Search for: All records

Physical Unclonable Functions (PUFs) are widely researched in the field of security because of their unique, robust, and reliable nature, PUFs are considered device-specific root keys that are hard to duplicate. There are many variants of PUFs that are being studied and implemented including hardware and software PUFs. Though PUFs are believed to be secure and reliable, they are not without challenges of their own. The efficient performance of PUF depends on various environmental factors, which leads to inefficiency. Bit flipping is one such problem that can bring down the reliability of the PUF. Memory-based PUFs are prone to unavoidable bit flips occurring in the hardware, similarly, sensor-based PUFs are prone to bit flips occurring due to temperature variation. The number of errors in the PUF response must be minimized to improve the reliability of the PUF in security applications. In this research we explore the Machine Learning (ML) model based on K-mer sequencing to detect and correct the bit flips in the PUFs, hence fortifying the PUF-based secure authentication system for authentication and authorization of Edge Data Centers (EDC) in a Collaborative Edge Computing (CEC) Environment.