Table of Contents
Fetching ...

MATTER: Multi-stage Adaptive Thermal Trojan for Efficiency & Resilience degradation

Mehdi Elahi, Mohamed R. Elshamy, Abdel-Hameed Badawy, Mahdi Fazeli, Ahmad Patooghy

TL;DR

This work addresses the vulnerability of mobile System-on-Chips to thermal attacks that exploit Dynamic Thermal Management (DTM). It introduces MATTER, a two-stage thermal Trojan that targets the DTM temperature-sensing interface and uses Thread Biasing to reveal and exploit DTM thresholds such as $Th_{Trigger}$, $Th_{Critical}$, and $Th_{Recovery}$. The authors demonstrate that MATTER can degrade DTM performance by $73\%$, raise power and core utilization, and remain stealthy against average-temperature detectors, highlighting a gap in current defenses. The findings emphasize the need to harden DTM interfaces and develop robust detection against threshold-targeted thermal attacks, with implications for hardware security and device longevity.

Abstract

As mobile systems become more advanced, the security of System-on-Chips (SoCs) is increasingly threatened by thermal attacks. This research introduces a new attack method called the Multi-stage Adaptive Thermal Trojan for Efficiency and Resilience Degradation (MATTER). MATTER takes advantage of weaknesses in Dynamic Thermal Management (DTM) systems by manipulating temperature sensor interfaces, which leads to incorrect thermal sensing and disrupts the SoC's ability to manage heat effectively. Our experiments show that this attack can degrade DTM performance by as much as 73%, highlighting serious vulnerabilities in modern mobile devices. By exploiting the trust placed in temperature sensors, MATTER causes DTM systems to make poor decisions i.e., failing to activate cooling when needed. This not only affects how well the system works but also threatens the lifespan of the hardware. This paper provides a thorough analysis of how MATTER works and emphasizes the need for stronger thermal management systems in SoCs.

MATTER: Multi-stage Adaptive Thermal Trojan for Efficiency & Resilience degradation

TL;DR

This work addresses the vulnerability of mobile System-on-Chips to thermal attacks that exploit Dynamic Thermal Management (DTM). It introduces MATTER, a two-stage thermal Trojan that targets the DTM temperature-sensing interface and uses Thread Biasing to reveal and exploit DTM thresholds such as , , and . The authors demonstrate that MATTER can degrade DTM performance by , raise power and core utilization, and remain stealthy against average-temperature detectors, highlighting a gap in current defenses. The findings emphasize the need to harden DTM interfaces and develop robust detection against threshold-targeted thermal attacks, with implications for hardware security and device longevity.

Abstract

As mobile systems become more advanced, the security of System-on-Chips (SoCs) is increasingly threatened by thermal attacks. This research introduces a new attack method called the Multi-stage Adaptive Thermal Trojan for Efficiency and Resilience Degradation (MATTER). MATTER takes advantage of weaknesses in Dynamic Thermal Management (DTM) systems by manipulating temperature sensor interfaces, which leads to incorrect thermal sensing and disrupts the SoC's ability to manage heat effectively. Our experiments show that this attack can degrade DTM performance by as much as 73%, highlighting serious vulnerabilities in modern mobile devices. By exploiting the trust placed in temperature sensors, MATTER causes DTM systems to make poor decisions i.e., failing to activate cooling when needed. This not only affects how well the system works but also threatens the lifespan of the hardware. This paper provides a thorough analysis of how MATTER works and emphasizes the need for stronger thermal management systems in SoCs.

Paper Structure

This paper contains 12 sections, 1 equation, 6 figures, 2 tables.

Figures (6)

  • Figure 1: Illustration of Attack on the System's Interface
  • Figure 2: MATTER's behavior across two defined stages while DTM is deactivated, activated, and under attack
  • Figure 3: Number and duration of IREs with and without MATTER across different benchmarks: PARSEC (with simmedium input size), SPLASH2 (with small input size), and SPEC CPU2006 (with millions of instructions)
  • Figure 4: Increased Power Consumption Pre- and Post-Application of the MATTER Attack
  • Figure 5: Relative core resource utilization with and without exposure to the MATTER attack
  • ...and 1 more figures