Numerical Investigation and Optimization of SBS-Based Slow-Light Using Filtered Incoherent Pump

The performance of stimulated Brillouin scattering （SBS）-based slow light using a novel spectrally-sliced broadband incoherent pump source is numerically studied. The profile of the pump-power spectrum is determined by the transmission spectra of the optical filter followed by the polarized broadband incoherent pump source. We also investigate the performance of Gaussian-type and super-Gaussian-type filtering under different spectrally-sliced bandwidths and pump power levels for 2.5Gbit/s return-to-zero pulse （50% duty-cycle）. The pulse broadening is characterized by the full width of half maximum （FWHM） and the rms pulse width, respectively. However,the results obtained by the two kinds of measurement methods deviate from each other with increasing pump power. Compared with the regular Gaussian-type filtering, the pulse broadening can be significantly reduced using super-Gaussian-type filtering at the cost of a small reduction in delay time. Furthermore, the maximum improvement in pulse broadening of ＆#8710; BFWHM =28.4% and ＆#8710; B RMS =10.4% is achieved by using a five-order super-Gaussian-type filter and a pump power of 500mW.