The next future of DWDM: 100 Gbps

With 40 Gbps DWDM now in deployment, carrier and technology developer attention has turned to 100 Gbps. Three key factors will determine the timing of 100-Gbps DWDM development and deployment.

Th 314349

by Niall Robinson

With 40 Gbps DWDM now in deployment, carrier and technology developer attention has turned to 100 Gbps. Three key factors will determine the timing of 100-Gbps DWDM development and deployment.

Some of the most asked questions in the DWDM industry today focus around 100 Gbps: What will be the technology winners? Do capacity growth rates still present a sustainable business case? How quickly will the development and deployment of 100-Gbps DWDM occur?

But there is no question that with 40-Gbps transmission now in wide-scale deployment in long-haul networks and beginning to appear in metro networks, the industry is turning its research focus and development budget towards 100 Gbps.

The optical transmission market has been growing dramatically for the last several years, driving up demand for ever higher DWDM transmission rates and system capacity. Deployment of 40 Gbps has been ongoing since 2006 and continues to expand in unit volumes, total revenue, and market applications. Initially deployed in long-haul networks to support high capacity demand and interconnects between 40-Gbps router ports, the market for 40-Gbps DWDM technology has now expanded to metro networks and soon to submarine networks.

Huge capacity-growth forecasts from Tier 1 carriers have led research and development groups from transport equipment manufacturers, optical subsystem developers, and optical component companies to invest in the next-generation transmission rate, 100 Gbps.

There are three major timelines involved in the 100-Gbps market:

  1. The market demand timeline: Lead carriers have expressed the need for 100-Gbps transmission as early as 2009 or 2010.
  2. The standards bodies' timeline: The IEEE High Speed Task Force (HSTF) with the ITU and OIF are initiating significant standards activities.
  3. The industry research and development timeline: The industry must deliver carrier-grade products that are manufacturable, cost-effective, and reliable.

The first timeline centers on the demand for broadband deployment in the U.S. and how critical it is to our economy. Internet traffic grows 50% to 60% a year, according to the Minnesota Internet Traffic Studies (MINTS), which has been tracking growth in Internet traffic from publicly available sites since 2002. While the rollout of broadband Internet connections to consumers in North America has fueled significant capacity growth, it remains to be seen how these traffic growth rates will be affected by the recent economic declines. However, even if the growth slows, the need for 40-Gbps and higher transmission rates will remain.

Th 314349

Figure 1. Proposed OIF timeline for development of a 100-Gbps DWDM MSA specification

Additionally, new government stimulus packages to improve the quality and penetration of broadband services are on the way. Supporters of the plans identify broadband deployment as critical to the competitiveness of the U.S. economy.

Assuming market demand can sustain the current pace of investment in 100-Gbps technology, the industry standards bodies will continue to find the support necessary to maintain their development timelines. The most recent standards initiative was the formation of the OIF group “100G Long-Haul DWDM Transmission Module,” focused on developing a 100-Gbps DWDM module multi-source agreement (MSA) specification. The target timeline for the development of this specification is shown in Figure 1.

Finally, perhaps the hardest timeline to predict is the R&D timeline of the necessary technologies required to develop cost-effective, manufacturable, reliable transmission equipment.

Technology investment for 100 Gbps is occurring in both the electrical and optical component areas. The leading modulation format candidate for 100-Gbps DWDM deployments is coherent polarization-multiplexed quadrature phase-shift keying (C-PM-QPSK). To realize this approach will require investments in forward-error correction and framing ASICs, electrical multiplexing ASICs and, most significantly, the coherent correction ASIC that rebuilds the distorted optical signal at the receiver, enabling 10-Gbps-like transmission performance.

The long pole in the tent in terms of availability is clearly the coherent detection ASIC. Several field trials of 100-Gbps DWDM systems made the news in 2008; however, all the coherent-based demonstrations required “offline” processing of the received signal. In offline processing, the received signal is saved electronically and post-processed using coherent corrective algorithms to determine what the performance level would have been had an ASIC been available to do it in real time. Nortel has released the only commercial coherent platform at 40 Gbps and is now looking to extend the technology to 100 Gbps. But achieving a single-wavelength approach will likely require a significant investment in ASIC and analog-to-digital converter (ADC) technology due to the higher speeds required.

On the optical front, the investment focus is on optical integration. To create cost-effective components suitable for use in an MSA-defined module, significant integration of multiple optical functions is required. The OIF has another project group focused in this area: “100G Long Distance DWDM Integrated Photonics.” Of primary focus is the definition of the internal optical building blocks that will form the transmitter and receiver functionality.

Examining the transmitter optics for example, it's clear there's an opportunity to integrate the four modulators required to generate the PM-QPSK signal. In consideration are such factors as whether the polarization splitting and combining optics should be added to this building block and should the transmit laser be integrated too. Similar questions arise on the receive side with respect to the 90° hybrid, local oscillator laser and photodetectors. Fortunately there are many potential technologies and a number of technology providers are involved in the effort, which enables a full and broad discussion of the most suitable approach.

Figure 2 shows a probable timeline for commercial development and deployment of 100-Gbps systems. At the current pace of technology investment and development of the necessary standards, initial commercial systems are likely to be available towards the end of 2010. Wide-scale deployment would then occur in 2012.

Th 314351

Figure 2. Probable/possible timeline for 100-Gbps development and deployment

So what does this all mean for 100 Gbps? Market demand is strong and traffic growth rates will justify wide-scale 100-Gbps commercial deployments in three to four years. While this is later than the timelines requested by the largest Tier 1 carriers, it does match the timelines of both the standards bodies and technology development.

Many companies that are currently successful in the 40-Gbps space are extending the capabilities of their existing high-speed investments to components suitable for 100 Gbps. It's sensible to expect that the technology investments made for 40-Gbps will provide a competitive advantage in both time-to-market and design quality for 100 Gbps.

As 100 Gbps enters the marketplace, the winners will be the many companies who are already investing in this next generation of technology. Growing capacity rates make a sustainable business case, and development and deployment will occur in a matter of a few years. With that, there is no question that the DWDM industry will continue to focus on 100 Gbps.

Niall Robinson is vice president, product marketing, Mintera Corp. ( He can be reached at

Wall Street Journal: Internet Providers Move to Shape Broadband Push

Optical Internetworking Forum: Current OIF Projects

The Lightwave Channel: High-Speed Networking

More in Optical Tech