Modified Euclidean algorithm for finite field inversions

Based on a reformulated Euclidean algorithm, systolic architectures for computing inverses in finite fields have been designed with small critical path delay, O(m) latency, and O(m/sup 2/) area-time product.