An eye diagram parameters measurement method based on K-means clustering algorithm

Abstract

Eye diagram and its parameters measurement are very important to represent and analyze the high-speed digital signal. In this paper, K-means algorithm, which is popular in data mine, is introduced into the eye diagram parameters measurement. With the advantage of K-means algorithm, the method proposed in this paper can work automatically and effectively. A differential line is measured by vector network analyzer (VNA) and several eye diagrams at different data rates are generated by the measured S parameters. The method is used to measure these eye diagrams. Experiment results show that this method works very well.