摘要
数据融合是将多个传感器的信息加以集成,进行综合利用,其性能优于单传感器检测。寻找分布式并联融合系统的全局最优解需要求解一组耦合的非线性方程,运算量随着系统中传感器数量的增加迅速增长,传统方法很难求解。文中分析了融合的结构,重点研究了基于遗传算法的分布式贝叶斯融合系统的全局优化。采用穷举法列举所有可行融合规则,用遗传算法搜索相应规则下的最优解,实现了系统解耦。通过比较各融合规则下的最优解,得出分布式并联融合检测系统的全局最优解。仿真结果表明,该方法求解有效,建立了全局最优的贝叶斯融合检测系统。
Data from multiple sensors were synthesized in data fusion detection system, whose performances were respected better than single sensor detection＇s. To obtain the global optimizing solution of distributed Bayesian fusion system, a group of coupled nonlinear equations need to be solved. As the number of sensors grows, the computational burden grows rapidly. It can＇t be resolved by traditional methods. Structures of data fusion system were presented. Distributed Bayesian detection fusion system was formulated. To establish the global optimizing Bayesian detection fusion system, combination of exhaustively enumerating search and genetic algorithm were introduced. The simulation results prove the validity of the algorithm.
出处
《计算机仿真》
CSCD
2007年第4期183-185,共3页
Computer Simulation
关键词
数据融合
贝叶斯检测
遗传算法
分布式
Data fusion
Bayesian detection
Genetic algorithm
Distributed
