基于FLOC的ARMA SαS模型α谱估计方法  

α-spectrum estimation method for ARMA SαS process based on FLOC

作　　者：王首勇[1] 朱晓波[2] 

机构地区：[1]空军雷达学院军队重点实验室 [2]空军雷达学院研究生管理大队,湖北武汉430019

出　　处：《通信学报》2007年第7期98-103,110,共7页Journal on Communications

摘　　要：分析了基于分数低阶矩(FLOM)估计ARMA SαS模型参数的不足,根据分数低阶协方差(FLOC)的概念,提出了一种基于分数低阶协方差系数估计ARMA SαS模型参数的方法。在此基础上,给出了ARMA SαS模型的α谱估计。通过对给定ARMA SαS模型的α谱估计、α稳定分布噪声中正弦信号的估计与分辨进行仿真,详细比较了基于FLOM的ARMA SαS模型α谱估计和基于FLOC的ARMA SαS模型α谱估计的性能。结果表明,α值较小时,基于FLOC的ARMA SαS模型α谱估计的性能明显优于基于FLOM的ARMA SαS模型α谱估计。The drawback of the parameter estimation method for ARMA SαS process based on fractional lower order moments （FLOM） was analyzed, and based on the conception of fractional lower order covariance （FLOC）, a new parameter estimation method of ARMA SαS process was proposed based on FLOC coefficient. Then, the α-spectrum estimation of ARMA SotS process was presented. By estimating the α-spectrum of ARMA SetS process and the sinusoidal signals that embedded in the α-stable noise, the α-spectrum estimation performances of ARMA SetS process based on FLOM and FLOC were compared in detail. Simulation results show that, for small values of alpha, the performance of the proposed α-spectrum estimation method for ARMA SetS process based on FLOC provides better performance than the method based on FLOM.

关 键 词：Α谱 ARMA SαS模型 α稳定分布 分数低阶协方差 分数低阶矩 

分 类 号：TN911.23[电子电信—通信与信息系统]