The periodic noise inside a truck cabin, which is mainly produced by the truck engine, is a combination of the harmonic noise components related to the engine rotational speed. Both the frequency and the sound pressure level of these harmonic components also vary according to the engine speed. With the aim of reducing these noise components, a multiple-channel narrowband feedforward active noise control system using a modified Filtered-x Least Mean Squares (FXLMS) algorithm has been developed and already proven to some extent to be effective to attenuate the periodic noise. However, since the spatial distribution of the secondary sources and the error sensors and the frequency-dependent convergence behavior, the overall performance of the active noise control system is degraded significantly. To optimize the attenuation of the control system in the whole frequency range, another approach based on the previous algorithm is adopted by way of an equalization process to the secondary path estimates. Experimental results show that, in comparison with the previous algorithm, this updated approach achieved a faster convergence speed and larger noise attenuation while keeping the stability and reliability of the active noise control (ANC) system. In order to reduce the periodic noise inside a truck cabin, a 3-channel narrowband feedforward ANC system was constructed. In addition, a 3lambda3lambda3 feedforward MR-MIMO FXLMS algorithm was developed and also optimized by way of an equalization process to the secondary path estimates. Then, some experiments with the recorded noise were conducted to evaluate the performance of the normal FXLMS algorithm and the equalized algorithm. In comparison to the normal FXLMS algorithm, the equalized FXLMS algorithm has faster convergence speed and improved noise attenuation. Especially, the equalized algorithm has achieved 15-times faster convergence speed and over 1 dB better performance in the overall frequency range than the normal FXLMS algorithm. As an offline process, it adds no computational burden to the normal algorithm when the ANC system is running. Because of the improved performance, the equalized FXLMS algorithm is more effective than the normal FXLMS algorithm for the practical use in truck cabins.