Advanced Voltage Variation Aware Timing Analysis and Reporting (AVATAR)

Abstract

This program developed a new methodology to accurately predict the voltage and timing of digital integrated circuits such that it would the designer may reduce the effective gate level induced timing slack to mitigate trojan insertion on critical nodes. This work utilizes modeling the voltage drop and voltage noise pre-fabrication, and with training a Neural Network post-fabrication, technique to improve the timing model collected during timing closure and produces a Neural assisted Golden Timing Model (NGTM) for side channel delay-signal analysis. The Neural Network acts as a process tracking watchdog for correlating the static timing data (produced at design time) to the delay information obtained from clock frequency sweeping test.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 2021
Accession Number
AD1149070

Entities

People

  • Avesta Sasan

Organizations

  • George Mason University

Tags

Communities of Interest

  • Advanced Electronics
  • Cyber
  • Space

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Artificial Intelligence Software
  • Circuits
  • Commerce
  • Computers
  • Data Mining
  • Electronics
  • Electronics Industry
  • Electronics Laboratories
  • Energy Consumption
  • Fabrication
  • Government Procurement
  • Governments
  • Information Science
  • Integrated Circuits
  • Intellectual Property
  • Machine Learning
  • Neural Networks
  • Semiconductors
  • Simulations
  • Standards
  • Supervised Machine Learning
  • Very Large Scale Integration

Readers

  • Computational Modeling and Simulation
  • Computer Engineering
  • Cybersecurity.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks