Modified sign algorithm for a constrained adaptive IIR notch filter

Modified sign algorithm (MSA) for a second-order constrained adaptive IIR notch filter (ANF) is presented in this paper. The proposed technique is based on minimizing the mean value of the product of the two signals, which are respectively produced by the FIR and the IIR sections of the ANF. Due to steep characteristic of the employed cost function, it thus makes gradient based adaptive algorithm converge to an optimum solution rapidly resulting in the faster convergence speed than of using the conventional method. In addition, extensive simulations are provided to confirm the performance of the proposed technique.