面向目标RCS仿真数据验证的FSV改进方法  

An Improved FSV Method for Targets' RCS Simulation Data Validation

作　　者：肖舒文[1] 李柏文[1] 陈晓盼[1] 薛非[1] 

机构地区：[1]中国国防科技信息中心,北京100142

出　　处：《计算机仿真》2016年第11期14-19,47,共7页Computer Simulation

基　　金：某军队基金重点项目(项目号:XXXXX009)

摘　　要：FSV方法是电磁仿真领域流行的数据验证技术。针对FSV方法应用于目标RCS仿真数据与专家视觉评价差异较大的问题,提出一种结合非线性变换和频域成分分解的FSV改进方法。上述方法以专家调查结果为基准,根据RCS数据特点以及专家视觉评估准则,采用S型函数和固定频域截点,减小低RCS值和背景噪声对评价结果的影响,确保数据量级、趋势和特征等专家视觉关注点的准确提取。以领域专家评估作为参照,改进方法在10组典型RCS数据验证中表现了良好的效果,评估误差均值得到降低。FSV is a widely used data validation method in computational electromagnetics modeling and simulation field. It reveals highly effectiveness in mimicking experts＇ visual evaluation on most electromagnetic-related data, however, fails when applied to targets＇ RCS simulation data. We show by comprehensive questionnaire survey that, there exist big gaps in the evaluating results between FSV and radar-related experts. Based on the results of expert investigation, this paper presents an improved FSV method, which combines the nature of visual assessment and the characteristics of RCS data. Experimental results validate the effectiveness of the proposed method. Specifically, compared with the FSV evaluation, the evaluation error of improved method is decreased.

关 键 词：计算机建模与仿真 雷达散射截面 数据验证 特征选择验证 

分 类 号：N945.12[自然科学总论—系统科学]