基于模糊神经网络的拥塞控制算法研究

本文介绍一种基于模糊神经网络的主动队列管理(AQM)算法,实现网络拥塞控制。利用神经网络来实现模糊推理,可自适应修正隶属函数的参数和加权系数,优化模糊逻辑控制器,从而达到某种性能指标的最优化。仿真结果表明,采用模糊神经网络进行流量速率预测的拥塞控制策略能够使缓冲器队列长度快速收敛到目标值,并且维持小的队列震荡。结果也表明该方法与传统的PD控制器相比具有更好的性能和鲁棒性。

Study on Congestion Control Algorithm Based on Fuzzy-neural Network

This paper presents an active queue management(AQM)scheme to provide congestion control using the combination of neural network and fuzzy logic cont rol approach.This proposed method can achieve fuzzy logic reasoning and modify in adaptive mode the membership function parameters and weights by using the neu ral network,as a result,the fuzzy logic controller can be optimized.The simul ation results demonstrate that it can lead to the convergence of the queue lengt h to the desired value quickly and maintain the oscillation small.The results a lso show that the scheme is very robust to disturbance under various network con ditions and outperforms the traditional PD controllers significantly.