摘要
对小波变换理论进行了分析,在利用该理论剔除野值的基础上,提出了一种结合3σ准则的新方法。利用小波变换将信号分解为近似分量和细节分量,通过3σ准则对细节分量进行野值判别,并确定野值剔除的标准,然后重构原信号。结合3σ准则的野值剔除方法能够保证在不损失原信号的前提下,实现野值剔除。在导弹试验中对遥测数据处理证明了该方法的有效性。
This paper analyzes the theory of wavelet transform. Based on this theory, it brings forward a new method of outliers elimimation with 3σ-rule. By the method of wavelet transform, the signal is disassembled with two parts, one is approximate components and another is detailed components. By applying the 3σ-rule, the outliers are eliminated in detailed components and the signal is reeonstruted. The new method can eliminate outliers in the signal without losing original information. The process of telemetry data in missile test proves the validity of this method.
出处
《航空兵器》
2008年第5期45-47,共3页
Aero Weaponry
基金
海军水下攻关重点科研课题
关键词
小波变换
3σ准则
野值
遥测数据
wavelet transform
3σ-rule
outlier
telemetry data
