Geometry-Based Stochastic Line-of-Sight Probability Model for A2G Channels Under Urban Scenarios

Qiuming Zhu, Fei Bai, Minghui Pang, Jie Li, Weizhi Zhong, Xiaohong Chen, Kai MaoIEEE Transactions on Antennas and Propagation 2022 · DOI · arXiv

Abstract

Line-of-sight (LoS) path is essential for the reliability of air-to-ground (A2G) communications, but the existence of the LoS path is difficult to predict due to random obstacles on the ground. Based on the statistical geographic information and Fresnel clearance zone, a general stochastic LoS probability model for 3-D A2G channels under urban scenarios is developed. By considering the factors, that is, building height distribution, building width, building space, carrier frequency, and transceiver’s heights, the proposed model is suitable for different frequencies and altitudes. Moreover, in order to get a closed-form expression and reduce the computational complexity, an approximate parametric model is also built with the machine-learning (ML) method to estimate the model parameters. The simulation results show that the proposed model has good consistency with existing models at low altitude. When the altitude increases, it has a better performance by comparing with that of the ray-tracing (RT) Monte-Carlo simulation data. The analytical results of the proposed model are helpful for the channel modeling and performance analyses such as cell coverage, outage probability, and bit error rate in A2G communications.

Introduction

In the evolving landscape of 5G and 6G networks, Unmanned Aerial Vehicles (UAVs) are increasingly utilized as critical communication nodes. A fundamental challenge in Air-to-Ground (A2G) links is the reliable prediction of Line-of-Sight (LoS) paths amidst complex urban obstacles. Existing empirical models, such as those from 3GPP or ITU-R, often fail at higher altitudes as they are primarily derived from ground-level data. Conversely, deterministic ray-tracing (RT) models, while accurate, are computationally expensive and require detailed 3D maps. This paper proposes a geometry-based stochastic model that leverages statistical urban information to predict LoS probability efficiently across various altitudes and frequencies.

Proposed Modeling Approach

The researchers developed a two-tiered modeling strategy to bridge the gap between empirical simplicity and deterministic accuracy.

The Analytical Model

Unlike traditional models that treat LoS as a simple optical line, this approach incorporates the Fresnel clearance zone. This elliptical volume around the direct path represents the region where the majority of radio wave energy propagates. If a building encroaches upon this zone, the signal degrades even without direct intersection of the center line. The model utilizes Rayleigh distributions to represent building heights and accounts for building width and spacing.

Figure 1: Geometry of the A2G communication scenario showing the Fresnel clearance zone between the UAV transmitter (TX) and ground receiver (RX).

Machine Learning Approximation

To reduce the mathematical complexity of the analytical model for practical engineering use, a Back Propagation Neural Network (BPNN) was employed. The BPNN "learned" the behavior of the complex analytical model to generate a simplified, closed-form parametric equation. This allows for near-instantaneous LoS probability calculations using basic parameters like transceiver height and distance.

Key Insights and Validation

The model was validated using extensive Ray-Tracing Monte-Carlo simulations across virtual urban scenes (Suburban, Urban, Dense Urban, and High-rise). Several critical insights emerged:

  • Frequency Dependence: Lower frequencies, characterized by larger Fresnel zones, are more susceptible to blockage than higher frequencies.
  • Impact of Building Width: Wider buildings significantly increase the likelihood of blockage, particularly at lower elevation angles—a factor often overlooked in simpler models.
  • Altitude Advantage: As UAV altitude increases, the "breakpoint" distance (where LoS probability begins to drop from 100%) extends significantly.

Results and Performance

The proposed model demonstrates high consistency with standard 3GPP/ITU models at low altitudes but offers superior accuracy at higher altitudes when compared to ray-tracing data.

Figure 2: LoS probability versus distance for different transmitter heights ($h_{TX}$), illustrating the increase in Maximum Communication Distance (MCD) as altitude rises.

As shown in Figure 2, to maintain a 60% LoS probability at 6 GHz, the Maximum Communication Distance (MCD) increases from approximately 70.2 m at a 30 m height to over 515.4 m at a 1500 m height. The ML-based approximation achieved a Mean Square Error (MSE) of less than 0.03, confirming its utility as a reliable surrogate for the complex analytical physics.

Conclusion

This geometry-based stochastic model provides a robust framework for analyzing A2G channel performance. By accounting for Fresnel zones and building dimensions, it offers more realistic predictions for high-altitude UAV communications than existing standards. Future work will aim to incorporate antenna radiation patterns and validate the model through extensive real-world field measurements in diverse urban environments.

Summary updated

Abstract

Line-of-sight (LoS) path is essential for the reliability of air-to-ground (A2G) communications, but the existence of the LoS path is difficult to predict due to random obstacles on the ground. Based on the statistical geographic information and Fresnel clearance zone, a general stochastic LoS probability model for 3-D A2G channels under urban scenarios is developed. By considering the factors, that is, building height distribution, building width, building space, carrier frequency, and transceiver’s heights, the proposed model is suitable for different frequencies and altitudes. Moreover, in order to get a closed-form expression and reduce the computational complexity, an approximate parametric model is also built with the machine-learning (ML) method to estimate the model parameters. The simulation results show that the proposed model has good consistency with existing models at low altitude. When the altitude increases, it has a better performance by comparing with that of the ray-tracing (RT) Monte-Carlo simulation data. The analytical results of the proposed model are helpful for the channel modeling and performance analyses such as cell coverage, outage probability, and bit error rate in A2G communications.
BibTeX
@article{zhu2022geometry,
  author = {Qiuming Zhu and Fei Bai and Minghui Pang and Jie Li and Weizhi Zhong and Xiaohong Chen and Kai Mao},
  title = {{Geometry-Based Stochastic Line-of-Sight Probability Model for A2G Channels Under Urban Scenarios}},
  journal = {IEEE Transactions on Antennas and Propagation},
  year = {2022},
  doi = {10.1109/tap.2022.3161277},
}
Figure