Cross-spectrum detector using a single acoustic vector hydrophone

Abstract

This paper aims to detect narrowband signal embedded in the noise background using a single acoustic vector hydrophone. In the isotropic noise field, processing on acoustic energy flow is the maximum likelihood ratio detector. Adopting ideas of the average power spectrum, we propose a narrowband detector based on the cross spectrum between acoustic pressure and acoustic particle velocity. Firstly, acoustic pressure and particle velocity are segmented respectively. Then different segments are averaged in the frequency domain. The real part of cross spectrum between averaged pressure and particle velocity is taken as the test statistic. According to the measurement model of vector hydrophone in the frequency domain, we can obtain the distribution characteristics of the test statistic, and further we can get receiver operating characteristic curve of the detector. The simulation and data processing results show that probability density of the test statistic is consistent with the theoretical derivation, and the detection performance of cross-spectrum detector proposed in this paper is superior than traditional average power spectrum detector.