Adaptive Control of Chaos in a Congestion Control Model
| Authors: | Jiang Kai, Shanghai Jiao Tong University, China
Wang Xiao Fan, Shanghai Jiao Tong University, China
Xi Yugeng, Shanghai Jiao Tong University, China
Li Xiang, Shanghai Jiao Tong University, China |
|---|
| Topic: | 7.4 Transportation Systems |
|---|
| Session: | Transportation Systems II |
|---|
| Keywords: | Congestion Control, Random Early Detection, Adaptive Control, Chaos Control |
|---|
Abstract
Random Early Detection (RED) is the most promising Active QueueManagement (AQM) mechanism. Its interaction with Transmission Control Pro-tocol (TCP) can be modeled as a discrete-time dynamical system, which mayexhibit complex bifurcation and chaos behavior. An improved RED algorithm isproposed by using a real-time, adaptive, model-independent (RTAMI) technique.Bifurcation analysis and numerical simulation results show that the new algorithmcan control the bifurcation and chaotic dynamic of a TCP-RED map. The situationof the presence of UDP tra±c is also studied.
