Lightpath routing for disaster survivability in optical networks

Abstract

Optical network serves as a core network with huge capacity and a multitude of high-speed data transmission. Natural disasters and physical attacks showed significant impacts on the optical networks such as damages the network nodes and optical links. This thesis aims to investigate and develop algorithms for the provisioning of risk-averse lightpaths to combat disastrous events or intentional attacks. Generally, network survivability is obtained by computing the backup path such that the nodes and the lightpaths are disjoint without considering how optical fiber cables are deployed within the physical plane. In contrast to many previous works, this research work has considered lightpaths, established over the fiber cables, as a series of line segments and not just a single line segment because real-world fiber paths are not always laid out as direct paths between cities or countries or even across the oceans. In this work, two novel disaster-resilient heuristic algorithms are proposed. First algorithm finds a pair of lightpaths with a maximum value of minimum spatial distance in order to enhance network survivability against spatial-based concurrent fiber failures, while second algorithm finds a pair of lightpaths in which length of primary lightpath is minimized but constrained by minimum spatial distance. Capacity exhaustion problem in post-disaster scenario is also addressed as a reactive compensation. In this regard, another novel congestion-aware lightpath routing algorithm is developed to tackle the provisioning and restoration of disrupted lightpaths in a post-disaster scenario. Selection of alternative lightpath is based on a criteria parameter for a lightpath to be least loaded and constrained by either the length or the spatial distance between primary and alternative lightpaths. The spatial distance between lightpaths enables to re-establish the disrupted connection request away from disaster proximity. Extensive simulations are performed to evaluate our proposed algorithms for several parameters like blocking probability, network resource utilization, connection success rate and minimum spatial distance, and compared with existing techniques proposed in the literature. Simulation results of proposed algorithms show an improvement through 50% reduced computation time by lowering blocking probabilities of lightpaths up to 10% and 3% to 21% enhanced capacity utilization. Moreover, 100% connection success rate is achieved for modest network load.