Statistical Channel Model for FSO Systems Assisted by a UAV-Mounted IRS

Integrating optical intelligent reflecting surfaces (IRSs) into aerial platforms, such as unmanned aerial vehicles (UAVs), has been proposed to relax the line-of-sight (LoS) constraint, extend coverage, and enhance deployment flexibility of free space optical (FSO) systems. However, misalignment errors induced by the UAV hovering, in both position and orientation, may degrade connectivity and impair the end-to-end channel quality. In this paper, we develop novel expressions for the electric fields incident on and reflected by an optical IRS, based on the Huygens-Fresnel principle. The resulting expressions are applicable for any combination of incident and reflected propagation directions. Building on this framework, we derive a closed-form statistical channel model that captures the geometric and misalignment losses of an FSO link assisted by a UAV-mounted IRS in the presence of random UAV fluctuations. In particular, we develop a statistical model for the beam misalignment at the receiver lens, assuming Gaussian fluctuations in both the UAV position and orientation. The proposed analytical model is validated through Monte Carlo (MC) simulations and is further used to provide practical design guidelines regarding the optimal placement of the UAV-mounted IRS for the minimization of the outage probability.

Machine Learning-based Path Loss Prediction in Suburban Environment in the Sub-6 GHz Band

Accurate path loss (PL) prediction is crucial for successful network planning, antenna design, and performance optimization in wireless communication systems. Several conventional approaches for PL prediction have been adopted, but they have been demonstrated to lack flexibility and accuracy. In this work, we investigate the effectiveness of Machine Learning (ML) models in predicting PL, particularly for the sub-6 GHz band in a suburban campus of King Abdullah University of Science and Technology (KAUST). For training purposes, we generate synthetic datasets using the ray-tracing simulation technique. The feasibility and accuracy of the ML-based PL models are verified and validated using both synthetic and measurement datasets. The random forest regression (RFR) and the K-nearest neighbors (KNN) algorithms provide the best PL prediction accuracy compared to other ML models. In addition, we compare the performance of the developed ML-based PL models with the traditional propagation models, including COST-231 Hata, Longley-Rice, and Close-in models. The results show the superiority of the ML-based PL models compared to conventional models. Therefore, the ML approach using the ray-tracing technique can provide a promising and cost-effective solution for predicting and modeling radio wave propagation in various scenarios in a flexible manner.