Design and automatic code generation of a two-dimensional fast cosine transform for SIMD DSP architectures

Author

Lehmann, A. ; Robelly, J.P. ; Fettweis, G.

Author_Institution

Dept. of Mobile Commun. Syst., Tech. Univ. Dresden, Dresden, Germany

fYear

2005

fDate

4-8 Sept. 2005

Firstpage

1

Lastpage

4

Abstract

Fast Algorithms for the computation of the two-dimensional Discrete Fourier Transform (DCT) can be described by means of elements of Multilinear Algebra. Multilinear Algebra offers not only a formalism for describing the algorithm, but it enables the derivation by pure algebraic manipulations of an algorithm that is well suited to be implemented in vector-SIMD signal processors with different levels of parallelism. The vector formulation of the two-dimensional DCT (2D-VDCT) can be implemented in a matrix oriented language and a suitable compiler generates code for the vector architecture. We show in this paper how important speedup factors can be achieved with this methodology.

Keywords

digital signal processing chips; discrete Fourier transforms; discrete cosine transforms; matrix algebra; parallel processing; program compilers; 2D DCT; 2D discrete Fourier transform; 2D fast cosine transform; SIMD DSP architecture; automatic code generation; matrix oriented language; multilinear algebra; vector-SIMD signal processor; Algorithm design and analysis; Discrete cosine transforms; Parallel processing; Program processors; Signal processing algorithms; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Conference, 2005 13th European

Conference_Location

Antalya

Print_ISBN

978-160-4238-21-1

Type

conf

Filename

7078364