This paper presents sequence-to-sequence (seq2seq) neural surrogates for high-speed transient response modeling and signal integrity analysis in fixed and variable design parameter spaces. To identify the structural requirements for prediction accuracy, multiple neural architectures were evaluated with respect to different training strategies such as sample-based and unit interval-based, as well as input configurations such as TX-only and TX+RX. For fixed design parameters, feedforward neural networks achieve accuracy comparable to recurrent models while requiring lower computational cost, as the static mapping can be approximated by fully connected and non-sequential representations. Variable design parameters alter the electromagnetic characteristics of the interconnect, which creates parameter-dependent temporal dynamics that require sequence-aware architectures to resolve the evolving interactions within the transient response. The results based on a 14-dimensional variable design parameter scenario show that while autoregressive TX+RX feedback suffers from severe phase drift and error propagation under variable conditions, non-autoregressive TX-only inference leads to stability. To overcome the structural limitations of standard decoders, a recurrent Early Fusion LSTM is proposed, which injects design parameters at the initial encoding stage. This approach achieves superior prediction accuracy, with an RMSE of 31.0 mV and an R2 of 99.51% for transient waveform prediction, an RMSE of 30.7 mV and an R2 of 99.57% for eye-height prediction, and an RMSE of 19.5 ps and an R2 of 99.94% for eye-width prediction. These results demonstrate that neural seq2seq surrogates are accurate for transient response modeling and signal integrity analysis in complex variable design parameter spaces.

翻译：暂无翻译