M2M流量特性分析

发布时间：2018-04-22 19:25

本文选题：物联网 + M2M　；参考：《西南交通大学》2014年硕士论文

【摘要】：网络流量特性随着网络的发展而变化,从早期的短相关性、马尔科夫性到现在的自相似性、长相关性等。随着物联网技术的发展,海量的物联网终端将会加入到网络中来,这会对网络产生巨大的冲击。下一代网络将是以物与物通信为主要特征的物联网,与目前以人为主要用户的互联网具有完全不同的特点,网络流量发生了根本性变化,对网络规划设计、性能评价和服务质量保障提出了新的挑战。本文首先对网络流量特性分析的发展、物联网的体系结构进行了分析。明确了需要重点研究的网络特性即自相似特性,以及研究方法。然后以典型的物联网业务为例,通过对业务特性的分析,得到不同业务的流量生成模型。在此基础上通过MATLAB仿真产生不同终端规模的业务流,对产生的业务流进行自相似特性分析。研究结果表明：不同业务的流量自相似性不同,有的具有自相似性,有的不具有,有的自相似特性很强,有的偏弱；同时不同业务随着终端规模的增加,自相似特性也呈现出不同的变化规律,有的增强有的很稳定,而也有的却会减弱。进一步,进行不同M2M (Machine to Machine)业务之间以及M2M与实际采集的互联网流量的聚合,分析聚合流量的特性,发现聚合流的流量特性因不同的M2M业务源而表现出不同的变化规律,同时分析、印证了流量源方差对聚合流的流量特性具有重大的影响。最后通过对实际采集的传感器网络流量特性分析,进一步印证了上面的分析结论。本文通过对M2M业务流以及聚合流量特性的分析研究,发现聚合流量会因M2M业务的不同而展现出不同的自相似特性,由此可以看出下一代网络流量的自相似特性将具有很大不确定性。
[Abstract]:Network traffic characteristics change with the development of the network, from the early short correlation, Markov to the present self-similarity, long-term correlation and so on. With the development of Internet of things (IoT) technology, a large number of Internet of things terminals will join the network, which will have a huge impact on the network. The next generation network (NGN) will be the Internet of things (IOT), which is characterized by material and material communication, which is completely different from the current Internet, which is mainly used by people. The network traffic has changed radically, and the network planning and design has taken place. Performance evaluation and quality of service assurance pose new challenges. Firstly, the development of network traffic characteristic analysis and the architecture of Internet of things are analyzed. The characteristics of the network that need to be studied, namely, self-similarity, and the research methods are defined. Then taking the typical Internet of things business as an example, the traffic generation model of different services is obtained by analyzing the service characteristics. On this basis, the traffic flow with different terminal size is generated by MATLAB simulation, and the self-similar characteristics of the generated traffic flow are analyzed. The results show that the traffic self-similarity of different services is different, some have self-similarity, some do not, some have strong self-similarity and others are weak. The self-similar characteristic also presents the different change law, some enhancement is very stable, and some will weaken. Furthermore, the aggregation between different M2m machine to machine services and between M2M and the actual collected Internet traffic is carried out, and the characteristics of aggregate traffic are analyzed. It is found that the traffic characteristics of aggregation flow show different changing laws because of different M2m service sources. At the same time, it is proved that the variance of flow source has great influence on the flow characteristics of aggregate flow. Finally, the above conclusions are confirmed by analyzing the traffic characteristics of sensor networks collected in practice. In this paper, the characteristics of M2m traffic and aggregate traffic are analyzed, and it is found that the aggregate traffic will exhibit different self-similar characteristics because of the different M2M traffic. It can be seen that the self-similar characteristics of next-generation network traffic will be very uncertain.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06

【参考文献】