TITLE:
Parametrically Optimal, Robust and Tree-Search Detection of Sparse Signals
AUTHORS:
A. T. Burrell, P. Papantoni-Kazakos
KEYWORDS:
Sparse Signals; Detection; Robustness; Outlier Resistance; Tree Search
JOURNAL NAME:
Journal of Signal and Information Processing,
Vol.4 No.3,
August
21,
2013
ABSTRACT:
We consider sparse signals embedded in
additive white noise. We study parametrically optimal as well as tree-search
sub-optimal signal detection policies. As a special case, we consider a constant
signal and Gaussian noise, with and without data outliers present. In the
presence of outliers, we study outlier resistant robust detection techniques.
We compare the studied policies in terms of error performance, complexity and
resistance to outliers.