Wireless devices nowadays support flexible formation of ad hoc networks. This allows users to establish virtual connections between their devices and devices they serendipitously encountered in their environment, and in a spontaneous manner. This process is known as spontaneous device association.

Spontaneous device association has been an ongoing research topic for many years. Much existing work examined technical aspects, such as suggesting new techniques using special hardware sensors and supporting robust security. However, users aspects were seldom investigated. By considering user aspects, researchers can create association techniques that are better-suited for users and enhance usability. This thesis examines aspects of how users are involved in device association, the action that people spontaneously choose for associating devices, and what researchers need to consider for creating a usable association process.

This thesis takes two research approaches: conceptual analysis and action elicitation. The first part of this thesis reviews existing work and examines the past and the current trends of spontaneous device association. It presents a taxonomy structure for classifying association techniques and an in-depth summary of existing device association research. It reveals the state of the art and discusses design considerations for building usable association techniques. The second part of this thesis elicits ideas for device association. By applying a methodology where people were asked to come up with actions for associating devices without the concern of technology, it answers the question of how people intuitively associate devices, in single-operator and group scenarios. The combined knowledge from the two parts forms the core contributions of this thesis.

The contributions include presenting new conceptual models, identifying a set of factors that influence usability, and discovering the rationale of how non-technical users associate devices. Together, they provide knowledge for considering users, which potentially help researchers to create usable techniques that are in-line with people's preferences and expectation in spontaneous device association.