Just How Many Radios Do You Need, Anyway?

Technology consolidation - previously unthinkable - is now the most likely solution

I recently attended an IEEE Communications Society event that focused on ZigBee, a very interesting set of radios and protocols mostly designed for telemetry and control applications. And during the entire presentation, my mind wandered back to a fundamental question - how many radios do we really need in a handset? ZigBee isn't yet big in the home, but imagine a residential-automation system using ZigBee, handling security, energy management ("smart grid" is a key focus for the ZigBee community), home entertainment, personal communications, etc. etc. Gosh, given such, wouldn't it be great to have a ZigBee radio in a handset? Will such come to pass?

I doubt it. In fact, let me throw this out: in the not-too-distant future, the only radios in most handsets will be 802.11n Wi-Fi (dual-band) and LTE. That's it, and that's really all that's required.

What, no Bluetooth? Nope. It will be replaced, as a radio, by Wi-Fi, with Bluetooth protocols and applications running as necessary over various forms of Wi-Fi connections. This is already happening - check out Bluetooth 3.0, based on Wi-Fi. Wi-Fi headsets, you ask? Sure - I first saw such as a prototype about five years ago. Concerned about power consumption, the usual argument against such a strategy? No need. Wi-Fi implementations can be made very power-efficient at both the chip and protocol level. The biggest issue here is having to buy a new headset, but Wi-Fi headsets should be pretty inexpensive, and you've probably misplaced a bunch of these already anyway. Similarly, by the way, Bluetooth keyboards and mice will be superseded by their Wi-Fi equivalents - Wi-Fi Direct and Wi-Fi PANs.

No UWB? No, sorry, no. UWB did not live up to its hype, and 802.11n is plenty fast. 60 GHz. is more than capable of supporting both USB 2.0 and 3.0. I would still like to see a "quick-connect" MAC mode for Wi-Fi, and I think we'll get that at some point. Thus Wi-Fi will be more usable than ever in sync and related applications, as well as serving as a carrier for USB. That's the theme - current implementations that go all that way to the PHY can stop at the DLL, and thus become upper-layer protocols only. Even ZigBee is being implemented on IP.

The only other radio I could see including is near-field communications (NFC), which may catch on in credit-card applications. Note even this can be replaced by Wi-Fi, assuming proper authentication protocols. But, then, that's really the message here. We can implement a broad range of upper-layer protocols on top of just a couple of MAC/PHY combinations with no compromise in application functionality. Thus we need Wi-Fi for short-range communications, and LTE for longer range. And, of course, we have the ability to dynamically hand off between the two. And there's no need for femtocells if one has Wi-Fi. Such a strategy maximizes the use of the spectrum, applying the best tool for the job, again with no limitations on the ultimate application(s). And, of course, gateways can be applied where necessary, as has always been the case in networking.

So I can't think of any reason, apart from the lobbying of chip-driven trade associations, why all of this doesn't make sense. Occam and Thoreau hard at work once again.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2010 IDG Communications, Inc.

SD-WAN buyers guide: Key questions to ask vendors (and yourself)