基于网络信息流的自适应 MWM 模型研究
On Adaptive MWM Model Based on Network Information Flow
-
摘要: 多重分形小波模型（MWM ）可以用于网络流量非严格自相似性的分析，但现有MWM及改进模型大多不能保证各个尺度下尺度系数的边缘分布都能拟合业务流数据，当边缘分布与源数据的统计特性有较大差异时，得到的排队分析结果也将出现较大差异。针对传统多重分形小波模型存在的缺陷，在分析实际视频流量多分辨率性能的基础上，提出了参数可调的自适应多重分形小波模型。该模型的改进之处在于它能够根据实际情况实时地调整小波参数和乘性系数，更好地适应实际网络流量的特性。对新模型的仿真流进行了尺度函数和多重分形谱的分析，结果证明了新模型的准确性。Abstract: After pointing out the limitation of the traditional multifractal wavelet model (MWM ) ,and ana‐lyzing multi‐resolution characteristics of real video traffic ,this paper has proposed an adjustable scale coef‐ficient MWM model .The improved MWM can adjust wavelet coefficients and multiplicative coefficients based on their distribution of the real network traffic ,which can suit the real network characteristics .Sub‐sequently the scale function and multifractal spectrum of the new model are analyzed . The conclusion proves the new model's accuracy .
-
-
计量
- 文章访问数: 787
- HTML全文浏览数: 407
- PDF下载数: 0
- 施引文献: 0
下载: