A Zero Chip Area Overhead Mapping Core Watermarking Algorithm

Abstract

The traditional IP core watermarking embedding algorithm usually completes the watermarking identification by writing the watermarking information to the unused hardware resources. When the watermarking information to be embedded is large, it may consume significant chip area. In this paper, we use the theory of multiplicative group of integers modulo n to establish a dual-projection security model. Then we propose an IP-core watermarking algorithm based on mapping rather than writing. The watermarking embedding is realized by establishing one-to-one mapping between the watermarking information and the original circuit information. The algorithm is immune to both removal and tampering attacks. Experiments show that the false alarm rate of the algorithm is superior to other algorithms.