Comparative Performance Evaluation of Federated Learning Frameworks for Edge-Oriented 6G Applications
Abstract
Federated learning (FL) enables privacy-preserving collaborative training at the network edge, a core capability envisioned for sixth-generation (6G) wireless systems. While surveys and scale-oriented benchmarks advance FL methodology, reproducible head-to-head comparisons of mainstream Python frameworks under identical FedAvg settings remain scarce. We benchmark six frameworks—Flower, TensorFlow Federated (TFF), FedML, NVIDIA FLARE, OpenFL, and PySyft—in a controlled two-phase study on a Proxmox virtualized testbed with containerized runners, formalize the FedAvg objective and communication-cost model, and position our contribution against prior surveys, scale benchmarks, and single-framework documentation. Phase 1 (MNIST) confirms protocol fairness (99.22–99.35% accuracy) but reveals up to 80% wall-time variation; Phase 2 (CIFAR-10) exposes stack-dependent accuracy gaps (TFF 71.74% vs. approximately 68% for PyTorch runners). We report per-round accuracy and loss curves, accuracy gains, wall-time comparisons, and simulated communication volume across nine figures. The experimental protocol, model topology, and hyperparameters are specified in full; per-round JSON metrics and global model checkpoints are published in . The study provides a documented baseline for 6G edge framework selection and for follow-on network-constrained and security experiments.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.