Виявлення RowHammer за допомогою подвійного класифікатора на основі ШІ

Purpose of this work is to show an algorithm for detecting anomalies and unauthorized changes in random access memory (RAM) using artificial intelligence methods. Originality of this approach is ensured by using access frequency analysis to detect RowHammer attacks with multiple machine learning algorithms in one detector. Proposed detection method can be used in modern information systems to increase the level of data security, in particular in systems with increased reliability requirements. Even though research is limited to the experimental environment and set of test-generated data, further research can be aimed at expanding the dataset, optimizing AI models, and analyzing the effectiveness of the proposed approach in real-world systems, as well as studying related memory attacks, in particular attacks through side channels