当前位置: X-MOL 学术IEEE Trans. Electromagn Compat. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Method for Accurate and Efficient Eye Diagram Prediction of Nonlinear High-Speed Links
IEEE Transactions on Electromagnetic Compatibility ( IF 2.1 ) Pub Date : 2021-06-08 , DOI: 10.1109/temc.2021.3074923
Dan Jiao , Yuhang Dou , Jin Yan , Jianfang Zhu , Adam Norman

Signaling analysis of nonlinear high-speed circuits is challenging because it cannot rely on linear time-invariant principles, whereas an exhaustive nonlinear simulation is computationally prohibitive for low bit error rates. To find the worst-case eye, an exhaustive approach requires nonlinear simulations of $2^m$ bit patterns for a channel of $m$ -bit memory. In this article, we represent the nonlinear responses to the $2^m$ inputs by a rank- $k$ matrix, where $k$ denotes the number of distinct waveforms in the responses for a prescribed accuracy, which is much smaller than $2^m$ . We further develop a fast full cross approximation algorithm to find the rank- $k$ model with a low complexity independent of $2^m$ . Simulations of large-scale real-world nonlinear circuits with an over 100-bit channel memory demonstrate the accuracy, efficiency, and capacity of the proposed work.

中文翻译:

非线性高速链路准确高效眼图预测的方法

非线性高速电路的信令分析具有挑战性,因为它不能依赖于线性时不变原理,而详尽的非线性仿真对于低误码率而言在计算上是令人望而却步的。为了找到最坏情况的眼图,详尽的方法需要非线性模拟$2^m$ 通道的位模式 百万美元 -位内存。在本文中,我们表示对$2^m$ 按等级输入 - $千$ 矩阵,其中 $千$ 表示指定精度的响应中不同波形的数量,远小于 $2^m$ . 我们进一步开发了一种快速的全交叉逼近算法来找到秩 - $千$ 具有低复杂度的独立于 $2^m$ . 具有超过 100 位通道存储器的大规模现实世界非线性电路的仿真证明了所提出工作的准确性、效率和容量。
更新日期:2021-06-08
down
wechat
bug