Odd even shifts in SIMD hypercubes

A linear-time algorithm is developed to perform all odd (even) length circular shifts of data in an SIMD (single-instruction-stream, multiple-data-stream) hypercube. As an application, the algorithm is used to obtain an O(M/sup 2/+log N) time and O(1) memory per processor algorithm to compute the two-dimensional convolution of an N*N image and an M*M template on an N/sup 2/ processor SIMD hypercube. This improves the previous best complexity of O(M/sup 2/ log M+log N).<
>