If 4G isn’t working, why stick to the same approach for 5G?

Analysis The start of 2015 is sure to bring an even greater intensity of interest in what "5G" might be. After all, some operators are saying they will start commercial deployments in 2020, and a five-year time-scale is short enough to induce panic attacks.

LTE has set a strong precedent here. The specifications that became LTE were first proposed in 2004 and official studies began in 2005, and TeliaSonera switched on the world’s first commercial network right at the end of 2009. But the five-year cycle ahead will be very different from that of 2004-9.

It was already known, before NTT Docomo and others made their proposals, roughly what LTE would be. It would be an air interface upgrade with more complex RAN and packet core elements than earlier standards, and with a new modulation scheme – but nevertheless, a network evolution of a familiar kind. It would be rolled out in a similar way to 3G and would address familiar targets such as improved speed and latency.

In the run-up to 5G, all the old certainties have dissolved. Will 5G need a new air interface at all, or should it have several? Is it more important to slash power consumption than push further increases in data rate? Will networks be deployed in a remotely recognisable way after 2020 and will there be any spectrum left for such deployments?

None of those questions can wait until 2020 to be answered, or at least addressed. Already, 4G is showing its limitations for the business models of the modern carrier. Early adopters, with their fairly conventional roll-out patterns and consumer propositions, have reaped little premium from LTE and are already seeing their prof-its being squeezed, as even the mighty Verizon Wireless warned this week.

Those cellcos which have tried to leverage 4G for new revenue streams, notably in the internet of things, have found a network - built for wide coverage, mobile voice and fast data – to be poorly suited to ultra-low power or deeply granular services.

So dramatic changes are underway even in the way LTE and LTE-Advanced are being planned – ever-smaller cells, integrated WiFi, virtualisation, a revised "LTE 0" better adapted to machine-to-machine usage, steps towards fully automated, self-managing networks and dynamic bandwidth sharing across many frequencies. Put all that together, in a software-defined framework, and you may well have 5G, or at least not need another disruptive and expensive step upgrade.

Copyright © 2014, 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.