doi:10.1016/j.jsv.2004.08.007 
Copyright © 2004 Elsevier Ltd All rights reserved.
Analysis and DSP implementation of an ANC system using a filtered-error neural network
Ya-Li Zhoua, 
, 
, Qi-Zhi Zhanga, Xiao-Dong Lib and Woon-Seng Ganc
aDepartment of Computer Science and Automation, Beijing Institute of Machinery, P.O. Box 2865, Beijing 100085, People's Republic of China
bInstitute of Acoustic, Academia Sinica, Beijing 100080, People's Republic of China
cSchool of Electronics and Electrical Engineering, Nanyang Technological University, Singapore 639798, Singapore
Received 16 October 2003;
revised 8 June 2004;
accepted 11 August 2004.
Available online 23 November 2004.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
In this paper, feedforward active noise control (ANC) using a neural network (NN) based on filtered-error back-propagation (BP) algorithm is considered. The filtered-error BP NN (FEBPNN) algorithm is first derived, and the difference between the FEBPNN algorithm and the filtered-X BP NN (FXBPNN) algorithm is given to show that the FEBPNN algorithm offers computational advantage over the FXBPNN algorithm. Computer simulations are carried out to compare the FEBPNN algorithm with the filtered-X least mean square (FXLMS) algorithm and the FXBPNN algorithm. The controllers based on the FEBPNN algorithm and the FXLMS algorithm are implemented on a Texas Instruments digital signal processor (DSP) TMS320VC33. The simulations and the experimental verification tests show that the FEBPNN algorithm performs as well as the FXLMS algorithm for a linear control problem, and better for a nonlinear control problem, at the same time, the simulations and the experimental verification tests also show that the convergence rate of the FEBPNN is acceptable, and the FEBPNN has better tracking ability under changes of the primal signal, the primary path and the secondary path. The experiments also lead to the conclusion that more work is required to improve the predictability and consistency of the performance of the NN controller based on the FEBPNN algorithm.
Fig. 1. Feedforward ANC system in a duct.
Fig. 2. Block diagram of ANC system using the NN algorithm.
Fig. 3. The NN controller.
Fig. 4. Block diagram of the FEBPNN algorithm.
Fig. 5. Power spectrum of active noise canceling errors for linear case.
Fig. 6. Power spectrum of active noise canceling errors for nonlinear case.
Fig. 7. The mean square error in error microphone versus the number of iterations.
Fig. 8. The mean square error in error microphone versus the number of iterations when the primary signal is changed.
Fig. 9. The mean square error in error microphone versus the number of iterations when the second path is changed.
Fig. 10. The mean square error in error microphone versus the number of iterations when the primary path is changed.
Fig. 11. The schematic diagram of ANC system.
Fig. 12. The experimental duct system of active noise attenuation. (A) The duct, (B) the DSP platform, (C) signal processing board.
Fig. 13. Frequency response of primary path P(Z).
Fig. 14. Frequency response of secondary path S(Z).
Fig. 15. The block diagram of the off-line secondary path identification.
Fig. 16. Amplitude of secondary path output. (a) The amplitude of e(n), (b) the amplitude of r(n), (c) the amplitude of e′(n).
Fig. 17. Power spectrum of secondary path S(Z).
Fig. 18. Impulse response of secondary path S(Z).
Fig. 19. The output signal of the error microphone.
Fig. 20. Error signal spectrum for linear case.
Fig. 21. Error signal spectrum for nonlinear case.
Fig. 22. Error signal versus number of iterations.
Fig. 23. Error signal versus number of iterations when the primary signal is changed.
Fig. 24. Block diagram of ANC system based on FEBPNN with on-line secondary path modeling using additive random noise.
Fig. 25. Error signal versus number of iterations when the secondary path is changed.
Abstract
Purchase PDF (477 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (5)
Purchase PDF (1368 K)
