Real-Time Detection of Electric Field Signal of a Moving Object Using Adjustable Frequency Bands and Statistical Discriminant for Underwater Defense

Abstract

The electric field method is robust in detecting underwater moving target objects in acoustically noisy conditions, such as a shallow water environment. However, it requires an enhanced signal processing algorithm to cope with the time-variant sea noise which distorts the target response and decreases data quality and detection performance. We propose a novel signal processing algorithm with superior performance to solve this problem. The proposed signal processing algorithm decomposes a measured signal into several frequency coefficients. It then extracts the target response using the coefficients in a frequency range chosen by a real-time statistical discriminant depending on the noise. We verify the effectiveness of the proposed signal processing algorithm through a field experiment, which can be a challenging task for conventional algorithms. We expect that this study will contribute to several fields of anomaly detection, especially in offshore defense and surveillance.