Search Results - roman+lysecky

3 Results

Sort By:

Portable Deep Neural Network based Side Channel Attacks

This technology uses deep neural networks to examine the power states of a device performing cryptographic operations and determine the underlying encryption key with high accuracy across varying processor architectures. Background: Embedded systems are among devices often the most vulnerable to attacks on their security. They are often used in public...

Published: 10/11/2023 | Inventor(s): Roman Lysecky, Janet Roveda, Manoj Gopale, Gregory Ditzler

Keywords(s):

Category(s): Technology Classifications > Engineering & Physical Sciences > Communications & Networking > Networking, Technology Classifications > Engineering & Physical Sciences > Communications & Networking > Wireless

TrustFlow

TrustFlow is a technology that enhances the security of systems based on the RISC-V architecture by protecting the integrity of sensitive data in memory. Background: Embedded systems and the Internet of Things are becoming more and more commonplace. As these devices become ubiquitous and applied to more critical applications such as medical uses and...

Published: 8/1/2024 | Inventor(s): Roman Lysecky

Keywords(s):

BackFlow renamed BackGuard

This invention, BackGuard, is a security extension for the compiler front end Clang/LLVM which uses the programming language C/C++. The extension is intended to facilitate the integration of software security in embedded applications and is intended to protect against software attacks including the remote execution of malicious code. BackGuard can...

Published: 8/1/2024 | Inventor(s): Roman Lysecky

Keywords(s):