PAM-4 eye-opening monitoring techniques based on Gaussian mixture model fitting

Abstract

Four-level pulse amplitude modulation (PAM-4) data formats are adopted to achieve next-generation high-speed data transmission standards. In this letter, a novel eye-opening monitoring technique based on machine learning is proposed to evaluate the received signal quality for the adaptive coefficients setting of a transmitter feed-forward equalizer for PAM-4 signaling. The monitoring technique employs a Gaussian mixture model (GMM) to classify the received PAM-4 symbols. Simulation and measured results of the coefficient adjustment using the GMM method are presented.