Design of Microwave Filters Using Swarm-Based Optimization Techniques

Abstract

The traditional and random drift particle swarm optimization algorithms have been applied to the design of microwave band-pass filter in this paper. The random drift particle swarm optimization has been tuned for best performance through proper choice of control parameter with respect to the defined cost function. The performance of the tuned random drift particle swarm optimization is further compared with that of the conventional particle swarm optimization algorithm. The study on performance comparison has been done with two different swarm sizes of 40 and 30. Random drift particle swarm optimization depicts rapid convergence rate and statistically superior performance compared to the traditional particle swarm optimization algorithm for both the population sizes. Although the thermal coefficient in random, drift particle swarm optimization is varied dynamically over iterations, the drift coefficient values of 1.2 and 1.3 prove to be optimal for population sizes of 40 and 30, respectively. The resultant filter based on the parameters obtained from the best run of the random drift particle swarm optimization shows a better frequency response compared to that from the traditional particle swarm optimization.