FPGA Implementation of Elliptic Curve Digital Signature Algorithm Based on PUFs
Hao Zhang, Xuankai Liu, Bing Li · 2024
In recent years, with the explosive growth of information, incidents threatening information security such as significant information leaks and personal privacy attacks have been constantly emerging. Research on public key cryptography has become an effective means to ensure security. Elliptic curve cryptography, due to its ability to provide higher security and lower computational complexity, has gradually become the mainstream asymmetric encryption method in public key encryption systems. Its corresponding Elliptic Curve Digital Signature Algorithm (ECDSA) has also gradually become a mainstream digital signature scheme. However, ECDSA often faces issues such as private key leakage and reuse of random numbers, leading to security risks. Physical Unclonable Functions (PUFs) with real-time reading, unpredictability, and non-clonability characteristics can effectively address the security issues caused by private key storage. This paper proposes a True Random Number Generator (TRNG) based on PUFs to address the security issues of ECDSA. The TRNG has passed NIST tests. The system adopts the spec256k1 curve, and experiments show that the entire system consumes 6.33ms for a single signature operation and 6.43ms for a single verification operation at a clock frequency of 100MHz. The entire system is implemented on Xilinx Zynq-7000 FPGA.