Performance Analysis of Generational GA over Microbial GA for Two-level Logic Minimization with Area and Power Trade-offs

Genetic Algorithms (GA) has been widely used for logic optimization and synthesis with a view to optimize area, power or testability and for various trades-offs. Traditional logic optimizers such as ESPRESSO targets area minimization only, while present day device scaling demand for extremely low power consumption in VLSI circuits. This paper compares the performances of two types of genetic algorithm namely, the generational GA and the microbial GA in the context of two-level logic optimization targeting area and power and their trade-offs. Result of experimentation suggests that although microbial genetic algorithms works only for one generation but the number of iterations required for the convergence of optimal point is relatively large for bigger circuits and hence needs more CPU time compared to generational GA.