文献类型：期刊文献

中文题名：用于多媒体网络节点的故障检测仿真

英文题名：A Fault Detection of Simulation for Multimedia Network Nodes

作者：崔军立[1]

第一作者：崔军立

机构：[1]北京联合大学商务学院

第一机构：北京联合大学商务学院

年份：2016

卷号：0

期号：8

起止页码：421-424

中文期刊名：计算机仿真

外文期刊名：Computer Simulation

收录：CSTPCD;;北大核心:【北大核心2014】；CSCD:【CSCD_E2015_2016】；

语种：中文

中文关键词：多媒体网络;故障检测;心跳策略

外文关键词：multimedia networks; fault detection; heartbeat strategies

摘要：对多媒体网络节点的故障进行准确检测,可以有效消除多媒体网络节点故障。在进行多媒体网络节点的故障检测时,需要对多媒体网络节点的状态变化进行动态预测,而传统神经网络的检测方法只能获取多媒体网络节点的静态状态参数,不能实时的对节点状态进行更新与预测,降低了多媒体网络节点故障检测精度。提出一种改进动态灰色预测算法的多媒体网络节点故障检测方法。建立了多媒体网络节点的故障检测模型,融合于灰色预测方法将心跳策略引入到体网络节点的故障检测中,动态地预测出多媒体网络节点的状态变化。利用模糊算法对多媒体网络节点故障信号延迟进行模糊化处理,计算出网络节点的故障观测参数和其参数残差,完成对多媒体网络节点的故障检测。仿真结果证明,改进方法能够有效地提高多媒体网络节点故障检测的准确率。
Exact fault detection can effectively eliminate the fault of multimedia network nodes. It needs to predict the status＇ changes of nodes dynamically during fault detection. Traditional detection method based on neural network can only acquire the static state parameter of nodes. It cannot update and predict the node state real-timely. So it reduces the fault detection accuracy. In this paper,we propose a detection method based on the modified dynamic gray prediction algorithm. We built a fault detection model to predict the status＇ changes of the nodes dynamically.The model introduces the heartbeat strategy to the fault detection of body network nodes by the gray prediction algorithm. We process the fault signal delay fuzzily via the fuzzy algorithm. Then we calculate the fault observation parameter and its residual error to finish the fault detection. Simulation result shows that the proposed method can improve the accuracy rate of fault detection for multimedia network nodes.