Load-Aware Nonlinearity Estimation for Elastic Optical Network Resource Optimization and Management

Elastic optical networks (EONs) have emerged as a promising technology to accommodate high-capacity and dynamic bandwidth demands of next-generation wireless networks. However, similar to the traditional
wavelength division multiplexing optical networks, there exists significant challenges to manage nonlinearity effects in EONs. In this paper, we first analyze state-of-the-art non linearity estimation solutions and propose a novel load- aware nonlinearity estimation method. We further present a resource allocation algorithm using the proposed nonlinearity estimation scheme. In case the new embedded light- path brings additional nonlinearity blocking the existing requests, we propose a mixed integer linear programming
model and two heuristic algorithms using the proposed nonlinearity model as the service reconfiguration scheme for efficient resource allocation in EONs. The objective of the solution is to minimize the spectrum resource usage while satisfying the bandwidth demands of the connection and ensuring the quality of transmission. The proposed solutions are evaluated using an extensive simulation with off-line traffic requests and incrementally loaded requests against the benchmark solutions for two types of trafficprofiles in a small network with six nodes and nine links and the National Science Foundation network. The resultspresented in this paper validate the benefits of the pro-posed nonlinearity estimation model and the corresponding algorithms to minimize the number of allocated frequency slots and service request blocking ratio while improving the overall network capacity.
Index Terms—Elastic optical networks; Fiber nonlinearity estimation; Mixed integer linear programming; Routing, modulation, and spectrum assignment.