Packet-Optical Technologies
Showing results for 
Search instead for 
Do you mean 

Cutting Through the Noise with Amplified Spans

by Juniper Employee ‎10-31-2015 11:19 AM - edited ‎11-30-2015 03:43 PM

Moving from the comfy land of client optics to the longer-reach world of packet-optical transport can be a daunting transition for the uninitiated. Fear not - a working knowledge of the fundamentals will help guide you towards a viable deployment.



Take it to the limit


When it comes to packet-optical transport, the most fundamental goal is connecting points A and B. In the short reach / access space of client optics (<80km) distances are modest enough that amplification is usually not needed. In this case, your maximum reach is most likely loss-limited.  Once you exceed that limit, the optical attenuation from long fiber runs requires optical amplification. In this situation, your link will most likely be noise-limited. There are other potential impairments (e.g. CD, PMD, etc.) that could limit you max reach, but let's focus on noise.




Optical amplification is most commonly implemented using EDFAs (erbium-doped fiber amplifiers). To increase the overall reach, amplification is added between each length of fiber. Each fiber+amplifier segment is referred to as a 'span', often in the range of 40-100km. Multiple cascaded spans make up a 'link'.


While amplification works well, it has its limits. Amplifiers are completely analog in nature - they amplify the input signal and noise in addition to adding extra noise of their own. As the number of spans increases, so does the accumulated noise.


Optical engineers use a metric called OSNR (Optical Signal to Noise Ratio). This figure of merit can be used to describe the optical quality of a link and also as a performance specification for packet-optical interfaces.


OSNR Cheat Sheet

  • For an optical link, larger OSNR is better: A larger value = a better quality / lower noise link
  • For a transmit interface specification, larger OSNR is better :A larger value = a higher quality signal launched onto the fiber
  • For a receive interface specification, smaller OSNR is better: A smaller value = more noise tolerance


In a noise-limited link, improving your reach can be a multi-dimensional problem. At a high level, there are two general areas where you can try to extend your reach:



Get in line


An optimized line system will enable maximum reach, improve system margin and/or reduce the constraints on your DWDM router interfaces. Dialing in the line system can truly be rocket science, but the concepts are within our grasp: 


Reduce Passive Losses

Reduced attenuation -> improved signal power -> higher OSNR -> greater reach and/or system margin


  • Reduce your total distance
  • Choose lower-loss optical fiber
  • Reduce lossy passive devices from the network (mux/demuxes, splitters, patchpanels, etc.)


Reduce Added Noise

Reduced noise -> higher OSNR -> greater reach and/or system margin


  • Use point-to-point links to eliminate in-line ROADMs
  • Eliminate passive dispersion compensation modules (DCMs)
  • User better quality amplifiers (lower noise figure), e.g. Raman amps vs. EDFAs


Shifting gear


Sometimes you can't touch the line system - it could be insanely expensive or perhaps it's a shared resource that is off-limits to tweaking. If that's the case, your only option may be to focus on the endpoint equipment, i.e. the packet-optical router interfaces.


Higher-performance equipment will allow you to take full advantage of a high-quality line system or will allow you to make do with a marginal one. The following components are key: 

  • Phontonics - The type and quality of electro-optical subcomponents will impact the signal transmission and reception performance
  • FEC - As discussed earlier, stronger FEC translates directly into longer reach. In fact, one figure of merit for FEC is its 'coding gain'. Increasing coding gain by 1dB roughly allows you to tolerate a 1dB reduction in the received OSNR. It's a beautiful thing.
  • Framing and encoding - Use of advanced techniques like non-differential encoding  can complement soft decision FEC to increase the net effective coding gain (NECG).
  • DSP - The strength of the signal processing not only improves reach but also improves tolerance to different impairments that can affect system stability and robustness.

Of course, the world's best submarine-rated ULH transport gear may not be cost-effective for your network. On the contrary, going overboard with the highest-performance standalone transport box can create power, space, and/or operational headaches that could otherwise be avoided with an optimized packet-optical solution.



Putting it all together


Like many engineering problems, optimizing your transport link is an exercise in tradeoffs. By better understanding your specific requirements, line system constraints and equipment options you can make better choices. Maybe you have just a handful of links to address or are planning a big new build-out. Either way, knowing the right questions to ask will help cut through the noise when building out your network.


by Juniper Employee
on ‎01-19-2016 10:11 PM



What are "large OSNR" values? (Typical examples)... I dont understand large. Can you provide numbers?

by Juniper Employee
on ‎01-22-2016 11:16 AM



A fair question. Like most things in life, it's all relative, but let's focus on 100G DP-QPSK:


A large OSNR (lower noise) value is going to correspond to a scenario where there is relareively little noise realteive to the singal. A value 30-50dB would be condered 'high OSNR'.


A small OSNR (higher noise) value is going to correspond to a scenario where there a lot of noise present relative to the signal. A value in the mid to low teens (e.g. >15dB) would be condered 'low OSNR'. 


Hope this helps,


Juniper Networks Technical Books
About the Author
  • I started work at a router company, moved to an optical company adding MPLS, moved to mobility company, moved to a packet optical company who got bought by a router company. Full Circle!!
  • David is a Distinguished Engineer in Juniper's Optical Engineering team, having joined Juniper as part of the acquisition of WANDL Inc in January 2014. David is working on routing and optimisation software for multilayer networks to help plan and design networks using the new generation of packet-optical technology from Juniper. In "previous lives", David worked on soliton propagation; diffractive optic device design and network design software and algorithms in the Optics Research Group in BT's Adastral Park Laboratories in Ipswich. He holds a BA and PhD in Mathematics from the University of Cambridge.
  • Steven Keck is a Distinguished Engineer within Juniper's Optical Engineering team where he is responsible for architecture and implementation of packet-optical solutions for routing and switching platforms. Steve has been designing telecom hardware and optical systems for nearly 20 years. Prior to Juniper, he has held positions at Stanford Telecom, StrataLight Communications and Cisco Systems. He holds a B.S. in Computer Engineering from University of the Pacific.
  • David Song is a Sr Staff Engineer within Juniper's Optical Engineering team where he is responsible for the design of packet-optical solutions for routing and switching platforms. He joined Juniper in 2004 and has been designing networking software in control plane and data plane on various platforms. Prior to Juniper, he held various software development positions at Ciena and Nortel Networks.