ABSTRACT
This chapter offers a literature survey of major classes of peer-to-peer (P2P) traffic detection methods. Intelligent combination is based on deep packet inspection (DPI) and semi-supervised learning is based on deep flow inspection (DFI). The chapter discusses the BitTorrent (BT) protocol and details the needs of detecting BT traffic. It presents the related studies, including the main class of the mechanisms to detect P2P traffic. The chapter describes new, improved approach, intelligent combination. BT is a P2P file-sharing protocol for distributing a large amount of data over the Internet. It is one of the most popular P2P protocols. The chapter examines the major techniques used to detect P2P traffic. There are several major classes of techniques: port-based, DPI, DFI and some combination of DPI and DFI techniques. The chapter explains the offline training module used for DFI classification. This step can quickly filter out some BT flows that used popular P2P ports, and therefore speeds up the average execution time.
