Home // UBICOMM 2012, The Sixth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies // View article

Exploring Efficient Methods to Extract Pedestrian Flows on a Mobile Adhoc Network

Authors:
Ryo Nishide
Hideyuki Takada

Keywords: Distributive Database; Bluetooth; Social Context; Pedestrian Flows; Mobile Devices; Adhoc Network.

Abstract:
This paper explores possibilities to extract pedestrian flows from Bluetooth detection logs in a distributive manner. Bluetooth devices are installed in mobile equipments such as laptops, tablet PCs, cell phones, and PDAs, which pedestrians carry with them in daily life. If these devices can be detected and logged, it may be possible to analyze the movements and density of surrounding pedestrians in real world. Moreover, we aim to build this system on a server-less adhoc network, in which the network can be built autonomously, and the data can be managed between mobile devices distributively. Our goal is to build this system as simple as possible, while avoiding the initial preparations to deploy sensors in real world. The results of experiments have revealed that detection logs implicitly record traces of surrounding pedestrian flows, which might provide possibilities to analyze and distinguish pedestrian flow patterns based on different situations. The paper has also discussed the related issues on network construction including methods to interpolate missing detections.

Pages: 29 to 34

Copyright: Copyright (c) IARIA, 2012

Publication date: September 23, 2012

Published in: conference

ISSN: 2308-4278

ISBN: 978-1-61208-236-3

Location: Barcelona, Spain

Dates: from September 23, 2012 to September 28, 2012