Design methodology for the implementation of multidimensional circular convolution

Author

Soudris, D.J. ; Paliouras, V. ; Stouraitis, T. ; Thanailakis, A.

Author_Institution

Dept. of Electr. & Comput. Eng., Democritus Univ. of Thrace, Xanthi, Greece

Volume

144

Issue

6

fYear

1997

fDate

12/1/1997 12:00:00 AM

Firstpage

324

Lastpage

328

Abstract

A systematic methodology for the parallel implementation of the multidimensional circular convolution algorithm is introduced. The methodology can be applied when at least one of the convolved sequences is separable. The multidimensional algorithm is described hierarchically by two levels of two-dimensional directed graphs. The top level corresponds to a graph, each node of which represents a one-dimensional convolution, while the lower level graph specifies the elementary operations within the one-dimensional algorithm. The array architectures result systematically from the application of novel partitioning and scheduling techniques onto the top-level graph. Depending on the proposed partitioning strategy, highly-pipelined architectures with either serial or parallel loading of data can be derived

Keywords

convolution; directed graphs; parallel architectures; pipeline processing; scheduling; sequences; array architectures; convolved sequences; highly-pipelined architectures; multidimensional circular convolution; one-dimensional convolution; parallel data loading; parallel implementation; partitioning; scheduling techniques; serial data loading; two-dimensional directed graphs;

fLanguage

English

Journal_Title

Circuits, Devices and Systems, IEE Proceedings -

Publisher

iet

ISSN

1350-2409

Type

jour

DOI

10.1049/ip-cds:19971548

Filename

646814