HSAS-MD Analyzer: A Hybrid Security Analysis System Using Model-Checking Technique and Deep Learning for Malware...

Area of Science:

• Cybersecurity
• Software Engineering
• Artificial Intelligence

Background:

• Established Internet of Things (IoT) platforms lack robust security analysis for applications.
• Existing security analysis systems (SAS) often rely on static or dynamic analysis individually, limiting accuracy.
• A hybrid approach combining static and dynamic analysis is crucial for comprehensive IoT app security.

Purpose of the Study:

• To propose a novel hybrid security analysis system (SAS) for Internet of Things (IoT) applications.
• To develop an analyzer that integrates static and dynamic analysis techniques with model-checking and deep learning.
• To enhance the detection of malware and verification of behavior in IoT applications.

Main Methods:

• Developed a hybrid (static and dynamic) SAS named HSAS-MD analyzer.
• Utilized model-checking for converting source code and verifying app behavior.
• Employed deep learning, specifically a CNN algorithm, for feature extraction and classification of IoT app data.

Main Results:

• The HSAS-MD analyzer demonstrated superior performance in detecting malware in IoT applications compared to existing SASs.
• Achieved high performance metrics: 95% accuracy, 94% precision, 91% recall, and 93% F-measure.
• The system provided effective results across various analysis criteria, outperforming other analyzers.

Conclusions:

• The proposed HSAS-MD analyzer offers a holistic approach to IoT application security analysis.
• Hybrid analysis, incorporating model-checking and deep learning, significantly improves malware detection and behavior verification.
• HSAS-MD represents a significant advancement in securing the Internet of Things ecosystem.