The 2G / 3G / 70th Birthday Sunset

I have a friend who recently reached 70, having worked in and around the telecoms industry.  Hallmark and Scribbler don’t have many cards aimed at that specific combination, so it felt like time to design a custom card.  Being of that annoying engineering bent, I wondered what 70 would appear as in different number bases.

Imagine my surprise when I found that 70 in Base 18 is 3G, and In Base 27, it’s 2G.  Both of which are being “sunsetted” around the world, as mobile operators refarm their spectrum allocation to increase revenue from their 4G and 5G networks.  I couldn’t believe the coincidence.  And so, a birthday card was born.

Of course, engineers should never be sunsetted.  Experience counts a lot, as the song says in The Boyfriend.  But for 2G and 3G mobile networks, the end is nigh (and has already come in many countries).  What’s driving this change is the basic commercial imperative to support more users and more data, as the world becomes more connected and increasinly data centric.  With a limited amount of radio spectrum available, for which Governments charge exorbitant licence fees, the income of every mobile operator is based on how many phones and connected devices they can support, along with the ability to send them the amount of data they require.

With limited radio spectrum available, this is all about spectral capacity, which comes down to the amount of data you can send per Hz of spectrum and how many devices each base station can interact with at the same time.  That’s why we have the evolving standards, which have moved us from the original 2G digital networks in the early 1990s, up to today’s 5G.  There are many factors affecting how each variant of the telecom standards perform, but a simple overview looks something like this:

	2G	3G	4G	5G
Data speed	64 kbps	8 Mbps	50 Mbps	10 Gbps
Connections per sq km	~ 200	~ 5,000	~ 100,000	> 1,000,000

Any accountant seeing that a 2G network could support fifty times as many 4G subscribers would take the view that it’s time to turn off the 2G network and use it for 4G, which is what is happening.  The new infrastructure will cost more to install, but calculating that trade-off is what accountants are meant to be good at.  (History may have a different view – the even numbered “G”s are reputed to be unprofitable.)

Sunsetting is a bit more complicated than that, which is why many countries around the world are turning off their 3G networks before they turn off their 2G ones.  It’s over 30 years since 2G networks appeared and 20 since we had 3G networks, so very few people will still be using phones that only run on these networks.  However, many industrial monitoring applications have built in 2G and 3G radios.  These include cars, smart meters, utility infrastructure and building control systems.  These have working lives that may exceed twenty years and it’s difficult and potentially expensive to update them.   2G and 3G have already gone in the US.  However, because Europe started the journey to digital comms earlier than the rest of the world, there’s more of these deployed on this side of the Atlantic.  This has led to an interesting cat and mouse game between the mobile operators and Governments.

Because many of these embedded, connected assets relate to Government funded schemes (think smart meters, transport, infrastructure), Governments have a vested interest in keeping these networks running as long as possible.  On the other side, from the mobile operator’s viewpoint, keeping their 2G and 3G networks operational means they are limiting how many customers they can serve, which potentially hits their revenues.  With applications like smart metering, the mobile networks have stung the utilities for longer term contracts, but we’ve now reached the point where there’s general agreement that this can’t continue.  As 3G devices generally work on 2G, the UK operators sunsetted 3G first, with most 3G services now turned off.  2G will go by 2033 at the latest, but potentially much earlier, leaving us with a 4G / 5G world.  That’s probably a good thing, but it may provide an interesting vulnerability, which I’ll discuss in a future article.

But back to my friend.  Most professions have little ageist aphorisms, like “Old accountants never die, they just lose their figures” and “Old composers never die, they just decompose”,  but the engineering ones all seem very laboured.  In general they don’t seem to acknowledge that the engineering profession has progressed much past train drivers, but that’s another issue.  Maybe it’s because most engineers keep on thinking up new ideas and keep on asking why?  I still don’t know what I want to be when I grow up.

So my friend got a sunset card with weird numbering bases.    He has neither a Government, nor a mobile operator interested in replacing him with a newer specification, so I wish him and every other 70 year old engineer a very happy birthday.