The Internet of Things (IoT) refers to an environment of ubiquitous sensing and actuation, where all devices are connected to a distributed backend infrastructure. The main benefit of the IoT is the ability to use myriad sensor data, leveraged into high-level information about the entities in the system for reasoning and actuation in context-aware applications. Significant growth in sensor deployment has lead to unregulated and diverse information being fed back to the system at large. A formal specification, or ontology, for data use provides regulation to the system. In addition, IoT middleware is required for context-aware applications to operate in an environment with constantly changing data, sources, and context. In this paper, we present a context engine for IoT applications founded on an ontology that specifies and reasons on context information. We explore and build upon related work on IoT needs and ontological principles. Our infrastructure leverages context information for learning and processing a changing environment. Finally, we implement two applications: one to demonstrate machine learning from heterogeneous, intermittent sources, and another with an end-to-end implementation of user-driven actuation using the IoT backend. In the former, we produce an output stream of context information 60x more accurate than either of the individual sensor streams alone. The latter exemplifies the ease of development and extension, with only 20% infrastructure-related overhead.
The authors of these documents have submitted their reports to this technical report series for the purpose of non-commercial dissemination of scientific work. The reports are copyrighted by the authors, and their existence in electronic format does not imply that the authors have relinquished any rights. You may copy a report for scholarly, non-commercial purposes, such as research or instruction, provided that you agree to respect the author's copyright. For information concerning the use of this document for other than research or instructional purposes, contact the authors. Other information concerning this technical report series can be obtained from the Computer Science and Engineering Department at the University of California at San Diego, firstname.lastname@example.org.
[ Search ]