Multi-Subcarrier Signal Processing for Fiber Optic Communication Systems

Abstract

Although fiber optic communication systems have enjoyed tremendous developments since their inception roughly fifty years ago, there are still improvements to be made. One such area for growth, relating more to long-haul applications of optical fibers, is the mitigation of nonlinear penalties. These nonlinear penalties appear due to the transmitted signal affecting the material properties of the fiber itself, and are more prominent when said signal has more energy in it—a conflicting effect from the traditional stance where more energy is considered beneficial. These penalties come in different forms, although they all have the same degrading effects for the system in question. Therefore, it is imperative to find new techniques that can mitigate these nonlinear penalties to ensure the optimal performance for our systems. One such technique, that we call multi-subcarrier processing, is the interest of the thesis here. The first section proposes a new method to compensate for the receiver side digital signal processing. More specifically, it introduces an algorithm that can remove the intermediate frequency offset that the signal acquires after passing through the system, which is necessary for coherent systems employed today. As some traditional algorithms used in the signal processing no longer work for multi-subcarrier systems, this new algorithm is necessary to compensate effectively for said offset. The second section introduces a new signal processing technique for the multiplexing and demultiplexing of multi-subcarrier signals. By making use of the well-known Fast Fourier Transform algorithm, the computational cost for generating said signals can be drastically reduced as the number of subcarriers grow larger. As systems continue the trend of operating at greater baud rates, the savings introduced here should be of even greater significance for commercial systems of the future.