基于正交小波变换合成拟1/ƒ过程

吉林大学学报(工学版) ›› 2003, Vol. ›› Issue (2): 69-74.

基于正交小波变换合成拟1/ƒ过程

曹坤勇, 于盛林, 李光

南京航空航天大学, 自动化学院, 江苏南京 210016

收稿日期:2002-10-08

To synthesize 1/ƒ process via orthonormal wavelet transform

CAO Kun-yong, YU Sheng-lin, LI Guang

College of Automation Engineering, Nanjing University of Aeronautics & Astronautics, Nanjing 210016, China

Received:2002-10-08

摘要/Abstract

摘要： 在正交小波基分析1/ƒ过程的Kahunen Loeve展开式的基础上,利用1/ƒ过程正交小波变换系数和尺度的关系,推导出一种拟1/ƒ过程的合成方法。仿真中,分析了所选用的正交小波基的消失距和数据长度对合成的拟1/ƒ过程的精确性的影响,同时说明了本合成方法的有效性。

关键词: 正交小波变换, 拟1/ƒ过程, 消失距, 滑动平均模型

Abstract: Orthonormal wavelet is used to analysis 1/ƒ process.A wavelet transform based method based on the relation between scales and variance of wavelet coefficients of 1/ƒ process to synthesize near 1/ƒ process is derived.In the numerical simulation,we changed the vanishing moments and the data length to observe their influences on the generation of 1/ƒ process.In order to prove its efficiency in precision and its predominance of synthesizing 1/ƒ process with any spectral exponent under the condition of Kahunen Loeve expression,we have compared it with the moving average model.

Key words: orthonormal wavelet, near 1/ƒ process, vanishing moments, moving average model

中图分类号:

O211

曹坤勇, 于盛林, 李光. 基于正交小波变换合成拟1/ƒ过程[J]. 吉林大学学报(工学版), 2003, (2): 69-74.

CAO Kun-yong, YU Sheng-lin, LI Guang. To synthesize 1/ƒ process via orthonormal wavelet transform[J]. 吉林大学学报(工学版), 2003, (2): 69-74.

参考文献

[1] KESHNER M S.1/ƒ noise［J］.Proc.IEEE,1982,70(2):212-218.
[2] VAN DER ZIEL A.On the noise spectra of semi-conductor noise and of fliker effect［J］.Physica,1950,16:359-372.
[3] BARTON R J,POOR V H.Signal detection in fractional gaussian noise［J］.IEEE Trans Inform.Theory,1988,34:943-959.
[4] WORNELL G W.A Kahunen-Loeve-Like expansion for 1/ƒ processes via wavelets［J］.IEEE Transaction on Information Theory,1990,36(4):859-861.
[5] WORNELL G W.Wavelet-based representations for the 1/ƒ family of fractal processes［J］.Proceedings of the IEEE,1993,81:1428-1450.
[6] SHUSTERMAN Eli,FEDER Meir.Analysis and synthesis of 1/ƒ processes via shannon wavelets［J］.IEEE Transaction on Signal Processing,1998,46(6):1698-1702.
[7] ANTAL T,DROZ M,GYORGYI G,et al.1/ƒ noise and extreme value statistics［J］.Physical Review Leters,2001,87(24):240601.
[8] WALDEN Andrew T.Wavelet analysis of discrete time series［Z］.http∥dmawww.epfl.ch/～choulat/3e.cycle.romand/percival.pdf,2001.
[9] ALIEVA Tatiana,BARBE Andre.Macroscopic 1/ƒ behavior of fractal signals generated by additive substitution［J］.IEEE Transaction on Information Theory,2000,46(6):2261-2268.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed