The focus of this work is on arrival time and amplitude estimation from acoustic signals recorded at spatially separated hydrophones in the ocean. A particle filtering approach is developed that treats arrival times as targets and tracks their location across receivers, also modeling arrival time gradient. The method is evaluated via Monte Carlo simulations and is compared to a maximum likelihood estimator, which does not relate arrivals at neighboring receivers. The comparison demonstrates a significant advantage in using the particle filter. It is also shown that posterior probability density functions of times and amplitudes become readily available with particle filtering.
All Science Journal Classification (ASJC) codes
- Arts and Humanities (miscellaneous)
- Acoustics and Ultrasonics