Autonomic Self Tunable Proactive Routing in Mobile Ad Hoc Networks

Proactive routing in MANETs induces high signaling overhead. Increasing the routing period size, while it reduces such an overhead, prevents to correctly track frequent changes in the topology and impacts the validity of routing as time goes farther from the start of the routing period. Routes' validity plays a central leveraging mission to enhance network performances as forwarding through incorrect routes not only results in traffic wondering inside the network without ever being able to be delivered to their ultimate destinations but also over consumes valuable network resources. In this paper, we propose an autonomic self tuning approach to dynamically gauge the size of the routing period in a way to properly calibrate between the amount of signaling overhead and the routing validity to yield better performances. First, we propose a distributed algorithm to collect the network cartography. We then study the validity of this cartography as a function of time and mobility. The validity of the cartography is then used to dynamically and locally self regulate the routing period size in a way to calibrate the signaling overhead and the routing pertinence. Simulation results show that our proposed scheme not only is capable of correctly tracking changes in network dynamics but also outperforms conventional proactive algorithms by doubling the network throughput at moderate to high workloads.