Dutch Telco KPN has waited for a standards approach based on the DVB Common Interface Plus module (CI Plus) to move towards a set top free IPTV service, with a prototype expected early in 2015.
This follows two key developments, firstly arrival earlier this year of the latest version of the module, CI Plus 1.4, adding ability to watch and record all channels at the same time and also enabling simultaneous access to OTT as well as broadcast content. Secondly there is now rapid progress toward an overall standard for set top free IPTV under a consortium led by Dutch non-profit applied science research group TNO.
KPN is a member of this consortium, along with Austria’s A1, Swisscom, TDC, Proximus, Boxer of Sweden and Bulgarian Telco M-Tel, as well as some industry partners, with the goal of building a common ecosystem for delivering IPTV services directly to the TV. Another key development closely related to CI Plus 1.4 was the OIPF (Open IPTV Forum) package for set top free IPTV published earlier this year, incorporating specifications for full service deployment.
As it happens, both KPN and the TNO consortium made their announcements at an IPTV Seminar sponsored by TNO at the Hague this week. TNO’s senior consultant for media technology Menno Bangma summarized the overall aim of IPTV with a set top to recreate the simplicity and ease of use of the analog TV days, but with the deeper user experience, content range and features of digital TV. This meant “no set top, no additional wiring and just a single remote control.”
KPN highlighted how these standards developments meant that it was now feasible to virtualize the set top within a premium IPTV service, although was of course referring to its own favored approach, based on the CI Plus module.
If the traditional set top is defined as a standalone box in the home it can be virtualized in four ways. One is to replace it with Internet connected smart TVs running a suitable app to access the service, as TeliaSonera has done in the Nordics. Secondly the set top can be shrunk down into a small form factor such that it is no longer physically an independent box, typically as an HDMI attached dongle communicating over WiFi as promoted by SoftAtHome for example. Thirdly the box can be replaced by a module that slots into compliant regular TVs, as KPN is doing with CI Plus, which has emerged as the industry standard here. Fourthly the set top can disappear completely into the cloud, including storage and media services, UI and DRM.
They all have pros and cons, with the dongle reliant on WiFi, while CI Plus will only work with compliant TVs, although there have been plenty of those for several years supporting evolving versions of the standard. The dedicated smart TV option suffers from lack of standardization in that area with the Smart TV Alliance having foundered largely through lack of Samsung’s support. TeliaSonera’s set top less service only works with Samsung TVs, via an app that recognizes the customer’s IPTV service when the TV is connected to the Internet.
With these different models the cloud platforms that enable set top virtualization should ideally support all the variants, as ActiveVideo does with its CloudTV. This will work with legacy set tops into the bargain, so that operators do not have to virtualize in one big bang. This has been chosen by Deutsche Telekom and Ziggo in the Netherlands as well now as KPN as part of their ecosystems involving set top virtualization.
KPN choose the CI Plus route because it enables virtualization and introduction of OTT with a hybrid model without having to ditch at a stroke the traditional set top approach to service delivery and security in particular. The CI Plus standard itself provides the foundation for this by cleanly separating the traditional CA (Conditional Access) functionality from the digital TV receiver and decoder. The CA can then be deployed in a removable module, the CAM, or potentially in software for a smart TV, executing the CA descrambling. The host TV then handles tuning and demodulation of the RF signal, while CAM is responsible for CA descrambling.
The latest version CI Plus 1.4 combined with the associated standards promoted by this TNO led consortium then enable advanced functions such as multi-stream PVR, with operators able to control ad trick mode control to prevent ad skipping on playback, for example. OTT access to both adaptive streams and file based content are now supported, with the crucial addition of DRM support alongside the traditional CA model. The DRM is then implemented in the CAM, which executes the decryption, effectively providing a form of hardware security that can be deployed within OTT streaming services. But it could also be used in place of a secondary IP client set top as part of a whole home DVR deployment.
In KPN’s case the CAM will be made by SMiT and incorporate Verimatrix content security. KPN is aligning the deployment with the TNO standards consortium it belongs to, which will yield a prototype designed to emulate in the CAM all existing set top functions, including network PVR, as well as the CA and IP routing. The main sticking point now, according to the TNO consortium, is reaching complete agreement with TV vendors over how all the additional features should be implemented. One interesting aspect here is that the standard is based on an app that in turn largely consists of the HbbTV browser, plus a few extensions, so it does have a very European flavor. Penetration so far is largely confined to Europe but it has also gained traction in the cable sector with some deployments from Liberty Global and Vodafone, in the latter case at subsidiary Kabel Deutschland.
The cable operators are less interested in set top virtualization at this stage. Telcos have been the derivers because for them with a box already in the home as a gateway or broadband router there is a strong incentive to avoid having another adding to both Opex and Capex.
Deutsche Telekom has been in the vanguard here but in this case not taking the CI Plus route. Instead Deutsche Telekom wants to reduce the set top client to dumb status with all apps and functions, except video decoding, executed in the cloud in a standard browser and transmitted to consumer devices as a video data stream, such as H.264 or H.265. Then the client can equally well be a standard TV equipped with an HDMI dongle for the decoding, a smart TV, a tablet, a PC or a smartphone.
DT has been running trials for several years with ActiveVideo to assess whether the UI can be rendered from the cloud with sufficient response and that the network can handle the number of unicast nPVR sessions without falling over. Both problems were solved effectively through a combination of intelligent caching and higher bandwidth, especially over the last mile, and now DT’s prototype has been successfully deployed in some of its operations, such as OTE in Greece.
Now DT is ready to roll out cloud TV and unlike some of the others, does not want to hang on to legacy set tops any longer than it has to. This is because a motive for cloud TV in this case is to avoid the impediment that legacy boxes present, both for performance and service upgradeability. By refocusing on UI rendering in a cloud HTML5 browser, the user experience is decoupled from the development process, which makes it easier and cheaper to deploy new functions and services.
KPN and its fellow CI Plus advocates will also enjoy lower costs and greater agility, but not the complete separation between the service and user device.