Title :
A new multidimensional recursive architecture for computing the discrete cosine transform
Author :
Elnaggar, Ayman ; Alnuweiri, Hussein M.
Author_Institution :
Dept. of Electr. Eng., Sultan Qaboos Univ., Muscat, Oman
fDate :
2/1/2000 12:00:00 AM
Abstract :
This paper presents a novel recursive algorithm for generating higher order multidimensional (m-D) discrete cosine transform (DCT) by combining the computation of 2m identical lower order (smaller size) DCT architectures. One immediate outcome of our results is the true “scalability” of the DCT computation. Basically, an m-D DCT computation can be constructed from exactly one stage of smaller DCT computations of the same dimension. This is useful for both hardware and software solutions, in which a very efficient smaller size m-D DCT core has been developed, and a larger DCT computation is required. The resulting DCT networks have very simple modular structure, highly regular topology, and use simple arithmetic units
Keywords :
data compression; digital arithmetic; discrete cosine transforms; multidimensional systems; pipeline processing; recursive estimation; transform coding; video coding; DCT architecture; DCT networks; DSP chips; RISC processors; arithmetic units; discrete cosine transform; hardware solution; higher order multidimensional DCT; modular structure; multidimensional recursive architecture; recursive algorithm; regular topology; software solution; video compression standards; Arithmetic; Computer architecture; Concurrent computing; Discrete cosine transforms; Hardware; Multidimensional systems; Tensile stress; Transform coding; Very large scale integration; Video compression;
Journal_Title :
Circuits and Systems for Video Technology, IEEE Transactions on