Machine Learning-based Advanced Localization Method in Wireless Communication

In existing localization research area, most of the localization methods suffer from propagation loss because of multi-path effects and sensitivity of the components of wireless technology, e.g. Received Signal Strength Indicator (RSSI). It is leading to miss estimation of localization methods and degradation of estimation accuracy. In this paper, a new advance localization method is proposed in different point of view. K-nearest neighbor (KNN) algorithm is adopted to get the best estimation accuracy and RSSI is used as main feature for estimating unknown position of mobile user. We extend to estimate a circle region instead of one coordinate position with the aid of circle theory as a new idea. The objective of this research is to improve the accuracy of localization methods and to mitigate the sensitivity of the existing methods. In this study, an advance localization method is proposed by considering existing finger printing method from a different point of view. In addition, Radius value optimization problem is investigated to be able to analyze the effects of different radius values on localization accuracy. On the other hand, the sensitivity of estimation accuracy of proposed method is analyzed in terms of different norm functions.

@article{Myint2018,
  title     = {Machine Learning-based Advanced Localization Method in Wireless Communication},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {9},
  number    = {1},
  year      = {2018},
  publisher = {The Science and Information Organization},
  author    = {San Hlaing Myint and Takuro Sato},
  doi       = {10.14569/SpecialIssue.2018.090109},
  url       = {https://doi.org/10.14569/SpecialIssue.2018.090109}
}