Machine understanding of domain computation for Domain-Specific System-on-Chips (DSSoC)

Richard Uhrie; Daniel W. Bliss; Chaitali Chakrabarti; Umit Y. Ogras; John Brunhaver

doi:10.1117/12.2519264

30 April 2019 Machine understanding of domain computation for Domain-Specific System-on-Chips (DSSoC)

Richard Uhrie, Daniel W. Bliss, Chaitali Chakrabarti, Umit Y. Ogras, John Brunhaver

Author Affiliations +

Proceedings Volume 11015, Open Architecture/Open Business Model Net-Centric Systems and Defense Transformation 2019; 110150O (2019) https://doi.org/10.1117/12.2519264
Event: SPIE Defense + Commercial Sensing, 2019, Baltimore, MD, United States

Abstract

Heterogeneous system-on-chips (SoC) can increase the energy-efficiency of domain-specific computation by orders of magnitude compared to scalar processors. High-performance systems can be generated procedurally through example-driven inference of a domain of computation to facilitate the design of domain-specific SoCs. This paper focuses on the domain of signal processing as it plays a recurring and important role in automation. The expertise required to build processors well-suited to a specific computation domain, rather than a single application or general computation, is inferred through the statistical analysis of computation, hardware, and their affinity for each other. This paper highlights the development of an ontological inference engine to achieve this goal.

Citation Download Citation

Richard Uhrie, Daniel W. Bliss, Chaitali Chakrabarti, Umit Y. Ogras, and John Brunhaver "Machine understanding of domain computation for Domain-Specific System-on-Chips (DSSoC)", Proc. SPIE 11015, Open Architecture/Open Business Model Net-Centric Systems and Defense Transformation 2019, 110150O (30 April 2019); https://doi.org/10.1117/12.2519264

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE