Signal structure information-based data association for maneuvering targets with a convolutional Siamese network

Chang Gaol, Hongwei Liu, Pramod K. Varshney, Junkun Yan, Penghui Wang

Research output: Chapter in Book/Entry/PoemConference contribution

1 Scopus citations

Abstract

Traditional data association methods usually suffer performance degradation caused by model mismatch when target maneuvers occur. In addition, the signal structure information (SSI) available from the employed signal processing techniques is rarely used in data association due to the lack of appropriate tools. To overcome the model mismatch problem brought by the target maneuvers, we resort to a data driven method to design the data association algorithm. To make full use of the SSI, a convolutional neural network is used to transform the SSI into a feature space, where the similarity of the SSI between two observations can be calculated using the Euclidean distance. The resulting structure is the convolutional Siamese network (CSN). By comparing the similarity with a predetermined margin parameter, the decision about association can be made. Simulation results are presented to show that the use of SSI in the designed CSN can improve the association probability of maneuvering targets.

Original languageEnglish (US)
Title of host publication2021 CIE International Conference on Radar, Radar 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages872-875
Number of pages4
ISBN (Electronic)9781665498142
DOIs
StatePublished - 2021
Externally publishedYes
Event2021 CIE International Conference on Radar, Radar 2021 - Haikou, Hainan, China
Duration: Dec 15 2021Dec 19 2021

Publication series

NameProceedings of the IEEE Radar Conference
Volume2021-December
ISSN (Print)1097-5764
ISSN (Electronic)2375-5318

Conference

Conference2021 CIE International Conference on Radar, Radar 2021
Country/TerritoryChina
CityHaikou, Hainan
Period12/15/2112/19/21

Keywords

  • convolutional neural network
  • data association
  • deep neural network
  • maneuvering target tracking

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Signal Processing
  • Instrumentation

Fingerprint

Dive into the research topics of 'Signal structure information-based data association for maneuvering targets with a convolutional Siamese network'. Together they form a unique fingerprint.

Cite this