The starry-eyed gave way to the hard-nosed at a recent conference in Chicago for sellers and buyers of wireless sensor networks.
Attendees directed tougher and more pointed questions at suppliers than at the debut Wireless Sensing Solutions Conference a year ago. Faced with extravagant promises about the benefits of temperature, vibration and other sensors married to low-power wireless mesh networks, attendees wanted to know about security, management, interference, reliability and total cost of ownership.
"We've been through the hype and buzz phase," said Jeff Bussgang, general partner with IDG Ventures, a venture capital firm in which Network World's parent company is a limited partner. "Then new technologies go through the reality phase. People are discovering these technologies are more complicated than anticipated."
Three engineers with Caterpillar's Systems and Controls Research group evaluated wireless mesh sensing at the show, which attracted about 400 attendees and about two dozen exhibitors, both increases from a year ago. Caterpillar would seem to be a perfect candidate for such technology, from asset management of heavy equipment in dealers' yards to monitoring earth-moving and mining machines.
But what the engineers found at the conference sparked questions and concerns, ranging from the viability of IEEE 802.14.5-based radios on big machines to possible interference when different wireless technologies are active in the same area.
"There's not real collaboration between the various standards bodies," said Tien Doan, one of the trio. "So you may have radio interference, and that could affect the reliability of the network."
Attendees were unimpressed, and impatient, with grandiose application possibilities.
Srini Krishnamurthy, vice president of business development for Airbee Wireless, outlined a wireless sensor mesh project under study for Ronald Reagan Airport in Washington DC. Every door in the airport could be outfitted with 900MHz wireless sensors and automated locks, networked to a central point where rules could be set for when a door could be opened, by whom, by time of day, without the need for guards.
This idea sparked a pragmatic response from Samuel Reed, an electrical engineer with Key Technologies: "Can a terrorist walk into the airport with a 900MHz jammer and shut the whole place down?"
Another attendee wanted to know how many products had shipped with ZigBee Alliance's mesh networking software stack.
The ZigBee Alliance, a group of start-ups, integrators and other manufacturers, has crafted higher-level specifications for security and networking, as well as a set of application-level profiles for use by programmers.
"There are some prototype projects with 100 to 150 nodes," said William Craig, program manager for wireless communications at ZMD America. "But you can't point to a ZigBee thermostat today." He said ZigBee-based products will start to hit the market late this year.
Over-filling the spectrum
Another questioner wanted to know whether large-scale wireless mesh networks could overwhelm the available radio frequency spectrum. None of the speakers answered with an unqualified No. They did point out that the ZigBee specification and the underlying IEEE 802.15.4 radio standard had various features to avoid collisions and find open channels; and that sensors could be set to transmit only occasionally and send small amounts of data when they did.
Faced with repeated questions about how easy such networks would be to deploy and run compared with traditional wired networks, vendors sounded almost plaintive.
"Replacing wires with wireless is a good thing," said Mark Goodman, director of sales for Crossbow Technologies. "But think about what you can do with wireless that you can't do with wires."
Listeners weren't convinced.
"The first question is 'how do these wireless sensor nets compare to wired systems?' '' said Michael Schell, president of EpiphanyTec, a California technology marketing consultancy that is working with several mesh companies targeting homeland security applications. "They look at the total costs and reliability of wireless systems compared to wired systems. If the industry doesn't address this question, it won't develop."
"Mesh offers the chance to save an order of magnitude or more in such costs," insisted Joy Weiss, president of Dust Networks, one of the pioneers in this market. She cited Teng & Associates, a design and construction company that installed a Dust sensor network in two hours at a site that Teng engineers estimated would take four weeks to set up using wires.
Another obstacle to the developing industry is the internal disagreement over the still-emerging specifications for wireless sensing networks. The IEEE 802.15.4 standard defines the basic direct sequencing spread spectrum radio for a low-power wireless mesh: the media access control layer and the physical layer.
Key Technologies' Reed said he was impressed by the capabilities outlined for ZigBee networks. "But it's the Cadillac," he said. "For something simple, there are lots of other alternatives that are cheaper and simpler."
"If you're not doing a network, there are lots of alternatives out there," said Airbee's Krishnamurthy. But "if you're doing a network, then ZigBee is key."
In several conference sessions, however, even vendors made it clear they were not wedded to ZigBee or 802.15.4. "The most important thing is to come up with applications that work, to validate the wireless sensing technology," said Goodman of Crossbow, which offers the Smart Dust wireless sensors. "Sometimes that's using ZigBee and 802.14.5, and sometimes it's not."
"Today, most protocol stacks are proprietary software over 802.15.4 radios," said Bernd Grohmann, vice president of alliances for Zensys, which has launched its own wireless mesh protocol through the Z-Wave Alliance. The company said more than 100 manufacturers, including Honeywell and Leviton, are incorporating Z-Wave into various products for home and building control applications.
Caterpillar's Doan was surprised at the level of disagreement. At a session about standards featuring four panellists, he said, "Two were for the standards and two were somewhat against them. Standards are definitely the way" to improve productivity and cut costs.
Questions to ask wireless sensor network vendors:
• How complex is deployment vs. that of conventional wired networks?
• How stable are standards like Zigbee?
• Why go with standard-based approaches vs. possibly more flexible proprietary mesh networking protocols?
• Will radio interference be a factor with multiple sensor nets with hundreds or even thousands of nodes?
• Can a deliberate jamming attempt shut down the entire net?
• What tools are available to manage these nets, and to treat them as part of an enterprise IP net?
• How can data from sensor nets be integrated with existing enterprise applications?
• How realistic are battery life projections of months or years?
• What are the total life-cycle costs of sensors nets, including battery replacement?