1. Extending 5G coverage – This is the single biggest challenge and I'm excited about the potential for progress. The good news is there are lots of options -- from low-band frequencies (including dynamic spectrum sharing), small cell densification, and in-building systems, through to RAN software features such as downlink carrier aggregation -- that can improve coverage. This is going to cost a lot of money, however, so it's about being smart and bold in RAN choices.

2. Standalone (SA) 5G and 5G core – Advanced operators will introduce 5G SA (standalone), alongside non-standalone (NSA), with the first commercial terminals available from mid-year. We generally expect 5G core to be deployed on NFV infrastructure using virtual network functions (VNFs), with cloud-native network functions (CNFs) starting to make an appearance later this year. It's probably too soon to go full "cloud native" on a 5G core production deployment.

3. Transport network investment – Operators have started investing in 5G-specific transport, most obviously by upgrading the backhaul connections to cell sites to 10 Gbit/s or higher and by running fiber to small cells. In 2020 we should also start to see transport network architectures that can support fronthaul, network slicing and edge connectivity, all areas of significant network planning research in recent years.

4. Cloud and edge – US operators, in particular, are emphasizing the importance of edge computing for 5G and are highlighting their nascent edge infrastructure deployments. Everyone is watching to see what that means in practice for application performance. In the meantime, expect the wave of partnerships between operators and cloud companies to continue. I can see this emerging as quite a dynamic part of the market, albeit focused more on announcements than actual services and revenues at this stage.

5. Proving the end-to-end architecture – It's important for the industry to demonstrate a working end-to-end 5G architecture that is capable of supporting at least some of the advanced services envisioned for 5G. Ultra-reliable low-latency communication (URLLC) services and network slicing are of particular interest and it would be encouraging if the industry can get beyond the proof-of-concept phase in 2020 and into small-scale pilots in the field.