The internet of things (IoT) is becoming an increasingly important area of research.  The ubiquity of interconnected sensors embedded in the physical world continues to grow; enriching the sensing fabric overlay that has the potential to increase our understanding of the interaction between people and their environments.  This is transforming various industries including building energy management and maintenance, industrial sector applications, connected cars, and healthcare.  We are also seeing it in our homes through smart thermostats and smart appliances.  Smart phones are more ubiquitous than ever and most new smart phones come with at least half a dozen sensors, providing information of the surrounding environment.  Data acquired using IoT devices will be massive in quantity, but to unlock its value the data needs to be processed and analyzed in reliable and useful ways. This presents researchers with the opportunity to explore how all the available data can be merged to get a holistic, time-varying understanding of the world like never before.

Fundamental challenges remain.  For example, we are seeing a growing use of cheap sensors transmitting physical readings over lossy links and stacks, often involving layers of transmission before reaching a data store for analysis.  The readings are noisy, missing, often uncalibrated, and temporally and spatially unaligned.  Moreover, the combination of modeling and data presents a family of new challenges as well.  In order to fulfill the potential of the internet of things we must explore ways to handle this complexity.