Routing and Spectrum Allocation in Elastic Optical Networks by means of Ant Colony Optimization
The current model of Optical Networks that use Wavelength Division Multiplexing (WDM) has been shown insufficient to meet the traffic demand from services such as IPTV, video on demand, real-time video links, among others which has increased by about 30% a year. One of the most promising solutions to extend the life of current Optical Networks without the need to change the structure already installed is the Elastic Optical Networks (EON). EONs make better use of the spectrum used when performing a better segmentation of bandwidths by allocating them according to traffic needs. With this new technology, we have the Routing and Spectrum Allocation (RSA) problem, which involves choosing a route and allocating spectrum for each request. In this work, we propose the use of a routing algorithm based on Ant Colony Optimization (ACO) to solve the RSA problem, comparing it with algorithms traditionally used in optical networks. Based on the positive partial results we have obtained by the computational simulations in this work, such as the decrease of the probability of blocking, the proposed ACO algorithm can become a feasible technique as a routing algorithm in EONs. For the conclusion of the thesis, it is also planned the analysis of other metrics, such as overhead protocol and connection establishment time. In addition, we intend to include the fragmentation-aware routing and the modulation choice in the RSA problem.