Publication Details
A SAT-based Fitness Function for Evolutionary Optimization of Polymorphic Circuits
multifunction logic, logic synthesis, genetic programming
Multifunctional (or polymorphic) gates have been utilized as building blocks for multifunctional circuits that are capable of performing various logic functions under different settings of control signals. In order to effectively synthesize the polymorphic circuits, several methods have been developed in the recent years. Unfortunately, the methods are applicable for small circuits only. In this paper, we propose a SAT-based functional equivalence checking algorithm to eliminate the fitness evaluation time which is the most critical overhead for genetic programming-based design and optimization of complex polymorphic circuits. The proposed approach has led to a 30% reduction of gates with respect to the solutions created using the polymorphic multiplexing combined with the optimization conducted by the ABC tool.
@INPROCEEDINGS{FITPUB9750, author = "Luk\'{a}\v{s} Sekanina and Zden\v{e}k Va\v{s}\'{i}\v{c}ek", title = "A SAT-based Fitness Function for Evolutionary Optimization of Polymorphic Circuits", pages = "715--720", booktitle = "Proc. of the 2012 Design, Automation and Test in Europe", year = 2012, location = "Dresden, DE", publisher = "European Design and Automation Association", ISBN = "978-1-4577-2145-8", doi = "10.1109/DATE.2012.6176563", language = "english", url = "https://www.fit.vut.cz/research/publication/9750" }