Portal de Periódicos da CAPES

WaaS (Wireless-as-a-Sensor): Conception, design and implementation on mobile platforms

2018; Elsevier BV; Volume: 88; Linguagem: Inglês

10.1016/j.sysarc.2018.05.009

1873-6165

Douglas Lautner, Xiayu Hua, Scott DeBates, Shangping Ren,

Energy Efficient Wireless Sensor Networks

With the tremendous growth of Internet-of-Things (IoT) technologies, both the number and the type of items that are sensible to computing systems increases exponentially [1]. As a result, wireless transceivers in mobile devices, such as Bluetooth or Radio Frequency Identification (RFID) readers, now have the potential to become contextual sensors by constantly interacting with nearby IoT wireless information beacons. In this work, we redefine these wireless transceivers as a new type of sensor, i.e., Wireless-as-a-Sensor (WaaS). However, current implementations of wireless transceivers are optimized as data transmitters for communication purposes, not for sensing, therefore directly implementing WaaS on a typical mobile device causes problems, such as an unacceptable reduction of battery life. To address this problem, we provide a novel smartphone-based architecture which utilizes an energy efficient sensor fusion core to execute always-on WaaS-related tasks. Based on this architecture, we further present a duty cycle determination algorithm that minimizes the power consumption while guaranteeing an application's Quality of Service (QoS) requirements. Both the architecture and the algorithm are implemented and evaluated on a Moto X testbed with Bluetooth and RFID sensors, respectively, as two case studies. A Bluetooth-Low-Energy indoor location service is also implemented to verify and evaluate the architecture in real applications. Data measured on the Moto X testbed depicts that using the WaaS architecture prolongs the battery life 5 times than that on smartphone architecture.