Resiliensi, Aliran Data, dan Survivabilitas Jaringan: Optimalisasi Komunikasi Radio dalam Operasi Militer Modern

Abstract

Modern military operations heavily rely on seamless real-time data exchange, yet dynamic electronic warfare activities frequently disrupt the stability of tactical communication links on the battlefield. Although conventional frequency-hopping technologies are widely deployed, their effectiveness in maintaining optimal data flow under dense, adaptive jamming threats remains severely limited due to processing constraints and overhead. Adopting a Research and Development (R&D) methodology to overcome this vulnerability, this study develops a novel predictive optimization framework that integrates dynamic spectrum allocation with intelligent, Deep Reinforcement Learning (DRL)-driven routing algorithms. This R&D model was comprehensively evaluated using a high-fidelity network simulator across L, S, and X-band frequencies under extreme electronic countermeasure scenarios. The simulation results demonstrate that this framework successfully increased data flow throughput by 42% and reduced latency by 35% during active jamming sequences compared to legacy AODV-based MANET protocols, while maintaining 88% core network connectivity despite losing 30% of physical nodes—specifically manpack radios, vehicular communication hubs, and UAV airborne relays—to enemy action. These findings prove that integrating predictive routing significantly enhances the resilience and survivability of military networks. This innovative approach offers a scalable, adaptive blueprint for future defense architectures, shifting the operator's role from manual frequency management to automated system monitoring, thereby reducing cognitive load while ensuring continuous command and control superiority.