FLYFDetect: A Smart Home Privacy Protection Framework via Federated Learning
Bing Chen, Yaping Liu, Shuo Zhang, Jie Chen, Zhiyu Han · 2022
With the advent of the Internet of Things (IoT), smart home devices and applications have become more widespread, and the privacy issues of smart homes have increasingly become more important. At present, the research on privacy protection technologies for smart home is to localize data as much as possible to reduce the outflow of users’ data. However, the data cannot be processed completely locally, and the private data of smart home devices that require intelligent services will still leak to the cloud. Therefore, we propose a smart home privacy protection communication protocol PPTrans based on federated learning and design a smart home privacy protection framework FLYFDetect that supports PPTrans, which implements a federated learning method based on YOLOv5, and verified the effectiveness of FLYFDetect in a real environment combined with a smart home flame warning application.