Related Experiment Video
Updated: Jun 28, 2025
09:47
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
1.0K
An Adaptive Temporal Convolutional Network Autoencoder for Malicious Data Detection in Mobile Crowd Sensing
Nsikak Owoh1, Jackie Riley1, Moses Ashawa1
1Department of Cyber Security and Networks, School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Cowcaddens Road, Glasgow G4 0BA, UK.
Sensors (Basel, Switzerland)
|April 13, 2024
Summary
This study introduces an adaptive model to detect malicious data in mobile crowdsensing (MCS) systems. The TCN-based model achieves 98% accuracy in identifying and mitigating threats to ensure data integrity.
Area of Science:
- Computer Science
- Cybersecurity
- Data Science
Background:
- Mobile crowdsensing (MCS) systems collect data from mobile devices, posing security risks due to potential data poisoning.
- Vulnerabilities in MCS systems can compromise data integrity and reliability.
- Existing detection methods may struggle against evolving adversarial tactics.
Purpose of the Study:
- To propose an adaptive and robust model for detecting malicious sensor data in MCS.
- To enhance the security and trustworthiness of mobile crowdsensing systems.
- To mitigate the impact of adversarial attacks on collected data.
Main Methods:
- Developed a Temporal Convolutional Network (TCN)-based model with an adaptive learning mechanism.
- Incorporated continuous evolution to detect novel malicious data patterns.
- Evaluated performance using the SherLock datasets for comprehensive analysis.
Main Results:
- The proposed TCN-based model demonstrated high effectiveness in detecting malicious sensor data.
- Achieved a detection accuracy score of 98% in performance evaluations.
- Successfully mitigated potential threats to MCS system integrity.
Conclusions:
- The adaptive TCN model significantly enhances the security of mobile crowdsensing systems.
- The model's ability to adapt to evolving threats ensures robust data integrity.
- This research contributes to developing more reliable and secure crowdsensing applications.
