Research On Watermarking And Blind Detection For IP Copyright Protection

In recent years,integrated circuits filed has been develop rapidly with the rise of mobile intelligent terminals and Internet of Things technology,the field programmable logic gate(FPGA)and application specific integrated circuit(ASIC)has been recognized by many enterprises.In order to reduce the costs and risks,shorten IC design cycle,IP core multiplexing technology is widely used.However,it brings convenience in IC design,at the same time it brings new problems.After IP core enter into the market,it is likely to face other manufactures illegal theft and use it to do re-development.The emergence of such acts will undermine the principle of fair competition between enterprises and bring enterprise immeasurable economic losses.In recent years,how to effectively protect the copyright of IP core has aroused the attention of many research scholars and research institutions at home and abroad.Many researcher have proposed a series of solutions at present.The method which using multimedia digital watermarking ideas to protect copyright of IP core has been a hotspot in this field.Researchers' works in the field of IP watermarking technology mainly focused on how to reduce the resource costs,improve the security of watermarking and easily track.The work of this paper is mainly focused on improve IP watermarking robustness and real-time performance of blind detection:(1)In order to solve the problem that the existing watermarking technology can not be effectively authenticated after the watermark has been attacked and destroyed,this paper proposes a high polymerization mutual mapping IP watermarking algorithm.The algorithm uses data redundancy to establish an inter mapping relationship between watermarks,allowing an attacker to destroy the watermark to some extent.When the watermark information is partially destroyed,the mutual mapping can be used for self recovery,thus enhancing the robustness of the core watermark.(2)In order to solve the problem that low real-time performance of existing IP watermarking blind detection scheme,this paper proposes a fast IP watermarking blind detection algorithm based on fuzzy location.By introducing the operation of fuzzy location,the algorithm breaks the reversibility of location scrambling algorithm,thus reducing the number of inquiry wheels in watermark blind detection,and effectively improving the efficiency of core watermark blind detection.The experimental results show that the high polymerization mutual mapping IP watermarking algorithm has strong ability to resist removal attacks.Moreover,compared with other core accounting methods,this method has more advantages in watermark concealment and resource cost.Through the experimental comparison with existing IP watermark blind detection algorithms,a fast IP watermarking blind detection algorithm based on fuzzy location can greatly reduce the computation which generated in inquiring process,and the real-time performance of blind detecting is better than others.