Road Segment Identification Using Hidden Markov Model and Travel Time Estimation Using Linear Regression Technique

This paper proposes a location determination method that uses data obtained from active cellular phones and the Hidden Markov Model (HMM) technique to identify specific segment on the Bangkok expressway network that contains the position of a given phone-equipped vehicle. The paper also proposes a method using linear regression model for estimating travel time for each of the segments on the expressway using the same set of cellular data. Cell Identification (CID) and Cell Dwell Time (CDT) are the two types of data collected from mobile phones that are used in the proposed methods. Experimental results show that segment identification using HMM gives 89.86% accuracy and the linear regression technique is able to give travel time estimation with absolute percentage error of under 20% for typical trips, and absolute error of 20 seconds for short trips lasting under two minutes.


  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; Maps; References; Tables;
  • Pagination: 11p
  • Monograph Title: ITS in Daily Life

Subject/Index Terms

Filing Info

  • Accession Number: 01146614
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Dec 16 2009 2:58PM