无线信道统计模型识别  

Statistical model recognition of radio channel

作　　者：曾明如[1] 刘公法[1] 宗胡亮[1] 刘亮[1] 

机构地区：[1]南昌大学信息工程学院,江西南昌330031

出　　处：《南昌大学学报（理科版）》2012年第6期577-580,共4页Journal of Nanchang University(Natural Science)

基　　金：国家自然科学基金资助项目(50747025)

摘　　要：为准确地找到最符合快衰落特性的统计分布模型,论文基于概率论和信息论设计了Kullback-Leibler距离及信息准则表达式,对几种主要的统计分布模型进行了仿真和实验,得到了各统计分布模型的正确识别率,通过统计的方式对比了各种识别模型,发现基于直方图的KL方法识别模型的可靠性比其他方法要高,而韦伯衰落模型更加适合用来对无线信道进行建模。To find the model which best coincides with the fast fading channel among the different statistical laws,The paper design the Kullback--Leibler distance and information criterion expression based on the probability theory and information theory, obtain the correct recognition rate for each candidate distribution model repeat by simulation and experiment, compare and analyze the performances of ours identification methods by statistical method. The experiments show the KL method based on Histogram than other methods to high reliability,that Webuill fading model is more suitable for modeling of the wireless channel in the environment we researched.

关 键 词：无线信道估计 模型识别 信息准则 动态直方图 混合模型 

分 类 号：TN925[电子电信—通信与信息系统]