A Method for Detecting Impulsive Noises in Chaotic Time Series (Special Section on Nonlinear Theory and its Applications)

概要

論文の詳細を見る
A method is presented for detecting impulsive noises in chaotic time series, based on a new nonlinear prediction algorithm. A multi-dimensional trajectory is reconstructed from a time series using delay coordinates. The future value of a point on the trajectory is predicted using a local approximation technique revised by adding the Biweight estimation method and then the prediction error is calculated. Impulsive noises are detected by examining the prediction errors for all points on the trajectory. The proposed method is applied to the time series of the pupil area and the refractive power of the lens in the human eye. The Lyapunov exponent analysis for these time series is conducted. As a result, it is shown that the proposed method is effective in detecting impulsive noises caused by blinking in these time series.
社団法人電子情報通信学会の論文
1996-10-25