| Title: |
Characterising the Statistics of Border Gateway Protocol (BGP) Behaviour |
| Venue: |
EN205, Level 2, EN Building |
| Abstract: |
The core routing protocol of the Internet, the Border Gateway Protocol (BGP) is remarkably vulnerable to accidental or deliberate misconfiguration. BGP is responsible for distributing routing information between autonomous systems. However, BGP generally assumes one autonomous system (AS) can trust what neighbouring ASes tell it about how to reach the rest of the Internet. Consequently, whether by malice or mistake, it is easy for individual service providers to inject erroneous changes having widespread, sometimes globally negative impact on Internet services. The aim of this project is to identify statistical markers that are robust early indicators of such anomalous BGP behaviour. In particular we are keen to make use of novel techniques of data analysis suitable for non-linear dynamic systems. In this talk, I will discuss how to characterise BGP behaviour using Autoregressive-moving-average (ARMA) model. |
| Biography: |
Djuro is enrolled in Bachelor of Engineering (Telecommunication and Network Engineering) / Bachelor of Science (Computer Science and Software Engineering) at Swinburne University of Technology and currently he is in his 4th year. Djuro is also completing a winter internship at CAIA. |
