Analysis of signal resampling effects on attention-driven SEI for IoT systems

The rapid growth of the Internet of Things (IoT) has connected billions of devices, many with minimal security, making them vulnerable to attacks. Specific Emitter Identification (SEI) offers a passive and reliable security solution by identifying devices through their unique hardware features, enabling serial number level distinction without altering the emitter. SEI can serve as the “something the entity is” factor in zero-trust multi-factor authentication frameworks. However, conventional SEI methods rely on high sampling rates, which are impractical for resource-limited IoT devices. This work evaluates an attention-based SEI model that maintains high identification accuracy at reduced sampling rates. The proposed approach achieves over 97% accuracy using only 2,500 signals sampled at 5 MHz and sustains above 90% accuracy under Rayleigh fading, reducing memory usage by 87.5% without compromising performance. These results highlight the potential of attention mechanisms for efficient, scalable IoT device identification.

Grantor dc:publisher

University of Tennessee at Chattanooga

Year dc:date.available

2027

Author dc:creator

• Mohamedkhir, Mohamedelfateh

Contributors dc:contributor

• Reising, Donald
• Weerasena, Lakmali; Fadul, Mohamed K. M.
• College of Engineering and Computer Science

• IEEE 802.11 (Standard)
• Internet of Things--Computer networks--Security measures
• Pattern recognition systems

Licence

• http://rightsstatements.org/vocab/InC/1.0/

Language dc:language

English, eng

Repository record dc:identifier

https://scholar.utc.edu/theses/1040

OAI identifier oai:identifier

oai:scholar.utc.edu:theses-2223