Абстрактный

Performance of Hybrid Genetic Algorithm on Digital Circuit Area Minimization

Rajine Swetha R , DR Sumithra devi KA

At present, evolution based Genetic algorithm is used for optimization of the digital circuits. In physical design backend of Chip designing, these algorithms play a major role in optimizing the various design constraints. The current age design automation tools provide the designer with a predefined optimized circuits without further choice in terms of backend algorithms. This paper explores a possible combination of move based algorithm of Fidducia Matthyses effectively used with minimum distance based genetic algorithm. This approach ensures the designer of maximum possible optimization of the circuit possible. The aim of this paper is to apply FM-Mincut hybrid algorithm on bench mark circuits and discuss the effect of it on the area of the design. This approach provides a scope to unify the stages of partitioning and placement and optimize the physical design process and floor area of the application.