[190566] in North American Network Operators' Group
Re: Interesting Article on Modulation Schemes
daemon@ATHENA.MIT.EDU (Rod Beck)
Fri Jul 8 18:38:11 2016
X-Original-To: nanog@nanog.org
From: Rod Beck <rod.beck@unitedcablecompany.com>
To: Eric Kuhnke <eric.kuhnke@gmail.com>
Date: Fri, 8 Jul 2016 22:15:15 +0000
In-Reply-To: <CAB69EHh-zL4OmMTE5GBVTeunCkbA+McXwa_sZ6FwT0vh+PFEHw@mail.gmail.com>
Cc: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
I don't see wide spread deployment. The most recently built TransAtlantic c=
able is Aquacomms It is QPSK and 130 100 gig waves per pair. Does one reall=
y need more? C-Lion, the new Finland/Germany cable is 18 terabits per fiber=
pair. I think that is 8 QAM. Is that a representative sample? I don't kno=
w. It is certainly a small sample and hence could be highly contained by ra=
ndom error. Since so many ISPs dropped their Layer 2 networks in favor of b=
uying cheap transit, the market for 100 gig waves is limited to Tier 1 ISPs=
, a few huge hosting companies, and the public Web giants in shopping, soci=
al media. I have been told that the video streaming guys like Netflix are m=
ore similar to Akamai than Telia. Dense local footprints.
Bottom line. I don't think the demand is sufficient or the interface costs =
on the customer side sufficiently low. Could be wrong about both.
Regards,
Roderick.
________________________________
From: Eric Kuhnke <eric.kuhnke@gmail.com>
Sent: Friday, July 8, 2016 11:45 PM
To: Rod Beck
Cc: nanog@nanog.org
Subject: Re: Interesting Article on Modulation Schemes
Not just "talking about" 16QAM is in active use for subsea high capacity ch=
annels... Both Xtera and Infinera are shipping DWDM terminals for installa=
tion at cable landing stations that use 16QAM for 100/200/400 Gbps supercha=
nnels.
http://www.xtera.com/home/technology/100g-and-400g/
100G and 400G Coherent | Xtera<http://www.xtera.com/home/technology/100g-an=
d-400g/>
www.xtera.com
Xtera's coherent technology support 100G, 400G and beyond optical channel r=
ates for high-capacity backbone networks.
http://www.xtera.com/home/products/nu-wave-optima/
Unless I'm grossly mistaken, Alcatel-Lucent and Huawei as well.
On Fri, Jul 8, 2016 at 2:24 PM, Rod Beck <rod.beck@unitedcablecompany.com<m=
ailto:rod.beck@unitedcablecompany.com>> wrote:
Apparently 40 gigs is the limit of simple laser flash equals 1, no flash eq=
uals 0. Above that threshold the signal becomes larger than an ITU 50 gigah=
ertz channel. Most new undersea cables are using QPSK or 8 QAM and talking =
about 16 QAM.
This companion piece explains it: http://digital.lightwaveonline.com/lightw=
ave/20130708/?pm=3D1&u1=3Dfriend&pg=3D19#pg19.
- Roderick.
________________________________
From: Eric Kuhnke <eric.kuhnke@gmail.com<mailto:eric.kuhnke@gmail.com>>
Sent: Friday, July 8, 2016 10:40 PM
To: Rod Beck
Cc: nanog@nanog.org<mailto:nanog@nanog.org>
Subject: Re: Interesting Article on Modulation Schemes
Essentially the transceiver optics are applying the same modulation and cod=
ing that have been used in point-to-point microwave for a long time... St=
arting from OOK, up to BPSK and then on to QPSK, 16QAM and possibly 64QAM w=
ith varying levels of FEC.
A singlemode fiber is just an extremely narrow diameter waveguide. Big diff=
erence in frequency between a 71-86 GHz FDD radio pair and optical at 191 t=
o 196 THz.
On Fri, Jul 8, 2016 at 1:50 AM, Rod Beck <rod.beck@unitedcablecompany.com<m=
ailto:rod.beck@unitedcablecompany.com>> wrote:
The new undersea cable systems are now capable of 18 terabits per fiber pai=
r. It is interesting how combinations of bits are being represented by comb=
inations of optical features.
http://www.lightwaveonline.com/articles/print/volume-30/issue-5/features/wh=
ich-optical-modulation-scheme-best-fits-my-application.html
Roderick Beck
Director of Global Sales
United Cable Company
www.unitedcablecompany.com<http://www.unitedcablecompany.com><http://www.un=
itedcablecompany.com>
