Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

전자전 환경에서 전이학습 기반 저피탐 레이더 변조 신호 분류 성능 분석Analysis of LPI Radar Waveform Classification Based on Transfer Learning in Electronic Warfare Environment

Other Titles
Analysis of LPI Radar Waveform Classification Based on Transfer Learning in Electronic Warfare Environment
Authors
서동호박지연윤우진백지현이원진남해운
Issue Date
Dec-2022
Publisher
한국통신학회
Keywords
LPI Radar; Intra-pulse Modulation; CNN; Transfer Learning; Deep learning
Citation
한국통신학회논문지, v.47, no.12, pp 2168 - 2171
Pages
4
Indexed
KCI
Journal Title
한국통신학회논문지
Volume
47
Number
12
Start Page
2168
End Page
2171
URI
https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/113177
DOI
10.7840/kics.2022.47.12.2168
ISSN
1226-4717
2287-3880
Abstract
최근 저피탐 레이더 기술의 발전에 따라 LPI 레이더 위협 신호의 정확한 탐지 및 변조 방식 분류 기술이 중요한 기술로 대두되고 있다. 특히 레이더 변조방식 분류에 관한 연구는 딥러닝 기반의 이미지 처리분야에서도 연구가 활발히 진행되고 있다. 하지만 이러한 딥러닝 기반 기술은 실제 무기체계에 적용하는데 있어 양질의 학습 데이터 확보에 어려움이 존재한다. 본 논문에서는 전자전의 신호 수집환경을 고려하여, 전이 학습을 이용하여 낮은 SNR 환경에서도 레이더 신호 분류 성능을 향상시키는 방법을 제안한다. 제안한 전이 학습 기반의 분류 방법은 –12 dB 환경에서 90% 이상의 분류 성공률을 보임을 확인하였다.
As the development of low detection radar technology, accurate detection and modulation classification technology for LPI threat signals is emerging as an important technology. In particular, research on the classification of radar modulation methods has been actively conducted recently by applying deep learning-based image processing technology. However, these deep learning-based approaches have difficulties in securing high-quality learning data when applied to weapon systems. In this paper, we propose a method to improve radar signal classification performance even in a low SNR environment using transfer learning considering the signal reception environment of electronic warfare. It was confirmed that the proposed transfer learning-based classification method showed a classification success rate of over 90% at -12 dB.
Files in This Item
Go to Link
Appears in
Collections
COLLEGE OF ENGINEERING SCIENCES > SCHOOL OF ELECTRICAL ENGINEERING > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Nam, Hae woon photo

Nam, Hae woon
ERICA 공학대학 (SCHOOL OF ELECTRICAL ENGINEERING)
Read more

Altmetrics

Total Views & Downloads

BROWSE