Binary Code Similarity Detection: State and Future

Binary code similarity detection (BCSD) is an important research direction in the field of computer network security. The application scenarios for BCSD are widespread and many related methods have been proposed to solve the problems of eliminating code redundancy, vulnerability mining, malware analysis, and code copyright protection. Based on the absence of a comprehensive overview of the BCSD method, the present article provides a review of the state of art on BCSD research. Firstly, this paper introduces the relevant concepts, general process, and challenges faced by BCSD, such as the compiler optimization diversity challenge. Secondly, based on different feature information and method principles, this paper classifies the current BCSD method into three categories: text-based, logic-based, and semantics-based methods. Then, the implementation procedures of multiple BCSD methods are described. Finally, this paper proposes prospects for the technology and application of BCSD.