Title page for etd-0801116-152823


[Back to Results | New Search]

URN etd-0801116-152823
Author Jang-Yo Wang
Author's Email Address No Public.
Statistics This thesis had been viewed 5576 times. Download 588 times.
Department Applied Mathematics
Year 2015
Semester 2
Degree Master
Type of Document
Language zh-TW.Big5 Chinese
Title Sensor Network Localization with Support Vector
Machine Learning
Date of Defense 2016-07-11
Page Count 68
Keyword
  • Kalman Filter
  • Triangulation method
  • Received Signal Strength Indication
  • Path Loss Coefficient
  • Support Vector Machine
  • Orthogonal frequency-division multiplexing
  • Abstract In the wireless localization system, the accuracy of the localization will be reduced due to environmental interference in the area. In this study, we apply Orthogonal frequency-division multiplexing (OFDM) technology and Kalman Filter to reduce the signal interference and enhance the Received Signal Strength Indication (RSSI). In addition, the signal transmissions with different Path Loss Coefficient are characterized using Support Vector Machine (SVM) to classify the signal data travel path in different circumstances. The proposed SVM localization algorithm is implemented using Matlab codes. The simulation results indicate our proposed sensor network localization with
    SVM learning can outperform the traditional RSSI/triangulation localization algorithm. In addition, the proposed algorithm is able to track the moving object with spline approximation to function and the simulation results show reasonable accuracy.
    Advisory Committee
  • Mong-Na Lo Huang - chair
  • Mei-Hui Guo - advisor
  • Yu-Jung Huang - advisor
  • Files
  • etd-0801116-152823.pdf
  • indicate access worldwide
    Date of Submission 2016-09-05

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have more questions or technical problems, please contact eThesys