Research On Simulation In Embedded Networks

Industrial networks have been an essential technical in industrial control system, which is realization of industrial control layer by computer networks, such as fieldbus. The aim of development of industrial control system is to build the architecture of integrated automation system based on the development on computer technology, communication technology and control technology. Researchers often use simulation method to study networks because of limitation of studying networks in real circumstance of networks. Simulation on computer networks and industrial networks is a convenient and effective method to be used widely.NS2 is a universal, visual and expandable software for network simulation, which is programmed on Unix and includes four packages:Tcl, Tk, OTcl and TclCL All simulations are driven by discrete event based on a virtual clock and use C++ and TCL as programming language. NS2 can not only support many kinds of network protocols, routing algorithms, congestion control algorithms, multicast algorithms, and MAC protocols, but also generate many types of network traffics, such as exponential distribution and Pareto distribution.The thesis studies simulation on information networks based on H1 fieldbus by NS2 and builds the models according to the network circumstances with different settings on definition of nodes and links. The structure of the thesis is as follows. First of all, a survey on development of industrial control networks is expounded. Secondly, the introduction to NS2 and its related tools is stated. Thirdly, the simulation on information networks based on H1 fieldbus by NS2 is studied and optimized. Lastly, the thesis rearranges and analyzes the collective data and studies congestion control, traffic control and self-similar distribution by using the collective data. On the basis of the settings and the data analysis, the thesis obtains the conclusion which is the distribution of blend traffic is self-similarity after a non-self-similar traffic is mixed by a self-traffic.