Enhancing Human Activity Recognition With FedPA: Focusing on Non-IID Data Challenges in Federated Learning
- 16 October 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (23) , 39230-39242
Abstract
Federated Learning (FL) revolutionizes distributed learning in Human Activity Recognition (HAR) by allowing clients to train models locally and share only model parameters, thus optimizing data usage and mitigating privacy concerns. However, the presence of Non-Independent and Identically Distributed (Non-IID) data across clients impedes FL efficiency between the server and clients, affecting HAR performance and communication efficiency. Acknowledging the significance of post-training parameters as intrinsic representations of knowledge acquired by client models, we introduce FedPA (Federated Parameters Averaging), a novel algorithm for model aggregation. This algorithm strategically assigns varying weight coefficients to clients during aggregation, accurately reflecting each client’s learning contribution and thereby enhancing the efficacy of both global and client models. We conducted empirical analyses using real-world datasets to assess FedPA’s effectiveness. The results demonstrate that FedPA not only preserves accuracy but also improves communication efficiency compared to existing FL aggregation algorithms, like FedAvg, FedMA, and FedCDA. These findings underscore FedPA’s superiority in addressing Non-IID challenges in HAR tasks and highlight its potential to improve overall model performance in FL settings.Keywords
Funding Information
- Henan Province Key Research and Development Project (241111312000, 231111221600)
- Backbone Teacher Support Program (GG202414)
- Henan Provincial Foreign Experts Programs (GZS2022012)
- Henan Province Key Technologies Research Development Program (242102211066)
This publication has 30 references indexed in Scilit:
- Federated PAC-Bayesian Learning on Non-IID Data2024
- SRFL: A Secure & Robust Federated Learning framework for IoT with trusted execution environmentsExpert Systems with Applications, 2023
- VPPFL: A verifiable privacy-preserving federated learning scheme against poisoning attacksComputers & Security, 2023
- Human Activity Recognition (HAR) Using Deep Learning: Review, Methodologies, Progress and Future Research DirectionsArchives of Computational Methods in Engineering, 2023
- A triboelectric gait sensor system for human activity recognition and user identificationNano Energy, 2023
- Federated machine learning for privacy preserving, collective supply chain risk predictionInternational Journal of Production Research, 2023
- A federated learning system with enhanced feature extraction for human activity recognitionKnowledge-Based Systems, 2021
- Illuminating the dark spaces of healthcare with ambient intelligenceNature, 2020
- Security and Privacy in the Industrial Internet of Things: Current Standards and Future ChallengesIEEE Access, 2020
- Human Activity Recognition Using Federated Learning2018