SDTP-VA: An AI-Resistant Visualization and Secure Data Transmission Framework for Wearable Consumer IoT in Sports Training

Wearable consumer electronics are transforming sports performance training by allowing real-time physiological monitoring, analysis, and feedback. Along with the increasing IoT integration of wearable devices comes a serious concern regarding the security and privacy of sensitive biometric data from...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on consumer electronics p. 1
Main Authors: Sun, Youzhi, Wu, TianCi, Li, Zixu, Wu, Xiangzhi
Format: Journal Article
Language:English
Published: IEEE 2025
Subjects:
Artificial intelligence
Artificial Intelligence (AI)
Biomedical monitoring
Consumer Electronics
Data visualization
Internet of Things
Monitoring
Real-time systems
Sensors
Sports
Sports Training
Training
Visualization
Wearable devices
ISSN:0098-3063, 1558-4127
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Wearable consumer electronics are transforming sports performance training by allowing real-time physiological monitoring, analysis, and feedback. Along with the increasing IoT integration of wearable devices comes a serious concern regarding the security and privacy of sensitive biometric data from AI-driven cyberattacks. The purpose of this work is to establish a novel framework that integrates Secure Data Transmission Protocols with Visualization Approaches (SDTP-VA) that is resistant to AI-driven adversaries. This work aims to improve both the security and interpretability of wearable IoT systems deployed in sports environments. The proposed system incorporates lightweight cryptographic algorithms designed specifically for resource-constrained wearable devices to protect the integrity and confidentiality of data while maintaining device performance. Our organization has applied visualization techniques to provide an AI-resistant transform that convey raw data in tamper-proof, robust representations to protect ourselves from adversarial attacks. Distinct pattern representations allow for human-in-the-loop validation and disrupt the automated AI systems from subtly exploiting the data for malign inference. Both of these goals could be achieved. The framework is tested across various sport contexts using wearable sensors that measure mobility, heart rate, and muscle activation. This is achieved through wireless technology. Furthermore, scaling efficiencies decreased by 95.81%, while reaction times decreased by 96.88%, resulting in significant gains. The experimental results indicate a strong amount of resistance against spoofing and data injection attacks, suggesting the ability to achieve secure data interpretation with accuracy above 95% while achieving real-time performance characteristics.
ISSN:0098-3063
1558-4127
DOI:10.1109/TCE.2025.3632385