RPSO Optimization with machine learning in WSN

This work emphasizes to increase the network lifetime by using an appropriate data collection scheme and machine learning technique. The routing mechanism is one of the best approaches to decrease energy consumption and increase the lifetime of the network as well. We have used PSO with an updated scheme where we are selecting the random values to find best fitness value then the final route will be calculated. Genetic methods like mutation and crossover are implemented over the final routes to get alternate routes and then performance will be calculated. We have compared the lifetime and stability of network with existing protocols like Low Energy Adaptive Clustering Hierarchy (LEACH), Power-Efficient Gathering in Sensor Information Systems (PEGASIS), and Ant Colony Routing (ACR). In this work, we have added active-sleep feature with our network to enhance the network lifetime and the machine learning technique is used to predict the data of the network in sleep state. MATLAB is used to validate our mathematical framework; we have performed analytical simulations by choosing the network area, the number of nodes in each cluster. The lifetime and stability period is analyzed and compared with other optimization methods.