资源描述
Vision 2030 Insights for Mid-band Spectrum Needs July 2021 Copyright 2021 GSM AssociationThe GSMA represents the interests of mobile operators worldwide, uniting more than 750 operators and nearly 400 companies in the broader mobile ecosystem, including handset and device makers, software companies, equipment providers and internet companies, as well as organisations in adjacent industry sectors. The GSMA also produces the industry-leading MWC events held annually in Barcelona, Los Angeles and Shanghai, as well as the Mobile 360 Series of regional conferences. For more information, please visit the GSMA corporate website at Follow the GSMA on Twitter: GSMA1 The full Coleago research report IMT spectrum demand: Estimating the mid-bands spectrum needs in the 2025-2030 timeframe is available on the GSMAs website at Mid-band spectrum needs Meeting mid-band spectrum needs is vital to 5Gs future and requires forward-planning from policymakers. Summary The speed, reach and quality of 5G services depend on mobile operators having timely access to the right amount and type of affordable spectrum. Mid- band spectrum is especially important as it offers a good mixture of coverage and capacity for 5G. This report presents the GSMAs vision for how much mid-band spectrum mobile operators will require between 2025 and 2030 and provides some options for operators to meet this demand. It includes an overview of research commissioned by the GSMA from Coleago Consulting 1 and helps policymakers understand how much 5G mid-band spectrum is required. It is designed to lay out a vision of how policymakers can make sure 5G reaches its full potential and maximise the socio-economic impact that it can deliver. The analysis looks at how much mid-band spectrum 5G networks will need for reliable high-speed mobile broadband services in heavily populated urban areas. This spectrum will allow MNOs to provide Fixed Wireless Access with fibre-like speeds. Commercial mobile operators can also support the needs of a wide variety of use cases from vertical sectors. Each vertical sector will generally have multiple different requirements (e.g., low latency, high throughput, long battery life, localised coverage, wide area coverage, etc). However, these differing requirements need different spectrum and network resources. MNOs are well placed to support such diverse requirements due to their expertise and wider spectrum assets. Mid-band Spectrum Needs 1 The GSMA recommends that governments and regulators: Plan to make 2 GHz of mid-band spectrum available in the 2025-2030 time frame. This is the average value needed to guarantee the IMT- 2020 requirements for 5G; Carefully consider 5G spectrum demands when 5G usage will be reaching its peak, and advanced use cases will carry additional needs; Base spectrum decisions on real-world factors including population density and extent of fibre rollout; and Support harmonised mid-band 5G spectrum (e.g., within the 3.5 GHz, 4.8 GHz and 6 GHz ranges) and facilitate technology upgrades in existing bandsThe research assesses how much spectrum is needed to deliver the ITU IMT-2020 requirements of reliable 100 Mbps end-user download speeds and 50 Mbps upload speeds based on population density metrics as its means of analysis. It excludes large commuting or tourist populations in some of the selected cities. Studying 36 cities in the 2025-2030 time frame shows that: Densely populated cities need, on average, a total of 2 GHz of mid-band spectrum. Precise spectrum demands vary depending on population density, fibre availability and other factors. This means there is no simple correlation between a countrys income level and its spectrum demand. However, additional spectrum is required beyond existing ITU, regional or national plans in all cases. High income countries will require from 1320 MHz to 3630 MHz Upper middle income countries will require from 1020 MHz to 2870 MHz Lower middle income countries will require from 1320 MHz to 3020 MHz IMT-2020 requirements will be at risk with less spectrum, and significantly more base stations would be needed without sufficient assignments. Where densification is possible, the total cost of networks would be 3-5x higher over a ten- year period if there is a deficit of 800-1000 MHz. This equates to $782mn-$5.8bn in extra investment in each city. Additional base stations will generate a carbon footprint 1.8-2.9x higher without sufficient spectrum. The additional network densification mentioned above would increase mobile network energy consumption in the cities by 1.8- 2.9x, as well as in the manufacturing process. Importantly, such a high level of densification may not even be feasible for other reasons (such as too much interference, site availability, restrictive electromagnetic field rules). This can be avoided through the timely availability of the right spectrum. Affordable fixed wireless access will raise demand. The additional spectrum in mid-bands will allow each cell site to support 3.5-6x more homes with 5G FWA. This would create significant cost- savings in network roll-out and drive affordable connectivity in areas where other broadband solutions are not economically viable (e.g., where fibre is not widely available or remains limited to bigger cities). Mid-band Spectrum Needs 2 WRC-23 is a crucial opportunity for mid-bands With WRC-23 approaching, positive engagement on mid-band solutions for IMT will provide vital support to the harmonisation of spectrum and give clear technical guidance for regulators. Coordinated regional decisions will lead to a WRC which enables the future of 5G and supports wider broadband take-up by increasing capacity and reducing costsSection 1: How much spectrum is needed in major cities between 2025 and 2030? Delivering reliable high-speed mobile broadband services is a particular challenge in densely populated areas such as major cities. The large number of people in relatively small areas can easily over-burden the capacity of mobile networks. As a result, mobile networks spectrum requirements are typically higher in these urban settings. Coleago analysed how much spectrum is needed to meet key ITU targets for 5G (IMT- 2020) in 36 cities around the world in the 2025-2030 timeframe. At this time, 5G take-up is expected to be significantly higher than today. The analysis calculates how much spectrum is needed to reliably deliver 100 Mbps download speeds and 50 Mbps upload speeds to end users. The Coleago analysis found that the amount of spectrum needed depended on a number of factors. These included: Population density (excluding temporary population); Predicted amount of available spectrum by 2025; Geographical separation of base stations (inter- site distance); 5G technology used in every band, with MIMO upgrades and both outdoors small and macro cells; Percentage of high-band, indoor small cells and Wi-Fi offload; and Cellular network activity factor. Activity Factor The activity factor is an assumption surrounding what percentage of human and machine connections require the 100 Mbps download and 50 Mbps upload connection at any one time during the busiest hours. This varies depending on 5G usage levels and is likely to be greater in higher-income markets initially, growing in time in every market, based on both mobile and FWA requirements. Table 1 (below) shows, as a range, the total amount of spectrum that is predicted to be needed in each city based on assumptions for the network activity factor and high-band offload. It shows that additional spectrum is needed in every city and that, on average, a total of 2 GHz in mid-bands will be required. While the amount of additional spectrum is lower in some markets and/or less populated cities, it is important to note that the amount required is still far greater than what is currently planned for release. This begs the question: where are policymakers going to find the right amount of mid- band spectrum? The activity factor is expected to rise over time and thus the projected amount of spectrum will rise as well. This is likely as 5G usage growth is driven by smart cities, including intelligent transport and connected video cameras, and by consumers and businesses for mobile use and for FWA. Mid-band Spectrum Needs 3Table 1: Total mid-band spectrum needed for 5G in the 2025-2030 time frame 3020 3690 3260 3130 2930 2870 2830 2810 2800 2520 2510 2480 2450 2450 2430 2380 2340 2250 2230 2100 2080 2050 2010 1950 1830 1810 1780 1780 1720 1690 1600 1570 1540 1480 1480 1470 1200 2440 2580 2420 2410 2350 2140 2340 2130 1850 2060 2040 2050 2050 2000 1980 1720 1880 1780 1580 1560 1690 1630 1540 1500 1500 1510 1450 1320 1350 1300 1300 1270 1260 1260 1020 Hong Kong Lagos New York Tokyo Sao Paulo Paris Yangon Moscow Beijing Mumbai Madrid Barcelona Riyadh Istanbul Jakarta Mexico City Ho Chi Minh City Bogata Bangkok Cairo Nairobi Johannesburg Berlin Rome Amman Makkah Lyon Milan Tashkent Hamburg Marseille Munich Amsterdam Baku Minsk Tehran 0 500 Lower estimate of mid-band spectrum needs 1000 1500 2000 2500 3000 3500 4000 Upper estimate of mid-band spectrum needs Mid-band Spectrum Needs 4Cost of densification If there is a shortfall in the amount of spectrum available, where possible, mobile operators will be required to densify their networks by deploying more small cells. However, this imposes additional costs, which are ultimately borne by mobile operators customers. Such a densification may not even be feasible for other reasons (e.g., interference scenario, site availability, restrictive electromagnetic field rules), while large numbers of cells will also increase the overall power consumption (see table 2) as well as having an aesthetical impact. Coleago calculated the environmental impacts for three cities assuming a conservative spectrum shortfall of 800-1000 MHz. Table 2: Financial and environmental costs in three cities with an 800-1000 MHz spectrum shortfall City # of additional small cells Cost of additional cells over 10 years Relative increase in total network costs Increase in network power consumption Paris 27,505 $782mn 3x 1.8x Mumbai 195,785 $5bn 4.3x 2.9x Mexico City 178,236 $5.8bn 4.9x 2.5x Densification alone will not solve the lack of spectrum and high levels of densification would not be physically possible in some of these cities. On top of this, after a certain level of densification, interference between sites would also trigger the need for additional spectrum. Mid-band Spectrum Needs 5Section 2: How much mid-band spectrum is needed for Fixed Wireless Access? 5G delivers broadband FWA with fibre-like speeds, but at a fraction of the deployment cost. This can have a profound impact in areas with limited access to high-speed fibre. Mid-band spectrum is well equipped to provide 5G FWA in population clusters such as towns and smaller urban areas, especially where other options are expensive or unavailable. Mid-bands complement lower bands (e.g. sub-1 GHz), which are useful for rural and remote areas where populations are more spread out, and mmWave bands, which can provide access in more densely populated areas with the very fastest 5G speeds. FWA connections typically place a much larger capacity burden on mobile networks than a smartphone. Homes or offices can have several concurrent users who often consume large amounts of video, including televisions. To show the economic impact of rolling out FWA in different environments, Coleago analysed how many homes could receive 100 Mbps download and 50 Mbps upload speeds using a single 5G FWA cell site given different amounts of spectrum (see figure 1). They found a single 5G mid-band cell site could cover local population clusters of 540 households, which equate to a small town or village. Coleago then applied these findings to assess the cost of extending broadband rollouts in India and Europe and compared these with the fibre alternatives. The model assumed a higher activity factor (50% rather than 10-25%) to represent the heavier usage of FWA connections. They found that meeting the proposed 5G spectrum needs in mid-bands allows a 5G cell site to support 3.5-6x more households when adding 1 GHz and 2 GHz to a standard 400 MHz upper mid-band spectrum baseline. This dramatically improves the business case for these services and thus the ability to widen broadband access through affordable FWA 5G. In India the cost of covering rural towns using 5G FWA was found to $9.8bn less than using fibre. In Europe, the savings were 42bn. Figure 1: Number of households supported per mid-band cell site with different amounts of spectrum 600 500 400 300 200 100 0 Homes supported per site + 1 GHz additional upper mid- bands 400 MHz upper mid-bands + 2 GHz additional upper mid-bands 90 315 540 Mid-band Spectrum Needs 6Section 3: MNOs can meet vertical needs with licensed mid-band spectrum 2 Read more about the impact of high spectrum prices at: 5G can have a profound impact on vertical sectors. It is designed to meet ambitious requirements in terms of speed, latency, security, and support for connected machines and sensors. Alongside small cells, network slicing, IoT, cloud computing, machine learning and robotics, the stage is set for what has been dubbed industry 4.0. However, network provision for vertical sectors spectrum needs to be carried out in a spectrally efficient way and set-asides for private networks can have a damaging effect. Limiting the amount of spectrum available to MNOs makes it harder to deliver on 5G targets and creates artificial scarcity, leading to high auction prices. Higher auction prices are strongly linked to reduced network investment and thus slower rollouts, worse coverage and slower data speeds. 2 Each vertical sector will generally have multiple different requirements including low latency, high throughput, long-battery life, localised coverage, wide area coverage, and others. However, these differing requirements need different spectrum and network resources. MNOs are well placed to support such diverse requirements due to their wider spectrum assets and broad expertise. Mid-band Spectrum Needs 7Section 4: How to meet the demand Meeting the 5G spectrum demand in the 2025-2030 time frame will be challenging and policymakers must already consider the best options for inclusion in their digital connectivity plans and spectrum roadmaps. Mid-bands, which offer a good mixture of coverage and capacity benefits, should be at the core of this effort. The majority of the first commercial 5G networks are relying on spectrum within the 3.3-3.8 GHz range. Other bands which may be assigned to, or re-farmed by, operators for 5G include 1.5 GHz, 1.8 GHz, 2.1 GHz, 2.3 GHz and 2.6 GHz. These are still not available in all countries. More spectrum will be needed to maintain 5G quality of service and meet growing demand in the longer term (e.g., 3.3-4.2 GHz, 4.8 GHz and 6 GHz). However, the number of options may be limite
展开阅读全文