More than a flash in the PAN

With Bluetooth, ZigBee, ultrawideband and near-field communications already existing in the short-range sector, Nokia's decision in October to introduce a new technology dubbed Wibree raised more than a few eyebrows in the industry.

Nokia officials were adamant that Wibree is not designed to compete with existing technologies, but initial descriptions of the proposed standard raised skepticism. Featuring robust throughput (1 Mb/s) and a range (10 meters) that rivals that of existing Bluetooth devices and better battery life than low-power alternative ZigBee, Wibree appeared to be a threat to both standards. Furthermore, many have questioned the market opportunity for Nokia's newest technology should Wibree not be positioned as a challenger to these standards.

But such skepticism is beginning to wane, as Nokia better identifies the niche it sees for Wibree in the personal area network (PAN) market. In a recent research brief, ABI Research principal analyst Stuart Carlaw forecasted a $513 million market for Wibree products by 2011.

Such a predicted market opportunity for Wibree does not portend the demise of other short-range technologies, Carlaw said. In fact, he said, the success of Bluetooth is critical to Wibree because both technologies utilize the 2.4 GHz band and use the same over-the-air coding, allowing Wibree to be integrated rather easily and cost-effectively into Bluetooth chips.

“There has been a lot of noise about [Wibree] being a Bluetooth-killer, but I think that's complete nonsense,” Carlaw said. “I think it's more of a symbiotic relationship, in that you've got Wibree sitting on top of Bluetooth and extending Bluetooth into applications that have been targeted by Bluetooth for a number of years but [have been unreachable] because of cost, power and size.”

Indeed, Bluetooth is impractical for communicating with some small devices because its fixed-packet-length protocol requires a relatively large amount of power, said Jani Tierala, Nokia Research Center business development manager. Wibree, which utilizes dynamic packet lengths, is an ideal solution in situations where bursty data transfers are needed, while Bluetooth is better equipped to handle larger data loads, he said.

“As the data amount you want to transfer grows, then the Wibree packets would reach their maximum and you don't get much power-consumption savings,” Tierala said. “All of the use cases we have targeted with Wibree are when you are infrequently sending small amounts of data.”

Wibree's power-efficient feature is expected to enable personal-area communications in devices such as watches, wireless keyboards, toys and sports sensors that have limited battery capacity. More important, this communication can occur even when Bluetooth is delivering data-intensive signals such as audio or video, said Jamey Hicks, director of the Nokia Research Center.

“Bluetooth is sending individual packets, and it's channel hopping,” Hicks said. “Wibree has been worked out so that, when you're not sending a Bluetooth packet, you can be sending or receiving a Wibree packet.”

As a secondary channel, Wibree could be used to communicate parameters such as encryption keys for a Bluetooth signal to make it more secure, Hicks said. Wibree also can provide a PAN link to transfer information from one device — a mobile phone or a sensor — to an alternative display area, such as a wristwatch.

In addition, Wibree's power-efficient characteristics and ability to locate devices quicker than Bluetooth could provide the foundation for another complementary application — alerting, or “waking up,” a Bluetooth device when it's time to become active, without draining valuable battery capacity.

“You could use Wibree mode to keep the connection alive and then, when you need to actually transfer more data … you could use Bluetooth or wireless LAN to transfer a lot of data,” Hicks said.

Such applications are particularly compelling because Bluetooth and Wibree can be implemented on the same silicon cost-effectively — about 20¢ per chip in incremental costs compared to a Bluetooth-only chip, Carlaw said. The notion appears to be compelling to chipmakers, several of which already have announced their membership in the newly formed Wibree Forum.

“The thing that makes Wibree and Bluetooth coexist so well is that they actually use the same coding in the air at the link level,” Carlaw said. “It's just a very simple modulation scheme in Wibree and Bluetooth, and … [they] only require a little bit of software to create a dual-mode chip.”

Wibree does not share such synergies with ZigBee, making integration impractical with that low-power technology, which is gaining traction in the building-controls and industrial-automation sectors. With Nokia touting Wibree as being more power-efficient than ZigBee and providing about four times the data throughput, Wibree would seem to be a significant threat to ZigBee.

In reality, that's probably not the case because ZigBee provides greater range and includes mesh-routing capabilities that would work well in a warehouse or manufacturing environment, Carlaw said. Although it may be possible to introduce a meshing component to Wibree, doing so likely would hamper the technology's power efficiency, he said.

“The way [Wibree] works at the moment is that it's got very long duty cycles. … If you start waking the device up and leaving it on to keep it integrated in a mesh, it's going to kill the battery life,” Carlaw said. “What makes Wibree power efficient is the fact that it's not communicating for long periods of time — it's waking up, firing off a message and going back to sleep again.”

Hicks agreed that Wibree generally will not be a threat to ZigBee, but both he and Carlaw noted the medical arena as one potential area of competition. In particular, Hicks said, Wibree could be a better low-power communications link in a facility — such as a hospital — that has plenty of backhaul capacity, which would make multi-hop mesh capability unnecessary.

Although ABI Research projects 800 million Wibree-enabled devices in use by 2011, Carlaw believes this only will occur if other mobile handset makers embrace the technology. To make this a reality, the Bluetooth Special Interest Group (SIG) must adopt Wibree, he said.

“It's dependent on adoption by the Bluetooth SIG — I don't think it will happen otherwise,” Carlaw said. “It's not a royalty issue. The issue comes in the fact that, in the highly competitive handset market, no one's going to want to line Nokia's pockets and give them the advantage of what is effectively a proprietary solution in the handsets.”

Nokia's Hicks disagreed.

“I think it would help if the Bluetooth SIG adopted it, but I actually think they will wait to see how it does in the marketplace,” he said. “I don't think [adoption is] a prerequisite [to success].”

Noting the interest from key chipmakers and the low incremental cost associated with deploying Wibree, Nokia is confident the new technology will gain market traction relatively quickly, Hicks said.

“Because it's compatible with Bluetooth, and it essentially costs nothing to add it to mobile phones or laptops that have Bluetooth support, we can get deployment very rapidly,” he said. “It should be a lot quicker than with other kinds of [technologies], where you need an add-on module for your phone and laptop.”