Boundaries of Four-Wave Mixing Based Optical Signal Processing in Optical Communications

This thesis considers methods of meeting the challenges that face the world of optical communications; the ever increasing demand for and use of data. As usage increases so too does the electricity consumption of the internet, and both electricity consumption and load threaten to grow beyond what can be handled by existing infrastructure. To meet these challenges, we investigate spectrally highly efficient communication systems, such as the ones carrying data between continents, and using a frequency comb we push the boundaries of data rates at long distances. Novel technologies using all-optical signal processing techniques are studied, focusing on their limitations and the conditions under which they can be deployed as practical systems. We find that optical time lenses can operate using coherent modulation formats, and that such systems can be deployed in dispersion uncompensated regimes, which resemble practical links. Further we find that -10 dB conversion efficiency maximizes the noise properties of optical wavelength converters. We show that dual-stage wavelength converters can be used for true any-2-any wavelength conversion within the band of operation of the nonlinear medium, and that such systems can be improved by using Raman amplification to operate beyond the limitations of stimulated Brillouin scattering.