ATR系统评价中的因素作用测算方法及应用  

A Factor Effect Measuring Method and Its Application in ATR System Evaluation

作　　者：何峻[1] 肖立[1] 刘峥[1] 付强[1] 

机构地区：[1]国防科技大学ATR实验室,湖南长沙410073

出　　处：《运筹与管理》2010年第2期56-62,共7页Operations Research and Management Science

基　　金：装备预先研究项目"××目标自动检测识别技术"(51301050102)

摘　　要：针对自动目标识别(Automatic Target Recognition,ATR)系统评价中存在的性能建模手段难以有效分析因素变化对系统性能作用的情况,提出了一种新的因素作用测算方法。该方法根据工作条件下ATR实验结果所具有的面板数据特点,基于Malmquist指数对因素变化进行定量度量,无需对ATR系统的代价进行严格限定。求解过程中采用了非参数的DEA方法计算距离函数,避免了对ATR性能随因素变化趋势的模型假设,并通过引入AHP约束锥实现了多指标情况下的权重限制。最后,通过一个应用实例说明如何运用该评价方法实现因素作用的测算与分析。To investigate the effect of factors in automatic target recognition （ATR） system evaluation where mod- eling methods are often impracticable, a new factor effect measuring method is proposed. According to the paneldata character of ATR testing results in operation condition, this new method uses Malmquist index to measure the factor effect without the necessity of holding the cost of ATR system constantly. The distance functions are cacluated by a nonparametric DEA approach so as to avoid supposing the models of the ATR system performance and the factors, and the multi-attributes weighting are achieved through an AHP restraint cone. At last, an application example is given to illustrate how to use the proposed method to quantify and analyze the factor effect in extended operation condition.

关 键 词：系统工程 评价 MALMQUIST指数 自动目标识别 数据包络分析 

分 类 号：N945.16[自然科学总论—系统科学] TN955[电子电信—信号与信息处理]