Implementation of traffic engineering in NGNs using hybrid genetic algorithms

Traffic engineering, particularly routing optimization, is one of the most important aspects to take into account when providing QoS in next generation networks (NGN). The problem of weight setting with conventional link state routing protocols for routing optimization has been object of study by a few authors. To solve this problem for big networks artificial intelligence heuristics have been used, in concrete genetic algorithms (GA). Some of the proposals incorporate local search procedures in order to optimize the GA results, in the so-called hybrid genetic algorithm (HGA) or memetic algorithm. This paper presents an inedited comparative analysis of the main hybrid genetic algorithms (HGA) proposals, as well as comparing them with other algorithms for the same problem by means of simulations. One of the HGA algorithms was chosen from the results analysis and was implemented over a real testbed with commercial routers with successful OSPFv3 routing optimization.