基于小波分解的色噪声预测

研究色噪声的预测．将小波分析理论与神经网络建模预测基本原理相结合，提出了基于小波分解的神经网络预测方法．通过对年平均太阳黑子数典型统计模型的预测，验证了该方法的预测效果．将该预测方法用于色噪声的预测研究，通过改变对色噪声的采样速率，分析了色噪声预测的可能性和效果．研究结果表明，色噪声是可以预测的；对其预测的误差随采样率的提高而减小；基于小波分解的神经网络预测方法的预测精度优于线性神经网络预测方法．

Colored Noise Prediction Based on Wavelet Decomposition

The prediction of colored noise is studied. A method of neural network prediction based on wavelet decomposition is proposed by combining the theory of wavelet analysis with the fundamental principle of neural network modeling and prediction. It has been verified that the prediction effect of the method is sound by predicting average sunspot numbers per year, which is a typical statistical model. Then the method is used in colored noise prediction. The possibility and effect of colored noise prediction is studied through changing the sampling rate for colored noise. The research results show that colored noise can be predicted, the prediction error decreases with the increase in the sampling rate and the precision of prediction of the method of neural network prediction based on wavelet decomposition proved to be better than the precision of prediction of neural network prediction.