LTE-Broadcast – In search of successful pioneers to make it happen

LTE Broadcast divides people as much as new US president Donald Trump. You either love it or you have long since discarded it as a technology. Whenever you find someone who loves it, they are prepared to forgive an extraordinary amount in order to see it as the promised land.

Talking this week to Claude Seyrat, CMO of French firm Expway, known for its sophisticated video software stack, we found some of that blind religious fervor that some companies reserve for their favorite technology, with Seyrat clearly showing a vested interest, and he says he has progress to announce at next week’s Mobile World Congress. We hope it’s a concrete deal for Expway’s sake.

The only change in LTE-B status since 2015 that we can find is that a few new operators and one or two vendors, have joined the LTE Broadcast Alliance late last quarter, with Hong Kong Telecom, CSL Mobile (also Hong Kong), Reliance Jio, Indonesian ISP Smartfren, Telecom Italia Mobile and Turkcell all joining long term members Verizon, Telstra, Korea Telecom and the UK’s EE. AT&T is also known to be working with the technology.

Each of those early supporters have had trials for well over a year, with Verizon saying that it has launched LTE-B services – for delivery of two Super Bowls and more NFL games over smartphones. But most of those guys had trials of the previous broadcast technology, MBMS, back in the mid noughties, and never launched services.

We asked Seyrat about the Verizon entry, “The stadium apps we always thought were a poor business case,” so his first comment seems to trash the one commercial deployment that LTE-B has.

Once Seyrat warms to the subject he dives straight in, “There is no one killer app for LTE-B, but once an operator has deployed one application which uses it, they will find many uses for it.” Then he almost contradicts himself by suggesting that Public Safety is the killer app.

Talking about such public safety standards as APCO P25 and the European Project MESA, there are few clues that they really will rely on video, broadcast or otherwise, in their first implementations. It is true that estimates of $6 billion of contracts have been made around P25, but early implementations are mostly about voice and video apps appear to be down the road somewhat.

“Not so,” says Seyrat, “Emergency personnel need to share videos, whether that is security based like nuclear power plants and train stations or drone video used to survey a natural disaster. These will all be mandated as being LTE-B based,” he said.

The use case is certainly there. If you are diagnosing an injured person or dampening a nuclear power reactor, you need multiple copies of the same video – and broadcasting it uses way less bandwidth. We can see that Seyrat MAY be right, but we have met these arguments before. Ericsson told us a decade ago that phones would all have the capability to view MBMS video built into them (the 3G version of Broadcast), and after that it would just be a matter of time before the particular version of 3GPP software was installed that supported it. Ten years on we see no-one using it.

But Seyrat is adamant that lots of operators are “about” to use it. He should know, or at least may know, if he is providing the software stack to control the video, either on device or in the network. “It will be mandatory for public safety networks to support LTE-B,” he proclaimed, not something widely known. “Then it will appear in the connected car” he added confidently. “Not immediately, but it will be the only logical way to send software updates to the connected car. If a car is hijacked or if there are software faults found, you will need these updates to be broadcast. In public safety evacuation maps, emergency alerts, must all be sent as a broadcast.”

The argument back in 2004/5 around multiple mobile TV technologies was that broadcast was the most efficient delivery mechanism, and that by 2010 there would be 100 million users of the technology. Well there are probably 100 million mobile video broadcaster users in just Japan and Korea, but few outside these areas, and none in those countries using MBMS or eMBMS, which is the basis for LTE-B.

Software updates should be sent this way, but currently they are not. Video should be sent in broadcast mode, but currently it is unicast. And the entire debate around video has been that it will take up 70% plus of all data to the smartphone, and it will choke the airways, so we MUST go to LTE-B. And still operators have not made the move.

So we asked Seyrat for reason behind this reticence on the part of MNOs to use broadcast technology? “We have to wait until the business case becomes mature. It does not have to be about choosing either Broadcast or Unicast. Once the software is installed in the base stations, then as soon as two copies of a video are being unicast across a particular cell, it would be more efficient for these to become multicast.”

Seyrat believes that MNOs will introduce it as a weapon to save bandwidth, and using it dynamically on a cell by cell basis, as and when it is needed. It could be a great way to allow customers zero rated video (see separate survey story in this issue about this) where they do not have to pay for data when video is used on a phone. “The customers don’t even have to be told they are on LTE-Broadcast, and it doesn’t have to be for the entire video transmission.”

His point it that it has taken the fear that their spectrum will be swamped with video to get them to look at this type of solution. What about getting programmatic advertising into broadcast streams? “Push the ads unicast ahead of schedule over network,” he counters, you will still save bandwidth on the video.”

“And what about sending TV services over fixed wireless? We know this will happen on 5G networks, but it is already happening in Mexico and Canada, and it is being tested as a broadcast using LTE-B.” We understand that Telus is one operator in Canada looking at this. But again, still only in testing.

“LTE-B could easily be used dynamically to send the top 4 or 5 channels in a cluster of cells, and all others can be sent unicast. It costs just $1,000 to upgrade each eNode B (base station). Think of LTE-B as part of a CDN, or instead of a CDN. If the content provider wants to ensure high quality, they may pay for their channels to be broadcast, and that would be like a CDN charge, around $50 per terabyte, or so.” Seyrat suggested that he knew of trials where this approach was being tried, but could give no operator names. It is interesting that it would be save the operator bandwidth but at the same time it would charge for the service? Isn’t that just a little greedy?

To us as Faultline Online reporter, we have seen the reticence towards LTE-B, and references to hidden costs built into it. Perhaps this technology is mature enough to push those costs aside, perhaps not. All we know is that until multiple operators use it for multiple services and talk openly in white papers about the bandwidth and money they have saved, the queue to be the pioneers in this market are very short.