3G has always had a higher cost position compared to 2G and this is partly due to licensing costs associated with registered patents to enable voice over the UMTS network. There have been a number of enhancements made to the modules available on the market to reduce costs by removing the voice functionality. This removes the ability for the equipment to make a receive voice calls however in the IoT market space voice is not generally a requirement and therefore removing it from the end solution removes the license fee for voice and reduces costs.
As the IoT market does not necessarily require high data rates the enhancements that are brought with the 3G network are normally not needed and as a result 2G has remained the favoured solution for IoT equipment developers. One signifiant improvement offered by the 3G network over the 2G network is the network connection latency between two endpoints. This improvement means that the first byte will be received at the destination much more quickly and this means that the entire system latency will reduce. This opens the door for more real time, safety critical applications and allows for a much faster response times when sending and receiving information across the network.
As with 2G the major problem facing the 3G network is that it is being slowly phased out across the world due to newer technologies becoming available such a 4G / LTE which offer enhanced connection speeds, lower power consumption and additional user services.
The 3G network has already been scheduled to be sunset by a operators in a number of territories across the globe and many of these will be staring as early as 2019. There are plans to continue running the 3G network in certain regions until a number of specified dates going out to 2020 and 2025. However the point is that, like 2G, the 3G network will not have any more investment from any vendor and will not be being updated when cell towers go down. Therefore it is not recommended as a solution for new designs or for upgrades.