Energy Efficiency in Optical Network Planning

Telecommunications operators are continuously exploring new lternatives to up-grade their networks in order to handle the ever increasing Internet traffic demand. However, increase in bandwidth have been accompanied with the increase in power consumption which is an important subject to be addressed. For the said purpose my research is focused on developing methods of bringing energy efficiency in the core telecommunications networks. The target of this thesis is the improvement of the energy efficiency in wired telecommunication networks. Indeed, several research studies have underlined that the already huge energy consumption is destined to increase in the next years since the traffic in telecommunication networks will continue to grow constantly due to the increasing popularity of new bandwidth-consuming applications, and the increase of population using them. This situation thus urges to find strategies that can mitigate the demand of energy in the networks. The main contributions concern the definition of algorithms and strategies that incorporate and exploit the knowledge of the network topology, the carried traffic and the power behavior of network devices to achieve energy efficiency in the network. In particular, during the three years, research work targeting the energy efficient designing of conventional fixed grid (Wavelength Division Multiplexing) WDM networks and Orthogonal Frequency Division Multiplexing (OFDM) based elastic optical networks is being carried out. The problem is tackled using Mixed Integer Linear Programming (MILP), greedy heuristics and sophisticated meta-heuristics. Results show that an appropriate design can increase energy efficiency in fixed-grid WDM and flexible-grid elastic optical networks. The problem of power-efficient design of IP-over-WDM networks is investigated through explicitly targeting the minimization of power consumption. Results, obtained over an extensive set of scenarios and networks, indicate that both the MILP and heuristics provide solutions for power-efficient networks. Moreover, it was observed that most of the power in an IP-over-WDM network is consumed by routers and line cards, even when high power consumption of OLAs and WDM terminals is assumed. The other topic covered during my research activity is the investigation of energy-efficient design of flexible-grid networks. The activity was mainly focused on the design of the logical layer, which is usually disregarded when dealing with flexible-grid networks. More precisely, the impact of introducing an energy-aware electronic traffic grooming in flexible-grid network design is evaluated. Two greedy heuristics for the network design are proposed, one exploiting traffic grooming, and their energy efficiency is compared. In summary, this thesis makes important contributions to the networking research community providing new methods and approaches for the definition of energy efficient networking techniques.