MULTIPASS: IoT scheduling over a multilayer NTN architecture

Abstract The 6G vision promotes the convergence of terrestrial and non-terrestrial networks (NTNs) to enable seamless internet of things (IoT) connectivity, particularly in remote and infrastructure-limited regions. However, direct satellite–IoT communication remains constrained by the low transmission power of IoT devices, intermittent satellite visibility, Doppler-induced signal distortion, and high collision probability, due to the large number of IoT devices, within the satellite coverage. To address these challenges, this work investigates a three-layer IoT–NTN architecture that introduces an intermediate layer composed of unmanned aerial vehicles (UAVs). The latter act as relay, and they forward the data collected data from the IoT devices to low-Earth-orbit (LEO) satellites through a data collection mechanism, mitigating Doppler effects and improving link reliability. Building on this architecture, we propose MULTIPASS, a novel scheduling method designed to coordinate packet transmissions across terrestrial, aerial, and satellite layers under different UAV flight scenarios, to avoid packet collisions, and thus, saving energy, while improving efficiency. Several UAV flight scenarios are proposed, namely sequential (S), adjacent (A), and direct relay (D) modes, differing in their data collection timing and satellite visibility coordination. Simulation results, considering LoRaWAN technology and sequential fly mode, show that MULTIPASS achieves up to 6 times higher data collection than direct satellite communications and 20 times greater energy efficiency than regular periodic (Aloha-based) transmission, while significantly reducing packet collisions.