Date of Award

2018

Publication Type

Master Thesis

Degree Name

M.Sc.

Department

Computer Science

Supervisor

Jaekel, Arunita

Supervisor

Bandyopadhyay, Subir

Rights

info:eu-repo/semantics/openAccess

Abstract

Security and attack management have become the prime concern for the network operators due to high data transfer rates and vulnerabilities associated with transparency in WDM networks. In the recent years, there is a substantial increase in perception to develop suitable mechanisms for subduing the adverse effects of malicious attacks such as high power jamming and tapping attacks.In transparent optical networks (TONs) traffic is carried over the optical fibers in the form of signals called lightpaths, creating a virtual topology over the physical interconnections of an optical fiber. This allows an exchange of an enormous amount of data at a very high speed. A fault or an attack on the network can lead to data tampering and data loss. Unlike faults, malicious attacks may not be localized and we cannot handle them with the standard fault-tolerance mechanisms in WDM networks. The Routing and Wavelength Assignment (RWA) problem assigns appropriate routes and wavelengths to all associated lightpaths in the network. Most the researchers considered the static traffic model, where the network requests (i.e. lightpaths to be established) are known in advance and last over long durations. In this thesis, we are solving the security-aware problem for dynamic requests by using protection strategy known as dedicated path protection (DPP). In the dynamic model, lightpaths are generated on-demand, and RWA must be performed based on available resources that are not being used by ongoing lightpaths. We propose an Integer linear programming (ILP) formulation to maximize requests satisfaction and reducing the disruption in the network due to malicious attacks (In-band and out-band).

Download

Share

COinS