Let the Wireless Wars Begin

It’s been an interesting week for the short range wireless standards.  The two terrible teenagers, ANT and ZigBee have both shown signs of their growing maturity, starting to position themselves as far more serious contenders in the market place.  In the wake of their move from adolescence, a new toddler has emerged in the form of Toumaz, with their announcement of their Telran chip.

What has been missing is any reaction, or in fact much sign of any action from their elder siblings – Bluetooth and Wi-Fi.  As large manufacturers continue to tighten their belts, one of the less noticed effects has been a steady withdrawal of engineering support from standards organisations.  In the past, many of these have been staffed with seconded experts from the big names in industry.  Increasingly those big names are withdrawing, relying largely on chip vendors to push their interests within the standards organisations.  That’s left Wi-Fi and Bluetooth battling to persuade industry members that either standard has a development future, with certain of their members considering that the job has been done.

Which opens up the field for the former competitors to claim some potentially interesting parts of the market.

ZigBee has been around for almost seven years, but has struggled to gain traction.  Part of its problem is that unlike Bluetooth and Wi-Fi, it hasn’t got a free ride from being incorporated into phones or PCs.  That’s meant it’s had to fight for every chip that its members have sold.  During those seven years it’s tried its arm in a number of different application areas before selling its soul to the smart metering movement (and to a lesser degree, home automation).  That Faustian pact looks as if it may be about to pay off.

Last week the ZigBee Alliance made the provocative move of holding its quarterly meeting in Seattle, home of the Bluetooth SIG.  In the event, it’s not obvious that they even disturbed the sleep of their neighbourly slumbering giant.  The meeting’s main purpose was to finalise and approve the latest Version 1.1 of the ZigBee Smart Energy Profile (SEP 1.1).  That happened, so the world now has SEP 1.1.  In addition, the meeting managed to do a host of useful work on moving towards the next version – SEP 2.0, and making progress on the Home Automation Profiles.

SEP 1.1 is an extremely important step for ZigBee, as it signifies a bridge between today’s form of ZigBee mesh and the future form, which is IP based, drawing on the work of the 6LoWPAN initiative.  Every certified ZigBee produce that exists in the market today uses ZigBee’s own mesh protocol.  That has been refined through a number of releases to be a robust mesh network.  However, it does not allow every node to be addressed from an IP network.  That requires a fundamentally new layer to be added, which is the core of SEP 2.0.  In the US, NIST has mandated IP addressability for future smart grid standards, and elsewhere in the world, many other governments and regulators are thinking whether they should follow that lead.

The problem is that the two protocols of pre- and post-IP ZigBee are not compatible.  For utilities that are currently starting smart meter deployments that’s a big concern, as if they standardise on meters with ZigBee SEP 1.0 today, they risk having millions of stranded assets that may be incompatible with their future infrastructure.  That’s where SEP 1.1 comes in, and why it’s so important.  The main addition that it contains is a standardised upload method, which will allow devices in the field to be upgraded over the air to a future ZigBee 2.0 standard.  That should give the smart metering industry the confidence to move ahead with deployments, with a much lower risk of stranded meters.  There are still some questions to be asked, not least of which is whether an SEP 1.1 device will have sufficient memory and computing resource to run a future 2.0 Smart Energy Profile?  And we won’t necessarily know the answer to that for another six months, when work on that profile will be drawing to a close.  But with that caveat, SEP 1.1 should be lifting a considerable weight from the minds of utilities contemplating a smart meter project.

ANT weren’t meeting this week, but released an “independent” analysis of the power consumption of different wireless standards, blasting ZigBee as “cumbersome, high power and fragmented“.   ANT has come from being a small scale, proprietary standard for sorts and fitness devices, to being a serious contender that is also beginning to make inroads in medical devices.

In some senses, ANT is still proprietary, as it’s owned and managed by Dynastream, a company in turn owned by Garmin.  They’ve resisted opening the standard up as a wider industry effort, as they claim that this would result in unnecessary committees, doubtless agreeing with the old adage that a committee is “something that takes minutes and wastes hours”.  Ironically that’s not dissimilar to the attitude of Bluetooth’s five founding members, who felt that they could do a more expedient job of getting a standard to market by keeping the work to themselves.  What goes around, goes around…

What has moved ANT out of the proprietary regime is a growing number of silicon vendors who are supporting, or looking to support it.  As well as the original devices from Nordic Semiconductor, you can now find ANT support in TI chips, including those that go into mobile phones.  ANT has been doing a very good job of allowing manufacturers to tell it what profiles they want and then implementing them, with the result that a growing number of health and fitness devices are sporting ANT.

In contrast, the ability of Bluetooth, Wi-Fi and ZigBee to get into health devices has been at best lacklustre.  Despite the Continua Health Alliance selecting both Bluetooth and ZigBee as its preferred wireless standards, ANT is the standard that’s making the running in the marketplace.

The newcomer, or the toddler showing promise is Toumaz – a UK start-up, which has been active for some years in low power, sub GHz Body Area Wireless Networks, notably with its Sensium platform.  Their latest chip – the TZ1053 “Telran” (and I have no idea what that means – the only reference I can find to Telran is an Israeli company selling set top boxes and food mixers), shows that they’re keen to move past their medical market and try to take a bite out of wireless sensor networks, environmental monitoring and smart metering.

Toumaz differs in that it’s not running at 2.4GHz, and offers higher range and lower power, such that it will run off a coin cell for years.  If you think you’ve heard that before, you’d be right – it’s what Bluetooth chip companies have been claiming for Bluetooth low energy.  I’ve certainly seen demonstrations that prove that they can achieve these figures.  But with no clear indication of when profiles will arrive for Bluetooth low energy, and with Nokia – the key driver of the technology, (dating back from when it was WiBree) having rather more important issues to deal with, the question arises as to whether the Bluetooth SIG is fiddling whilst Rome burns?

A few years ago the short range wireless hegemony looked fairly well established.  Wi-Fi would do internet access to static laptops, Bluetooth would do voice and anything connected to a phone, ZigBee would probably do smart energy, and anything else was up for grabs.  Bluetooth low energy and Wi-Fi then came along and threatened to change the status quo, but most of that early enthusiasm seems to have disappeared as the desire to grow and own new ecosystems has waned.  ZigBee now looks as if it will repel any boarders on the smart energy front, and ANT may well be the eventual winner for healthcare.  There’s still a healthy business for low cost proprietary wireless in everything from keyboards to toys to burglar alarms, and it looks as if that might stay around for a lot longer than anyone thought.

The only question now is whether any of the standards will fail?  Most of them have the ability to replace some of the other applications, but don’t necessarily appear to have the stomach for a battle.  Instead they may play a waiting game, hoping that something upsets one of the markets, giving them a chance to come in and pick up the pieces.   The one thing that has changed for all of them is the level of industry participation.  To move forward, they’re having to do more with fewer resources.  It may well be that is the key factor that maintains the status quo we seem to have arrived at.  If so, it will be interesting to see whether new or derivative standards are successful, or whether the future is limited to muddling along with the current incumbents?