Exploitation des techniques de l’intelligence artificielle pour l’optimisation de la QoS et l’efficacité spectrale dans les réseaux de radio cognitive.

Abstract

The dynamic spectrum access concept addresses the spectral scarcity problem, resulting from the wireless communications services and standards proliferation, through the key technology cognitive radio. This latter is considered to be an intelligent and autonomous device capable of making decisions to adapt to its dynamic environment in real time in order to satisfy the needs of the user in terms of quality of service while respecting the regulatory framework in force. In this thesis, we propose new algorithms based on bio-inspired approaches that have been developed during this last decade, namely: the firefly algorithm, the Coucou search and the bat algorithm, for the adaptation of the transmission parameters which has been formulated as a multi-objective optimization problem. The results obtained through a series of tests and simulations demonstrate a clear superiority of our algorithms in terms of quality of solutions and speed of convergence compared to the most used techniques in the literature, namely the genetic and the particle swarm optimization algorithm.