基于最小二乘拟合的外弹道测量数据野值剔除方法

针对莱特准则在外弹道测量数据中剔除野值的不足,提出了一种结合最小二乘拟合的外弹道测量数据野值剔除方法。将外弹道测量数据的长程变化规律看作趋势项,采用最小二乘法对趋势项进行提取,将测量数据减去趋势项,进一步利用莱特准则进行野值剔除,将剔除野值后的数据与最小二乘提取的趋势项相加,得到剔除野值后的外弹道测量数据。最后利用计算机仿真和实测数据处理证明了这一方法可在保证不丢失原信号有用信息的前提下获得较好的野值剔除结果。

Outliers Elimination Method of Ballistic Measurement Data Based on Least Square Fitting

A new method of outlier elimination of ballistic measurement data was proposed in this paper to overcome the shortage of White rule.The ballistic measurement data was viewed as the trend changes of long-term,and it was extracted by means of least square method.Then it subtracted the measured data tendency,and furthermore the White rule was adopted to eliminate the outliers.Finally the result could be obtained by summing the trend and the data after eliminating outlier.The results of computer simulation and ...