Spectrum has always been the lifeblood of the mobile industry, an asset fought over in terms of what represents fair allocation between operators in a market and what mix of licensed and unlicensed allocations best serve the technology needs of populations. Strategy Analytics views the 6 GHz band as a critical resource for the mobile industry, one that will enable 5G operators to deliver on their longer-term value-generating strategies for both consumer and enterprise markets. It is a band that we believe regulators and standards bodies should prioritise in the run up to WRC-23.
Regulatory approaches to spectrum awards have had an impact on the early success with 5G around the world.
- Countries that have awarded large allocations of mid-band spectrum are typically enjoying early leads in 5G user experience and adoption, while countries where operators have been forced to turn to low band or mmWave spectrum have been weaker.
Mid-band spectrum will be critical to meeting the long-term capacity needs of 5G, built on scenarios of 250 GB/user/month driving by high-quality video, FWA, and enterprise and industrial Internet use.
- Operators will struggle to sustain 100 Mbps city-wide 5G service experiences using existing low, mid and high-band spectrum resources, with a share of the 1.2 GHz of available 6 GHz spectrum crucial for long-term industry growth.
As governments turn to telecoms operators to deliver Gigabit Societies and enable post-pandemic economic recovery, they need to provide the tools: more mid-band 5G spectrum is a key resource and the 6 GHz band the primary candidate here.
- Regulators should assess the opportunity presented by IMT identification of the 6 GHz band at WRC-23, involving research into the coexistence of 5G with current incumbents in the band and field testing IMT deployment. These activities will be essential for the cultivation of the 6GHz ecosystem for 5G.
The influence of spectrum policy can be seen in the performance of today’s 5G operators.
Mid-band spectrum has been a key part of many 5G launches globally, with the 3.5 GHz band (n78) most widely used, but also activity in early markets at 2.5 GHz (n41) and 4.5 GHz (n79). There has been growing activity for 5G in 2020 in low bands, in particularly using dynamic spectrum sharing (DSS) in sub-1 GHz to provide a coverage layer and in mmWave bands (n257-n260). However sub-6 GHz mid-band spectrum has been key to delivering a strong 5G user experience. It is interesting to note that, when we look at Opensignal data on the average 5G user download speeds experience in Q3 2020, it is arguably the countries that have a more fragmented 5G spectrum licensing regime that have so far failed to really deliver on a high-quality 5G user experience.
While operators in Saudi Arabia and South Korea are all using 100 MHz of 3.5 GHz spectrum (with the exception of LG Uplus’s 80 MHz allocation), the picture is less impressive at the bottom end of the experience curve, often linked to regulatory delays in clearing or licensing sufficient new spectrum:
- The US is using 3.55-3.7 GHz for shared access CBRS, with an auction of 3.7-3.98 GHz starting in December. While T-Mobile is rolling out 5G in the 2.5 GHz band, both it, AT&T and Verizon have generally had to weigh up the benefit of some urban mmWave 5G (with limited coverage) and sub-1 GHz DSS (with limited speeds);
- In the Netherlands, the government has yet to auction the 3.5 GHz with operators launching DSS into existing low and mid-band spectrum;
- In Germany, while all three operators hold 100 MHz of 3.5 GHz spectrum, they have accelerated 5G coverage with liberal use of DSS;
- And in Hong Kong and the UK, initial 5G spectrum awards have generally seen the incumbents holding 40-60 MHz of 3.5 GHZ spectrum.
The experience in China mirrors that in South Korea and Saudi Arabia: large allocations of mid-band spectrum for 5G services are enabling the operators to aggressively roll out services and rapidly sign up users. China Mobile holds 260 MHz of spectrum for 5G (2515-2675 MHz and 4800-4900 MHz), while China Telecom and China Unicom are effectively sharing 300 MHz (3400-3600 MHz, plus 3300-3400 MHz for indoor use). According to the
Ministry of Industry and Information Technology, by October the three operators had deployed 690,000 5G base stations and were serving 160 million 5G devices. 5G’s share of subscriptions stood at 10% in China and 14% in South Korea in October, significantly in excess of levels seen in other markets where, prior to October’s iPhone 12 launch, few operators have added more than 2% of their base to 5G.
Mid-band spectrum represents a sweet spot for operators looking to manage 5G coverage and capacity requirements balanced against deployment and operational costs. 3.5 GHz has delivered on this and in many deployment scenarios has also allowed 5G to be added to the existing base station grid with minimal need for additional site infill. Strategy Analytics’ recent conversations with Switzerland’s
Sunrise on its nationwide 3.5 GHz 5G deployment experience highlighted the mid-band benefits of upgrading 4G sites with a single compact 4G and 5G antenna configuration, with strategic use of 5G small cells to improve cell-edge coverage (outdoor and indoor) and provide a supplementary capacity layer in heavy traffic locations
Today’s mid-band spectrum allocations will only get 5G operators so far.
More mid-band spectrum will be needed to meet the long-term capacity requirements for 5G. Strategy Analytics predicts that by 2030 57% of the world’s mobile connections will be on 5G, accounting for 78% of industry value. The traffic growth will follow the volume there: one long-term forecast scenario from the ITU had average global mobile traffic per subscription per month at more than 250 GB/month by 2030. This is clearly a dramatic leap from the 7.2 GB/month carried in Q3 2020, however it is worth noting that:
- The top 5% of 5G subscribers in South Korea generated 134 GB/month in Q3 2020. Five years’ ago, the top 5% of 4G subscribers in South Korea generated 26 GB/month, less than overall 5G averages today.
- The move to UHD and 360o video content will drive significant mobile traffic increases. In South Korea, 3D AR and VR video typically consumes 400-600 Mbytes of traffic per minute, so 250 GB of data could easily be consumed with 15-20 minutes’ daily use of the current generation of XR experiences. Even at the ‘lower end’ of the market we are seeing greater use of 720p video multiplying traffic loads in developing regions, and believe there is dramatic upside potential in video use across the board.
- FWA services on 4G today, but 5G tomorrow, will drive up averages. For example, 4G FWA use in Tunisia averaged 151 GB/month in Q3 2020 up from 67 GB/month a year earlier. Fixed broadband use is a useful benchmark here, with UK averages growing from 57 GB/month to 315 GB/month over the last five years according to Ofcom.
- Enterprise and Industrial Internet requirements could be an order of magnitude higher than consumer. For just one use cases, connected cars are forecast to generate anywhere between 8 GB in mobile data traffic (Cisco by 2023) and 4TB of total data by all sensors (Intel) per DAY.
So where will the next for mid-band spectrum opportunities come from?
While there are clearly longer-term opportunities for operators to re-farm sub-3 GHz mid-band spectrum to 5G, this will deliver marginal gains compared with the need to identify significant new swathes of spectrum. The 6 GHz band (5925-7125 MHz) represents perhaps the only large new block of mid-band spectrum that could help support 5G capacity requirements.
In a detailed white paper (pdf), Coleago Consulting modelled how much 6 GHz spectrum would be needed to deliver a 100 Mbps user experience across dense urban environments in five cities. Against a baseline ‘activity factor’ (concurrent use in a cell) of 5-10% today, the use of 700 MHz of 6 GHz spectrum would allow networks to deliver 100 Mbps with a 10% activity factor in Lagos and Moscow, a 15% activity factor in Paris and Sao Paulo, and a 20% activity factor in Tokyo. Using the full 1200 MHz of 6 GHz spectrum would increase those activity factors by 5 percentage points in each city.
Availability of the 6 GHz band varies by region, with a number of countries, most notably the USA and South Korea, already having released the entire 1200 MHz for unlicensed use. In Europe, the bottom 500 MHz (5925-6425 MHz) has also been released for unlicensed use and consultations are taking place in a number of other countries and regions. While there is a good argument to be made for releasing additional unlicensed spectrum, with
Wi-Fi 6E touting wider channels and less interference for low-latency high-capacity connectivity, there is also a need for the 6 GHz band to deliver high quality urban 5G user experience.
R&D and standardisation work for 5G in the 6 GHz band is a priority for 2021.
The good news for mobile operators in some regions is that the 6 GHz band has been included in Agenda Item 1.2 of WRC-23, specifically 6425–7025 MHz (for Region 1) and 7025–7125 MHz (globally). It is rising up the agenda at a range of organisations covering spectrum standardization and is also covered in a
study item at 3GPP, examining both licensed and unlicensed NR use in the band.
While regulatory momentum may be building in Europe, it is perhaps China where vendors, operators and regulators should look for inspiration in the run up to WRC-23. During WRC-19, the Chinese delegation actively promoted 6 GHz as a candidate band for IMT services and last month China’s IMT-2020 Promotion Group confirmed that it will launch research and testing of 6 GHz in 2021. This group co-ordinates 5G technology research and international standards co-operation in China and has been the main organisation driving the 5G R&D testing so critical to the maturity of China’s 5G ecosystem. The tests will be a significant milestone for the IMT industry, highlighting the growing industry focus on the 6GHz band and the accelerating industry progress in this area.
(China’s IMT-2020 Promotion Group announced the launch of 6GHz Tests in 2021 at PT Expo China in October)
Long-term spectrum planning is critical for regulators looking to support the success of their 5G industries. If policy expectations of a Gigabit Society were not already enough, then the prominence of 5G and AI as platforms for industrial transformation and growth in a growing number of governments’ post-coronavirus ‘Digital New Deal’ plans for economic recovery add further weight to the need for more 5G spectrum. Delivering a fair share of the 6 GHz band for 5G services will help to hit the longer-term sweet spot where the demand for high-volume and ubiquitous high-performance 5G can neither be met by lower mid-band spectrum nor mmWave.