Design of the adaptive noise canceler using neural network with backpropagation algorithm

Abstract

In this paper, an adaptive noise canceller using a neural network with backpropagation is designed. The adaptive noise canceller using the least mean square algorithm has the large correlativity of the reference signal and shows the limitation of the big signal to noise ratio. The system proposed in this paper plays an important role in denoising these signals. In addition, the experiments are carried out to analyze the effects of the number of hidden layers and nodes about the system. The performance of the proposed adaptive noise canceller is compared with that of the system which is used the least mean square algorithm.