A reduced complexity adaptive legendre neural network for nonlinear active noise control

This paper proposes a novel low complexity nonlinear active noise control (ANC) system. The nonlinear controller is composed of an adaptive Legendre neural network (LeNN), updated using a filtered-l least mean square (FlLMS) algorithm. The computational complexity of the proposed scheme has been further reduced by incorporating the principle of partial update adaptive algorithms. Simulation study demonstrates comparable performance of the new ANC method with that of the conventional nonlinear ANC schemes, with reduced computational complexity.