Development of dynamic service delivery model in wavelength routed optical networks using GMPLS framework

Wavelength routed optical networks has been claimed as the main technology to underpin the future telecommunication networks and therefore it has garnered widespread interests from industrial and academic researchers alike to investigate key issues that affect its implementation and deployment. The main challenge within this domain is primarily to provide a dynamic service delivery technique to match the growing demand in network capacity due to the increasing popularity of end-users applications which include the omnipresent internet connectivity, video on-demand, online gaming and multimedia services. The proliferation of next generation optical networks also known as the all optical networks has paved the way for the study in this thesis. Network functions previously performed in the electrical domain have now gradually migrated to the optical domain, thus giving a huge upgrade in terms of
the network capacity and capability. While a number of works has focused on the developing network models to measure the qualitative and quantitative performances of such networks, the combination of routing algorithms, signaling and wavelength assignment schemes employed proves to have substantial effects. Thus, this research attempts to address the issue of providing dynamic service delivery for users in the optical domain from modeling perspective. A new optical network model is developed using the GMPLS based framework for the control plane of the overall network operation. The key feature in this model is the lightpath provisioning process for accommodating service request in dynamic fashion. The performance of this model is then investigated by means of simulations using OMNeT++ simulation platform. Experiments through simulations for this research have produced results showing that the network model gives consistent performance in terms of the network behaviour. Furthermore, results obtained from the simulation are also used to validate and verify the accuracy of the developed network model in this study. In addition to that, this thesis proposes a benchmarking for delivering such on-demand requests for future network services especially on the correlation between the desired network utilisation and its baseline network setup and resources.