Effects of optical communication receiver circuits on Q and BER estimation accuracy

Q factor and bit error rate (BER) estimation based on synchronous and asynchronous sampling have been established as an important method for performance monitoring in optical networks. This study presents a simple model that estimates Q and BER from experimental data using curve fitting under the Gaussian assumption. The model improves estimation accuracy by optimising the receiver analog front end and correcting for sampling errors. The effects of front-end analog amplifiers on the sampling method accuracy have been investigated experimentally by using different optical receivers and post-amplifiers. The results show that the estimated Q and BER track the experimental values over a wide power range. In addition, post-amplifiers operating in the linear region with carefully optimised gain result in a more accurate Q and BER compared to automatic gain control amplifiers.