The Parallel Achievement of Expanded-Dimension Discrete Polynomial Transform on Tight Coupling Multiprocessor

Author

Hu, Hui ; Ye, Xinhua ; Xu, Luchao

Author_Institution

Sch. of Inf. Eng., East China JiaoTong Univ., Nanchang

fYear

2008

fDate

21-22 Dec. 2008

Firstpage

723

Lastpage

726

Abstract

In this paper, a parallel discrete polynomial transform (DPT) algorithm on expanded-dimension parallelism is put forward. One dimension DPT can be decomposed into two dimension DPT. The study on achievement of the parallel DPT algorithm is carried out on tight coupling multiprocessor platform-TMS320C80. The algorithm reduces programming complexity, data relativity and limitation of processor unit´s on-chip memory capacity. Experiment results are consistent to theory analysis, and the speed and precision of the algorithm could be ensured.

Keywords

discrete transforms; multiprocessing systems; parallel algorithms; polynomials; coupling multiprocessor; data relativity; expanded-dimension discrete polynomial transform; on-chip memory capacity; parallel discrete polynomial transform; programming complexity; Algorithm design and analysis; Convolution; Discrete Fourier transforms; Discrete transforms; Fast Fourier transforms; Fourier transforms; Parallel processing; Parallel programming; Polynomials; Programming profession; expanded-dimension parallelism; parallel processing; polynomial transform; tight coupling multiprocessor;

fLanguage

English

Publisher

ieee

Conference_Titel

Knowledge Acquisition and Modeling Workshop, 2008. KAM Workshop 2008. IEEE International Symposium on

Conference_Location

Wuhan

Print_ISBN

978-1-4244-3530-2

Electronic_ISBN

978-1-4244-3531-9

Type

conf

DOI

10.1109/KAMW.2008.4810592

Filename

4810592