Efficient Analytics and Fuzzy Call Admission Control With Adaptive Scheduling for Enhanced Quality of Service in 6G IoT for Multimedia Streaming

ABSTRACT The increasing demand for stringent quality of service (QoS) guarantees and low latency in mission and time‐critical 6G internet of things (IoT) applications necessitates advanced call admission control (CAC) mechanisms. Although 6G networks offer enhanced capabilities, resource limitations like bandwidth and processing power persist. Consequently, efficient resource allocation strategies are crucial to balancing the needs of diverse services, particularly for real‐time streaming applications. This paper introduces a novel Analytics & Fuzzy Call Admission Control (AF‐CAC) algorithm combined with an Adaptive Leaky Bucket (ALB) scheduling mechanism. The AF‐CAC algorithm integrates predictive analytics and fuzzy logic to make informed, intelligent admission decisions, ensuring reliable communication and optimal resource utilization. It prioritizes critical data transmission and prevents network overloading by controlling the number of admitted calls. Concurrently, the ALB mechanism dynamically adapts to changing network conditions, user mobility, and traffic patterns, efficiently allocating resources to meet diverse QoS requirements. The proactive resource allocation is enhanced with a convolutional neural network (CNN) and reinforcement learning (RL) agents, enabling dynamic and efficient resource management to ensure high‐quality multimedia streaming, support mission‐critical applications, and maintain uninterrupted service during mobility. The average throughput for the AF‐CAC technique is 1350 packets/slot, and the average delay over the range of 300 simulated devices is 840 ms. Hence, it exhibits significant enhancements in admitting connections and overall QoS compared to existing approaches for managing multimedia traffic in 6G networks.