Aug. 25, 2003 – Like many engineering-focused companies, Applied Wireless Identification Group (AWID) hasn’t spent a lot of money on a glitzy Web site or a slick marketing campaign. Instead, the company, located in Monsey, N.Y., puts all of its resources into developing new products. That strategy may soon pay off. AWID is designing a
multi-protocol UHF reader in the form of a PCMCIA card that can be plugged into any handheld computer or bar code scanner with the appropriate slot. And it’s planning to shrink its fixed reader into a set of microchips, which would cut costs dramatically.
AWID was founded by Donny Lee, an engineer who worked at a number of high-tech companies, including Fairchild and General Instrument. In the mid-1990s, Lee was VP of the electronics system division for Aeroflex, a military contractor. He was working on advanced radar systems in 1996, when IBM asked Aeroflex to design an RFID reader. Lee spent seven months developing a reader that operated at 2.45 GHz. That sparked his interest in the potential of RFID technology.
In 1998, Lee left Aeroflex to launch AWID with two partners. Eventually, he bought out his partners and got some funding from silent partners in Taiwan, where he grew up. He developed a line of RFID readers for access control, which currently accounts for 80 percent of AWID’s revenue. The company makes low-frequency (125 KHz) systems and one gate-control reader that operates in the UHF band (920 to 928 MHz) for security companies, including DoorKing, Identicard, Lenel Systems International and PCSC.
Lee says AWID’s access control business is doubling every year, but the overall market is not growing rapidly. So the company must win business away from HID Corp., which dominates the access control industry. But he sees more long-term opportunity in providing RFID readers for use in the supply chain. “There’s explosive growth potential in the asset management market,” Lee says. “About 90 percent of our engineering resources are now being put into developing new products for this market.”
Earlier this year, Lee hired Jeffery Jacobsen, former CEO of Alien Technology, as president. Jacobsen raised tens of millions of dollars for Alien and helped to position the Morgan Hill, Calif., startup to take advantage of the adoption of EPC technology. Lee is hoping that Jacobsen can do the same for AWID.
AWID’s strategy is to develop multi-protocol readers, so companies don’t have to buy one reader for one application and another for a different one. Right now, AWID has a competitive advantage in the handheld market, because it has developed a 13.56 MHz reader in a PCMCIA card and is working on a UHF version, which will be on the market next year.
The PCMCIA readers are designed to plug into handheld bar code scanners. Companies can buy the cards, which cost between $500 and $600, and plug them into their existing bar code scanners. Then they can read either the bar code or the RFID tag without having to spend tens of thousands of dollars on new handheld RFID readers.
Most older bar code scanners use the PCMCIA card slot to provide wireless connectivity, so companies that swap out the wireless network card for the RFID reader will have to download RFID data in batch mode. Newer handheld scanners have 802.11 capabilities built in, so companies can get a wireless bar code and RFID reader all in the same package.The PCMCIA card will also work with most PDAs, which can be used in stores. Some companies, including Tesco in the United Kingdom, have experimented with smart-shelf technology. AWID has designed readers to be installed in displays for one Auto-ID Center sponsor. But in the near term, Lee thinks smart shelves will be too expensive and a UHF handheld reader will be a more practical solution. “You lose the
real-time visibility of what’s on the shelf and the anti-theft capabilities if you use a handheld instead of a smart shelf,” he says. “But outfitting an entire store with smart shelves isn’t going to be possible for most supermarkets, which operate on a 1- percent margin.”
AWID has also developed a line of fixed readers that operate at 13.56 MHz, 915 MHz and 2.45 GHz. These readers can communicate with a variety of tags using different protocols. The Sentinel-Sense MPR-2010 is a wall-mounted UHF reader that covers the Electronic Product Code (EPC) specifications, ISO-18000 Type A, Type B, Dura-label from SCS and Supertag from BiStar. The MPR-2030 wall-mounted reader covers the major 13.56 MHz tags, including ISO-15693, I-Code, Gemplus, MicroID, Performa and Texas Instrument’s Tag-It.
AWID has taken a different approach to fixed readers than most other companies. Typically, a UHF reader has several channels, so you can connect four or more antennas. This lets you install one reader over a dock door and install four antennas around the door. The antennas are activated in sequence, increasing the chances of reading all of the tags on a pallet stacked with cases of product. But there are some shortcomings to this approach, according to Lee.
The coaxial cable connecting the antenna to the reader cable can fail. Moreover, RF energy is lost as it travels through the cable. If you have to string your cable over 10 feet or more on a dock door, the cable loss can degrade system performance, particularly in Europe where reader output is restricted to 500 milliwatts of power. “To minimize the loss, you have to use half-inch diameter cable with connectors, which is expensive,” says Lee.
Lee’s solution: build a fixed reader with one internal antenna. In the long term, he believes that this will be more reliable, more economical and provide better system performance. “Reader prices will come down to the point where you can put four readers on a dock door and connect them through a network cable, which is much cheaper that coaxial cable,” he says.
This setup also means that if one reader fails, you wouldn’t lose complete visability at that location, as you would if you used one reader with several antennas. The readers could be linked to a local area network wirelessly or through a wired TCP/IP connection, so if one went down, the others would continue to feed data through the network.
Lee also believes that end users should not adopt the portal or “choke point” approach to installing readers that many vendors are advocating. This approach aims to minimize the number of readers by installing them only at key points in the supply chain where product must pass—for instance, a receiving bay in a warehouse or the doorway that separates the storeroom from the retail floor.
He says companies need to create a “spider web” of readers throughout their supply chain. That’s because no one reader will pick up every item every time under real-world conditions. In a web-like setup, a product that isn’t read at one point will be picked up at another. For instance, if a read was missed at the receiving bay but the pallet was scanned when a forklift picked it up, the system would update inventory automatically.
This approach is more expensive than simply putting readers at key choke points, but Lee says it’s the only reasonable approach. “The industry is still misleading the end users,” he says. “When you see a demonstration where they set up a portal and push a pallet through and read all the cases, it’s staged. It’s not going to work like that in the real world.”AWID wants to shrink its existing reader design down to a microchip or set of microchips. It’s well positioned to do that because, unlike many other companies, AWID doesn’t use a delay line or circulator in its readers. These are separate components that enable the reader to divert the weak incoming signal from the tag, while the much more
powerful signal is being transmitted from the reader, so the incoming signal can be read.
When AWID was designing its first readers, it chose not to use a delay line or a circulator because reducing the number of components would help keep the cost of the readers down. Instead, it engineered its readers with transistors and capacitors, basic components in all microchips. If the entire reader can be incorporated in a single chip, or a set of microchips, it would be possible to create readers that sell for a couple of hundred dollars. “RFID isn’t going to be pervasive until the readers cost $100 to $200,” says Lee. “We have the only IC-ready design that makes it possible to get to that level.”
But creating readers in a chip won’t be an inexpensive process. It can cost upward of $1 million to design a new microchip. Lee says AWID is close to closing a new round of funding that would provide the money needed to create a low-cost UHF reader and even a multi-channel 13.56 MHz reader-in-a-chip, which could control several antennas in a smart shelf.
In the longer term, Jacobsen envisions AWID being a leader in developing a new generation of readers that are designed into sheets that can be rolled out on existing shelves. The idea is to transfer CMOS or gallium arsenide circuits, which are typically less than one micron thick, from the surface of silicon wafers to large surface materials like plastics, polymers and metal foils to create high-performance, low-cost flexible antennas.
The goal would be to print RFID transmission and reception circuitry—amplifiers, digital signal processors, filters and so on—right on plastic sheets with embedded antennas using low-cost, high-speed roll-to-roll manufacturing techniques. Some of the new funding in the works will go toward preliminary research into the feasibility of these low-cost smart shelf systems.
AWID currently sells its asset management readers directly to customers. Lee says that most systems integrators and value added resellers don’t have the expertise to support the readers, and distributors simply add to the final cost of the product. That will change over time, but right now he is more interested in developing products for the future than marketing alliances. If the company succeeds in bringing products based on new chipsets to market in the next two or three years, the lower cost could accelerate adoption of RFID technology and turn AWID into a major player in the asset management market.