The aim of this paper is to show the experimental results achieved in the attenuation of periodic disturbances inside a vehicle with two Active Noise Control algorithms implemented on the TMS320C6701 DSP and to compare the computational complexity of both strategies: (1) Modified FxGAL: Modified filtered-x gradient adaptive lattice algorithm. This technique is based on the signal orthogonalization carried out by an adaptive lattice predictor in a previous stage. (2) Gì-FxSLMS: Filtered-x sequential least mean square algorithm with step-size gain. This strategy is based on partial updates of the weights of an adaptive filter as well as on the controlled increase in step-size of the algorithm. This work illustrates by means of two different algorithms the tradeoff established among computational costs, convergence rate, stability and mean- square error excess when DSP-based strategies are used in control systems focused on the attenuation of acoustic disturbances.
Cite as: Ramos, P., Vicente, L., Torrubia, R., López, A., Salinas, A., Masgrau, E. (2005) On the complexity-performance tradeoff of two active noise control systems for vehicles. Proc. Biennial on DSP for In-Vehicle and Mobile Systems, paper M1-2
@inproceedings{ramos05_dspinv,
author={Pedro Ramos and Luis Vicente and Roberto Torrubia and Ana López and Ana Salinas and Enrique Masgrau},
title={{On the complexity-performance tradeoff of two active noise control systems for vehicles}},
year=2005,
booktitle={Proc. Biennial on DSP for In-Vehicle and Mobile Systems},
pages={paper M1-2}
}