🤖 AI Summary
This study addresses the challenge of achieving ubiquitous communication coverage in low-altitude airspace, where neither standalone satellite nor terrestrial 5G networks suffice. To this end, the paper proposes a partially integrated satellite–5G cooperative networking architecture, featuring an innovative location-aware adaptive synchronization mechanism and a joint time–frequency–spectrum sharing framework. Relying solely on large-scale channel state information, the approach significantly reduces system overhead and complexity under coarse synchronization conditions. Furthermore, by integrating link-characteristic clustering with a divide-and-conquer optimization strategy, the solution effectively circumvents NP-hard computational bottlenecks. The proposed scheme not only ensures efficient communication coverage for low-altitude aerial vehicles but also offers a practical pathway toward the evolution of integrated space–terrestrial networks.
📝 Abstract
Driven by both technological development and practical demands, the low-altitude economy relying on low-altitude aircrafts (LAAs) is booming. However, neither satellites nor terrestrial fifth-generation (5G) networks alone can effectively satisfy the communication requirements for ubiquitous lowaltitude coverage. While full integration of satellites and 5G networks offers theoretical benefits, the associated overhead and complexity pose significant challenges for rapid deployment. As a more economical and immediately viable alternative, this paper investigates partially-integrated networks where satellites and 5G systems operate with coarse synchronization yet achieve coordinated spectrum sharing, pooling their capabilities to jointly serve LAAs. Leveraging the inherent position-awareness of LAAs, we propose a framework for joint time-frequency spectrum sharing with an adaptive synchronization time scale, where only large-scale channel state information (CSI) is required. To avoid solving the NP-hard optimization problem directly, link-feature-aided clustering is employed following a divide-andconquer strategy. The proposed framework achieves substantial performance gains with low overhead and complexity, enabling swift advancement of low-altitude applications while paving the way for future integrated satellite-terrestrial network evolution.