arXiv subjects

Giovanni Schembra

Publications and source records attributed to Giovanni Schembra.

2 recordsLinked to original sources

arXiv2026

Experimental Evaluation of a UAV-Mounted LEO Satellite Backhaul for Emergency Connectivity

Reliable connectivity is critical for Public Protection and Disaster Relief operations, especially in rural or compromised environments where terrestrial infrastructure is unavailable. In such scenarios, NTNs, and specifically UAVs, are promising candidates to provide on-demand and rapid connectivity on the ground, serving as aerial base stations. In this paper, we implement a setup in which a rotary-wing UAV, equipped with a Starlink Mini terminal, provides Internet connectivity to an emergency ground user in the absence of cellular coverage via LEO satellites. The UAV functions as a Wi-Fi access point, while backhauling the ground traffic through the Starlink constellation. We evaluate the system via both network simulations in ns-3 and real-world flight experiments in a rural environment, in terms of throughput, latency, coverage, and energy consumption under static and dynamic flight conditions. Our results demonstrate that the system can maintain a stable uplink throughput of approximately 30 Mbps up to approximately 200 meters, and with minimal impact on the UAV battery lifetime. These findings demonstrate the feasibility of deploying commercial LEO satellite terminals on UAVs as a practical solution for emergency connectivity.

cs.NI↗

arXiv2024

DRL-based Latency-Aware Network Slicing in O-RAN with Time-Varying SLAs

The Open Radio Access Network (Open RAN) paradigm, and its reference architecture proposed by the O-RAN Alliance, is paving the way toward open, interoperable, observable and truly intelligent cellular networks. Crucial to this evolution is Machine Learning (ML), which will play a pivotal role by providing the necessary tools to realize the vision of self-organizing O-RAN systems. However, to be actionable, ML algorithms need to demonstrate high reliability, effectiveness in delivering high performance, and the ability to adapt to varying network conditions, traffic demands and performance requirements. To address these challenges, in this paper we propose a novel Deep Reinforcement Learning (DRL) agent design for O-RAN applications that can learn control policies under varying Service Level Agreement (SLAs) with heterogeneous minimum performance requirements. We focus on the case of RAN slicing and SLAs specifying maximum tolerable end-to-end latency levels. We use the OpenRAN Gym open-source environment to train a DRL agent that can adapt to varying SLAs and compare it against the state-of-the-art. We show that our agent maintains a low SLA violation rate that is 8.3x and 14.4x lower than approaches based on Deep Q- Learning (DQN) and Q-Learning while consuming respectively 0.3x and 0.6x fewer resources without the need for re-training.

cs.NI↗