Search papers, labs, and topics across Lattice.
This paper introduces CFD-HAR, a feature disentanglement technique for user-controllable privacy in Human Activity Recognition (HAR) using IMU sensor data. CFD-HAR separates activity and sensitive attributes in the latent space, providing tunable privacy protection. The authors compare CFD-HAR against autoencoder-based few-shot HAR, analyzing their privacy guarantees, data efficiency, and security implications in continual IoT settings, finding that neither approach fully satisfies next-generation IoT HAR requirements.
Achieve user-controlled privacy in human activity recognition by disentangling sensitive attributes from activity features in sensor data.
Modern wearable and mobile devices are equipped with inertial measurement units (IMUs). Human Activity Recognition (HAR) applications running on such devices use machine-learning-based, data-driven techniques that leverage such sensor data. However, sensor-data-driven HAR deployments face two critical challenges: protecting sensitive user information embedded in sensor data in accordance with users'privacy preferences and maintaining high recognition performance with limited labeled samples. This paper proposes a technique for user-controllable privacy through feature disentanglement-based representation learning at the granular level for dynamic privacy filtering. We also compare the efficacy of our technique against few-shot HAR using autoencoder-based representation learning. We analyze their architectural designs, learning objectives, privacy guarantees, data efficiency, and suitability for edge Internet of Things (IoT) deployment. Our study shows that CFD-based HAR provides explicit, tunable privacy protection controls by separating activity and sensitive attributes in the latent space, whereas autoencoder-based few-shot HAR offers superior label efficiency and lightweight adaptability but lacks inherent privacy safeguards. We further examine the security implications of both approaches in continual IoT settings, highlighting differences in susceptibility to representation leakage and embedding-level attacks. The analysis reveals that neither paradigm alone fully satisfies the emerging requirements of next-generation IoT HAR systems. We conclude by outlining research directions toward unified frameworks that jointly optimize privacy preservation, few-shot adaptability, and robustness for trustworthy IoT intelligence.
