A Study On The Self-Similarity And Simulation Of Cps Traffic

Cyber-Physical Systems (CPS) realizes the real-time sensing of the physical environment and resources, information fusion and dynamic control of physical processes through the deep integration of computation, communication and control technology, so as to make the physical world integrate the cyber world more closely.As CPS technology progresses, CPS network needs to carry more and more application services. In order to achieve reliable data transmission and a reasonable allocation of network resources in CPS applications, a deep research of the complex behavior characteristics of network traffic is required. However, there is little research on CPS traffic in the current literature. A set of theories and methods hasn’t been proposed as guidance to design CPS networks.This research project is originated from Beijing University of Posts and Telecommunications fundamental research fund subject:"Traffic Analysis and Modeling in Complex Hybrid Network Environment". The innovation of this research is that we focus on CPS traffic characteristics, classify CPS application scenarios, and study traffic characteristics in the different CPS application scenarios by theoretical analysis and simulated verification. This research can help to better plan and design CPS networks.In this paper, an overview of the key contents of Cyber-Physical Systems is firstly presented. Then we study the definition and properties of self-similarity, compare the advantages and disadvantages of main self-similarity analysis methods. In addition, CPS application scenarios are analyzed in details and classified. CPS traffic characteristics is analyzed theoretically for different CPS application scenarios. At last, we utilize NS-2to simulate CPS network. Based on the emulated platform, we design the corresponding simulation scenarios for different CPS application scenarios, simulate the CPS traffic, and analyze the simulation results. The analysis result indicates that, the traffic of CPS sink node is self-similar both in the event-drive application scenarios and the time-drive application scenarios.