摘要
针对无线传感器网络中汇聚节点的拥塞,设计一种基于灰色预估神经网络控制队列的控制器。利用RBF神经网络的自学习能力解决网络实时变化时算法参数的在线整定问题,并利用灰色GM(1,1)预测器有效地解决了在汇聚节点易发的大时滞对网络性能的影响。仿真结果表明该算法具有较好的鲁棒性。
For sink node congestion in wireless sensor network,in the thesis we design a controller which is based on gray prediction neural network＇s controlling queue.The controller uses self-learning ability of RBF neural network to cope with the online setting issue of algorithm parameters in time-varying network.By using gray GM（1,1） predictor,the impact on network performance by incidental large delays in large cluster nodes has been effectively overcome.Simulation results show that this algorithm has better robustness.
出处
《计算机应用与软件》
CSCD
北大核心
2012年第11期272-274,301,共4页
Computer Applications and Software
