Sparse signal separation methods for target detection in sea-clutter

This paper investigates two methods of sparse signal separation known as morphological component analysis and basis pursuit denoising. These techniques have both been demonstrated as effective in separating targets from sea-clutter, but rely on the tuning of different parameters. In the first part of this paper, we study the variation of the regularisation or penalty parameter and propose a value which achieves good separation. Then we present a detection scheme which relates the probability of false alarm to the choice of penalty parameter. The performance of this detection scheme is then demonstrated with Monte-Carlo simulation using the Ingara medium grazing angle sea-clutter data set.