Determining Minimum Embedding Dimension in Short Time Series using Precisely Averaged False-nearest-neighbors Approach

摘要：

A novel method is proposed to determine theminimum embedding dimension on the basis of the falsenearest nodes (FNN) and the averaged false neighbors (AFN).We take all the false nearest nodes into account and developtwo parameters G1 and G2 instead of E1 and E2 defined by AFNmethod for analyzing time series.Several short chaotic timeseries were tested.The proposed method shows more precisethan AFN method and requires less subjective parameters thanFNN method.The method is effective even if a time seriessuffers the corruption with a white Gaussian noise of 20％.Itcan deal with the experimental chaotic signals from the fiberring laser based on a semiconductor optical amplifier (SOA),while AFN method fails.

关键词： embedding dimension FNN chaos

作者: Li Zhao Nian Fang Lutang Wang Ying Wang Zhaoming Huang

作者单位: The Key Lab of Specialty Fiber Optics and Optical Access Network,SCIE,Shanghai University 149 Yanchang Road,Shanghai,200072 China

会议类型: 国际会议

会议名称: 2008 China-Japan Joint Microwave Conference(2008年中日微波会议)

会议地点: 上海

会议语种:英文

页码: 554-557

在线出版日期: 2008-09-10（万方平台首次上网日期，不代表论文的发表时间）