Now that we have officially launched into the age of 5G (at least, according to the carriers) we are beginning to see the results of the rubber meeting the road.
As was expected, as well as having been cautioned, by the technological community, early deployments of 5G did not, exactly, set the bar. These early deployments do not offer much in the way of 5G enhancements other than the 5G icon showing up on the smartphone’s home screen.
Any nascent technology’s early deployment often shows little in the way of improvements over existing platforms. It takes some maturing before benefits are realized. Unfortunately, the hype of 5G set the bar so high that it will be several years before both the technology and the verticals approach the lines drawn in the sand. However, those in the know are aware of all of this so the present focus is on overcoming the technical challenges of 5G. The hype is over simply because one cannot hype reality.
But what is reality are the challenges. During the pre-launch era we had a pretty good idea of what was happening, sub 6 GHz. Above that, especially way above that. Not so much. Therefore, most of the hard challenges will be in mmWave territory.
Obviously, one of the more significant challenges in this spectrum is 5G coverage. Remember, 5G hype promised ubiquitous coverage. It did not specify in what spectrum. However, to get that ubiquitous coverage, in mmWave territory, means antennae everywhere. I have discussed this, and the challenges that accompany this, in previous missives so I will not rehash it here. But, know there are significant challenges in putting an antenna ever couple of hundred meters.
Beyond just getting coverage, most of the SMEs I know agree that the only way to accomplish effective and efficient management of this coverage, once it happens, and across a disparate set of technologies, is through advanced platforms such as virtualization and software-defined networks (SDN). Then, throw in a dash of AI, a pinch of self-management, and a spoonful of self-configurability. Then add a few pounds of fiber for data hauling, some redundancy layers, Stir well and simmer for a while until thoroughly cooked and hope it tastes the way the recipe says it should.
Interesting analogy. But the truth is that making 5G work, particularly at mmWave, is akin to having certain key ingredients in a recipe that without which, neither will turn out well. In this case, the main ingredients would be AI, virtualization and SDN. And, since SDN and virtualized networks are not really ready for prime time yet, any large scale mmWave deployments will be a bit further into the future than some have predicted. Therefore, the first challenge in mmWave is simply getting it deployed.
In the sub-6 GHz world, “5G” or some semblance thereof, is a bit easier to deploy. However, much of what 5G promises (low latency, wide bandwidths) are difficult to achieve here. There are a handful of reasons but two of the major ones are that sufficient bandwidths are difficult to come by at these frequencies and advanced technologies, such as network slicing, do not scale well without lots of bandwidth. The challenge here is to figure out how this spectrum can be reworked to obtain wider bandwidths and make next-generation technologies work.
There are other significant challenges but until the industry can overcome these major ones, 5G is going to stay nascent at lower frequencies and spotty at mmWave.
The other major slice of the pie is the business case. While there is a slew of potential business cases on the books, until the promised advanced technologies of 5G arrive, most of these cases will be limited, or lie dormant because a large percentage of them rely on new spectrum to accommodate the data – and lots of it.
One of the most hyped business cases is having real-time access to scads of data. That promises a whole new world of marketing opportunity. It raises the bar an order of magnitude in the ability to analyze and understand customers’ interests, needs and expectations. Millennials and later generations expect instant, relevant, tailored content, which must occur in real time before their short attention spans drift, for example.
For this, and so many other number crunching (autonomous vehicles, smart “X”) apps AI, coupled with bandwidth and low latency, is the only real and tangible solution (until quantum computing becomes mainstream).
AI-driven analytics are the only way organizations can ingest the real-time data tsunami that 5G networks have the potential to generate and turn them into meaningful insights. Such insights are what will allow companies to interpret customer behavior and experience and generate recommended actions.
Eventually, machine learning will complement AI and take all of this to the next level. But I digress. Until we actually have the bandwidth, technology, and 5G wireless infrastructure, this will stay in the potential category as a major challenge.
Now, moving on to the rest of the challenges. 5G promises new apps and enhancements to existing apps. Some include:
But all of these require advanced platforms as well as available bandwidths. We are a ways away from single-digit latency and 100-megahertz channels. Therefore, the challenge of developing such apps lies in first developing the infrastructure. But tell me something new.
Perhaps the biggest challenge is not technology or commerce. It is momentum. We all know that the infrastructure must come first. However, it, by itself, will not be a significant source of revenue – that, for all intents and purposes, has become a given.
And business is not likely to bet on the come and lay out a bunch of money on the promise that, eventually, apps and the infrastructure will become available. That means that the deployment of 5G will be funded by, largely, the carriers. And, it will be expensive – another challenge.
All this means there are significant challenges still facing both technology and business. There is a lot of excitement around 5G. In the end, when all is said and done, years down the road, the picture may not resemble what is on the drawing board today. But we have to start somewhere and lay out plans. But challenges, as they arise, have a way of modifying plans. And such will be the case for 5G as much as anything else.