A Random Access Scheme to Balance Energy Efficiency and Accuracy in Monitoring Applications

In this paper we consider a monitoring application where the process of interest is the time integral of measurements provided by some sensors, e.g., the volume of fluid crossing a pipe, the total amount of fertilizer/insecticide dispersed onto a certain cultivation area, and so on. In this context, a tradeoff arises between the accuracy obtained in the monitoring process and the energy consumption for the transmission of the sensors' measurements, which is critical since many devices are expected to be battery powered. In this scenario, we propose an adaptive sampling and random access strategy that adapts the sensing accuracy and transmission probability based on the desired Quality of Service (QoS) level, expressed in terms of reconstruction accuracy of the monitored process, while minimizing the energy consumption of the sensor nodes. We also present a thorough analytical study of this scheme and of its optimization. The latter procedure involves a two-step process which considers separately the interference in the network and the constraint on the reconstruction accuracy and iterates until convergence.