Hunter Akins
Title: Mode estimation to perform ignorant localization
Abstract: Passive source localization (just listening, no "pinging") is possible in underwater acoustics due to the complex interaction of sound with the environment. The major caveat is that you have to accurately model the environment to take advantage of the complexity. In shallow water environments and at low frequency, sound is modeled with normal modes. A model for the seafloor (geoacoustic model) is required to calculate these modes. However a seafloor model can be hard to obtain. I present a method for using the water-column sound speed profile (SSP) and noise from a ship of opportunity to extract the normal modes, effectively replacing the need for a geoacoustic model. The method is demonstrated on SWellEx-96 experiment data, and the extracted normal modes are used for localization.
Bio: Hunter is a fourth-year student in the AOS program at Scripps, advised by Bill Kuperman and Bruce Cornuelle. His favorite movie is Roadhouse (1989), his favorite flavor of ice cream is pistacchio, and his research is on passive sonar signal processing methods for source localization and environment estimation. Contact: fakins@ucsd.edu