Comparison of various greedy approaches for CS detection

Tutor: Henning Schepker
Type of Thesis: Project (MSc)
date of end: 04/2013
Student: Ashish Sharma
Status: finished

Compressive Sensing (CS) is a new method of detection that has emerged in recent years. It allows the detection of sparse signals from under-determined systems. While the detection was originally defined for convex optimization, various variants of the greedy Matching Pursuit (MP) algorithm have been created to solve the problem more efficiently.

The goal of this thesis is to first get an overview of the variants of MP, that have defined in the literature (e.g. OMP, CoSaMP, StOMP, ROMP, PrOMP, LSOP, SOMP, and OLS), and compare how they differ from one another. Later these algorithms should be implemented in MATLAB and their performance compared for a few specific sample scenarios.

Last change on 30.04.2013 by H. Schepker
AIT ieee GOC tzi ith Fachbereich 1
© Department of Communications Engineering - University of BremenImprint / Contact