Author

Mingook Kim

Date of Award

3-4-2009

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Electrical and Computer Engineering

First Advisor

Stuart Kurkowski, PhD

Abstract

Captain Nathan C. Stuckey implemented the idea of the stochastic estimation and control for network in OPNET simulator. He used extended Kalman filter to estimate packet size and packet arrival rate of network queue to regulate queue size. To validate stochastic theory, network estimator and controller is designed by OPNET model. These models validated the transient queue behavior in OPNET and work of Kalman filter by predicting the queue size and arrival rate. However, it was not enough to verify a theory by experiment. So, it needed to validate the stochastic control theory with other tools to get high validity. Our goal was to make a new model to validate Stuckey’s simulation. For this validation, NS-2 was studied and modified the Kalman filter to cooperate with MATLAB. Moreover, NS-2 model was designed to predict network characteristics of queue size with different scenarios and traffic types. Through these NS-2 models, the performance of the network state estimator and network queue controller was investigated and shown to provide high validity for Stuckey’s simulations.

AFIT Designator

AFIT-GCS-ENG-09-04

DTIC Accession Number

ADA497702

Share

COinS