A bioinspired twin inverted multiscale matched filtering method for detecting an underwater moving target in a reverberant environment

Abstract

To this day, biological sonar systems still have great performance advantages over artificial sonar systems, especially for detection in environments with clutter, strong reverberation, and a low signal to noise ratio (SNR). Therefore, mammal sonar systems, for instance, bats and toothed whales, have many characteristics worth learning from. This paper proposes a bioinspired twin inverted multiscale matched filtering method to detect underwater moving targets. This method can be mainly divided into three parts. Firstly, a hyperbolic frequency modulation (HFM) continuous wave (CW) multiharmonic detection signal was adopted after analyzing signals from bats and dolphins. This signal combines the advantages of CW and HFM signals and has excellent time measurement and speed measurement performance when detecting a moving target. Secondly, the twin inverted waveform was introduced to suppress strong linear reverberation and highlight moving targets. The pulse interval was determined by assessing the reverberation reduction time. Thirdly, when processing echoes, a multiscale matched filtering method was proposed to make use of multiharmonic information and improve detection performance. Finally, a channel pool experiment was carried out to test the performance of the proposed method. The experimental result demonstrates that the proposed method has better performance when detecting a moving target in a reverberant environment compared to the conventional matched filtering method. Related results can be applied to small underwater platforms or sensor network platforms for target detection and coastal defense applications.