Intelligent Noise Cancellation Using Adaptive RNN-Based Filtering
DOI:
https://doi.org/10.65405/.v10i37.597Keywords:
adaptive filter, dynamic neural network, NARXAbstract
This study details the design and implementation of an Adaptive Noise Canceller (ANC) using a Recurrent Neural Network (RNN), especially a Nonlinear Autoregressive model with exogenous inputs (NARX). The suggested approach is designed to efficiently remove non-stationary and colored noise from voice signals, a prevalent issue in settings like airplane cockpits. The dynamic recurrent architecture of the NARX network facilitates the improved modelling of intricate noise patterns for precise prediction and cancellation. The process was modelled in MATLAB/Simulink, showing that the RNN-based adaptive filter attains superior noise suppression, as shown by a low Root Mean Square (RMS) error. This method has been compared with the standard method. The findings validate that the recurrent network methodology provides outstanding accuracy, rendering it especially appropriate for applications where precision in noise cancellation is paramount, albeit it requires extended computational processing time.
Downloads
References
[1] Konar, Amit, Artificial intelligence and soft computing, CRC Press LLC ISBN 0-8493-1385-6,2000.
[2] A. Rubaai and R. Kotaru, “Online identification and control of a DC motor using learning adaptation of neural networks,” IEEE Trans. Industry Application, Vol. 36, 2000, pp. 935-942.
[3] Hagan, M.T., H.B. Demuth, and M.H. Beale, Neural Network Design, Boston, MA: PWS Publishing, 1996.
[4] Rosenblatt, F., Principles of Neurodynamics, Washington, D.C.: Spartan Press, 1961.
[5] Widrow, B., and S.D. Sterns, Adaptive Signal Processing, New York, Prentice-Hall, 1985.
[6] K.J. Hunt, D. Sbarbaro, R. Zbikowski and P.J. Gawthrop, “Neural networks for control systems: A survey,” Automatica, Vol. 28, No. 6, 1992, pp. 1083-1112,1992.
[7] Dennis, J.E., and R.B. Schnabel, Numerical Methods for Unconstrained Optimization and Nonlinear Equations, Englewood Cliffs, NJ: Prentice-Hall, 1983.
[8] H.B. Demuth, M. Beale, Users’ Guide for the Neural Network Toolbox for Matlab, The
Mathworks, Natica, MA, 1998.
[9] EUGEN DIACONESCU, The use of NARX Neural Networks to predict Chaotic Time Series .Electronics, Communications and Computer Science Faculty University of Pitesti Targu din Vale, Nr. 1ROMANIA .eugend@upit.ro.
[10] M. T. Hagan, O. D. Jesus, R. Schultz, Training Recurrent Networks for Filtering and Control, in (editors) L.R. Medsker, L.C. Jain, Recurrent Neural Networks – Design and Applications, CRC Press, 2001.
[11] Language of technical computing version 7.5 .0.342(R2007b).
[12] Haykin, S. Adaptive Filter Theory. 3rd ed. Englewood Cliffs, NJ: Prentice Hall, 1996.
[13] prof. Ali A. Hassan - Artificial neural networks for Identification of nonlinear industrial processes .AUFOEE.2000.
[14] Abdulla, A. H., Ahmed, A. A., & Aldeeb, A. N. (2025). Modeling and Dynamic Analysis of Pitch Motion Control of an Aircraft System. Eurasian Journal of Theoretical and Applied Sciences (EJTAS), 1-8.
[8] H.B. Demuth, M. Beale, Users’ Guide for the Neural Network Toolbox for Matlab, The
Mathworks, Natica, MA, 1998.
[9] EUGEN DIACONESCU, The use of NARX Neural Networks to predict Chaotic Time Series .Electronics, Communications and Computer Science Faculty University of Pitesti Targu din Vale, Nr. 1ROMANIA .eugend@upit.ro.
[10] M. T. Hagan, O. D. Jesus, R. Schultz, Training Recurrent Networks for Filtering and Control, in (editors) L.R. Medsker, L.C. Jain, Recurrent Neural Networks – Design and Applications, CRC Press, 2001.
[11] Language of technical computing version 7.5 .0.342(R2007b).
[12] Haykin, S. Adaptive Filter Theory. 3rd ed. Englewood Cliffs, NJ: Prentice Hall, 1996.
[13] prof. Ali A. Hassan - Artificial neural networks for Identification of nonlinear industrial processes .AUFOEE.2000.
[14] Abdulla, A. H., Ahmed, A. A., & Aldeeb, A. N. (2025). Modeling and Dynamic Analysis of Pitch Motion Control of an Aircraft System. Eurasian Journal of Theoretical and Applied Sciences (EJTAS), 1-8.
[15] D.P. Mandic, J.A. Chambers, Recurrent Neural Networks for Prediction, John Wiley & Sons, 2001.
[16] Jos_e Maria P. J_unior and Guilherme A. Long-Term Time Series Prediction with the NARX Network: An Empirical Evaluation.Barreto Department of Teleinformatics Engineering Federal University of Cear_a, Av. Mister Hull, S/N CP 6005, CEP 60455-760, Fortaleza-CE, Braziljosemenezesjr@gmail.com, guilherme@deti.ufc.br.
[18] Dorf R.C. and Bishop R.H., “Modern Control System”, AddisonWesley Logman, Inc., 1998.
[19] Hagan, M.T., and M. Menhaj, "Training feed-forward networks with the Marquardt algorithm," IEEE Transactions on Neural Networks, No. 6, 1999, pp. 989-993, 1994.
[20] prof. Mustafa Alfellany - Comparison between the non-recurrent and recurrent networks. .COET/BANI-WALID.206.
[21] Parks, T.W., and C.S. Burrus, Digital Filter Design, John Wiley & Sons, 1987, pp.226-228.
[22] Prof. Robert Morley , Digital Active Noise Cancelling Headphones By Michael Jezierny Brenton Keller Kyung Yul Lee.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
