MacLeR: Machine Learning-Based Runtime Hardware Trojan Detection in Resource-Constrained IoT Edge Devices

Faiq Khalid, Syed Rafay Hasan, Sara Zia, Osman Hasan, Falah Awwad, Muhammad Shafique

Introduction

The globalization of the semiconductor supply chain has introduced significant security risks, notably Hardware Trojans (HTs). These malicious circuit modifications can leak sensitive data, cause system failures, or reduce device lifespan. Detecting HTs in resource-constrained IoT edge devices is particularly challenging due to the lack of "golden" reference chips and the high overhead of traditional monitoring solutions.

The MacLeR Framework

MacLeR (Machine Learning-based Run-time HT detection) is a lightweight framework designed to address these challenges. It operates on the principle that malicious activity in an untrusted Intellectual Property (IP) block will manifest as subtle power fluctuations in the trusted processor it interacts with.

The MacLeR detection pipeline, illustrating the flow from untrusted fabrication to ML-based runtime monitoring.

Fine-Grained Power Profiling

Unlike traditional methods that monitor total chip power, MacLeR analyzes power consumption at the individual microprocessor pipeline stage level (e.g., Fetch, Decode, Execute, and Store). By focusing on specific instructions, the system can detect minute "glitches" caused by Trojans that would otherwise be lost in the noise of the entire System-on-Chip (SoC).

Fine-grained power signatures and percentage changes (δP) for MOV, ADD, and INC instructions across various Trojan states (S1-S7) relative to the clean state (S0).

Low-Overhead Data Acquisition

To minimize hardware impact, MacLeR utilizes a Single Power-Port Current Acquisition Block (SP-CAB). This circuit uses current mirrors and Time-Division Multiplexing (TDM) to sample different pipeline stages sequentially through a single port, significantly reducing area and power overhead. The captured data is then processed by a lightweight Multi-Layer Perceptron (MLP) classifier with only two hidden layers.

Evaluation and Results

The framework was evaluated using a LEON3-based SoC integrated with standard HT benchmarks from Trust-Hub, including AES-T100, RS232-T1000, and EthernetMAC10GE-T700.

  • Detection Accuracy: MacLeR achieved a peak accuracy of 96.26%, outperforming existing machine learning-based techniques by approximately 10%.
  • Resource Efficiency: The hardware overhead is minimal, occupying only 0.025% of the SoC area and consuming less than 0.07% of total system power—a 7x reduction compared to prior methods.
  • Environmental Robustness: The system maintains high accuracy despite manufacturing process variations and circuit aging. Even under a simulated 10-year aging scenario, the accuracy drop was nearly 10 times more resilient than competing models.

Conclusion

MacLeR provides a robust, low-overhead solution for real-time Hardware Trojan detection in IoT devices. By leveraging fine-grained pipeline-level power signatures and efficient ML models, it offers a practical defense against supply chain vulnerabilities without compromising device performance or battery life.

Summary updated

Abstract

Traditional learning-based approaches for run-time Hardware Trojan detection require complex and expensive on-chip data acquisition frameworks and thus incur high area and power overhead. To address these challenges, we propose to leverage the power correlation between the executing instructions of a microprocessor to establish a machine learning-based run-time Hardware Trojan (HT) detection framework, called MacLeR. To reduce the overhead of data acquisition, we propose a single power-port current acquisition block using current sensors in time-division multiplexing, which increases accuracy while incurring reduced area overhead. We have implemented a practical solution by analyzing multiple HT benchmarks inserted in the RTL of a system-on-chip (SoC) consisting of four LEON3 processors integrated with other IPs like vga_lcd, RSA, AES, Ethernet, and memory controllers. Our experimental results show that compared to state-of-the-art HT detection techniques, MacLeR achieves 10\% better HT detection accuracy (i.e., 96.256%) while incurring a 7x reduction in area and power overhead (i.e., 0.025% of the area of the SoC and <0.07% of the power of the SoC). In addition, we also analyze the impact of process variation and aging on the extracted power profiles and the HT detection accuracy of MacLeR. Our analysis shows that variations in fine-grained power profiles due to the HTs are significantly higher compared to the variations in fine-grained power profiles caused by the process variations (PV) and aging effects. Moreover, our analysis demonstrates that, on average, the HT detection accuracy drop in MacLeR is less than 1% and 9% when considering only PV and PV with worst-case aging, respectively, which is ~10x less than in the case of the state-of-the-art ML-based HT detection technique.
BibTeX
@article{khalid2020macler,
  author = {Faiq Khalid and Syed Rafay Hasan and Sara Zia and Osman Hasan and Falah Awwad and Muhammad Shafique},
  title = {{MacLeR: Machine Learning-Based Runtime Hardware Trojan Detection in Resource-Constrained IoT Edge Devices}},
  journal = {IEEE Trans. Comput. Aided Des. Integr. Circuits Syst.},
  year = {2020},
  doi = {10.1109/tcad.2020.3012236},
}
Figure