Efficient routing and spectrum assignment in elastic optical networks with time scheduled traffic

Elastic optical networks (EONs) employ dynamic routing and spectrum assignment (RSA) algorithms to support diverse services and heterogeneous requests. However, these RSA algorithms may possibly induce spectrum fragments when allocating spectrum to accommodate different service requests. Therefore, such induced spectrum fragments should also be regarded as spectrum consumption besides the allocated spectrum by RSA algorithms. In this paper, by additionally considering the holding times of lightpaths and service connections, we first introduce a comprehensive spectrum consumption model to simultaneously investigate both the allocated and the fragmented spectrum consumptions. Then we solve this model in both static and dynamic traffic scenarios, by either formulating the RSA problem with time-scheduled traffic or introducing a time-aware spectrum-efficient heuristics algorithm. Since no defragmentation is executed in spectrum allocation, the proposed RSA algorithm requires no traffic disruption and can be realized more easily in reality. Simulation results show that the proposed algorithm reduces the comprehensive spectrum consumption and has lower bandwidth blocking probability than the typical first-fit RSA algorithm.