A generic sensing and processing framework for smartphones for heterogeneous sensing
Shahnawaz Alam, Keshaw Dewangan, Arijit Sinharay, Avik Ghose · 2016
Nowadays smartphones have become crucial part of our everyday life. Smartphones come with several built-in sensors, along with powerful processing units and easy internet connectivity which make them very attractive tools in pervasive sensing space. While there are several research tracks coming up with various interesting apps which enable such pervasive sensing/ analytics through smartphones, they lack having a common framework that supports heterogeneous sensing and processing. Here we propose a common framework that can be exploited to support heterogeneous sensing, with power management and adaptive real-time performance as an added feature. The key feature is to partition the processing between the smartphone and a cloud as well as designing the software architecture to uniformly support various sensor data logging and displaying requirements. In this work, we particularly investigate feasibility of our proposed framework by actually implementing and testing with several collaborative sensing applications that are inherently heterogeneous in nature. Our results show promise as the CPU overhead and power consumption are improved quite a bit across all the solutions we have tested.