Private Data Leakage in Federated Human Activity Recognition for Wearable Healthcare Devices

TL;DR

This paper investigates privacy leakage risks in federated human activity recognition models for wearable devices, revealing high membership inference attack success rates and highlighting significant privacy concerns.

Contribution

It introduces a federated learning architecture for HAR on wearable devices and demonstrates privacy vulnerabilities through membership inference attacks.

Findings

Membership inference attack accuracy reaches 92%

Federated HAR models exhibit substantial privacy risks

Study provides new insights into privacy leakage in wearable federated learning

Abstract

Wearable data serves various health monitoring purposes, such as determining activity states based on user behavior and providing tailored exercise recommendations. However, the individual data perception and computational capabilities of wearable devices are limited, often necessitating the joint training of models across multiple devices. Federated Human Activity Recognition (HAR) presents a viable research avenue, allowing for global model training without the need to upload users' local activity data. Nonetheless, recent studies have revealed significant privacy concerns persisting within federated learning frameworks. To address this gap, we focus on investigating privacy leakage issues within federated user behavior recognition modeling across multiple wearable devices. Our proposed system entails a federated learning architecture comprising $N$ wearable device users and a parameter server, which may exhibit curiosity in extracting sensitive user information from model parameters. Consequently, we consider a membership inference attack based on a malicious server, leveraging differences in model generalization across client data. Experimentation conducted on five publicly available HAR datasets demonstrates an accuracy rate of 92\% for malicious server-based membership inference. Our study provides preliminary evidence of substantial privacy risks associated with federated training across multiple wearable devices, offering a novel research perspective within this domain.