With disputes looming in many of the millimeter wave bands (see preceding item), the cellular industry is scrambling to make the argument for most of this spectrum being turned over to 5G. The CTIA, in the USA, has proposed to squeeze satellite incumbents into part of the 50 GHz band; while the GSMA is wooing regulators and goverments with claims that mmWave 5G could boost the total growth in GDP, enabled by 5G, by 25% – and boost tax coffers.
Harnessing high frequency spectrum to support innovative 5G services could add $565bn to global GDP and $152bn in tax revenue from 2020 to 2034 , according to the GSMA-commissioned report, entitled ‘Socio-Economic Benefits of 5G Services Provided in mmWave Bands’. However, this is contingent on governments making policy to release this spectrum quickly, and to present strong plans at the next ITU World Radio Conference in 2019 (WRC-19).
“The global mobile ecosystem knows how to make spectrum work to deliver a better future,” said Brett Tarnutzer, head of spectrum at the GSMA. “Mobile operators have a history of maximizing the impact of our spectrum resources and no one else has done more to transform spectrum allocations into services that are changing people’s lives. Planning spectrum is essential to enable the highest 5G performance and government backing for mmWave mobile spectrum at WRC-19 will unlock the greatest value from 5G deployments for their citizens.”
Applications identified by the GSMA, which would benefit from the fast data rates and high capacity enabled by high frequency spectrum, included telemedicine supported by Tactile Internet capabilities; next generation robots and drones; and (inevitable in any report seeking to influence government policy) autonomous cars.
The GSMA recommends supporting the 26 GHz, 40 GHz and 66-71 GHz bands for globally harmonized mobile broadband, though it acknowledges that 28 GHz, because of the influence of the USA in particular, is emerging as an important band in the commercial world, even if it was not identified as a candidate at WRC-15. As well as the USA, regulators in Canada, South Korea, Japan and India plan to use 28 GHz.
(The report can be found here. It was produced by consulting firm TMG.)
The applications for mmWave 5G are mainly distant in time, either because of commercial and regulatory uncertainties (autonomous cars or robotic surgery), or because, for now, they can be supported using 4G or sub-6 GHz 5G, giving operators breathing space to hold off mmWave deployment until the ecosystem is mature and various technical issues, such as power consumption, addressed.
The exceptions are applications which hark back to the old days of fixed wireless, when the 28 GHz and 39 GHz bands were called LMDS. Fixed wireless access (FWA) itself can be a short term opportunity to harness high frequency spectrum, as Verizon has proved, and could be especially useful to broadband providers to replace fiber in the last few feet of a home connection. However, in most markets (the USA, with its peculiar regional structure, being an exception), FWA revenues will be incremental. For non-wireline providers, FWA only makes commercial sense in unserved areas – elsewhere, it will always be in a price war with fiber. And for wireline operators, FWA can lower the cost of getting gigabit broadband to every home, but it does not generate additional revenue.
The other, related application with some short term promise is 5G backhaul or fronthaul. Again, the connectivity is fixed, so no need to worry about the handset ecosystem. The high capacity and increased density of 5G networks will open up opportunities to use wireless backhaul, since fiber to every small cell will be impractical in most scenarios; and in small-footprint virtualised networks, even fronthaul (links between the central baseband and the remote radio head) could be wireless.
ETSI has an ISG (Industry Specification Group) focusing on mmWave transmission (the mWT ISG), and it has released new reports and guidelines on how high frequency microwave and mmWave spectrum could be used for 5G backhaul. In its latest publication, the group does not envisage mmWave backhaul being required until after 2020, but argues that the technical and regulatory groundwork must be put in place right now.
It is assuming that very dense networks will be needed, at least in the second phase of 5G deployment, to satisfy rising quality of experience expectations, and is looking at a typical 5G area traffic capacity target of 10Mbps per m² over the next few years. Backhaul and fronthaul for networks of this density will require technical advances like
- carrier and band aggregation spanning many bands
- advanced interference cancellation techniques such as those being developed by wireless full duplex pioneers like Kumu
- higher directivity enabled by ever-smarter antennas
- software-defined networking to automate inter-connectivity and resource allocation in a dense network.
It will also require more backhaul spectrum, argues the ISG, including new frequencies in the W-band, D-band, and other mmWave or microwave bands, plus wider channels for microwave bands.
The W-band ranges from 75 GHz to 110 GHz with wavelengths of 2.7-4 millimeters. It sits above the V-band (40-75 GHz), which has already been used for small cell backhaul products, as has the overlapping E-band (60-90 GHz), though with limited commercial success so far. That was because of the limited deployment of outdoor small cells to date – something that ETSI, among others, expects to change in 5G – and because the early solutions were expensive and complex. The D-band ranges from 110 GHz to 170 GHz, with wavelengths of 1.8-2.7 millimeters.
ETSI calls on regulators to formulate policy, especially with regard to the cost of spectrum to support these emerging use case requirements, saying this regulatory preparedness is of “paramount importance” to the future of dense network X-haul (backhaul or fronthaul supporting various RAN architecture splits), and therefore of 5G.