Section: New Results

Sensor networks

Participants : Nizar Bouabdallah, Sofiane Moad.

Wireless Sensor Networks (WSNs) are composed of tiny sensor nodes, which are capable of sensing and processing data from inaccessible environments and communicating them to the end-user for further analysis. WSNs are characterized by the limited capacity of their sensor node batteries, making energy efficiency a critical issue. Once a WSN is deployed, sensor nodes must self-organize and live as long as possible, based only on their initial energy stores. Consequently, techniques minimizing energy consumption are required to improve network lifetime. Our research on WSNs [72] revolves around two main directions: 1) clustering, and 2) radio diversity. Regarding clustering, we first developed a Connectivity Degree-Based Energy Efficient Clustering Protocol for WSNs (CDEEC) [55] , resulting in better topology management and decreased energy consumption compared to the well-known clustering protocol HEED. Then, we integrated a compression mechanism within a cluster-based architecture to develop a Compression Cluster-based scheme in a Spatial Correlated Region protocol ($CC\_SCR$) [56] , with the goal of further decreasing the energy consumption. In the direction of radio diversity, we first proposed the WETX metric [58] which uses a minimum-energy radio while routing, then we proposed the BL metric [57] , on top of WETX, which allows energy-balancing inside a network in order to further extend its lifetime. The validation of our contributions was carried out with analytical analysis, and simulation using TOSSIM.