Since the appearance of various P2P IPTV systems which timely broadcast live streaming to peers, they have attracted millions of users from all over the world. It is reported that the online audience have reached more than 1.2 million in peak time by the official website of PPStream, one of the most popular IPTV system in China. However, at the same time the popularity of these systems make the amounts of video traffic grow exponentially. In order to study the global playback performance, users' behaviors and network characteristics as well, we developed our dedicated crawler of PPStream. Based on the measurements, we make some extensive performance evaluation on this commercially successful P2P IPTV system, and some characteristics on geographic clustering, connection stability, arrival/departure pattern, playback quality, sharing ratio and topology have been revealed. We think these findings can help other researchers model such streaming systems and system operators make further optimizations.
With the wide spread of P2P streaming systems, there appear some hackers who try to pollute the system by inserting fake data chunks into the system. These "dirty" chunks will then be propagated to many normal peers, which will result in the failure of the whole system. Considering the playback quality and the burden of the client, most systems nowadays don't adopt any measures to protect the video content, such as encryption, CRC, or other methods.We think it's more important to eliminate the polluters actively than to prevent the peers from polluted passively. In the paper, we propose a new scheme which tries to leverage the P2P nature of the system. We piggyback some "check" bytes in the data chunks exchanged between different peers. On one hand, these "bytes" can verify the authenticity of the data chunks, which prevent the diffusion of the polluted chunks. On the other hand, they can also help to detect and identify the "polluters" of the system. By simulation, the effects of our schemes are evaluated and the results are very inspiring. We think this light-weight solution against the streaming attacks is promising and can be deployed in real world.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.