Abstract

Smartphones have become popular mainly because of the large variety of applications they can run. In contrast, most devices in the phone's environment -- e.g., household appliances or environmental sensors -- are much less flexible because their functionality is hardcoded at the design time. In order to realize the vision of the Internet of Things (IoT), where all devices communicate with each other to realize joint tasks, it is necessary that these devices are able to extend and adapt their functionalities on-the-fly based on their surrounding. To realize smart functionalities for IoT devices, we propose "hybrid Apps", the concept of Smartphone "Apps" applied to small embedded systems. In contrast with current packaged "smart home" solutions, where all appliances have to be changed to their smart counterparts at the same time, hybrid Apps permit an incremental and hence feasible deployment of the IoT vision. In this paper we discuss the challenges and opportunities associated with this approach. We argue code interpretation as a candidate reprogramming method for IoT devices and analyzed its feasibility with real-world measurements of key parameters such as computational and energy overhead. While in general, code interpretation incurs a large energy-overhead, we show that for typical IoT applications executed every few seconds, it is as low as 1%.