Sleep Mode Operation of Optical Networks for Power Consumption Minimization: a Case Study of Ethio Telecom Backbone Optical Transport Network

Due to the growth in data traffic volume and diversification of applications that use telecommunication network infrastructure more power consuming telecommunication network equipments have been deployed. This scenario, although efforts are made to minimize the power consumption of the deployed equipments, has led to an increase in the power consumption of the sector. When it comes to backbone networks this increase in power consumption is more dependent on the traffic level, which is the aggregate traffic from all access segments, and on the number of subscribers on the access side. In the future it is expected that one of the bottlenecks in the development of the telecommunication sector will be its power consumption. In order to overcome this problem, stake holders in the sector are striving to come up with ways to minimize this power consumption. In this thesis, the sleep mode operational strategy which enables some components in the network to operate in sleep or low power consumption state after rerouting traffic on the remaining components, is assessed. After investigation of different algorithms proposed under this strategy, an approach which takes the Physical Layer Impairment (PLI), power loss/attenuation, together with other constraints is proposed. An evaluation of the approach is carried out on an optimization platform and a simulation environment and results analyzed taking two Ethio Telecom backbone Optical Transport Network (OTN) segments as a case study. These optimizations & simulations help to analyze impacts on Quality of Service (QoS) of applying this approach in addition to its main goal of power consumption minimization. Results reveal that up to 51% and 44% of power saving can be achieved at maximum link utilization thresholds of 70% and 50% for AA backbone and N-E backbone segments respectively. In addition, the evaluations show the trade-off between power saving and QoS using different parameters.