Energy-Aware Cluster-Based Routing with Federated Learning Integration for Scalable IoT Environments

Abstract

As a result of the Internet of Things' (IoT) explosive growth secure routing, energy optimization, and privacy preservation in resource-constrained environments have become major challenges. High overhead, static decision-making and susceptibility to malevolent attacks are common problems with traditional routing protocols. Federated Learning-Assisted Encrypted Routing based on Cost Function (FL-ERCF), an improved routing protocol that combines encrypted transmission with intelligent, privacy-preserving cluster head selection, is proposed in this paper to address these issues. Three main steps make up the proposed protocol: (i) link quality evaluation using metrics like RSSI, SNR, and variance; (ii) trust-based cluster formation directed by a federated learning model trained across dispersed IoT nodes to dynamically choose the best CHs; and (iii) secure data transmission using a lightweight symmetric encryption mechanism enhanced with digital certificates. Federated learning protects node-level privacy while improving flexibility and resilience to changing network conditions. The NetSim simulator is used to implement and assess the suggested FL-ERCF protocol, and it is compared to more established protocols like Energy Efficient Secure Routing (EESR) and Hybrid Secure Routing (HSR). Even in the face of an adversarial attack, experimental results show increased packet delivery ratios, decreased routing overhead, and enhanced throughput. A scalable and secure IoT routing framework that is appropriate for mission-critical applications, smart cities, and healthcare is established by this work.