FEATURE FIBRE OPTIC NETWORKS
Overcoming Distortions in Multimode Fibre Smooth Sailing
NETCOMMS.EUROPE: Can you tell us about the nature of your recent investigations?
At Comms Expo this
May, Panduit exclusively announced the discovery they have made in the field of multimode fibre. We spoke to Panduit’s Dr Rick Pimpinella to find out more.
DR. RICK PIMPINELLA: In 2004, we began studying channel performance in multimode fibre, and trying to correlate actual measured performance with fibre bandwidth. We found that the correlation was extremely poor. In many cases, there was no correlation at all, which is not what you would expect. This initiated a six year study where we tried to identify the root cause of the lack of correlation between bandwidth and performance. This led us to make some extremely interesting discoveries. Two distortions govern multimode
fibre performance: modal dispersion and chromatic dispersion. Up until now people assumed these two signal distortions were independent. But we have discovered that in new high speed systems you can no longer consider them to be independent. They are in fact very much connected. Based on this new discovery we can now design links that have much higher performance than multimode fibre has ever had before.
NE: Can you explain the different types of distortion that affect multimode fibre?
RP: In multimode fibre, light takes multiple paths (modes) through the fibre. Modal dispersion is the spreading out of light (signal) which occurs due to the different paths taken when traversing the fibre. Chromatic dispersion is the
Neil ‘Sullivan spoke to Dr Rick Pimpinella, Chief Engineer, Panduit
Actual VCSEL Optical Coupling Into Multimode Fiber
Typical Transmi-er Op0cal Sub‐Assembly
λ0 λ2
Multimode Fibre VCSEL (TO-packaged) lens ferrule
plastic or metal sub-assembly
Short wavelengths couple to high-order modes Long wavelengths couple to low-order modes
λ0 > λ1 > λ2 SM Actual VCSEL Optical Coupling into Multimode Fibre Panduit Confidential and Proprietary
spreading of the pulse due to difference in colours. Every transmitter has a spectrum and the red colours travel at different speeds to blue. It is the same effect as when you shine light through a prism. Up until now, experts assumed that the chromatic dispersion effect, which is the dispersion due to differences in colour, was negligible. It was not really considered when calculating fibre bandwidth. We found out that this assumption was incorrect when it comes to higher speeds.
NE: What new discoveries have you made?
RP: We discovered that the VCSELs (vertical-cavity surface-emitting lasers, the type of laser used in multimode
channelling) don’t emit light in the way we had previously assumed. It is well understood that the VCSEL emits a spectrum of light but what we didn’t recognise was that different wavelengths are emitted at different angles. Therefore, when you couple the light into a multimode fibre, there is a radial dependence of the wavelengths. This means that each of the modes actually carries a different set of wavelengths that was previously neglected. Because of this, the chromatic dispersion effect has a very strong influence on the modal dispersion, which was the only thing that people really calculated for bandwidth. When you include the chromatic dispersion effect, you can now correlate the actual performance with the calculated bandwidth.
NE: How have you used this knowledge?
RP: One consequence of not taking this phenomenon into account is that there is a 20% failure probability that fibres won’t support the link length that is specified in the 10G Ethernet standard (10GBASE-SR). But by designing the fibre to have
Modal Dispersion Differential Mode Delay (DMD) 12 NETCOMMS europe Volume I, Issue 6 2011
a certain characteristic, we can compensate for this optical property, the chromatic dispersion effect, and we can increase the total bandwidth significantly and design the links to perform at very long distances. We have actually demonstrated that we are able to get a 25G VCSEL for Next
www.netcommseurope.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52