Wi-Fi Based Small-sample-size Positioning using Rayleigh Distribution

  • 藪圭輔
    School of Engineering, Nagoya University
  • 榎堀優
    Graduate School of Information Science, Nagoya University
  • 間瀬健二
    Graduate School of Information Science, Nagoya University

Bibliographic Information

Other Title
  • レイリー分布を用いたWi-Fi電波強度分布に基づく少数標本時位置推定

Search this article

Description

In Wi-Fi based positioning, RSSI distribution matching is required to realize robustness for RSSI noise that is occurred by humidity change, human movement, and so on. However, RSSI distribution matching is not effective in small-sample-size positioning if it use common distribution such as Gaussian. Thus, in this paper, we propose use of Rayleigh distribution in order to solve the issue. The Rayleigh distribution is a model of scattering signals arrival with multiple paths in communication theory, and is able to estimate its distribution with small-size samples. As an evaluation result, in two samples, the F-measure is 0.770±0.235 with our proposed method, and it is 0.055 ± 0.054 with Gaussian; and in five samples, the F-measure is 0.871 ± 0.174 with our proposed method, and it is 0.647 ± 0.195 with Gaussian. In addition, our proposed method is able to estimate position in only one sample with 0.522 ± 0.235 in F-measure.

Journal

Keywords

Details 詳細情報について

  • CRID
    1570854177859788800
  • NII Article ID
    110009676732
  • NII Book ID
    AA11851388
  • ISSN
    09196072
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top