Enhanced broadband matched field processing through high resolution techniques

One of the important problems in ocean acoustics is to localize an underwater wideband sound source from the array hydrophone data. Among the algorithms developed, estimation of source parameters via maximum likelihood or beamforming techniques have been greatly exploited. This paper introduce two algorithms for estimating the wideband parameters. These algorithms are based on the statistics of the received signals. Firstly, a blind algorithm initially introduced by Friedlander et al. (1997) is extended to this problem. Secondly, a maximum likelihood estimator is developed. The performance of the proposed algorithms are studied on the real data recorded during an underwater acoustic experiments.