The Lightwave Blog

Posted by Meghan Fuller Hanna

This week, I've been trying to wrap the ol' cerebral cortex around dual-polarization quadrature phase-shift keying, or DP-QPSK to those of us in the know. (Since it took me most of the week just to memorize the acronym, I probably shouldn't count myself among that esteemed company.)

At any rate, I started my research with the Optical Internetworking Forum, which recently announced that its 100G project would focus on DP-QPSK as the advanced modulation format for 100G long-haul transmission. When I asked Joe Berthold, former president of the OIF and one of the authors of the current 100G project, why the organization has decided to standardize on DP-QPSK, he was quick to note that this is not a standard, per se.

Instead, the OIF's goal is to "create a critical mass of vendors in the market to buy technology components of a particular type so that we can create an ecosystem," he explained. "We are not looking at standardizing the be-all-and-end-all. We're looking at selecting a particular implementation that looks attractive enough that we think a lot of us are going to follow [it], and we think it makes a lot of sense for component companies to invest in the various piece-parts for it."

In other words, the OIF is aiming to 1) mitigate the risk inherent in component R&D;, and 2) drive down the cost of the resultant components by creating a critical mass of buyers.

On the one hand, I applaud the OIF for trying to ease the financial burden on the component vendors by creating an ecosystem of buyers. In an article I wrote summarizing the mood at this year's OFC Conference, "No relief in sight for optical components sector," I noted that the component sector was still too fragmented, gross margins remained tight at 20% to 30%, and no one seemed to know which part of the telecom food chain should bear the brunt of the R&D; burden going forward. In that environment, what impetus does a component vendor have to develop the high-speed electronics and integrated photonic components that DP-QPSK will require? These companies poured hundreds of millions of R&D; dollars into 40G, and . . . well, we all know what happened there.

But here's the other side of this particular conundrum: While Berthold says the OIF was perfectly willing to "back another horse" should someone come up with a viable alternative to DP-QPSK, some wonder whether it's simply too early for the OIF to be backing any horse in this particular race.

DP-QPSK still presents formidable challenges. For real 100G transmission, digital signal processing should be at least two times faster than for 40G, and the industry just isn't there yet. And it will take far more photonic integration than is currently available. You can build a system out of discrete components, but it's cost-prohibitive. Speaking of which, several people have told me that it's still at least--AT LEAST--six times more expensive than currently available 40G options, which, frankly, isn't going to fly with any of the carriers I know.

To be fair, no one I interviewed for my story disputed the fact that DP-QPSK holds a lot of promise, but some wonder whether we can conclusively say it holds the most promise at this point.

Because here's another interesting thing about DP-QPSK: it's relatively immature. As one industry insider told me, there is usually a three- to five-year lag before research concepts are commercialized, but "this is like lab-to-the-field almost immediately." There are still a lot of folks in research labs and university settings working on variations of APSK (amplitude and phase shift keying), for example, and there seems to be a recent groundswell of interest in OFDM (optical frequency domain multiplexing).

So what do you think? Is now the right time for the industry to rally around DP-QPSK, or has the OIF jumped the gun?

(FYI: My article--tentatively titled "Is DP-QPSK the end-game for 100 Gbits/sec?"-- is slated to appear in Lightwave's November issue.)