New source localization method using heteroscedasticity time series in passive sonar

In this paper we propose a new source localization method using underwater ambient noise modeling based on heteroscedasticity time series in array signal processing for a passive SONAR. In this application, measurement of ambient noise in natural environment shows that noise can sometimes be signifcantly non-Gaussian. Besides in many applications, such as those sensors having non-ideal hardware, involving sparse hydrophones with prevailing external noise, the assumed noise model may be simplifed by different sensors noise variances. Generalized Autoregressive Conditional Heteroscedasticity (GARCH) time series are feasible for heavy tailed probability density function (PDF) (as excess kurtosis) and time varying variances (a type of heteroscedasticity) of stochastic process. We use GARCH noise model in the Maximum Likelihood Approach for the estimation of Direction-Of-Arrivals (DOAs) of impinging sources. Through simulation, we show that the GARCH modeling is suitable for high-resolution source localization and noise suppression in an underwater environment.