Spectrum is expensive, and as this Light Reading article points out, anything that promises additional capacity for a given chunk of spectrum sounds good. The current focus is orthogonal time frequency space (OTFS) modulation, which some are promoting as a fundamental technology for (gasp!) 6G. But is sounding good enough; is this yet another way of pushing 6G hype just as people have finally decided that 5G is over-hyped beyond redemption? You’ve got to start by asking what OTFS-generated extra capacity would actually be used for.
Roughly doubling the capacity of current spectrum could mean doubling the potential per-user connection speed. I don’t have a shred of doubt that will be one of the big PR hooks sunk into this debate, because “faster” always sounds better…even when it isn’t. 5G demonstrated that given the essential limitations of what we can do with mobile devices, users may not see any difference between a “fast” and a “normal” connection. The one place where I think we could argue that OTFS could contribute would be FWA, where it could offer a better Internet connection to more people than the older 5G technologies.
Connecting more devices is another question mark. In most markets, the number of smartphone users has plateaued, so the only way that “more devices” becomes relevant is if we assume IoT connections would explode. They probably will, but not IoT connections using mobile technology. Recall, too, that the notion that 5G would create an IoT land rush was one of the oft-cited justifications for 5G, and it didn’t happen.
So how about the value of OTFS in reducing the cost of spectrum? If you could do twice as much per unit of RF, couldn’t you buy half as much spectrum? That sounds logical too, but it ignores two basic truths. The first is that most of the suitable spectrum has already been bought. Does anyone think that the companies who bought spectrum at auctions could go hat-in-hand to the governments involved and get a rebate because they don’t need as much?
But couldn’t they sell it off? Wouldn’t OTFS allow more competitors to acquire (directly or third-party) spectrum, increasing the number of players? Yes for sure it could, but you have to wonder why those who already have spectrum would sell it off to facilitate competition.
Finally we have the “high-mobility” justification, and here at least we have a glimmer of value. “High mobility” here really means “moving fast”, faster than an automobile could drive and faster than most trains. The problem with really fast-moving users is that they create the Doppler effect on signals, and the faster they move the more likely it would be that the signal couldn’t be decoded properly. OTFS is much more tolerant of that.
But what, besides very fast aircraft, really challenges current mobile technology? Nothing, and unless we believe we’ll all have personal superfast flying cars in the near term, we can assume that we’d be packed into a commercial aircraft. Given that, couldn’t we provide a single fast link, perhaps an OTFS satellite, to the aircraft itself, then WiFi to the passengers? This would have the added benefit of making the new approach compatible with current smartphones.
And so we introduce the question of phones. In order to use OTFS you’d need a specialized smartphone/device capable of receiving and sending it. The phone manufacturers would love that, but the problem is that unless there’s a real benefit to the average smartphone user, OTFS wouldn’t likely push everyone to upgrade their phones in the near term. That would mean that OTFS spectrum might have to be new spectrum, so the older devices could still be supported.
It seems to me that we have two possible OTFS scenarios on tap. In one, we get limited OTFS deployment for the FWA and high-mobility applications, but we don’t push it aggressively for the average smartphone user. The other is that we push OTFS, via the 6G connection, as the next step in mobile service evolution.
We’re already talking about 6G, absent any clear technology decisions relating to it, much less actual deployment. That’s happening because we’ve used up 5G as a way of generating clicks and ad serves to users. So do we really believe that the same thing won’t happen to 6G? There are realistic missions for things like the only relevant 5G Core feature, network slicing, but who wants realism? Everyone wants clicks, so we shouldn’t expect 6G realism of the kind our first option would require.
But even setting my hype cynicism aside (reluctantly, of course) there’s the question of whether those missions would really justify any deployment. We have FWA now using 5G, and we have satellite broadband for thin areas and fiber for higher demand densities. What niche would an OTFS FWA service fill?
So are we left with our second option, making OTFS and 6G the explicit next step in mobile services? It would seem so, but that has its own challenges.
Top of the list is the inevitable delay in getting something out there in the real world. How long did it take for real 5G to evolve? I think that given our 5G experience, we could reasonably say that 6G could be real by about 2028 and pervasive by 2030. Why then would telcos be tossing money at the technology now? Are the niche applications so interesting? I don’t think so. We have here another example of how Field of Dreams leads inevitably to UFO-think.
Operators don’t want to believe that they can’t somehow go back to Alexander Graham Bell and Thomas Augustus Watson. What they want is a technology that can promise that, which is very different from one that can certainly deliver it. For one thing, there is no technology that could do that, but for another what the telcos really need now is the promise, something they can use to help buoy up their stock for a while. 5G did that, but it’s feet of clay are getting clear even to Wall Street. What’s after “5” if not “6”? And 6G, being a current glimmer in a standards-writer’s eye, has the advantage of a UFO. Since it’s not likely to land immediately and present itself for inspection, you can assign any properties to it that you find helpful.
I’m not saying that OTFS isn’t a good idea (it is), but that it’s an idea that isn’t likely to create any near-term payoff for operators. That’s bad because they need things with a near-term payoff. You can’t be asking for subsidies from big tech on one hand, and on the other focusing your attention on technology shifts that cannot possibly alleviate your profit per bit issues. The first requirement any telco asking for subsidization in any form must meet is the requirement to do everything possible to put their own house in order. I don’t believe that OTFS can help with that, and in fact focusing on it likely perpetuates a service focus that’s destructive to profits in the long run.