DWDM Has Reached the Shannon Limit—Now What?
When the Shannon Limit was set forth in the mid-1900s governing the theoretical limit of information carrying capacity of a communications channel, it was impossible to imagine that this would someday be applied to fiber optics communications that was not even conceived of at that time.
<< Watch our latest interview with Light Reading’s Ray LeMaistre called: The Era of Agile Optics Has Arrived >>
But the reality is that dense wavelength division multiplexing (DWDM) optical transport networks are closing in on the Shannon Limit and the telecom industry must find alternatives to simultaneously increasing both the speed and distance of data transmission, in spite of the mathematical barrier.
The Shannon Limit Explained
The Shannon Limit hinges on the fact that there is inherent noise on any given communications channel. When a signal is sent over the channel, it mixes progressively with this ever-present noise until it arrives at the destination. The problem is that for higher speed transmission using complex encoding schemes, the signal becomes increasingly tangled in the noise until it reaches a point that it can no longer be recovered at the destination. The disentangling of the signal from the noise becomes impossible.
This is the Shannon Limit—the point where we can no longer simultaneously increase speed and distance. Higher signal speeds can only be achieved by reducing link spans, which comes with the penalty of additional signal regeneration costs. (For full disclosure, we can also increase signal power but this quickly reaches thermal limits.)
If you think about the evolution of DSL lines, there came a point where DSLAMs had to move closer and closer to the subscribers for transmission speeds to meet demands. Optical transmission is in this same situation. QAM16 modulation schemes can travel about 800km without regeneration, but surpassing this performance would require lines to be cut nearly in half.
Increased performance is possible, but the necessary regeneration certainly isn’t free. Rather than wasting money trying to work around the mathematical Shannon Limit, the telecom industry needs to improve the 30% operational utilization of existing optical transport networks. In the short-term, there are two ways you can make this happen.
How to Optimize Optical Transport Networks to Manage the Shannon Limit
The utilization of existing optical transport networks breaks down to about 30% fill, 30% protection and 40% waste due to expensive wavelength conversion processes. Inside-the-box thinking might lead you to add more fibers or add regeneration, but these approaches merely mask the growing Shannon Limit issue and significantly increase your costs.
If you want to stop wasting your existing communications channels, there are two more effective ways to manage the Shannon Limit—agile optics and flexible spectrum:
- Agile Optics: Transceiver pairs are no longer relegated to static transmission speeds of 100G, 200G or even 400G. Instead, you can have optics that can adapt to line conditions, providing the maximum transmission speed for given distances and noise levels. By combining agile optics with feedback from dynamic Optical Signal-to-Noise Ratio (OSNR) monitoring, you can invest in one set of line cards and know that you’ll get the maximum capacity level.
- Flexible Spectrum: Ten years ago, the DWDM spectrum was divided into 40 to 50 100GHz wide channels. When this wasn’t efficient enough, the spectrum was broken down to about 100 50GHz wide channels. Now, we can deal with the Shannon Limit by creating channels for optical transmission in increments of 6.25GHz that can dynamically adjust to transmission requirements.
Don’t Try to Surpass the Shannon Limit—Just Manage It Better
Discussing the Shannon Limit is just one piece of a larger discussion the industry must have regarding improved packet optical transport networks. As communications service providers lose revenue to more agile OTTs, optical transport networks must improve to deliver new revenue streams.
Buying our way out of performance challenges is no longer a sufficient response—service providers must innovate to find new ways to utilize packet optical networks to support more complex, agile services.
In the bigger picture, managing the Shannon Limit will require a more elastic packet optical network and a range of multi-tiered services. If you want to learn more about how ECI can help you make this a reality, watch our latest interview with Light Reading’s Ray LeMaistre.>