基于本征模函数能量的外测数据误差分析方法

受测量环境、测量手段和弹道特性影响，外弹道测量数据误差复杂，存在随机误差具有不可观测性、强相关性和非平稳时变特性，以及系统误差具有潜伏性和不易识别的问题。为了有效估计随机误差特性，准确修正系统误差影响，提出了一种基于本征模函数(IMF)能量拐点的外测数据误差分析方法。根据外测数据随机误差、系统误差和真实数据的频率特征，采用IMF能量拐点方法将分解得到的IMF分成高频随机误差、混合信息和有效信息共3个集合。将高频随机误差集合直接去掉，有效信息集合保留，混合信息集合采用改进的阈值函数进行小波滤噪，重构滤噪后得到外测有效数据。经数据验证，该方法可合理补偿系统误差值，与真实弹道测量值差别最小，提高了定位精度。

Error Analysis Method of External Measurement Data Based on Energy of IMF

Due to the influence of measurement environment,measurement means and ballistic characteristics,the error of external ballistic measurement data is complex.There are problems that random error has non observability,strong correlation and non-stationary time-varying characteristics,and system error is latent and difficult to identify.In order to effectively estimate the characteristics of random error and accurately deduct the influence of systematical error,an external measurement data analysis method based on energy inflection point of intrinsic mode function (IMF) is proposed.According to the frequency characteristics of random error,systematical error and real data,the IMF energy inflection point method is used to divide the decomposed IMF into three sets,including high-frequency random error,mixed information and effective information.The high-frequency random error set is directly removed,and the effective information set is retained.The mixed information set adopts the improved threshold function for wavelet noise filtering,and the filtered data is reconstructed to obtain the external effective data.The data verify that the method can reasonably compensate the systematical error, with the minimum difference from the real ballistic measurement value,which improves the positioning accuracy.