Genetic adaptive IIR filtering algorithm for active noise control

This paper presents an adaptive infinite impulse response (IIR) filtering algorithm using genetic algorithm for active noise control (ANC). Feedforward ANC systems are equivalent to system identification configuration whose reference model has infinite impulse response and input signal is colored. Stochastic gradient algorithms such as filtered-x least mean square (FXLMS) algorithm are conventionally used for their simplicity and stability. But these algorithms have disadvantages of local minimum and large eigenvalue disparity of input signal's autocorrelation matrix. To solve those problems we propose an IIR structure ANC algorithm with genetic algorithm which is well known as global optimization tools in control and signal processing area. Computer simulation shows that genetic IIR algorithm is superior to LMS IIR algorithm for insufficient case.