Refined numerical simulation of signal propagation in soils for wireless underground sensor networks

Wireless underground sensor networks (WUSN) have promising applications in the field of underground structural health monitoring. Signal path loss prediction is a great challenge in the design and onsite installation of WUSN. Existing wireless signal propagation models cannot well predict the total path loss due to the neglection of node detailed structure. This paper develops a numerical model in FEKO to simulate the signal propagation in soils considering the detailed structure of buried sensor node including linear shaped antenna, coaxial feeder, batteries and protecting PVC box. The developed numerical model is validated using the Friis equation in free space and results of outdoor signal propagation experiments. Compared to analytical propagation models, the developed numerical model can well consider the path loss induced by sensor node itself. The proposed numerical model is able to simulate the signal propagation in heterogeneous soils. Signal path loss fluctuates in heterogeneous soils due to the coupling of direct, reflected and refracted waves. Parametric studies are conducted to investigate the effects of soil properties, signal frequency, node orientation and soil heterogeneity on total path loss. An optimal frequency foptimal exists to minimize the signal path loss and foptimal is in the range of 0.56-0.65 GHz for the node in this paper.