QoS management in wireless ATM networks using large deviation principle

Author

Ahmed, C.B. ; Boudriga, N. ; Obaidat, M.S.

Author_Institution

Tunis Univ., Tunisia

Volume

1

fYear

2000

fDate

2000

Firstpage

271

Abstract

This paper presents a new QoS management scheme for wireless ATM networks using the large deviation concepts. We consider an ATM network, and assume that all switches are modeled by stochastic servers fed by sets of parallel buffers. The dynamics of such systems evolve in discrete-time using an on demand processor sharing (ODPS) discipline that we introduce. The arrival process in each buffer is an arbitrary stochastic process. Using an extension of the large deviation theory, we developed predictive techniques that allow the management of QoS in the presence of mobiles. Our approach is based on the determination of overflow probabilities. The proposed scheme has important applications for QoS management in wireless ATM networks

Keywords

asynchronous transfer mode; buffer storage; discrete time systems; land mobile radio; packet radio networks; probability; quality of service; stochastic processes; telecommunication network management; QoS management; arrival process; congestion control; discrete-time systems; large deviation principle; mobile radio; on demand processor sharing; overflow probabilities; parallel buffers; predictive techniques; stochastic process; stochastic servers; switches; wireless ATM networks; Asynchronous transfer mode; Computer network management; Intelligent networks; Optimal control; Probability; Random variables; Stochastic processes; Switches; Traffic control; Wireless networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Wireless Communications and Networking Confernce, 2000. WCNC. 2000 IEEE

Conference_Location

Chicago, IL

ISSN

1525-3511

Print_ISBN

0-7803-6596-8

Type

conf

DOI

10.1109/WCNC.2000.904641

Filename

904641