Title:  Multichannel Myopic Deconvolution in Underwater Acoustic Channels

Committee: 

Dr. Romberg, Advisor,

Dr. Sabra, Co-Advisor  

Dr. Davenport, Chair

Dr. McClellan

Abstract:

The objective of this study is to develop new techniques for solving the multichannel blind deconvolution problem and to implement these techniques in an acoustic waveguide environment. We are revisiting this classical problem by investigating channel models and recovery methods. We will show how to use a priori information about channels to build appropriate channel models that, in turn, can be incorporated into our methods. Both linear and bilinear channel models will be investigated in our study. Our first method views solving the multichannel deconvolution problem as solving a system of bilinear equations that, in turn, can be recast as recovering a low-rank matrix from a set of linear observations. We will also further our knowledge in the classical subspace method for blind deconvolution and develop an efficient and guaranteed algorithm that will show more stability in applications. Moreover, we will investigate our channel models and methods in a multiple-source multiple-channel convolution scenario and develop a source separation framework.