Assuring accuracy on low penetration rate mobile phone-based traffic state estimation system

Quang Tran Minh, Eiji Kamioka

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper investigates the effect of the penetration rate on the effectiveness of the mobile phone-based traffic state estimation. As a result, the acceptable penetration rate is identified. This recognition is useful for the investigating to bring the traffic state estimation using mobile phones as traffic probes into the real world application. In addition, an adaptive velocity-density estimation model, namely the velocity-density inference circuit, is proposed to improve the accuracy of the average velocity and the density estimations in cases of low penetration rate. Furthermore, a neural network-based prediction model is introduced to assure the effectiveness of the velocity/density estimation when the penetration rate degrades to zero. The experimental evaluations reveal the effectiveness as well as the robustness of the proposed solutions.

Original languageEnglish
Title of host publication2011 IEEE Vehicular Technology Conference Fall, VTC Fall 2011 - Proceedings
DOIs
Publication statusPublished - 2011
EventIEEE 74th Vehicular Technology Conference, VTC Fall 2011 - San Francisco, CA, United States
Duration: 2011 Sept 52011 Sept 8

Publication series

NameIEEE Vehicular Technology Conference
ISSN (Print)1550-2252

Conference

ConferenceIEEE 74th Vehicular Technology Conference, VTC Fall 2011
Country/TerritoryUnited States
CitySan Francisco, CA
Period11/9/511/9/8

Keywords

  • ANN-based prediction
  • low penetration rate
  • mobile probes
  • traffic state estimation
  • velocity-density inference circuit

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Assuring accuracy on low penetration rate mobile phone-based traffic state estimation system'. Together they form a unique fingerprint.

Cite this