Federated Learning for Authentication in IoT-Enabled Mobile Ad Hoc Networks: A Systematic Review

The rapid proliferation of Internet of Things (IoT) devices within Mobile Ad Hoc Networks (MANETs) presents significant challenges for authentication and identity management. Traditional centralized authentication mechanisms are inadequate for dynamic, resource-constrained, and infrastructure-less IoT-MANET environments. Federated Learning (FL) has emerged as a promising paradigm enabling distributed, privacy-preserving machine learning without centralizing sensitive data. This systematic review analyzes the current state of FL-based authentication approaches for IoT-enabled MANETs. Following PRISMA guidelines, we reviewed 87 studies published between 2019 and 2024 from IEEE, ACM, Springer, and Elsevier databases. We propose taxonomy classifying approaches by authentication mechanism, FL architecture, privacy technique, and attack resilience. Our analysis reveals that while FL demonstrates significant potential for decentralized authentication, critical gaps exist in certificate-less cryptography integration, mobility-aware protocols, and Byzantine resilience. We identify seven future research directions and provide recommendations for advancing this field. This review offers researchers and practitioners a structured understanding of FL-authentication for IoT-MANETs and highlights promising avenues for future investigation.