欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
网络流量整形、调度、异常检测、管理与控制及保障QoS需求等都需要了解业务流的局部变化特性．本文给出离散小波及其模极大值的网络流量奇异谱估计算法及影响因素，并通过真实的网络业务数据对算法进行了评估和比较．实验结果表明，两种方法的奇异谱估计能有效刻画网络业务流的局部变化特征，并且能通过奇异谱特征参数之间的差别描述不同业务流之间的差异性，也表明了在一定条件下，离散小波模极大法更加优越．
英文摘要：
The analysis of network traffics plays a significant role in many aspects of network engineering, such as network traffic shaping, scheduling, intrusion detection, traffic monitoring, accounting, quality of service （QoS） guarantee, etc. In this paper, we present the singularity spectrum estimation based on discrete wavelet transform and discrete wavelet transform modulus maxima technology including the principle, procedure and condition parameters. We apply them to real network in order to demonstrate the capability of two methods on different network traffics. The experimental results show that both of them are efficient in studying the singularities of the network traffics. And the discrete wavelet transform modulus maxima is more accurate and efficient in detecting singularities of different network traffics by acquiring more different characteristic parameter of the singularity spectrum.
关于余顺争：
同期刊论文项目
同项目期刊论文
Generating regular expression signatures for network traffic classification in trusted network manag
Modeling Oscillation Behavior of Network Traffic by Nested Hidden Markov Model with Variable State-D
位置：