| Forward error correction (FEC) and line-coding techniques are attracting more and more attention in the rapid development of optical fiber communication technologies. Previous studies are mostly based on standard FEC codes and line-coding schemes, such as Hamming codes, RS codes, AMI codes, and Manchester codes, which were initially developed in wireless communications and older communications systems with the additive white Gaussian noise (AWGN) and/or binary symmetric channel (BSC) assumptions. There has been little effort to optimize the choice of codes and design new codes by taking into account the physical mechanisms behind the particular impairments in optical fiber transmission lines and systems.; The amplified spontaneous emission (ASE) noise in optical amplifiers, however, is a major source of errors in optical fiber transmission systems, but it has non-Gaussian asymmetric statistics. The chi-square distribution model is currently the most accurate theoretical approximation of ASE noise. However, for simplicity, chi-square distributions are usually approximated with Gaussian distributions having the same means and variances. In this dissertation, we do a three-level study of FEC code performance in optical fiber channels with dominant ASE noise using three different noise statistics models—chi-square asymmetric, Gaussian asymmetric, and Gaussian BSC models. The high-level study investigates the lower bound on general FEC code performance. The middle-level study investigates the upper bound on linear code performance. The low-level study investigates the possible performance improvement of Turbo-code with the more accurate optical fiber channel model. We show that, in all three level studies, using the chi-square asymmetric ASE model instead of Gaussian or BSC models may significantly improve the accuracy of the performance evaluation and/or the code performance.; Nonlinear inter-channel interference is a main source of errors in optical WDM systems that use traditional solitions. Based on the understanding of the data pattern dependence of the nonlinear inter-channel interference caused errors, we develop a new linecode, the sliding window criterion (SWC) code, to mitigate this kind of errors. We show, by simulations, that the proposed SWC code can significantly enhance data rate and reduce channel spacing in WDM optical fiber transmission systems.; Our results demonstrate that more accurate FEC code performance evaluation, significant improvement of FEC code performance, and highly efficient effective line-coding schemes can be obtained when the physical characteristics of the optical fiber transmission systems are taken into account. |
