More Journal Articles
A Novel Federated Learning Approach for Enhancing QoS and Scalability in IoT Networks
Jamal Et-Tousy, Abdellah Zyane
Cluster Computing, Vol. 29 (14)
This paper proposes FL-OneM2M, a federated learning framework integrated with the oneM2M standard to provide decentralized, scalable, privacy-preserving, and QoS-aware intelligence for IoT networks. The approach evaluates federated aggregation strategies, including FedAvg and FedProx, across distributed IoT clients while analyzing communication performance using HTTP and MQTT. Experimental results show that FedAvg achieves 80% accuracy, while MQTT reduces RTT by 98.4% compared with HTTP. The framework also demonstrates robust performance under client dropout, highlighting its suitability for scalable and dynamic IoT environments.
Adaptive QoS Management in oneM2M Standard: Machine Learning and Deep Learning for IoT Network Optimization
J. ET-TOUSY, A. ZYANE, S. SHARIF
Journal of Network and Systems Management, Vol. 34 (3)
This study proposes an adaptive quality-of-service management approach based on machine learning and deep learning to optimize performance in oneM2M-based IoT networks.
Comprehensive Analysis, Modeling, and Optimization of QoS in oneM2M for IoT Applications
J. ET-TOUSY, A. ZYANE
International Journal of Communication Systems, Vol. 38 (14)
This article presents a comprehensive analysis and modeling framework for evaluating and optimizing quality of service in oneM2M-based Internet of Things applications.
Cost and QoS Analysis in IoT: An Optimization Approach Based on the oneM2M Standard
J. ET-TOUSY, S. ABOURRICHE, A. ZYANE
Engineering Proceedings, Vol. 112 (1)
This paper analyzes the relationship between deployment cost and quality of service in IoT environments and proposes an optimization approach based on the oneM2M standard.