Research On Traffic Characteristics Of Iot Under Different Traffic Density And Human Particapation Degree

A lot of measurement data indicates that traditional internet flow has self-similarity(or long range dependence)characteristic,and the characteristic has a significant impact on network performance evaluation and business modeling techniques.Since the Internet of Things has different characteristics compared with the traditional Internet,including low mobility,superior occupancy,no human participation,the Internet traffic characteristics will undergo a great change.The study of network traffic characteristics is the basis of network planning and design and performance evaluation,and it is very important to understand the running rules of network and ensure network security.Firstly,four typical types of business in the Internet of Things are analyzed based on the analysis of the development,architecture and application way of the Internet of Things,including telemedicine,intelligent agriculture,autopilot and vending.Based on the results of the above analysis,the corresponding traffic models are put forward,from which the simulation traffic is generated and then the Hurst parameters and variance characteristic can be analyzed under different time scales,different traffic densities and human participation,and the result shows that the network traffic has different traffic characteristics in different circumstances.For the same service under different flow densities,the self-similar characteristic will increase while in other occasions,it will keep stable.For the same kind of service in different timescales,some businesses have self-similar characteristic,and some businesses do not have.Furthermore,after analyzing the characteristics of traffic when different kinds of businesses are aggregated,it is found that there is still no self-similar characteristic after two businesses who both have no self-similar characteristic are aggregated,and the flow still have self-similar characteristic after two businesses who both have self-similar characteristic are aggregated.When two businesses are aggregated,one of them has self-similarity and the other does not.Weather the aggregative traffic has self-similar characteristic or not due to the Hurst and variance parameters that make up the aggregative traffic.Then,the characteristics of the flow rate under different variance and Hurst parameters are analyzed.It is found that the flow variance has a great influence on the polymerization flow.Finally,the influence of network traffic parameters on network delay characteristic is analyzed using OPNET simulation tool.It is found that the variance of network traffic and Hurst parameters have a great impact on network performance.