Resource Aware Gaussian Regressive Jarvis Patrick Clustering for Reliable Data Transmission in E-Health using WSN-IoT

Abstract

Wireless sensor network (WSN) includes numerous sensor nodes (SN) integrated with Internet of Things (IoT) play a crucial role in numerous applications. IoT links physical devices as sensor and forms whole network for sharing information.  The IoT has been used in different domains. In this scenario, patients utilize wearable medical sensors to monitor medical parameters. This medical sensor is equipped with batteries and has limited energy. Therefore, the network lifetime enhancement is a major challenging issue. To prolong network lifetime, novel technique called Resource-efficient Gaussian process regressive Jarvis Patrick clustering (REGPRJPC) is introduced. At first, the IoT devices are used in SN for sensing and collecting the patient data. After data collection process, SN is grouped into diverse clusters using Jarvis Patrick clustering technique. Jarvis Patrick clustering is graph-based clustering to partition SN with help of Gaussian process regression function. The regression function analyzes the SN and performs the clustering process based on the estimated energy and bandwidth. After clustering process, cluster head (CH) is selected to enhance data transmission and minimizes delay. Source node transmits gathered data to their CH. Then CH finds nearest CH using time of flight method. Followed by, data transmission is performed from source to sink node via the cluster head. In this way, resource-efficient data transmission is performed in WSN.  Numerical analysis indicates that the REGPRJPC technique efficiently improves the reliable patient data packet delivery and minimizes the loss rate, delay.