EDFAs enable providers to satisfy end-user data-intensive needs
With ever-increasing subscriber demand for high-bandwidth services such as streaming HDTV, 5G Internet, and Unified Communications tools, global optic networks are under intense pressure to deliver data-intensive applications over long distances continuously.
Many access networks are turning to coherent Dense Wave Division Multiplexing (DWDM) optics as a solution. Coherent optics transmit high data rates (i.e., 100G, 200G, 400G, etc.) over long-haul and ultra-long-haul optical communication links. Their ability to carry multiple data streams on a single optic fiber coherent optics supports increased network traffic capacity with superior signal processing and low error rates.
But with these solutions come new challenges. In high-performance data networks, transmit power is essential for achieving signal quality and ensuring long-distance signal reach. As access networks try to implement the newer 400ZR and Open ZR+ coherent optics technology standards, they may find it hard to achieve the necessary transit power levels. According to the minimum specifications, the output optical power starting point for 400ZR and OpenZR+ is -10 decibel milliwatts (dBm). Still, many access network providers will find that they need output optical power of 0 dBm, especially for architectures using DWDM passive optical muxes and demuxes. Having an optical power of 0 dBm allows for extended reach with the possibility of eliminating amplifiers (EDFAs) as well as connecting to existing Optical Line Systems (OLS) or Reconfigurable Optical Add-Drop Multiplexers (ROADM).
Integrating embedded micro EDFAs into coherent optics enables access networks to achieve the required transit power levels. Using erbium, a rare-earth element, micro EDFAs amplify optical signals, providing power optimization to deliver data-intensive content, enhance system performance, and extend the reach of long-distance data transmissions.
DCI vs. access networks
Network operators are adopting two common standards for coherent optics: 400ZR by the Optical Interconnect Forum (OIF) and OpenZR+, a separate multisource agreement (MSA). These specifications (especially 400ZR) were primarily developed for giant cloud hyperscalers in their Data Center Interconnect (DCI) networks.
DCIs are active high-bandwidth, point-to-point networks that transmit data between multiple facilities across a metro environment. DCI networks have often used external tools such as EDFAs to amplify their signals, which enables transmission over long distances, from Point A to Point B, for example.
Access network operators, including MSOs, Mobile Operators and fiber infrastructure providers, do have a use for DCI-style interconnection, with its high data transmission capabilities. But access networks also have a highly distributed architecture, where different and often pre-existing links (e.g., A to B, A to C, B to C) have different power transmit requirements. If one link on the network has a -10 dBm coherent transceiver, it cannot interface with an OLS or ROADM, which typically have power input requirements of 0 dBm. Even though OpenZR+ gives more flexibility in data rates and use cases (i.e., enabling metro, regional and long-haul connectivity beyond DCI), it too falls short of the 0 dBm need for access networks.
Often, access network operators need to equalize optical output powers at 0 dBm, which is becoming the unofficial new minimum requirement even as the market trends towards OpenZR+ and all the flexibility and capability that it brings (including requests for +3 dBm and +5 dBm). Many access networks feature passive architectures; adding external amplification components would be an antithetical solution in these cases. After all, the more a network operator can reduce dependence on amplification, the better the optical signal-to-noise ratio one can achieve in a link.
Though OpenZR+ enables more functionality for access network operators, its typical TX output power of around -10 dBm has constrained its use. In some ways, the legacy of 400ZR and DCI, which has often relied on external amplifiers, has impacted OpenZR+, intended to open new doors for access network operators seeking to use coherent optics for purposes beyond DCI. As we’ve noted in a previous whitepaper, OpenZR+ can be considered a continuation of 400ZR, and from a power output perspective, this has created difficulties for modern access networks. After all, though the use of Open FEC (O-FEC) in OpenZR+ enables better compensation for chromatic dispersion than 400ZR’s use of Concatenated FEC (C-FEC), OpenZR+ modules still require external amplification to achieve their greater reach. Consequently, OpenZR+ optics have needed further optimization for the access network.
Enhancing transmit power
Most 400G coherent optics are based on silicon photonics, where components such as modulators, switches, and detectors are created on a silicon wafer. This provides high integration density and potential cost reductions over standard indium phosphide photonics.
But with silicon photonics, specifically the 400ZR and OpenZR+ standards, the minimum output optical power is -10 dBm, which is not a good starting point for access network operators that have a goal of unamplified links (it’s possible to reach long distances with EDFAs). The solution for both standards has been to innovate to create transceivers with a higher TX output power at around ~0 dBm.
For example, some suppliers in the industry have started presenting some variations of OpenZR+ modules with higher TX output powers from +3 to +5 dBm. That said, it’s essential to remember that none of these optics are plug-and-play. In some scenarios, the higher TX output power may not be the most advantageous solution. Choice in coherent optics depends on much more than output power. It’s necessary to contextualize one’s selection regarding the entire network design, from the link budget to the type of mux/demux used, the fiber span, the OSNR, and any need to interconnect to an OLS or ROADM. Doing the latter, for example, will generally require a 0 dBm input as +3 to +5 dBm may present too high of a power level.
But when overcoming the -10 to 0 dBm hurdle, the most modern and forward-thinking innovation has been integrating embedded micro-erbium-doped Fiber Amplifiers (EDFAs) into coherent optics. There are, however, a few other ways of also getting around the power output gap, including:
· External Solutions: Use the -10 dBm optic as-is and add external EDFAs along the network. However, this defeats the access network operator’s goal of not adding external amplification devices to a passive network architecture.
· Pluggable Solutions: Use the -10 dBm optic as-is and add a second, pluggable optic that functions as an EDFA. However, with this solution, the switch/host platform needs two ports instead of one, which wastes valuable space.
As you can probably surmise, the embedded micro EDFA approach provides the most efficient solution for boosting transmit levels from -10 dBm to 0 or greater. However, it is also complicated to achieve. For manufacturers, miniaturizing the EDFA to 1/10th the size of a standard EDFA product is no easy feat. More importantly, fibers are sensitive to a tight bend radius, but achieving this in a tiny form factor package is essential. Miniaturization is an innovation that will pay dividends for access network operators. Still, a network operator’s choice of coherent optics will depend on various factors, not just power output.
Nevertheless, as the market shifts toward 400G ZR and Open ZR+ standards, 0 dBm is fast becoming the unofficial minimum requirement for optical output levels service providers need. Embedded micro EDFAs hold the most significant potential to meet future transmit power needs in access networks, as mini EDFAs can boost transmit levels to 0 dBm or higher. (Of course, this enhanced transmit power comes with a higher power consumption and a complex design to fit the micro EDFA into a tight space within QSFP-DD form factors. So, there will be trade-offs to be made.)
Custom access network solutions
While integrating micro EDFAs into coherent optics offers excellent potential, no off-the-shelf products provide a one-size-fits-all solution to the need for high power transmit levels on access networks. Coherent optics are more complicated than anything network operators have encountered before, and different networks have different transmit power needs. Achieving a higher optical output power may not be the right solution for your network.
As they try to upgrade their network to achieve the 0 dBm standard, access network providers might be unsure of the exact kind of solution they require. The best way to overcome optical transmission issues is to partner with a network integration specialist. A vendor offering systems integration and network engineering expertise and robust and customized testing approaches can help you find the right solution to help your network meet its optical transmit power needs.
Chris Page is the SVP of Engineering and CTO of Precision OT. He oversees the company’s Engineering and R&D organization of technical experts, innovators, and industry thought leaders.
For related articles, visit the Optical Tech Topic Center.
For more information on optical components and suppliers, visit the Lightwave Buyer’s Guide.
To stay abreast of optical communications technology, subscribe to Lightwave’s Enabling Technologies Newsletter.