Study on Forecasting Method of Nonlinear Signal Processing

摘要

The forecast of nonlinear time series is very important in the field of signal processing. Grey system model, time series model and neural network model are becoming the main methods in this field. In order to be used suitably, the three methods are compared and analyzed comprehensively on their brief introduction. Their merits and demerits, applying range are showed, and their sameness and difference are also analyzed. The result shows that the grey system modal can only analyze the sequence, which has the exponential law, the time series modal can only analyze the random sequence and the neural network modal can almost be used in any sequences.