Section: New Results

Routing

Participants : Nathalie Mitton, Julien Vandaele.

Wireless sensor and actuator/robot networks need some routing mechanisms to ensure that data travel the network to the sink with some guarantees. The FUN research group has investigated different routing paradigms.

Geographic routing has gained much attention as a basic routing primitive in wireless sensor networks due to its memory-less, scalability, efficiency and low overhead features. Greedy forwarding is the simplest geographic routing scheme, it uses the distance as a forwarding criterion. Nevertheless, it may suffer from communication holes, where no next hop candidate is closer to the destination than the node currently holding the packet. For this purpose, a void handling technique is needed to recover from the void problem and successfully deliver data packets if a path does exist between source and destination nodes. Many approaches have been reported to solve this issue at the expense of extra processing and or overhead. [19] proposes GRACO, an efficient geographic routing protocol with a novel void recovery strategy based on ant colony optimization (ACO). GRACO is able to adaptively adjust the forwarding mechanism to avoid the blocking situation and effectively deliver data packets. Compared to GFG, one of the best performing geographic routing protocols, simulation results demonstrate that GRACO can successfully find shorter routing paths with higher delivery rate, less control packet overhead and shorter end-to-end delay.

Betweenness centrality metrics usually underestimate the importance of nodes that are close to shortest paths but do not exactly fall on them. In [16], [41], we reevaluate the importance of such nodes and propose the ρ-geodesic betweenness centrality, a novel metric that assigns weights to paths (and, consequently, to nodes on these paths) according to how close they are to shortest paths. The paths that are just slightly longer than the shortest one are defined as quasi-shortest paths, and they are able to increase or to decrease the importance of a node according to how often the node falls on them. We compare the proposed metric with the traditional, distance-scaled, and random walk betweenness centralities using four network datasets with distinct characteristics. The results show that the proposed metric, besides better assessing the topological role of a node, is also able to maintain the rank position of nodes overtime compared to the other metrics; this means that network dynamics affect less our metric than others. Such a property could help avoid, for instance, the waste of resources caused when data follow only the shortest paths and reduce associated costs.

To illustrate the data routing over a real demo, in [39], we show a webcam view of the testbed with remotely controlled lighting (ceiling LEDs and a mobile robot carrying a torch). A tight grid of 256 sensors will be used to collect light information. We display live updates of the resulting heatmap, live energy profiles and other performance metrics.