Theses and Dissertations

Weiwei Hu

Issuing Body

Mississippi State University

Advisor

Younan, H. Nicholas

Committee Member

Fowler, E. James

Committee Member

Du, Q. Jenny

Committee Member

Abdelwahed, Sherif

Date of Degree

5-1-2010

Document Type

Dissertation - Open Access

Major

Electrical Engineering

Degree Name

Doctor of Philosophy

College

James Worth Bagley College of Engineering

Department

Department of Electrical and Computer Engineering

Abstract

Optical networks employing wavelength division multiplexing technology have been well recognized as the core networks for the next generation Internet. In such networks, any fiber cut or node failure may lead to huge data loss. Thus, reliability is of great importance in the design of modern high-speed networks. At the same time, traffic grooming is another important design objective since it addresses multi-granularity traffic. The traditional routing approaches with differentiated services do not consider the traffic grooming case or reconfiguration method. Therefore, they are not resource-efficient for the next generation Internet. In this dissertation, an effective reconfigurable traffic grooming with differentiated reliability scheme is proposed to efficiently use network resources. Compared with the conventional rerouting method, the proposed scheme makes the network more robust and immune from service interruptions. An integer linear programming (ILP) formulation is presented first. By solving the ILP formulation, an optimal solution is obtained for each incoming connection request. However, the solution is so time consuming, a heuristic algorithm is introduced to get an approximate optimal solution. The performance evaluation indicates that the connection blocking probability can be decreased greatly by the proposed scheme.

URI

https://hdl.handle.net/11668/15076

Comments

survivable traffic grooming||shared risk group||routing and wavelength assignment||wavelength division Multiplexing||differentiated reliability||integer linear programming||wavelength retuning

Download

Share

COinS