Harvesting of Location-Specific Information Through WiFi Networks

Abstract

Ubiquitous computing requires ready access to information that is relevant to users’ context – especially information relevant to their current location. Applications on our personal devices should be able to autonomously and continuously harvest the information provided at that location and interrupt us only when it is important to do so. Currently, client devices are designed for explicit querying for information rather than continuous background harvesting of relevant information. To enable ubiquitous access to location-specific information, we can take advantage of the widespread deployment of WiFi networks. There is a wealth of location-specific information that network providers are willing to make publicly available to any users. However, today’s models for accessing wireless networks do not easily support this due primarily to concerns over security and bandwidth utilization. In this paper, we present and compare the different methods that can be applied to solve the problem of continuous background access to location-specific information. Specifically, we compare client-pull and server-push models and show how tradeoffs can be made involving privacy, power consumption on devices, and utilization of wireless bandwidth. We also present three applications and discuss how the tradeoffs affect their design.