基于模糊逻辑的ABR显式流量控制算法

面向连接的ATM网络通过一系列的流量管理机制为各种应用提供纯粹的服务质量(QoS)保证，其中ABR业务流量控制的作用尤显重要．显式速率(Explicit Rate,ER)控制是一种有效的机制，有效性、公平性和算法复杂度是各种显式速率控制算法所面临的主要问题．该文将ABR业务流基于速率的端到端闭环反馈控制抽象为控制理论中典型的调节系统，运用不依赖对象模型的模糊控制理论设计了一种新的流量控制算法——模糊速率控制算法(FRCA)．仿真试验表明FRCA的综合性能良好：保证了最大—最小公平性的实现；在LAN和WAN上均有较好的适应能力；算法的复杂度低；队列长度的控制能力强；易于硬件实现．

ABR Explicit Rate Control Algorithm Based on Fuzzy Logical Control

Connection oriented ATM network provide pure QoS(Quality of Service)for diversified services through a series of traffic management mechanism.The ABR flow control is especially important among these approaches. Explicit rate control is an effective mechanism. Among the existing algorithms, the effectiveness, fairness and complexity are the main unsolved problems. Therefor the end-to-end closed-loop feedback rate control used in ABR flows is modeled as a typical regulation system in control theory, and then a novel flow control algorithm based on fuzzy logic control, named fuzzy rate control algorithm (FRCA), is designed, which is independent of the model of the controlled plant. The simulation results show that the FRCA algorithm is able to fairly and effectively allocate the available bandwidth among all active ABR connections; it is applicable to all network configurations, including LAN and WAN; the FRCA algorithm is very powerful in adjusting the queue length to the reference value, and is uncomplicated to be implemented in hardware.