Comparison of methodologies for signal detection in underwater acoustic noise in shallow tropical waters

Abstract

Signal detection is important in sonar and underwater digital communication. Optimum signal detection in underwater acoustic noise (UWAN) can be achieved with the knowledge of noise statistics. The assumption of Gaussian noise allows the use of linear correlation (LC) detectors. However, the non-Gaussian nature of UWAN results in the poor performance of such detectors. This study presents an empirical model of the characteristics of UWAN in the shallow waters of an acoustic underwater channel. Four detectors for the detection of known signals in UWAN are compared: locally optimal (LO) detector, sign correlation (SC) detector, LC detector, and the proposed Gaussian noise injection detector (GNID). The performance of the detectors is evaluated according to the detection probability (PD) and receiver operating characteristic curves. Given a probability of false alarm (PFA) rate of 0.01 and PD of 90 percent, the energy-to-noise ratios of the LO detector, SC detector, GIND, and LC detector are 8.854, 10.8, 10, and 12 dB, respectively. Among the four detectors, the LO detector achieves the best performance, whereas the LC detector shows the weakest performance in the presence of non-Gaussian noise.