Solving Difficult SAT Problems by Using OBDDs and Greedy Clique Decomposition

* Final gross prices may vary according to local VAT.

Get Access

Abstract

In this paper, we propose an OBDD-based algorithm called greedy clique decomposition, which is a new variable grouping heuristic method, to solve difficult SAT problems. We implement our algorithm and compare it with several state-of-art SAT solvers including Minisat, Ebddres and TTS. We show that with this new heuristic method, our implementation of an OBDD-based satisfiability solver can perform better for selected difficult SAT problems, whose conflict graphs possess a clique-like structure.

Supported by the Beijing Forestry University Young Scientist Fund No. BLX2009013, the Chinese National 973 Plan (No.2010CB328103), the ARC grants FT0991785 and DP120102489, the National Natural Science Foundation of China under Grant No. 60833001, and the CAS Innovation Program.