparedminSize, and to DminAssignment:

D m i n = m i n ( D m i n , D n p ) - - - ( 6 )

(3)np=np+1;

(4) during circulation performs this step (2)~(3), until np=nmax+ 1 terminates, and the most just obtains Dmin

(5)np=1;

(6) intersection pointThe intersection point dense degree following formula of present position calculates

γ n p = D m i n D n p - - - ( 7 )

From this formula,It is the least,The biggest;The i.e. distance of intersection point and the least, the intersection point of this intersection point present position is intensive Degree is the biggest;

(7) ifWillIt is considered as outlier, and noteOn the contrary, ng=ng+1;

χ0To choose same N closely related, total principle is: N is the biggest, number of intersections n obtainedpThe biggest, therefore χ0Should The dense degree of the larger ability valid metric intersection point chosen;

(8)np=np+1;

(9) during circulation performs this step (6)~(8), until np=nmax+ 1 terminates;

Now the dense degree of all intersection points is all measured, and obtained ngThe intersection point set that individual dense degree is higher;

Step 6: estimate target location

Target state estimator positionCalculate with following formula

x ~ e = 1 n g Σ i = 1 C N 3 x ^ i y ~ e = 1 n g Σ i = 1 C N 3 y ^ i - - - ( 8 )

I.e. take ngThe higher intersecting point coordinate average of individual dense degree, as the estimation position of target, thus completes the location to target.

Accompanying drawing 2 is that each radar is to the cluster positioning result after target single measurement.5 radars to target when common property in prelocalization RawIndividual intersection point, the inventive method identifies 1 intersection point and processes as outlier.In terms of direct result, the outlier identified Point is significantly different with the trend that other intersection points are presented.

Accompanying drawing 3 is the inventive method and traditional positioning result comparison diagram based on 3 σ criterions after outlier identification.Two kinds of methods are equal Carrying out 200 Monte Carlo emulation, wherein figure (a) is the result figure of the present invention, and figure (b) is knots based on 3 σ criterions Fruit figure.As seen from the figure, the intersection point of figure (a) substantially concentrates near target location.

Accompanying drawing 4 is root-mean-square error (RMS) comparison diagram of accompanying drawing 3.Under current simulated conditions, the positioning precision of the present invention Relatively 3 σ criterions improve about 600m.Therefore, compared with traditional outlier recognition methods, the method for the present invention is more real-time, accurate, And then it is effectively improved the positioning precision of target.

Claims (1)

1. a pure distance radar network composite single goal clusters localization method, it is characterised in that comprise the following steps:
Step 1: by a certain moment N (N > 3) portion radar receiver Ri=(xi,yi)T(i=1,2 ..., N) record target range letter Breath diSend into Radar Signal Processing computer;
Step 2: initialize
npFor number of intersections, initialize np=1;
nmaxFor number of intersections maximum,
χ0For cluster detection threshold, take χ0∈(0,1);
ngFor the number of intersections that dense degree is high, initialize ng=0;
Step 3: calculate intersecting point coordinate
(1) three radar R are arbitrarily selectedm=(xm,ym)T、Rn=(xn,yn)T、Rk=(xk,yk)T
(2) intersecting point coordinate about target is recordedFor:
x ^ n p = 1 2 L 1 [ ( y m - y k ) L 2 - ( y m - y n ) L 3 ] - - - ( 1 )
y ^ n p = 1 2 L 1 [ - ( x m - x k ) L 2 + ( x m - x n ) L 3 ] - - - ( 2 )
Wherein, L1=-xmyk-xnym+xnyk+ymxk+ynxm-ynxk
L 2 = d n 2 - d m 2 + x m 2 - x n 2 + y m 2 - y n 2 ;
L 3 = d k 2 - d m 2 + x m 2 - x k 2 + y m 2 - y k 2 ;
(3)np=np+1;
(4) during circulation performs this step (1)~(3), until np=nmax+ 1 terminates;
(5)np=1;
Step 4: calculate intersection point distance and
(1) intersection pointTo the distance of other all intersection points be:
D n p = Σ j = 1 C N 3 - 1 ( x ^ n p - x ^ j ) 2 + ( y ^ n p - y ^ j ) 2 , ( j ≠ n p ) - - - ( 3 )
(2)np=np+1;
(3) during circulation performs this step (1)~(2), until np=nmax+ 1 terminates;
(4)np=1;
Step 5: outlier detection based on clustering method
(1) initialize minimum range and
(2) D is comparedminSize, and to DminAssignment:
D m i n = m i n ( D m i n , D n p ) - - - ( 4 )
(3)np=np+1;
(4) during circulation performs this step (2)~(3), until np=nmax+ 1 terminates;
(5)np=1;
(6) intersection pointThe intersection point dense degree following formula of present position calculates
γ n p = D m i n D n p - - - ( 5 )
(7) ifWillIt is considered as outlier, and noteOn the contrary, ng=ng+1;
(8)np=np+1;
(9) during circulation performs this step (6)~(8), until np=nmax+ 1 terminates;
Step 6: estimate target location
Target state estimator positionCalculate with following formula
x ~ e = 1 n g Σ i = 1 C N 3 x ^ i y ~ e = 1 n g Σ i = 1 C N 3 y ^ i - - - ( 6 )
Thus complete the location to target.
CN201610031441.7A 2016-01-18 2016-01-18 Range-only radar networking single-target clustering positioning method CN105891814A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610031441.7A CN105891814A (en) 2016-01-18 2016-01-18 Range-only radar networking single-target clustering positioning method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610031441.7A CN105891814A (en) 2016-01-18 2016-01-18 Range-only radar networking single-target clustering positioning method

Publications (1)

Publication Number Publication Date
CN105891814A true CN105891814A (en) 2016-08-24

Family

ID=57002546

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610031441.7A CN105891814A (en) 2016-01-18 2016-01-18 Range-only radar networking single-target clustering positioning method

Country Status (1)

Country Link
CN (1) CN105891814A (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103605110A (en) * 2013-12-03 2014-02-26 北京理工大学 Indoor passive target positioning method based on received signal strength
CN104898104A (en) * 2015-06-10 2015-09-09 中国西安卫星测控中心 Target combined positioning method based on Euler's distance means clustering

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103605110A (en) * 2013-12-03 2014-02-26 北京理工大学 Indoor passive target positioning method based on received signal strength
CN104898104A (en) * 2015-06-10 2015-09-09 中国西安卫星测控中心 Target combined positioning method based on Euler's distance means clustering

Similar Documents

Publication Publication Date Title
CN103402258B (en) Wi-Fi (Wireless Fidelity)-based indoor positioning system and method
CN103945533B (en) Wireless real time position localization methods based on big data
CN102821465B (en) WLAN (Wireless Local Area Network) indoor positioning method based on subregion information entropy gain
CN102123495A (en) Centroid location algorithm based on RSSI (Received Signal Strength Indication) correction for wireless sensor network
CN103139907A (en) Indoor wireless positioning method by utilizing fingerprint technique
CN104200236A (en) Quick target detection method based on DPM (deformable part model)
CN104703143B (en) A kind of indoor orientation method based on WIFI signal intensity
CN103455797B (en) Detection and tracking method of moving small target in aerial shot video
CN104635203B (en) Radio interference source direction-finding and positioning method based on particle filter algorithm
CN103220777A (en) Mobile device positioning system
CN103338516B (en) A kind of wireless sensor network two step localization method based on total least square
CN104330721B (en) IC Hardware Trojan detecting method and system
CN103760542B (en) A kind of based on multimodal variability index CFAR object detection method
CN104007419B (en) About residence time and the radar time resource combined distributing method of heavily visiting interval
CN103648108A (en) Sensor network distributed consistency object state estimation method
CN103167606B (en) Based on the WLAN indoor orientation method of rarefaction representation
CN103369466B (en) A kind of map match assists indoor orientation method
CN105004340B (en) With reference to inertial navigation technology and the position error modification method of fingerprint location technology
CN101466145B (en) Dual-base-station accurate orientation method based on neural network
CN104076355B (en) Tracking before Dim targets detection in strong clutter environment based on dynamic programming
CN103760556B (en) Based on the multi-target cognitive tracking of centralized MIMO radar
CN102298142A (en) ht-stc-tbd Multi weak target detection radar ranging fuzzy conditions
CN103969622B (en) A kind of time difference positioning method based on the receiving station that does more physical exercises
CN104766175A (en) Power system abnormal data identifying and correcting method based on time series analysis
CN104853435B (en) A kind of indoor orientation method based on probability and device

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination