• DocumentCode
    3742368
  • Title

    Collision-aware composite hypothesis testing in random-access WSNs with sensor censoring

  • Author

    Seksan Laitrakun;Edward J. Coyle

  • Author_Institution
    School of Information Technology, Mae Fah Luang University, Thailand
  • fYear
    2015
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    We consider a distributed composite hypothesis testing problem in which sensor nodes share a collision channel to send their decisions and the fusion center (FC) has a limited time to collect these decisions. When the FC does not have enough time to collect all local decisions successfully, we propose a transmission protocol called sensor censoring random access as the multiple access scheme used by sensor nodes to send their decisions to the FC. By using this protocol, the collection time is divided into frames, where each frame consists of a number of time slots. The sensor nodes whose observations are within a specific range will send their decisions in a specific frame by using slotted ALOHA. Thereafter, we derive a Rao test used by the FC to decide whether the event is happening. Since this Rao test is aware of packet collisions and exploits them to make a global decision, we call it a collision-aware Rao test (CA-Rao test). Its asymptotic performance (the probabilities of detection and false alarm) is determined. The receiver operating characteristics of the CA-Rao test are evaluated and compared to those of a Rao test of distributed detection using parallel access channels.
  • Keywords
    "Protocols","Testing","Wireless sensor networks","Probability density function","Electronic mail","Receivers","Learning systems"
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Engineering Conference (ICSEC), 2015 International
  • Type

    conf

  • DOI
    10.1109/ICSEC.2015.7401430
  • Filename
    7401430