基于B样条小波的异常数据辨识

为解决靶场测量数据中系统误差引起的大量连续奇异数据的辨识难题,同时针对传统和目前国内外较先进的异常值识别方法的不足,提出了基于B样条小波的异常数据辨识。首先描述了异常值,然后分析了小波异常值识别原理和物理意义,构造了用于异常数据辨识的B样条小波分析方法。通过对不同情况仿真数据的计算分析,证明了该方法科学、合理,实现了测量数据异常值的辨识和重构。

Identifying the Exceptional Data Based on B-Splice Wavelet

In Order to find a solution to identify a mass of the exceptional data caused by systematic error in range measure,and,aiming at the shortage of the advanced and traditional methods adopted by the domestic and foreign researchers,this paper puts forward a method based on B-splice wavelet to identify the exceptional data. At first,it describes the exceptional data,then analyses the theory and physical meaning of the exceptional data identification based on B-splice wavelet,and constructs the B-splice wavelet used. The result of computing the simulative data in the different condition proves the method is scientific and reasonable,realizing identify and reconstruct the exceptional data.