Pebbles and Urns: A Tangible, Presence-Based Service Delivery Framework

Wireless and pervasive computing research continues to study ways the Internet of Things (IoT)
can make lives easier and more productive. Areas of interest include advances in new
architectures and frameworks that support large-scale IoT deployments beyond research
prototypes, simple and inexpensive human-to-device and device-to-device interfaces, and user
decision making support with opportunistic information services.

This dissertation investigates the design and implementation of a general-purpose framework
upon which IoT and opportunistic computing (OC) systems can be built.

The result of this work is Pebbles and Urns (P&U), a casually accessible system designed to
deliver information to a person that is pertinent and beneficial to them with respect to their
current activity, location and other contexts. P&U is a proximity-based information delivery
framework that leverages a simple, inexpensive tangible interface and context-rich, physicallysituated,
distributed information repositories. By its proposed use of enforced proximity, local
context, and location-specific services, P&U can support the situated interaction between user
and place.

The P&U framework is based on a layered architecture consisting of an isolated physical
communication layer, a data repository supporting opportunistic service composition and
delivery, and a controller/interface providing user feedback. Serving as a potential IoT design
pattern, P&U application developers can use the framework API's and software tools to build
and deploy P&U systems.

As validation of this work, P&U prototypes are constructed using the framework, API's and
software tools. The prototypes are based on use cases depicting a person engaged in the day-today
activities of attending class, going to the gym and grocery shopping. Performance
measurements are performed on the prototypes profiling core components of the framework.
Results indicate proper functioning of P&U tangible interfaces, communication connections,
service request and delivery, and internal framework operations.

Contributions of this research include a general-purpose framework, a simple IoT interface and
an opportunistic engine.