Distributed Localization for Participatory Sensing Systems

Abstract

The focus of this paper is on participatory sensing in which every participant carrying smartphone senses its environment and shares it with server. Most of the applications require location information to perform sensing activity. But, GPS drains considerable amount of energy if used for localization. There exists a centralized localization approach to detect a set of devices for the role of broadcaster which must turn on GPS. Its neighbouring devices rely on them to calculate their position. In this paper, we provide a novel distributed solution for finding roles for the participating mobile devices. It is important in real scenario to alleviate server from the burden of assigning roles. Our proposed approach does not make any assumption about the presence of infrastructure or device capabilities, other than the availability of WiFi and the ability to know devices within the WiFi range. To the best of our knowledge, this is the first protocol for distributed localization in participatory sensing system. Simulation results using real dataset demonstrate that our algorithm is capable of assigning role in 70-85% less time compared to the centralized algorithm but consumes 12-15% more energy as it does not find the optimal set of broadcasters which requires global information.