Radial Basis Tactical Generalized Bagging Ensemble Clustering For Energy-Efficient Data Aggregation In Wsn-Iot

Abstract

Wireless sensor networks (WSNs) consist of a huge number of tiny-sized sensor nodes, whose most important task is to sense the environmental conditions in a particular region of interest. In WSN, data transmission is the multi-hop approach where each node forwards its data to the neighbor node which is nearer to the sink. Due to the higher density of sensor nodes in networks, data aggregation is a challenging task while transmitting the packets from source nodes to sink nodes. Researcher’s still faced difficulty to select an efficient and appropriate data aggregation technique. In order to improve the energy efficient data aggregation, an ensemble technique called Radial Basis Tactical Generalized Bagging Ensemble Clustering-based Energy-Efficient Data Aggregation (RBTGBEC-EEDA) is introduced in IoT-enabled WSN.  The main aim of the RBTGBEC-EEDA technique is to improve the energy-efficient data aggregation with minimum delay.  Initially, the IoT devices are used to sense and collect the data.  Then the energy of each sensor node is measured to improve the energy-efficient data aggregation in WSN. The Radial Basis Tactical Generalized Bagging Ensemble Clustering technique uses the weak learners as a Radial Basis Kernelized RBF clustering technique with the number of sensor nodes. The Radial Basis Kernelized RBF clustering technique is applied to partition the network into number of sensor nodes based on the energy. By applying the Tactical voting scheme, the bagging technique provides the final clustering results with higher accuracy and minimum error. After the clustering process, the cluster head is chosen from each cluster based on higher residual energy.  The sensor nodes in the cluster send the aggregated to their head. Then the cluster head sends the data to sink via the neighboring cluster head. In this way, efficient data aggregation is performed with higher accuracy.  Simulation of the proposed RBTGBEC-EEDA technique is carried out with different parameters such as packet delivery ratio, packet loss rate, energy consumption, throughput, and end-to-end delay. The results observed that the RBTGBEC-EEDA technique effectively improves the data aggregation with minimum data loss rate, energy consumption, and delay than the conventional approaches.