一种基于组合型模糊控制的主动队列管理算法

计算机网络具有的复杂性和动态特性使传统控制理论难以进行主动队列管理（Active Queue Management, AQM）算法的设计和分析．本文在模糊集合和模糊系统理论的基础上设计了一个主动队列管理算法CF（Combination Fuzzy control）．其中模糊控制器I根据瞬时队列的长度和变化值计算控制量；模糊控制器II根据系统负载因子计算控制增益．通过选择模糊控制器参数，模糊控制系统与使用PI（Proportional Integral）控制器的系统具有相同的局部稳定性．最后通过仿真对CF、PI和单模糊控制器的性能进行了比较．

An Active Queue Management Algorithm Based on Combination Fuzzy Control

Due to the complexity and dynamic characteristics of computer network,traditional control theory is difficult to apply to the algorithm design and analysis of AQM(Active Queue Management).In this paper,an AQM algorithm named CF(Combination Fuzzy control) is presented,which is based on fuzzy sets and fuzzy system theory.Fuzzy controller I computes the control value based on the instantaneous queue length and the derivation of queue length,and fuzzy controller II computes the control gain based on the queue load factor. Proper choice of fuzzy controller parameters endows the fuzzy control system with the same local stability as the PI control system.Simulations are made to verifiy the presented algorithm and to compare the performances of CF,PI(Proportional-Integral) and single fuzzy controllers.