In this paper, we cover the fundamentals of neural networks and Bayesian learning with a focus on signal and power integrity problems arising in packaging. Rather than only focus on mathematical formulations, we explain the important concepts and the intuition behind them, thereby demystifying the use of machine learning for these problems. We also share some of the recent developments in this area along with future research directions in the context of packaging. Links to open source downloadable software for some of the methods discussed are also provided.
1 January 1996 in IEEE Transactions on Components, Packaging, and Manufacturing Technology: Part B
The importance of inductance and inductive coupling for accurate delay and crosstalk prediction in on-chip interconnections is investigated experimentally for the top three layers in a five-layer wiring structure and guidelines are formulated. In-plane and between-plane crosstalk and delay dependence on driver and receiver circuit device sizes and line lengths and width are analyzed with representative CMOS circuits. Simplified constant-parameter, distributed coupled-line RLC-circuit representation that approximates the waveforms predicted with frequency-dependent line parameters is shown to be feasible.
12 August 2020 in IEEE Transactions on Electromagnetic Compatibility
One of the favorable tools for signal integrity evaluation is eye diagram analysis. This is traditionally performed with a lengthy transient simulation, which can be prohibitively time consuming for complex high-speed channels with a low bit error rate. Methods for eye estimation exist; however, they are either only applicable to linear time-invariant systems or have lack in accuracy or efficiency. In this article, an optimization-based approach is proposed to quickly obtain the worst-case eye diagram characteristics. This approach focuses on the inter-symbol interference since its effect can span over many symbols and include crosstalk, making it challenging to model. In this article, the data patterns leading to the lowest voltage corresponding to a high symbol, the highest voltage corresponding to a low symbol, and the times of minimum and maximum level crossing points are calculated. Then, eye height, eye width, and the worst-case eye opening are estimated using these points. To reduce complexity, the proposed approach includes a mapping algorithm that exploits the Gray code. Additionally, Bayesian optimization is used because of its efficiency and good performance on non-linear and non-convex problems. Finally, the application of the proposed approach to high-speed SerDes channels, and channels in system-on-package designs is evaluated with numerical examples, where the results show its accuracy and efficiency.