A proactive, fragmentation-aware spectrum management algorithm for routing and spectrum assignment in elastic optical networks

Abstract

Spectrum fragmentation significantly affects spectrum utilization, leading to increased blocking in elastic optical networks (EONs). Understanding fragmentation and implementing effective management strategies are essential to improve network performance and efficiency. In this paper, we propose a fragmentation avoidance routing and spectrum allocation (RSA) algorithm, referred to as the proactive, fragmentation-aware spectrum management RSA (PFA-SM-RSA) algorithm. The algorithm employs a two-stage approach: in the first stage, it applies a strict exact-fit spectrum allocation strategy to efficiently assign resources to incoming requests. If this allocation fails, the algorithm proceeds to a second stage, which performs two key tasks simultaneously: first, it intelligently allocates resources to requests, ensuring that sufficient space is reserved for incoming large requests; second, it actively prevents the generation of slot fragmentation during resource allocation, thus addressing and mitigating the spectrum fragmentation issue within the network. To achieve these tasks, we propose a novel spectrum block fitness metric, termed Quality_score. Using this metric, our proposed approach dynamically assesses the Quality_score value of each spectrum block during resource allocation and selects the one that best satisfies both tasks. By utilizing Quality_score based selection, the proposed algorithm significantly reduces fragmentation and optimizes spectrum usage, leading to improved network performance and greater spectral efficiency. Simulation results confirm that the PFA-SM-RSA algorithm consistently outperforms existing benchmark strategies in terms of blocking probability, bandwidth blocking probability, and spectrum utilization.