Related Experiment Video
Updated: Aug 1, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
3.9K
A Dynamic Trust-Related Attack Detection Model for IoT Devices and Services Based on the Deep Long Short-Term Memory
Yara Alghofaili1, Murad A Rassam1,2
1Department of Information Technology, College of Computer, Qassim University, Qassim 51452, Saudi Arabia.
Sensors (Basel, Switzerland)
|April 28, 2023
Summary
This study introduces a deep learning model using long short-term memory (LSTM) to detect trust-related attacks in Internet of Things (IoT) services. The model effectively identifies and isolates untrusted entities, enhancing IoT security.
Area of Science:
- Cybersecurity
- Network Security
- Artificial Intelligence
Background:
- The proliferation of cloud and Internet of Things (IoT) technologies drives innovation in smart applications but also introduces severe security threats.
- Trust-based attacks exploit IoT vulnerabilities and inherent complexities like heterogeneity and dynamic behavior, hindering adoption.
- Existing trust management solutions struggle with large data volumes and evolving attack patterns.
Purpose of the Study:
- To propose a dynamic model for detecting trust-related attacks in IoT devices and services.
- To enhance the security and reliability of IoT ecosystems through effective trust management.
- To identify and isolate untrusted entities within IoT services.
Main Methods:
- Development of a dynamic trust-related attack detection model.
- Utilizing the deep long short-term memory (LSTM) technique for pattern recognition in IoT data.
- Evaluation of the model's effectiveness using diverse data samples of varying sizes.
Main Results:
- The model achieved 99.87% accuracy and 99.76% F-measure in normal conditions.
- Demonstrated high efficacy in detecting trust-related attacks, with 99.28% accuracy and 99.28% F-measure.
- Successfully identified and isolated untrusted entities in evaluated scenarios.
Conclusions:
- The proposed LSTM-based model offers a robust solution for dynamic trust-related attack detection in IoT.
- The model significantly improves IoT security by accurately identifying and mitigating threats.
- This approach is crucial for ensuring the long-term viability and trustworthiness of IoT applications.
