Rendezvous points based energy-aware routing using hybrid neural network for mobile sink in wireless sensor networks

Abstract

In wireless sensor networks (WSN), the data are collected from the sensor using the mobile sink for preventing the energy-hole or hotspot problem through traversing the network periodically. The mobile sink permits the node to visit only the fewest number of nodes or locations called rendezvous points (RPs) to minimize the energy utilization and delay by visiting all the cluster heads (CHs). Further, the CHs transmit the packets to its adjacent RP. Several approaches are employed for enhancing the network lifetime and reducing the energy utilization. This paper presents a new hybrid neural network based energy-efficient routing strategy through RPs. Initially, the sensor nodes are clustered utilizing the mean shift clustering methodology. Then, the new Bald Eagle Search algorithm selects the cluster head (CH) for the clustered nodes. Consequently, RPs are selected instead of visiting all the cluster heads. Here, RPs are elected based on the weights evaluation among number of transmitted data packets and hop distance. Finally, a hybrid neural network with Group Teaching Algorithm is introduced to determine the best path through the selected RPs that moderates the energy utilization in WSNs. The implementation of the introduced methodology is performed in the Matlab platform. The simulation results proves that the presented methodology provides better outcomes than the previous techniques in regards of energy utilization, throughput, packet delivery ratio, delay, packet loss ratio, jitter, latency and network lifetime.