New & improved models for SAT-based bi-decomposition

Boolean function bi-decomposition is pervasive in logic synthesis. Bi-decomposition entails the decomposition of a Boolean function into two other simpler functions connected by a simple two-input gate. Existing solutions are based either on Binary Decision Diagrams (BDDs) or Boolean Satisfiability (SAT). Furthermore, the partition of the input set of variables is either assumed, or an automatic derivation is required. Most recent work on bi-decomposition proposed the use of Minimally Unsatisfiable Subformulas (MUSes) or Quantified Boolean Formulas (QBF) for computing, respectively, variable partitions of either approximate or optimum quality. This paper develops new group-oriented MUS-based models for addressing both the performance and the quality of bi-decompositions. The paper shows that approximate MUS search can be guided by the quality of well-known metrics. In addition, the paper improves on recent high-performance approximate models and versatile exact models, to address the practical requirements of bi-decomposition in logic synthesis. Experimental results obtained on representative benchmarks demonstrate significant improvement in performance as well as in the quality of decompositions.

[1]  H. A. Curtis,et al.  A new approach to The design of switching circuits , 1962 .

[2]  Jie-Hong Roland Jiang,et al.  Bi-decomposing large Boolean functions via interpolation and satisfiability solving , 2008, 2008 45th ACM/IEEE Design Automation Conference.

[3]  Sarma B. K. Vrudhula,et al.  BDD Based Decomposition of Logic Functions with Application to FPGA Synthesis , 1993, 30th ACM/IEEE Design Automation Conference.

[4]  Christoph Scholl.  Functional decomposition with applications to FPGA synthesis , 2001 .

[5]  Tsutomu Sasao,et al.  On bi-decomposition of logic functions , 1997 .

[6]  Hans K. Buning,et al.  Propositional Logic: Deduction and Algorithms , 1999 .

[7]  Sze-Tsen Hu.  ON THE DECOMPOSITION OF SWITCHING FUNCTIONS , 1961 .

[8]  Niklas Sörensson,et al.  An Extensible SAT-solver , 2003, SAT.

[9]  Kartik Mohanram,et al.  Bi-decomposition of large Boolean functions using blocking edge graphs , 2010, 2010 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[10]  Bernd Steinbach,et al.  An algorithm for bi-decomposition of logic functions , 2001, Proceedings of the 38th Design Automation Conference (IEEE Cat. No.01CH37232).

[11]  Joao Marques-Silva,et al.  Improvements to satisfiability-based boolean function bi-decomposition , 2011, 2011 IEEE/IFIP 19th International Conference on VLSI and System-on-Chip.

[12]  Mikolás Janota,et al.  QBf-based boolean function bi-decomposition , 2012, 2012 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[13]  Jie-Hong Roland Jiang,et al.  To SAT or Not to SAT: Scalable Exploration of Functional Dependency , 2010, IEEE Transactions on Computers.

[14]  Shih-Chieh Chang,et al.  Technology mapping for TLU FPGAs based on decomposition of binary decision diagrams , 1996, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[15]  Jie-Hong R. Jiang,et al.  To SAT or not to SAT: Ashenhurst decomposition in a large scale , 2008, ICCAD 2008.