A Bio‐Inspired Quality of Service (QoS) Routing Algorithm

A Mellouk (1) , S. Hoceini (1) , S. Zeadally

1 CIR

We propose in this letter a bio-inspired Quality of Service (QoS) routing algorithm that is based on the trial/error paradigm combined with a continuous adaptive function to optimize three QoS criteria: static cumulative cost path and dynamic end-to-end delay and residual bandwidth. Our proposed approach uses a model that combines a stochastic planned pre-navigation for the exploration phase and a deterministic approach for the backward phase. We adopt a unified framework of online learning to develop a cost function. We evaluated the performance of our QoS-routing algorithm and the simulation results demonstrate substantial performance improvements for networks with dynamically changing traffic.

Mots clés: Quality of service routing; bio-inspired reinforcement learning; optimization; performance evaluation

Domaines: Réseaux et télécommunications [cs.NI]

A Mellouk, S. Hoceini, S. Zeadally. A Bio‐Inspired Quality of Service (QoS) Routing Algorithm. IEEE Communications Letters, 2011, 15 (9), pp.1016-1018. ⟨hal-01678452⟩