Abstract :

The Rapid Growth Of Internet Of Things (IoT) Devices Has Significantly Increased The Vulnerability Of Modern Networks To Sophisticated Cyber Threats Such As Botnets, Cryptominers, Infostealers, Ransomware, And Zero-day Attacks. Traditional Centralized Malware Detection Systems Face Major Challenges Including Privacy Leakage, Communication Overhead, Limited Scalability, And Vulnerability To Adversarial Manipulation In Large-scale IoT Environments. Although Federated Learning (FL) Enables Collaborative Learning Without Sharing Raw Device Data, Existing FL-based Approaches Often Rely On Static Aggregation Mechanisms And Lack Adaptability Against Evolving Cyberattacks. To Address These Limitations, This Paper Proposes CyberForce, A Privacy-Preserving Federated Reinforcement Learning Framework With Dynamic Aggregation For Intelligent Malware Detection And Adaptive Cyber Defense In IoT Environments. The Proposed Framework Integrates Federated Learning For Secure Distributed Model Training, Reinforcement Learning (RL) For Dynamic Defense Optimization, And Robust Aggregation Algorithms Such As FedAvg, Krum, And Trimmed Mean To Mitigate Poisoning And Adversarial Attacks. Deep Learning Models Including CNN, LSTM, And BERT Are Employed To Extract Static And Dynamic Malware Features From Network Traffic And System Activities. Experimental Results Demonstrate That The Proposed Framework Achieves High Malware Detection Accuracy, Low Detection Latency, Improved Scalability, And Strong Resilience Against Sophisticated Cyber Threats, Making It A Reliable Cybersecurity Solution For Next-generation IoT Systems. Keywords: IoT Security, Malware Detection, Federated Learning, Reinforcement Learning, Dynamic Aggregation, Deep Learning, Privacy Preservation.

Published:

08-8-2026

Issue:

Vol. 26 No. 8 (2026)

Page Nos:

818 - 824

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite