Ranking of Sensor Nodes by Optimizing Sensor Data in Energy Harvesting Wireless Sensor Network

Abstract

Wireless Sensor Networks should be self-automated and there must be a continuous power supply for the proper functioning of sensor networks. The Energy Harvesting Wireless Sensor Network plays an important role when engaging in long-term ecological monitoring, when sensor nodes are established, and when data from the environment is meant to be collected and relayed to a base station. The Internet of Things (IoT) has sparked interest in the present era, so there is a huge demand for low-power energy-harvesting wireless sensor networks in a variety of industries, such as healthcare, the military, and transportation. These networks are assessed by conducting tasks such as data collection, process monitoring, and autonomous activity control. The use of batteries to power wireless sensors limits their life and functionality in these sensor networks. By harvesting energy from the sensor’s local environment to power the device, it is possible to increase the sensor’s lifespan while simultaneously making it more ecologically friendly. The use of energy harvesting in sensor nodes allows them to be powered by batteries, dramatically lowering the cost of battery replacement. The research proposes a method for collecting sensor data from a simulator utilising six different sensors like Temperature, Wind, Humidity, Vibration, Pressure, and Light in each node and optimising the sensor nodes using the Naive Bayes machine learning approach. The final data will be represented graphically.