The broadcasting industry has refocused its efforts away from defending spectrum against invading cellular services, towards ensuring that those networks allocate some of the frequencies they have gained to video services.
This looks like a case of robbing Peter to pay Paul because the underlying technologies for carrying audiovisual (AV) traffic will be essentially the same with the change being over who owns and manages the networks and allocates underlying spectrum. There is another angle though, in that a growing amount of video traffic ends up reaching viewing devices over wireless connections, which can be WiFi or cellular depending where the user is. Therefore, convergence between WiFi 6, the latest WiFi standard, and 5G, is also a concern for the AV industry.
However, the underlying technology is less of a concern than ensuring its implementation in both emerging 5G infrastructure and the devices that will access it.
That is the remit of the 5G Media Action (5G-MAG) launched last month at IBC 2019, which has just held its first General Assembly at the EBU (European Broadcasting Union) headquarters in Geneva. It has representation across the media industry with French encoder maker Ateme, test and monitoring group Rohde & Schwarz, satellite platform operator Eutelsat and video infrastructure provider Enensys among over 30 members so far, along with broadcasters Bayerischer Rundfunk, Cyfrowy Polsat Group, France Télévisions, Italy’s RAI and the EBU itself. Huawei is the notable representative from the cellular infrastructure side.
As Antonio Arcidiacono, EBU director of technology & innovation, said at the Assembly, the question is not so much whether 5G networks can play a central role in future media delivery, but whether they will. “One of the big challenges is to ensure that new 5G features such as broadcast capabilities are supported by network infrastructure and user devices,” he said.
Essentially the same SFN (Single Frequency Network) operation will be used to optimize transmission efficiency over large areas with multiple transmitters, using underlying OFDM (Orthogonal Frequency Division Multiplexing). This is the basis of the high-power high-tower (HPHT) model optimized for downlink only multicast traffic whose operation over cellular networks but without requiring SIM cards is defined under 3GPP Release 14.
This HPHT model has been trialed in Germany as part of the research project ‘5G Today, funded by the Bavarian Research Foundation and supported by Telefónica Germany alongside Bayerischer Rundfunk, the Bavarian state broadcaster and one of the 5G-MAG members. The broadcaster provided two HPHT sites near Munich for this trial, claiming this to be the first of its kind in the world.
This trial then is providing input for the 5G-MAG and helping meet one core objective that does have a technical aspect, ensuring that 5G wide area network coverage even outside metropolitan areas reaches as close to 100% of the population and territory as possible, in line with coverage already achieved by DTT but not always mobile services.
However, this will not achieve anywhere close to 100% coverage indoors and that is where the convergence with WiFi comes in. Although cellular and WiFi have often presented as competitors, even more so with 5G and WiFi 6 as they come closer together, they still really have distinct use cases. Essentially, WiFi is the medium for indoor coverage and cellular outdoors.
Because more video is consumed indoors and WiFi supports higher device density, it will continue to carry the most traffic, even though in the 5G era cellular will narrow the gap. Cisco in its Visual Networking Index predicts that by 2022, 201 exabytes of monthly data will be transmitted worldwide on a combination of fixed IP and WiFi networks, against 116 exabytes a month on wireline-only networks and 77 exabytes on cellular networks.
Even with 5G closing the gap, WiFi 6, the simplified name for the latest 802.11ax version of the IEEE standard, increases bandwidth by several orders of magnitude and brings on new use cases. There is also increasing momentum behind unified cellular/WiFi services with transparent handover between the two, which has been on the agenda for several years but never quite delivered. One stumbling block has been lack of a standard for discovering WiFi networks that a device has not encountered before, so that most users have continued having to do that manually.
But now the Wireless Broadband Alliance, representing mobile or broadband operators like AT&T and Comcast, along with WiFi access point (AP) makers including Cisco, CommScope’s Ruckus and HP, has partnered with the Next Generation Mobile Networks (NGMN) Alliance to produce a white paper highlighting and investigating “the importance of existing and future WiFi and cellular convergence”.
The paper’s premise is that emerging individual wireless use cases, as well as a growing number of vertical market segments, will require access to both cellular and WiFi networks. There are various drivers such as reduced latency, higher throughput, greater connection density, availability, reliability and above all ubiquitous coverage. It will generally improve the experience for smartphone users, while generating new revenue streams for operators on both sides.
This translates into technical requirements, with a fundamental one being to enable the many WiFi-only devices, such as most tablets and laptops, to connect to the 5G core without requiring SIM cards or equivalent. This means adding support for additional device identity types based on WiFi centric subscription and authentication, with corresponding alternative authentication credentials in the 5G Core network.
Currently, the 3GPP standard requires presence of SIM credentials, so work is needed here to enable access to the 5G Core from untrusted non-3GPP access networks such as WiFi.
Then there is obvious demand from the cellular side for improved visibility into and control of WiFi access networks for configuration and management. Here, it may be possible to capitalize on work already done by broadband operators for improving QoS over WiFi as that has become effectively their new last mile. There is similarly a need from the WiFi side to set policies in the 5G core for handling traffic, or requesting network slices to ensure end to end QoS, as well as smooth handover.
Another challenge sometimes forgotten is cost, which may be one reason cellular/WiFi convergence has enjoyed poor traction so far and some of the early vendors offering products for this have fallen by the wayside or gone bust. This is also the reason so much traffic goes over WiFi. Despite improvements, the cost per bit of cellular traffic is still an order of magnitude higher than fixed network traffic. This will have to be addressed if converged WiFi/cellular operation with 5G service delivery is to take off.
Despite these caveats, it is only a matter of time before greater WiFi cellular/convergence finally does happen, because the two complement each other well and demand will be driven not just by emerging use cases around the Internet of Things but existing ones such as video delivery. 5G-MAG is important for broadcasters because increasingly WiFi will be their last mile while 5G will provide core transmission to a growing number of homes, as well as for viewing devices on the road. For 5G-MAG, ability to access 5G cores from WiFi only devices will be one of the major objectives.