Model-driven adaptive wireless sensing for environmental healthcare feedback systems

Energy-aware and quality-driven sensor management for green mobile crowd sensing

“…The results show that the overall energy consumption is significantly reduced in the case of our G-MCS approach when compared to a baseline approach which acquires all generated sensor data. Also, the G-MCS approach achieves savings in terms of the number of transmitted messages compared to the model-driven approach (Nikzad et al, 2012) when the model error rate is higher than 10%. As expected, the largest reduction is achieved during the day period when users are active and moving through the city.…”

“…we present an integrated G-MCS framework which relies on a cloud-based IoT architecture centered around a QoS sensor management component and publish/subscribe middleware, we generalize the energy savings model to include variable data requirements within cells of interest, we discuss and assess sensor quality valuation functions adequate for our G-MCS approach, we include novel results based on an evaluation performed on the real data set collected in Seoul, South Korea for the purpose of understanding a participatory MCS application and which is thus in this paper adjusted to be applicable for evaluating opportunistic MCS, as explained in Section 5, and we compare the G-MCS approach with the model-driven approach proposed in Nikzad et al (2012) in terms of their energy blueprint.…”

“…Another approach is presented in Nikzad et al (2012) where the authors introduce a model-driven adaptive sensing technique which sends sensor readings from mobile devices to a back-end server only in cases when readings improve or correct the pollution map. This approach is based on utilizing a dynamic local model of the monitored environmental conditions to predict upcoming sensor readings and determine whether they should be communicated through the network.…”

“…Next we discuss the top-k approach to continuously identify the k-best data sources for an MCS task (refereed to as the top-k/w approach). Finally, we compare the G-MCS approach to the utility-driven (Riahi et al, 2013) and model-driven (Nikzad et al, 2012) MCS solutions.…”

“…The problem of energy efficiency in MCS applications is addressed in Nikzad et al (2012) where the authors present an approach for model-driven adaptive environmental sensing. They aim to reduce the amount of redundant sensor readings by maintaining local models of expected sensor readings on mobile devices.…”