Carleton University
Finite Length Analysis of Verifcation-Based Message Passing Algorithms in Compressed Sensing
Abstract
dc:description.abstractCompressed sensing (CS) is a newly developed area in signal processing which generally aims at compressing sparse signals. The most challenging part of a CS problem can be the recovery process in which a signal will be reconstructed from its measurements. In this thesis, we address the finite length analysis of verification-based message-passing algorithms (VB MPAs), analytically. In contrary to asymptotic regime analysis, the finite length analysis is fairly an untouched problem. We begin the analysis by characterizing the problematic combinatorial structures, defined over the measurement graph of the CS system in which VB algorithms are trapped. Then, based on this characterization, we derive exact expressions for the average probability of error of a given ensemble of graphs, and also for a given individual graph. The analysis for the individual graph is also shown to be useful for the introduction and estimation of error floors in VB MPAs in CS.
Degree
thesis:*- Name thesis:degree_name
- Master of Applied Science (M.App.Sc.)
- Level thesis:degree_level
- Master's
- Discipline thesis:degree_discipline
- Engineering, Electrical and Computer
- Grantor dc:publisher
- Carleton University
- Year dc:date.issued
- 2015
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Farhangdoust, Seyed Mohammad Ebrahim
Subjects
dc:subject × 2Rights
dc:rights- Statement dc:rights
-
- Copyright © 2015 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, research, scholarship, and teaching. Theses may only be shared by linking to Carleton University Institutional Repository and no part may be used without proper attribution to the author. No part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.
- Language dc:language.iso
- en