Microsense: Sensor Framework for IoT System-on-Chip

Related Work

In the current scenario, mobile phones not only serve as communication means, but also as a source of data from sensors that the distributed human centric sensing applications can collect and exploit. Among them, environmental monitoring systems and emergency response systems will specifically benefit from human-based sensing. Owing to the restriction on resources of mobile devices, data sensed is normally uploaded to the cloud. Nonetheless, modern solutions lack an integrated approach to support varied applications suitably, whilst lowering the power consumed on the mobile device.

In this context, Fakoor, Raj, Nazi, Di Francesco and Das (2012) put forward an integrated framework to store; process and deliver sensed data to human-centric applications installed in the cloud. The integrated platform forms the backbone of a novel delivery model, namely, Mobile Application as a Service (MAaaS) that permits the development of human-centric applications covering many disciplines, along with mobile social networks and participatory sensing. It particularly addresses a case study of an emergency response system for raising alarm in the event of fire. The framework showed reduction in the energy consumed on the mobile devices by the way of a prototype test bed implementation, while satisfying the requirements of the application.

Additionally, Sharma and Ghose (2009) presented an integrated comprehensive security framework that provides security services for all services of sensor network. Additional components i.e. Intelligent Security Agent (ISA) to evaluate degree of security and cross layer interactions in many components like Intrusion Detection System, Trust Framework, Key Management scheme and Link layer communication protocol.

Sensors attached to the smart phones and smart buildings makes possible mobile sensing and users’ behavior modeling that opens the door for cutting-edge applications such as customized intelligent computing, activity prediction, behavior intervention and health monitoring. Privacy is a major hurdle in mobile sensing and behavioral modeling. The end users invariable have less trust in the processing and storing for their personal information in the public cloud. The framework presented here uses hybrid cloud to control the computation between the devices, public cloud and personal cloud. Personal cloud is used for storing the raw sensor data and users are given complete control over it physically. Analytical widgets (e.g., health monitor or marketing survey) can collect user-authorized data on approval only. We evidently show that with this approach, there is considerable reduction in the users’ anxiety and increase in the acceptance rate of the mobile sensing technology from 23% to 60% (Zhang, Wu, Zhu et al., 2013).