Practical metropolitan-scale positioning for GSM phones

  • Authors:

  • Mike Y. Chen;Timothy Sohn;Dmitri Chmelev;Dirk Haehnel;Jeffrey Hightower;Jeff Hughes;Anthony LaMarca;Fred Potter;Ian Smith;Alex Varshavsky

  • Affiliations:

  • Intel Research Seattle;University of California at San Diego;University of Washington;Intel Research Seattle;Intel Research Seattle;University of Washington;Intel Research Seattle;University of Washington;Intel Research Seattle;University of Toronto, Canada

  • Venue:
  • UbiComp'06 Proceedings of the 8th international conference on Ubiquitous Computing
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper examines the positioning accuracy of a GSM beacon-based location system in a metropolitan environment. We explore five factors effecting positioning accuracy: location algorithm choice, scan set size, simultaneous use of cells from different providers, training and testing on different devices, and calibration data density. We collected a 208-hour, 4350Km driving trace of three different GSM networks covering the Seattle metropolitan area. We show a median error of 94m in downtown and 196m in residential areas using a single GSM network and the best algorithm for each area. Estimating location using multiple providers' cells reduces median error to 65-134 meters and 95% error to 163m in the downtown area, which meets the accuracy requirements for E911. We also show that a small 60-hour calibration drive is sufficient for enabling a metropolitan area similar to Seattle.