|
Activity Number:
|
435
- SPEED: Sports to Fire: Fascinating Applications of Statistics
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 3:05 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #332654
|
|
|
Title:
|
An Application of Machine Learning for 3D IC Defect Detection
|
|
Author(s):
|
Meihui Guo* and Yu-Jung Huang
|
|
Companies:
|
National Sun Yat-Sen University and I-Shou University
|
|
Keywords:
|
Through Silicon Via;
HFSS Scattering parameters;
Machine learning
|
|
Abstract:
|
The TSV process is a key technology for 3D chip stacking by providing a route with the shortest and vertical interconnection path. Defective TSVs can be introduced during TSV fabrication process and can potentially cause signal integrity problems and fail the entire 3D IC along with all known-good dies. As the technology node scales, the defective TSV will face more severe reliability issues. In this presentation, we provide HFSS data for various types of TSV defect model and design various statistical models for TSV defect analysis. The proposed methodology can be cross-validated to achieve the TSV fault detection in stacked die structure. The objective of the present proposed machine learning based method to predict TSV faults is expected to reduce the testing time and cost to ensure the signal integrity reliability of the 3D stacked dies.
|
Authors who are presenting talks have a * after their name.
