Radar Pulse Sequence Loss Position Estimation Based on KNW

Le résumé fourni par la source

In radar signal sorting, obtaining a complete pulse descriptor words (PDWs) sequence is essential to accurately compute the inter-pulse information. A single-site reconnaissance receiver is unable to perceive missing pulses. In contrast, multi-site reconnaissance receivers can achieve data fusion of pulses, enabling the estimation of the positions of the missing pulses. In this paper, a method named K-Means-Needleman-Wunsch (KNW) is proposed to estimate the missing positions of the PDWs. Initially, an unsupervised clustering approach is employed to cluster the PDWs from various stations, ensuring the correlation between sequences when comparing the temporal order of sequences across multiple stations. Then, an improved Needleman-Wunsch algorithm is used to pair and align the clustered results, identifying sequences corresponding to the same radiation source and forming a matrix with missing positions. Our experimental results demonstrate that the proposed KNW algorithm exhibits strong robustness against interference. This low computational complexity approach removes barriers to applying matrix completion algorithms to PDWs sequence completion in multi-site reconnaissance receivers.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.