Application of Machine Learning for Energy Efficient Clustering in Wireless Sensor Networks
As the sensor nodes are energy constrained in Wireless Sensor Networks (WSNs), several algorithms proposed for energy efficiency especially at network layer. To reduce the energy consumption in WSNs, clustering is performed in which the whole network partitioned into different clusters. However, the selection of optimal CH nodes in WSNs is the main challenge and NP-Hard optimization problem. In this research work, we proposed the application of Machine Learning (ML) technique for efficient clustering and data transmission in WSNs without loss of generality. The optimal CH selection and cluster formation is performed using the supervised ML technique called Artificial Neural Network (ANN). The problem of optimal CH selection for each cluster is formulated according to the architecture of ANN (input layer hidden layer, and output layer) in which the every sensor node properties such as residual energy, distance from the BS, and bandwidth allocated are processed as input to ANN. The functionality of the hidden layer performed to select CH in the output layer using the adaptive learning of ANN. After the clustering, the inter-cluster and intra-cluster data transmissions performed. The simulation results show the proposed clustering protocol delivered improved performances compared to existing ML-based clustering methods.