Last decade Intel decided to change the wireless industry by putting its considerable might behind WiMAX, the first of the so-called 4G standards. That was a bet that didn’t pay off. But Intel apparently hasn’t given up on its dreams of becoming a powerhouse in mobile networking, though it’s now casting its gaze much further into the future.
The silicon giant has created a group called the Intel Strategic Research Alliance (ISRA), bringing university labs and carriers together to focus on the development of 5G technologies. Other 5G exploratory groups like METIS have sprung up recently, funded by euros from the European Union, but ISRA is the first U.S.-company effort I’ve seen emerge. Intel is kicking off ISRA with $3 million in funding, which is pretty small considering the billions in R&D dollars it will take to fully develop future 5G technologies. But trust me, Intel has plenty of time to invest more.
5G is really just the merest of concepts today. Many engineers would argue we’ve yet to see our first 4G network, making the idea of 5G a lab rat’s fantasy. We’re talking technology well beyond LTE-Advanced (which itself is much further away than the mobile industry would have us believe) on the 2020 horizon if not more distant. I applaud Intel for not abusing the term 5G, though. Unlike other mobile vendors — cough, Samsung — Intel in its blog post announcing the initiative positioned 5G exactly as it should be: an aspirational and still unidentified set of technologies, not a network that is being developed today.
What would 5G networks look like? Intel has some ideas. According to Intel Principal Engineer Shilpa Talwar, 5G networking would move debate beyond the tired discussion of raw speed and focus on pervasive connectivity — getting a fast, resilient link to the internet whether you’re in a subway tunnel, at the top of skyscraper or in a conference center vying with 50,000 other people for a network connection.
Talwar believes 5G networks should scale to support billions of devices, not just our smartphones and tablets but the billions of appliances, sensors and objects that make up the internet of things. Finally, Talwar proposes 5G networks adapt to the tasks that they’re given, tailoring their speeds, latencies and quality of service to the applications and devices that connect to them at any given moment. Your HD video conference call needs extremely low latency and adequate bandwidth to work properly, but your wearable’s upload of health data to the cloud does not. Why give the same type of connection to each?
Those are all lofty goals, but there’s one thing Talwar failed to mention that I believe really belongs front and center in any mobile networking conversation: 5G networks need to make the cost of mobile data dirt cheap. If we’re still paying $10 to $20 a gigabyte to connect to the mobile network in 2020, then what’s the point of innovation? Our current networks can handle that load easily, and the mobile carriers are free to become broadband equivalent of electrical utilities. If driving down the cost of mobile data isn’t one on ISRA’s goals, then Intel is just throwing its $3 million investment away.
ISRA has already attracted an impressive array of research institutions including the USC, NYU, Princeton, Stanford, the University of Illinois, the University of Texas, the University Fabra Pompeu, Purdue, Cornell, IIT Delhi, UCLA, Rice University, and Macquarie University. Verizon has also signed on as its first carrier member.
I’m curious to see how far Intel takes this effort. Usually these kinds of standards initiatives are spearheaded by the big wireless equipment vendors – the Ericssons, Huaweis, Alcatel-Lucents and Nokias of the world – but Intel seems dead set at becoming a company that matters in the world of mobile networking. WiMAX may have been a bum investment, but it’s pursued multiple other mobile avenues since then.
Intel bought Infineon in 2010, making it an instant player in mobile device radio chipsets. And it’s been trying to position as its Xeon processors as the brains of the mobile radio network. Maybe with 5G, Intel’s luck will change.