Outlook of the WDM Netorks
The throughput in optical networks is increasing inexorably, the backhauling traffic generated by triple-play and video-on-demand services being the main drivers. Therefore, the very first application for 100G systems for Ethernet transport is anticipated in metro networks and storage-area networks. A serial transmission approach is effective in the case of large capacities that require WDM transmission, because of the reduced efforts for wavelength management.
However, the way 100GbE transponders will be realized depends on the network environment. What’s needed for 100GbE transport are different for point-to-point connections, meshed networks, or meshed networks with ROADMs. For meshed networks, the transmission reach is critical due to the PMD of the fibers forming that network. In meshed networks with ROADMs, the singal in addition must have a high robustness against narrowband filtering brought on by cascaded ROADMs.
In a WDM network the commonly used frequency grid is either 50 or 100 GHz (inside a DWDM MUX, 200 GHz is also available in Fiberstore), which will have a major effect on which from the above-described approaches can be applied. Also a mixed bit rate operation with 10 or 40 Gbps channels and 100 Gbps channels must be studied.
Of course, 100GbE can invariably be realized with a WDM approach (e.g., 10×10 Gbps), but this will not increase the actual throughput. Such a WDM approach could be attractive for low-cost application, but for networks with dimensions of a MAN or WAN the wavelength management effort could be complex. For instance, if we assume a guy with ROADMs in case of a WDM approach several channels need to be dropped at a time instead of a single channel. We accept that the serial processes for realizing a 100GbE will mainly be suitable for MAN an WAN. If so such 100GbE transmission systems have to fufill several other requirements of carriers.
Ethernet has in principle the requtation of being a low-cost technology. Besides the advantages of serial approach, the serial approaches will have to compete with WDM approaches regarding the cost for the terminal equipment. This possible only when components are matured and integration is progressing indicated in the chapter on components for 100G systems.
Optical Amplifier’s Impact on WDM Netwoks
The power of light, harnessed together with components to create, modulate, manipulate, and detect it, and supported by low-loss optical fiber for transmission that ushered inside a new era of information transmission systems in the 1970s, is an incredible gift to mankind. You could have hardly expected more, however the invention and development of the optical amplifier within the late 1980s and early 1990s completed the technology suite, unleashing the full potential and power optics for communication networks. The resulting cost-effective, robust, high-capacity optical networks, along with packet-based data networks that ride over them, enabled the world-wide web which has dramatically revolutionized our daily lives.
The global growth of WDM optical networks over the last 10 years has been remarkable. Some optical networks commonly are not directly visible towards the typical consumer, the very visible internet could be impossible without them. Spanning continents, crossing oceans, reaching across metropolitan areas and now also providing direct fiber by connections, commercially deployed optical transmission systems with per fiber capacity up to several Tbps supply the enabling high-capacity connectivity that underpins the WWW (World Wide Web). No longer simple point-to-point links, today’s optical networks are flexible, switchable wavelength routed networks, both ring and mesh, that provide wavelength granular networked pipes inside the physical fiber with all-optical on and off ramps in much the way time slots are used in time-division-based transport networks.
None of this would be possible without the optical fiber amplier. The optical amplifier is truly a gift of nature that is as close to ideal as you could expect. It’s spectrally matched to fiber’s low-loss window and provides highly efficient, broadband, low noise gain. Crucial for its enabling of WDM, it features a temporal response that enables essentially unlimited signal data rates while allowing multiple wavelengths to be amplified without cross-talk between independent communication signals carried by neighboring wavelengths.
The potential of WDM to tap the bandwidth of fiber, without requiring superhigh bit rates and also the necessary enabling high-speed electronics, have been well known for some time. But, WDM was not a cost-effective solution for high-capacity systems as long as each wavelength channel had to be separeted and regenerated individually with a discrete electronic regenerator. However, the optical amplifier, using its ability to amplify multiple wavelengths simultaneously, first and foremost, made DWDM the cost-effective approach to building very-high-capacity optical transmission systems. That capability alone was revolutionary (first demonstrated in commercial products within the mid-1990s).
What at the time was far less obvious to most, even if a few could foresee it, was that in enabling DWDM transmission, the optical amplifier seemed to be preparing the way to fundamentally new network architecture using wavelengths as the networked parameter, the common unit of “currency” for enabling and building a network. These WDM networks, while they also depended on an array of other new technologies, including, most importantly, optical switching elements to construct the OADM and OXC networking elements, depended on optical amplifiers not just to enable WDM networks but additionally to compensate for that losses in these switching elements. There are both CWDM OADM and DWDM OADM in Fiberstore, where you can buy the 4CH CWDM OADM that equivalent CWDM-OADM4-1 with excellent compatibility to Cisco. WDM networks offered the possibility to provision, manage, and protect capacity according to wavelength “chunks” via fully flexible, switched wavelength networks.
While to many this vision appeared far-out, it was actually a very natural consequence of adopting WDM for transmission systems. Nevertheless, a tremendous worldwide research effort was required to provide the base of knowledge needed to answer key questions, invent and develop new technologies, and refine and demonstrate the worth proposition of WDM networks to convince providers around the globe to deploy these networks for both long-haul and metro networks. Not restricted to fiber amplifiers, for example DWDM EDFA and Raman fiber amplifier, additionally they address the possibility role of semiconductor amplifiers with its somewhat less ideal temporal characteristics but possible cost advantages, particularly when integrated on one photonic integrated circuit along with other optical functions. That role seems particularly interesting for future metro and access applications.