Traffic modeling, prediction, and congestion control for high-speed networks: A fuzzy AR approach

In general, high-speed network traffic is a complex, nonlinear, nonstationa ry process and is significantly affected by immeasurable parameters and var iables. Thus, a precise model of this process becomes increasingly difficul t as the complexity of the process increases. Recently, fuzzy modeling has been found to be a powerful method to effectively describe a real, complex, and unknown process with nonlinear and time-varying properties. In this st udy, a fuzzy autoregressive (fuzzy-AR) model is proposed to describe the tr affic characteristics of high-speed networks. The fuzzy-AR model approximat es a nonlinear time-variant process with a combination of several linear lo cal AR processes using a fuzzy clustering method. We propose that the use o f this fuzzy-AR model has greater potential for congestion control of packe t network traffic. The parameter estimation problem in fuzzy-AR modeling is treated by a clustering algorithm developed from actual traffic data in hi gh-speed networks. Based on adaptive AR-prediction model and queueing theor y, a simple congestion control scheme is proposed to provide an efficient t raffic management for high-speed networks. Finally, using the actual ethern et-LAN packet traffic data, several examples are given to demonstrate the v alidity of this proposed method for high-speed network traffic control.