Software watermarking method based on collatz conjecture

Abstract

The invention relates to a software watermarking method based on collatz conjecture. The method comprises a software watermark embedding and extracting process. In the embedding process, a part of conditional branch structures are selected as embedding points by analyzing an execution path of a host program; and applying control flow confusion based on collatz conjecture at each embedding point, constructing a control function to guide the confusion component to express the embedded watermark information under a specific condition, and finally compiling the transformed host source code into aprogram instance carrying a watermark. The extraction process inputs a predetermined secret parameter configuration to a detected program and monitors an execution flow of the detected program, identifies a special condition jump mode generated by controlling the mixing assembly in the process, and reduces watermark information according to the definition of the collatz conjecture. Compared with the prior art, the method is highly fused with code obfuscation transformation, a control flow structure of a host program is used as a watermark carrier, and the method has the advantages of being high in data rate, concealment and resistance.

Description

technical field [0001] The invention relates to software watermark technology. More specifically, the present invention relates to a method for embedding a software watermark in the form of a special dynamic behavior generated by a control flow obfuscation structure by utilizing the unique control flow pattern of the Collats function in the Collats conjecture and later when needed A new software watermarking method to identify them. Background technique [0002] In the Internet age, as the distribution of digital products becomes easier, the status quo of related copyright protection is not optimistic. As the market share of software products increases, the urgency of software copyright protection becomes increasingly prominent. According to the report [1], as of 2017, 37% of the running software in the world has not been properly authorized, and the corresponding commercial value is as high as 46.3 billion US dollars. Software watermarking technology is an important part...

AI Technical Summary

Problems solved by technology

[0005] (1) Existing software watermarking methods such as literature [13, 14, 15, 18], etc., often expose exploitable weaknesses in the watermark payload code or identification process, allowing attackers to use these weaknesses to find the embedding position of the watermark , or make the watermark unrecognizable by tampering with the host software

[0006] (2) Existing software watermarking methods such as literature [5, 6, 12, 16, 17, 20] cannot be closely integrated with the host program, and the structure of the watermark payload cannot make full use of the existing content in the host program. On the one hand, the volume of the program embedded with the software watermark becomes significantly larger; on the other hand, there is a clear boundary between the software watermark and the carrier program

Another example is the CT algorithm used in literature [12, 17, 20], the generation process of the graph data structure used to hide the software watermark and the normal process of the program lack the necessary connection, and it is easy to be suspected by reverse attackers.

[0007] (3) Existing software watermarking algorithms can only cope with certain specific attacks, and lack immunity to many normal program deformations, such as optimization, obfuscation, compression and encryption, etc. to 11] etc. are particularly prominent

[0008] (4) There are design flaws in the watermark recognition and extraction process in a considerable number of software watermarking methods, so that attackers can use these flaws to invalidate the watermark directly by tampering with the host program embedded with the watermark

[0009] (5) Many software watermarking algorithms such as literature [12, 14, 17, 23] have low efficiency and cannot embed large information

Images