Detailed Information

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

BiLSTM-Attention과 LRP를 활용한 시계열 LPI 신호분류 및 중요도 분석Time-Series LPI Signal Classification and Relevance Analysis Using BiLSTM-Attention with LRP

Other Titles
Time-Series LPI Signal Classification and Relevance Analysis Using BiLSTM-Attention with LRP
Authors
Park, KiwanNam, Haewoon
Issue Date
Dec-2024
Publisher
Korean Institute of Communications and Information Sciences
Keywords
BiLSTM; Electronic Warfare (EW); Layer wise Relevance Propagation; LPI Signal Classification; Self Attention
Citation
Journal of Korean Institute of Communications and Information Sciences, v.49, no.12, pp 1695 - 1697
Pages
3
Indexed
SCOPUS
KCI
Journal Title
Journal of Korean Institute of Communications and Information Sciences
Volume
49
Number
12
Start Page
1695
End Page
1697
URI
https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/125521
DOI
10.7840/KICS.2024.49.12.1695
ISSN
1226-4717
2287-3880
Abstract
본 연구는 BiLSTM-Attention 모델과 Layer-wise Relevance Propagation (LRP)를 적용하여 저피탐지성(LPI) 신호의 분류와 중요도 분석을 수행하였다. 본 연구의 주요 목적은 시계열 데이터로 학습된 모델의 예측 근거를 LRP를 통해 해석하고, 이를 통해 모델의 예측과정에서 의미있는입력특성을효과적으로 식별하는 것이다. LRP를 활용한 해석 결과, 모델이 시계열 도메인에서 학습된 상태임에도 불구하고, 예측근거가 FFT(고속 푸리에 변환)로 변환된 주파수 도메인에서도 높은 일관성을 유지하며 주요 주파수 성분을 정확하게 탐지할 수 있음을 확인하였다. 다양한 SNR(신호 대 잡음비) 환경에서의 실험을 통해 본 모델은 신뢰도 높은 분류 성능을 유지할 뿐만 아니라,LRP 기반 해석을 통한 중요 지점탐지성능역시안정적임을 확인하였다.
This study applies a BiLSTM-Attention model and Layer-wise Relevance Propagation (LRP) to classify and analyze the importance of low probability of intercept (LPI) signals. The goal is to interpret the predictions of a time-series trained model using LRP and effectively identify meaningful input features. The analysis shows that the model maintains high consistency in its prediction rationale even in the frequency domain, transformed through Fast Fourier Transform (FFT). Experiments across various Signal-to-Noise Ratio (SNR) conditions confirm that the model delivers reliable classification performance while ensuring stable detection of key features through LRP-based interpretation. © 2024, Author. All rights reserved.
Files in This Item
There are no files associated with this item.
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