2022/05/13 by Prakash, P. Suman, Kavitha, D., Reddy, P. Chenna
paper · doi:10.24423/cames.383
The wireless sensor networks (WSNs) and their extensive characteristics and applicability to a wide range of applications attract researchers attention. WSN is an emerging technology where the sensor nodes are its major elements used to monitor and control physical and environmental systems. Clustering in wireless sensor networks groups all the nodes in a region, uses a single node as a cluster head, and communicates with the sink. However, the resource-constrained nodes’ lifetime reduces in the communication process. To improve the network lifetime, an efficient cluster head selection process is widely adopted. Similarly, identifying energy-efficient routing reduces the node energy requirements and enhances the network lifetime. Considering these two characteristics as objective, this research work proposes a fuzzy neural network-based clustering with dolphin swarm optimization routing and congestion control (FNDSCC), where an energy-efficient cluster head selection using a deep fuzzy neural network (DFNN) model and an energy-aware optimal routing using an improved dolphin swarm optimization (DSO) enhance the network lifetime by reducing the energy consumption of the nodes. Moreover, novel rate adjustment techniques to overcome the congestion inside the network are introduced. Proposed model performance is experimentally verified and compared with conventional methods such as genetic based efficient clustering (GEC), hybrid particle swarm optimization (HPSO), and artificial bee colony (ABC) optimization and rate-controlled reliable transport (RCRT) protocol in terms of latency, reliability, packet delivery ratio, network lifetime and efficiency. The results demonstrate that the proposed multi-objective approach performs better than conventional models.