The dust is clearing from WRC-15’s last frantic days at the end of November. But all players are already beginning to look ahead to the next WRC four years from now, and in particular the technical groups’ preparations to assess bands for IMT2020 (which will identify additional spectrum for 5G).
Delegates at WRC-15, when deciding the future agenda for 2019, chose to focus on spectrum bands above 24 GHz. This is consistent with the formal exclusion of any discussion on UHF spectrum until 2023. Sub-1GHz spectrum for signal propagation and in-building penetration will also be necessary for 5G, although any existing IMT spectrum can be refarmed as 2G and 3G networks are shut down. There is also general consensus spectrum between 6 GHz and 24 GHz will also be necessary for 5G, but at the last minute WRC-15 removed this item from the agenda.
From the research done to date, there is general agreement very large quantities of spectrum will be needed, with 5G’s ideal bandwidth being at least 500 MHz, and perhaps twice this, as compared to the 20 MHz LTE best-performing bandwidth (an amount which can be boosted several times over by Carrier Aggregation). So each operator in each country will need many times more spectrum for 5G than they currently hold for 2G, 3G and 4G combined. The only place this can come from is the higher bands, with WRC-15 deciding to set up a task group to research 11 bands between 24.25 and 86 GHz, providing a combined 33 GHz of spectrum, and report back to WRC-19:
• 24.25 – 27.50 GHz (3,250 MHz)
• 31.80 – 33.40 GHz (1,600 MHz)
• 37.00 – 40.50 GHz (3,500 MHz)
• 40.50 – 42.50 GHz (2,000 MHz)
• 42.50 – 43.50 GHz (1,000 MHz)
• 45.50 – 47.00 GHz (1,500 MHz)
• 47.00 – 47.20 GHz (200 MHz)
• 47.20 – 50.20 GHz (3,000 MHz)
• 50.40 – 52.60 GHz (2,200 MHz)
• 66.00 – 76.00 GHz (10,000 MHz)
• 81.00 – 86.00 GHz (5,000 MHz)
This will not be an easy task as many of the bands are already in use for a large variety of purposes, with none of the incumbents likely to move without a fuss.
It is unlikely that 5G will have gained any scale by 2020 given the fact that key decisions on the assignment of the required large spectrum chunks will take place only by the end of 2019.