一种时滞网络自适应主动队列管理算法研究

主动队列管理(AQM)算法的自适应能力和克服滞后性不良影响的能力是该文研究的重点。在分析AQM采用传统PID存在的问题的基础上，提出了一种时滞网络的自适应主动队列管理(FAGPID)算法。由模糊控制器实现PID参数对动态网络环境的在线自适应调整；成功引入灰预测算法实现反馈数据的超前预测，补偿滞后。仿真对比AQM环境中FAGPID，传统PID以及基于模糊免疫PID(FIGPID)的算法，可知FAGPID相对于FIGPID复杂度低，但FAGPID与FIGPID性能相当，均能克服滞后的影响，能快速稳定地适应动态网络环境，收敛于期望队列长度，具有较小的丢包率，优于传统PID算法。

Research on Active Queue Management of Delay Network Based on Adaptive Control

Enhancing Active Queue Management (AQM) algorithm’s self-adapting and overcoming network delay’s poor effect are research emphases. After analyzing traditional PID control algorithm’s limitation, a novel active queue management algorithm for delay network based on Fuzzy Adaptive PID control and Gray-prediction (FAGPID) is proposed, which can achieve PID parameters’ on-line self-adapting by fuzzy control under the dynamic delay network circumstances. And, a gray-prediction algorithm is successfully introduced into feedback data’s advanced prediction to compensate delay. Contrasted with traditional PID and FIGPID (Fuzzy Immue Gray-prediction PID) by simulations, FAGPID has equivalent performance to FIGPID and has better performance than traditional PID control. Both FAGPID and FIGPID can converge to queue size-setting value rapidly and stably, and get lesser packets loss rate, but FAGPID’s algorithm complexity is lower.