Abstract: This paper introduces a semi-Markov model to character Internet traffic, which is applied to traffic prediction. Some practical trace data with different time scales from WAN and LAN is employed to verify the model and the prediction method respectively. The prediction accuracy of all data is smaller than the confidence set beforehand. The result shows that the model is well matched to real network traffic and the prediction method based on the model has good performance.

Key words: Network traffic model, Traffic prediction, Semi-Markov model

摘要： 引入了半马尔柯夫模型描述网络流量特性，并在该模型的基础上分析推导了相应的流量预测方法。分别对广域网和局域网不同时间尺度统计的实际流量数据进行分析和短期、长期预测，所有数据的实际预测精确度均小于预先设定的置信度。说明引入的模型能真实反映网络流量特性，基于该模型的流量预测方法具有良好的预测性能且适用于不同长度的预测。

关键词: 网络流量模型, 流量预测, 半马尔柯夫模型

CLC Number: 

• TP393