On designing QoS for congestion control service using neural network predictive techniques

With the ever-increasing data transmission appli- cations recently, considerable efforts have been focused on the design of congestion control scheme for data transmission service to guarantee the quality of service (QoS). The main difficulty in designing an efficient congestion controller for data transmission service stems from the heterogeneous receivers, especially those with large propagation delays in data transfer, which also mean the feedbacks arriving at the source are somewhat outdated and can be harmful to the control actions. This usually leads to a mismatch between the network resources and the amount of admitted traffic. To attack this problem, the present paper describes a novel congestion control scheme that is based on a Back Propagation (BP) neural network technique. We consider a general computer communication model with multiple sources and one destination node. This network-assisted property is different from the existed control scheme in that the data source can predict the dynamic of buffer occupancy of the bottleneck node for which the back control packets experience very long propagation delay and probably cause irresponsiveness of a data flow. This active scheme makes the control more responsive to the network status. Thus the rate adaptation can be in a timely manner for the sender to react to network congestion quickly. We analyze the theoretical aspects of the proposed algorithm, show how the control mechanism can be used to design a controller to support the data transmission based on feedback of explicit rates' (ER's), and verify this agreement by the simulations. Simulation results show the efficiency of our scheme in terms of quickly response and excellent predictive accuracy.