Empirical performance of RSSI-based Monte Carlo localisation for active RFID patient tracking systems

William P.L. Cully, Simon L. Cotton, W.G. Scanlon

    Research output: Contribution to journalArticleAcademicpeer-review

    12 Citations (Scopus)
    8 Downloads (Pure)

    Abstract

    The range of potential applications for indoor and campus based personnel localisation has led researchers to create a wide spectrum of different algorithmic approaches and systems. However, the majority of the proposed systems overlook the unique radio environment presented by the human body leading to systematic errors and inaccuracies when deployed in this context. In this paper RSSI-based Monte Carlo Localisation was implemented using commercial 868 MHz off the shelf hardware and empirical data was gathered across a relatively large number of scenarios within a single indoor office environment. This data showed that the body shadowing effect caused by the human body introduced path skew into location estimates. It was also shown that, by using two body-worn nodes in concert, the effect of body shadowing can be mitigated by averaging the estimated position of the two nodes worn on either side of the body.
    Original languageUndefined
    Pages (from-to)173-184
    Number of pages12
    JournalInternational journal of wireless information networks
    Volume19
    Issue number3
    DOIs
    Publication statusPublished - 5 Sept 2012

    Keywords

    • EWI-22243
    • Localisation
    • ShadowingOff-body
    • RSSI
    • Wearable
    • Tracking
    • Personnel
    • METIS-296088
    • IR-81539
    • Body centric
    • Monte Carlo

    Cite this