CN108345864A - Random set mould assembly radar emitter signal parameter high frequency mode method for digging based on weighted cluster - Google Patents

Random set mould assembly radar emitter signal parameter high frequency mode method for digging based on weighted cluster Download PDF

Info

Publication number
CN108345864A
CN108345864A CN201810182612.5A CN201810182612A CN108345864A CN 108345864 A CN108345864 A CN 108345864A CN 201810182612 A CN201810182612 A CN 201810182612A CN 108345864 A CN108345864 A CN 108345864A
Authority
CN
China
Prior art keywords
cluster
random
mould assembly
similarity
signal parameter
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810182612.5A
Other languages
Chinese (zh)
Other versions
CN108345864B (en
Inventor
徐欣
刘伟峰
张桂林
赵真
赵真一
饶佳人
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.)
CETC 28 Research Institute
Original Assignee
CETC 28 Research Institute
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 CETC 28 Research Institute filed Critical CETC 28 Research Institute
Priority to CN201810182612.5A priority Critical patent/CN108345864B/en
Publication of CN108345864A publication Critical patent/CN108345864A/en
Application granted granted Critical
Publication of CN108345864B publication Critical patent/CN108345864B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Signal Processing (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a kind of random set mould assembly radar emitter signal parameter high frequency mode method for digging based on weighted cluster, for the signal parameter data of random set mould assembly airborne radar radiant source target, build similarity distance matrix, compare the similarity between random set mould assembly measured value two-by-two, random set mould assembly cluster is merged based on hierarchy clustering method iteratively on the basis of similarity distance matrix, until it can not find the cluster for meeting random collection distance threshold, calculate to Simultaneous Iteration formula the support and random collection value of each new cluster, it finally excavates and meets the corresponding random collection value of support thresholding cluster as high frequency mode output.Advantages of the present invention：(1) the similarity-rough set method for random set mould assembly radar emitter signal parameter is proposed；(2) it is based on similarity distance matrix, can further excavate high-frequency signal parameter mode；(3) at low cost, implementation method engineering is calculated.

Description

Random set mould assembly radar emitter signal parameter high frequency mode based on weighted cluster is dug Pick method
Technical field
The present invention relates to random set mould assembly radar emitter signal supplemental characteristic analyzing processing field more particularly to a kind of bases In the random set mould assembly radar emitter signal parameter high frequency mode method for digging of weighted cluster.
Background technology
It is well known that the signal characteristic of current Radar emitter is more and more, and other than traditional type, airborne radar radiation Source has also appeared the random set mould assembly signal parameter being made of several arbitrary measures.Most of traditional signal characteristic type is Continuous type, also there is discrete type.Nowadays, random set mould assembly signal characteristic also becomes the important kind of radar emission source data. This is because as Radar emitter complexity increases, the type and form of signal characteristic are also more and more diversified.In addition, by It is usually no longer in the progress of signal characteristic measuring technique and the influence of measuring environment (such as noise), the measured value of signal characteristic One fixed value, but a set being made of several random values.Existing radar emitter signal characteristic analysis method and Radar emitter kind identification method, can't be specifically for random set mould assembly almost both for the signal characteristic of fixed value Signal characteristic is described and analyzes.With the development of science and technology random set mould assembly Radar emitter data resource can also increase, it is anxious Need corresponding processing method.
Invention content
Goal of the invention：In view of the above problems, the present invention proposes a kind of random set mould assembly radar emission based on weighted cluster Source signal parameter high frequency mode method for digging.
Technical solution：To achieve the purpose of the present invention, the technical solution adopted in the present invention is：One kind being based on weighted cluster Random set mould assembly radar emitter signal parameter high frequency mode method for digging, for random set mould assembly airborne radar radiation source mesh Target signal parameter data build similarity distance matrix, compare the similarity between random set mould assembly measured value two-by-two, in phase Like degree distance matrix on the basis of based on hierarchy clustering method it is iterative merge random set mould assembly cluster, until can not find satisfaction with Until the cluster of machine aggregate distance thresholding, the support and random collection value of each new cluster are calculated to Simultaneous Iteration formula, is finally dug It excavates and meets the corresponding random collection value of support thresholding cluster as high frequency mode output.
Random set mould assembly radar emitter signal parameter high frequency mode method for digging based on weighted cluster, including：
(1) Radar emitter target is pressed into random collection size SORDDescending sequence；
(2) similarity distance matrix, matrix size n × n are initialized；
(3) setting current radar radiant source target indexes s；
(4) Radar emitter target to be compared is set and indexes t；
(5) judge that target index to be compared continues step if being unsatisfactory for t≤n and S (ORD [t])/S (ORD [s]) >=ε 6, continue step 8 if met；
(6) current goal index value is updated；
(7) judge that current goal indexes, if meeting s >=n, continue step 10, otherwise, return to step 4；
(8) the similarity distance dist between Radar emitter is calculated, and updates similarity matrix；
(9) Radar emitter index to be compared, return to step 5 are updated；
(10) n initial clustering is built；
(11) judge the two cluster C that similarity distance is minimum in distance matrixuAnd Cv(u<V) whether meet dist (Cu,Cv) ≤ ε continues step 12, if being unsatisfactory for skipping to step 14 if met；
(12) agglomerative clustering CuAnd CvNewly to cluster Cw, and be updated；
(13) it is based on new cluster and updates distance matrix, newly the distance definition of cluster and other clusters is to constitute the original newly clustered There are two the minimum value of cluster and other clustering distances, return to step 11；
(14) cluster that all supports meet thresholding minsup is found, it is exported and corresponds to random collection as radar parameter High frequency mode.
Further, random collection distance threshold ε is used to compare the similarity between random measurement value set, support door Limit minsup is for judging whether signal parameter pattern belongs to high frequency.
Further, airborne radar emitter Signals measured value of parameters be one be made of several arbitrary measures it is random Set.
Advantageous effect：Advantages of the present invention：(1) it proposes for the similar of random set mould assembly radar emitter signal parameter Spend comparative approach；(2) it is based on similarity distance matrix, can further excavate high-frequency signal parameter mode；(3) cost is calculated It is low, implementation method engineering.
Description of the drawings
Fig. 1 is the work flow diagram of the present invention.
Specific implementation mode
Technical scheme of the present invention is further described with reference to the accompanying drawings and examples.
The present invention is directed to the signal parameter data of random set mould assembly airborne radar radiant source target, and structure similarity is apart from square Battle array, compares the similarity between random set mould assembly measured value two-by-two, and hierarchical clustering side is based on the basis of similarity distance matrix Method merges random set mould assembly cluster iteratively, until it can not find the cluster for meeting random collection distance threshold, simultaneously repeatedly The support and random collection value for for formula calculating each new cluster, finally excavate that meet support thresholding cluster corresponding random Rendezvous value is exported as high frequency mode.
Assuming that airborne radar emitter Signals parameter data set is by several airborne radar radiant source targets { o1, o2 ..., on } It constitutes, by a random set mould assembly, (measured value is a collection being made of several random numbers to each airborne radar radiant source target Close) measurement set of signal parameter A at.Specifically, Radar emitter target oiMeasured value on signal parameter A is represented by Random collection Si={ Sip}p, wherein SipIndicate target oiA measured value on signal parameter A in several arbitrary measures. Moreover, it is assumed that arbitrary measures similarity thresholding is δ, random collection distance threshold is ε (ε values are between zero and one), support Thresholding is minsup, measured value minimal weight thresholding is minw.
As shown in Figure 1, the method for the present invention specifically includes following steps：
(1) goal ordering；
Radar emitter target is pressed into the descending sequence of random collection size, is denoted as ORD so that SORD[1]≥SORD[2] ≥…≥SORD[n], wherein SORD[1]The random collection size for indicating first Radar emitter target in ORD sequences, with such It pushes away.
(2) similarity distance matrix initializes；
Similarity Distance matrix D istArray is initialized, size is n × n, and ranks index value corresponds to ORD sequences Index value, diagonal line and lower unit value are sky, and upper triangular unit value is infinity.
(3) current goal is arranged；
Current radar radiant source target is set and indexes s=1.
(4) target setting to be compared；
Radar emitter target index t=s+1 to be compared is set.
(5) object judgement to be compared；
Judge target index value t to be compared, if being unsatisfactory for t≤n and size (ORD [t])/size (ORD [s]) >=ε, after Continuous step 6 continues step 8 if met.
(6) current goal is changed；
Update current goal index value s=s+1.
(7) current goal judges；
If current goal index value s >=n, skips to step 10, otherwise return to step 4.
(8) similarity distance calculates；
It calculates the distance of the similarity between Radar emitter ORD [s] and ORD [t] and updates similarity matrix The computational methods of DistArray, similarity distance value are：
Wherein, SORD[s]And SORD[t]The random collection size of Radar emitter ORD [s] and ORD [t] are indicated respectively, MatchSet(oORD[s],oORD[t]) indicate matching set between Radar emitter ORD [s] and ORD [t] random collection, then With set be defined as by respectively from two random collections and measured value difference be no more than measured value similarity thresholding δ measured value The set being composed of is：
MatchSet(oi,oj)={<Sip,Sjq>|Sip∈Si,Sjq∈Sj,|Sip-Sjq|≤δ} (2)
(9) target update to be compared；
Update Radar emitter index t=t+1 to be compared, return to step 5.
(10) clustering initialization；
According to n Radar emitter random collection, n initial clustering is built, the random collection difference each clustered is initial Turn to the random collection of corresponding Radar emitter, i.e. C1=S1, C2=S2..., Cn=Sn；Each cluster CkRandom collection member The weights initialisation of element is 1, is denoted as Wk1=Wk2=...=Wk|Ck|=1, wherein | Ck | indicate cluster CkRandom collection it is big It is small；The support each clustered is initialized as 1, i.e. Sup1=Sup2=...=Supn=1；The affiliated member set each clustered MemsetkIt is initialized as by the set of corresponding Radar emitter target configuration, i.e. Memsetk={ ok}。
(11) cluster judges；
Judge the two cluster C that similarity distance is minimum in Distance matrix D istArrayuAnd Cv(u<V) whether meet dist (Cu,Cv)≤ε continues step 12, if being unsatisfactory for skipping to step 14 if met.
(12) Cluster-Fusion；
Agglomerative clustering CuAnd CvNewly to cluster Cw, cluster CwRandom collection element and its weight respectively according to formula (3) and (4) it is updated：
The support newly clustered is updated to Supw=Supu+Supv；The affiliated member set Memset newly clusteredkIt is updated to Memsetw=Memsetu+Memsetv。
(13) distance matrix updates；
Based on new cluster update Distance matrix D istArray, newly the distance definition of cluster and other clusters is that composition is new poly- There are two the minimum value of cluster and other clustering distances, return to step 11 for the original of class.
(14) high frequency mode exports；
The cluster that all supports meet thresholding minsup is found, it is exported and corresponds to height of the random collection as radar parameter Frequency pattern.
The method of the present invention can enhance the analysis ability of random set mould assembly signal characteristic, can preferably complete radar emission Identifing source task.Illustrate the random set mould assembly airborne radar radiation based on weighted cluster of the present invention below by an example Source signal parameter high frequency mode method for digging.
Assuming that detecting 3 samples of a kind of Radar emitter, the PRI signals of each sample are random set mould assembly signal ginsengs Number, value is as shown in table 1, unit MHz.
Table 1
Sample PRI signal parameters (MHz)
o1 {60,69,85,100}
o2 {61,70,84,99}
o3 {70,86,101}
Arbitrary measures similarity thresholding δ values are 0.1, and random collection distance threshold ε values are 0.6, and support meets Thresholding minsup values are 2, and measured value minimal weight thresholding minw is 0.3, and its step are as follows：
Step 1, Radar emitter target presses the descending sequence of random collection size, ORD=o1＜ o2＜ o3；
Step 2, similarity Distance matrix D istArray is initialized, size is n × n, as shown in table 2：
Table 2
Sample o1 o2 o3
o1 - ∞ ∞
o2 - - ∞
o3 - - -
Step 3, current goal is arranged, and setting current radar radiant source target indexes s=1；
Step 4, Radar emitter target index t=s+1=2 to be compared is arranged in target setting to be compared；
Step 5, judge that target index value t to be compared meets t≤n and size (ORD [t])/size (ORD [s]) >=ε, after Continuous step 8；
Step 8, similarity distance calculates, and calculates Radar emitter o1With o2Between similarity distance and update similarity Matrix D istArray, as shown in table 3：
Table 3
Sample o1 o2 o3
o1 - 0 ∞
o2 - - ∞
o3 - - -
Step 9, target update to be compared updates Radar emitter index t=t+1=3 to be compared, return to step 5；
Step 5, object judgement to be compared judges that target index value t to be compared meets t≤n and size (ORD [t])/size (ORD [s]) >=ε continues step 8；
Step 8, similarity distance calculates, and calculates Radar emitter o1With o3Between similarity distance and update similarity Matrix D istArray, as shown in table 4：
Table 4
Sample o1 o2 o3
o1 - 0 0.25
o2 - - ∞
o3 - - -
Step 9, target update to be compared updates Radar emitter index t=t+1=4 to be compared, return to step 5；
Step 5, object judgement to be compared judges that target index value t=4 to be compared is unsatisfactory for t≤n and size (ORD [t])/size (ORD [s]) >=ε, continues step 6；
Step 6, current goal is changed, update current goal index value s=s+1=2；
Step 7, current goal judges s<N, return to step 4；
Step 4, Radar emitter target index t=s+1=3 to be compared is arranged in target setting to be compared；
Step 5, object judgement to be compared, judge target index value t=3 to be compared meet t≤n and size (ORD [t])/ Size (ORD [s]) >=ε continues step 8；
Step 8, similarity distance calculates, and calculates Radar emitter o1With o3Between similarity distance and update similarity Matrix D istArray, as shown in table 5：
Table 5
Sample C1 C2 C3
C1 - 0 0.25
C2 - - 0.25
C3 - - -
Step 9, target update to be compared updates Radar emitter index t=t+1=4 to be compared, return to step 5；
Step 5, object judgement to be compared judges target index value t=4 to be compared, is unsatisfactory for t≤n and size (ORD [t])/size (ORD [s]) >=ε, continues step 6；
Step 6, current goal is changed, update current goal index value s=s+1=3；
Step 7, current goal judges, current goal index value s >=n skips to step 10；
Step 10, clustering initialization builds 3 initial clusterings according to n Radar emitter random collection, each to cluster Random collection be initialized as the random collection of corresponding Radar emitter respectively, measured value weight identifies in bracket；
Table 6
Step 11, cluster judges, judges the two cluster C that similarity distance is minimum in Distance matrix D istArray1And C2It is full Sufficient dist (C1,C2)≤ε continues step 12；
Step 12, Cluster-Fusion, agglomerative clustering C1And C2Newly to cluster C1', random collection element and its weight are updated, is surveyed Magnitude weight identifies in bracket, as shown in table 7：
Table 7
Sample PRI signal parameters (MHz) Sup MemSet
C1’ {60.5(1),69.5(1),85(1),100(1)} 2 {o1,o2}
C3 {70(1),86(1),101(1)} 1 {o3}
Step 13, distance matrix updates, and updates Distance matrix D istArray, return to step 11 based on new cluster；
Table 8
Cluster C1 C3
C1 - 0.25
C3 - -
Step 11, cluster judges, judges the two cluster C that similarity distance is minimum in Distance matrix D istArray1' and C3It is full Sufficient dist (C1',C3)≤ε continues step 12；
Step 12, Cluster-Fusion, agglomerative clustering C1' and C3Newly to cluster C1", random collection element and its weight are updated, is surveyed Magnitude weight identifies in bracket, as shown in table 9：
Table 9
Sample PRI signal parameters (MHz) Sup MemSet
C1” {60.5(0.67),69.67(1),85.3(1),100.3(1)} 3 {o1,o2,o3}
Step 13, distance matrix updates, and newly clusters C1" respective distances matrix it is as shown in table 10, return to step 11；
Table 10
Sample C1”
C1” -
Step 11, cluster judges, no corresponding cluster skips to step 14；
Step 14, high frequency mode exports, and finds the cluster C that all supports meet thresholding minsup=21", correspond to Machine set { 60.5,69.67,85.3,100.3 } is high frequency mode, output.
The achievement in research of the present invention is conducive to improve the analysis ability of random set mould assembly radar emitter signal feature, favorably In the type identification ability for further increasing Radar emitter.
The research work of the present invention has obtained state natural sciences fund (No.61771177) subsidy.

Claims (4)

1. a kind of random set mould assembly radar emitter signal parameter high frequency mode method for digging based on weighted cluster, feature exist In：For the signal parameter data of random set mould assembly airborne radar radiant source target, similarity distance matrix is built, is compared two-by-two Similarity between random set mould assembly measured value is melted on the basis of similarity distance matrix based on hierarchy clustering method iteratively Random set mould assembly cluster is closed, until it can not find the cluster for meeting random collection distance threshold, is calculated to Simultaneous Iteration formula each The support and random collection value newly clustered is finally excavated and meets the corresponding random collection value of support thresholding cluster as high Frequency pattern exports.
2. the random set mould assembly radar emitter signal parameter high frequency mode according to claim 1 based on weighted cluster is dug Pick method, it is characterised in that：Including：
(1) Radar emitter target is pressed into random collection size SORDDescending sequence；
(2) similarity distance matrix, matrix size n × n are initialized；
(3) setting current radar radiant source target indexes s；
(4) Radar emitter target to be compared is set and indexes t；
(5) judge that target index to be compared continues step 6, such as if being unsatisfactory for t≤n and S (ORD [t])/S (ORD [s]) >=ε Fruit, which meets, continues step 8；
(6) current goal index value is updated；
(7) judge that current goal indexes, if meeting s >=n, continue step 10, otherwise, return to step 4；
(8) the similarity distance dist between Radar emitter is calculated, and updates similarity matrix；
(9) Radar emitter index to be compared, return to step 5 are updated；
(10) n initial clustering is built；
(11) judge the two cluster C that similarity distance is minimum in distance matrixuAnd Cv(u<V) whether meet dist (Cu,Cv)≤ε, Continue step 12 if met, if being unsatisfactory for skipping to step 14；
(12) agglomerative clustering CuAnd CvNewly to cluster Cw, and be updated；
(13) it is based on new cluster and updates distance matrix, newly the distance definition of cluster and other clusters is to constitute original two newly clustered The minimum value of a cluster and other clustering distances, return to step 11；
(14) cluster that all supports meet thresholding minsup is found, it is exported and corresponds to height of the random collection as radar parameter Frequency pattern.
3. the random set mould assembly radar emitter signal parameter high frequency mode according to claim 2 based on weighted cluster is dug Pick method, it is characterised in that：Random collection distance threshold ε is used to compare the similarity between random measurement value set, support Thresholding minsup is for judging whether signal parameter pattern belongs to high frequency.
4. the random set mould assembly radar emitter signal parameter high frequency mode according to claim 2 based on weighted cluster is dug Pick method, it is characterised in that：Airborne radar emitter Signals measured value of parameters be one by several arbitrary measures constitute with Machine set.
CN201810182612.5A 2018-03-06 2018-03-06 Random set type radar radiation source signal parameter high-frequency mode mining method based on weighted clustering Active CN108345864B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810182612.5A CN108345864B (en) 2018-03-06 2018-03-06 Random set type radar radiation source signal parameter high-frequency mode mining method based on weighted clustering

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810182612.5A CN108345864B (en) 2018-03-06 2018-03-06 Random set type radar radiation source signal parameter high-frequency mode mining method based on weighted clustering

Publications (2)

Publication Number Publication Date
CN108345864A true CN108345864A (en) 2018-07-31
CN108345864B CN108345864B (en) 2020-09-08

Family

ID=62957435

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810182612.5A Active CN108345864B (en) 2018-03-06 2018-03-06 Random set type radar radiation source signal parameter high-frequency mode mining method based on weighted clustering

Country Status (1)

Country Link
CN (1) CN108345864B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110135450A (en) * 2019-03-26 2019-08-16 中电莱斯信息***有限公司 A kind of hotspot path analysis method based on Density Clustering

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002014896A1 (en) * 2000-08-11 2002-02-21 Alenia Marconi Systems Limited Method of interference suppression in a radar system
CN102749616A (en) * 2012-06-29 2012-10-24 北京市遥感信息研究所 Fuzzy-clustering-based Aegis system signal sorting method
CN102930255A (en) * 2012-11-13 2013-02-13 中国电子科技集团公司第二十八研究所 Signal analysis method based on radiation source category pairs
CN104794431A (en) * 2015-03-25 2015-07-22 中国电子科技集团公司第二十八研究所 Radar radiation source pulse-to-pulse mode excavation method based on fuzzy matching
CN105005029A (en) * 2015-07-17 2015-10-28 哈尔滨工程大学 Multi-mode radar signal sorting method based on data field hierarchical clustering
CN106056098A (en) * 2016-06-23 2016-10-26 哈尔滨工业大学 Pulse signal cluster sorting method based on class merging
CN106405518A (en) * 2016-12-07 2017-02-15 中国船舶重工集团公司第七二四研究所 Complex system radar signal grade correlating, clustering and sorting method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002014896A1 (en) * 2000-08-11 2002-02-21 Alenia Marconi Systems Limited Method of interference suppression in a radar system
CN102749616A (en) * 2012-06-29 2012-10-24 北京市遥感信息研究所 Fuzzy-clustering-based Aegis system signal sorting method
CN102930255A (en) * 2012-11-13 2013-02-13 中国电子科技集团公司第二十八研究所 Signal analysis method based on radiation source category pairs
CN104794431A (en) * 2015-03-25 2015-07-22 中国电子科技集团公司第二十八研究所 Radar radiation source pulse-to-pulse mode excavation method based on fuzzy matching
CN105005029A (en) * 2015-07-17 2015-10-28 哈尔滨工程大学 Multi-mode radar signal sorting method based on data field hierarchical clustering
CN106056098A (en) * 2016-06-23 2016-10-26 哈尔滨工业大学 Pulse signal cluster sorting method based on class merging
CN106405518A (en) * 2016-12-07 2017-02-15 中国船舶重工集团公司第七二四研究所 Complex system radar signal grade correlating, clustering and sorting method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JANUSZ DUDCZYK: "Radar Emission Sources Identification Based on Hierarchical Agglomerative Clustering for Large Data Sets", 《HINDAWI PUBLISHING CORPORATION JOURNAL OF SENSORS》 *
孙盼杰 等: "基于PRI熵的雷达信号聚类方法研究", 《电子信息对抗技术》 *
郭强 等: "基于距离度量方法的雷达辐射源信号识别研究", 《中国电子科学研究学报》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110135450A (en) * 2019-03-26 2019-08-16 中电莱斯信息***有限公司 A kind of hotspot path analysis method based on Density Clustering
CN110135450B (en) * 2019-03-26 2020-06-23 中电莱斯信息***有限公司 Hot spot path analysis method based on density clustering

Also Published As

Publication number Publication date
CN108345864B (en) 2020-09-08

Similar Documents

Publication Publication Date Title
Liang et al. A visual detection method for nighttime litchi fruits and fruiting stems
CN110349160B (en) SAR image segmentation method based on super-pixel and fuzzy C-means clustering
CN109409400A (en) Merge density peaks clustering method, image segmentation system based on k nearest neighbor and multiclass
CN107945215B (en) High-precision infrared image tracker and target rapid tracking method
CN110135450A (en) A kind of hotspot path analysis method based on Density Clustering
CN112684427B (en) Radar target recognition method based on serial secondary reinforcement training
CN109271427A (en) A kind of clustering method based on neighbour&#39;s density and manifold distance
CN108154157A (en) It is a kind of based on integrated quick Spectral Clustering
CN104850867A (en) Object identification method based on intuitive fuzzy c-means clustering
Tavenard et al. Balancing clusters to reduce response time variability in large scale image search
CN108154158A (en) A kind of building image partition method applied towards augmented reality
CN108873715A (en) Intelligent home control system based on gesture identification
CN108345864A (en) Random set mould assembly radar emitter signal parameter high frequency mode method for digging based on weighted cluster
Yu et al. Multi‐scale mean shift tracking
CN104463227A (en) Polarimetric SAR image classification method based on FQPSO and target decomposition
CN108344975A (en) A kind of joint cluster scaling method declined using gradient with included angle cosine
Wagner et al. Modifications of the OPTICS clustering algorithm for short-range radar tracking applications
Novitasari et al. Classification of house buildings based on land size using the K-nearest neighbor algorithm
Matthews et al. An improved storm cell identification and tracking (SCIT) algorithm based on DBSCAN clustering and JPDA tracking methods
CN107564008A (en) Rapid SAR image segmentation method based on crucial pixel fuzzy clustering
Keyvanpour et al. Feature weighting for improving document image retrieval system performance
CN105574363A (en) Feature selection method based on SVM-RFE (Support Vector Machine-Recursive Feature Elimination) and overlapping degree
Gong et al. Maximum entropy multi-threshold image segmentation based on improved particle swarm optimization
CN106327527B (en) Target profile tracing method based on Online Boosting
KR101551237B1 (en) Method and system for recognizing outdoor place based on straight line and geometric relation

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