Implementation of Rate Control in Distributed Wireless Multicast by Neural Network Prediction

Recently considerable efforts have focused on the design of self-adaptive flow control schemes for wireless multicast service. This attention is significantly necessary due to the large scale heterogeneous wireless multicast receivers, especially those with large propagation delays, which means the feedbacks arriving at the source node are somewhat outdated and harmful to the control actions.To solve the above problems, the paper describes a novel, autonomous, and predictive wireless multicast flow control scheme, the so-called proportional, integrative plus neural network (PINN) predictive technique. The final sending rate of the multicast source is the expected receiving rates computed by PI controller based on the consolidated feedback information. The link bandwidth is fairly shared among multiple multicast sessions from different sources, and also shared between multicast flow and CBR flow. They analyze the theoretical aspects of the proposed algorithm, simply show how the control mechanism can be used to design a controller. In this paper, we describe more details on how this controller can support wireless multi-rate multicast transmission based on feedback of explicit rates, and give relevant simulation analysis. Simulation results demonstrate that this PINN scheme optimizes the QoS of wireless multicast networks in terms of fast response, scalability, intra-session fairness, inter-session fairness, and stability of buffer occupancy. Thus, the presented scheme makes the wireless multicast system achieve reliable performance and scalable application in the large scale heterogeneous wireless multicast system.