A hybrid early warning system for the prevention of mobile asset theft: case of laptops in South African government buildings

dc.contributor.authorMatsemela, Thabiso John
dc.date.accessioned2026-03-24T10:06:01Z
dc.date.issued2025
dc.descriptionMaster of Engineering in Electrical Engineering
dc.description.abstractThe growing threat of mobile asset theft in South African government establishments presents serious risks to information security and public sector performance. Mobile assets such as laptops and similar devices increasingly store sensitive and operational data, ensuring their protection has become critical. This study addresses the problem by proposing a hybrid early warning detection and prevention system designed to monitor and respond to unauthorised access and unlawful handling of mobile assets in real time. The system integrates a biometric fingerprint authentication, intelligent camera surveillance, infrared motion detection, a microcontroller-managed limit switch, smart lock, and wireless notification devices, forming a multi-layered security system. Developed using the Design Science Research Methodology (DSRM), the system was tested under simulated conditions to evaluate its reliability and responsiveness. MATLAB was used extensively for component-level simulation, while MikroC facilitated the programming of a PIC16F628A microcontroller to coordinate all system functions. A backup power supply capable of 48-hour operation was also included to ensure uninterrupted monitoring during power outages. Analyses, including chi-square testing, were applied to assess the significance of biometric authentication and system performance across two government buildings, with additional validation using theoretical scoring and multivariable performance models under varied conditions. The results showed over 97, 5% accuracy, a response time of 95 milliseconds during breach events, and consistent behaviour across test environments. The intelligent camera only activates under threat conditions, preserving user privacy, and a manual override provides operational flexibility. These findings confirm the system’s viability as a scalable, intelligent security solution for mobile asset protection in government and potentially private sector environments.
dc.description.sponsorshipSupervisor: Prof ED Markus
dc.identifier.urihttp://hdl.handle.net/11462/2805
dc.language.isoen
dc.publisherCentral University of technology
dc.subjectmobile asset theft
dc.subjectearly warning detection
dc.titleA hybrid early warning system for the prevention of mobile asset theft: case of laptops in South African government buildings
dc.typeThesis

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Matsemela TJ.pdf
Size:
23.62 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: