Multiobjective Genetic Algorithms for Solving the Impairment-Aware Routing and Wavelength Assignment Problem

In future transparent (all-optical) WDM networks, the signal quality of transmission (QoT) will degrade due to physical layer impairments. In this paper, we propose two genetic algorithms for solving the static impairment-aware RWA (IA-RWA) problem by accounting for the impact of physical impairments in the optimization process when searching for the optimum routing path and wavelength channel. The first algorithm indirectly considers the physical impairments through the insertion of the path length and the number of common hops in the search process, using classical multiobjective optimization (MOO) strategies. The second algorithm is a single-objective genetic algorithm (GA) that uses the Q factor for the evaluation of the feasibility of the selected RWA solution. The Q factor is used in each iteration of the algorithm in a self-learning mode in order to evaluate the fitness of each solution to the RWA problem and trigger the evolution of the population. Performance results have shown that considering path length and number of common hops for indirectly handling impairments provide an efficient solution to the IA-RWA problem.