导引头在线野值剔除与平滑方法研究

Studied on On-Line Outliers Elimination and Smoothing Method of Seeker

针对导引头信号野值频发的问题,提出了一种基于可变阈值方差的信号野值识别和基于牛顿插值法的野值剔除与缺失数据补偿的方法。测试对比结果表明,该方法可以同时克服孤点野值和连续野值的不利影响,保证了数据处理后的导引头信号的完整性和平滑性,满足应用需求。

Aiming at the problem of outliers identification based on variable threshold was of a seeker for KKV, a variance of seeker signals outlier proposed, and an outlier compensation algorithm based on Newton interpolation was adopted for outlier elimination and missing data compensation. The test result shows that the method can overcome the detrimental effect of isolated outliers and continuity outliers, ensure the signal of seeker is complete and smooth when data was preprocessed, and meets the requirement of engineering application.