An approach to identification of unknown IIR systems using crossover cat swarm optimization

Summary Traditional learning techniques creates stability problem in infinite impulse response (IIR) systems identification. Additionally the performance significantly degrades if reduced order adaptive models are used for such identification. In this paper identification of IIR system is formulated as an optimization problem. This paper also proposes a modification to the cat swarm optimization algorithm i.e. crossover cat swarm optimization which always tries to explore the search space for improved solutions without getting trapped in the local optima and diverse situations. The results of actual and reduced order identification for standard system by new method exhibit superior performance as compared to cat swarm optimization and particle swarm optimization in terms of mean square error, convergence speed and estimation of coefficients.