Optical CDMA Product Stirs Debate

BALTIMORE – NFOEC – Yesterday, APN Inc., a small startup based in Quebec, became the latest in a long line of companies to announce a product based on optical code-division multiple access (CDMA) technology, a coding scheme originally used in cellular telephones.

Big deal? Yes, if it works. But the history of optical CDMA is filled with failed experiments and is looked upon skeptically by many experts in the industry.

APN hopes to change that. The company officially came out of stealth mode yesterday and announced its first product, the APN-1008, at the NFOEC show in Baltimore (see APN Launches). The APN-1008, which the company says is available for customer trials, uses optical CDMA to increase network capacity up to 20 times, at a third of the cost of adding new Sonet equipment or upgrading to higher-capacity wavelength-division multiplexing (WDM) systems, says Rick Whittaker, vice president of business development for APN.

Traditionally, CDMA has been used in cellular phone networks to add more capacity to a wireless network by giving each user a specific code in a section of the partitioned service area. For at least a decade, researchers have been working to adapt the technology for use in fiber-based optical networks to pack more capacity over wavelengths. So far, most of the activity surrounding OCDMA has been in research.

“I haven’t seen anyone doing CDMA over fiber,” says Greg Mumford, president of the optical Internet division at Nortel Networks Corp. (NYSE/Toronto: NT). “Why? It’s really hard to do. So if someone gets it to work, it’s a big deal.” Krishna Bala, co-founder and chief technology officer for Tellium Inc. (Nasdaq: TELM), an optical switch company, says he remembers working on OCDMA in the early 1990s, when he was a senior scientist for the optical networking group of Bellcore, one of the first research organizations to work on it.

“It’s very difficult technology,” he says. “I can’t imagine anyone would have a commercial product even close to being ready. I haven’t really looked into it for a couple of years, but I was never bullish on it.”

Over the years, other companies have also worked to develop a commercial OCDMA product. So far there has been little success. CodeStream Technologies Corp., the only subsidiary of holding company CodeStream Holdings Inc., announced its OCDMA product at the Supercomm trade show in 1998. Three years later, the company filed for Chapter 7 bankruptcy protection. When it officially closed its doors in June of this year, it was still at least 10 months away from having customer trials (see CodeStream Goes Under). Templex Technology Inc., a components firm funded by Intel Corp. (Nasdaq: INTC), is also working on an OCDMA tool, but the company is actually more focused on providing fiber Bragg gratings for WDM systems and hasn’t shipped an OCDMA product yet.

What makes APN different from the rest? Whittaker says the company’s engineers are taking a completely different approach to OCDMA than any of these other companies. For one, CodeStream built and marketed its product as an alternative to WDM technology. APN says its technology is complementary to WDM and Sonet. In fact, Whittaker sees APN’s target market as incumbent providers looking to dramatically increase the capacity of their current WDM or Sonet networks. The key here is that APN’s OCDMA implementation can be applied to one channel at a time, regardless of whether or not WDM is being used. It also can be used with Sonet technology, he claims, adhering to all performance standards.

APN also uses less expensive passive optical components like broadband signaling LEDs for input; and fiber Bragg gratings are used to encode the signal-using frequency and time parameters. This is in contrast to other systems, which use expensive electrical components and lasers, along with sophisticated encoding and decoding components.

The company, which has 32 employees (including 22 engineers, six with PhDs), officially launched in January of 2001. But Habib Fathallah, the founder and CEO, had been working on the technology for several years as a senior research engineer at Laval University in Canada. To date, the company has raised less than $10 million in a single seed round of funding from Innovatech, a Canadian venture capital firm. The current cash should take the company through to its first commercial shipment, set for April of 2002, says Whittaker.

Whittaker, himself, used to be a director of venture investments at Nortel, where he discovered APN. Nortel had supposedly been looking into the company late last year and Whittaker ended up leaving Nortel and joining the startup.

“I saw what Habib was working on and was very impressed,” he says.

He will not say how far Nortel got with APN, or why they didn't make an investment. But he claims the companies are still having ongoing discussions.

Even if APN is able to get the technology to work, some people in the industry still think that it may have trouble finding a home in carrier networks.

“Increasing bandwidth is not the hottest topic right now,” says Tellium’s Bala. “Transport has become a commodity. It’s more about managing bandwidth. Service providers installing metro DWDM gear have enough capacity for awhile. Why would they need to increase the capacity?”

- Marguerite Reardon, Senior Editor, Light Reading

For more information on NFOEC, please visit the Light Reading NFOEC Site.

Be the first to post a comment regarding this story.
Sign In