Frank Siegemund, Michael Rohs
Communication platforms for ubiquitous computing need to be flexible, self-organizing, highly scalable and energy-efficient, because in the envisioned scenarios a large number of autonomous entities communicate in potentially unpredictable ways. Short-range wireless technologies form the basis of such communication platforms. In this paper we investigate device discovery in Bluetooth, a candidate wireless technology for ubiquitous computing. Detecting new devices accounts for a significant portion of the total energy consumption in Bluetooth. It is argued that the standard Bluetooth rendezvous protocols for device detection are not well suited for ubiquitous computing scenarios, because they do not scale to a large number of devices, take too long to complete, and consume too much energy. Based on theoretical considerations, practical experiments and simulation results, recommendations for choosing inquiry parameters that optimize discovery performance are given. We propose an adaptive rendezvous protocol that significantly increases the performance of the inquiry procedure by implementing cooperative device discovery. Also higher level methods to optimize discovery performance, specifically the use of sensory data and context information, are considered.