Enhanced Security of Digital FIR Filters Using Dynamic and Multi-layer Obfuscation Technique

Abstract

Recent advancements in digital signal processing have highlighted the critical need for robust protection mechanisms for digital Finite Impulse Response (FIR) filters against various sophisticated security threats such as hardware Trojans, differential power analysis (DPA), and fault injection attacks. This research introduces an enhanced FIR filter design that incorporates novel dynamic and multi-layer obfuscation techniques to improve security. The proposed model is benchmarked against a baseline FIR filter model to evaluate its effectiveness. Comprehensive simulations demonstrate that the proposed obfuscation techniques significantly enhance the security and robustness of the FIR filters without compromising performance. Key metrics such as Signal-to-Noise Ratio (SNR), filter response accuracy, and error rates are analyzed to illustrate the superior performance of the proposed model. The results indicate a notable improvement in security metrics, including increased resistance to SAT-based and machine learning-based attacks, showcasing the proposed model as a viable solution for secure digital signal processing applications. This study contributes to the field by providing a detailed analysis of the security enhancements and demonstrating the practical feasibility of implementing dynamic and multi-layer obfuscation techniques in digital FIR filters.