【摘要】 大量网络通信量测量研究表明,通信量在长时间尺度上的突发性规律可以用分形或自相似模型来描述。为准确评估CSMA/CD协议性能,基于RMD算法产生近似的FBM,并利用近似FMB构造了自相似通信量模型。基于自相似通信量,采用离散事件仿真技术对CSMA/CD协议建立了排队模型,对网络性能进行了分析,通过与通信量服从泊松到达的仿真结果进行对比,说明了自相似特性对网络性能分析的影响。更多还原

【Abstract】 A number of traffic measurements studies have shown that the burstiness of packet traffic is associated with long-range correlations that can be efficiently described in terms of fractal or self-similar models,e.g.fractional Brownian motion(FBM).In order to estimate performance of CSMA/CD effectively,consider an approximate FBM generation method known as the random midpoint displacement(RMD) algorithm,and generate the cumulative arrival process by the approximate FBM trace which is come from the RMD algorithm.Based on the self-similar traffic modeling,a CSMA/CD protocol simulation system is designed and implemented by using the queueing analysis and discrete event simulation.The simulation result based on similar traffic modeling is analyzed and compared with that based on Poisson arrival process,and the impact of similar nature on network performance is described.更多还原