The old cliché that ‘life is a journey not a race’ should certainly apply to 5G. The technologies that are currently being deployed are not ‘full 5G’ but the first step in a platform evolution which will take place over the coming decade, encompassing continual change in the RAN, core, service architecture, spectrum and use cases. Meanwhile, vendors, operators’ marketing departments, and (unfortunately) governments are turning it into a race – who can be first to deploy even half-baked 5G, not because there is an urgent need for a new network, but for bragging rights over rival MNOs or countries.
The artificial race can be seen as a relatively harmless bit of self-promotion, except when it damages the operator’s choices about its route and timings for the overall 5G journey. Some MNOs are complaining – most privately, some publicly – that they have felt railroaded into deploying before they see mature platforms or firmed-up business cases. The result is a sub-par roll-out which has little commercial impact, and does little more for the user experience than a bit of 4G enhancement or a new spectrum carrier would have done. And that, in turn, may delay the point at which the operator has the funds to build out ‘true 5G’, or may impose an awkward, expensive and avoidable upgrade process.
The most glaring example, of course, is the looming need for operators to move from their 5G New Radio Non-Standalone (NR NSA) towards Standalone 5G with a full cloud-native 5G core. This will be a far more challenging migration than adding 5G RAN to the LTE core, and should have a far more transformational effect on the business, if implemented well. But many operators would, ideally, have waited a couple of years for these solutions to mature and then rolled out the new core and RAN together.
Most will make do with NR NSA for several years before taking the plunge with the 5G core, but a few have more aggressive timescales to migrate, such as China Mobile and Three UK, while from 2020 onwards, there will more MNOs which did not join the race to deploy in the first year of 5G, but will go to Standalone from day one, once they do make the move.
The 5G ‘race’ can be even more damaging to rational deployment strategies when it takes place at a national or geopolitical level. The rivalry between the USA and China covers many areas of trade, national security and politics, but 5G has come to be a proxy for the bigger war. And that race to be the 5G ‘world leader’ is certainly in danger of putting the longer 5G journey at risk, by triggering sanctions and patents battles that will split the globe into two conflicting zones in terms of standards development and technology production. That will deny each zone access to one another’s innovations and kill the dream of a common global standards base with the benefits of interoperability, economies of scale and a broad ecosystem (see lead article below).
Whatever the individual decisions MNOs and governments make about 5G, there will be many other enhancements to plan to push the network’s capabilities forward. Some of these are already finding their way into commercial equipment – most vendors are offering 5G cores, though with varying levels of real cloud-native support; Ericsson has just announced its equipment to support the new set of 3GPP standards, Release 16.
Further ahead, the 3GPP has started work on Release 17, which will include some important improvements to latency, security, network slicing and other features. And there are more and more inputs to the broader 5G platform from other sectors, including cable and satellite (see Key Issues).
And a few operators are already starting to consider ‘6G’, which has been the subject of academic projects for a few years, mainly with regard to very high frequency spectrum up to terahertz bands, but is now making its first steps into the labs of MNOs such as SK Telecom of Korea.
Huawei ban threatens global 5G standards, and risks costly patents wars
Many companies, even in the USA, are lobbying hard to get the stepped-up US restrictions on Huawei reversed or eased, raising concerns that excluding the Chinese giant from the global mobile ecosystem will erect high trade barriers which will damage western firms too; that this will have a knock-on effect on the cost and effectiveness of 5G; that security risks will get worse, not better, as cooperation breaks down; and that the world will be split into two technology islands, ending hopes of common standards for 5G and future platforms.
Many of the concerns of the companies lobbying for a new approach to Huawei are short term:
- In countries where bans on Huawei selling 5G kit are being considered, operators fear an even less competitive supply chain as they start to roll out 5G, rising higher costs and lower innovations.
- Companies, such as chip providers, which count Huawei as a major client fear the loss of revenue, and that the bans on procurement of US components might be extended to their other Chinese customers. This does not just concern US firms, but others which might be forced to choose between continuing to supply Huawei and still being welcome as US trade partners.
- Google and others claim that the ban on Huawei using Google services in Android (which the search giant did voluntarily impose after Huawei was placed on the US ‘entity list’) will accelerate China’s existing initiative to create its own operating system and mobile software platforms. That could not only split the smartphone software ecosystem into two camps, but increase the risk of security breaches if the Chinese operating system were used worldwide.
- The risk of the world splitting into two 5G camps is even greater in the chip industry, where bans on procurement of US components are intensifying China’s national bid to become self-sufficient in semiconductors and core technologies. A temporary block on ZTE almost brought the company to its knees last year; Huawei is less reliant on western chips because of its HiSilicon subsidiary and its cultivation of some other local suppliers, but it still needs access to US innovations to ensure it can develop cutting edge equipment and devices. China has made huge investments, and great progress, in its silicon industry, but it will take years to achieve equivalence with the USA at the leading edge.
These concerns are mainly about near term impact on 5G cost and viability, in the operators’ case, and on the next few quarters of financial results, for vendors like Broadcom (which has already issued a warning). But they also cast a long shadow over the longer term future and economics of the mobile industry. Several questions are critical to the way the future standards-based platforms will look:
- Will the world, indeed, split into two technology camps, one based around the USA and one around China?
- What impact will that have on the development of future standards, including forthcoming releases of 5G and eventually 6G, plus other relevant standards processes in the IEEE, IET and other global initiatives?
- Will a geopolitical split in developing new technologies result in higher patent costs for companies working across the divide, or simply a reduced access to innovation on a global basis?
- And how far will a split drive costs up in the near term as Chinese companies – which have amassed a huge patent mountain in the mobile world, including in standards-essential patents (SEP) – may chase payments more aggressively. China is determined to reduce its dependence on western IPR, and if the industry thought it was hard to define and impose Frand (fair reasonable and non-discriminatory) pricing practices on Qualcomm or Samsung SEP, try doing the same for angry and defensive Chinese firms.
The patents issue will be at the heart of the matter if the US continues to place Huawei on its entity list, which means that US firms cannot trade with it, except with special permissions. Chinese firms, especially ZTE and Huawei, are estimated to have doubled their share of SEP in 5G compared to 4G and to have applied for about a third of major 5G patents so far. According to IPlytics, which analyzed applications (not awarded patents, so the exact percentages may change), Chinese firms have applied for 34% of major 5G patents, compared to 25% for South Korea, and 14% each for the USA and Finland. Nokia and its ecosystem seem to be outgunning Ericsson, since Sweden came in at 8% and Japan at 5%, with Taiwan, Canada, the UK and Italy completing the top 10, but with less than 1% each.
The analysis related to 74,500 patents (so many will not be SEP of course), and it found that China’s share of 5G patents so far is about 50% higher than its 4G share, where it matched South Korea, with about 22% each. The USA has a lower share than in 4G, which highlights one reason both for its defensiveness towards China over 5G trade and technology, and the fact that it is in a relatively weak position once patents get dragged into trade wars, especially as Huawei also has IPR in existing 3G and 4G systems – which it may choose to monetize aggressively as a tactic in its battle against the US ban.
Indeed, Huawei signalled how that could pan out for western operators and vendors when it told Verizon that the telco owes Huawei “more than $1bn in patent licensing fees”. According to reports, the claim relates to equipment made by other vendors, but incorporating Huawei IPR. Reports by the Wall Street Journal and Reuters said “the patents cover network equipment for more than 20 of the company’s vendors including major US tech firms” and involve about 230 patents in all.
According to another patent analytics provider, Relecura, “Huawei has been granted more than 69,000 patents globally related to everything from data transmission to network traffic management”, and another 49,379 patent applications are pending. Of those granted, over 57% are in China, and nearly 18% in the USA.
This shows how Chinese technology is ubiquitous, even in the USA where Chinese firms have been barred from supplying equipment for critical infrastructure – which includes national mobile networks – for years. In the past, Huawei will have engaged in cross-licensing deals with other suppliers in the usual way, but it will now see limited opportunity to secure patent fees from US vendors or telcos, or from any firm that supplies them, because of the bar on trade.
Reuters said that representatives of Huawei and Verizon had met “to discuss some of the patents at issue and whether Verizon is using equipment from other companies that could infringe on Huawei patents”. The telco’s main mobile network suppliers are Ericsson and Nokia, and it is also using some Samsung equipment for 5G.
But despite the talks, the mood was not one for compromise at the end of this week. Senator Marco Rubio proposed a ban on Huawei initiating any patent litigation or IPR claims against US firms should the Commerce Department find, in an imminent ruling, that the Chinese firm poses a national security risk in telecoms networks. Rubio, with Democrat Senator Mark Warner, had previously called on the US Director of National Intelligence, Dan Coats, to “issue a comprehensive and unclassified report on China’s participation in the international standard-setting bodies for … 5G. This report would allow companies in the US to fully assess any existing threats to fair competition and push back against them.”
President Trump’s main economic advisor, Larry Kudlow, director of the National Economic Council, said recently: “I don’t want certain countries to run away with certain standards.”
Meanwhile Huawei, which has not been at all litigious in the past, indicated that it would look beyond Verizon. CEO Ren Zhengfei said: “Over the past years, we were not aggressive seeking IPR royalties to companies that use our IPR — that’s because we were busy pursuing our business growth. Once we have more time off, we may try to get some money from those companies who use our IPR.”
This new climate will not just affect the balance of patents licensing power in future, and the costs for vendors and operators. It could also destroy the global nature of mobile standards-setting if Chinese firms are excluded from international bodies, or choose to pursue their own path.
Last week, a group of 26 technical specifications and standards bodies wrote to Wilbur Ross, secretary of the US Department of Commerce, seeking clarification on whether Huawei is allowed to contribute to their work while on the entity list. The organizations included the Broadband Forum, Ethernet Alliance, MulteFire Alliance and Video Electronics Standards Association (VESA), which were not included on the list of standards bodies that the US government said could continue to work with Huawei until August 19 (see inset).
More broadly, they argued that excluding Huawei from participation in standards bodies would lead to “standards wars”, and called for an amendment to the current language used in the entity list judgement, which would enable “business as usual” for standards activities.
If the current situation persists, they warned, it will create “a serious risk that specifications developed by US-based consortia will fail to achieve the goal of adoption as formal or de facto international standards, and that future necessary standardization efforts will be led elsewhere. Further, if companies with large market presence are excluded from US-based standards efforts, a possible result is the launch of competing standards, leading to lengthy and destructive ‘standards wars’.”
Some industry bodies, such as the WiFi Alliance and SD Alliance, already restricted Huawei’s rights to participate; and conversely, there were notable no-shows from some US firms when the 3GPP had a plenary meeting in Shenzhen, China in April. Executives from three US operators – Sprint, AT&T and the public safety network FirstNet – failed to attend the meeting at which the ‘final drop’ of Release 15 standards was released (though Verizon and T-Mobile representatives were there). Sprint, AT&T and FirstNet have sent representatives to every plenary meeting of the 3GPP’s RAN Technical Specification Group (TSG) for more than two years.
Mike Thelander, an analyst with Signals Research Group, who attended the Shenzhen meeting, wrote in a client note: “Unfortunately, the question seemed to have been, ‘given the political climate, what are the optics associated with my company attending a Huawei-hosted 3GPP meeting in Shenzhen, China?’ When a mobile operator doesn’t attend a standardization meeting then it isn’t present to further its agenda, which could be of strategic importance to the operator’s network evolution.”
This all points to the risk of China going its own way on core technologies – as it did when it was the sole adopter of TD-SCDMA as the basis of 3G, a move which disadvantaged China Mobile, when the MNO was forced to use the technology, and created a damaging split in R&D funds and vendor markets.
Now there is the risk of a bigger split, closer to that seen in the GSM/CDMA and the 3G eras, which would also put at risk the scale of the global ecosystem of equipment and devices, achieved in LTE and envisaged for 5G. That would be another blow to the hoped-for economic efficiencies of 5G networks, to add to the potential additional cost that will be incurred by operators in countries where Huawei and ZTE are barred from 5G deals, reducing the MNO’s choice of suppliers, for macrocells at least, to three.
All this comes at a crucial time. The 3GPP is engaged with Release 16 and recently began work on Release 17 of the 5G standards. But it is doing so under the shadow of uncertainty about how far Huawei can take part. Thelander wrote, after attending a recent 3GPP RAN meeting in Newport Beach, California: “The unfolding saga this time pertained to how Huawei delegates and delegates from US companies can interact with each other. As we understand it, rules were put in place by the companies themselves, likely to protect themselves from outside scrutiny and/or to comply with the US government’s actions being taken against Huawei.”
He also commented on Huawei’s approach, adding: “We assume they don’t want to further jeopardize their ability to resolve matters. We also wouldn’t rule out a bit of spite on their part. Separately, we heard from a western operator that Huawei doesn’t want any US citizens working for the operator to have any interaction with Huawei. This restriction also includes non-US citizens from having a conference call with Huawei while located in the US. We know multiple US citizens who currently work for non-US operators so if this Huawei policy is universally applied then the impact isn’t trivial.”
A spokesman for the 3GPP insisted, in a statement to LightReading: “We have a business as usual approach in 3GPP. We hope that cooperation on standards will continue as we are used to.”
But Adrian Scrase, head of the 3GPP’s permanent support team at ETSI, told the UK Financial Times: “If the current situation prevails, this could have a dramatic impact on future standardization”; while Wall Street analyst Jefferies told investors: “The risk that 3GPP is going to break down is pretty high. If Chinese companies are prohibited from licensing technologies that they have already seen – and which in many cases they have heavily influenced – they could just use it.”
Despite the lack of communication between certain attendees, the meeting in Newport Beach produced the first proposals for what might be included in Release 17 (Release 16 is expected to be finalized early next year, and Release 17 about a year later). Particular focuses for Rel17 include Industrial IoT, spectrum above 52.6 GHz, V2X scenarios, lower power WAN applications, and ways to make 5G more effective for virtual reality.
Some of these are clearly extension of work done in current releases for new 5G use cases, while others are new. An important addition is work to make the 5G standard agnostic to whether spectrum is licensed or unlicensed. A key feature of Rel16 will be standards for 5G in unlicensed bands, which will enhance the ability of non-MNO deployers, such as neutral hosts and private network operators, to harness 5G. Rel17 aims to go a step forward and allow any kind of spectrum to be used by 5G seamlessly. These proposals, where accepted, will be turned into workgroup items at the December RAN meeting.
And already, the industry is starting to think about 6G. This has, so far, largely applied to academic research projects focused on very high frequency spectrum, above 100 GHz and even into the terahertz range. But now operators are starting to include ‘6G’ (in the sense of the next generation after 5G, not in any 3GPP definition) in their research programmes, which will be slightly less bluesky than most of the university initiatives in the USA, China, Japan, Finland and elsewhere.
SK Telecom of South Korea recently signed memorandums of understanding with Ericsson, Nokia and Samsung to conduct joint R&D in advanced 5G technologies, and how those might evolve into 6G. Technologies include URLLC (ultra-reliable low latency communications) and enhanced MIMO and 28 GHz approaches. The next stage will be to
draft technical requirements and new business models for 6G.
And in the current climate, the most striking aspect of that announcement for the industry was not SK’s thoughts on what 5G might constitute, but the fact that there are no Chinese firms on its MoU list.
The US pressure to ease Huawei ban ramps up:
Many players in the US mobile industry stand to lose business, or increase costs, should Huawei remain on the ‘entity list’ (see main text). These examples show the rising tide of opposition to the ban, and the possibility of it being extended to others. After putting Huawei on the list, the government relented slightly and allowed a grace period, until August 19 this year, for companies to continue to work with Huawei, particularly to conclude ongoing projects. But after that date, the gates will close completely unless there is a change of policy.
- Three major US chip suppliers, Qualcomm, Intel and Xilinx, are calling for the US government to lift the ban on them selling to Huawei. Reuters cited inside sources who said that executives from Intel and Xilinx attended a meeting with the Commerce Department in late May, while Qualcomm has also presented its concerns to the same agency.
The Commerce Department is due to make a final judgement on whether Huawei constitutes a risk to national security because it could place spyware in telecoms equipment at the behest of the Chinese government. This is the official reason for the bar on the Chinese firm, though it fiercely denies all allegations and no evidence has yet been presented. Of course, the actions against Huawei are also seen as part of the trade, intellectual property and general technology leadership wars which have escalated between the USA and China throughout the past year.
The chipmakers are particularly focusing on the components they sell for Huawei’s smartphones and wearables, since these are unlikely to present serious national security risks. And they point out that, in 2018, Huawei spent nearly $11bn on components from US firms.
- As Huawei dusts off its own mobile operating system project to counter the ban on its use of Google services, it is likely to produce a hybrid version of Android based on the open source code alone. Google is arguing to the US authorities that this OS will be far more vulnerable to hacking, even by Chinese authorities, than the standard version. Even though Huawei handsets will not be available in the USA if the ban persists, Chinese handsets will be able to communicate with US phones.
- The head of a White House agency, the Office of Management and Budget, has called for a two-year delay in implementing the ban on Huawei. Russel Vought, acting director of the agency, was specifically referring to government contractors and organizations receiving federal grants, which already have a longer grace period than vendors do, because of the complexity of unpicking a supplier from a major contract. Currently the ban, imposed under the National Defense Authorization Act (NDAA), comes into operation in 2021. Vought wants to extend that to 2023 to give organizations which rely on Huawei gear time to adjust, especially rural telcos (which have not been restricted in buying Huawei kit until now, unlike the national operators).
Jacob Wood of the OMB told the Wall Street Journal: “This is about ensuring that companies who do business with the US government or receive federal grants and loans have time to extricate themselves from doing business with Huawei and other Chinese tech companies listed in the NDAA.”
Will the stay of execution be extended?
Many hopes for some normalization of relations in the 5G industry rest on the possibility that the US government will extend its stay of execution for Huawei, and that repeated reprieves might turn into an effective reversal of the ban. This would be somewhat typical of the current administration’s pattern of coming down very hard – very publicly – on perceived enemies, and then backing away in reality. Examples of this range from ZTE last year to North Korea. Such behaviour would not alleviate uncertainties for the industry, or reduce China’s need to end its reliance on such an unpredictable technology partner. But it could enable vendors and operators, and standards organizations, to continue to work with Huawei.
Shortly after saying it would place Huawei on its entity list – which forbids US firms to trade with the blacklisted company without a special licence – the US government announced a 90-day reprieve, which lasts until August 19. It also said that this could be renewed, giving hope of an outcome which would be messy but at least not catastrophic to companies on both sides.
The temporary licence allows:
- Continued operation and maintenance of existing networks and equipment, including software patches, if these are part of contracts signed with Huawei before May 16 2019.
- Support for existing Huawei handsets which were available to the public before May 16.
- Disclosure to Huawei of information regarding security vulnerabilities in items owned, possessed or controlled by Huawei, if this is necessary to supporting the equipment and devices.
- Engagement, as necessary, for development of 5G standards by recognized standards bodies (IEEE, IETF, ISO, ITU, ETSI, 3GPP, TIA and GSMA).
Russia comes to Huawei’s aid with 5G deal and smartphone OS offer:
The current crisis is not just about China becoming a technology island, as it was, in many ways, in the 1990s. It is about building two zones of influence. The pressure the USA has put on its allies to ban Chinese vendors from 5G networks shows that the Trump government regards 5G technology as a geopolitical issue, one in which countries will stand together to support friendly nations.
The operators do not see it like that of course – European MNOs want freedom to choose the best and cheapest kit, and claim they can take responsibility for securing their networks effectively against spyware. And the bid to build an international alliance against Huawei and ZTE plays both ways – China also has powerful alliances, and may encourage them to choose Chinese technology, at a cost to US suppliers like Cisco, but also to Nokia and Ericsson.
A hint of the way this hi-tech Cold War could play out was seen when 5G issues were mentioned at a recent summit between the leaders of China and Russia, Xi Jinping and Vladimir Putin. At the same time, Huawei signed an agreement with Russian operator MTS to supply it with 5G networks, while Russia offered support in developing an alternative mobile operating system to Google Android.
The CEO of MTS, Alexei Kornya, said the deal would accelerate 5G roll-out in Russia, and would also “contribute to the further development of economic ties between Russia and China”.
And according to Russian news agency TASS, the country has offered Aurora, a Russian smartphone OS, and its ecosystem of search engine, social media and other software, to China as its Android alternative. Aurora is based on the Linux mobile OS created by Finnish start-up Sailfish. China is likely to want to develop its own platform in this highly strategic area, but the offer of Aurora/Sailfish indicates how major multinational resources could be pumped into ensuring that an alternative to US technology is not confined to China alone.
Ericsson upgrades prepare for wide 5G coverage and cloud-native core
The most important near term stage in the operator’s 5G journey is to migrate from an initial deployment of 5G NR Non-Standalone (NSA) to the Standalone (SA) version, which can harness the 5G core rather than relying on the LTE packet core. The 5G core is the heart of ‘true 5G’ and the only way to achieve many of the promises made for the new platform, from end-to-end slicing to ultra-reliable low latency communications (URLLC).
Ericsson has announced commercial availability of its 5G NR SA software to allow operators to deploy a pure 5G network on Ericsson equipment for the first time. The Swedish vendor’s RAN can be upgraded to SA with a software update, it said, and it expects to see the start of deployments later this year.
Upgrading the RAN to be able to work with a 5G core does not, of course, mean that those operators will deploy the new core immediately. Many MNOs are saying they will wait until fully cloud-native 5G core offerings are more mature, and the impact on their services and organizations are better understood, before taking the plunge into a powerful platform which will, nevertheless, represent a very different way of working.
Some operators have virtualized elements of their 4G cores, but not in a cloud-native way, while many have held back, so that 5G will be their first experience of implementing a critical network element, at scale, in software on commoditized hardware, rather than in specialized appliances. The major network and core vendors are all claiming full cloud-native implementations, in which virtual network functions (VNFs) are built specifically for the cloud (not repurposed from physical functions). The cloud-native approach disaggregates the various functions of the core into microservices that can be combined and re-combined, Lego-style, to suit different services and operating scenarios. But in most cases, only certain attributes of cloud-native are available, and many operators will want to see these platforms mature some more before making their choice.
Ericsson also released other enhancements to its 5G network offering, including inter-band carrier aggregation software, two new 5G radios, and new NFV infrastructure (NFVi) optimized for distributed cloud and edge workloads, which is multi-tenant and multi-purpose.
“We foresee in the future more complex and more advanced 5G networks than these initial deployments,” Hannes Ekstrom, head of Ericsson’s 5G RAN product line, told SDxCentral, with faster response times, better coverage and low latency.
He believes that, in addition to the 5G core, the other much-needed enhancement to the 5G platform will come from low-band spectrum. All eight of the commercial 5G networks Ericsson has deployed to date are in mid-band or millimeter wave spectrum, which are challenging when it comes to coverage.
“As you move to Standalone architecture and no longer rely on the LTE connectivity, we need to introduce low-band spectrum as well,” he said.
Ericsson’s Inter-band NR Carrier Aggregation software will extend the coverage and capacity of 5G NR on midrange and high bands when combined with 5G NR in low-band spectrum. Meanwhile, its new MIMO 5G radios, AIR 1636 and AIR 1623, are designed for midrange bands like 3.5 GHz, the workhorse of most early 5G roll-outs. AIR 1636 has been engineered to improve coverage with optimized performance on longer inter-site distances; while AIR 1623 has been designed for easy site build with low total cost of ownership.
Ericsson is also launching a partner certification programme for VNFs. This is open to all VNF vendors and grants a certification on the Ericsson NFVI platform using Ericsson Labs.