基于无监督学习的P2P流量识别技术的研究-计算机技术专业论文.docxVIP

基于无监督学习的 基于无监督学习的 P2P 流量识别技术的研究 西华大学硕士学位论文 西华大学硕士学位论文 PAGE PAGE IV Abstract With the development of network technology of P2P,P2P technology is become more widely, and the identification of P2P traffic has been pursued by P2P technology researchers. As more and more applications, the identification of P2P traffic has become increasingly difficult. This paper introduces the P2P technology and analyzes several typical P2P traffic identification technologies. From the advantages and disadvantages of these technologies, this paper presents an improved algorithm; the improved algorithm is based on unsupervised learning of a kind of clustering algorithm. This paper from the data packet level and data flow level analyzes the statistical characteristics of P2P traffic, which the stream is selected average variance of packet size, the duration of flow, the flow of the packet size conversion ratio, the data flow of the average number of bytes in the packet, and download and upload speeds ratio. Secondly, the paper briefly describes the advantages and disadvantages of K-means algorithm and DBSCAN algorithm, and then to be improved (DBK algorithm). Besides the finding the process of initial point of the algorithm, this paper adds the BIC which obtain BIC core points as the initial node, and then through the K-means clustering algorithm. Finally, this algorithm comparing with K-means algorithm and DBSCAN algorithm for the experiment. The concludes were carried out from the accuracy and false positive rate. From these concludes：DBK algorithm running time is long but it is relative to the other two algorithms of CRT visits and its average accuracy is better, and the misjudgment rate is relatively low. From these ,we can illustrate that the algorithm has a better accuracy rate and low false positive rate, so as to arrive an improved algorithm of this paper, and this algorithm is effectively and feasibility. Keywords: P2P; K-means algorithm; DBSCAN algorithm; DBK algorithm; Baye