Energy-Efficient Framework Clustering and Routing in WSN Using Federated Deep Q-Network with Improved Fossa Optimization Algorithm

Today's major goals in sensor network research are to extend the life of wireless sensor networks (WSNs) and reduce power consumption. IoT-based WSN are widely used in a range of applications, including military, healthcare, and industrial monitoring. WSN nodes often have limited battery capacities, making energy efficiency an important consideration for clustering and routing. Data is transferred from the source SNs to the destination SNs. These are likely to be completed in a secure manner and in less time. Energy-efficient data transmission is a significant challenge for WSNs coupled with IoT. This research provides an optimal clustering and routing paradigm for increasing network lifetime, reducing energy usage, and ensuring reliable data transfer. Cluster creation is carried out using a Trusted Energy-Efficient Fuzzy Logic-Based Clustering (TEEFLC) Algorithm, which takes into account node trustworthiness, residual energy, and network density. The Improved Fossa Optimization Algorithm (FOA) is used to choose the ideal Cluster Head (CH), maintaining balanced energy distribution and reducing the number of CH replacements. To provide efficient data transmission, a Federated Deep Q-Network (FDQN) based routing strategy is used, which optimizes next-hop selection based on energy efficiency and link quality. Simulation findings show that the proposed method outperforms standard clustering and routing protocols in terms of energy efficiency, packet delivery ratio, and network longevity, indicating that it is a viable solution for WSN-IoT applications.