The short and simple answer is yes, 6G is needed. Internet traffic continues to grow exponentially, and with that proportionally more capacity will be needed for the foreseeable future. Of course, it does follow a certain logic; after 1, 2, 3, 4 and 5, one would be right expect 6. But what with all of the hype around 5G could 6G possibly do to top it? Well, to begin, 5G’s 100 Gbps will begin to look pretty paltry against 6G’s promised Terabits per second. But this doesn’t begin to describe the scale of the iceberg jutting above the surface. A simple change or upgrade to the air interface does not a mobile generation make. Rather, a generational shift must be witnessable across the entire network.
One of the defining features of so called “Telecom Grade” is that of the so called “Five Nines” or 99,999% availability or roughly an acceptable downtime rate of 10 hours per year.
This was relatively easy to deliver and guarantee back in the good old days of circuit switch connections. But as we moved to the wonderful world of packet switched connections, things became suddenly, simultaneously much less certain and much more complicated. To achieve that high availability number in a packet switched world, mobile networks need custom-built hardware. That is, until, the arrival of 5G. And as a necessary result, the price tag on that hardware, justified by its “Telecom Grade Reliability” stamp, has come at a very, very high. Add to that, and hand-in-glove with reliability, interoperability which must exist within several layers, requiring a great deal of standardization and co-operation between parties.
As any telecom observer knows and to much his chagrin, “Best Effort” has been the operative paradigm of mobile networks since the arrival of 3G. Ironically, Best Effort is precisely the opposite of Telecom Grade, and a non-Telecom Grade operative standard previously associated with things like Ethernet type networks. As these packet switched and IT type networks developed and evolved concurrently, today it’s quite difficult to tell the difference between them, though their cost structures on the hardware side have been quite considerably different. And so, with the arrival of 5G, that cost gap will, as well, have disappeared. While radio devices will have a big price difference between 5G and WiFi, that has much more to do with regulation than with Telecom Grade Reliability.
IT networks have many advantages. Typically, and quite unlike Telecom UEs, IT devices are generally backward compatible. For example, they will feature routers capable of dealing with both IPv4 and IPv6 addresses. Contrarily, there are no Telecom Grade products capable of dealing with 3G and 4G, excluding of course phones that must be able to hop them. As such, IT interfaces and protocols are much easier to use and their patenting schemes are also quite different and all come in at a lower cost.
In case it hasn’t already become quite apparent, despite their minimal aspirational differences, as seen through the lens of technological development (think web 3.0 vs. 4G), they are quite odd bedfellows. Yes, this a story of Kane vs. Abel; of incremental improvement vs. generational overlays; agile vs. waterfall development; of revolution vs. evolution.
As it happens, and despite the hype, the 5G air interface is actually very similar to that of 4G. But not with hardware. This is not where we find the uniqueness of these two generations. Rather, it’s all about the software, and here, the difference is to say the least major.
Virtualization not only radically changes the name of game in a quantum leap seen in no other generational step. It also means that there is there is no longer any longer what can be called Telecom Grade hardware. Rather, the very same Cloud infrastructure that runs the rest of the show will run the telecom network. And here’s the rub. This will provide us an opportunity to break out from telecom generation thinking. We no longer need to make investment decisions worth billion every decade, but we can have backwards compatible functionality and incremental improvement as a constant state. That is massively disruptive at very least. The idea that the very same Policy Control Function (PCF) can be used for 5G and 6G radio interfaces, and that the very same AMF can provide service for both “generations”, is to say the least, very compelling for Telecom entrepreneurs.
In college, at the end 80s, I remember asking our professors which one they thought would win, Ethernet or Telecom. This was after I realized that there are two cables doing basically the same thing in every office. The answer they gave was, frustratingly, we know. With 40 year of hindsight don’t know, an with very good reason. What are the strengths or benefits of either network.
Now I think we have an answer.
Telecom Grade may be gone, but new generations don’t mean changing everything, including (for good or ill) the acronyms. Fortunately, it will only mean being more granular in general development but most particularly in mobile networks. This will not only make network investment decisions easier, but also increase the pace of technological development in the mobile space. Yes there will be a 6G air interface, but thanks to 5G, it will not need any new network functions, or at least to say no new basic functionality. Rather it will be able to be simple improved in the very same way that today’s clouds services are.