Ever since the first "wired" BAS was installed, it seems that manufacturers have been looking for ways to make them wireless. Wireless solutions would definitely make installation easier, especially in retrofit situations in which it is difficult, if not impossible, to run wire to all parts of a building.

Having no wires to deal with would also result in less labor costs and more flexibility, since wireless sensors could be placed in a variety of previously inaccessible locations. This would result in better climate control and improved IAQ for building occupants, as well as lower energy costs for building owners.

While wireless technology has been a reality in small applications (such as home networks) for several years, it has usually been deemed as too unreliable or too expensive for use in commercial building applications. In addition, connectivity issues and battery-hungry wireless devices have not made wireless solutions all that attractive.

But a new wave of wireless products promises better results for the building automation market. Some products are available now, and others are expected to be on the market by mid-2004.

What's Available Now

For the last two years, Kele (Memphis) has offered a wireless solution that can be used with a variety of BAS. The Kele "Frontier Wireless I/O System" is a line of transmitters, receivers, repeaters, and survey tools that allows users to bypass traditional, hard-to-wire applications, such as asbestos ceilings, marble walls, and parking lot light retrofits.

Eric Karl, product manager for Kele, said, "Our wireless system can work with anybody's automation system because it uses the signal levels typically found in automation and control systems."

The Frontier system takes a standard DDC/IO and transmits that to a receiver. The receiver can pick up a 4-20 mA input, voltage input, and digital input, as well as produce digital and analog outputs.

The system is an alternative for applications where it is difficult to run the last bit of wire in a system, said Guy Zebrick, regional account manager, Kele. "Often, it's easy enough to drop a pair of wires down the wall, but if you're trying to run a pair of wires across the street to pick up a portable classroom at a school, even though it's only 200 feet away, it's a very expensive and labor-intensive 200 feet. You've got to trench through the parking lot or run wires overhead. Our product operates at a minimum 200 to 300 feet level indoors, or 2,700 feet open air. Repeaters extend the range even further."

Kele's wireless system uses a mix of battery and nonbattery powered devices. The batteries can last over five years when the transmission rate is every five minutes. With a transmission rate of every minute, batteries usually last more than two years.

Interference issues aren't a problem, said Zebrick, because the Frontier uses a technology called direct sequence spread spectrum (DSSS). DSSS is a form of radio transmission in which a digital data signal is spread over many frequencies all at once and then reassembled at the receiver. The receiver hears only the properly encoded digital signals for which it was programmed to listen, and it interprets the signal as a digital noise burst. Since the transmitted signal appears as nothing more than noise, it will not interfere with other systems, such as mobile communications and wireless LANs. The inherent digital encoding ensures secure data transmission within the wireless network.

"The bottom line is the world is going wireless," said Zebrick. "Customers are looking to do a tremendous amount of I/O in the building without doing it with wires."

Kele's next wireless release will be a LON gateway that will connect directly to the BAS front end.

What's Coming Next

The next wave of wireless technology is creating quite a buzz in the building automation industry. Several companies are working to develop what they call wireless mesh networks, which won't be available until mid-2004. The manufacturers say these networks overcome some of the barriers that have previously been encountered in the wireless world, such as connectivity and reliability issues.

Rob Conant, vice president of business development and cofounder of Dust, Inc. (Berkeley, CA), noted that it has been difficult to install wireless systems, because the RF signal strength is almost impossible to predict. "You end up with dead spots, so in some places you can communicate hundreds of meters and in others you can't even communicate a few feet. It's really difficult to predict, and it depends a lot on the environment."

If an environment changes frequently, there are more problems. Simple things like a person walking into a room or the moving of a file cabinet can knock some sensors out of communication. "We quickly learned that in order to make a robust system, each device had to be able to talk not only to the central gateway but to every other device in the network as well," said Conant.

In a mesh network, for example, there is a sensor and a hub that the sensor reports back to. There may be a 1% chance that the sensor won't be able to communicate because of interference problems. If that device can talk to two other devices (e.g., peer devices such as other temperature sensors), then there's a one-in-10,000 chance that that sensor's information won't be able to get back to the hub through this multipath networking, in which data is sent to a peer device, which then sends it forward to the hub.

Conant stated that if three or more devices are communicating, it's to the point where it's possible to drop the devices in place, and the devices will automatically establish communication in the network in a very reliable, predictable way.

"These robust networks are self-configuring networks, so each device can talk to its peers, get the data back to the head end, and that overcomes the problem of installation," said Conant. And because all the peer devices can talk to one another, the reliability of the networks becomes much better.

Smaller Is Better

Tod Riedel, cofounder and vice president of business development, Millennial Net (Cambridge, MA), said some of the benefits to Millennial's self-organizing wireless sensor networks are that the devices are extremely small. "The device itself is about the size of a dime, and it runs on a coin battery. Depending on the sampling - how often you transmit - that coin-sized battery can last upwards of 10 years," noted Riedel.

The technology portion of a Millennial star/mesh network really consists of three components: An extremely small form factor, very long battery life due to ultra-low power consumption, and a robust and secure protocol, so that these devices can self-organize and basically self-heal in the event of interference or disturbance in the network.

Millennial Net manufactures devices called "i-Bean endpoints," which are connected to a sensor, such as a temperature sensor. The i-Bean endpoints require an extremely low amount of power, which helps lower the cost of ownership. "Our customers are able to put these devices into the environment, process, or building and forget about them," said Riedel. "They organize themselves, so there's no need for anyone to administer the network. They simply turn on an i-Bean device, and it automatically finds its way back to the i-Bean gateway. It does this at a distance of between 30 and 100 meters."

Anything over 30 to 100 meters requires an i-Bean router, which "hops" the signal an additional 30 to 100 meters to the i-Bean gateway, which can be the control panel, a PC, or even a handheld device. "The i-Bean routers enable us to extend the range of the network, but they also build in full tolerance, because through the use of the routers, multiple paths back to the gateway have been established. These multiple paths can be used in an ad hoc basis. If a path from an i-Bean endpoint to the gateway is blocked, the i-Bean endpoint will find another path automatically," said Riedel.

Setting Up Standards

Of course all this talk about wireless sensor networks brings up the question of communication protocols. How will the sensors communicate with each other and with the BAS? Some of these issues are still being sorted out.

The "ZigBee" Alliance seems to be the main avenue by which manufacturers are working to resolve the standards issues. The ZigBee Alliance states that it is an association of companies working together to enable reliable, cost-effective, low-power, wirelessly networked monitoring and control products based on an open global standard.

Robert Poor, CTO, chairman and co-founder of Ember Corp. (Boston), said that the ZigBee Alliance is defining application profiles, "So we're talking about a lighting profile, and an HVAC profile, and other things like that. The purpose of the alliance is to make sure that devices that are built on top of IEEE 802.15.4 (the published standard for wireless communication) chips will work with each other for specific applications."

This is necessary in order to make interoperable devices, because building owners want to be able to buy, for example, a thermostat from one company and a wireless-compliant air conditioner from another and have them talk to each other.

Ember is actively involved in the ZigBee Alliance and all its standards efforts. "While ZigBee is still in development, we have software stacks that work now," said Poor. "Our customers are buying our 802.15.4 compliant radios with the ‘EmberNet' software stack on top. Whether or not these customers will convert to the ZigBee profiles when they become available, the market will dictate."

Poor added that in and of themselves, these wireless networks will not be BACnet® or LonWorks® compliant, but there will almost certainly be systems that allow interchange among those systems, perhaps at a gateway level. "You can't translate BACnet directly to a wireless protocol; it's not adapted to that. Neither is LonWorks. I think wireless will make it easier," said Poor.

Riedel stated that Millennial Net has three different ways customers can order products, depending on what they're looking for. First, the company offers i-Bean devices that use radios in the ISM band, which is the license-free band in the 868 MHz, 916 MHz, and 2.4 gigahertz frequency ranges. These are narrow band radios, so for customers who don't need 802.15.4 compliance, this may be the best solution for them.

Second, the company offers devices that support an 802.15.4 radio, so Millennial builds its protocol on top of the physical layer that 802.15.4 provides. Basically, it's the 802.15.4 physical and medium access layers, but it's a Millennial Net proprietary networking layer. Third, the company is involved with the ZigBee Alliance, so they are working toward having a ZigBee-compliant device for customers desiring an open protocol.

Millennial Net, Dust-Inc., and Ember are all negotiating with existing building automation firms and expect to have their wireless products available within the next few months. As Conant noted, "We see the building automation market as a tremendous opportunity for these wireless sensor networks, and we want to provide the technology to a company that understands the markets and specifics of the sensors that need to be integrated into the architecture of the overall system."

It looks like 2004 is shaping up to be the year of the wireless network. ES

Wireless Technology: Why Now?

There are several reasons why wireless technology has "suddenly" become available. The federal government is one reason, while laws of the industry are another.

In the late 1990s, researchers at the Defense Advanced Research Projects Agency (DARPA) proposed the so-called "smart dust" concept. Smart dust is a tiny grain-of-rice size device with microprocessor, sensor, and communication capabilities. The military wanted to be able to drop hundreds of these devices from airplanes onto battlefields, so they could sense when tanks were driving by, how fast they were going, where they were, etc.

As Rob Conant noted, the challenge to distributing these devices is that they have to be able to set up a network with each other whichever way they are dropped. "You can imagine dropping thousands of these out of an airplane onto a battlefield, and they've got to be able to establish communication with each other, they need to be able to figure out which sensors are attached to which devices, and then understand what they have to do with that information. It's really creating a real distributed computation platform," he said.

Basically, the funding from DARPA offered researchers at universities such as U.C. Berkeley and M.I.T. the ability to focus on building reliable robust wireless networks.

Robert Poor, while acknowledging the benefits of DARPA funding, stated that Moore's Law and Metcalfe's Law are the reasons why wireless technology is suddenly becoming available. Moore's Law stated that the number of transistors on a microprocessor doubles approximately every 18 months, while Metcalfe's Law states that the usefulness, or utility, of a network equals the square of the number of users.

"If you consider those in the context of wireless device networks, here's how it plays out," said Poor. "To network devices, you want to use silicon radio technology, rather than linking them using wires, connectors, and labor, because it will continue to get cheaper. Metcalfe's Law says that you really want to make networks of these devices, not point-to-point connections, not wire replacements for the network, because as more devices become available, the more motivation there will be to make them networked."

Either way, it's a good thing that wireless technology is here.