Lightpath Update Model in Multi-Core Fiber Optical Networks Considering Inter-Core Crosstalk

In recent years, with the advancement of 5G and the emergence of applications such as virtual reality, networks that can accommodate a large number of connection requests are expected to be realized. In order to establish such connection requests, elastic optical networks have attracted attention because of their potential to improve the efficiency of network resource utilization. On the other hand, in elastic optical networks, where new optical path requests arrive one after another or existing optical paths are deleted, it is necessary to consider updating existing lightpaths in order to cope with spectrum fragmentation. When considering lightpath update in elastic optical networks with multi-core fiber, the inter-core crosstalk generated during update is a problem. When crosstalk exceeding the crosstalk threshold is generated during lightpath updating, it causes a degradation of communication quality. In this paper, we propose the lightpath update model for elastic optical networks with multi-core fiber. In the proposed model, the inter-core crosstalk generated during lightpath update is taken into account; the lightpath update model using a two-stage renewal method is considered and formulated as an integer linear programming problem. Also, we develop a heuristic algorithm to reduce the computation time. In the performance evaluation, an approach that solves the integer linear programming problem (the optimization approach), an approach that uses the heuristic algorithm (the heuristic approach), and an approach that does not consider the inter-core crosstalk (the benchmark approach) are evaluated based on two aspects: the number of rounds to complete the update and the computation time. Evaluation results show that, compared with the benchmark approach, the two approaches in the proposed model require up to 66.7% more rounds to complete the update. The heuristic approach requires up to 2.6% more rounds to complete the update compared with the optimization approach.