A Yuletide Request from Optics Researchers: All I want for Christmas is…
As the festive season drags us kicking and screaming into rounds of seasonal silliness (and if you’ve never seen a man in a false beard and Santa Suit surrounded by fake snow in 40 degree heat (100F for the renegade non-metrics out there) you haven’t lived) it’s time to hope you’ve been a good person and send that letter to the North Pole and hope that the man in red can make it through the air conditioning vents and deliver something for all those of us who work at the optical layer. I’m therefore making three Christmas wishes for developments from our researchers worldwide.
Better fibres, please.
Silica is a wonderful material and we all love it dearly for its incredible transparency, strength and (now we know how to do it) ease of processing. However, it still isn’t ideal. It does have some loss and, possibly even worse, has a level of non-linearity. The combination of the two means that we still don’t have the perfect transmission medium. So my first Christmas wish is for a fibre which has zero loss and no non-linearity. Maybe some of the hollow-core Photonic Crystal Fibre designs that have been proposed will get us there in the end, but I haven’t seen any signs of real progress towards practical, manufacturable designs which could be useful in a real-world transmission system. Since we will therefore always have some form of loss, this leads to my second wish…
More Gain.
Since we don’t have zero loss fibres, we need optical amplifiers. As an industry, we have been incredibly fortunate that the EDFA gain region and the minimum loss window of silica coincide so well. This has supported the development of DWDM systems in all their shapes and sizes over the past 25 years. However, as we keep pushing system capacity, we’re running out of gain in the EDFA (even allowing for L-band) and having to drag Raman amplifiers kicking and screaming into the mainstream of technology. However, Raman amplification has its own issues (particularly noise figure) and what we’d really like is an amplifier with 0dB noise figure, 30+ dB of gain, saturated output powers of >+20dBm and operating over a 200nm bandwidth. Phase Sensitive Amplifiers show some promise here, although the technology is still very immature and is unproven at a network level.
Polarisation be gone.
With the imminent introduction of coherent transmission systems, Polarisation Mode Dispersion (PMD) will no longer be the bugbear it was for high-speed 40G systems. However, our old friend PDL (Polarisation Dependent Loss) now becomes a limiting factor, so my final wish is for optical components with no PDL (or Polarisation Dependent Gain – PDG).
I am sure there are other things we could wish for– let me know your thoughts and suggestions.
In the meantime, have a cool yule and a restive festive.
Industry Outlook, Optics, Out of His Depth | Simon Poole | 
10:20 am
I’m intersted in your mention of the trade-offs between...
By Andrew Schmitt, December 15, 2010 @ 11:36 am
While you are at it, how about CMOS transistors with no gate leakage.
How about a multi-sourced tunable XFP?
By Gene, December 17, 2010 @ 8:55 pm
I think all these amenities turn into realities, only if electronics is much expensive than optics to meet technical requirements. I hope that day will come…soon.
By Simon Poole, December 19, 2010 @ 2:33 pm
Hi Andrew
I *know* there ain’t no thing as zero-gate-leakage-current CMOS but for the multi-sourced Tunable XFP? Well, maybe some wishes do come true…
Merry Christmas
Simon
By Simon Poole, December 20, 2010 @ 4:13 pm
Hi Gene
I think some of what I want is pushing the friendship with the laws of physics but I do think that electronics wil play an ever-increasing role in our photonic world. As a ’30 years at the mast, man-and-boy’ optics guy (albeit with an electronics background) I use to think that photonics would conquer all, but in my grizzled old age I’m becoming increasingly of the belief that it’s the electronics which will be our saviour as we continue to try and head down the cost per bit per km curve. The introduction of Forward Error coding about 10 years agowas the first step here whilst the advent of coherent systems means that the electronics cost in a transceiver may equal (or maybe exceeed) that of the optics… Merry Christmas
Simon