Application of an ANFIS algorithm to narrowband active noise control system

This paper proposes a new narrowband active noise control (ANC) system where an ANFIS (adaptive network-based fuzzy inference system) is utilized as an adaptive controller. The ANFIS is a combination of a neural network and a fuzzy inference system. For the purpose of computational cost reduction, the nonlinear premise parameters in the ANFIS are fixed and only its linear consequent parameters are adjusted based on a gradient descent method rather than the hybrid learning rule. The proposed system may looks somewhat more complicated than the conventional FXLMS-based ANC system, but it does not require reference signal filtering at all. Simulation results reveal that the proposed ANC system enjoys sufficient effectiveness in suppressing noise signals of sinusoidal nature.