Abstract—Wireless sensor networks (WSNs) and their applications have received considerable interest in the last few years. In WSNs, accurate path loss models should be considered to achieve a successful distribution of several nodes. In this work, two path loss models are proposed to evaluate the distance between two ZigBee WSNs. First, a path loss model based on conventional Log-Normal Shadowing Model (LNSM) is derived using the collected received signal strength indicator (RSSI) of the ZigBee in real time. Second, a new path loss model based on Particle Swarm Optimization (PSO) algorithm hybridized with Polynomial Equation (PE) is proposed. The PSO algorithm is used to select the optimum coefficients of PE. These coefficients can be utilized to optimize the distance estimation error based on the curve fitting. Therefore, the new path loss model called hybrid PE-PSO is innovated in this work. The hybrid PE-PSO model considerably improves the distance estimation accuracy compared with the LNSM. Results show that the hybrid PE-PSO achieves 85% improvement in distance error compared with the traditional LNSM. The mean absolute error of 0.77 m is obtained for distance estimation, which outperforms that by state of the arts.
 

Index Terms—Indoor environment; LNSM; measurement; PSO; radio propagation; RSSI; WSN; ZigBee

Cite: Huda Ali Hashim, Salim Latif Mohammed, and Sadik Kamel Gharghan, "Path Loss Model-Based PSO for Accurate Distance Estimation in Indoor Environments," Journal of Communications, vol. 13, no. 12, pp. 712-722, 2018. Doi: 10.12720/jcm.13.12.712-722