A stochastic evolution algorithm for the graph covering problem and its application to the technology mapping

A stochastic evolution algorithm is applied to solve the graph covering problem in which a set of patterns that fully covers a subject graph with a minimal cost is sought. This problem is a typical constrained combinatorial optimization problem and is proven to be NP-complete. Many branch-and-bound algorithms with different heuristics have been proposed but most of them cannot handle practical sized problems like the technology mapping problem from the VLSI synthesis area. Our stochastic evolution is based on a problem-specific encoding scheme to reduce the size of the search space and incorporates the tree matching algorithm at the initial solution generation stage for speed-up. Experimental results show that the proposed algorithm produces good solutions within a reasonable amount of time.