Highly Efficient Test Architecture for Low Power AI Accelerators
- Title
- Highly Efficient Test Architecture for Low Power AI Accelerators
- Author
- 박성주
- Keywords
- Electrical and Electronic Engineering; Computer Graphics and Computer-Aided Design; Software
- Issue Date
- 2021-09
- Publisher
- IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- Citation
- IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, Page. 96700-96710
- Abstract
- Low power AI accelerators are being developed to
support the battery-operated edge devices at a minimum expense
of classification error. However, the testing of such large AI
accelerators with traditional techniques is inefficient in achieving
the required certifications for Autonomous Driving Assistant
Systems (ISO 26262). ISO 26262 sets very stringent requirements
on the testing time and fault coverage during diagnosability of
faults leading to system level failures during in-field testing. This
paper proposes a test architecture for low power AI accelerators
by reusing the existing data paths for large AI accelerator arrays.
As compared to the full scan-DFT, the proposed test architecture
reduces the test time and peak test power, which enhances the
reliability of the test responses. The proposed technique reduces
(1) the switching power by 87%, (2) testing times by 72% on
average for cases upto 32 × 32 and (3) the peak power by 59%.
Further, there is an average reduction in area by 10% for the
accelerator.
- URI
- https://ieeexplore.ieee.org/document/9530455https://repository.hanyang.ac.kr/handle/20.500.11754/170582
- ISSN
- 1937-4151; 0278-0070
- DOI
- 10.1109/TCAD.2021.3110739
- Appears in Collections:
- ETC[S] > 연구정보
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML