QColony: A multi-pheromone best-fit QoS routing algorithm as an alternative to shortest-path routing algorithms

This paper introduces a new generic routing algorithm called QColony for packet-switched communications networks that support real-time flows. In this context, we introduce both a novel path-selection scheme, namely the best-fit scheme, and a novel routing technique, namely the multi-pheromone technique, inspired by observations of biological ant colonies. Simulation was carried out for our algorithm and two other routing algorithms under various traffic scenarios and different irregular network topologies. Simulation results show that we can achieve good performance for the QColony algorithm. According to our experiments, QColony is able to provide smooth performance for all operating conditions, especially with large networks and under traffic scenarios with failure conditions and improperly functioning nodes. We also demonstrate through simulation that the best-fit scheme is able to deliver higher routing performance than what the shortest-path scheme can achieve.