DocumentCode
1795730
Title
A first look at machine-to-machine power grid network traffic
Author
Sang Shin Jung ; Formby, David ; Day, Carson ; Beyah, Raheem
Author_Institution
Commun. Assurance & Performance (CAP) Group, Georgia Inst. of Technol., Atlanta, GA, USA
fYear
2014
fDate
3-6 Nov. 2014
Firstpage
884
Lastpage
889
Abstract
The purpose of network traffic characterization is to explore unknown patterns in different types of network communications to help improve many aspects of the network. While many previous studies have explored the characterization of many different networks (e.g., university networks), the power grid network (and other SCADA networks) characterization has not yet been studied. In this paper, we provide a characterization of the power grid network to answer questions like the following: i) how stable is the communication based on configurations?; ii) are there different observable traffic patterns in different vendor equipment?; iii) are there trends in the network traffic?; iv) can information be gathered from the traffic characterization to help secure the power grid network? To address these questions, we have collected power grid network traffic in a live substation for two months and conducted an empirical study to identify network traffic behaviors in the live substation. Our empirical study shows different behaviors between the devices and vendors when they communicate with each other.
Keywords
SCADA systems; carrier transmission on power lines; power grids; power system security; security of data; substations; telecommunication traffic; machine-to-machine power grid network traffic; network communications; network traffic characterization; power grid network security; Internet; Ports (Computers); Power grids; Protocols; Security; Substations; Sun;
fLanguage
English
Publisher
ieee
Conference_Titel
Smart Grid Communications (SmartGridComm), 2014 IEEE International Conference on
Conference_Location
Venice
Type
conf
DOI
10.1109/SmartGridComm.2014.7007760
Filename
7007760
Link To Document