Exact NFDM Transmission in Presence of Fiber-Loss

Nonlinear Fourier transform (NFT) based fiber optic communication techniques so far have been applied to the nonlinear Schroedinger equation (NLSE) that models signal propagation in a lossless optical fiber. Conventionally, the lossy propagation of signal is approximated by a lossless propagation using the path-average approach which makes the propagation model suitable for NFT. The path-average model so far has been used for many demonstrations of NFT-based transmission systems. Although the errors due to the path-average approximation can be small, it can also notably degrade the performance of NFT based systems and impose challenges on designing high data rate NFT based systems. Previously, we proposed the idea of using dispersion decreasing fiber (DDF) for nonlinear frequency division multiplexing (NFDM) systems to make the transmission exact. We have shown in simulations that complete nonlinearity mitigation can be achieved in lossy fibers by designing an NFDM system with DDF along with an adapted NFT for it. We reported performance gains by avoiding the aforementioned path-average error in an NFDM system by modulating the discrete part of nonlinear spectrum. In this paper, we extend the proposed idea to the modulation of continuous spectrum. We do a comparative study of the performance of NFT based transmission systems designed with dispersion decreasing fiber to that of systems designed with a standard fiber with path-average model. We quantify the improvement in the performance of NFDM systems that use DDF through numerical simulations.