1st International ICST Workshop on Wireless Networking for Intelligent Transportation Systems

Research Article

An architectural framework for the support of intelligent vehicular network monitoring

  • @INPROCEEDINGS{10.1145/1577769.1577771,
        author={Angelos Lenis and Vassilis Chatzigiannakis and Mary Grammatikou and Symeon Papavassiliou},
        title={An architectural framework for the support of intelligent vehicular network monitoring},
        proceedings={1st International ICST Workshop on Wireless Networking for Intelligent Transportation Systems},
        publisher={ACM},
        proceedings_a={WIN-ITS},
        year={2007},
        month={8},
        keywords={Vehicular network monitoring Sensor Grids Data fusion Incident detection},
        doi={10.1145/1577769.1577771}
    }
    
  • Angelos Lenis
    Vassilis Chatzigiannakis
    Mary Grammatikou
    Symeon Papavassiliou
    Year: 2007
    An architectural framework for the support of intelligent vehicular network monitoring
    WIN-ITS
    ACM
    DOI: 10.1145/1577769.1577771
Angelos Lenis1,*, Vassilis Chatzigiannakis1,*, Mary Grammatikou1,*, Symeon Papavassiliou1,*
  • 1: Network Management & Optimal Design Laboratory (NETMODE), School of Electrical & Computer Engineering National Technical University of Athens (NTUA) 9 Iroon Polytechniou str. Zografou 157 73, Athens, Greece
*Contact email: anglen@netmode.ntua.gr, vhatzi@netmode.ntua.gr, mary@netmode.ntua.gr, papavass@mail.ntua.gr

Abstract

In this paper we propose an efficient, scalable and secure architectural framework for the monitoring of multiple fixed and mobile sensors, designed for road traffic incident detection. The Grid provides the means to control the sensors and gather information with security and reliability. The system includes a Decision Support Service that fuses multi-metric data from heterogeneous sensors to produce a global and comprehensive view of the vehicular network state. The adopted fusion algorithm is based on the application of Principal Component Analysis on multi-metric data, and provides an efficient way of taking into account the combined effect of the correlated observed data, for incident detection purposes. Finally, the performance and operational effectiveness of the proposed approach and system is evaluated via modeling and simulation.