Radiation Effects of Brain Inspired Computing Systems

Abstract

Non-von Neumann computing systems, especially the so-called brain-inspired systems (also called cognitive and neuromorphic systems), have potential to be game changing in computing capability per unit power. The underlying technologies of computer chips have advanced to the point that functional neuromorphic architectures of increasing complexity are being reported; continuing evolution of technologies is expected to expand the practicality and functional levels of neuromorphic architectures. Such computational capability at very low power levels has tremendous potential benefit for DoD applications. A great deal is known about the response of semiconductor technologies (esp. silicon-based) to radiation; somewhat less so for emerging technologies (non-silicon, 2D materials, non-charge based memories, etc.), but in many cases is the subject of ongoing research. However, very little is known about the impact of radiation environments on the robustness of neuromorphic circuit architectures implemented in advanced technologies. This multi-disciplinary project will bring together researchers with expertise in non-conventional computer architecture, device engineering, novel circuit design and radiation effects, and leverage unique access to hardware and simulation capabilities, to advance the understanding of the radiation science of this computing paradigm. Moving forward, novel hardware/circuit specifically designed and optimized for neuromorphic computing are required to unlock the ultra-low power and massively parallel processing power of such brain-inspired systems. Among the many options, novel non-volatile mutlilevel memory elements are essential for the construct of artificial synapses. A novel charge trapped transistors (CTT) and the associated artificial synapse circuit are currently being developed at UCLA in collaboration with GlobalFoundries. In this program, we will examine the radiation effects on these CTT and quantitatively examine the effects of radiation on the neuromorphic computing architecture realized using these novel memory elements.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 10, 2017
Source ID
HDTRA11710035

Entities

People

  • Subramanian Iyer

Organizations

  • Defense Threat Reduction Agency
  • University of California, Los Angeles

Tags

Readers

  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.
  • Parallel and Distributed Computing.

Technology Areas

  • Microelectronics