Complete SAT solver based on set theory

  • Authors:
  • Wensheng Guo;Guowu Yang;Qianqi Le;William N. N. Hung

  • Affiliations:
  • University of Electronic Science and Technology, Chengdu, Sichuan, China;University of Electronic Science and Technology, Chengdu, Sichuan, China;University of Electronic Science and Technology, Chengdu, Sichuan, China;Synopsys Inc., Mountain View, California

  • Venue:
  • ICICA'12 Proceedings of the Third international conference on Information Computing and Applications
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

SAT problem is a NP-complete. Many SAT benchmarks that come from the different real life SAT problems are proposed to verify the performance of solvers. Our research focuses on the Model RB benchmark which can be mapped by the coloring problem and others. We propose a translating method based on set for Model RB instances of CNF formulas, and a complete search algorithm. We use the weight of clauses based on the set to determine the order of the search. The results show our solver has the best runtime for the mostly instances and is comparable to the best SAT solvers.