Improved polar decoding for optical PAM transmission via non-identical Gaussian distribution based LLR estimation.

Abstract

In this paper, an improved polar decoder based on non-identical Gaussian distributions is proposed and experimentally demonstrated for optical pulse amplitude modulation (PAM) interconnection. The principle of the polar coded PAM system is illustrated theoretically and the non-identical Gaussian distributions based log-likelihood ratio (LLR) estimation is introduced in the polar decoder to mitigate nonlinearity. Transmission systems of 28-Gbaud 4-level pulse amplitude modulation (PAM-4) and 8-level pulse amplitude modulation (PAM-8) based on commercial 10-GHz directly modulated laser (DML) are both demonstrated over 10-km standard single-mode fiber (SSMF) in C-band without dispersion compensation. Experimental results show that, aided by the improved polar decoder, the channel nonlinearity can be taken into consideration and additional sensitivity gains of 0.7 dB and 1 dB are respectively achieved compared with traditional polar decoder for PAM-4 and PAM-8 systems.