Empirical Study of Near Ground Propagation in Forest Terrain for Internet-of-Things Type Device-to-Device Communication

• Published in: IEEE access Vol. 6; pp. 54052 - 54063
• Main Authors: Hejselbaek, Johannes, Odum Nielsen, Jesper, Fan, Wei, Pedersen, Gert Frolund
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Antenna measurements; Antennas; Clutter; clutter loss; Continuous radiation; Device-to-device communication; deviceto- device (D2D); Elevation; excess loss; Foliage; forest; Forestry; Internet-of-Things (IoT); Leaves; Loss measurement; narrow band IoT (NB-IoT); near ground propagation; path-loss; Power measurement; Propagation; Propagation losses; Radio propagation; Receiving antennas; Terrain; vegetation; Wave power
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2018.2871368

This paper presents continuous-wave power measurements conducted in forest terrain with a focus on path-loss in a device-to-device (D2D) communication scenario. The measurements are performed at 917.5 MHz with measurement ranges extending to more than 2.5 km, using a purpose-developed measurement system with a dynamic range of 180 dB. The impact of different antenna heights has been investigated by placing antennas at 1.5-, 2.5-, and 3.5-m elevation over the terrain. The measurements show that the antenna elevation in the given scenario has no significant impact on the received power. The measurements also show that the dominant path of propagation is through the foliage for the first approximately 1000 m, resulting in foliage excess loss being the dominant loss mechanism in this region. After approximately 1000 m, the measured received power tends to follow the fourth-power law, indicating that the dominant loss mechanism in this region is the distance-dependent path-loss. The measurement results have been compared to models proposed in the literature based on empirical data, as well as models proposed by ITU and COST. The comparison shows that only two models predicted the limit in the foliage excess loss in the measurement data. By applying these excess loss models in combination with path-loss models, it is possible to model the measured total path-loss with a root-mean-square error of less than 10 dB. This is achieved by either applying a model proposed by Tewari or a combination between the two-ray path-loss model and the ITU-R P.2108 clutter loss model.