Jul 2026· Italian National Conference on Sensors· Vol 26· 0 citations· 35 references
Medicine
TL;DR
Simulation results demonstrate that the proposed digital twin (DT)-based dynamic FL aggregation method not only accelerates model convergence and improves accuracy but also significantly reduces training latency and energy consumption costs compared with baseline FL algorithms.
Abstract
Federated learning (FL) enables collaborative model training without sharing raw data, but it faces challenges due to client heterogeneity, leading to inefficiency and reduced accuracy. This paper proposes a digital twin (DT)-based dynamic FL aggregation method to address these issues. The framework integrates a DT layer on the server side to perform preaggregation evaluations, simulating various aggregation strategies to select the optimal approach before actual global aggregation. An adaptive clustering method based on K-means is employed to group clients with similar characteristics, and a hierarchical aggregation evaluation strategy is designed to optimize both intra-cluster and inter-cluster aggregation, with the goal of minimizing latency and energy consumption while maximizing model accuracy. Simulation results on the MNIST and CIFAR-10 datasets demonstrate that the proposed method not only accelerates model convergence and improves accuracy but also significantly reduces training latency and energy consumption costs compared with baseline FL algorithms. This DT-assisted approach delivers a practical and effective optimization solution for federated learning deployment over large-scale heterogeneous IoT sensor networks.
Accurate energy forecasting is essential for grid stability, demand-side management, and efficient renewable integration. However, energy consumption data collected from smart meters may expose sensitive user information, thus raising privacy concerns. Federated Learning (FL) offers a privacy-preserving mechanism for collaborative model training without sharing raw data. However, conventional synchronous FL suffers from training delays caused by heterogeneous client availability and computational capabilities, while frequent exchange of model parameters can lead to communication overheads. To address these challenges, this paper proposes an asynchronous federated learning framework for energy forecasting that enables continuous global model updating without waiting for all clients to complete local training. We introduce a federated asynchronous adaptive aggregation mechanism, where client-specific learning rates are dynamically adjusted based on both update staleness and model performance contribution. A partial aggregation strategy is defined for a Long Short-Term Memory (LSTM) forecasting model that splits the local models’ layers, allowing clients to exchange only a subset of the weights with the server. The proposed solution is evaluated using real-world energy consumption data from multiple consumers. Experimental results demonstrate that the proposed asynchronous adaptive strategy outperforms the classic FedAvg approach and maintains prediction accuracy relative to personalised FedAvg, while reducing communication costs. Additionally, the proposed method outperforms the classic FedAsync algorithm across all client groups, with statistically significant improvements in most cases.
Liana Toderean, Mara Mesesan, T. Cioara et al.· Science in progress· 0 citations
Federated Learning (FL) has emerged as a promising paradigm for privacy-preserving distributed machine learning across mobile and edge devices. Nevertheless, the presence of non-independent and identically distributed (non-IID) data in wireless networks often leads to severe gradient divergence, slow convergence, and degraded model performance. To mitigate these challenges, this paper proposes FedCoalition, a coalition game-based clustered federated learning framework that organizes clients into collaborative coalitions according to the similarity of their model updates. The proposed method integrates a coalition formation mechanism with similarity-aware clustering to reduce gradient divergence among participating clients. In addition, each coalition designates a Coalition Head (CH) responsible for intra-cluster aggregation and model validation using auxiliary data, thereby improving model reliability and communication efficiency. The experimental results on MNIST and CIFAR-10 under non-IID data distributions demonstrate that FedCoalition significantly improves learning performance, achieving higher accuracy and faster convergence compared with existing clustered federated learning approaches.