We propose a mobile peer to peer personal cloud architecture which allows users to capture, store, analyse, interact with and share different types of personal and context data with no privacy leakage. Our mobile personal cloud can host multiple different services which are intelligent, distributed, dynamic and operate in real time. In this paper we describe one service that we designed and deployed on our mobile personal cloud called Mobile Wellbeing Companion Cloud (MWCC). Using low-cost, off-the-shelf hardware components and open-source software, our MWCC combines several sensor network technologies to allow users to monitor and interact with their personal data and environment in real time without privacy leakage. MWCC augments heterogeneous sensors data with state of the art machine learning algorithms for signal filtering, fast classification and analysis and provides interactive data visualisation for transparent user interaction. We show that our MWCC is easy to use and highly accurate while managing to keep resource costs low.
Radenkovic, M., & Huynh, V. S. H. (2016). Low-cost mobile personal clouds. https://doi.org/10.1109/IWCMC.2016.7577166