Mohammad Hossein Yaghmaee - Azad University of Quachan
Farzaneh Fatemipour - Azad University Mashhad, Iran
Elham Maleki - Azad University Mashhad, Iran

Network congestion control is a hi gh priority and critical issue in today S networks. End-toend congestion control mechanisms such as those in TCP are not enough to prevent congestion and they must be supplemented by control mechanisms inside the network. A number of Active Queue Management (AQM) algorithms such as RED, BLUE and GREEN are presented for this purpose. In this paper, we will present a new Active Queue Management scheme to provide congestion control in TCP/IP best ejfort nebvorks by using a ji~zzy logic approach, This approach computes the packet drop probability according to preconjigured fuzzy system using the instantaneous queue length and number of packets dropped in a period of time as input variables. Our goal is to improve system performance and resource zltilization in congested networks by a simple method. The proposed approach is simple and it is no/ depended on any other existing AQM algorithms Simulation resz~ltss how that proposed Fzazy AQM can eflectively control the queue length around the desired value. Compared to existing AQM algorithms such as RED, BLUE and GREEN, Fuzzy AQM obtains higher resource utilization and lower packet loss rate.

Active Queue Management (AQM), Fuzzy Logic Controller, RED, BLUE, GREEN