论文部分内容阅读
针对FlashP2P技术,对其RTMFP协议进行了深入分析,提出了一种基于RTMFP包检测的FlashP2P流量识别算法,并采用该算法对国内主流视频网站的FlashP2P流进行了有效的识别。在此基础上,对FlashP2P流量特征进行分析并证明其具有自相似性。最后,提出了一种基于ARIMA模型的经验模式分解预测自相似网络流量的方法,而且进行了仿真验证。结果表明,该模型不仅降低了算法的复杂度,并且对短期预测精度较高。
According to FlashP2P technology, the RTMFP protocol is analyzed in depth. A FlashP2P traffic identification algorithm based on RTMFP packet inspection is proposed and used to identify the FlashP2P stream in domestic mainstream video websites. Based on this, the FlashP2P traffic characteristics are analyzed and proved to be self-similar. Finally, a method based on the ARIMA model empirical mode decomposition to predict the self-similar network traffic is proposed, and the simulation is carried out. The results show that this model not only reduces the complexity of the algorithm, but also has higher accuracy for short-term prediction.