Nouveau modele pour le positionnement des senseurs avec contraintes de localisation.

摘要

Wireless sensor networks continue to be without doubts a major research area. The objective of the wireless sensors network planning problem is to locate the sensors while respecting a set of performance constraints. In this work we consider coverage and connectivity constraints. Moreover we impose that each mobile station be located by the sensors (i.e. location constraints).;The purpose of this work is to propose a planning strategy of wireless sensor networks. It will ensure full coverage and increased connectivity. The approach begins by proposing a formalism of combinatorial optimization model with coverage, location and connectivity constraints. The search space is discrete and the choice of potential sites is directly related to the environment of deployment characteristics. Thus, we assume known the coordinates of positions for mobile stations and potential sites, the characteristics of radio propagation, the receiver sensitivity for mobiles as well as for sensors to be installed, the minimum number of signals for the location, the maximum hops allowed for routing between sensors. An initial greedy heuristic is proposed as well as search metaheuristic. Solutions found are compared to a lower bound obtained by a relaxed version of the model. The simulations show the validity of the planning approach. However, some limitations arise especially for the modeling of radio propagation. Indeed, we propose an improvement which is reflected in the integration of measures or estimates for the level of interference signals. From our perspective, this approach will essentially guarantee an exact location of mobiles and a better connectivity of the nodes of wireless sensor networks.