Section: New Results

Routing in FUN

Participants : Thierry Delot, Tony Ducrocq, Nicolas Gouvy, Nathalie Mitton, Enrico Natalizio, David Simplot-Ryl, Tahiry Razafindralambo, Dimitrios Zormpas.

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 geographic routing paradigms. Georouting assumes that every node is aware of its location, the one of its neighbors and of the destination(s).

In this context, we first propose the first k-anycating georouting protocol, ie in which a node wishes to send a message to $k$ sinks in the network [13] . Then, we tried to relax some of the assumptions. For instance in [12] , we introduce HECTOR which is the first position based routing protocol which relies on virtual positions, is energy-aware and guarantees the data delivery.

In [46] , [21] , we assume that only a part of nodes is aware of its position and proposes a hybrid approach between position-based greedy approach and traditional on-demand routing. Indeed, geographic routing protocols show good properties for WSNs. They are stateless, local and scalable. However they require that each node of the network is aware of its own position. While it may be possible to equip each node with GPS receiver, even if it is costly, there are some issues and receiving a usable GPS signal may be difficult in some situations. For these reasons, we propose a geographic routing algorithm, called HGA, able to take advantages of position informations of nodes when available but also able to continue the routing in a more traditional way if position information is not available. We show with simulations that our algorithm offers an alternative solution to classical routing algorithm (non-geographic) and offers better performances for network with a density above 25 and more than 5% of nodes are aware of their position. [46] analyses the impact of nodes topology on network performances. We show that different topologies can lead to a difference of up to 25% on delivery ratio and average route length and more than 100% on overall cost of transmissions.

In [24] , [25] , [26] , [3] , we consider that nodes are able to move by themselves and we try to take advantage of this feature to improve the network performance. In sensor networks, there is often more than one sensor which reports an event to the sink in WSN. In existing solutions, this leads to oscillation of nodes which belong to different routes and their premature death. Experiments show that the need of a routing path merge solution is high. As a response, [24] , [25] introduce the first routing protocol which locates and uses paths crossing to adapt the topology to the network traffic in a fully localized way while still optimizing energy efficiency. Furthermore the protocol makes the intersection to move away from the destination, getting closer to the sources, allowing higher data aggregation and energy saving. Our approach outperforms existing solutions and extends network lifetime up to 37%.

Using nodes location, position-based routing protocols generally apply a greedy routing that makes a sensor forward data to route to one of its neighbors in the forwarding direction of the destination. If this greedy step fails, the routing protocol triggers a recovery mechanism. Such recovery mechanisms are mainly based on graph planarization and face traversal or on a tree construction. Nevertheless real-world network planarization is very difficult due to the dynamic nature of wireless links and trees are not so robust in such dynamic environments. Recovery steps generally provoke huge energy overhead with possibly long inefficient paths. In [26] , we propose to take advantage of the introduction of controlled mobility to reduce the triggering of a recovery process. We propose Greedy Routing Recovery (GRR) routing protocol. GRR enhances greedy routing energy efficiency as it adapts network topology to the network activity. Furthermore GRR uses controlled mobility to relocate nodes in order to restore greedy and reduce energy consuming recovery step triggering. Simulations demonstrate that GRR successfully bypasses topology holes in more than 72% of network topologies avoid- ing calling to expensive recovery steps and reducing energy consumption while preserving network connectivity.

[31] relaxes the assumption that nodes are aware of their neighbors and considers that dynamic energy sources could be available. It introduces MEGAN (Mobility assisted Energy efficient Georouting in energy harvesting Actuator and sensor Networks), a beacon-less protocol that uses controlled mobility, and takes account of the energy consumption and the energy harvesting to select next hop. MEGAN aims at prolonging the overall network lifetime rather than reducing the energy consumption over a single path. When node s needs to send a message to the sink d, it first computes the ideal position of the forwarder node based on available and needed energy, and then broadcasts this data. Every node within the transmission range of s in the forward direction toward d will start a backoff timer. The backoff time is based on its available energy and on its distance from the ideal position. The first node whose backoff timer goes o is the forwarder node. This node informs its neighbor- hood and then moves toward the ideal position. If, on its route, it finds a good spot for energy harvesting, it will actually stop its movement and forward the original message by using MEGAN, which will run on all the intermediate nodes until the destination is reached. Simulations show that MEGAN reduces energy consumption up to 50% compared to algorithms where mobility and harvesting capabilities are not exploited.

Additionally, according to a wide range of studies, (Informatics Technologies) IT should become a key facilitator in establishing primary education, reducing mortality and supporting commercial initiatives in Least Developed Countries (LDCs). The main barrier to the development of IT services in these regions is not only the lack of communication facilities, but also the lack of consistent information systems, security procedures, economic and legal support, as well as political commitment. In [18] , we propose the vision of an infrastructureless data platform well suited for the development of innovative IT services in LDCs. We propose a participatory approach, where each individual implements a small subset of a complete information system thanks to highly secure, portable and low-cost personal devices as well as opportunistic networking, without the need of any form of infrastructure. [18] reviews the technical challenges that are specific to this approach.