6 Comments

Summary:

By multicasting popular content over cellular networks, carriers figure they can conserve valuable 4G capacity. But as consumers use their smartphones and tablets to personalize their multimedia consumption, the ship may have already sailed on multicast’s potential.

Verizon CEO Lowell McAdam’s CES 2013 keynote on Tuesday night wasn’t the news-extravaganza T-Mobile pulled off nearby, but he did let one interesting tidbit drop. While chatting with NFL Commissioner Roger Goodell, McAdam mentioned Verizon hoped to have the technology in place to “broadcast” the biggest U.S. sporting event, the Super Bowl, in 2014.

By broadcast, McAdam was referring to LTE-broadcast, one of the many multicast technologies that’s been kicking around the wireless industry for years. LTE-broadcast would turn cell towers into the equivalent of mini-digital TV towers that could multicast video, audio and even data to multiple users simultaneously.

verizon-4g-lteRight now mobile multimedia works through an on-demand unicast model. Every time you stream a video or a song to your smartphone, you get your own dedicated portion of the cell’s capacity to deliver your content, even if the guy right next to you is watching the same program. That unicast model and video’s intensive bandwidth demands explain why mobile video is such a network hog.

LTE-broadcast, however, would turn a portion of a network’s bandwidth into a multicast network, sending a single video or audio stream to multiple devices similar to the way TV and radio towers broadcast their programming.

If this all sounds familiar, you’re probably recalling Qualcomm’s FLO TV service of the last decade, which shut down in 2010 for lack of subscribers, devices and compelling content. Or perhaps the TV broadcasters’ own Dyle mobile digital TV initiative, which appears to be going nowhere very slowly. But there are some pretty key differences between those efforts and the LTE-broadcast technology that McAdam is talking about.

Qualcomm’s FLO technology required (and Dyle requires) a special receiver and therefore a dedicated TV handset to receive their respective transmissions. That pretty much doomed them from the beginning. But LTE-broadcast is based on the evolved Multimedia Broadcast Multicast Service (eMBMS) technology being standardized for LTE. Chipmakers like Qualcomm have already committed to supporting eMBMS in their future radio silicon. That means future handsets will be pretty much eMBMS-ready whether carriers chose to use the technology or not.

eMBMS also uses the same LTE radio infrastructure, requiring only upgrades to the network core. So if a carrier decides to get into the broadcast business, the equipment is largely in place. The barriers to entry are much lower for LTE-broadcast, but there’s still one big question: will consumers actually use it?

The age of personalized multimedia

hbo goThe problem is that an increasingly technically savvy public is moving away from broadcast models completely when it comes to digital content. Consumers are personalizing their radios with Pandora and Spotify. The reason HBO Go rocks is we don’t have to be at home a pre-determined hour –- or set our DVRs –- to watch the next episode of Game of Thrones. We just pull content out of the air whenever we please.

There are still plenty of people consuming broadcast video and audio on their TVs and car stereos, but on smartphones and tablets streaming is king. By imposing a broadcast model, carriers would be going against mobile data trends.

That’s why McAdam highlighted the Super Bowl as the ideal use case for LTE-broadcast. Blockbuster live events would attract hundreds of thousands of simultaneous viewers that would best make use of the technology. Verizon already streams entire NFL games through its NFL Mobile app, so being able to multicast those games would save it enormous amounts of network capacity — or so you might think.

There are a lot of cells out there

The thing about mobile networks is that they’re much denser than TV broadcast networks. Instead of using a single tower to cover a whole city, hundreds if not thousands of towers — each sporting multiple sectors — blanket any given metropolis with mobile broadband. Even if thousands of people in the same city are watching the same game on their phones, chances are few of them are going to be in the same cells at the same time. Multicasting effectively becomes unicasting if there is only one person receiving the transmission.

Nokia Siemens Networks' conception of a heterogeneous network

Nokia Siemens Networks’ conception of a heterogeneous network

What’s more, cells will start shrinking and multiplying as carriers begin deploying small cells and heterogeneous network (HetNet) architectures. The more cells in the networks, the less chance you’ll have users simultaneously streaming the same content in any given cell, unless you’re talking about big events. But playoff games and the State of the Union Addresses don’t occur everyday.

According to a new research report from iGR, carriers are weighing those factors, and some of them are leaning towards deploying LTE-broadcast selectively, targeting venues where people are most likely to stream the same content. Airports would be a good example, but so would a sports arena. Ticketholders might be watching the same games live, but they could all view the same replay videos simultaneously.

The iGR report also proposes that LTE-broadcast could turn our phones and tablets into mobile DVRs. We could subscribe to particular TV programs on apps like HBO Go. At set times, the LTE-broadcast network would schedule the download of various shows, beaming them down to thousands if not millions of devices simultaneously and caching them for later consumption. There’s nothing to prevent LTE-broadcast from being used for other types of media or data like digital magazines or device OS updates.

iGR projects that mobile video will account for 71 percent of mobile network data traffic in 2016. By utilizing LTE-broadcast, the study concludes, carriers could reduce capacity demand on their networks by 12.5 percent overall and by 15 percent at peak hours, the study found. The bottom line is unicast on-demand video will remain supreme, but a 15 percent capacity savings when the network needs it most is certainly nothing to scoff at.

Photo courtesy of Shutterstock user Peter Sobolev

  1. Nicholas Paredes Thursday, January 10, 2013

    It is also reminiscent of some of the cheaper cellular alternatives from the late 80s and early 90s. I suggested putting music on some of the these channels back then.

    It makes no sense to treat networks as one dimensional channels. They should adapt to usage requirements.

    Share
  2. why not just put regular over the air TV receivers in cell phones?

    Share
    1. The industry keeps trying that, but so far there doesn’t appear to be much interest in just regular free-to-air programming. It seems to work in other countries, but cab drivers are free to watch TV while driving in Asia and ours are not. :)

      Share
  3. In Korea, my former client SK Telecom made a major investment in 2002 to launch a satellite to beam broadcast TV to millions of their cell phones in S. Korea. Phones had been equipped with a sat receiver. The service launched, called TU media…the flopped.

    Turns out, as Kevin wrote, that people don’t prefer to use a personal communication device for watching linear broadcast streams. There are specific exceptions, such as the stadium concept…but there are still another 364 days in the year.

    If we move the conversation to batch-sync of programs, turning the phone into a PVR, well, that’s great. But that could be done even more cheaply over WiFi when the user is at home overnight.

    The eMBMS solution is OK, and has some possible opportunities as a content upsell, but I don’t see it putting a significant dent in unicast mobile video traffic.

    @derekkerton

    Share
  4. Lynnette Luna Monday, January 14, 2013

    The thinking around eMBMS has quickly turned from mitigating video traffic on the network to monetizing it. The monetization ideas are intriguing, but requires a new LTE Broadcast ecosystem consisting of eMBMS infrastructure, capable devices, enabling partners and programming that is significantly differentiated from OTT options from a quality and content perspective. I could see many forms of LTE Broadcast content being zero rated.

    Share
  5. Traditional broadcast on mobile exists and it works. It should not be considered as a new service per se, it is an extension of the residential TV business. It adds few minutes per day of viewership and is great for local news and sports (if you have the rights). In Japan, NTT Docomo launched recently a new generation of broadcast to mobile (called NOTTV) where all the programs are actually specifically designed for the mobile. This premium content is a mix of 3 live TV channels and some push content: and this really works. MMBI (the service provider) has gathered a great amount of subscribers in the last 9 months at 4$/months, growing fast.

    The big benefits of eMBMS, compared to traditional Mobile Broadcast, is that you don’t need an other chipset in the terminal. You can simply use the LTE one. Most recent LTE chipsets can even be upgraded through a firmware update!

    Today, as we see it, most of our customers will not use eMBMS for “saving bandwidth”. They will use it for new business models, to reach users they can’t reach today.

    I recommend also to have a look at the NOTTV business case, which is really interesting: a new brand is created, with original content for mobile, social media integration and a mix of pushed and live content.

    Share

Comments have been disabled for this post