A Neural-Network-Based Indoor Positioning System by Using Sectored Antenna Array

Chih-Yung Chen, Yu-Ju Chen, Ya-Chen Weng, Rey-Chue Hwang

This paper presents the development of sectored antenna array and modified probabilistic neural network positioning algorithm for an indoor positioning system (IPS). Firstly, a new hexagonal IPS station is composed of six printed-circuit board Yagi-Uda antennas and Zigbee modules. It is designed to obtain the signals between an object and the station. Then, a modified probabilistic neural network (MPNN) is applied to estimate the accurate position of the object with the signal strength. From the experimental positioning results shown, the developed IPS system has the outperformance in an 8x8 square meters indoor scene. The proposed indoor positioning technique not only has a high positioning accuracy, but also is an effective solution to solve the difficult issue of positioning station deployment.