广义神经网络在数据流量预测中的应用仿真  

Application Simulation of Generalized Neural Network in Data Traffic Prediction

作　　者：姚迎乐 冯乃勤 YAO Ying-le;FENG Nai-qin(Software Department,Institute of Zhengzhou Industrial Application Technology,Xinzheng Henan 451150,China;Department of Computer and Information Engineering,Henan Normal University,Xinxiang Henan 453000,China)

机构地区：[1]郑州工业应用技术学院软件学院,河南新郑451150 [2]河南师范大学计算机与信息工程学院,河南新乡453000

出　　处：《计算机仿真》2024年第4期382-386,共5页Computer Simulation

基　　金：2021年第一批产学合作协同育人项目(202101101016)2021省级教改项目:“新工科”背景下产学研多方合作大数据应用型人才培养研究与实践,项目编号:2021SJGLX616教育部产学研协同育人项目:基于云仿真平台Simdroid的“双师型”师资培训:项目编号:220906517024221。

摘　　要：在超大数据量的流量预测中,数据易出现缺陷、错误或不完整问题。广义神经网络因其对数据处理具有较强的鲁棒性和容错能力,因此研究广义神经网络在数据流量预测中的应用,并对应用效果完成验证。将数据流量预测作为研究目标,分析数据流量特征,获取数据流量时空特征和空间维度两者对应的空间相关性特征。选择和被测网络相关性最大的数据流量作为广义神经网络的输入,构建基于广义神经网络的数据流量预测模型。为验证广义神经网络的应用效果,设计对比测试实验。结果表明,广义神经网络在数据流量预测中具有可行性,且算法应用下数据流量预测误差更小。In the traffic prediction of massive data,data are prone to defects,error or incomplete problems.And the generalized neural network has strong robustness and fault-tolerant ability for processing data.In this paper,data traffic prediction was taken as the research objective,and the characteristics of data traffic were analyzed to obtain the spatial characteristics corresponding to the spatial-temporal characteristics and spatial dimensions of data traffic.Moreover,the data flow that correlated the most with the tested network was selected as the input of the generalized neural network.Finally,a model of predicting data flow was built based on the generalized neural network.In order to verify the application effect of generalized neural network,a comparative test was designed.The simulation results show that the generalized neural network is feasible in predicting data traffic,with lower prediction error of data traffic.

关 键 词：广义神经网络 数据流量 时空特征 空间维度 流量预测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]