摘要
对于α稳定分布的随机过程,由于没有有限的二阶矩,因此不能利用功率谱来进行相应的频域分析.基于分数低阶统计量,提出了新的谱分析方法,并研究了分数低阶协方差谱的几种估计方法.理论分析和仿真实验表明,分数低阶协方差谱是一种有效的α稳定分布过程的谱分析方法,三种分数低阶协方差谱估计方法中,加权交叠平均法由于采用加窗和分段使得估计方差明显变小.
α stable distribution random processes can not be conduct for frequency analysis using power spectrum due to the less of limited second moments.A new spectrum analysis method is proposed based on fractional lower order statistics,and several estimation methods of fractional lower order covariance spectrum are discussed.Theory analysis and simulation results show that the fractional lower order covariance spectrum is an effective spectrum analysis method of α-stable distribution processes,and the estimating variance of weighted overlapped segment average method is the lowest in error among the three spectrum estimation methods.
出处
《大连交通大学学报》
CAS
2010年第4期9-12,共4页
Journal of Dalian Jiaotong University
基金
辽宁省科学技术基金资助项目(20071067)
辽宁省教育厅高等学校科学研究计划资助项目(2008106)
关键词
稳定分布
谱分析
共变谱
分数低阶协方差谱
stable distribution
spectrum analysis
covariation spectrum
fractional lower order covariance spectrum
