Improved variable step size (IVSS) LMS for active noise control system

Adaptive filters based on the least-mean-square (LMS) algorithm in active noise control (ANC) system have an important limitation that the selection of a certain value for the step size implies compromise between speed of convergence and steady-state misadjustment. Traditional variable step size (VSS) LMS algorithms aiming at this problem have little efficiency in the high filter order case because of the small step variable range limited by the convergence condition. A proposed algorithm based on a novel principle is established to overcome this shortcoming. By increasing the variable range of the step size, it raises the convergence speed and tracking ability in the steady state of the system and lowers the steady-state misadjustment error in the high filter order case in ANC system.