Modeling and security analysis of hardware trojan attack based on GSCPN

Addressing the issue that research on hardware Trojan (HT) has predominantly focused on specific strategies of HT design and defense, with a deficiency in quantitative evaluation models for system security in the context of HT attacks, a HT attack and security assessment model based on the generaliz...

Full description

Saved in:
Bibliographic Details
Main Authors: MEI Bo, GUO Wei, CAO Zhipeng, LI Peijie, ZHANG Wenbo
Format: Article
Language:English
Published: POSTS&TELECOM PRESS Co., LTD 2024-08-01
Series:网络与信息安全学报
Subjects:
Online Access:http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2024054
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Addressing the issue that research on hardware Trojan (HT) has predominantly focused on specific strategies of HT design and defense, with a deficiency in quantitative evaluation models for system security in the context of HT attacks, a HT attack and security assessment model based on the generalized stochastic coloring Petri net (GSCPN) was proposed. The GSCPN model representing the HT attack of a single basic unit was constructed, founded on the temporal relationship between HT attack behaviors and the states of a single basic unit. A general method for constructing the GSCPN model of HT attack for the entire system was provided, based on the communication dependency relationships between the basic units during the system’s operational state. The analytical solution of the Markov chain isomorphous to the GSCPN model was derived, alongside the simulation solution achieved using the Monte Carlo simulation method. The discrepancy between the two methods was a mere 0.00183, which verified the precision of the modeling approach. The security of the system under various scenarios was analyzed through a straightforward case study. The experimental results indicate that the intensity of the HT attack, HT recovery time, and task execution time all exert varying degrees of influence on system security.
ISSN:2096-109X