Obfuscation-Resistant Hardware Malware Detection: A Stacked Denoising Autoencoder Approach

The increasing reliance on Machine Learning (ML) for malware detection has enhanced security across various computing environments. However, these models remain vulnerable to adversarial manipulations such as code obfuscation, which alters program structure and execution patterns to evade detection while maintaining malicious functionality. This challenge is particularly critical in Hardware Malware Detection (HMD), where obfuscation can alter execution behavior, affecting performance counter-based features and weakening ML classifiers; yet it remains largely overlooked. In this paper, we examine the impact of code obfuscation techniques on hardware malware detection, reducing the effectiveness of ML-based security solutions. Through diverse obfuscated malware variants and an extensive evaluation across multiple ML models, we demonstrate a substantial reduction in detection performance, highlighting the evasive impact of code obfuscation in HMD systems. In response, we propose ObfusGate, an intelligent obfuscation-resistant malware detection framework that leverages feature representation learning with a stacked denoising autoencoder to enhance ML models' performance against both obfuscated and unobfuscated malware. Experimental results show that ObfusGate improves detection rates by up to 24% across various ML models, reinforcing the resilience of malware detection systems in real-time execution environments against adversarial obfuscation techniques.