Jitter and Eye Estimation in SerDes Channels Using Modified Polynomial Chaos Surrogate Models

Abstract

Estimation of data-dependent jitter and the resulting eye diagram in modern SerDes channels using state-of-the-art conventional simulation techniques are either computationally expensive or have limits in their application. Therefore, this paper proposes an approach based on uncertainty quantification, where a surrogate model using the Polynomial Chaos (PC) theory is developed and used to predict jitter, eye height, and eye width including the statistical variation. The accuracy and efficiency of this approach is demonstrated using a high-speed SerDes channel topology and specialized SerDes simulation tool, which shows about 100X speedup in channel simulation costs for this example.