Digital stable IIR low pass filter optimization using particle swarm optimization with improved inertia weight

In this paper, a recently proposed global heuristic search algorithm, namely, particle swarm optimization with improved inertia weight (PSOIIW) approach is considered for the design of the 8th order infinite impulse response (IIR) low pass (LP) digital filter. Unlike conventional PSO, inertia weight component in the proposed technique is calculated separately for each particle and modification is based on the iteration dependent velocity and position components. With this approach better global and local exploration is achieved which resulted in faster convergence to near global optimal solution. Performance of the proposed PSOIIW based approach is compared with some well accepted evolutionary algorithms such as particle swarm optimization (PSO) and real coded genetic algorithm (RGA). From the simulation study it is established that the proposed optimization technique PSOIIW outperforms RGA and PSO, not only in the accuracy of the designed filter but also in the convergence speed and solution quality, i.e., the stop band attenuation, transition width, pass band and stop band ripples. Further, the pole zero analysis justifies the stability of the designed optimal IIR filter.