网络流量的神经网络直接自适应控制

Neural Network Direct Adaptive Control for Flow Rate in Networks

设计了在线学习的网络流量神经网络直接自适应控制器，其适应能力及高速并行处理能力能很好地满足了网络控制所必需的适应性、鲁棒性和实时性的要求。从而使信源的发送速率能快速响应网络状态的变化，保证了网络处于良好的运行状态。其理论分析得到了仿真计算结果的验证。

The paper designs the neural network direct adaptive control for flow rate in networks, which is self-learning on line. The abilities of adaptation and high-speed parallel processing can meet the requirements of adaptability, robustness and realtime which is necessary to network control. Thus the source rates can respond to the changes of network status rapidly, and the network can be under the good condition at operation. Moreover, the theoretical analysis is verified by simulation.