An Efficient DRL‐Dual‐Conceptualized Firefly‐Weighted Grey Wolf Optimization Approach for Clustering and Routing in IoT Cloud Environmental Wireless Sensor Network

ABSTRACT This paper presents an effective clustering and routing framework for IoT cloud‐based wireless sensor networks using a newly proposed DRL‐dual‐conceptualized firefly weighted grey wolf optimization (DRL‐DCFF‐GWO) algorithm. The main goal of this work is to increase the throughput, the network lifetime, the routing reliability and the energy efficiency of large‐scale WSN systems while decreasing latency and packet congestion. This work proposes an integration of deep reinforcement learning (DRL) and a metaheuristic combination of firefly algorithm (FA) weighting and grey wolf optimization (GWO) weighted by a cluster head choosing factor (CCF) to select the best cluster heads and the best routing paths at each sampling interval based on the online measurements of signal‐to‐interference‐and‐noise ratio (SINR), network congestion, and node survivability. Simulation results under NS‐2 demonstrate that the proposed DRL‐DCFF‐GWO significantly outperforms existing clustering and routing protocols, including chaotic GWO with slime mold algorithm (CGWOSMA), GWO–based sleep scheduling (GWOSS), low‐energy adaptive clustering hierarchy with ant colony optimization (LEACHACO), firefly algorithm based mobile sink routing (FIREFLYMS), and continuous particle swarm optimization (ContPSO). Specifically, the proposed DRL‐DCFF‐GWO achieves 8%–20% higher throughput, 10%–18% longer network lifetime, 12%–15% lower energy cost, and 10%–25% lower end‐to‐end delay than other techniques. The performance improvements indicated the effectiveness of the algorithm in minimizing energy consumption, reducing latency, and enhancing the reliability of data delivery over scalable IoT‐cloud WSN scenarios. DRL‐DCFF‐GWO can be used as a robust solution in various applications such as smart city monitoring, industrial IoT‐cloud applications such as smart cities and industrial IoT, environmental monitoring, and healthcare systems, providing flexible and efficient communication for the large‐scale deployment of WSNs.