Ensemble Regression Machine Learning Based Detection of Intrusion Packets in Healthcare Wireless Sensor Networks

Wireless sensor networks (WSNs) have arisen as a feasible technology in healthcare applications, due to recent developments in wireless networking, wireless sensors, and low-power systems with networks. Securing these networks is essential for practical deployment in healthcare applications since the data transferred in these wireless body area networks (WBAN) frequently consists of sensitive and important patient health and personal information. The objective of this study is to create and develop sophisticated intrusion detection methods that will increase protection in WBAN. In this study, we offer an intrusion detection system (IDS) that utilises Ens_learnHealth to offer the best identification of attacks for such environments. The suggested approach ensures that the intrusion detection process uses exactly the features required for identifying a particular attack, reducing the difficulty of computation. As a result, it is found that Ens_learnHealth achieves 99.7% of accuracy, 98.4% of precision, 95% of recall,96% of F1-score, and 21% of miss rate.