ABSTRACT

In active noise control (ANC) systems, there exists an inherent feedback from the loudspeaker to the primary microphone. Adjoint least mean square (ALMS) algorithm is known to be an alternative to the widely used filtered x LMS (FxLMS) for reducing the computational complexity and memory requirements, especially in the case of multi-channel systems. Further FxLMS algorithm is based on the assumption that the order of the weighing filter and secondary path can be commuted which is not always true in practice. Though ALMS do not make such an assumption, neither FxLMS nor the ALMS algorithms consider the feedback path effect that is inherent in ANC systems. We propose a feedback ANC system based on ALMS algorithm which is analogous to the system based on FxLMS. Detailed computational complexity analysis for addition and multiplication requirements is presented and are compared with those of its counterpart to establish its usefulness. Simulation results show the convergence characteristics of the ALMS based ANC with feedback path modeling is on par with that based on FxLMS.