Performance of nonlinear queue management algorithms in best-effort networks
| Authors: | Lehto Timo, Tampere University of Technology, Finland
Koivisto Hannu, Tampere University of Technology, Finland
Ekola Teemu, Tampere University of Technology, Finland
Laurikkala Mikko, Tampere University of Technology, Finland |
|---|
| Topic: | 3.1 Computers for Control |
|---|
| Session: | Networked Control, SW for Real-Time Control |
|---|
| Keywords: | Traffic control (communication), simulation, fuzzy control, communication system performance, Internet |
|---|
Abstract
This paper presents a system theoretic approach to, and performanceanalysis of improved congestion control methods i.e. the RED(Random Early Detection)-like AQM (Active Queue Management)algorithms. We revisit the principles of Fuzzy-RED and its linearcounterparts and examine their advantages and possibleshortcomings from a control theoretical point of view. Theperformance comparisons are based on extensive simulations. Themajor interest is on the requirement of a nonlinear control law.
