Passive localization based on local observations in wireless sensor network using finite element method and Kalman filter

This paper deals with the recently investigated physical field model in wireless sensor network without additional infrastructure for location estimation, i.e. passive localization, in the form of partial differential equation (PDE) by employing the temperature distribution in the field. This novel framework can be defined into two steps. In the first step, the estimation algorithm of sensor node localization problem using finite element method (FEM) has been employed to obtain the solution based on physical phenomena (e.g., temperature) governed by discretizing the 1-D heat equation. In the second step, under consideration of uncertainty both occurring in the system and arising from noisy measurements, a refinement technique, i.e. Kalman filter, is incorporated to improve the accuracy of error analysis by reconstruction and prediction of temperature distribution. The computational results illustrate that the significant effect and better accuracy using elegant finite element approach and Kalman filter technique by means of simulation results.