Problematic Issues of a Program Representations Genetic De-evolution for Search Vulnerabilities and Recommendations for Its Resolution
Израилов Константин Евгеньевич · Proceedings of Telecommunication Universities · 2025
The relevance of the topic is justified by the software reverse engineering methodology lack required to resolve the following scientific contradiction in the field (as a contrast between need and opportunity): on the one hand, vulnerability search is most effective in those program representations in which they were implemented (e.g. source code, algorithms or architecture); on the other hand, as a rule, only machine code is available for analysis, which is poorly suited for identifying high-level vulnerabilities (i.e. from earlier representations). The main author's study, the final stage of which is given in the article, is devoted to the creation of the this methodology elements (concept, model, method, algorithms, metrics, as well as their implementations). The purpose of this article is to discuss 25 problematic issues (the so-called scientific discussion) that arose in the main study devoted to the software reverse engineering development based on genetic algorithms. The main application of the research results is both obtaining a program representation suitable for expert (and other) analysis for vulnerabilities, and their direct search using the built-in signature method. At the same time, resolving even a part of the issues will significantly increase the efficiency of such genetic reverse engineering. The following methods were used in the work: the main research results analysis to identify problematic issues, ways to resolve them synthesis, as well as issues systematization and scoring from the standpoint of ways to eliminate them for an overall assessment of the scientific work completeness. A causes of each issue detailed research allowed us to determine ways to resolve them, the feasibility of which also justifies the main research results. In particular, problematic issues are based both on some absence of theoretical tools necessary for genetic reverse engineering, and on the insufficient practical efficiency of others. The scientific novelty of the issues lies in the fact that almost each of them is voiced for the first time. The theoretical significance lies in the fact that the development of each problematic issue can both open a separate scientific study (or even a direction), and obtain new significant results. The practical significance lies in the possibility of creating software solutions to resolve identified issues, which can also be applied to related tasks.