The transformational change is that the technology is being developed that will allow the industry to centralize the brains and the computing functions of the network. Today there are nearly 200,000 cell phone towers in the US and each cell phone tower requires a full set of switching electronics. Much of the smarts of the cellular network is done at these cell sites. That makes the cellular network somewhat unique in that most other types of networks have been able to centralize the brains and computing power of the network into hubs rather than to leave everything on the edge.
There are several groups now working on ways to start migrating the brains of cell sites back to regional data centers. Some people have called this moving the cellular network into the cloud, but that is really not a great description of it. Rather, this is a migration of computing and processing power from the edge back to a core like has happened with all other kinds of networks. The cable industry called this migration ‘headend consolidation’ where they created huge headend that can serve millions of customers.
This will be a transformational change because today it costs a literal fortune for a cellular carrier to implement a technology upgrade since they have so many cell sites. And this matters because upgrades are hitting the industry at a fast and furious pace. With a centralized cellular network, a cell company could upgrade the core software and electronics at only a few hubs since the cell sites will become little more than a transmitter site with little brains.
The second big change is that in 2014 we are seeing the cell industry adding tens of thousands of small cell sites. For a few years there have been network extenders called femtocells, but now the vendors in the industry have developed mini cell sites that are not a whole lot more than a cell site on a card. These small sites don’t have the same power as a full cell site, but they can be placed in areas where there is currently network congestion.
These small cell sites can be deployed in stadiums, downtown districts, convention centers and commuting corridors to provide extra call and data capacity where it is most needed. For example, I have a friend who I talk to regularly during his morning commute and I always lose him when he is crossing the 14th Street Bridge into DC. These mini cell sites ought to be able to fix the holes and dead spaces in the existing cellular network.
Finally is the change that will get all of the hype. Rumor has it that one or more of the cellular companies are going to start talking about 5G cellular networks this year. As I have discussed in the past, there are not even any networks today that are close to being able to call themselves 4G networks. The 4G standard begin with ability for a cell site to deliver 1 gigabit data speeds and there aren’t any sites today who can do 1/20th of that speed.
Sprint and T-Mobile coined the word 4G to promote some incremental enhancements to their HSPA+ and LTE networks. And so 4G was used as a marketing phrase to try to distinguish their technology from the competition. Then the whole industry followed suit and we now have 3G and 4G phones which use the same networks and have essentially the same capabilities.
There are dozens of little improvements to cellular technology being developed in vendor labs, and every time there is a new little tweak that makes speeds a little faster or that somehow enhances the customers experience the cell companies have been itching to say that they now have 5G. And it will happen. One of them will pull the trigger as marketing hype and the rest will follow. Ironically, by the time we finally get real 4G technology we will probably be selling phones in the market labeled as 10G.
So while the marketers make a lot of hype out of little changes in the network, the really huge change is the possibility to centralize the networks into hubs. Once that is done, a company could upgrade a few hubs and introduce a new technology improvement overnight. But that doesn’t sound sexy and is hard to market, so it will just quietly get implemented in the background.