J 2024

Multistage Malware Detection Method for Backup Systems

NOVÁK, Pavel, Václav OUJEZSKÝ, Patrik KAURA, Tomáš HORVÁTH, Martin HOLÍK et. al.

Basic information

Original name

Multistage Malware Detection Method for Backup Systems

Authors

NOVÁK, Pavel (203 Czech Republic, belonging to the institution), Václav OUJEZSKÝ (203 Czech Republic, guarantor, belonging to the institution), Patrik KAURA (203 Czech Republic, belonging to the institution), Tomáš HORVÁTH (203 Czech Republic, belonging to the institution) and Martin HOLÍK (203 Czech Republic, belonging to the institution)

Edition

TECHNOLOGIES, SWITZERLAND, MDPI, 2024, 2227-7080

Other information

Language

English

Type of outcome

Článek v odborném periodiku

Field of Study

20203 Telecommunications

Country of publisher

Switzerland

Confidentiality degree

není předmětem státního či obchodního tajemství

References:

Impact factor

Impact factor: 3.600 in 2022

Organization unit

Faculty of Informatics

UT WoS

001172262100001

Keywords in English

backup; detection; hashes; malware; model; machine learning; system

Tags

International impact, Reviewed
Změněno: 29/5/2024 14:25, doc. Ing. Václav Oujezský, Ph.D.

Abstract

V originále

This paper proposes an innovative solution to address the challenge of detecting latent malware in backup systems. The proposed detection system utilizes a multifaceted approach that combines similarity analysis with machine learning algorithms to improve malware detection. The results demonstrate the potential of advanced similarity search techniques, powered by the Faiss model, in strengthening malware discovery within system backups and network traffic. Implementing these techniques will lead to more resilient cybersecurity practices, protecting essential systems from hidden malware threats. This paper’s findings underscore the potential of advanced similarity search techniques to enhance malware discovery in system backups and network traffic, and the implications of implementing these techniques include more resilient cybersecurity practices and protecting essential systems from malicious threats hidden within backup archives and network data. The integration of AI methods improves the system’s efficiency and speed, making the proposed system more practical for real-world cybersecurity. This paper’s contribution is a novel and comprehensive solution designed to detect latent malware in backups, preventing the backup of compromised systems. The system comprises multiple analytical components, including a system file change detector, an agent to monitor network traffic, and a firewall, all integrated into a central decision-making unit. The current progress of the research and future steps are discussed, highlighting the contributions of this project and potential enhancements to improve cybersecurity practices.

Links

VK01030030, research and development project
Name: Systém pro zálohování a ukládání dat s integrovanou aktivní ochranou proti kybernetickým hrozbám
Investor: Ministry of the Interior of the CR, Data backup and storage system with integrated active protection against cyber threats