PBS: Periodic behavioral spectrum of P2P applications

Abstract

Due to the significant increase of peer-to-peer (P2P) traffic in the past few years, more attentions are put on designing effective methodologies of monitoring and identifying P2P traffic. In this paper, we propose a novel approach to measure and discover the special characteristics of P2P applications, the periodic behaviors, from the packet traces. We call this the “periodic behavioral spectrum” (PBS) of P2P applications. This new finding, learning the characteristics of P2P traffic from a new angle, could enhance our understanding on P2P applications. To show the effectiveness of our approach, we not only provide justifications as to why P2P applications should have some inherent periodic behaviors, but also conduct hundreds of experiments of applying the approach on several popular P2P applications.