A comparison of methodologies for the stationary analysis of data gathered in the simulation of telecommunication networks

Author

Francini, A. ; Neri, F.

Author_Institution

Dipartimento di Elettronica, Politecnico di Torino, Italy

fYear

1996

Firstpage

116

Lastpage

122

Abstract

The paper compares methodologies proposed in the literature for the stationary analysis of simulation data, focusing on the Monte Carlo simulation of telecommunication networks. We consider estimation rules for both the duration of the initial transient of the simulation experiment, when the simulation model reaches stationarity, and for the accuracy, in terms of confidence level and confidence interval, of the performance figures that are usually computed as suitable averages on simulation data. These rules are compared on some simple simulation models. Two rules showing robustness against different models and different model configurations are selected

Keywords

Monte Carlo methods; data analysis; digital simulation; estimation theory; queueing theory; simulation; statistical analysis; telecommunication computing; telecommunication networks; M/M/1 queue; Monte Carlo simulation; accuracy; confidence interval; confidence level; estimation rules; initial transient duration; model configurations; performance figures; robustness; simulation data; stationary analysis; telecommunication networks; Analytical models; Computational modeling; Data analysis; Performance analysis; Power system modeling; State estimation; Statistical analysis; Steady-state; Telecommunication computing; Transient analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, 1996. MASCOTS '96., Proceedings of the Fourth International Workshop on

Conference_Location

San Jose, CA

Print_ISBN

0-8186-7235-8

Type

conf

DOI

10.1109/MASCOT.1996.501003

Filename

501003