Date of Award

Spring 2004

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Computer Engineering - (Ph.D.)

Department

Electrical and Computer Engineering

First Advisor

Symeon Papavassiliou

Second Advisor

Edwin Hou

Third Advisor

Sirin Tekinay

Fourth Advisor

Filippos I. Vokolos

Fifth Advisor

Lev A. Zakrevski

Abstract

The explosive growth of the Internet and the continued dramatic increase for all wireless services are fueling the demand for increased capacity, data rates, support of multimedia services, and support for different Quality of Services (QoS) requirements for different classes of services. Furthermore future communication networks will be strongly characterized by heterogeneity. In order to meet the objectives of instant adaptability to the users' requirements and of interoperability and seamless operation within the heterogeneous networking environments, flexibility in terms of network and resource management will be a key design issue. The new emerging technology of mobile agent (MA) has arisen in the distributed programming field as a potential flexible way of managing resources of a distributed system, and is a challenging opportunity for delivering more flexible services and dealing with network programmability.

This dissertation mainly focuses on: a) the design of models that provide a generic framework for the evaluation and analysis of the performance and tradeoffs of the mobile agent management paradigm; b) the development of MA based resource and network management applications. First, in order to demonstrate the use and benefits of the mobile agent based management paradigm in the network and resource management process, a commercial application of a multioperator network is introduced, and the use of agents to provide the underlying framework and structure for its implementation and deployment is investigated. Then, a general analytical model and framework for the evaluation of various network management paradigms is introduced and discussed. It is also illustrated how the developed analytical framework can be used to quantitatively evaluate the performances and tradeoffs in the various computing paradigms. Furthermore, the design tradeoffs for choosing the MA based management paradigm to develop a flexible resource management scheme in wireless networks is discussed and evaluated. The integration of an advanced bandwidth reservation mechanism with a bandwidth reconfiguration based call admission control strategy is also proposed. A framework based on the technology of mobile agents, is introduced for the efficient implementation of the proposed integrated resource and QoS management, while the achievable performance of the overall proposed management scheme is evaluated via modeling and simulation. Finally the use of a distributed cooperative scheme among the mobile agents that can be applied in the future wireless networks is proposed and demonstrated, to improve the energy consumption for the routine management processes of mobile terminals, by adopting the peer-to-peer communication concept of wireless ad-hoc networks. The performance evaluation process and the corresponding numerical results demonstrate the significant system energy savings, while several design issues and tradeoffs of the proposed scheme, such as the fairness of the mobile agents involved in the management activity, are discussed and evaluated.

Share

COinS