The domination of the 802.16 process by a handful of vendors is often criticized, but at least it has resulted in a measure of unity that is looking increasingly enviable as the other key IEEE wireless efforts fall prey to dangerous fragmentation, which could significantly devalue the role of the standards body itself. We have examined the deep splits in the 802.11 camp over the approach to the upcoming 802.11n standard for 100Mbps-plus Wi-Fi, and the even more damaging battle over the proposed 802.15.3a specification for high data rate, short range communications based on UltraWideBand. Now 802.15.4, or ZigBee, the standard for low data rate personal area networking, is also threatening to shatter into conflicting pieces.
The problem, according to a new study from West Technology Research Solutions, is one already familiar in Wi-Fi – the race by vendors to build market share by releasing proprietary, ‘prestandard’ products. This means that, “instead of becoming a standard for low data rate network environments, ZigBee is in danger of evolving into simply one among many proprietary options. The proliferation of proprietary 802.15.4 solutions in advance of the availability of the ZigBee standard has effectively marginalized the overall market opportunity for ZigBee”, says West.
To reflect this risk, West has downgraded its predictions for ZigBee chip shipments and now expects gradual growth from 19m in 2006 to 123m units in 2009. This would put ZigBee at under 20 per cent of the total market, with proprietary variants controlling the rest.
More sanguine is ABI Research, which believes around 80m ZigBee chips will ship by the end of 2006. ZigBee is generally associated with industrial control applications but its biggest opportunity lies in the digital home, ABI believes. It says that the market for wireless home controls and other short range applications will explode in the home from 2006, and this is an easier sector, technologically, for vendors to enter than industrial networks.
Bob Metcalfe, founder of 3Com and so-called ‘father of the internet’, commented: "I see parallels between ZigBee and Ethernet. As Ethernet has connected millions of computers, [ZigBee] has the potential to net-work-enable billions of devices that previously couldn't communicate.” He was speaking on taking the chairman’s role at ZigBee mesh start-up Ember, which has raised $25m in funding. In April, Ember, maker of sensor-based meshes based on pre-standard ZigBee, acquired technology and personnel for ZigBee from the UK’s Cambridge Consultants. The company is working on a single-chip device for the personal area standard and expects the technology to be running on an UWB physical layer by 2006, which will increase speed and power efficiency.
This shows how the innovation and release of products is running well ahead of the yet-to-be finalized standards. This hints at a future – echoed in 802.15.3a and 802.11 – where the impetus behind a new market will be sufficient to create a mass market for a technology, even in de facto variants, where real standards cannot be agreed in a sufficiently rapid timescale to meet market demand.
UltraWideBand in ZigBee
The next ZigBee challenge will be devising the proposed extension to the 802.15.4 standard, ‘4a’, which could be based on UltraWideBand too. This would make UWB the common technology for both low and high data rate short range networks. However, the continuing standards war, and severe limitations on where and how UWB is regulated, could hold it back so badly that its place in its largest potential market, the digital wireless home network, will be usurped by 802.11n (assuming that can overcome its own splits).
While chipmakers and analysts fret over that clash, start-up Pulse~Link, which has already demonstrated UWB over wires and over Lan distances, was taking an inclusive approach, showing off a chip that can support both UWB and narrowband wireless, including Wi-Fi, simultaneously. At this week’s UltraWideBand Conference in California, the company showed UWB communications across in-home electrical wiring, on a 750MHz cable television network, and using narrowband carriers at 2.4GHz and 5GHz. All transmissions were demonstrated in simultaneous operation from a single chip.
Again, the market predictions are so divergent as to be irrelevant. In-Stat/MDR predicts that UWB devices will grow at annual rate of over 400 per cent between 2005 and 2008 because their high speeds (up to 480Mbps) will see off challenges from fast Wi-Fi and will attract OEMs even if standards are only de facto. But the Diffusion Group is in the camp that believes 802.11n will be ‘good enough’ should the industry fail to sort out standards for 802.15.3a soon.
Copyright © 2004, 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.