Revamped LTE-B with DTT-like overlay ready for another mobile TV assault

We can count at least three major attempts to get mobile broadcast over cellular off the ground since the turn of the Millennium and all flopped dismally. This might seem good reason to be skeptical over prospects for the fourth effort now underway on the back of 5G, especially as at first sight it is just a revision of the third system, evolved Multimedia Broadcast Multicast Service (eMBMS), otherwise known as LTE Broadcast, or more correctly LTE Multicast, which was first trialed by Verizon among others in 2014.

However, there are some substantial differences in the latest version, Further evolved Multimedia Broadcast Multicast Service (FeMBMS), specified in 3GPP Release 14 in June 2017 and also confusingly referred to as LTE Broadcast. The most important enhancement is support for High Power High Tower (HPHT) transmitters alongside cellular cells, to create overlay networks able to provide the required coverage in a more cost effective way for popular linear content in both rural and urban areas.

Alongside this is support for larger greater Inter-Site Distance (ISD) at high spectral efficiency, with ability to allocate all of the eMBMS carrier bandwidth if required to create dedicated broadcast networks – this was limited to 60% of total capacity under eMBMS. The new model also enables hybrid services where on demand content can still be unicast over the cellular network.

Even these improvements by themselves would not suddenly make mobile TV successful, especially as they also call for substantial investment. It is worth first glimpsing briefly back at the past failures to assess why mobile TV might be ready to break through at last.

The first was DVB Handheld (DVB-H) developed and published as an ETSI standard in 2004, specifying an additional module in compliant mobile devices. Commercial services were launched in some European countries, but DVB-H failed for various reasons, including cost of upgrading the infrastructure and implementing DVB-H stacks in devices, neither of which happened to a great extent through lack of a sufficiently compelling profitable business model. Its prospects were also dented by arrival of DVB-T2 with support for portable receivers in the home, which was a primary use case for DVB-H since mobile handsets were then incapable of playing back good quality video.

Meanwhile, Qualcomm had developed its MediaFLO technology with a different approach based on a dedicated wireless network designed to avoid taxing cellular operators’ infrastructure while offering higher quality pictures. This too foundered on the costs involved, compounded again by lack of consumer demand or a compelling use case. It was hindered by lack of available TV shows and other content over the Internet for access via smart phones or tablets either free or at least for the price of a data service plan.

Then the 3GPP came out with Multimedia Broadcast Multicast Service (MBMS), extending 3G UMTS network capabilities to support point-to-multipoint (PMP) distribution, but this only gained interest from operators when the evolved version eMBMS came along and transmissions were delivered over an LTE network, giving birth to LTE Broadcast. Qualcomm itself had rejoined the mobile TV fold with renewed enthusiasm and teamed up with Vodafone Germany along with Ericsson as infrastructure provider and Samsung with handsets to conduct the first live tests in Europe in 2014.

eMBMS had several perceived advantages over those earlier technologies, primarily being integrated with existing 4G/LTE networks so that no major enhancements were required. Support for SFN (Single Frequency Network) operation allowed the same bandwidth to be reused for a given channel in every cell. It also had what seemed like a compelling launch use case for delivering simultaneous video, such as instant replays, to large numbers of users in and around stadiums at live sporting events. These were indeed the use cases for the first trials, with subsequent traction then anticipated from push VoD, where assets would be cached on the device for later playback, monetized either via a traditional pay model or ad-supported. The idea was that operators could push common titles or bundles to devices on the basis of user profiles and intelligent algorithms, DRM-supported with expiry and content protection.

But again, this really never took off because there was not really enough demand for such simultaneous access to content in stadia, while the push VoD to mobile devices such as smartphones has been hamstrung by lukewarm consumer acceptance combined with lack of on-board storage. While a set top with a 1 TB drive has plenty of capacity for push VoD, this is not the case for a phone where 16 GB has to be shared with apps and other user data.

With support for HPHT (High Power High Tower) combined with dynamic re-allocation of downstream bandwidth between broadcast and unicast operation, eMBMS is now better placed to support converged mobile/broadcast services. The MBMS operation on demand (MOOD) mechanism allows dynamic switching from unicast delivery to multicast when the number of devices accessing the same content within a given cell exceeds a defined threshold. It could be that when 4 or more users are watching the same content via unicast in an area, it may be more convenient or economical to deliver that content just once via multicast, avoiding wasting resources, and transparently migrate those devices and users to this overlay stream.

Until now, mobile linear TV services over eMBMS have not scaled sufficiently in revenue to justify investments in the additional infrastructure, whether with ad-funded or subscription models. It has still been more profitable to deliver via unicast OTT given the potential user base, while even though the eMBMS API is now included in Android, the number of devices supporting those capabilities is limited.

However, one other factor now favors mobile multicast and that is the transition of the 700 MHz spectrum from DTT to mobile broadband, which in the European Union must be finished by 2022 at the latest. This is driving broadcasters towards mobile TV as an alternative to DTT. It is notable that the European Broadcasting Union (EBU), along with several of its members, played a significant role in development of the latest enhancements to eMBMS, including the HTHP overlay model. The EBU is also working on QoS and security for mobile TV as part of an ongoing 3GPP Feasibility Study on Audio-Visual Service Production.

At this stage, eMBMS is still hitched to LTE and so one future development will be to align it fully with 5G, but that has been postponed until at least 2020. There will though be multicast/broadcast support in Release 16 of 5G due for release later in 2019. Meanwhile, further enhancements are now under consideration, such as support for mixed mode multicasting, bringing the potential to exploit both downlink and uplink unicast, with configurable dynamic coverage ranging between a single cell and a large area, along with multiplexing and what the 3GPP calls seamless switching between broadcast and unicast traffic.

Along with these developments there has been mounting interest in use of 5G networks for transport of audiovisual content in production and contribution. There is the prospect of reducing costs for live events with remote production, which currently often requires complex expensive setups including wireless cameras with video relays to OB (Outside Broadcast) vans at the venue, alongside satellite or fixed links to studios. This gels with the trend towards IP based workflows, with the possibility also of avoiding the need for wired infrastructures in studio production.

Given these various options and the underlying business drivers, there is good reason for betting on 5G making a much greater impact on broadcasting at some level than its predecessors.