Performance bounds on traffic predictor models for dynamic routing

Author

Soong, B.H.

Author_Institution

Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore

Volume

3

fYear

1994

fDate

14-18 Nov 1994

Firstpage

1018

Abstract

The paper studies the problem of identifying and tracking the unknown and possibly time-varying parameters λ (arrival rate) and 1/μ (mean holding time) of a traffic stream. Soong (see SICON/ICIE93 conf., Singapore, Sept. 1993) developed two main predictor models. It is the purpose of this paper to: (i) motivate several modifications to the models (concentrating mostly on M/M/n/n queue) to provide class of models within this category: (ii) provide several variants of the current recursive least squares estimators: and finally. (iii) evaluate via simulation studies, the performance of each individual estimation algorithm under different conditions

Keywords

least squares approximations; prediction theory; queueing theory; recursive estimation; telecommunication network routing; telecommunication traffic; M/M/n/n queue; arrival rate; dynamic routing; estimation algorithm; mean holding time; performance; performance bounds; recursive least squares estimators; simulation; time-varying parameters; traffic predictor models; traffic stream; Adaptive control; Filtering; Large Hadron Collider; Least squares methods; Predictive models; Reduced instruction set computing; Routing; Signal processing; Steady-state; Traffic control;

fLanguage

English

Publisher

ieee

Conference_Titel

Singapore ICCS '94. Conference Proceedings.

Print_ISBN

0-7803-2046-8

Type

conf

DOI

10.1109/ICCS.1994.474249

Filename

474249