Four wave mixing-based wavelength conversion in silicon nanowire waveguide

number: 
2739
English
Degree: 
Author: 
Mina Talal Ahmed Al- Saad
Supervisor: 
Dr. Raad Sami Fyath
year: 
2012
Abstract:

Silicon-on-insulator (SOI) nanowire waveguides with high refractive-index contrast are attractive for all-optical signal processing due to their enhanced nonlinear optics. This thesis addresses theoretically wavelength conversion based on four-wave mixing (FWM) in these advanced waveguides. Comprehensive numerical model is developed to investigate the characteristics of FWM in silicon nanowire waveguides. The model is based on a set of coupled nonlinear Schrödinger equations describing the propagation of the pump, signal, and converted waveform through the waveguide. The effect of dispersion, free-carrier absorption, two-photon absorption, and state of polarization are taken into account. Analytical solutions are also deduced from the model under simplified conditions to get approximated description for the key roles played by various system parameters. Both numerical and analytical models are applied to four silicon nanowire waveguides, differ in their cross section dimensions, to assess their FWM-based wavelength conversion characteristics. The polarization-dependent parameters of these waveguides are included in the modeling. The effect of pump power, waveguide length, and wavelength shift on conversion efficiency are addressed in details. The investigation is carried further by simulating the performance of these wavelength converters in optical communication systems using Opti system software. The results indicate clearly that there is an optimum length for the nanowire waveguide which maximizes ηc. The maximum value of ηc is -18.39dΒ, at pump power=100mW, which is achieved when the (285nm×650nm) cross section waveguide is used in TM-mode operation with optimum waveguide length of 1.03cm. Further, the performance of wavelength conversion is degraded by both laser phase noise and the intrinsic noise associated with FWM conversion. Generally, TM operation yields higher conversion efficiency compared with TE operation due to its higher input coupling efficiency.