Secure hardware IP of GLRT cascade using color interval graph based embedded fingerprint for ECG detector

Area of Science:

• Biomedical Engineering
• Hardware Security
• Digital Design

Background:

• Electrocardiogram (ECG) detectors rely on Generalized Likelihood Ratio Test (GLRT) hardware for accurate QRS complex detection.
• Intellectual Property (IP) piracy of GLRT cores threatens ECG detector reliability and patient safety.
• Existing security measures for hardware IP are insufficient against sophisticated threats.

Purpose of the Study:

• To propose a security-aware design methodology for GLRT hardware IP cores in ECG detectors.
• To protect against IP piracy and fraudulent ownership claims.
• To ensure the integrity and reliability of ECG data analysis.

Main Methods:

• Developed an integrated design flow using High-Level Synthesis (HLS) for securing GLRT hardware IP.
• Implemented a Colored Interval Graph (CIG) framework for fingerprint biometric integration.
• Designed secure Register Transfer Level (RTL) datapath with embedded vendor fingerprint.

Main Results:

• Achieved superior tamper tolerance and a significantly lower probability of coincidence compared to other security methods.
• Embedded fingerprint biometric successfully secured the GLRT hardware IP core.
• Demonstrated zero design cost overhead for the proposed security approach.

Conclusions:

• The proposed methodology effectively secures GLRT hardware IP for ECG detectors against piracy.
• Fingerprint biometric integration provides robust protection for critical medical device hardware.
• The secure design ensures reliable ECG data estimation, enhancing patient safety without additional cost.