基于局部加权回归的低信噪比外弹道数据预处理  

Preprocessing of Low-SNR External Trajectory Measurement Data based on LOESS

作　　者：骆雨桐 苏楠 马斌[1] Luo Yutong;Su Nan;Ma Bin(The 39th Research Institute of CETC,Xi'an 710065,China)

机构地区：[1]中国电子科技集团公司第39研究所,西安710065

出　　处：《测控与通信》2024年第1期41-45,共5页

摘　　要：连续波靶场测量雷达受多重复杂因素的影响,可能出现回波信噪比较低的情况,导致弹道参数测量结果包含较大的随机误差。通常采用最小二乘拟合、分段拟合等预处理算法修正弹道参数的测量误差,但在低信噪比条件下效果不佳。为解决该问题,提出了1种基于局部加权回归(LOESS)的外弹道数据预处理算法。采用四阶龙格-库塔算法解算外弹道模型,通过仿真对比了低信噪比条件下局部加权回归法与最小二乘法和分段拟合法的测量精度,并在实测数据上验证。结果表明局部加权回归法精度更高,有效提高了连续波雷达事后处理环节输入数据的准确性。The continuous wave range measurement radar may be affected by multiple complex factors,which may lead to low signal to noise ratio,resulting in large random errors in the measurement results of ballistic parameters.Preprocessing algorithms such as least square fitting and piecewise fitting are usually used to correct the measurement error,but the effect is not good under low SNR.In order to solve this problem,a new algorithm based on local weighted regression(LOESS)is proposed for external trajectory data preprocessing.The four order Runge-Kutta algorithm is used to calculate the external trajectory model,and the measurement accuracy of LOESS is compared with the least square method and piecewise fitting method under the condition of low SNR,and test on measured data.The results show that LOESS has higher precision which effectively improves the accuracy of the input data of CW radar post-processing.

关 键 词：外弹道测量 四阶龙格-库塔算法 局部加权回归 

分 类 号：TN9[电子电信—信息与通信工程]