A new variable tap-length and step-size FxLMS algorithm

The filtered-X least mean-square (FxLMS) algorithm is widely used for active noise control (ANC). A long tap-length is usually required for some FxLMS applications, and consequentially the convergence rate becomes very slow. In this letter, a new variable tap-length and step-size FxLMS algorithm is proposed, especially suited to a long tap-length filter. Taking into account the lowpass filter effect in the secondary path of ANC, the new algorithm is developed for the control filter with an unsymmetric and two-sided exponential decay envelop over its impulse response. Simulation results show that the new algorithm provides faster convergence and cancellation performance compared to previously proposed variable-step-size FxLMS algorithms.