The Influence of Terrain Scattering on Radio Links in Hilly/Mountainous Regions

Abstract

Current rural propagation models only consider paths (direct, ground reflected, terrain diffracted, troposphere scattered) that lie in the vertical plane containing the transmitter and receiver. This paper shows that nonspecular scattering from terrain can give significant additional contributions to the received signal in hilly/mountainous environments. To show this, we have carried out Monte Carlo simulations of scattering in many different types of rural regions. These simulations model scattering using the bi-static radar equation with a Lambert's law coefficient. In order to permit simulations for many radio links and many terrain databases, we have developed a novel algorithm to rapidly search for terrain scattering elements that are visible to both the transmit and receive antennas. Results show that in sufficiently rugged terrain, the scattered paths can give a greater contribution to received power than vertical plane paths in more than 80% of links. Conclusions are drawn for radio channel characteristics, such as angle spread and short-term spatial fading. Statistical models dependent on terrain variation for path loss, and temporal response are also developed.Index Terms-Mountainous terrain, path loss model, rural propagation, tapped delay line, terrain scattering, viewshed.

Spatial Diversity Deconstruct of C-band Scatter Components in Multistatic RaDaR Datasets using Machine Learning Techniques

“…This means that even if clustering is repeated for the same value of 'k', say n = 5, the resulting clusters are different each time. To address this issue, the method employed has been to iterate clustering multiple times for each value of 'k', say 20, and find the most optimum cluster by selecting the cluster that has the lowest Euclidean distance [6]. Each time the clustering process iterates, the centroids are randomized within the range of the dataset, ensuring that all of the possible combinations get explored.…”

“…A geometrical model was developed by Lu et al [6] in which a 6 • 10 4 × 6 • 10 4 m 2 box had been centered is the setup within the 10 4 × 10 4 m 2 test region. The antennae were placed randomly within the box.…”

Radio propagation models for differential GNSS based on dense point clouds

“…Finally, few papers focus on large‐scale maps, where detailed point clouds captured from ground are typically replaced by airborne lidar scans or radar elevation maps. Lu, Han, and Bertoni (2013) show that the terrain‐scattering effect in mountainous areas with sufficiently rough terrain is a significant contributor to the received signal power, apart from the ones which are dominant in flat‐terrain areas. They base their findings on simulations processing data from artificial terrain databases.…”

Microwave Tomography Data Deconstruct of Spatially Diverse C-Band Scatter Components Using Clustering Algorithms

“…The work led to the development of a model that performs significantly better when compared to previously developed 2-D models. 3) A geometrical model was developed by Lu et al [9], and simulations were carried out using different artificial terrain databases. 4) Various model-driven approaches have been proposed…”