Comparison of Multilayer Perceptron and Generalized Regression Neural Networks in Active Noise Control

Passive methods such as silencers and isolation are large, costly and ineffective at low frequencies. Active cancellation of noise was presented because of these problems. In this paper, performance of multilayer perceptron (MLP) and generalized regression neural networks (GRNN) is evaluated in active cancellation of sound noise. The performance of these networks is compared for ANC. In order to compare the networks, training and test samples are similar. Noise signals from a SPIB database are used for simulation procedures. Simulation results show that MLP neural network is more effective in canceling sound noise than GRNN.