An SDN-based framework for detection of illegal rebroadcasting of channels in P2PTV

Nowadays, mesh based live P2P streaming systems (P2PTV) is a popular means of streaming audio and video content over the Internet. The main reason being that it can accommodate large amounts of users with low cost compared to dedicated servers, or content distribution network solutions. However, one of the issues and impediments associated with this technology is the rebroadcasting of illegal or inappropriate material. This work presents an early architecture that relies on capabilities provided by Software Defined Networking (SDN) to address this issue. We adaptively partition the outgoing P2PTV traffic and distribute detection across a range of detection nodes, which use some of the existing video identification techniques to detect suspicious content. Moreover, due to processing requirements of vector graphics we suggest an assistive scheme so that controllers can push traffic to other controllers' network and take advantage of available resources under their control. To automate this process in a network we propose an extension to distributed SDN controllers architectures so they can actively cooperate. To the best of our knowledge, there is no work that has studied detection of illegal or inappropriate rebroadcasted stream through P2PTV technology and our proposed scheme is novel.