基于改进GM(1,1)模型-马尔科夫残差修正的网络性能预测  

Prediction of Network Performance Based on the Improved GM(1,1)Model-the Markov Residual Error Correction Model

作　　者：孙明玮 齐玉东 王晓虹[2] SUN Mingwei;QI Yudong;WANG Xiaohong(Naval Aviation University,Yantai 264001;107th Hospital of the PLA,Yantai 264001)

机构地区：[1]海军航空大学,烟台264001 [2]中国人民解放军第107医院,烟台264001

出　　处：《计算机与数字工程》2020年第4期878-882,894,共6页Computer & Digital Engineering

摘　　要：随着计算机网络的快速发展,网络规模变得极为庞大和复杂,网络性能参数的略微波动就有可能对整个网络服务造成影响,此时对网络性能的准确预测显得格外重要。论文将改进的灰色预测模型应用于网络性能的预测,并采用马尔科夫残差修正模型对预测结果进行修正以提高预测精度。实验结果表明,经马尔科夫残差修正后的预测值与实际值的误差较传统的灰色预测模型有明显减小,结果更接近于实际情况。With the rapid development of computer network,the scales of the network are so vast and sophisticated.The slight fluctuation of network performance parameters may affect the whole network service.At this time,the accurate prediction of network performance is particularly important.This paper applies the Improved Grey Forecast Model to the prediction of network performance,and uses the Markov Residual Error Correction model to correct the predicted values to increase the forecast accuracy.The experimental results show that the error between the actual values and the predicted values corrected by Markov Residuals Error Correction model is significantly reduced compared with the traditional Grey Forecast Model,and the results are closer to the actual situations.

关 键 词：性能预测 改进灰色预测模型 马尔科夫残差修正模型 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]