Efficient backward-propagation using waveletbased filtering for fiber backward-propagation

Abstract

With the goal of reducing the number of operations required for digital backward-propagation used for fiber impairment compensation, wavelet-based filtering is presented. The wavelet-based design relies on signal decomposition using time-limited basis functions and hence is more compatible with the dispersion operator, which is also time-limited. This is in comparison with inverse-Fourier filter design which by definition is not time-limited due to the use of harmonic basis functions for signal decomposition. Artificial, after-the-fact windowing may be employed in this case; however only a limited amount of saving in the number of operations can be achieved, compared to the wavelets-base filter design. Wavelet-based filter design procedure and numerical simulations which validate this approach are presented in this paper.