How “flexible” will OFC/NFOEC 2013 be?
By Dr. Ioannis Tomkos | Posted: 13 December 2012 8:54:42 AM
Twitter | LinkedIn | Facebook | Reddit
The world we live in is constantly changing. We see big transformations taking place across the globe that affect our lives. In such turbulent times, we all need to be flexible and adapt quickly to new conditions. As Darwin said, it is not the strongest that will survive but the fittest.
Like everything else in life, optical communication systems need to be flexible. Flexibility is needed in order to adapt to new traffic conditions and network characteristics (such as degradation of the quality of fibers and other physical network elements). But what led us to require such increased level of flexibility from the optical networks?
Over the past decade we have experienced major changes in the traffic characteristics of access and backbone networks. For a start, data traffic surpassed voice traffic, and then data traffic grew even more dramatically with the emergence of video-based applications/services (i.e. such as YouTube) that are huge bandwidth-generating sources. In addition, more recently, traffic variations have become even more unpredictable through the emergence of 3G/4G data services to and from fast moving mobile users. In the future, such conditions will be further exemplified via the wide-spread deployment of cloud services to (mobile) consumers. All of these new traffic sources lead to new characteristics such as requirements for massive network bandwidth capacity, rapidly changing traffic patterns (i.e. high traffic churn), high peak-to-average traffic ratio, increasing high Quality of Service (QoS) traffic, etc.
Therefore, due to the new traffic conditions, optical communication networks have some very difficult problems to face. On the one hand, the quest for ever-increasing channel bit-rates and higher capacity of single-fiber-core optical communication systems is becoming more and more challenging; while on the other, the rigid fixed bandwidth allocation of WDM channels, as defined by current ITU-T standards, is not quite appropriate to accommodate the traffic churn without significant over-provisioning and waste of resources.
The available fiber bandwidth cannot be considered an abundant resource anymore and as a result, the next generation of the optical systems should operate with spectral efficiencies close to the maximum possible values (as determined by Shannon’s theorem) in order to achieve the maximum capacity that can be offered in the C-band. Recently some scientists have proposed the deployment of new fibers that have multi-cores or/and can take advantage of multi-modes per core as a solution to this problem. These are certainly forward-looking solutions, as I am convinced that network operators would only abandon their currently deployed fiber infrastructures as a very last resort. Other scientists propose a more short-term solution which tries to utilize the available fiber spectrum more efficiently/wisely by properly adapting the channel capacity to the actual traffic requirements at any given point in time (as this is the case in wireless networks where bandwidth was always considered as a scarce resource).
Consequently, in the last few years, the OFC/NFOEC research community has focused on new spectrum-efficient optical networking techniques that have been proposed as a way to offer efficient utilization of the available resources. "Flexible." "elastic," "tunable," "grid-less," or "adaptive" are few examples of the terms used in the literature to describe solutions that migrate from the fixed WDM single line rate systems to systems that provide support for the most efficient (and dynamic) bandwidth utilization.
All such solutions with the term “flexibility” refer to the ability of the network to dynamically adjust its resources (wavelength channels, bandwidth, transmission format, data rate, etc.) in an optimum and elastic way (taking advantage of coherent receivers and software-defined network elements) according to the continuous varying traffic conditions, while taking into consideration the QoS and Quality of Transmission (QoT) requests of both the pre-established and newly assigned network connections.
Based on all these developments, OFC/NFOEC 2013 has also become “flexible” -- with several workshops, invited talks and contributed presentations covering the topic of flexible/elastic optical communication systems and networks. Certainly there are many research challenges that need to be tackled by all of us in order for Flexible Optical Networks to become a reality. Some of these challenges will be indeed discussed at the OFC/NFOEC workshop titled “Will Flexgrid Networks be Worth the Investment for just 30% Improvement?”, while several invited talks (e.g. “Elastic Transponder and Regenerator” by Masahiko Jinno - NTT, “ROADM Options in Optical Networks: Flexible Grid or Not?” by Sheryl Woodward - AT&T, “Can We use Flexible Transponders to Reduce Margins?” by Jean-Luc Auge - Orange Labs, “Flexgrid, SDM, Multi-fiber: Which is Best for Networks?” by Jim Benson - Nokia Siemens Networks, and “Elastic and Green Optical Access based upon Coherent Interleaved Frequency Division Multiple Access” by Ken-ichi Kitayama - Osaka University) will address the relevant technical issues.
When the final program becomes available and throughout the conference I will comment on the most important recent developments in this field and discuss the content of the most relevant presentations. I hope that all readers of this blog can contribute to the relevant discussion pointing out the important new research directions and developments.
Ioannis Tomkos (B.Sc., M.Sc. Ph.D.), has been with AIT since September 2002 (serving as Professor, Research Group Head and Associate Dean). In the past he was Senior Scientist at Corning Inc., USA (1999 – 2002) and Research Fellow at University of Athens, Athens, Greece (1995 - 1999).
Posted: 13 December 2012 by
Dr. Ioannis Tomkos
| with 0 comments