Abstract
Signal processing with the Matching Pursuit algorithm requires a parallel, scalable and well balanced implementation in order to be effective. Starting from a careful analysis of execution times, a scalable parallelization, based on explicit message passing, is presented. It is based on a regular scattered decomposition of the dictionary, which allows workload balancing. An experimental implementation is discussed and related timings are presented.
This work has been partially supported by the Italian National Research Council (CNR) as part of the Strategic Project “Mathematics for Industry and Society”.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Bergeaud F., Mallat, S.: Matching pursuit of image, Proc. ICIP, 1 (1993) 53–56.
Corvi M., Dodero G., Gianuzzi V., Moscati M., Parallelization Analysis for the matching pursuit, Proceedings SIMAI Conference, Salice Terme (Italy) 1996, 558–560.
ftp://ftp.cs.nyu.edu/pub/wave/software/mpp.tar.Z
v.Hanxleden R., Ridgway Scott L.,: Load balancing on message passing architectures, J. Parall. Distrib. Computing, 13 (1991) 312–324.
Mallat S., Zhang Z.,: Matching pursuit with time-frequency dictionaries, IEEE Trans. Signal Proc., 41 (1993) 3397.
Moscati M.,: Parallelization of the matching pursuit algorithm with Gabor dictionary for signal analysis, Master Thesis in Computer Science, University of Genova (in Italian) (1997).
Moscati M., Sacchetto G.,: Halting criteria and reconstruction quality: a comparison of GMP and WPMP. Tech. Rep. DISI-TR-97-09 (in Italian) (1997).
Sunderam V.S.,: PVM a framework for parallel distributed computing, Concurrency Pratice and Experience, 2(4) (1990) 315–339.
Uhl A.,: Adapted wavelet analysis on moderate parallel distributed memory MIMD architectures, Proc. IRREGULAR'95, LNCS 980, (1995), 275–283.
Zemla A., Wavelet transform computing on PVM, in Proceedings Parallel Scient. Comp. 94, LNCS 879, (1994), 534–546.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Dodero, G., Gianuzzi, V., Moscati, M., Corvi, M. (1998). A scalable parallel algorithm for matching pursuit signal decomposition. In: Sloot, P., Bubak, M., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1998. Lecture Notes in Computer Science, vol 1401. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0037173
Download citation
DOI: https://doi.org/10.1007/BFb0037173
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64443-9
Online ISBN: 978-3-540-69783-1
eBook Packages: Springer Book Archive