Provision of Creative QoS for Adaptive Multimedia over Wireless Networks

In the future, wireless networks will provide adaptive multimedia by varying the capacity of each continuous flow in response to variations in traffic load. Quality-of-service (QoS) must be provided in this adaptive multimedia system by combining a bandwidth adaptation algorithm with a call admission control mechanism. This study suggests a novel method for QoS provisioning that can maximise network revenue under a variety of specified QoS constraints. It uses average reward reinforcement learning and stochastic approximation. Because our technique does not require explicit state transition probabilities, unlike other model-based algorithms (like linear programming), the assumptions underlying the underlying system model are more realistic than they were in earlier schemes. When we consider the state of neighbouring cells, the suggested technique is also capable of dynamic adaptation to changes in traffic conditions. Additionally, the strategy effectively controls the bandwidth adaptation frequency by accounting for the cost of bandwidth change in the model. Using simulated outcomes, adaptive multimedia wireless networks demonstrate the value of the suggested strategy.