Indoor Localization by using Particle Filtering Approach with Wireless Sensor Nodes

Hakan Koyuncu, Ahmet Çevik

Abstract


Jennic type wireless sensor nodes are utilized together with a novel particle filtering technique for indoor localization. Target objects are localized with an accuracy of around 0.25 meters. The proposed technique introduces a new particle generation and distribution technique to improve current estimation of object positions. Particles are randomly distributed around the object in the sensing area within a circular strip of 2 STD of object distance measurements. Particle locations are related to object locations by using Gaussian weight distribution methods. Object distances from the transmitters are determined by using received RSSI values and ITU-R indoor propagation model. Measured object distances are used together with the particle distances from the transmitters to predict the object locations.

Keywords


Wireless sensor node (WSN), received signal strength indicator (RSSI), Gaussian weight distribution, Standard deviation (STD), particle filter, cumulative distribution

Full Text:

PDF


DOI: http://dx.doi.org/10.24138/jcomss.v9i1.159



Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.