Виявлення малих літальних апаратів за акустичним випромінюванням

Abstract

In this paper explores acoustic emission analysis as a method for drone detection, emphasizing the quieter electric engines of remotely controlled UAVs. Through examining the spectral power density (SPD) of acoustic emissions from drones and considering the impact of the Doppler effect on perceived sound frequency, this study develops a mathematical model for identifying drone characteristics. The research utilizes neural networks to analyze acoustic signals, enhancing the detection accuracy even in noisy environments or when signals are weak. The findings demonstrate the potential of acoustic methods to improve the security of critical infrastructures and public events by providing detailed insights into drone trajectory and behavior.