论文部分内容阅读
众所周知,业务流量特点建模有助于设计能效低、稳定性强的网络协议.近年来新兴业务如雨后春笋一般涌现,重新考察这些新业务的特点也就至关重要了.本文以即时消息业务为例,了解当前移动互联网下的流量新特点.为了得到可信的结果,采集了大量的蜂窝网即时消息实测数据,并从用户层面和基站层面两个维度去研究其统计规律.首先在用户层面,通过用户级的数据包包长和到达间隔建模发现,相关结果同传统上包长符合几何分布、到达间隔满足指数分布的3GPP报告明显不同;与之相反,微信包长和到达间隔分别遵循幂律和对数正态分布.其次在基站层面,通过随机选取基站发现,α-稳定模型能更好地刻画基站层面基站流量——这一结果同传统固定宽带网络业务研究相吻合.最后,建立了用户层面数据包同基站层面流量的理论联系.
As we all know, the modeling of business traffic features help to design low energy efficiency and stability of the network protocol.In recent years, emerging business springing up in general, re-examine the characteristics of these new business is also crucial.In this paper, instant messaging business In order to get credible results, a large number of measured data of instant messages in cellular networks are collected, and the statistical rules are studied from the two dimensions of user level and base station level.First, at the user level , It is found through the modeling of user-level packet length and arrival interval that the correlation results are consistent with the traditional packet length and the 3GPP reports whose arrival interval satisfies the exponential distribution are obviously different. On the contrary, the length and arrival time of the micro-channel packets are respectively followed Power law and logarithm normal distribution.Secondly, at the base station level, the α-stability model can better depict the base station layer base station traffic by randomly selecting the base station discovery - This result is consistent with the traditional fixed broadband network service research.Finally, Established a theoretical level of user-level data packets and base station-level traffic.