AI-DRIVEN CROSS-LAYER OPTIMIZATION OF WI-FI 7 AND 5G FWA HYBRID BROADBAND SYSTEMS

Abstract

The convergence of Wi-Fi 7 and 5G Fixed Wireless Access (FWA) technologies offers unprecedented opportunities for hybrid broadband systems that leverage complementary strengths of both access technologies. However, realizing this potential requires intelligent orchestration across multiple protocol layers and radio interfaces. This research presents an AI-driven cross-layer optimization framework that coordinates physical layer parameters, MAC scheduling, and application-level QoS to maximize performance in hybrid Wi-Fi 7/5G FWA deployments. Our system employs machine learning models for traffic prediction, intelligent failover decision-making, and adaptive routing that considers latency, bandwidth, and reliability requirements across heterogeneous access technologies. The framework implements dynamic load balancing, application-aware traffic steering, and predictive maintenance capabilities that anticipate connectivity issues before they impact users. Experimental validation across residential and enterprise scenarios demonstrates 43% improvement in aggregate throughput, 58% reduction in tail latency for critical applications, and 91% reduction in service disruptions during access link failures. The proposed architecture addresses fundamental challenges in multi-radio environments including asymmetric link characteristics, varying propagation conditions, and diverse application requirements, providing a comprehensive solution for next-generation hybrid broadband systems.