Over recent years, a major shift has occurred in piracy of paid-for content services toward illegal redistribution of live content in real-time over the Internet.
This paper will provide insight into pirate content platforms, covering the various architectures and protocols used, from peer-to-peer protocols adapted for live streaming to more traditional Web streaming protocols.
More specifically, it will focus on the methods generally employed to set up and scale ad-based illegal services using some of the above-mentioned protocols with streaming media platforms, while securing streaming servers, enabling these sites to remain hidden.
A thorough analysis of the used architectures and protocols makes it possible to measure the actual audience viewing illegal streams, typically leveraging peer-to-peer networks data.
This enables content service providers to assess the piracy threat level of any content, while illustrating the need for a business intelligence tool that provides relevant information on viewers’ behavior.
The history of pay-TV [1, 2, 3], considering the business at stake, is unsurprisingly tightly coupled with the history of content services piracy, effectively proving the saying that “security is a process, not a product” in this industry.
Content services piracy has evolved year after year, mainly adapting to both the solutions developed by content security vendors and technologies available to circumvent them while offering an alternative solution intended to generate parallel business.
Piracy forms have ranged from video receivers and smartcard piracy, to the sharing of subscription rights through service access credentials (login and password) or conditional access smartcard sharing to content decryption keys (known as Control Words) redistribution both over the Internet and satellite feeds to cover wide distribution regions.
Those Control Words feeds have even been delivered over the very satellites whose capacity was legitimately used by operators to broadcast their channel signals.
A major shift in paid-for content piracy has occurred in recent years. There has been, in particular, an increase toward illegal direct redistribution of live content in real-time over the Internet, most notably for content with very high “live value” such as sports events.
THE PIRACY LANDSCAPE
Illegal live content redistribution over the Internet has been following two main distinct approaches: peer-to-peer (P2P) live streaming and Web streaming with and without the use of Content Delivery Networks (CDN).
The origin of P2P as a technology dates back from the early days of Napster in 1999: people used their “own” bandwidth to share content (music back then) long before CDNs had become the standard for content delivery.
The origin of P2P streaming, although based on the same idea of limiting servers’ bandwidth, is somewhat unclear. The Chinese P2PTV protocols (e.g., PPLive or PPStream)  began to deliver content in the second part of the 2000 decade.
Most of these protocols were real-time-enabled BitTorrent derivatives. These protocols have evolved, with new improved implementations having emerged, paving the way for today’s landscape.
Different P2P streaming platforms are used nowadays, with the two main ones being SopCast [5, 6] and AceStream (formerly known as TorrentStream).
SopCast is a Chinese university project that became a widespread piece of software, loosely maintained. AceStream is certainly the latest and most innovative method of P2P streaming, apparently maintaining a legitimate side to its operations.
Other P2P protocols have emerged and gained a certain level of exposure, most notably BitTorrent Live announced by BitTorrent Inc. at the Consumer Electronics Show (CES) in 2012, until a complete shutdown of the trials in February 2014. BitTorrent Live’s new target seems to be mobile live broadcast through P2P.
P2P streaming principles are similar to P2P file exchange ones: P2P users sharing the same content form a loosely connected mesh network compared with a full mesh network where all peers are connected to all other peers.
This structure gives the P2P network reliability and resilience. If one network node stops working, the remaining nodes can still work together, provided they are able to reconnect, if needed.
The benefits of P2P streaming, from a viewers’ standpoint, are essentially twofold. Firstly, the size of a P2P network is virtually “limitless”. The capacity to deliver content to a significant amount of viewers, widely acknowledged for P2P file sharing, therefore also applies to live streaming P2P protocols.
Some of our recent measures over such P2P streaming networks indeed confirm audiences of over 30,000 viewers on each of selected streams on a regular basis. Secondly, the quality of streamed content is usually noticeably higher in terms of achievable bitrate.
Whereas the majority of direct Web streaming bitrates are lower than 600Kbps, our knowledge base shows that 60 percent of the P2P streams have bitrates below 2Mbps, 30 percent between 2Mbps and 4Mbps, and the remaining streams with bitrates above 4Mbps.