QEMSS: A selection scheme for participatory sensing tasks

The new generation of smart devices, equipped with a large variety of sensors, enhances the Participatory Sensing of data. However, many issues arise when selecting participants to perform the sensing tasks. These issues are necessarily related to the limited energetic resources of devices, the impact of users mobility as well as the quality of collected data, recently defined as “Quality of Information” (QoI). In this context, we propose QEMSS (QoI and Energy aware Mobile Sensing Scheme) as a selection scheme for participatory sensing tasks, taking into consideration the quality of sensed data, QoI, and the dedicated energy for their acquisition. The aim of our model QEMSS is to select, among all participants in the sensing campaigns, the subset of users who maximizes the QoI of non redundant information while minimizing the overall energy consumption. To do so, we illustrate our selection scheme based on the Tabu Search algorithm in order to achieve a sub-optimal solution. Simulation results were compared to two other State of The art schemes: the Random Selection (RS) and a method based on a greedy search (DPS). Our scheme is proved to be as performing as the two other methods. Particularly, our scheme achieves a very high quality of information in challenging scenarios such as low dense areas and/or low energetic resources.