JoWUA

Abstract

Adaptive load balancing in heterogeneous wireless mobile networks is crucial for optimizing resource utilization, ensuring quality of service (QoS), and adapting to changing traffic patterns and diverse network infrastructures. In this research, we present a comprehensive, adaptive load-balancing framework that intelligently distributes traffic over disparate access technologies, including LTE, 5G, and Wi-Fi, in real-time. We demonstrate how the integration of context-aware decision making, mobility prediction, and machine learning techniques allows our approach to dynamically assess the network's conditions, user behavior, and service demand, thereby optimizing resource allocation. Our model considers key performance metrics, including latency, bandwidth, energy consumption, and handover frequency, to ensure minimal user service interruption and seamless connectivity. Through simulation-based results, we have shown that improving throughput by considering the above metrics also minimizes the number of lost packets and enhances user experience over static and threshold-based load balancing schemes. The results demonstrate the importance of adaptations in network management, particularly when a combination of user mobility, fluctuating loads, and heterogeneous network infrastructure is encountered.