The news that Apple had joined the NGMN (Next Generation Mobile Networks) Alliance came a week after Intel signed up to the board of governors of 4G Americas. Both were examples of the changing face of wireless power, and hinted at how different will be the big influencers of ‘5G’ standards, compared to its predecessors.
Intel famously tried to create its own ‘4G’ platform around WiMAX, but is now taking the same view as most companies entering the mobile world from IT or the web – live with 3G and 4G for now, however alien to one’s heritage in terms of IPR and structures, but make sure you have a hefty influence on emerging platforms, so that the norms change over time and shift away from those set by the cellular sector.
This is starting to be seen in the ongoing discussions about what ‘5G’ may consist of. Some are still focusing on the conventional air interface step upgrade, defined by the rarefied processes of the 3GPP. Hossein Moiin, CTO of Nokia Networks, told FierceWireless that he thought a new air interface would be necessary, though with some caveats – but of course, Nokia is one of the companies with the greatest interest in prolonging the 3GPP model a bit longer, especially now it does not have to worry about the impact of the traditional IPR royalties on its handset business.
Other, less traditional mobile heavyweights are, predictably, focusing on aspects which play to their own strengths and heritage – virtualization for the enterprise giants like Cisco and Intel; user experience and non-carrier models, for Google and Apple; ultra-densification and ultra-low power for the internet of things and vertical influencers.
Apple, which has tended to build its own platforms rather than participating in standards to a large extent, clearly feels it must stamp its mark on 5G, and has taken the rare step of joining the NGMN Alliance, particularly to participate in its 5G Workshop.
This is just one example of how, whichever bodies contribute to the mesh of specifications and concepts which is likely to form ‘5G’, the voices will be very different. The NGMN Alliance itself has traditionally been led heavily by cellcos (its board consists of 20 CTOs from leading operators), but is starting to become far more diverse. Cableco Ziggo, part of the Liberty Global Group, was another new member, along with Apple, as was Peking University of China, indicating not only the rising influence of Chinese organizations on future architectures, but also the fact that all network operators and service providers, not just MNOs, are interested in wireless now. Ziggo is an aggressive deployer of homespots and hotspots to support its moves towards a quad play, and the ability to create a seamless pool of capacity using many air interfaces and spectrum bands will certainly be a key theme of 5G developments.
At a workshop in Beijing, the NGMN Alliance presented the latest results of its 5G Initiative, which aims to present end-to-end requirements to the industry in a white paper as well as be a talking shop for potential solutions. A global team is working on the definition of these consolidated operator requirements for 2020 and beyond, but will increasingly be taking into consideration the needs of non-MNOs.
And at an event across the world in Silicon Valley, Intel was talking about new air interfaces for 5G too – two of them, indeed, one specifically for the millimeter wave bands such as 60GHz and above, and one for IoT access. This was the view of the company’s chief wireless technologist, Kenneth Stewart, speaking at an even hosted by gigabit WiFi chip pioneer Quantenna. So rather than trying to shoehorn existing designs into radically new frequency bands and applications, Stewart said many devices would benefit from a tailored air interface in future – such as devices “that don’t have to sync before they transmit”.
This is another example of the different standpoints which the 5G discussion will encourage, as will the new players in the mobile chain. Intel is talking about standards “targeted to specific use cases” rather than generic platforms which should be able to support almost any application. Stewart’s concept would see wireless chips supporting GSM, 3G, 4G, WiFi, WiGig and other future PHY level standards. The baseband would also include on-chip location processing; options for ultra-low power operations; and inter-RAT connection management for tasks like WiFi offloading and multi-RAT aggregation. The companion RF chip would also have to support many enhanced functions such as up to 60 frequency bands and new levels of multi-antenna operations, envelope tracking and auto interference support.
“We don’t see a fundamental obstacle to delivering this device,” said the executive optimistically, but pointed out that the cleverest chip on the planet still needed a killer device which would transform the user experience as the iPhone and others did for 3G and 4G.
At the same event, the father of massive MIMO, Thomas Marzetta of Bell Labs, described research in which 1Gbps speeds were delivered to each of 10 users via 675 antennas in relatively small amounts of spectrum. Marzetta, as quoted in EETimes, boasted that his technology could reduce the need for small cells and would replace today’s point-to-point approaches to MIMO. “I believe LTE and WiFi are ready for early retirement if they do not convert to massive MIMO,” he said, though this technology is very much in its infancy, like many predicted 5G elements.
Like most 5G discussions, the emphasis soon shifted to high frequency spectrum. “Most of the interest and benefits of 5G lie in millimeter wave bands around 28GHz to 90GHz,” said Stewart, who sees the high capacity and data rates of these bands being necessary to hit the ITU’s goal of 10Gbps data rates for some 5G services. WiFi is already heading this way via 60GHz WiGig, and Stewart predicted that LTE will hit gigabit data rates by the end of the decade too.
However, there is little consensus on which high frequency bands will be most feasible for wireless standards around 2020. The ITU will publish a report looking at the technical feasibility of IMT in the bands above 6GHz in June, to accompany its WRC-15 World Radio Conference, which will looking at some of the 5G spectrum issues, though the WRC-18 event may be more impactful.
The 28GHz band is popular with early ‘5G’ demonstrations, like Samsung’s at this year’s Mobile World Congress, though at a London conference, called 5G Huddle, last week, the China Academy of Telecoms Technology said the main demand in 2020 would still be in conventional bands like 1.8GHz, though its speaker, Shaoli Kang, kept the options open for later dates, saying the range under investigation was “6GHz to 100GHz”.
In the US, the FCC and the government’s NIST (National Institute of Standards and Technology) are investigating more licence-exempt options, notably the U-NII-2B and U-NII-4 bands, and the FCC has issued a notice of inquiry into the feasibility of millimeter wave spectrum above 24GHz. In a blog post, FCC chairman Tom Wheeler wrote that “early studies show that these new technologies – what some are calling ‘5G’ – can ultimately facilitate a throughput of up to 10Gbps, a speed that is orders of magnitude greater than that available today. Our effort here is to learn about the technology and ensure a regulatory environment where these technologies can flourish.”