信号奇异性计算方法在电噪声信号中的应用  

Computational method of signals singularity in electrical noise

作　　者：赵健[1] 谢端[2] 肖云[1] 彭进业[1,2] 谢瑜[1] 

机构地区：[1]西北大学信息科学与技术学院,西安710069 [2]西北工业大学电子信息学院,西安710072

出　　处：《计算机应用研究》2009年第10期3824-3826,共3页Application Research of Computers

基　　金：陕西省自然科学基础研究资助项目(2006F42;2007F38);中国博士后科学基金资助项目(20060401007)

摘　　要：为了快速准确地计算电噪声奇异性,在介绍信号奇异性计算方法的基础上,将其引入到电噪声信号分析中,提出一种新的基于多重分形奇异性指数计算信号电噪声的方法。新方法利用多重分形来提取电噪声中可以表达信号内在细节特征的奇异点Hlder指数,通过计算电噪声中Hlder指数的差异来进行噪声分析。通过对实测电迁移和光耦电噪声的计算分析表明,电迁移后期奇异性指数会发生突变;而良品光耦器件和次品光耦器件在信号噪声的平均Hlder指数方面差异明显。实验结果证明本方法是一种快速可靠的电噪声奇异性计算方法。In order to computing electrical noise singularity fast and reliable, proposed a new multi-fractal based electrical noise singular computational method. Used new methods to extract muhi-fractal noise, could be expressed in the signal characteristics of the internal details of singular points Hoelder index by analyzing electrical noise in Hoelder index for the difference in noise analysis. This method of migration and optocoupler electrical noise analysis shows that migration, the late singularity index will happen and mutation will be after the opening of aluminum ; good optocoupler devices and faulty devices in the average noise Hoelder index significant difference. The experimental results show that this method in the electrical noise signals in the computation and analysis of very practical. It is a fast and reliable electrical noise singularity computational method.

关 键 词：电噪声 多重分形 奇异性 Hoeder指数 

分 类 号：TN911[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]