Energy/coverage quality trade-off based tasks allocation for opportunistic real time mobile crowdsensing

摘要

Opportunistic real time mobile crowdsensing pertains to the instantaneous monitoring of large scale phenomena by leveraging human mobility and smartphones' sensors. However, while urban applications concern more about coverage quality and timeliness of detected data, mobile users should care about their consumed energy for performing a sensing task. In this paper, we introduce the Real time OPportunistic Scheduler Framework for Energy aware mobile Crowdsensing (Re-OPSEC). Especially, we present in deep the off-line tasks allocation approach operating on top of energy saving model. The methodology consists of two main processes. Re-OPSEC builds firstly connectivity patterns called online episodes and extracts mobility information jointly with those patterns. Secondly, it allocates, in an off-line mode, online episodes to send jointly sensing tasks in real time. Allocated patterns should increase the total expected sensing revenue and protect volunteers from embarassed situations such as their smartphones turning off. The performance of the approach is evaluated using realistic mobility datasets. The results demonstrate firstly the superiority of our approach as compared to an other existing work. Secondly, they prove that coverage rate can achieve a high level in spite of energy saving mechanisms restriction.