Optical transport network design beyond 100 Gbaud [Invited]

Optical line interface technology has been the key enabler to reduce the cost per bit transported, thus cost-effectively scaling optical transport networks and mitigating or even avoiding the need to roll out or lease additional optical fibers. However, this technology is reaching fundamental limits, hampering the expectation of significant gains in spectral efficiency in the foreseeable future. State-of-the-art line interfaces already exploit symbol rates that are roughly twice those available with the preceding generations to increase per-channel capacity, and this trend is likely to continue. In the short term, harvesting the benefits of introducing these interfaces mostly depends on the installed reconfigurable optical add/drop multiplexer infrastructure. In the longer term, the impact of further increases in the symbol rate also depends on the evolution of the dominant client data rates and on the channel format selection strategies. Considering a reference transport network and extrapolating how client traffic rates and line interface baud rates will evolve, this work presents a preliminary assessment of the potential benefits and shortcomings of state-of-the-art and future generations of line interfaces.