Joint Energy Predication and Gathering Data in Wireless Rechargeable Sensor Network

Abstract

Wireless Sensor Network (WSNs) is an infrastructure-less wireless network deployed in an increasing number of wireless sensors in an ad-hoc manner. As the sensor nodes could be powered using batteries, the development of WSN energy constraints is considered to be a key issue. In wireless sensor networks (WSNs), wireless mobile chargers (MCs) conquer such issues mainly, energy shortages. The proposed work is to produce an energy-efficient recharge method for Wireless Rechargeable Sensor Network (WRSN), which results in a longer lifespan of the network by reducing charging delay and maintaining the residual energy of the sensor. In this algorithm, each node gets sorted using the K-means technique, in which the data gets distributed into various clusters. The mobile charges execute a Short Hamiltonian cycle opposite direction to reach each cluster’s anchor point. The position of the anchor points is calculated based on the energy distribution using the base station. In this case, the network will act as a spare MC, so that one of the two MCs will run out of energy before reaching the BS. After the current tours of the two MCs terminate, regression analysis for energy prediction initiates, enabling the updating of anchor points in the upcoming round. Based on the findings of the regression-based energy prediction model, the recommended algorithm could effectively refill network energy.