Akademik Veri Yönetim Sistemi
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING, cilt.50, sa.24, ss.1-15, 2025 (SCI-Expanded, Scopus)
This paper introduces a novel multi-objective optimization framework for dynamic load balancing in large-scale Very Low Earth Orbit (VLEO) satellite networks, addressing critical challenges in next-generation Non-Terrestrial Networks (NTNs). A graph-theoretic Maximum Clique-based Scheduling (MC) algorithm with polynomial-time approximation is proposed alongside a Load-Aware Assignment (LA) heuristic, rigorously comparing them against conventional maximum elevation and random assignment strategies. The key contributions include: (1) A high-fidelity simulation of a 1,584-satellite Walker-Delta constellation serving 10,000 uniformly distributed users across Europe, incorporating orbital mechanics, ITU-R compliant Ka-band propagation with spatial rain correlation, and terrain-aware path loss; (2) a novel visibility graph formulation with approximate maximum clique detection that exploits spatial correlations among satellites to maximize resource utilization while minimizing handovers; and (3) a comprehensive 24-hour performance evaluation at 5-minute intervals, quantifying trade offs between throughput (1.37 Gbps), fairness (Jain’s index = 0.92), and satellite utilization (82.3%) including computational complexity analysis. Simulation results demonstrate that the proposed LA algorithm achieves 18.7% higher throughput and 40% better resource efficiency than traditional methods, with a 44.7% reduction in handovers. The approximate MC approach achieves 85% of optimal performance with polynomial complexity, demonstrating practical feasibility for real-time systems. This work bridges the gap between theoretical scalability and practical deployability, offering actionable insights for mega-constellation operators like Starlink and OneWeb. Furthermore, the work also includes analysis of Inter-Satellite Links integration potential, demonstrating 30% additional handover reduction and 9% service availability improvement.