摘要
简要分析了离散动态贝叶斯网络用于目标识别时对缺失数据处理方法的研究现状,提出了把数据修补技术用于变结构离散动态贝叶斯网络。首先在对空中目标识别模型分析的基础上,构建了一种基于数据修补的离散动态贝叶斯网络模型。进而构建了修补算法的数学模型,并给出了修补过程。待缺失数据修补完整后,再运用修正后的直接推理算法计算目标类型。通过仿真结果对比,表明该方法可以有效地提高识别系统的准确性、可靠性和鲁棒性。
The current research status of how to handle the missing data on Discrete Dynamic Bayesian Networks （DDBNs） was reviewed. It was proposed to apply the data repairing technique to the variable structure DDBNs. First based on the analysis of the air target identification mode/, a kind of DDBNs model was constructed with the data repairing. Then the repairing algorithm in theory was deduced, in addition, the process of the repairing was given. After repairing the missing data, the modified direct reference algorithm was used to compute the target type. By the comparison of the simulation results, it shows that this method can effectively improve the accuracy, reliability and robustness of the system.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2010年第3期678-681,690,共5页
Journal of System Simulation
基金
国家自然科学基金(60774064)
关键词
贝叶斯估计
离散动态贝叶斯网络
数据修补
超参数
Bayesian estimation
discrete dynamic Bayesian networks
data completion
hyper oararneter
