Abstract:Accurate and efficient signal integrity modeling methods are crucial in the iterative design process of high-speed links. While data-driven deep learning exhibits robust capabilities for temporal transient modeling, it often ignores the correlations among high-speed links sharing similar structures, necessitating separate retraining for each distinct case. In this study, we propose a new transient modeling approach for high-speed links employing a transfer learning-enhanced deep simple recurrent unit (TL-DSRU) method. The deep simple recurrent unit architecture overcomes the challenges in handling sequential data and parallel processing found in traditional recurrent neural networks, enabling efficient modeling. Our technique leverages the initialization principles of transfer learning, utilizing a pretrained model of a basic high-speed link to enhance the comprehension of more intricate cases. The proposed TL-DSRU model combines parallelization and transfer learning initialization methods to balance training speed and accuracy, thereby enhancing the practicality and generalization potential of neural network-based transient simulation for high-speed links. Comparative experiments demonstrate that the transfer learning-based initialization method substantially outperforms typical neural network random initialization techniques, delivering markedly improved time-domain waveform prediction accuracy across various channels and equalization in high-speed links, as well as yielding more precise predictions of eye diagram parameters.

Fast bottom up algorithm to infer link delay distributions

Inference of Link Loss Rates by Explicit Estimation

A Fast Link Delay Distribution Inference Approach under a Variable Bin Size Model.

A Fast Bottom-Up Approach To Identify The Congested Network Links

A fast path-based approach to infer link loss rates by explicit estimation

On Delay Tomography: Fast Algorithms and Spatially Dependent Models

A Fast Adaptive Routing Algorithm Based on Delay

A Fast Method To Estimate Loss Rate

A bottom-up inference of loss rate

Distributed estimation over complex networks

Delay Minimization for Federated Learning Over Wireless Communication Networks

Network Reconstruction from Binary-State Time Series in Presence of Time Delay and Hidden Nodes

Fast Parameter Estimation in Loss Tomography for Networks of General Topology

End-to-end Delay Analysis for Networked Systems

Transient Modeling of High-Speed Links Using Transfer Learning-Based Neural Network Initialization

Lossy link identification based on maximum pseudo likelihood estimation

Method of Fiber Transfer Delay Measurement Based on Phase Quantization and Delay Synthesis

The Inference of Link Loss Rates with Internal Monitors.

FAD and SPA: End-to-end Link-Level Loss Rate Inference Without Infrastructure

Fast Uplink Bandwidth Testing for Internet Users.

FABMon: Enabling Fast and Accurate Network Available Bandwidth Estimation