Distributed memory parallel architecture based on modular linear arrays for 2-D separable transforms computation

A framework for mapping systematically 2-dimensional (2-D) separable transf orms into a parallel architecture consisting of fully pipelined linear arra y stages is presented. The resulting model architecture is characterized by its generality, high degree of modularity, high throughput, and the exclus ive use of distributed memory and control. There is no central shared memor y block to facilitate the transposition of intermediate results, as it is c ommonly the case in row-column image processing architectures. Avoiding sha red central memory has positive implications for speed, area, power dissipa tion and scalability of the architecture. The architecture presented here m ay be used to realize any separable 2-D transform by only changing the coef ficients stored in the processing elements. Pipelined linear arrays for com puting the 2-D Discrete Fourier Transform and 2-D separable convolution are presented as examples and their performance is evaluated.