Ultra HDTV will bring the best viewing experience possible to consumers, with unmatched video quality and greatly improved audio quality coming soon. However producing and transmitting UHD content adds an order of magnitude to the overall cost for broadcasters and service providers. As a result Ultra HDTV raises the bar for revenue security even higher than it was already, adding further value so that illegal redistribution of UHD on-demand and live content has an even bigger financial impact on content providers and operators. This was recognized early by Movie Labs which issued its specific mandate for UHDTV security in April 2014, notable for including forensic watermarking as a compulsory component of future video services that deliver its members’ content at the higher 4K resolutions. Since then both owners and distributors of premium live content, primarily sports, have also indicated that watermarking will be required in future for services delivering their assets.
Not surprisingly therefore watermarking has been a major preoccupation for the Ultra HD Forum’s Security Group, given its role in developing guidelines for the whole delivery ecosystem. This group has been working hard to bring out the first guidelines specifying some baseline degree of interoperability among the principle forensic watermarking systems ready for deployment in pay TV services. These come from Content Armor, Nagra, NexGuard and Verimatrix. Recent announcements demonstrate that watermarking technology and use is moving from marginal to mainstream: ContentArmor, whose watermarking technology was originally conceived within Technicolor for post-production services to Hollywood studios, was hived off through a management buyout in March 2016 and in July 2016, Kudelski group, owner of Nagra, strengthened its position by acquiring all of NexGuard Labs BV.
Join the UHD Forum Masterclass at IBC 2016 as they host a free-to-attend session covering this topic and much more on Monday 12th September. Full session details
The objective of forensic watermarking is to provide a new line of defence against illicit redistribution, especially over the Internet, of content that may initially have been acquired legally by apparently legitimate subscribers. Existing content protection mechanisms are bypassed, possibly by direct camcording from a screen, although a greater threat is through direct capture of streams using illegal tools that circumvent the protection provided by HDCP (High-bandwidth Digital Content Protection) over the HDMI interfaces to TVs.
It has been acknowledged not just by MovieLabs but widely within the content security industry that watermarking, admittedly in conjunction with other techniques such as network forensics, is uniquely placed to combat illicit redistribution. This is because the threat can only be dealt with by identifying individual streams and tracing them back to their source, which must require insertion of some unique identifier at the source or during distribution. A watermark can be designed to be tamper resistant and at the same time to make minimal impact on the video or audio, although there is a balance to be struck between robustness, performance and transparency to the user. It is worthwhile to note that visible marking, also known as visible fingerprinting, is already used in some countries, but pirates are finding ways to remove the mark before redistributing the content. Therefore, one of the merits of forensic watermarking is that pirates cannot know if the forensic mark is (still) present before redistributing the content.
It is up to the watermarking technology vendors to make sure their products fulfil these requirements and can resist attempts by pirates to circumvent the protection in some way such as by transforming the marks to make them unrecognisable. The task of the Ultra HD Forum is to make sure that watermarking in general can be deployed as easily as possible by operators and that the different products are interoperable, given that for various reasons such as mergers there may be two or more within a given delivery infrastructure.
An underlying challenge here is that watermarking has not been deployed widely on Pay TV. Although forensic watermarking has been used for protection of movie content in Digital Cinema for some years to identify individual theatres that are sources of pirated camcordings, for pay TV operators it did nothing to combat their traditional threats such as card cloning and sharing. It was only with the rise of content redistribution over the Internet that watermarking came in the frame for pay TV.
“The challenge is the lack of maturity for pay TV, the fact that only a handful of Pay TV operators have done it,” noted Laurent Piron, Principal Solution Architect at Kudelski’s Nagravision and member of the Ultra HD Forum security group. Piron suggested that version one of the Forum’s security guidelines due to be unveiled at IBC will set the stage for more widespread deployment of watermarking. “Our aim is to find a common ground level where all the watermarking companies can agree on something they can put their technology on top of, so that any company, such as an encoding vendor, can integrate with the API. Then they will know it is going to work and can implement one technology or another one. It will also ease the life of operators who want to introduce watermarking, by saving them the pain of integrating all technical components themselves, benefitting from pre-defined and pre-tested integration frameworks.”
Although watermarking has stolen the limelight in the content security sphere recently it is only part of the overall solution, with common encryption being just as important to protect assets in flight or during storage. The Forum will be publishing its encryption guidelines separately from the watermarking in version one, driven it seems by pressure from members to get these out as quickly as possible. Then in version 2 due out hopefully by the end of 2016, the watermarking and encryption guidelines will be integrated as part of the drive to provide a coherent framework for end to end UHD security. At the same time the Forum will revisit earlier work on CAS (Conditional Systems) and DRMs for version 2, which at that stage will still be a separate document from the Ultra HD Forum’s general guidelines.
Piron emphasized here that there was no appetite at all for attempting to converge towards a common DRM as was once a widespread ambition. This idea has been blown out of the water by the proliferation of device platforms each raising somewhat different security issues, served in many cases by major vendors seeking competitive advantage. As Piron pointed out the Forum’s role is to ensure that every DRM can work within the UHD ecosystem and that operators can switch readily from one to another. This again means establishing a baseline to define the encryption and how to identify the licence keys.