Particle-Filter-Based Based Detection and Tracking of an Moving Node in Wireless Sensor Networks
In this paper, the authors investigates the performance of the Particle Filter (PF) algorithm for tracking a moving object using a Wireless Sensor Network (WSN). It is well known that the PF is particularly well suited for use in target tracking applications. However, a complete analysis on the effect of various design and calibration parameters on the accuracy of the PF has been overlooked. This paper outlines the results from such a study. In particular, they evaluate the effect of various design parameters and calibration parameters (such as the gain, path loss factor, noise variations, and non-linearity constant) on the tracking accuracy and computation time of the particle -filter-based tracking system.