A hybrid approach based on tissue P systems and artificial bee colony for IIR system identification

摘要

This paper presents a hybrid approach for infinite impulse response (IIR) system identification, called ABC-PS, that combines artificial bee colony (ABC) and tissue P systems. A tissue P system with fully connected structure of cells has been considered as its computing framework. A modification of ABC was developed as evolution rules for objects according to fully connected structure and communication mechanism. With the control of the object's evolution-communication mechanism, the tissue P system designed can effectively and efficiently identify the optimal filter coefficients for an IIR system. The performance of ABC-PS was compared with artificial bee colony and several other evolutionary algorithms. Simulation results show that ABC-PS is superior or comparable to the other algorithms for the employed examples and can be efficiently used for IIR system identification.