Optimal Feedback Control of ABR Traffic in ATM Networks

Bo-Kyoung Kim and Charles Thompson
University of Masschusetts - Lowell

Abstract

In this work we will examine to the problem of regulating available bit rate (ABR) traffic in an ATM network. A unique feature of ABR service is the requirement that sources be regulated using information from the distant ATM nodes. This feature is problematic for source control schemes when time-delays incurred in the feedback path and the temporal variation in the link capacity at the ATM node are considered. A source rate control algorithm that allows for the coordination of the traffic between sources sharing a congested node is presented. It is shown that by correlating the traffic rates between sources one can optimize preformance. An auto-regressive moving-average (ARMA) model will be used as the control-law for the aggregate input rate at the congested node. The optimal coefficients of the model are chosen subject to the minimization of a quadratic performance index. The results of the method are also given.

Bo-Kyoung Kim
Center for Advanced Computation and Telecommunications
University of Masschusetts - Lowell
Lowell, MA 01854
tel: (508)934-3661
bkim@cs.uml.edu

Charles Thompson
Center for Advanced Computation and Telecommunications
University of Masschusetts - Lowell
Lowell, MA 01854
tel: (508)934-3661
thompsonc@woods.uml.edu

pages by
ecewebmaster@ecs.umass.edu