Millimetre wave.. omigerd it's going nowherrr.. Apple, you say?
Guess who's joined the bandwagon...
Comment Apple may not be the invincible force it once was in mobile, but it is still unrivalled in its ability to scatter stardust over new technologies – just ask the companies which struggled to push Wi-Fi Calling or wireless charging into the mainstream before the iPhone maker came along. Now it has kindled new sparks of enthusiasm, or perhaps hype, around millimetre wave spectrum by applying for an experimental licence to test high frequency bands in the US.
The mmWave bands are being closely watched as a potential new source of plentiful spectrum capacity. However, despite some interesting trials in Japan and South Korea, the major commercially oriented activity is confined to the US and fixed wireless access. Also, the challenges of deploying in these bands are starting to be fully appreciated, risking a bursting of the bubble.
Apple plans mmWave tests in California
So it is timely for mmWave enthusiasts that Apple has helped restore the faith. It has asked the FCC for permission to test in 28 GHz and 39 GHz bands – the main ones which are envisaged for commercial services from Verizon and AT&T from next year. Verizon is, so far, the major holder of these airwaves, having made several acquisitions to secure licences in both bands. AT&T has 39 GHz licences and T-Mobile gained some 28 GHz spectrum when it bought MetroPCS. Further mmWave airwaves will be auctioned at a future, as-yet undecided date.
Apple aims to carry out the tests at its Silicon Valley headquarters and in Milpitas, California. "Apple Inc. seeks to assess cellular link performance in direct path and multipath environments between base station transmitters and receivers using this spectrum. These assessments will provide engineering data relevant to the operation of devices on wireless carriers' future 5G networks," the company said in its filing.
Light Reading reported in November that Apple was advertising for “multi-gigabit” mmWave chip designers, though there are many issues with implementing these high frequency radios in mobile devices at affordable cost and acceptable battery life. Complex antenna design and beamsteering will be required to address the limited propagation of low power equipment in high frequencies – the most prominent trial of mobile services, outside the lab, is being undertaken by Softbank and Ericsson in Japan.
Millimetre wave bands are in use already in wireless – for point-to-point "wireless fibre" high speed links including backhaul, and in the 60 GHz band, which supports the Wi-Fi-like WiGig. This has been implemented in chipsets suited to handsets although its main commercial use so far has been for high speed, short range connectivity between devices – to transfer video between TVs and tablets in the home, for instance, or to link PCs to peripherals.
However, Apple could get into the 28 GHz/39 GHz game early by designing a home gateway, Apple TV or iPad which could link to the operators’ fixed wireless services (the tablet could still have a sub-6 GHz radio for mobility purposes). Apple has tested its DirecTV services in 39 GHz.
Samsung is in pole position to deliver the first mmWave smartphone, but it has suggested this will not happen until mid-2019. The Korean vendor has been a significant investor in mmWave R&D and has been involved both in operators’ network trials and on the device side. Meanwhile, Intel and Qualcomm are promising mmWave chip samples later this year and production chips in 2019.
Verizon encouraged by its own trial results
And while mmWave may not yet be catching the imaginations of all mobile operator CTOs – European MNOs and regulators, in particular, remain cool – the enthusiasts are getting more confident about the ability to deploy commercial services in 2019 or even earlier. Verizon CEO Lowell McAdam told a JP Morgan investor conference this month that small cells in high frequency spectrum were starting to make sound economic sense – both for densification, compared to buying new spectrum; and compared to building fiber-to-thehome out everywhere for fixed access.
He said: “The cost is minuscule to be able to address a very large market outside the Washington to Boston corridor, so as we have built that architecture, 5G on a fixed broadband perspective was the best application for us. We did not need to wait for all of the mobile standards, we did not have to wait for it to be crammed into an iPhone or a Samsung Galaxy device. You can use basically your home router that you have today, just put some different chips in it and we’re working with Intel on that, and you are then a broadband provider and a TV provider outside of your franchise footprint.”
He said Verizon was expecting “less than 1,000 feet of propagation to deliver 1 Gig”, and that some tests in Korea and elsewhere were showing far better distances, because of the significant advances in beamforming and Massive MIMO antennas, which are helping engineers address the traditional problems of high frequency spectrum, such as foliage blocking line of sight.
He said that, in one test in Samsung’s technology park, the system was delivering 1.8Gbps over 2,000 feet. He added: “We said OK, take that truck, drive it around the back side of the building so there is no possible way you will have a direct line of sight—2,000 feet away, it delivered 1.4Gbps of throughput. In the field today, we’re doing heavy foliage, we’re doing downtown urban areas, we’re doing residential, we’re doing long haul where you’re out in a rural area and you don’t have anything for five miles before you get to the next house.”
It is not just McAdam who is pointing out the huge leaps in antenna technology in recent years, which have made the high frequency bands look economically viable for wireless applications. Even Seizo Onoe, the influential CTO of NTT Docomo, has become more optimistic as he has seen some of the new developments. He previously did not foresee mainstream deployments in mmWave spectrum before 2022 or 2023, but now thinks that timescale will come forward by a couple of years. Even the fast-tracked subset of the 3GPP 5G New Radio standard, NR Non-Standalone, will support high frequency bands.
Copyright © 2017, Wireless Watch
Wireless Watch is published by Rethink Research, a London-based IT publishing and consulting firm. This weekly newsletter delivers in-depth analysis and market research of mobile and wireless for business. Subscription details are here.