6533b872fe1ef96bd12d3eb0

RESEARCH PRODUCT

Optimization of flat area coverage under connectivity constraint in wireless sensor networks

subject

description

A wireless sensor network consists of a set of small autonomous units that interact via a network built by their communication modules. They observe their environment, capture information, then manage this information according to their computing and/or storage capacity. To effectively accomplish their task(s), they need to cover as much of the area of interest as possible. It is therefore essential to quantify the quality of their coverage. In this thesis, we therefore seek to best cover an area of interest, with a precise number of sensors. While taking into account the possible overlaps between sensors, we first deploy in a zone of regular dimensions and evaluate the exact coverage using a genetic algorithm. The proposed evaluation function allows us to calculate without bias the exact coverage of the sensors. Once we know how to cover and assess the coverage of a regular area, the second step is to deploy the sensors on an area of irregular shapes. The objective here will be to manage not only to effectively cover the zone taking into account the overlapping of the sensors, but also to favor the inside of the zone of interest compared to the outside of it. We then create inside the deployment zone, areas that we call "zones of no interest", where the sensors should not be deployed. Finally, if it is important to ensure good coverage of a area of interest, it is just as important to ensure that once the data has been captured, it can be sent to the other sensors for processing and/or relay to the processing node(s). To the objectives previously stated, we therefore add in the second case, the constraint of total connectivity between the sensors.