基于非参数回归方法的城市电磁干扰消除研究

Research on Eliminate Urban Electromagnetic Interference in Non-parametric Regression Methods

随着近年来城市交通运输的发展,高铁、地铁、轻轨等轨道交通线路不断新建运营,对城市的电磁环境产生了一定的影响,为尝试解决城市电磁干扰的消除问题,本文选用了两种基于加权线性最小二乘和n阶多项式模型的局部回归方法与它们的稳健形式,通过调节窗宽系数,去除噪声信号,并通过与未受干扰的数据比对,计算分析相关系数和欧氏平方距等数据指标,使之尽量接近于未受干扰时的数据形态,从而达到有效消除电磁干扰的结果。

With the development of urban transport in recent years, the high-speed rail, metro and rail lines continued expansion, the urban electromagnetic environment have been suffered a certain impact. For trying to solve this problem, this paper based on two locally weighted regression scatter plot smoother method by n-order polynomial model and their robust method, through adjusting the width of the window, remove the noise signal, compare the undisturbed data, analysis of the calculated correlation coefficient and the squared Euclidean distance and other data indicators, to be as close as possible data shape which is undisturbed, obtained effective elimination results。