Abstract
Internet-of-Things (IoT) can allow healthcare professionals to remotely monitor patients by analyzing the sensors outputs with big data analytics. Sleeping conditions are one of the most influential factors on health. However, the literature lacks of the appropriate simulation tools to widely support the research on the recognition of sleeping postures. This paper proposes an agent-based simulation framework to simulate sleeper movements on a simulated smart bed with load sensors. This framework allows one to define sleeping posture recognition algorithms and compare their outcomes with the poses adopted by the sleeper. This novel presented ABS-BedIoT simulator allows users to graphically explore the results with starplots, evolution charts, and final visual representations of the states of the bed sensors. This simulator can also generate logs text files with big data for applying offline big data techniques on them. The source code of ABS-BedIoT and some examples of logs are freely available from a public research repository. The current approach is illustrated with an algorithm that properly recognized the simulated sleeping postures with an average accuracy of 98%. This accuracy is higher than the one reported by an existing alternative work in this area.
Highlights
Internet-of-Things (IoT) has allowed people to collect and analyze information from many environments, devices and objects integrated in common daily activities
DISCUSSIONS The current approach has presented a mechanism for allowing researchers to test different sleeper posture recognition algorithms in a simulated smart bed with IoT and big data generated from their sensors
WORK The current work has presented an agent-based simulators (ABSs) for supporting the development of algorithms for detecting sleeping postures by simulating a sleeper in a smart bed with IoT and load sensors
Summary
Internet-of-Things (IoT) has allowed people to collect and analyze information from many environments, devices and objects integrated in common daily activities. ABSEM [13] is an ABS that simulates the emotions and the bodily sensation maps of some meditators following specific mindfulness interventions In this context, this work presents a framework for simulating sleeping postures for promoting and facilitating the research area about sleeping posture recognition through smart beds with sensors and IoT. We developed an ABS that simulates different kinds of sleepers in a smart bed with load sensors This simulator is called ABS-BedIoT, and its underlying framework provides support for the development of sleep posture recognition algorithms. The ABS generates logs with big data about the simulated signals of the smart bed sensors with different kinds of sleepers, so that researchers can explore big data analytics for sleep posture recognition.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
