Caching Techniques for Reducing the Communication Cost of Federated Learning in IoT Environments

2025-07-19

AI Summary

To address the high communication overhead and resource constraints of edge devices in Internet-of-Things (IoT)-based federated learning (FL), this paper proposes a cache-aware model update selection mechanism. We deploy FIFO, LRU, and priority-driven caching strategies at edge nodes to intelligently filter and forward high-value model updates—marking the first systematic integration of caching into FL communication optimization. Experiments on CIFAR-10 and a real-world medical dataset demonstrate that our approach reduces total communication volume by up to 62%, incurs negligible accuracy degradation (<0.8%), and improves memory utilization and training scalability. The method is particularly suited for latency-sensitive, high-concurrency edge applications such as smart cities and telemedicine. By leveraging caching to decouple communication efficiency from model quality, this work establishes a novel paradigm for lightweight, scalable federated learning in resource-constrained IoT environments.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Federated Learning (FL) allows multiple distributed devices to jointly train a shared model without centralizing data, but communication cost remains a major bottleneck, especially in resource-constrained environments. This paper introduces caching strategies - FIFO, LRU, and Priority-Based - to reduce unnecessary model update transmissions. By selectively forwarding significant updates, our approach lowers bandwidth usage while maintaining model accuracy. Experiments on CIFAR-10 and medical datasets show reduced communication with minimal accuracy loss. Results confirm that intelligent caching improves scalability, memory efficiency, and supports reliable FL in edge IoT networks, making it practical for deployment in smart cities, healthcare, and other latency-sensitive applications.

Problem

• Reducing communication cost in Federated Learning for IoT
• Optimizing model update transmissions using caching strategies
• Maintaining accuracy while lowering bandwidth usage in FL

Innovation

• Caching strategies reduce FL communication cost
• Selective forwarding of significant model updates
• Improved scalability and memory efficiency