Keynote 2 — Ioannis Tomkos

Recent advances in backbone optical networks The traffic carried by core optical networks as well as the per-channel interface rates required by routers are growing at a remarkable pace year-over-year. Optical transmission and networking advancements have so far satisfied these traffic requirements by delivering the content over the network infrastructure in a cost and energy efficient manner. However we are approaching fundamental spectral efficiency limits of single-mode fibers and the growth capabilities of conventional WDM networks operating on a fixed frequency grid are quite limited. Over the recent years, a large number of significant innovations that are able to offer a significant improvement in capacity increase (compared to legacy WDM systems at 10 Gb/s on a 50-GHz spacing) have emerged. Initial efforts targeted innovative modulation/coding techniques and flexible frequency allocations, in an effort to increase the spectral density in optical fiber links, leading eventually to the definition of spectrally flexible/elastic optical networks utilizing optical superchannels together with spectrally flexible/elastic multiplexing schemes (e.g. OFDM and Nyquist WDM), and advanced modulation formats which enable the dynamic and adaptive allocation of end-to-end demands with variable connection characteristics (e.g. requested data rates). Along these lines, novel all-optical processing schemes (e.g. realizing all optical add/drop of sub-channels out of super-channels and all-optical traffic grooming) assist in realizing transparentoptical networking with increased capacity and energy efficiency. However, while the spectrally flexible/elastic super-channel transmission/networking approach can optimize network resources through increased spectral utilization, it has limited growth potential due to the nonlinear Shannon limit imposed on the transport capacity of single-mode optical fiber within the limited gain bandwidth of optical amplifiers. Multi-band amplificati- n technologies (e.g., C+L+S-band amplifiers) may yield temporary relief, but the only evident long-term solution to extend the capacity of optical communication systems relies on the use of the spatial domain. The simplest way to achieve spatial multiplexing is to deploy multiple systems in parallel. However, by simply increasing the number of systems, the cost and power consumption also increases linearly. In order to limit the increase in cost and power consumption, component sharing and integration have to be introduced. To this extent, significant research efforts have focused on the development and performance evaluation of few-mode fibers (FMF) and multi-core fibers (MCF), which can be seen as ‘integrated fiber’ media, for space division multiplexed (SDM) systems. This line of work is further supported by the development of integrated optical amplification systems, as well as the significant development efforts in the field of Tb/s integrated transponders for SDM systems. For such systems, the use of spatial superchannels, which are groups of same-wavelength sub-channels that are transmitted on separate spatial modes but routed together, are being investigated. However the benefits of SDM systems and the potential for deployment are yet to be fully understood. Finally it should be noted that in the optical networking evolution, significant efforts are being made on the development of the proper control plane framework to orchestrate the operation of such spectrally and spatially flexible networks in order to bring out their full-potential (i.e. besides capacity increase to support also other capabilities like network virtualization) and in that framework new solutions based on the software defined networking (SDN) paradigm are considered. This talk will survey the recent research developments in the field of core optical networking and will present in a tutorial form the main solutions that are currently under investigation.