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.
Location: Ballroom G
Date: Wednesday, January 29
Time: 2:00pm - 2:45pm
Track: 14. Machine Learning for Microelectronics, Signaling & System Design, 07. Optimizing High-Speed Serial Design
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
Follow up on our previous paper on accelerating channel optimization using principal component analysis. We create families of surrogate models in the reduced dimension PCA space based on various channel conditions. When a new system topology is encountered, random points within the PCA space is sampled. The resulting performance is compared against the families of surrogate models and the closest solution is considered the nearest channel condition model. The optimal operating can then be easily set based on precomputed optimal setting for that surrogate model. Alternatively, the precompute settings can be used as seed values for circuit level auto tuning.
Automatic channel identification
Automatic channel tuning
Principal component analysis vector space
Polynomial chaotic expansion surrogate models
Principal component analysis, Surrogate model , PAM-4 SerDes, channel optimization