Adaptive Filtering Techniques using Neural Networks

Abstract

An adaptive filter automatically adjusts its own impulse response. In this paper adaptive noise canceller and adaptive signal enhancer systems are implemented using feedforward and recurrent neural networks using back propagation algorithm and real time recurrent learning algorithm respectively for training. Their performances are compared with conventional adaptive filtering techniques using LMS and RLS algorithms. The recurrent neural network employing RTRL algorithm which functions better than the other algorithms is studied further by varying the number of nodes, adding a bias to the neurons, adding a momentum term for learning and varying the momentum term and learning rate for better convergence.

Additional information

Notes on contributors

S Selvan

S Selvan, obtained his BE in Electronics and Communication Engineering in 1977 and ME in Communication Systems in 1979 both from University of Madras. He worked as a Lecturer at PSG College of Technology, Coimbatore from 1979 to 1985. Then he joined PSNA College of Engineering and Technology, Dindigul in 1985 and he is now working there as Professor of Electronics and Communication Engineering. He has submitted his PhD thesis to Madurai Kamaraj University. His areas of interest include neural networks, adaptive filters, biomedical signal processing and adaptive arrays. He has 22 publications in International and National Conferences and Journals. He is a senior member of IEEE.

R Srinivasan

Rengaramanujam Srinivasan obtained his BE in Electrical & Electronics Engineering in 1962 (university of Madras), ME Power Electrical in 1964 (llSc) and PhD in Electrical in 1970 (IIT, Kharagpur). He joined Thiagarajar College of Engineering, Madurai in 1964 as Associate Lecturer and retired as the Principal of that College in 1998. Currently he is working as the Principal of Dr MGR College of Engineering, Chennai. He has more than two dozen publications in National and International Journals. He is presently working in neural networks. The other areas of his interest are energy audit & energy conservation.