A New Evolutionary Algorithm based BDD Optimization for Area and Power

Anirban Dutta,K. L. Baishnab,Saurabh Chaudhary
Abstract:Reduced Ordered Binary Decision Diagram (ROBDD) is the most popular data structure for efficient representation and manipulation of Boolean functions. However, this novel data structure is very sensitive to the variable ordering, i.e. the size may vary from linear to exponential. Since finding the optimal variable ordering is an NP-complete problem and the best known algorithm has exponential runtime so in the last few years many heuristics have been presented for finding good variable orderings. But all these algorithms for input variable ordering for OBDD focus primarily on area minimization. Here we present a Genetic algorithm based approach for the computation of an efficient variable ordering heuristically, which reduces not only the size of the BDD but also the power dissipation (switching activity power and leakage) of the circuit. A trade-off has also been done for combined area and power (switching and leakage) minimization. It has been found that the proposed GA method minimizes the area by 33.16%, dynamic power by 57.28% and leakage by 29.23%.
What problem does this paper attempt to address?