MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING

MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING

Finding potential security weaknesses in any complex IT system is an important and often challenging task best started in the early stages of the development process. We present a method that transforms this task for FPGA designs into a reinforcement learning (RL) problem. This paper introduces a me...

Full description

Bibliographic Details
Main Author: Michael Vetter
Format: Article
Language:English
Published: CTU Central Library 2019-11-01
Series:Acta Polytechnica
Subjects:
fpga, it security, model-driven design, reinforcement learning, machine learning.
Online Access:https://ojs.cvut.cz/ojs/index.php/ap/article/view/5030
id doaj-850d2ef798ff4e0eb51cec4bcbc7f306
record_format Article
spelling doaj-850d2ef798ff4e0eb51cec4bcbc7f3062020-11-24T21:51:48ZengCTU Central LibraryActa Polytechnica1210-27091805-23632019-11-0159551852610.14311/AP.2019.59.05184668MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNINGMichael Vetter0University of West BohemiaFinding potential security weaknesses in any complex IT system is an important and often challenging task best started in the early stages of the development process. We present a method that transforms this task for FPGA designs into a reinforcement learning (RL) problem. This paper introduces a method to generate a Markov Decision Process based RL model from a formal, high-level system description (formulated in the domain-specific language) of the system under review and different, quantified assumptions about the system’s security. Probabilistic transitions and the reward function can be used to model the varying resilience of different elements against attacks and the capabilities of an attacker. This information is then used to determine a plausible data exfiltration strategy. An example with multiple scenarios illustrates the workflow. A discussion of supplementary techniques like hierarchical learning and deep neural networks concludes this paper.https://ojs.cvut.cz/ojs/index.php/ap/article/view/5030fpga, it security, model-driven design, reinforcement learning, machine learning.
collection DOAJ
language English
format Article
sources DOAJ
author Michael Vetter
spellingShingle Michael Vetter
MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING
Acta Polytechnica
fpga, it security, model-driven design, reinforcement learning, machine learning.
author_facet Michael Vetter
author_sort Michael Vetter
title MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING
title_short MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING
title_full MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING
title_fullStr MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING
title_full_unstemmed MODEL-BASED SECURITY ANALYSIS OF FPGA DESIGNS THROUGH REINFORCEMENT LEARNING
title_sort model-based security analysis of fpga designs through reinforcement learning
publisher CTU Central Library
series Acta Polytechnica
issn 1210-2709
1805-2363
publishDate 2019-11-01
description Finding potential security weaknesses in any complex IT system is an important and often challenging task best started in the early stages of the development process. We present a method that transforms this task for FPGA designs into a reinforcement learning (RL) problem. This paper introduces a method to generate a Markov Decision Process based RL model from a formal, high-level system description (formulated in the domain-specific language) of the system under review and different, quantified assumptions about the system’s security. Probabilistic transitions and the reward function can be used to model the varying resilience of different elements against attacks and the capabilities of an attacker. This information is then used to determine a plausible data exfiltration strategy. An example with multiple scenarios illustrates the workflow. A discussion of supplementary techniques like hierarchical learning and deep neural networks concludes this paper.
topic fpga, it security, model-driven design, reinforcement learning, machine learning.
url https://ojs.cvut.cz/ojs/index.php/ap/article/view/5030
work_keys_str_mv AT michaelvetter modelbasedsecurityanalysisoffpgadesignsthroughreinforcementlearning
_version_ 1725878444169887744

Similar Items