Comparison of Cat Swarm Optimization with particle swarm optimization for IIR system identification

Infinite impulse response (IIR) adaptive filters have been developed to identify IIIR systems, but system identification is challenging due to non-unimodality of the error surface and the non-linear relationship between the error signal and the system parameters. Cat Swarm Optimization (CSO) was recently introduced to solve optimization problems with a new learning rule to achieve better performance than particle swarm optimization (PSO). Also, it has been used for IIR system identification. This paper examines the parameters of CSO to optimize them for IIR system identification with a few benchmarked IIR plants. Results demonstrate better performance for the CSO algorithm when compared to the inertia-weighted PSO algorithm.