In this work, we propose a non-terrestrial vehicle communication network where the communication between the satellite and the terrestrial source is assisted by a unmanned aerial vehicle (UAV) used as a relay. In particular, a reconfigurable intelligent surface (RIS) is used in the RF channel to reflect the signals of the terrestrial user to the relay with a fixed amplification gain, while free-space optical (FSO) is applied to the relay-satellite link to obtain a high speed transmission. For such a dual-hop system, assuming that the FSO channel experiences $\mathcal {M}$ -distributed fading with pointing errors, the expressions for the outage probability (OP) and average bit error rate (ABER) are evaluated in closed-forms. In addition, the high signal-to-noise ratio (SNR) analyses for the OP and ABER are developed and the lower and upper bounds on the average channel capacity (ACC) are calculated to obtain further insights. Results show that the diversity order of the proposed system is $\min \lbrace k_{w},m_{w}\rbrace$ , where these two parameters are related to the Nakagami- m distribution parameters and the number of RIS elements. Finally, we take the shadowing effects into consideration, and it can be seen that the shadowing effects significantly degrade the system performance.

Air-to-ground (A2G) channel modeling supports low-altitude wireless networks for unmanned aerial vehicle (UAV) services in 5G/6G systems. Existing data-driven models are often benchmarked against weak empirical baselines, decoupled from the underlying measurement campaigns, and prone to overfitting on small UAV datasets. This paper presents PACE-A2G, a calibrated deep-tree ensemble on physics-aware features, together with a stricter evaluation methodology. A quadrotor-mounted software-defined-radio measurement system produces five datasets covering 433, 740, 1500, and 3300 MHz in overwater, suburban, and mountainous environments. The deep tabular backbone combines periodic numerical embeddings, feature-group attention, and a parameter-efficient ensembling head; propagation physics enters through the features rather than through the architecture. At inference, the deep estimate is blended with a tree ensemble, and a split-conformal quantile head provides calibrated prediction intervals. Evaluation under random, chronological, and spatial-block protocols shows that random partitions overestimate accuracy: the chronological-split root-mean-square error (RMSE) of the strongest tree-ensemble baseline is 1.9 to 8 times its random-split value. In distribution, the hybrid predictor is statistically on par with the strongest tree ensembles. Across nine cross-scenario and cross-frequency protocols, it transfers more robustly than the tree ensemble on balance, with a mean paired RMSE reduction of 0.34 dB, although the fitted log-distance baseline remains the strongest single predictor under shift. Split-conformal calibration reaches near-nominal 90% coverage in distribution but collapses under domain shift; a feature-space out-of-distribution gate detects the shift on every source–target pair, and ten labelled records of the new domain restore near-nominal coverage.

This paper investigates a measurement-driven rotary-wing unmanned aerial vehicle (UAV)-enabled wireless communication system serving multiple ground users. Unlike conventional studies relying on simplified theoretical channel models, we derive the air-to-ground (A2G) path loss and shadow-fading models from software-defined radio (SDR) channel measurements at 3.3 GHz. To maximize the minimum average user rate, we formulate a joint optimization of three-dimensional (3D) trajectory, velocity, transmit power, and bandwidth, subject to quality-of-service (QoS) and UAV kinematic constraints. Owing to the strong coupling and non-convexity, we reformulate it as a constrained Markov decision process (CMDP) with a continuous-action space and solve it via proximal policy optimization (PPO), where a constraint-preserving action parameterization enforces power, bandwidth, and motion constraints by construction. Simulations over real digital-elevation-model terrain show that the proposed approach attains 95.7% of a search-based rate reference at $K=3$ and surpasses an offline AO/SCA benchmark in max–min rate at all user densities, with 100% QoS satisfaction at $K=3$, near-zero constraint violations, and $0.142$–$0.181$ ms online inference per decision.

In recent years, the utilization of unmanned aerial vehicles (UAVs) has surged in a variety of applications, including weather monitoring, emergency search and rescue operations, and smart agriculture. In these applications, UAVs are instrumental in executing data collection tasks. However, due to the scarcity of resources such as battery capacity, the duration of work is limited, necessitating the optimization of UAV trajectories and resource allocation. In this paper, we introduce a hierarchical task scheduling and resource allocation scheme for UAV data collection tasks, which incorporates the deep reinforcement learning (DRL) technology into a two-layer training framework. The continuous actions, including UAV trajectories, UAV communication power, and UAV CPU main frequency, are first optimized in the lower layer, while the upper one concentrates on the discrete action space for the target allocation. In particular, the well-trained lower-layer network is integrated into the upper layer, facilitating rapid global reward feedback between the two-layer network and accelerating the training process of the upper-layer network. Experimental results demonstrate that the proposed approach achieves superior performance than the baseline methods in terms of the amount of collected data as well as UAV energy consumption.