A Parallel and Reconfigurable Architecture for Efficient OMP Compressive Sensing Reconstruction

Abstract:

Compressive Sensing (CS) is a novel scheme, in which a 
signal that is sparse in a known transform domain can be
reconstructed using fewer samples. However, the signal re-
construction techniques are computationally intensive and
power consuming, which make them impractical for em-
bedded applications. This work presents a parallel and re-
configurable architecture for Orthogonal Matching Pursuit
(OMP) algorithm, one of the most popular CS reconstruc-
tion algorithms. In this paper, we are proposing the first re-
configurable CS reconstruction architecture which can take
different image sizes with sparsity up to 32. The aim is
to both minimize the hardware complexity, area and power
consumption, and improve the reconstruction latency. First,
the accuracy of reconstructed images is analyzed for different
sparsity values and fixed point word length reduction. Next,
efficient parallelization techniques are applied to reconstruct
signals with variant signal lengths of N. The OMP algorithm 
is mainly divided into three kernels, where each kernel
is parallelized to reduce execution time, and efficient reuse
of the matrix operators allows us to reduce area. The pro-
posed architecture can reconstruct images of different sizes
and measurements and is implemented on a Xilinx Virtex
7 FPGA. The results indicate that, for a 128x128 image
reconstruction, the proposed architecture is 2.67x to 1.8x
faster than the previous non-reconfigurable work which uses
much smaller sparsity.

Paper

Paper

  PDF (3.0 MB),   (c) Copyright, 2015, IEEE

Reference

Amey Kulkarni, Houman Homayoun and Tinoosh Mohsenin " A Parallel and Reconfigurable Architecture for Efficient OMP Compressive Sensing Reconstruction", 24th GLSVLSI 2014,Houston, Texas, USA,May2014

BibTeX Entry

@inproceedings{Kulkarni:2014:PRA:2591513.2591598,
 author = {Kulkarni, Amey M. and Homayoun, Houman and Mohsenin, Tinoosh},
 title = {A Parallel and Reconfigurable Architecture for Efficient OMP Compressive Sensing Reconstruction},
 booktitle = {Proceedings of the 24th Edition of the Great Lakes Symposium on VLSI},
 series = {GLSVLSI '14},
 year = {2014},
 isbn = {978-1-4503-2816-6},
 location = {Houston, Texas, USA},
 pages = {299--304},
 numpages = {6},
 url = {http://doi.acm.org/10.1145/2591513.2591598},
 doi = {10.1145/2591513.2591598},
 acmid = {2591598},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {FPGA, OMP, compressive sensing, high performance and reconfigurable architecture},
}