Reliable ranging technique based on statistical RSSI analyses for an ad-hoc proximity detection system

  • Authors:

  • Keiichi Nakamura;Masato Kamio;Tetsushi Watanabe;Shinsuke Kobayashi;Noboru Koshizuka;Ken Sakamura

  • Affiliations:

  • Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Japan;Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Japan;Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Japan;Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Japan;Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Japan;Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Japan

  • Venue:
  • PERCOM '09 Proceedings of the 2009 IEEE International Conference on Pervasive Computing and Communications
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a proximity detection system that supports pedestrian traffic. Active radio tags are utilized for building the system to reduce the overall system cost. But accurate ranging based on utilizing active radio tags only is difficult to achieve, since the RSSI is affected by small-scale fading of a radio signal in real-world environments. In this paper, 4 simple statistical RSSI filtering methods and a packet exchange protocol for reducing the fading effect are proposed. Combinations of proposed methods are thoroughly experimented in real-world environments and discussed for determining the best combination for the system. Our results show that the variation of detecting distance is successfully reduced to 4 meters for particular combinations.