Non-linear prediction using a three-layer neural network

A nonlinear predictor is investigated for a differential pulse code modulation (DPCM) encoder using artificial neural networks (ANNs). The predictor is based on a three-layer perceptron with three input nodes, 30 hidden nodes, and one output node. The backpropagation learning algorithm was used for the training of the network. Simulation results are presented to evaluate and compare the performance of the proposed neural-net-based nonlinear predictor with that of an optimized linear predictor. Success in the use of the nonlinear predictor is demonstrated through the reduction in the entropy of the differential error signal as compared to that of a linear predictor. It is shown that the ANN predictor is more robust for encoding noisy images compared to the linear predictor