In this article, we propose a novel statistical eye-diagram estimation method considering nonlinear power/ground noise induced by parallel simultaneous switching output (SSO) buffers. Probability density functions (pdfs) of output step-responses affected by the SSO are derived based on the occurrence probability of each SSO pattern. From the derived output response pdfs, main-cursor (MC) pdfs and intersymbol interference (ISI) pdfs are defined. Statistical output responses are derived based on recursive convolution between the MC-pdfs and ISI-pdfs affecting each other. The proposed method, for the first time, considers steady-state responses affected by the SSO to estimate more accurate statistical eye-diagrams. The proposed method is verified by comparing the derived statistical eye-diagrams with eye diagrams obtained using full transient simulation. Also, bit error rate (BER) bathtub curves derived using the proposed method and full transient simulation are compared for verification. The proposed method is compared with the previous study to emphasize importance of the steady-state responses. Finally, the proposed method is expanded to consider previous bit patterns, which are also affected by the SSO.

中文翻译：

---

考虑同步开关输出 (SSO) 缓冲器引起的电源/地噪声的统计眼图估计方法

在本文中，我们提出了一种新的统计眼图估计方法，该方法考虑了并行同步开关输出 (SSO) 缓冲器引起的非线性电源/接地噪声。受 SSO 影响的输出阶跃响应的概率密度函数 (pdf) 是基于每个 SSO 模式的发生概率导出的。根据导出的输出响应 pdf，定义了主光标 (MC) pdf 和符号间干扰 (ISI) pdf。统计输出响应是基于相互影响的 MC-pdf 和 ISI-pdf 之间的递归卷积得出的。所提出的方法首次考虑了受 SSO 影响的稳态响应，以估计更准确的统计眼图。通过将导出的统计眼图与使用全瞬态仿真获得的眼图进行比较来验证所提出的方法。此外，将使用所提出的方法导出的误码率 (BER) 浴盆曲线与完整的瞬态仿真进行比较以进行验证。将所提出的方法与先前的研究进行比较，以强调稳态响应的重要性。最后，将所提出的方法扩展到考虑先前的位模式，这些位模式也受 SSO 的影响。