摘要
实际网络流量具有明显的分形特性,流量的多重分形特性对网络性能有着非常重要的影响,建立一个基于多重分形特性的实际流量的合成业务模型很有必要。利用小波变换能够去除实际流量相关性的特点,结合实际研究的过程中所发现的不同类型的流量具有不同的分形特性的结论,建立新的合成模型,提高合成流量的精度。同时,通过对合成后的流量进行尺度刻画和性能评价,验证了新模型的正确性。
Fractal is a ubiquitous property of real-traffic, and the muhi-fractal of traffic has great impact on network performances. Therefore, it is important to build the reconstructed model based on muhi-fractal for the real network traffic. In this paper, a new reconstruction model based on wavelet transform that could release the correlation of traffic was build in combination with the conclusion that different traffic had different fractal characteristics in the past research, and more accuracy was got. Then, through depicting the new traffic and performance evaluation, the correctness of the new model is certificated.
出处
《计算机应用》
CSCD
北大核心
2008年第3期586-587,590,共3页
journal of Computer Applications
基金
国家自然科学资金资助项目(60572143)
西南交通大学科学研究基金资助项目(2005A03)
关键词
多重分形
小波
合成
性能评价
muhi-fractal
wavelet
reconstruction
performance evaluation
