TY: THES T1 - Laser phase noise mitigation for subcarrier-multiplexing optical transmission systems A1 - Neves, Manuel dos Santos N2 - Multi-subcarrier optical systems have recently attracted the attention of the scienti c and industrial communities, owing to their inherent advantages in terms of ber propagation performance as well as their implementation complexity, when compared with legacy single-carrier systems. With the imminent introduction of next-generation transceivers supporting aggregate symbol-rates beyond 100 Gbaud, the achievable gains provided by multisubcarrier modulation are becoming increasingly more important, thereby attracting the attention of several leader optical manufacturers, which are now including this technology in their portfolio. However, the transition from single-carrier to multi-subcarrier modulation still requires signi cant optimization of digital signal processing subsystems, in order to maximize the inherent potential of subcarrier multiplexing. In this thesis, techniques for carrier phase estimation are explored. The penalty associated with the usage of the methods used in single-carrier systems when changing to a multi-subcarrier system is quantitatively studied and the need for a change in such methods is evidenced. Several jointly processing carrier phase estimation approaches are employed to improve the performance of phase recovery algorithms, resulting in an improvement in the power requirements for multi-subcarrier systems. In particular, a novel approach to carrier phase estimation is developed and assessed. This new method increases the tolerance of the system to chromatic dispersion and makes viable the usage of multi-subcarrier systems with a high number of subcarriers, highly outperforming any other carrier phase estimation approach for such systems. Finally, a novel low-complexity implementation of a digital monitoring system is suggested. It is capable of measuring the laser line width of the transmitter and local oscillators lasers. This implementation is only made possible by the nature associated with the novel algorithm presented. UR - https://ria.ua.pt/handle/10773/29172 Y1 - 2020 PB - No publisher defined