CN109444897B - Multi-array track association method based on multiple features - Google Patents

Multi-array track association method based on multiple features Download PDF

Info

Publication number
CN109444897B
CN109444897B CN201811068127.1A CN201811068127A CN109444897B CN 109444897 B CN109444897 B CN 109444897B CN 201811068127 A CN201811068127 A CN 201811068127A CN 109444897 B CN109444897 B CN 109444897B
Authority
CN
China
Prior art keywords
track
association
target
information
array
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.)
Active
Application number
CN201811068127.1A
Other languages
Chinese (zh)
Other versions
CN109444897A (en
Inventor
龚轶
祝献
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
715th Research Institute of CSIC
Original Assignee
715th Research Institute of CSIC
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 715th Research Institute of CSIC filed Critical 715th Research Institute of CSIC
Priority to CN201811068127.1A priority Critical patent/CN109444897B/en
Publication of CN109444897A publication Critical patent/CN109444897A/en
Application granted granted Critical
Publication of CN109444897B publication Critical patent/CN109444897B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/66—Sonar tracking systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention provides a multi-array track association method based on multiple characteristics, which comprises the steps of firstly, carrying out track association processing on target azimuth information acquired by different matrixes after data normalization by adopting a gray association algorithm to obtain target azimuth track association degree; performing track correlation processing on target beam output energy information acquired by different matrixes after energy compensation by adopting a gray correlation algorithm to obtain target energy track correlation degree; performing line spectrum extraction on target DEMON spectrums acquired by different matrixes to obtain envelope spectrum frequency information of a target, performing difference comparison, and taking the minimum value of absolute values in each frequency difference as a comparison value of multi-matrix frequency to obtain track association degree of the DEMON spectrum information; then, the association degrees of the three are fused through a Dempster combination rule to obtain the multi-array track joint association degree; and finally, selecting the associated flight path according to the joint association degree. Compared with a pure azimuth track association method, the multi-feature track association algorithm can optimize association quality and improve correct association rate.

Description

Multi-array track association method based on multiple features
Technical Field
The invention relates to the field of multi-array target track association in a sonar system, in particular to a multi-array track association method based on multiple characteristics.
Background
The multi-matrix data association is a key technology for generating a target unified situation of a multi-matrix sonar system, and in the multi-matrix passive sonar system, the data association is mainly directed at the problem of track association among multi-target tracks acquired by each matrix sonar. The quality of the performance of the data association algorithm directly influences the generation performance of the target unified situation of the sonar system. Many expert scholars have studied data association for many years, and have developed many classical algorithms. The method can be mainly divided into a statistical correlation method, a fuzzy correlation method and the like.
The statistical association method is characterized in that the statistical distance between a plurality of arrays of flight paths is used as an association judgment criterion, and when the statistical distance between one flight path and the other flight path is smaller than a certain threshold, the two flight paths can be considered to be from the same target; the fuzzy association method is that fuzzy membership between two tracks is calculated as an association decision criterion, and when the fuzzy membership of the two tracks is larger than a certain threshold value, the two tracks can be considered to be from the same target.
The traditional methods are all used for carrying out track association on pure azimuth information, and under a complex multi-target marine environment, the multi-array track association performance is greatly influenced only by the pure azimuth information. When the passive sonar detects a target, besides the direction information of the target, the passive sonar can also obtain the following information: and forming target characteristic information such as an output energy value and a DEMON spectrum by using the beam. Therefore, combining with other characteristic information is one of effective ways for improving the multi-array track correlation performance, the method combines and applies the azimuth, DEMON spectrum and beam output energy to the multi-characteristic target track correlation method, and can effectively improve the multi-array track correlation performance.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a multi-array track association method based on multiple characteristics.
The object of the present invention is achieved by the following technical means. Firstly, performing track association processing on target azimuth information acquired by different matrixes after data normalization by adopting a gray association algorithm to obtain target azimuth track association degree; performing track correlation processing on target beam output energy information acquired by different matrixes after energy compensation by adopting a gray correlation algorithm to obtain target energy track correlation degree; performing line spectrum extraction on target DEMON spectrums acquired by different matrixes to obtain envelope spectrum frequency information of a target, performing difference comparison, and taking the minimum value of absolute values in each frequency difference as a comparison value of multi-matrix frequency to obtain track association degree of the DEMON spectrum information; then, the association degrees of the three are fused through a Dempster combination rule to obtain the multi-array track joint association degree; and finally, selecting the associated flight path according to the joint association degree.
Furthermore, the target azimuth track information is regarded as a time sequence of the target azimuth information, and after data normalization, a grey correlation algorithm is adopted for track correlation processing.
Furthermore, the target beam output energy track information is regarded as a time sequence of the energy information, and after energy compensation is carried out on energy sequences acquired by different matrixes, track association processing is carried out by adopting a gray association algorithm.
Furthermore, demodulation DEMON spectrum analysis is carried out on the received broadband signals, line spectrum extraction is carried out on the obtained target DEMON spectrum, envelope spectrum frequency information of the target is obtained, difference comparison is carried out on the frequencies obtained by multiple arrays, the minimum value of absolute values in the frequency difference values is taken as a comparison value of the multiple arrays of frequencies, and the track correlation degree of the DEMON spectrum information is obtained.
The invention has the beneficial effects that: compared with a pure azimuth track association method, the multi-feature track association algorithm can optimize association quality and improve correct association rate.
Drawings
FIG. 1: and (3) a flow block diagram of a multi-feature target track association method.
FIG. 2: and (4) a sea trial data azimuth course map.
FIG. 3: and (4) tracking a track map by the marine test data target.
Detailed Description
The invention will be described in detail below with reference to the following drawings:
the invention mainly provides a method for performing multi-array track association by fusing characteristic information such as multi-array target azimuth, beam domain energy, DEMON spectrum and the like, which comprises the following steps:
firstly, regarding target azimuth track information as a time sequence of the target azimuth information, and after data normalization, performing track association processing by adopting a gray association method;
secondly, the energy track information output by the target beam is regarded as a time sequence of the energy information, and after energy compensation is carried out on the energy sequences obtained by different matrixes, a grey correlation method can be adopted for track correlation processing;
thirdly, performing demodulation DEMON spectrum analysis on the received broadband signal, performing line spectrum extraction on the obtained target DEMON spectrum, obtaining envelope spectrum frequency information of the target, performing difference comparison on the frequencies obtained by multiple arrays, and taking the minimum value of absolute values in the frequency difference values as a comparison value of the multiple arrays of frequencies to obtain the track association degree of the DEMON spectrum information;
fourthly, the association degrees of the three are fused through a Dempster combination rule to obtain the multi-array track joint association degree;
and fifthly, selecting the associated flight path according to the joint association degree.
1. Assuming that two arrays exist, the track information of the array 1 is selected as a reference array, and the track information of the array 2 is selected as a comparison array, which are respectively marked as Ai＝{αi(k) 1,2, L, L and Bj＝{βj(k) 1,2, L, where i is 1,2, L, N obtains N targets for array 1, j is 1,2, L, M obtains M targets for array 2, and L is track time point number;
2. in order to ensure comparability between data, before grey correlation analysis is carried out on azimuth track information and energy track information of different matrixes, a data transformation technology is required to be adopted for the data, data normalization processing is carried out on the azimuth sequence information, and energy compensation processing is carried out on the beam output energy sequence information;
orientation sequence normalization:
Figure BDA0001798825580000031
② compensation processing of beam output energy sequence
1) Because the manufacturing process and the preprocessing circuit of different array elements are different, the amplitude response of the array element receiving data is different, amplitude compensation is needed, and the compensation of sensitivity and amplification coefficients among different arrays is mainly involved;
factor of amplitude compensationA step of:
Figure BDA0001798825580000032
wherein A is0Representing the circuit amplification factor of the reference matrix, B0Representing the scaled value of the sensitivity of the reference array, AcomRepresenting the circuit amplification factor of the comparison matrix, BcomThe scaled values of the sensitivity of the comparative matrices are shown.
2) Because the frequency band ranges of different array processing are different, the output energy reference of the wave beam is not consistent, the output energy under different frequency bands needs to be unified to the frequency band of 1Hz to carry out subsequent correlation processing, and here, the spectral levels of the target signal and the sonar background noise are supposed to be reduced along with the frequency by the slope of about-6 dB/Oct or-20 dB/10 Oct.
Band compensation factor:
Figure BDA0001798825580000033
wherein the content of the first and second substances,
Figure BDA0001798825580000034
to reference the highest frequency of the array processing band,
Figure BDA0001798825580000035
processing the lowest frequency of the frequency band for the reference array;
Figure BDA0001798825580000036
to compare the highest frequencies of the matrix processing bands,
Figure BDA0001798825580000037
the lowest frequencies of the fundamental frequency bands are compared.
3) Because the array elements of different matrixes are different, the space gain obtained by the conventional summation beam forming processing is different, and energy compensation is needed.
Spatial gain compensation factor:
Figure BDA0001798825580000038
wherein M is0Showing a referenceNumber of elements of matrix, McomIndicating the number of array elements of the comparison matrix,
Figure BDA0001798825580000039
which is indicative of the energy of the target signal,
Figure BDA00017988255800000310
representing the noise energy.
Then the total compensation factor of the output energy of different array beams can be considered as:
Figure BDA0001798825580000041
3. respectively carrying out grey correlation algorithm processing on the processed azimuth sequence information and the beam output energy sequence information to obtain respective grey track correlation matrix gammabAnd Γe；
Gray correlation algorithm:
the gray correlation algorithm measures how close the data columns are based on how similar or dissimilar the data column factors are. When performing data column correlation analysis, a reference number column must be determined first, and then the proximity of other number columns to the reference number column is compared to make a judgment.
Calculating a relation coefficient between the reference sequence and the comparison sequence as shown in the following formula:
Figure BDA0001798825580000042
where ρ ∈ (0, + ∞) is called a resolution coefficient, and the smaller ρ is, the larger the resolution is, and is usually 0.5. minjmink|αi(k)-βj(k) L is called the two-level minimum difference, maxj maxk|αi(k)-βj(k) L is called the two-level maximum difference, | αi(k)-βj(k) L is called the absolute difference.
From the calculation of the correlation coefficient, the correlation coefficient value of each comparison number series and the reference number series relative to each detection point number is obtained, the result is more, the information is too dispersed, the comparison is inconvenient, and the correlation coefficient value needs to be concentrated on one value, namely the grey correlation degree. The gray-scale association of a comparison array to a reference array is commonly referred to as
σij＝σ(αi,βj) (7)
The calculation method of the correlation degree usually adopts an average value method, namely, the method is
Figure BDA0001798825580000043
Thus obtaining the gray correlation degree of the multi-target track between the two matrixes, namely
Figure BDA0001798825580000044
Is a gray correlation matrix whose elements σij(i-1, 2, L N; j-1, 2, L M) is AiAnd BjThe grey correlation degree of the navigation tracks reflects the correlation degree between the navigation tracks.
4. Frequency difference value comparison is carried out on the acquired DEMON spectral line spectrum information, the minimum value of the absolute values in the frequency difference values is taken as a comparison value of two frequencies, and the comparison value is recorded as fdSetting a threshold value f of degree of associationTAnd calculating the target track association degree based on DEMON spectrum information:
Figure BDA0001798825580000051
then the incidence matrix gamma of the multi-target track based on DEMON spectral information can be obtainedd。
5. And then fusing the multi-feature association degrees based on the azimuth and the energy and the DEMON spectrum through a Dempster combination rule to obtain a joint association degree.
Dempster combination rule:
the Dempster combination rule obtains a new combined track association degree as output by integrating the track association degrees with multiple characteristics. The combination rule is called an orthogonal sum rule and is denoted by #. The method applies Dempster combination rule to the fusion of multi-feature information to obtain the multi-array target track association degree under the common action of multiple features so as to judge the association problem between the target tracks detected by different matrixes.
Is provided with
Figure BDA0001798825580000052
Two arrays of correlation degree matrixes of the characteristic 1 and the characteristic 2 respectively,
Figure BDA0001798825580000053
for two arrays of correlation matrix after two features are fused, then
Figure BDA0001798825580000054
Firstly, the gamma-ray is processed1And Γ2Performing row-based normalization to obtain
Figure BDA0001798825580000055
Then
Figure BDA0001798825580000056
The Dempster combination rule can be generalized to three and more features for fusion:
Figure BDA0001798825580000057
Figure BDA0001798825580000061
M
Figure BDA0001798825580000062
6. finally, the obtained fusion track association degree needs to be judged through a decision criterion to select and obtain the assumed track related to the same target in each array. Because a processing means of row-by-row normalization is adopted when the Dempster combination rule is used for carrying out association degree fusion, the processing is carried out according to rows when the track related decision is judged, and the specific description is as follows:
Figure BDA0001798825580000064
wherein i is 1,2, L, N; j, k is belonged to [1, M ∈]，cij＝max(Γi)，cik＝max(ΓiK ≠ j), if there is one
Figure BDA0001798825580000063
Then c isijI.e. the judgment result, i and j are respectively the ith track of the array 1 and the jth track of the array 2, epsilon1And ε2Is a preset threshold.
Analyzing sea test data:
and analyzing the fusion performance of the multi-array target track by adopting primary sea test data of a certain sea area. The receiving array is two linear arrays, the distance between 1 array element of the array is 1.2m, the processing frequency band is 375-625 Hz, the number of the array elements is 64 elements, the distance between 2 array elements of the array is 0.6m, the processing frequency band is 375-1250 Hz, the number of the array elements is 128 elements, and the distance between the two arrays is 10 m. And respectively detecting and tracking the two arrays of data.
Fig. 2 is a diagram showing an azimuth history formed by two matrixes, and fig. 3 is a diagram showing a target tracking track of the two matrixes. And marking the target track from 0 degree to 180 degrees, wherein the array 1 has No. 1-9 targets, and the array 2 has No. 1-10 targets. And (3) grouping the No. 5-7 targets of the array 1 and the No. 6-8 targets of the array 2 into an assumed track set, and performing target track association and fusion analysis.
The table 1 and the table 2 are the pure azimuth correlation and the multi-array track correlation result based on multiple features for the target track in the marine test data, respectively, and it can be seen from the comparison of the two tables that the correct correlation rate of the multi-array track correlation algorithm based on multiple features, which integrates the target azimuth information, the target azimuth beam output energy information and the DEMON spectrum frequency information, is higher than that based on the pure azimuth gray track correlation. Under a complex multi-target marine environment, a multi-feature track association algorithm based on azimuth, energy and DEMON spectrum information has better performance than pure azimuth track association.
TABLE 1 two-array target grey correlation rate of sea test data based on azimuth information
Target Array 1 Array 2 Correct correlation rate Error correlation rate
1 Track 4 Track 5 0.58 0.42
2 Track 5 Track 6 0.58 0.42
3 Track 6 Track 7 0.95 0.05
TABLE 2 sea test data two-array target association rate based on multi-feature information
Target Array 1 Array 2 Correct correlation rate Error correlation rate
1 Track 4 Track 5 0.72 0.28
2 Track 5 Track 6 0.70 0.30
3 Track 6 Track 7 0.96 0.04
It should be understood that equivalent substitutions and changes to the technical solution and the inventive concept of the present invention should be made by those skilled in the art to the protection scope of the appended claims.

Claims (4)

1. A multi-array track association method based on multiple features is characterized in that: firstly, performing track association processing on target azimuth information acquired by different matrixes after data normalization by adopting a grey association algorithm to obtain target azimuth track association degree; performing track correlation processing on target beam output energy information acquired by different matrixes after energy compensation by adopting a gray correlation algorithm to obtain target energy track correlation degree; performing line spectrum extraction on target DEMON spectrums acquired by different matrixes to obtain envelope spectrum frequency information of a target, performing difference comparison, and taking the minimum value of absolute values in each frequency difference as a comparison value of multi-matrix frequency to obtain track association degree of the DEMON spectrum information; then, the association degrees of the three are fused through a Dempster combination rule to obtain the multi-array track joint association degree; and finally, selecting the associated flight path according to the joint association degree.
2. The multi-feature based multi-array track association method according to claim 1, wherein: and regarding the target azimuth track information as a time sequence of the target azimuth information, and performing track association processing by adopting a gray association algorithm after data normalization.
3. The multi-feature based multi-array track association method according to claim 1, wherein: and the energy track information output by the target beam is regarded as a time sequence of the energy information, and after energy compensation is carried out on the energy sequences acquired by different matrixes, track association processing is carried out by adopting a gray association algorithm.
4. The multi-feature based multi-array track association method according to claim 1, wherein: the method comprises the steps of demodulating a received broadband signal, carrying out DEMON spectrum analysis, carrying out line spectrum extraction on an obtained target DEMON spectrum, obtaining envelope spectrum frequency information of a target, carrying out difference comparison on the frequencies obtained by multiple arrays, and taking the minimum value of absolute values in the frequency difference values as a comparison value of the multiple arrays of frequencies to obtain the track correlation degree of the DEMON spectrum information.
CN201811068127.1A 2018-09-13 2018-09-13 Multi-array track association method based on multiple features Active CN109444897B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811068127.1A CN109444897B (en) 2018-09-13 2018-09-13 Multi-array track association method based on multiple features

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811068127.1A CN109444897B (en) 2018-09-13 2018-09-13 Multi-array track association method based on multiple features

Publications (2)

Publication Number Publication Date
CN109444897A CN109444897A (en) 2019-03-08
CN109444897B true CN109444897B (en) 2021-04-13

Family

ID=65530556

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811068127.1A Active CN109444897B (en) 2018-09-13 2018-09-13 Multi-array track association method based on multiple features

Country Status (1)

Country Link
CN (1) CN109444897B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110057353B (en) * 2019-03-20 2023-03-14 西安电子科技大学 Method for interrupting track association based on communication signal assistance
CN110221307B (en) * 2019-05-28 2022-12-13 哈尔滨工程大学 Multi-passive sonar non-cooperative multi-target line spectrum information fusion method
CN110361006B (en) * 2019-06-28 2022-07-19 哈尔滨工程大学 Selective track state estimation fusion method for local track dimensionality division
CN110596458B (en) * 2019-07-16 2021-02-02 西北工业大学 DEMON spectrum harmonic line spectrum and fundamental frequency automatic estimation method
CN110991539B (en) * 2019-09-05 2023-10-31 北京无线电测量研究所 Space target high-frequency repetitive behavior recognition method
CN111854729B (en) * 2020-05-29 2022-03-01 西北工业大学 Track association method based on motion information and attribute information
CN112017171B (en) * 2020-08-27 2021-10-26 四川云从天府人工智能科技有限公司 Image processing index evaluation method, system, equipment and medium
CN112684454B (en) * 2020-12-04 2022-12-06 中国船舶重工集团公司第七一五研究所 Track cross target association method based on sub-frequency bands
CN112733948B (en) * 2021-01-15 2022-09-06 重庆紫光华山智安科技有限公司 Full-target associated index automatic testing method, system, medium and terminal
CN115220002B (en) * 2022-06-02 2024-05-17 深圳大学 Multi-target data association tracking method and related device for fixed single station

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101214851B (en) * 2008-01-10 2010-12-01 黄席樾 Intelligent all-weather actively safety early warning system and early warning method thereof for ship running
US20120005149A1 (en) * 2010-06-30 2012-01-05 Raytheon Company Evidential reasoning to enhance feature-aided tracking
CN101944234B (en) * 2010-07-23 2012-07-25 中国科学院研究生院 Multi-object tracking method and device driven by characteristic trace
CN103699106B (en) * 2013-12-30 2016-03-30 合肥工业大学 Based on the multiple no-manned plane cotasking planning simulation system of VR-Forces emulation platform
CN108168564A (en) * 2017-12-04 2018-06-15 上海无线电设备研究所 A kind of Data Association based on LHD grey relational grades

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
A gray fixed cluster method based on the Dempster-shafer evidential theory;Hua Er-tian; Chen Ying; Cao Wei-wei; Xiao Jun-jun;《2011 International Conference on Electronics, Communications and Control (ICECC)》;20111103;全文 *
Estimation of mass function in evidence theory for fusion of gray level based images;Moualhi Wafa; Ezzeddine Zagrouba;《ICSES 2010 International Conference on Signals and Electronic Circuits》;20101007;全文 *
Information system security risk assessment based on grey relational analysis and Dempster-Shafer theory;Wei Miao；Yanhua Liu;《2011 International Conference on Mechatronic Science, Electric Engineering and Computer (MEC)》;20110923;全文 *

Also Published As

Publication number Publication date
CN109444897A (en) 2019-03-08

Similar Documents

Publication Publication Date Title
CN109444897B (en) Multi-array track association method based on multiple features
CN109085531B (en) Near-field source arrival angle estimation method based on neural network
Yi et al. Particle filtering based track-before-detect method for passive array sonar systems
Aboutanios et al. A STAP algorithm for radar target detection in heterogeneous environments
CN109597046B (en) Metric wave radar DOA estimation method based on one-dimensional convolutional neural network
CN110764069B (en) Sparse recovery STAP color loading method based on knowledge assistance
CN111126318A (en) Parameter-adjustable double-subspace signal detection method under signal mismatch
CN107656238B (en) Novel high-frequency ground wave radar direction finding method based on double lookup table method
Tian et al. Comparison of single‐trace and multiple‐trace polarity determination for surface microseismic data using deep learning
Tang et al. A tunable detector for distributed target detection in the situation of signal mismatch
CN111090089B (en) Space-time adaptive detection method based on two types of auxiliary data
Suleiman et al. Search-free decentralized direction-of-arrival estimation using common roots for non-coherent partly calibrated arrays
CN112699992B (en) Hybrid source positioning method based on convolutional neural network
CN110988856B (en) Target detection trace agglomeration algorithm based on density clustering
CN109444899B (en) Track association method based on pure angle information
CN115356718A (en) Cross entropy fuzzy clustering-based multi-target tracking data association method
Azimi-Sadjadi et al. Detection and classification of buried dielectric anomalies using neural networks-further results
US6683820B1 (en) Method and apparatus for tracking sonar targets
KR101715903B1 (en) Method and Apparatus for robust detection of target bearings for triplet sensor arrays against to noise signals, arrangement distortions, and angle errors
Ge et al. Multi-radar hybrid detection algorithm based on information entropy
US20070244952A1 (en) Signal analysis methods
Wang et al. A novel failure detection circuit for SUMPLE using variability index
CN113189550B (en) Array radar anti-noise interference method and system based on multi-unknown signal extraction
CN112630724B (en) High-resolution target azimuth estimation method suitable for UUV platform
CN113640840B (en) Non-compressed GNSS deception jamming detection and suppression method based on antenna array

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