One of the most visible impacts of Covid-19 (coronavirus) has been the increased usage of digital technologies, making high-speed internet access more important than ever. Building robust infrastructure, maintaining resilient networks, attaining wide 4G coverage and preparing 5G networks have become essential components of fulfilling today’s network requirements.

To execute on these requirements, operators need access to spectrum in new and existing bands. Only if it was this easy. With only about 40 markets opening access to new spectrum bands for 5G by the end of February 2021, operators must explore new ways to find the required bandwidth.

One such way is to trade-off existing spectrum bands from older generations of technology to newer generations: shutting down an older generation network, a network sunset, becomes a way to support a newer generation.

GSMA Intelligence data shows from now to 2025, more than 55 2G and 3G networks will be closed, allowing operators to plan for 4G and 5G.

With network sunsets becoming one of the key ways to support 5G launches and 4G expansions, it is important to understand the benefits operator derive

Mobile broadband coverage. The key frequency bands to provide 2G and 3G services (900MHz, 1800MHz and 2100MHz) are the low- and mid-bands, thereby offering greater coverage and capacity. This makes them ideal candidates for network sunsets as operators can use these bands, combined with mid- and high-bands, for coverage and expansion of their 4G and 5G networks. Some operators, with large 2G/3G footprints, for example, are already refarming these bands to deploy and expand higher technologies
Expenditure efficiencies. Ageing 2G/3G networks eat up a significant portion of an operator’s capex and opex, with investments and expenses not justified by the ARPU. Hence operators around the world have preferred to focus their investments and resources on increasing their 4G coverage on fully digital technologies like VoLTE by sunsetting 2G/3G and using those frequency bands.
Roadmap for vendors and manufacturers. With an increasing number of operators using sunsets to support new launches and the expansion of existing networks, it is imperative for operators to have a well-defined strategy and roadmap around this. Timely announcements by operators also acts as a guide for vendors and manufacturers, driving their strategies, R&D and investment decisions, as well as manufacturing decisions based on regional advancements.

This makes a strong case for networks sunsets across all the markets. But does this mean the trends are consistent across the globe? Not at all. Different regions depict different trends (see chart, below, click to enlarge).

[1]

Europe: 3G eclipse
Europe is experiencing more sunsets in 3G than 2G. A total of 19 operators in 14 nations plan to switch off 3G by 2025, whereas only eight operators in eight countries are planning a 2G switch off by same time. The sunset saga in the region started with Net1 being the first operator to completely shut down its 3G networks in Denmark and Sweden in 2015, followed by VodafoneZiggo in 2020, then Telenor Norway and Swisscom in 2021. Though 3G is newer, it is still retiring in the region because 2G has dished out a notable edge, especially for M2M and IoT services in countries including Germany, Finland, and Belgium.

Americas: Goodbye 2G
The Americas is on a different path from Europe, focussed towards turning 2G off: there has not been a single 3G closure. The region, an early adopter of new technology generations, has seen 13 operators across five countries launching their 5G networks, making it vital to have the required spectrum to foster the services. Here we expect 15 operators in seven countries to switch off their 2G network by the end of 2025. Putting the best foot forward, the operators are re-using their existing spectrum from 2G to fill in the demand for 4G and 5G networks

Asia: 2G departure
Service providers in the region are vying to retain their 3G networks and shutting down 2G instead to employ the infrastructure for 4G, which has high adoption. By the end of 2025, we expect 29 operators to shut down 2G and 16 operators to close 3G with Taiwan currently the only market which has witnessed both 2G and 3G sunsets, in 2017 and 2018 respectively.

Africa: Preserving 2G and 3G
In this region, 2G markets outnumber 3G twofold, basic feature phone still comprise 42 per cent of all devices and end-users are incentivised to remain on them given lower costs. In turn, this drives lower digital uptake, explaining why a negligible number of sunsets have been announced. Of course, as demand rises, we would expect operators would have to plan for them in future.

Oceania: Precedence for 2G
The way to 2G in Oceania was paved when Australia and New Zealand launched the service in 1993, while 3G knocked on the door in 2004. The region has been quick enough to migrate to newer generation networks as 5G services are available in four countries, whereas others are testing the network for commercial deployments. Today, 2G networks account for only 5 per cent of total connections, having very small amount of traffic across the region. As a result of this trend, Australia, the biggest market in the region, completed its 2G network sunset in 2018. Though operators have not announced their network sunset plans, we expect the region to see more 2G than 3G closures, based on representative shares.

While we expect network sunsets to be an important strategy for operators, one approach is not appropriate for all everyone. Europe, for example, is the only region focused on switching 3G off, while others have a preference to shut 2G. The uptake and usage of different generations of technologies in respective countries and regions, in turn, defines sunset plans for operators. For vendors and manufacturers, timely announcements of these sunsets are integral, as they act as a guide to help them make manufacturing decisions and roaming infrastructure deployment.

And, as 5G gains mass adoption, all eyes will be on how operators shut down the older generations and switch to newer generation of technology.

– Akanksha Hira and Saksham Agarwal – research analysts, GSMA Intelligence

The editorial views expressed in this article are solely those of the author and will not necessarily reflect the views of the GSMA, its Members or Associate Members.

[1] https://www.mobileworldlive.com/wp-content/uploads/2021/03/GSMAi_network_sunsets.jpg

Satellite broadband has been around for a while, though mainly confined to enterprise or military uses. Consumers have largely used it in settings such as extreme adventure sports where connectivity can help save the mission and lives. However, a new cohort of satellite players is betting they can expand the consumer use case to home broadband as well by focusing their efforts on LEO satellite broadband.

Starlink, owned by SpaceX, is building a vast constellation of LEO satellites for the direct-to-consumer (D2C) market.
Amazon through its LEO satellite venture Project Kuiper is poised to join SpaceX in this market.
OneWeb is taking a different approach, deploying its LEO constellation to address market opportunities via the B2B market in partnership with MNOs, which could use the satellites to offer broadband to their subscribers.

Given this building momentum, now is a good time to analyse the prospects for LEO satellite home broadband on both its viability and competitiveness. For this analysis we use Starlink as a guide given its D2C satellite broadband rollout is at the most advanced stage, which also means relatively good information availability.

Viability has improved, but it is still early days
Here we address the question of viability by looking at the three key component areas: service availability; the addressable market; and service performance.

Service availability: All LEO satellite broadband providers whether they are focussed on D2C or B2B have a stated aim of reaching global coverage. For both, the main reason for pursuing this target is the maximisation of their revenue growth opportunity through having access to the full prospective customer base. Starlink, which currently has the greatest number of LEO satellites in orbit will have covered 100 per cent of US and Europe by the end of 2021, followed by the rest of the world in 2022. OneWeb will have achieved its global coverage by 2022.

Addressable market: Fixed broadband penetration in many developed markets around the world has reached 80 per cent to 90 per cent of households. However, that still leaves 10 per cent to 20 per cent without access. The European Commission reports 6 million households across the whole of the European Union remain unserved by any broadband technology. In the US, the figure stood at almost 15 million in 2018. For many of those that do have access, speeds are highly variable, undermining the range of applications that can be used. Both of these problems are concentrated in rural and remote regions and are the impetus for a range of national broadband plans, particularly in Europe. As satellite can reach these areas, it is a viable option for plugging their connectivity gap.

Performance: Video streaming is one of the most popular uses of home broadband. A good quality streaming experience without buffering or other technical glitches is a reliable indicator of a broadband connection which scores well on both speed and latency. Based on the feedback Starlink beta customers have been providing, this service handles Netflix well (of course once out of beta, the actual performance will need to be monitored). With most other household uses of broadband being less demanding than video streaming, it stands to reason that LEO satellite broadband is at least technically ready to make a play for the home broadband market. Starlink, advertises its broadband service as offering speeds between 50Mb/s and 150Mb/s and latency between 20 milliseconds and 40 milliseconds. OneWeb, has demonstrated under lab conditions, and also with BMW, full HD (1080p) streaming video at latency of less than 40 milliseconds with speeds of more than 400Mb/s. Covid-19 (coronavirus) has suddenly brought broadband upload speeds into focus as well due to home working and schooling, and here Starlink received mixed reviews for activities such as video calls. Apart from limited upload performance, some other risk factors to satellite broadband service quality are dishes needing line of sight to the sky, capacity overload at peak hours, and interference due to rain and adverse weather. A combination of constellation build up, more ground stations being installed and improvements in the networking software should see the overall broadband performance including uploads improve further, but to what extent remains to be seen.

Competitiveness
Satellite broadband is pricier but could be the only option for many rural households.

Starlink and Eutelsat’s broadband prices could be a pointer to satellite broadband services being relatively expensive compared with fixed wireless access and wireline broadband options including DSL, cable and FTTP/B (see chart, below, click to enlarge).

[1]

The caveat with this comparison is it applies to mainly urban areas where FWA and wired broadband services are available. For many rural households these services are not available, or they might have a poor-quality DSL connection and for them the quality of connection that LEO satellite broadband can now provide is nothing short of a game changer.

One Starlink beta user based in a rural area of the US described moving to Starlink from DSL as “a spiritual experience”. To improve affordability, satellite broadband providers are likely to introduce cheaper broadband packs in the future offering lower speeds or capped data usage or a combination of the two. In addition, most governments with universal service obligations in place are interested in how LEO satellite broadband could help close the digital divide, which could open the purse strings of various funds and financing packages also helping improve the affordability of LEO satellite broadband. Considering the specific issue of relatively high prices for satellite broadband consumer premise equipment (CPE), these are likely to reduce over time given the importance placed on achieving this result by the satellite broadband providers (Starlink currently charges £439 for its CPE in the UK compared with typical 5G FWA charges of zero to £100 with a contract, though Vodafone UK charges £325 out-of-contract).

Starlink is building a new factory in US to manufacture its CPE which should help bring its unit costs down and lower consumer prices. A good parallel is how Tesla’s plants helped lower the cost of Tesla cars to mass-market levels in a very short period of time.

Economics are uncertain
Given the variety of system architectures and the lack of publicly available information on the costs of LEO satellite broadband from providers, reaching any definite conclusions on the viability of this business is difficult. Building a basic cost-benefit model using sanguine figures for the various parameters (such as satellite build and launch costs, annual opex, CPE costs, satellite capacity and service life) it is apparent cost reduction will be crucial for the economic viability of satellite broadband.

Build and launch are the major costs here. Starlink is able to take advantage of SpaceX’s Falcon rocket to reduce costs. The company’s planned Starship, which could carry multiple the number of satellites that Falcon currently lifts, should help bring launch costs down much further.

OneWeb is now producing satellites which cost a fraction of previous technology. Starlink has also indicated it has been able to bring LEO build costs down from the current market average, but to what extent remains unclear. Technology developments such as very high throughput satellites, improving spectrum use, and network optimisation techniques, improving predictive analysis and advances in active antennas and processing will help bolster the benefits side of the satellite broadband business model as well further improving its viability.

The sweet spot
Given the current prices for satellite broadband, it looks likely consumer uptake will probably be highest amongst rural households in developed countries. As the economics of satellite broadband improve, it will lead to lower prices making this service more accessible to consumers in developing countries, especially if the technology is championed by the International Monetary Fund or the World Bank.

Considering the broadband technology mix, countries where DSL is a high percentage of total broadband connections will be relatively more attractive markets for satellite broadband. Globally, governments are working on bridging the rural-urban connectivity divide through various measures such as financial subsidies and setting speed and coverage targets. The net impact of these policies on satellite broadband’s business prospects remains to be seen.

– Anshu Goel – senior analyst, Fixed Broadband, Video and Convergence, GSMA Intelligence

The editorial views expressed in this article are solely those of the author and will not necessarily reflect the views of the GSMA, its Members or Associate Members.

 

[1] https://www.mobileworldlive.com/wp-content/uploads/2021/03/GSMAi_UK_home_bb_price_comparison.jpg