Split Learning-Enabled Framework for Secure and Light-weight Internet of Medical Things Systems

2025-10-31

AI Summary

To address the challenges of malware detection in resource-constrained Internet of Medical Things (IoMT) devices under non-independent and identically distributed (Non-IID) data, this paper proposes a lightweight split learning framework integrating image-based malware representation, edge-coordinated training, and game-theoretic joint optimization. Unlike conventional federated learning (FL), the framework partitions the model between clients and edge servers, substantially reducing communication overhead and local computational burden while preserving data privacy. It pioneers the application of split learning to IoMT security and introduces a novel game-theoretic formulation to jointly optimize resource allocation and convergence efficiency. Experimental results demonstrate that, compared to state-of-the-art FL approaches, the proposed framework achieves a 6.35% improvement in accuracy, a 5.03% gain in F1-score, a 14.96% acceleration in convergence speed, and a 33.83% reduction in resource consumption.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Application Domains: Internet of Things, Sensor Networks & Smart Cities

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
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

The rapid growth of Internet of Medical Things (IoMT) devices has resulted in significant security risks, particularly the risk of malware attacks on resource-constrained devices. Conventional deep learning methods are impractical due to resource limitations, while Federated Learning (FL) suffers from high communication overhead and vulnerability to non-IID (heterogeneous) data. In this paper, we propose a split learning (SL) based framework for IoT malware detection through image-based classification. By dividing the neural network training between the clients and an edge server, the framework reduces computational burden on resource-constrained clients while ensuring data privacy. We formulate a joint optimization problem that balances computation cost and communication efficiency by using a game-theoretic approach for attaining better training performance. Experimental evaluations show that the proposed framework outperforms popular FL methods in terms of accuracy (+6.35%), F1-score (+5.03%), high convergence speed (+14.96%), and less resource consumption (33.83%). These results establish the potential of SL as a scalable and secure paradigm for next-generation IoT security.

Problem

• Securing IoMT devices against malware attacks with limited resources
• Reducing computational burden and communication overhead in IoT systems
• Addressing non-IID data vulnerability in distributed learning frameworks

Innovation

• Split learning divides neural network between clients and server
• Game-theoretic approach optimizes computation and communication costs
• Image-based malware detection ensures privacy and reduces resource usage