Improved MIMO Active Noise Control System with Integrated Adaptive Algorithms for Vehicle Powertrain Noise

Active noise control (ANC) technique has been gaining popularity for treating the interior powertrain noise in the automotive industry. In order to yield a large control area inside the cabin, a multiple-input-multiple-output (MIMO) ANC system that uses multiple loudspeakers and multiple error microphones is normally required. However, the traditional MIMO system configured with the filtered-x least mean square (FXLMS) algorithm often leads to prohibitively huge computational burden and slow convergence rate in some frequency ranges due to the dynamic characteristic of the secondary paths. To improve the computational efficiency and reduce the secondary path effect, an improved MIMO ANC system is proposed for powertrain noise control. The proposed system is tested with several advanced algorithms including the frequency domain FXLMS algorithm, channel equalization FXLMS algorithm (CE-FXLMS) and convex combination FXLMS algorithm (CC-FXLMS). The study includes a comprehensive computational complexity analysis to understand the computational cost for different algorithms. Numerical simulation are conducted to validate the salient features of these algorithms by designing a MIMO ANC system with two secondary loudspeakers and one error microphone to control the powertrain noise.