Spectral Defragmentation And Dynamic Routing Algorithm In Flexible Raster Optical Networks

With the rapid development of broadband services, cloud computing and data center, optical network is under immense pressure bought by the explosive growth of data traffic. Large scale and dynamic are the requirement of future optical network. Currently, the wavelength-division multiplexing (WDM) technology has significant drawbacks due to its inflexible grid and coarse bandwidth granularity. Meanwhile, the flexgrid optical networks are attracting high interest recently due to their higher spectrum efficiency and flexibility compared to traditional fixed grid WDM networks. In flexgrid optical networks, lightpath with flexible bandwidth and adaptive frequency spacing can be constructed. Since the transition to EON will be a significant leap forward, the conception also brings challenges at the network level, such as complicated routing and spectrum allocation problem, spectrum fragmentation and network resource optimization. To solve these problems, with the support of National973project, this paper mainly focuses on the key technologies in flexible grid optical networks, especially on the problem of resource allocation and optimization. The main innovative contributions are listed as follows:Firstly, to resolve the problems of spectrum fragmentation, we propose two new concepts called ’Sepctrum Fusion Degree (SFD)’and ’Fragment Fusion Degree (FFD)’to assess the level of spectrum fragmentation in flexible grid optical networks. They are calculated based on the availability of the spectrum fragments among neighboring links. Moreover, different defragmentation algorithms are porposed to optimize the resource utilization in different network scenarios. Under dynamic traffic, the network adaptively reroute specific lightpaths to enable the fusion of the available spectrum fragments. Moreover, we propose a genetic algorithm for solving the spectrum fragmentation problem with the objective of compacting occupation of the spectrum in flexgrid optical networks. The SFD and FFD are introduced as the fitness functions to conduct the evolution in genetic algorithm. As a result, the genetic algorithm provides a lightpath reconfiguration map, which identifies the candidate lightpaths to be reallocated, their reconfiguration sequence and new locations. The proposed algorithms are compared with commonly used approach and the results demonstrate the ability of the proposed algorithm to efficiently solve the problem of spectrum defragmentation in flexgrid optical networks.Secondly, to describe the spectrum utilization among links, we introduce the notions of spectrum allocation window (SAW), spectrum allocation point (SAP) and spectrum fusion array (SFA) in the following sections. Two dynamic fragmentation-aware RSA algorithms are proposed based on the proposed notions:spectrum fusion oriented routing and spectrum allocation (SF-RSA) algorithm and maximizing spectrum fusion routing and spectrum allocation (MSF-RSA) algorithm. The aim of these algorithms is to minimize the blocking probability by means of maximizing available spectrum consecutiveness among links. The simulation results show that they can reduce the average block probability and the network performance can be improved significantly by them.Thirdly, we propose different heuristic algorithms for time varying routing and spectrum allocation algorithm (TV-RSA) considering different network and traffic conditions. An iterative heuristic called BFSR-LB (Best Fit Shortest Routes&Load Balance RSA algorithm) for the shortest path routing in static traffic is proposed. The results suggest that it is feasible to choose routes, among the shortest paths, minimizing the number of frequency slots. For dynamic scenarios, we propose Maximum Shared Spectrum algorithm. By timely adjusting the number of occupied spectrum and the spctrum location of the connetion, more spectrum resource can be shared for the fine granularity and short duration traffic.