DesignCon is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.


Welcome to the DesignCon 2020 agenda and presentation download site. Here you can view and download conference and/or Chiphead Theater presentations before, during, and after the event. If you’re looking for a presentation from a specific session that you’re unable to find here, it is likely because the presenter has not provided permission for external use or has not yet shared their presentation with us. Please check back after the event for a more complete catalogue of available presentations.

Self-Evolution Cascade Deep Learning for SerDes Adaptive Equalization

Bowen Li (PhD candidate, North Carolina State University)

Brandon Jiao (Senior Staff Transceiver Engineer, Xilinx, Inc.)

Chih-Hsun Chou (Senior FPGA Design Engineer, Xilinx)

Romi Mayder (Senior Director, Xilinx, Inc.)

Paul Franzon (Distinguished Professor, North Carolina State University)

Geoff (Geoffrey) Zhang (Distinguished Engineer, Xilinx, Inc.)

Location: Ballroom G

Date: Thursday, January 30

Time: 2:50pm - 3:30pm

Track: 14. Machine Learning for Microelectronics, Signaling & System Design, 06. System Co-Design: Modeling, Simulation & Measurement Validation

Format: Technical Session

Pass Type: 2-Day Pass, All-Access Pass, Alumni All-Access Pass - Get your pass now!

Vault Recording: TBD

Audience Level: All

IBIS-AMI technology has become the de-facto methodology to model SerDes behavior for end-to-end link simulations. Meanwhile, machine learning (ML) techniques can facilitate the computer to learn a black-box system. This paper proposes the Self-Evolution Cascade Deep Learning (SCDL) model to show a parallel approach to modeling adaptive SerDes behavior effectively. Specifically, the proposed self-guide learning methodology uses its own failure experiences to optimize its future solution search according to the prediction of the receiver equalization adaptation trend. After the basic introduction of SCDL model, the paper presents results that are highly correlated with the IBIS-AMI simulations.

Takeaway

With self-guide information, the proposed SCDL model can accurately track the CTLE and DFE equalization adaptations over 50 various cases within 20 seconds based on only 150 training cases.

Intended Audience

1. Basic concept of high-speed serial link systems
2. Basic idea of IBIS-AMI modeling of high-speed serial links
3. Basic understanding system performance evaluation and analysis
4. Basic knowledge of machine learning/deep learning algorithms