Application of Hidden Markov Chains for Comparison of Traffic Traces

OData support
Dr. Molnár Sándor
Department of Telecommunications and Media Informatics

With the increasing popularity of the Internet, lot of different applications, such as video sharing portal , and Internet radio have appeared and spread widely. I selected the analysis of these traffics as the topic of my thesis based on packet inter-arrival time and on packet size with help of the given method. With help hidden Markov models there is a procedure based on the aforementioned two statistics, that we can comparing two traffic sample, and can express the similarity between them with only one number. My task was comparing traffic samples generated in different ways with each other, documenting the outcomes of this analysis, and then analyze the results. Naturally, first of all the examination of the suitable literatures meant the capital goal. My chosen examination areas was video sharing sites, internet radio, Skype, file downloads, browsing and live performance videos. In the analysis I received an answer to the question of whether the aforementioned traffic similar to each other in point of packet inter-arrival time and packet sizes and between these values exist a relationship.


Please sign in to download the files of this thesis.