Cold-Start of Decision-Aided Maximum Likelihood in Coherent Optical Receivers

In coherent optical communication systems, a novel cold-start method is proposed for decision-aided maximum likelihood (DA-ML) phase estimation. With adoption of the non-data-aided Mth-power algorithm, the initial transmitted data symbols can be effectively utilized to start up the DA-ML receiver without using long training sequences. Simulation results verify the feasibility of cold-start. Compared with conventional schemes, DA-ML with cold-start can effectively achieve similar system bit error rate performance, facilitating efficient transmission especially under burst-mode and multi-receiver scenarios.