A New Approach for Path Loss Prediction in Wireless Underground Sensor Networks

The Wireless Underground Sensor Networks (WUSNs) become an attractive field due to its applications. On these applications, links between nodes are made through wireless underground communication of electromagnetic waves. However, due to soil complexity, the wave signal is attenuated as it travels across the ground. The prediction of the path loss due to signal attenuation across the ground depends on several parameters like soil composition, soil particle sizes, moisture or temperature. Designing an accurate path loss model for WUSNs remains a key issue on Internet Of Underground Things (IOUT). In this paper, we proposed a new approach for path loss prediction in WUSNs. In order to achieve it, we firstly analyze the existing path losses in literature. Secondly, we evaluated and compared the Complex Dielectric Constant (CDC) derivation schemes. Thirdly, we proposed a new path loss model that uses a better CDC prediction than the existing path losses. To validate our approach, measurements in real experimental field were made and real sensor nodes are used. The conducted experimentations show that our proposed path loss is more accurate than the existing path loss models with the lowest errors. Furthermore, we show the efficiency of our approach by considering +/- 3% error of the soil moisture sensor.