Reliance’s 5G claim: A reality check

Print edition : August 14, 2020

Mukesh Ambani, chairman of Reliance Industries, a file photograph. Photo: PRAKASH SINGH/AFP

5G apparatus, manufactured by Ericsson AB, sits on the roof of a commercial building in Barcelona, Spain. Photograph taken on July 16. Photo: Angel Garcia/Bloomberg

Reliance’s claim that it has developed fully “indigenous” 5G capability appears dubious given the way the technology has evolved across the world. Its pole position in the Indian telecom market raises fears that this may limit India’s capabilities in telecommunications in the future.

Company chieftains, like politicians, are known to make outlandish claims. And, like politicians, they often get away with them. When Mukesh Ambani, chairman of Reliance Industries Ltd. (RIL) India’s largest company, recently made the stunning announcement that its telecom venture, Jio, had developed “world class 5G services”, not a single media outlet stirred in surprise at the enormity of the announcement’s implication. Instead, a prominent business daily’s header led thus: “RIL builds indigenous 5G tech, says ready to take on the world”.

Addressing RIL’s annual general meeting on July 16, Ambani said that Jio was ready with “a complete 5G solution from scratch”. Even more astonishingly, he claimed it would be based “100 per cent on home-grown technologies and solutions”. Media reports a few days later indicated that Jio had applied to the government for 5G spectrum for test operations, even though 5G spectrum auctions are nowhere in sight. Reliance’s claims are breathtaking because 5G rollouts across the world have been sporadic since the first rollouts began in South Korea last year. Across the world, telecom equipment suppliers, operators and regulators are struggling with the still evolving 5G technology ecosystem. They are also disturbing for another reason—the threat of what its growing oligopolistic clout implies not just for competitors in general but also for how the adoption of nascent 5G standards and technologies progresses in India.

Even by the standards of a company that has the habit of making at least one “dhamaka” announcement at its annual general meeting since 2016, when it launched Jio services, this year’s event was truly spectacular, marked as it was by a string of such announcements. There was the announcement that Google would take a 7.7 per cent stake valued at Rs.33,737 crore in Jio Platforms, RIL’s technology venture. Then there was the declaration that Google would partner Jio to develop an Android-based phone, exclusively for Jio. The biggest of all was the announcement that Reliance was now completely debt-free, a year ahead of schedule.

Indigenous 5G technology

Ambani made no further reference to how Reliance proposed to lay out its 5G network. Instead, he said Jio would be ready for trials as soon as the relevant spectrum was made available, with its “made in India 5G solution”. “We are ready for field deployment next year,” he said. Making vague references to Jio’s “all-network architecture”, he let on that this would enable the operator to quickly “upgrade 4G networks to 5G”. Instead of offering any substantive details about the kind of 5G network Jio was planning, he drifted away and focussed on something completely different—Jio’s platform and the “solutions” available on it. He claimed that Jio was well positioned to emerge “as an exporter of 5G solutions to other telecom operators globally as a complete managed service”. He suitably peppered his speech with references to the usual IT buzzwords—artificial intelligence, big data analytics, cloud computing, block chain, and so on. Finally, in keeping with the ‘national mood’, he dedicated Jio’s 5G solution to Prime Minister Narendra Modi’s “highly motivating vision of Atmanirbhar Bharat”.

Ambani’s announcement that Jio is ready with an indigenously developed telecom network is spectacular because, across the world, telecom operators are struggling to roll out 5G. This is partly explained by the high cost of the rollout; it is evident that most operators are planning a phased rollout, targeting premium customers first before spreading the 5G network wider. Users of 4G would recall that this is how it happened when that technology made an appearance initially, including in India.

But even more crucially, 5G is not just about faster downloads, it is about setting up a network infrastructure that would enable large-scale networks to function in a mission critical mode on a real time basis. Managing a metropolis-wide network—or at the regional or national scale—which maps and gathers data on traffic patterns, pollution levels, electricity grids or telemedicine on a continuous and real-time basis— requires it to be fail-safe with low latency, which is what 5G is expected to facilitate. This is the qualitative shift that 5G marks over 4G. Reliance is not known to have invested anything in building such a network. The “solutions” that Ambani talked about are a distraction in this context. They are what float on a network; but the network itself is the real thing that remains to be built.

IPlytics, a Berlin-based company that monitors technology trends, reported that Huawei held the most number of patents among “5G patent families” in the world. Chinese companies, including Huawei and ZTE, commanded the largest number of such patents, closely followed by Korean companies, including Samsung and LG. As of November 2019, Huawei held 3,325 such patents; its nearest rival was Samsung, with 2,846 patents. Nokia’s alliance with Alcatel-Lucent was the most significant non-Asian rival with a combined tally of 2,308 patents. What is obvious is that a lot of grunt work has gone into the development of 5G. Samsung, for instance, started work on 5G networks several years ago; by 2013 it announced it had developed its own 5G system, and progress since then has been slow, even if steady. Incidentally, Reliance is reported to have sourced some 5G gear from Samsung. The fact that established industry leaders have taken many years to develop 5G technologies warrants a sceptical assessment of Ambani’s claim.

There was another giveaway, just two days before Ambani spoke. United Kingdom Prime Minister Boris Johnson announced that he was giving British telecom companies time until 2027 to rip out and replace all Huawei equipment from British networks. What does this extended seven-year time frame suggest? It implies that telecom companies would either risk, at the very least, a partial blackout across their networks or that the capital costs would be prohibitively high if the time frame were to be abridged. So, when an Indian telecom company, even if it happens to be the biggest, with a mere four-year track record and no significant patents in the relevant domain, makes the claim that it is positioned to establish a fully indigenous 5G network, a coherent explanation is needed.

Of course, given the precarious state of the finances of operators in many parts of the world, which is partly the result of the thin margins with which they work, telecom companies are also trying to build a “5G-like” network. In effect, because of the nascent stage of the technology ecosystem, the limited prospective subscriber base in the short term and the high costs associated with a full-scale rollout, industry participants are exploring options in the interim until the ecosystem matures. The idea is that if a full-scale 5G rollout is not immediately possible or feasible, patches based on software solutions as well as open source hardware could be built in order to give a “5G-like” network performance, at least in limited markets or demographies.

It is obvious that telecom operators are aiming to put in place specific features of “5G-like” performance without having to invest the humungous amounts that a full-scale 5G deployment demands. Verizon, the largest telecom operator in the United States, for instance, deployed its proprietary version a couple of years ago. It promised its customers high-speed data, but its protocols were not in compliance with the 5G NR (New Radio) standards laid down by the 3rd Generation Partnership Project (3GPP) that defines 5G standards. Although Verizon did roll out full-scale 5G services in select cities later, the deviation from globally defined 5G standards implies the adoption of a lower technology standard. This is tantamount to rolling out a network that necessarily precludes the potential that a fully compliant 5G network potentially offers. Conversely, from a user-perspective, the adoption of a lower standard severely limits possibilities that the shift to genuine 5G promised in terms of standards and quality of service.

Indeed, 3GPP, which set international telecom standards and protocols from 2G to 5G, has set benchmarks for an intermediate stage, what in industry parlance is referred to as 5G networks operating in Non-Standalone (NSA) mode. In this mode, initial 5G NR launches would ride on the existing LTE (4G) infrastructure. Worldwide, this is expected to act as a bridge to a full-blown 5G network. When that happens, the software as well as the hardware of the network will be completely independent of 4G infrastructure. The global telecom industry is still struggling to establish a full-scale 5G network that conforms to standards for a network core architecture set by 3GPP. This involves building a core that comprises switching, signalling, data management and a cloud interface involving both software and hardware domains, which would enable 5G in its true sense.

Network equipment companies and their vendors have recognised this and have been active on both fronts—“5G-like” as well as the real thing. All the top-tier telecom equipment vendors—from Huawei to Cisco—have their hands in both segments, realising that grabbing the markets immediately while neglecting the still-evolving 5G game would pose serious risks to their fortunes in the future.

In early July, Vodafone UK announced that it had conducted trials on a Standalone 5G network at Coventry University. It said the network was not limited by a 4G platform, which provided a breakthrough for faster speeds, reduced latency and better utilisation of spectrum resources across the network. Telecom equipment major Ericsson participated in the project at Coventry, the first such in a British university. Other companies such as Qualcomm also participated. Significantly, Oppo, the Chinese company that is known in India as a maker of lower-end smartphones, provided software solutions that enabled “slicing” of network switching, which results in better network efficiency. More recently, on July 19, Huawei announced that it has provided equipment and solutions for the launch of 5G data services by South African telecom operator rain in the Cape Town area, the first 5G service in the continent. Huawei said the solution was Standalone 5G, not one built as a patch on a 4G platform.

Nowhere in the race

Meanwhile, across the world, telecom operators and smaller equipment suppliers are exploring other options for an interim solution. Indeed, India’s two biggest operators, Jio and Airtel, are both members of the O-RAN Alliance, a body formed in 2018. Its primary objective is twofold. The first is to enable the building of radio access networks (RAN) that are based on an open interface, which allows smaller vendors to custom-build networks that are suited to specific classes of users. The second addresses the specific issues arising from the enormous tide of data flows, which imply that human interventions to monitor and direct network traffic in an optimal and “intelligent” manner are simply impossible. This requires harnessing computing power for artificial intelligence in networks.

If and when Reliance rolls out a 5G network it is most likely to ride on a version of the NSA. It is also obvious that the company is nowhere in the race to build a true-blue 5G network if and when technologies and markets are ready. Just two days after Ambani’s speech, Jio asked the Department of Telecommunications (DoT) to assign 800 megahertz (MHz) spectrum each in both the millimetre wave bands (26 gigahertz, and 24 GHz). It also sought 100 MHz in the 3.5 GHz band for field trials of its new network in a few metros. Millimetre bands waves (mmWave) bands are preferred for 5G operations because they offer much higher bandwidths when compared to existing 4G networks, although they are restricted by the limited range of signals. This, in turn, requires a much closer cluster of cells in a service area, which has cost ramifications for telecom operators.

The dismissal by the Supreme Court on July 20, in what is popularly referred to as the “AGR case”, of pleas from Jio’s two main rivals—Bharti Airtel and Vodafone Idea—for a revision or a staggered payment of their dues to the DoT also skews the pitch. While Airtel’s dues amount to Rs.25,976 crore, Vodafone’s amount to Rs.42,545 crore. This is bound to affect their ability to participate in not just the auction of 5G spectrum but in undertaking the huge capital investments that a 5G rollout would demand. Reliance clearly occupies pole position in the 5G race. But this could come at a significant cost to society. The development of 5G and associated technologies in India would be limited by Reliance’s own restricted capabilities in the field. In effect, Reliance’s effective monopoly would act as a stumbling block to further development in communication technologies in India. Thus, when the world moves on to full-blown 5G systems, India, because of its reliance on Reliance, would be restricted by the choices made today by its biggest private enterprise. That would be a travesty of the promise of 5G.