Quality of transmission assured joint RMLSA for elastic optical networks

The routing and spectrum assignment (RSA) presents a significant challenge in the elastic optical networks (EONs). Integrating adaptive modulation formats into the RSA problem, i.e. routing, modulation level and spectrum assignment (RMLSA), increases allocation options heightening complexity. The conventional RSA approach involves pre-determining a fixed number of paths and subsequently allocating spectrum within them separately. However, expansion of path set for optimality may not be advisable due to the substantial increase in paths with network size expansion. This paper explores a novel RMLSA, proposing a comprehensive solution addressing route determination and spectrum assignment concurrently. Finding an optimal solution for dynamic traffic is not feasible due to the unknown future connections. However, allocating resources with better utilization and lower average fragmentation enhances the likelihood of successfully accommodating future connections. So, an objective function has been chosen and designated as ABACUS, adaptive balance of average clustering and utilization of spectrum. This nomenclature highlights the objective function’s capability to adjust and assign significance to ”average clustering” (lower fragmentation) and ”spectrum utilization”. Our approach involves formulating an integer linear programming (ILP) model with a simple relationship between path and spectrum constraints. The model also integrates physical layer impairments (PLIs) to guarantee end-to-end quality of transmission (QoT) for requested connections while upholding existing ones. Towards this goal, we adopted a structured formulation approach where essential information is determined beforehand, minimizing the need for online computations. The simulation results indicate that this approach reduces the spectrum resource utilization while improving network’s average fragmentation which contributed to a lower bandwidth blocking probability across all arrival rates.