A semi-Gray encoding algorithm for low-power state assignment

In sequential circuit synthesis, state encoding is to assign binary codes to given symbolic states such that a specific objective function such as area or power dissipation can be minimized in the final implementation. For low power design, reducing the switching activity of state bits is of special interest. In this paper we present a near-optimal semi-Gray encoding technique to minimize the switching activity for any given state transition graphs. Experiments with standard benchmarks show that there is a significant improvement over previous algorithms in terms of both speed and result quality.