A new publicly verifiable blind detection scheme for intellectual property protection

Existing zero-knowledge-based intellectual property (IP) watermarking techniques require a third party to identify suspected IP ownership. However, the credibility and controllability of the third party raise a new challenge to watermark detection in field programmable gate array (FPGA) IP cores. In this work, a public verifiable scheme for watermark embedding and blind detection in FPGA IP cores is proposed. The scheme uses the key threshold of interactive zero-knowledge protocol as the blind detection evidence for secure watermark authentication. The identification information of the IP owner and the key detection threshold is public in the zero-knowledge proof protocol. The security of public watermark detection for IP watermark adopts the idea of zero knowledge in cryptography. The scrambling algorithm will be used to scramble the authentication information in FPGA IP cores. In this case, the security of public watermark detection can be guaranteed. Without the original IP circuit and the third party, watermarks in FPGA IP cores can be blindly detected on site. The experiments show that our scheme improves the credibility of the watermarks detected by trading partners in public. The performances in terms of resource overhead, robustness and security are encouraging.