SFNB-Sensor Data Optimized by Fuzzy-Neuron with Bat Computation Based Machine Learning Routing Enabled in Wireless Network

Abstract

The machine learning system has been created to conserve energy and extend sensor life based on changes in the network and the type of the data perceived in the network. Additionally, the sensor network is constructed by merging machine learning with the BAT computational approach to decrease network redundancy by aggregating the same type of data perceived in the sensor network within a coverage region, hence reducing network redundancy. Additionally, the neighbours are chosen so that the sensors' energy is used wisely throughout the network, and feature sets are intercepted in fuzzy-neuron machine learning mode. In addition, the sensor's energy consumption is computed in the network based on the speed of the packets being monitored and the detecting interval between them. Furthermore, the data is opened and aggregated using the neural network approach in the aggregation method, which is estimated to save energy. The noise in the data is removed, and the same data is aggregated as a result. This conserves energy while maximising the utilisation of network resources. In addition, BAT computation is used to develop a consistent and energy-efficient routing path.