IEEE Future Directions
Related Articles
In many Countries around the world spectrum for the 5G wireless system is being auctioned with quite a significant spread of license fees. In Italy the auction just closed with a promise from several Operators to pay over 6 billion € over the next 5 years to get spectrum licenses. A first spectrum slice will be available starting next year and more will become available in the following years.
Does it make sense (any sense) to pay that much money to get this spectrum? Well, since Operators have bid I guess the answer is yes, so I am not going to discuss the pricing they are paying. I just hope they are right, and I also hope that the Government by making all those revenues out of the spectrum is not, as a side effect, diminishing the investment that will be required to exploit that spectrum…
What I want to discuss, I did already voice some thoughts in the past, is if 5G is really the silver bullet that some claim it is.
Depending on who you are listening to you will get different reasons why there is such a need for 5G (I remember hearing the very same reasons to justify the need for 4G…).
Let me take 5 of them in this and the next post, summarised in a nice article just few months ago, 5 Reasons why 5G is the Future:
- Autonomous cars
The article claims that autonomous cars will need to talk with one another and with the infrastructure and they need very low latency, in the order of a few ms, something that only 5G can deliver.
OK, let’s consider that. First 5G is not going to be ubiquitous, it will start deployment in urban areas and then will slowly extend to rural areas. Let’s say, with the experience of penetration of previous Gs, that by 2030 we should have an almost complete coverage by 5G. That’s nice. It is probably in synch with a real significant penetration of self driving cars. However, the question is: what will autonomous cars do before 5G will provide full coverage? What will an autonomous car do if for several years it will have to drive in areas not covered by 5G? What will cars do if there will be a network problem that makes 5G unavailable?
Well, it is easy to understand that because of the above car manufacturers are not betting on 5G, they are not designing autonomous cars with the assumption that 5G will be there, everywhere and at any time. They are designing autonomous cars that can be “autonomous”, i.e. independent of any specific infrastructure. Will 5G help? Sure, Will it be the silver bullet for autonomous cars increasing their safety, decreasing their cost? NO.
By the way, all this talking about few ms delays is simply not true, unless you are considering just the radio link. As soon as you involve the main network, and services in the cloud the ms increase rapidly. There are architectures being proposed to have services at the edges (in the fog, or clouds at the edges) but it is quite unlikely to see these clouds at the edges in rural environment, just because once in a while an autonomous car may need low latency.
In a nutshell: I don’t believe autonomous cars need 5G, nor that Operators can make money out of autonomous cars connectivity on 5G. - Smart cities
The article claims, correctly, that smart cities will rely heavily on connected devices many of which will have to interact with autonomous cars. Additionally, smart cities will have to rely, and leverage, on a variety of data to understand what is going on and take appropriate measures.
I cannot agree more with these statements. The only problem I have is that all of that can be done using current wireless systems, for sure LTE (but also 3G is pretty good for those things). It is obviously a matter of coverage (and for at least 5 years LTE will have a greater footprint than 5G, I quite frankly do not see 5G reaching a greater footprint that present wireless systems). Yes, 5G is expected to be deployed using smaller cells (but that will take time and will not happen everywhere), it will have a broader spectrum, hence higher capacity, but what you have today is good enough for smart cities.
I spoke with many people in municipalities round the world and although many were exited by 5G (because newspapers tell them they have to be) I was not able to discover a single smart city service that cannot be delivered using LTE (and many of them will be pretty good on 3G).
Will 5G be exploited in smart cities? Of course! Do smart cities have to wait for 5G to deliver any service? NO. Will Operators make significant additional money from smart cities services thanks to 5G? No way.I’ll continue tomorrow. I know that many of you may not like what I said but please, let me know where I am wrong by explaining services that can only be provided with 5G in the above two areas or that would result in significant lower cost if delivered on 5G. Comments are welcome.
