Comparison of Indoor localization techniques by using reference nodes and weighted k-NN algorithms

Positioning accuracy with Landmarc localization systems in the literature is around 2 meters. In this study, a reference node topology is introduced and 3-NN weighted algorithm is utilized to determine the unknown target locations. Firstly, target localization is carried out by using simple Landmarc system. Secondly, two stage 3-NN algorithm is deployed to estimate the target location. Finally, 3-NN algorithm is utilized with a new weight mechanism related to both physical and Euclidean distances. Overall comparative results reveal that classical Landmarc technique introduces 1.2 meter error distance while two stage 3-NN algorithms introduces 0.7 meter and new weighted system introduces 0.4 meter error.

[1]  Steve Hinske.  Determining the Position and Orientation of Multi-Tagged Objects Using RFID Technology , 2007, Fifth Annual IEEE International Conference on Pervasive Computing and Communications Workshops (PerComW'07).

[2]  Yunhao Liu,et al.  LANDMARC: Indoor Location Sensing Using Active RFID , 2004, Proceedings of the First IEEE International Conference on Pervasive Computing and Communications, 2003. (PerCom 2003)..

[3]  Shuang-Hua Yang,et al.  A ZigBee-based mobile tracking system through wireless sensor networks , 2008 .

[4]  Ming Zhu.  Novel positioning algorithms for RFID-assisted 2D MEMS INS systems , 2008 .

[5]  Abdel-Badeeh M. Salem,et al.  RECENT ADVANCES in INFORMATION SCIENCE , 2012 .

[6]  Shuanghua Yang,et al.  A 2 D positioning system using WSNs in indoor environment , 2011 .

[7]  Dirk Timmermann,et al.  Position Estimation in Ad hoc Wireless Sensor Networks with Low Complexity , 2005 .