Potential of Particle Swarm Optimization and Genetic Algorithms for FIR Filter Design

This article studies the performance of two metaheuristics, particle swarm optimization (PSO) and genetic algorithms (GA), for FIR filter design. The two approaches aim to find a solution to a given objective function but employ different strategies and computational effort to do so. PSO is a more recent heuristic search method than GA; its dynamics exploit the collaborative behavior of biological populations. Some researchers advocate the superiority of PSO over GA and highlight its capacity to solve complex problems thanks to its ease of implementation. In this paper, different versions of PSOs and GAs including our specific GA scheme are compared for FIR filter design. PSO generally outperforms standard GAs in some performance criteria, but our adaptive genetic algorithm is shown to be better on all criteria except CPU runtime. The study also underlines the importance of introducing intelligence in metaheuristics to make them more efficient by embedding self-tuning strategies. Furthermore, it establishes the potential complementarity of the approaches when solving this optimization problem.