As 5G networks evolve, we’re told, coverage will branch out from densely populated places to cover wide areas, while use cases such as industrial IoT, C-V2X (cellular vehicle-to-everything), smart agriculture and mobile XR (eXtended Reality) will mature thanks to faster speeds, lower latency and greater reliability. At the same time, the network infrastructure itself will become more sustainable. Yet the current experience of 5G for many people is very different. I’m writing this article from my home office in a rural but hardly remote UK location (I’m 40 miles from central London, 12 miles from Milton Keynes, 15 miles from Bedford, 14 miles from Luton). Entering my postcode into the coverage checkers for the UK’s four mobile networks reveals a 5G desert: O2 – ‘Sorry, we do not have 5G coverage at this location’; Vodafone - ‘No coverage yet’; EE - ‘5G is not available at this location yet’; Three - ‘Unavailable. Unfortunately, our 5G network isn’t currently available in this area’. Yet the UK is regarded as a pioneer in 5G deployment in Western Europe. Clearly there’s a long way to go before 5G fulfils its promise in many places, even though smartphones supporting the technology are now widely available. The journey to 5G fulfilment will hopefully involve the disposal of conspiracy theories and health scares, but also requires more legitimate concerns, such as airport safety (in some regions), to be addressed. Let’s look further at where 5G is going, and how it will get there. For more detail, check out the other articles in ZDNet’s special report.
5G network rollouts
As of the end of December 2021, according to the GSA (Global Mobile Suppliers Association), 487 operators in 145 countries or territories had invested in 5G mobile or 5G FWA/home broadband networks. Of these, 200 operators in 78 countries/territories had announced 3GPP-compatible 5G service launches (mobile or FWA), with 187 offering commercial 5G mobile services and 83 providing 5G FWA services. A recent development is the advent of standalone (SA) 5G networks, which combine 5G New Radio (NR) in the Radio Access Network (RAN) and 5G core infrastructure at the back end – in contrast to non-standalone (NSA) 5G networks that piggy-back on existing 4G LTE deployments. Backed by cloud-native microservices-based infrastructure, 5G SA networks can offer low latency, massive device support, and network slicing to accommodate use cases and customers with different service level requirements. GSA has identified 99 operators in 50 countries that are “investing in 5G standalone for public networks in the form of trials, plans, paying for licences, deploying, or operating networks”. At least 20 operators in 16 countries have launched public 5G SA networks, according to GSA, while a further five have deployed 5G SA technology but have yet to launch services, or have only soft-launched them. Add the 25 operators that are in the process of deploying or piloting public 5G SA and 27 that are planning to do so, and it’s clear that “the launches catalogued by the GSA so far will be the first of many”. Private 5G networks designed for specific needs are an attractive option for many organisations and enterprises – especially those that prefer, or need, to keep their data securely on-premises and/or require reliable low-latency connectivity for use cases like smart manufacturing. According to GSA, 166 organisations are using 5G technology for private mobile networks or pilots, of which 32 are already working with 5G SA. Here’s a summary of these numbers: Data: Global Mobile Suppliers Association (GSA) Although there’s plenty of noise around private 5G networks, Omdia analyst Pablo Tomasi has cautioned that 2022 may not be the year that sees lift-off: “5G will play an increasingly important role, but despite what everyone wants, this won’t be the year of 5G. What we still see in the market is that private LTE is a good enough technology to address the majority of the use cases. There are still a lot of uncertainties regarding what private 5G can do that LTE cannot deliver,” he said in a recent interview.
5G spectrum
The main use cases envisaged for 5G – enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), massive machine-type communications (mMTC) and fixed wireless access (FWA) – require different combinations and amounts of radio frequency spectrum. There are three main 5G bands: low (<1GHz), mid (1-6GHz), and high (24-100GHz, a.k.a. mmWave). Low-band spectrum offers wide coverage and good indoor penetration, but speeds and latencies are little better than those delivered by 4G LTE networks. Mid-band spectrum offers a good combination of speed (~100Mbps) and coverage, while high-band mmWave spectrum can deliver very fast speeds (≥1Gbps) over short distances. A new development in 5G networks is the widespread use of Time Division Duplex (TDD), which works well for mid- and high-band spectrum, as opposed to Frequency Division Duplex (FDD), which works well for low-band frequencies. Under TDD, base stations and devices transmit and receive traffic using the same frequency channel at different times, while FDD is a duplex mode that uses different channels to transmit and receive. A potential issue with TDD is interference within and between networks, the solution to which is typically synchronisation of networks operating within the same frequency range and geographical area. However, as the GSMA – an industry association representing the interests of mobile operators worldwide – has noted, there are implications for the use cases that can be supported under these conditions. “Regulators need to consider this when deciding how to make spectrum available in 5G TDD bands and technical conditions for use,” the GSMA said in a March 2021 report. Up to now, mid-and low-band spectrum – often called ‘sub-6GHz’ – has been most widely used in 5G networks, particularly C-Band spectrum which sits between the 2.4GHz and 5GHz Wi-Fi bands. However, mmWave, which can provide high data bandwidth in geographically restricted but densely populated places like city districts, shopping centres and sports stadiums, is beginning to roll out around the world. According to GSA, 140 operators in 25 countries/territories hold public licences enabling them to build 5G networks using high-band spectrum between 24.25GHz and 48.2GHz. Of these, 28 operators in 16 countries/territories have already deployed 5G networks using mmWave spectrum. Cell coverage and peak data rates in 5G networks can be improved by a technique called Carrier Aggregation, which was first introduced in 4G LTE by the 3GPP. Here, different portions of spectrum are divided into primary and secondary component carriers (PCCs and SCCs), with uplink, control and user data being sent on the PCCs. Network cells are similarly divided into PCells and SCells. With carrier aggregation, the uplink can operate on a low band and the downlink on a mid or high band, resulting in better overall performance. When it comes to making spectrum available for 5G, regulators have mostly adopted the conventional practice of auctioning off exclusive nationwide licences, although some have also set aside portions for organisations to build private 5G networks. In March 2021, the GSMA outlined its positions on 5G spectrum allocation in a ‘roadmap for regulators’:
The 5G device ecosystem
As 5G networks roll out around the world, so the number of devices that can make use of them is increasing. According to the latest GSA data (January 2022), a total of 1,257 5G devices in 22 form factors from 180 vendors have been announced. Of these, 857 (68.2%) are ‘understood’ to be commercially available (up 21.7% quarter-on-quarter). 2. 5G needs spectrum across low, mid and high spectrum bands to deliver widespread coverage and support a wide range of use cases. 3. Governments and regulators should support new harmonised bands on the international stage to help 5G services grow over the longer term (e.g. UHF, 3.3-4.2GHz, 4.8GHz and 6GHz). This includes engaging in the WRC-23 process to ensure sufficient mid- and low-band spectrum is available. 4. Exclusively licensed spectrum over wide geographic areas is vital to the success of 5G. 5. Spectrum sharing and unlicensed spectrum can play a complementary role. 6. Setting spectrum aside for local or vertical usage in priority bands (i.e. 3.5/26/28GHz) could jeopardise the success of public 5G services and may waste spectrum. Sharing approaches like leasing are typically better options in these situations. 7. Governments and regulators should avoid inflating 5G spectrum prices as this is linked to slower broadband speeds and worse coverage. Key concerns are excessive reserve prices, annual fees, limited spectrum supply (e.g. through set-asides) and poor auction design. 8. Regulators should carefully consider 5G backhaul needs including making additional bands available and supporting wider bandwidths in existing bands. Measures should also be taken to ensure licences are affordable and designed effectively. 9. Regulators should carefully consider the right 5G spectrum licence terms, conditions and awards approach and consult industry to maximise the benefits of 5G for all. 10. Governments need to adopt national spectrum policy measures to encourage long-term heavy investment in 5G networks (e.g. long-term licences, renewal process, spectrum roadmap etc.) Here’s how the GSA’s 5G device numbers break down:
- drones, head-mounted displays, robots, TVs, cameras, femtocells/small cells, repeaters, vehicle OBUs, a snap-on dongle/adapter, a switch, a vending machine and an encoder Data: GSA As you might expect, the biggest category of 5G devices is phones (48.8%), followed by FWA CPE (Customer Premises Equipment) devices (16.7%) and 5G modules (13.8%): “Based on vendors’ previous statements and recent rates of device release, we might expect to see the number of commercial devices getting close to the 1000 mark by the end of Q1 2022,” the GSA report said. The result of increasing network and device availability is, of course, more connections. In May 2021, analyst firm CCS Insight forecast that connections to 5G networks would reach 670 million worldwide by the end of the year and were on track to reach 3.6 billion in 2025. By the middle of the decade, CCS Insight expects over 75% of mobile phones in North America, Western Europe and Asia-Pacific to have moved to 5G networks, while South Korea and China stand out for their current speed of 5G deployment (both countries will have passed the 20% mark during 2021). “Risks to the exact mode and speed of 5G adoption remain, ranging from the unpredictable nature of the global pandemic and the high uncertainties facing the world economy, to the short supply of components for smartphones and other smart devices,” the analyst firm said. “However, the mobile industry has firmly stepped on the path of upgrade to 5G and short-term challenges will do little to hinder its long-term progress.”
Evolution of 5G standards
The 3GPP (3rd Generation Partnership Project) is the consortium of telecoms organisations that develops and maintains standards for GSM (2G), UMTS (3G), LTE (4G) and 5G – and, in due course, 6G. The 3GPP’s 5G standards are an ongoing multi-release process. The first stage, Release 15, concentrated on enhanced mobile broadband(eMBB) and basic ultra-reliable low-latency communication (URLLC), while the focus of Release 16, which was completed after a three-month pandemic-related delay in July 2020, was on expanding the range of use cases, including enhanced URLLC, Industrial IoT and non-public (private) networks. Rel 16 also included support for unlicensed spectrum (NR-U), 5G-based indoor positioning, NR-based sidelink for cellular vehicle-to-everything (C-V2X) communication, and integrated access and backhaul (IAB), which allows mmWave base stations to act as both wireless access and backhaul. Enhancements were also made in Rel 16 to multi-user MIMO, service-based architecture (SBA) and network slicing. Release 17, which the 3GPP describes as “perhaps the most versatile release in 3GPP history in terms of content”, was initially expected to be completed by 2021, but the pandemic has pushed this back to March 2022 (freeze) and June (protocol coding freeze). Its primary aim is to enhance existing use cases such as mobile broadband, industrial automation, C-V2X and private networks, while introducing new capabilities covering areas like edge computing, satellite connectivity (non-terrestrial networks, or NTNs), drones and support for mmWave in the 52.6-71GHz band. Release 18, with a functional freeze set for March 2023, is considered significant enough to get its own ‘5G Advanced’ branding. According to 3GPP, over 500 presentations were reviewed to identify topics for inclusion in Rel 18/5G Advanced, which it has grouped under ’evolved mobile broadband’ (eMBB), ’non-eMMB evolution’ and ‘cross functionalities for both eMBB and non-eMBB driven evolution’. Notable eMBB developments in Rel 18 focus on beamforming/MIMO, network efficiency in LTE-5G transitions and network power savings; non-eMBB advances focus on reducing device cost and power consumption (RedCap), eXtended reality (XR) services, and improved drone control and rogue drone detection; functionality in Rel 18 that targets both eMBB and non-eMBB use cases includes the use of AI/ML to improve the performance of both the physical and radio-access layers, and the feasibility of full duplex on TDD frequency bands.
The road to 6G
As standalone networks roll out, new spectrum becomes available and the device ecosystem grows, 5G deployments are moving beyond enhanced mobile broadband and fixed wireless access, and are beginning to enable a wider range of use cases. But there’s still a long way to go. In 2022, according to the GSA, developments will include “5G carrier aggregation in SA networks, URLLC capabilities to support machine-to-machine communications in 5G SA systems, increasing mmWave support, network slicing in 5G networks and the introduction of voice-over new radio (VoNR) in 5G SA networks.” 5G Advanced (a.k.a. Release 18) will add further enhancements and capabilities in 2023. Beyond that, towards the end of the decade, is 6G. What will that look like? At a recent (December 2021) workshop involving key stakeholders from government, industry and academia, participants took the view that “6G will be enabled and defined by software, more so than hardware, with intelligence built into the network. However, the hardware network will be highly diverse, with processing capability embedded throughout the network. As such, 6G will require particular emphasis on standards which enable flexibility and interoperability.” No doubt 6G, when it arrives, will enable mind-boggling use cases, surpassing the autonomous vehicles and remote/robotic surgery examples so often cited for 5G. In the meantime, I’ll wait patiently for any sort of 5G connection here in my home office, just 40 miles from London.
