Week in Review: FCC’s new spectrum heralds end of MNOs’ business model
While US operators will be first to harness millimeter wave bands, the real 5G beneficiaries will not be traditional mobile players
Yesterday, the FCC voted unanimously to open up high frequency spectrum bands for 5G networks, as proposed earlier in the year.
This certainly sees the US taking a lead in pushing millimeter wave wireless networks forward, rather than just talking about them. But while Verizon, AT&T and T-Mobile are racing ahead, for most operators round the world, millimeter wave is a distant prospect. Senior MNO executives continue to issue warning statements, calling on the industry to slow down and ensure that LTE continues to be enhanced for a decade or more to come. And even for those operators which believe they know what 5G will be, and aim to deploy it in the early 2020s, the first focus – as with LTE – will be on conventional sub-3 GHz bands and conventional mobile broadband and fixed wireless services.
The radical changes in applications and network behaviour, which millimeter wave will allow, will follow later – even the 3GPP is not expected to complete its standards for these high frequency bands in its first wave of specifications in 2018.
The hype around millimeter wave needs to stop for a while, and the industry needs to put a brake on pushing aspects of 5G which, for most MNOs, are not yet urgent. But looking ahead to the dense zones of small, high capacity cells which spectrum above 24 GHz will support, we can start to glimpse the disintegration of the established mobile order, which dramatic changes in the networks will hasten.
Some of that is happening already. Facebook’s OpenCellular and Terragraph, and Google’s Loon, show how open technologies can be applied to access networks, breaking the stranglehold of the traditional vendors with their closely guarded magic, but also lowering the barriers to new entrant operators. MNOs from Bouygues to Saudi Telecom to Deutsche Telekom discussed the divestment of their towers this week – this is just one small step to a deconstructured market where infrastructure is shared, capacity flexibly allocated, and core technology placed into open source.
The first great enabler of change will be virtualization rather than 5G. The ability to slice up a network and allocate virtual chunks to individual enterprises or MVNOs is promising, especially for vertical industries, and AT&T and Orange are among those leading the way. Orange Business Services is planning an international expansion of its network-on-demand offering, which incorporates some of the thinking, in an early form, which will go into dynamic network slicing at the 5G stage.
If the mobile operators have their challenges, the handset makers are also being forced to adjust quickly to harsh new realities as their smartphone growth engine sputters and slows. Like the MNOs, which have to keep improving their data rates even as ARPUs fall, the device vendors still have to keep rolling out ever-more powerful products to try to reverse the lengthening upgrade cycles – even though margins are shrinking and price wars are setting in.
Qualcomm is pushing the next handset cycle with the release of its fastest Snapdragon processor, the 821 – a successor for the 820, which will possibly make its commercial debut in the next Google Nexus devices. With the 820/821 system-on-chip designs the firm has really put the dark days of the overheating 810 behind it, as well as supply issues which dogged it a couple of years ago.
Of course it has other huge challenges to its mobile dominance, most of them coming from China, but it is looking to harness some of the wave of Chinese semiconductor innovation too – this week saw its recently formed joint venture, Huaxintong Semiconductor Technology, taking a multiyear ARM v8-A architectural licence to make server chipsets, mainly for the Chinese market. Qualcomm is contributing design expertise as well as funding and looks like it has the chance to attack the US server makers with its own new server SoC platform, while addressing the Chinese OEMs via a proxy.
The world has changed dramatically for Qualcomm and increasingly it will have to move into Intel strongholds to achieve growth, which may prove uncomfortable. Even its famous IPR business will have to adapt rapidly – to the rise of open source in mobile networks; the potential overshadowing of 3GPP standards in 5G, by IT technologies; and of course to the decline of the CDMA networks which Qualcomm utterly controlled. A wave of announcements is under way, from operators planning their CDMA shutdowns – Verizon will do this by 2019, India’s CDMA services are likely to be gone by next year. Like the MNOs and handset makers, Qualcomm needs to take tough decisions to survive when it can no longer set the rules.