Coded Modulation Techniques in Fiber-Optical Communications

Today’s high demand for increasing the data transmission rate motivates a great challenge to improve the spectral efficiency of fiber-optical channels. In order to achieve a higher spectral efficiency, exploiting an advanced coded modulation scheme is inevitable. Since a general fiber-optic link is a non-Gaussian channel with nonlinear behavior, new coded modulation schemes need to be designed for these non-Gaussian channels. The performance of many binary classic codes such as Reed-Solomon and capacity-achieving codes such as low density parity-check codes and turbo codes, originally designed for the additive white Gaussian noise channel (AWGN), has been evaluated in fiber-optical channels. However, the design of error-correcting codes for such a non-Gaussian fiber-optical channel is complicated and is not well investigated in the literature. Multilevel coded modulation (MLCM) uses low complexity multistage decoding, which is a suitable structure for a very high-rate fiber-optical communication system. We propose a new rate-allocation method for the MLCM scheme [Paper A] based on the minimization of the total block error rate. The proposed approach uses Reed–Solomon component codes and hard decision multistage decoding. A multidimensional MLCM system with an N-dimensional constellation constructed from the Cartesian product of N identical one-dimensional constellations is introduced in [Paper B]. According to our analysis, the multidimensional scheme shows better trade-off between complexity and performance than a one-dimensional MLCM. Exploiting the results in Papers A and B, we present a novel MLCM scheme for a non-Gaussian dispersionmanaged fiber channel [Paper C]. This MLCM scheme is designed with a ring constellation and nonlinear post-compensation of the self-phase modulation produced via the Kerr effect. In this scheme, a new set partitioning based on the Ungerboeck approach is introduced to maintain unequal error protection in amplitude and phase. In contrast to AWGN channels, increasing the minimum Euclidean distance is not a valid criterion to design a coded system for such fiber-optical channels. Finally, the joint probability density function of the received amplitudes and phases of a dispersion-managed fiber-optical channel is derived in [Paper D]. This analysis is performed for dual-polarization transmission with both lumped and distributed amplifications. The derived statistics can be used to design an ML receiver for data transmission systems in these channels.