I speak on a lot of panels at tradeshows and often get the question of what is the best technology for M2M:2G, 3G or 4G? I usually answer, “it depends.” Well what does it depend on? It depends on a variety of factors, four of which I will describe below.
It depends on the operator strategy: some operators are going hard straight to 4G (LTE). Some operators are guaranteeing their 2G networks for 10 years or more. Other operators, because of spectrum challenges are pushing for everyone to get off of their 2G (GSM) network now and move up to 3G (HSPA) as fast as possible.
Why do they have different strategies? It could be that they are constrained by the amounts and types of spectrum they own or they may need to make room for their new technologies. It could be that they’ve completely paid off their 2G network. It could be that it is cheaper for them to move bits through 4G.
It depends on how cost sensitive the hardware is: if you need to shave pennies off of your hardware, today, you would probably try and use a 2G module, which is cheaper than a 3G module and much, much cheaper than a 4G module. But remember, hardware is usually only a small portion of the total cost of ownership of an M2M deployment.
It depends on the application: how dependent is the application to latency and how large are the average monthly or daily data requirements? It is very challenging to do large software updates over 2G.
It depends on the lifecycle of the asset: how long is the M2M asset going to be in the field? How expensive is it to change the modem? If it is only going to be out there for two or three years, your data and reliability requirements can be met, and you aren’t worried about the life of the network, then install the cheapest modem in you can find. However, if it is hard/expensive/impossible to change, do you really want to go with an obsolete technology out of the gate? Probably not. Future proof your application as much as possible.
Hopefully it is becoming clear that when you are planning for an M2M deployment, you need to think about it holistically and not just focus on the cost of the modem. You need to take into account how you expect your applications to behave now and how they might evolve over the next three or five years.
You need to understand where your carrier partners, customer needs and competitors are going strategically, so you don’t get stuck. And you really need to look at the total cost of ownership of deploying an M2M solution. If the differential costs of a module between 2G and 3G is $20, but you are spending $10 a month on your data costs, how much difference does that differential make over three or five years of operations?
Like I said, when it comes to choosing between wireless generations, it depends.