论文部分内容阅读
针对单个主机单个协议流量的自相似性和非宏观上流量的自相似性,分析了端到端(P2P)网络流量的自相似性.对常见的端到端(P2P)应用进行分析后发现,其应用层数据存在自相似性,且在时间尺度与行为尺度的比较中,P2P应用层流量在行为尺度上的自相似性表现得更加明显和稳定.为了将行为尺度上的自相似性应用到业务感知领域,提出了一种新的P2P流量识别算法,该算法通过计算网络流量不同行为尺度下的容量维,再辅以主动系数来识别P2P流量.实验结果证明,新算法在P2P流量识别方面的准确率高于同类算法,在加密流量的识别上表现尤为突出.
Aiming at the self-similarity of single protocol traffic and the self-similarity of non-macro traffic, this paper analyzes the self-similarity of P2P traffic.After analyzing the common P2P applications, The self-similarity of application-layer data and the comparison of time scale and behavior scale show that the self-similarity of P2P application layer traffic behavior is more obvious and stable in behavioral scale.In order to apply the behavior-scale self-similarity to In the field of business perception, a new P2P traffic identification algorithm is proposed, which identifies the P2P traffic by computing the capacity dimension under different behavioral scales of network traffic, and then with the active coefficients. The experimental results show that the new algorithm has good performance in P2P traffic identification The accuracy of the algorithm is higher than that of similar algorithms, especially on the recognition of encrypted traffic.