[en] Human Activity Recognition (HAR) has seen remarkable advances in recent years, drivenby the widespread use of wearable devices and the increasing demand for personalized healthcare andactivity tracking. Federated Learning (FL) is a promising paradigm for HAR that enables the collaborativetraining of machine learning models on decentralized devices while preserving data privacy. It improvesnot only data privacy but also training efficiency as it utilizes the computing power and data of potentiallymillions of smart devices for parallel training. In addition, it helps end-user devices avoid sending users’private data to the cloud, eliminates the need for a network connection, and saves the latency of back-and-forth communication. FL also offers significant advantages for communication by reducing the amount ofdata transmitted over the network, alleviating network congestion and reducing communication costs. Bydistributing the training process across devices, FL minimizes the need for centralized data storage andprocessing, leading to more scalable and resilient systems. This paper provides a comprehensive surveyof the integration of FL into HAR applications. Unlike existing reviews, this paper uniquely focuses onthe intersection of FL and HAR, providing an in-depth analysis of recent advances and their practicalimplications. We explore key advances in FL-based HAR methodologies, including model architectures,optimization techniques, and different applications. Furthermore, we highlight the major challenges andfuture research questions in this domain, such as model personalization and robustness, privacy concerns,concept drift, and the limited capacity of edge devices.
