An Optimized Clustering and Improved Elman Recurrent Neural Network Based Data Aggregation Protocol For IOT Enabled WSN
K. Hemalatha, K. Somasundram · 2022
Internet of Things (IoT) integrated wireless sensor networks (WSN) supports data collection in several applications. The integration of IoT makes use of big data in different fields like weather forecasting, disaster prediction and disease prediction etc. But, the sensor node lifetime is the major issue due to its battery resource in WSN. Further, the integration of IoT makes complications in energy efficiency. The strategy or algorithm developed for energy efficiency should consider the features of both IoT and WSN. In this work, the metaheuristic optimization-based clustering and data aggregation protocol are presented to increase the energy efficiency. The optimal cluster head is selected using Dingo Optimizer (DO) with the consideration of distance and energy. The data redundancy classification is performed by an Improved Elman recurrent neural network (IERNN) for effective data aggregation. The performance results of proposed method is compared with an existing methods in terms of energy, delay, packet delivery ratio and data classification accuracy.