Intelligent Adaptive Noise Cancellation using Cascaded Correlation Neural Networks

Abstract

A novel adaptive noise cancellation algorithm using cascaded correlation neural networks is described. In the proposed algorithm the objective is to filter out an interference component by identifying the non-linear model between a measurable noise source and the corresponding immeasurable interference. In many situations a linear model performs outstandingly. However a linear model does not perform well for situations where nonlinear phenomena occur. Hence there is a need of nonlinear filtering approach. The neural networks have been a predominant technology for intelligent control for many years. The cascaded correlation neural network algorithm has the powerful capabilities of learning and adaptation. By virtue of the learning ability, neural networks can be adapted to constantly changing environments. Two inputs, single output cascaded neural networks are used to develop the system, which eliminates the random noise, which is mixed with the test signal. Results of simulation studies using different noise sources and noise passage dynamics show that superior performance can be achieved using the proposed techniques