IoT Healthcare Monitoring System with Predictive Modeling and Network Security: A Hybrid MANET Approach
B. Mary Havilah HaqueDepartment of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore, Tamil Nadu, India. havilah737@gmail.com0000-0002-9077-7395
Dr. K. Martin SagayamDivision of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore, Tamil Nadu, India. martinsagayam.k@gmail.com0000-0003-2080-0497
Dr. D. Jackuline MoniDivision of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore, Tamil Nadu, India. jackulinemoni91@gmail.com0000-0001-7314-023X
Dr. Mandalapu Kalpana ChowdaryDepartment of Electronics and Communication Engineering, BVRIT Hyderabad College of Engineering for Women, Hyderabad, Telangana, India. kalpanachowdary.m@bvrithyderabad.edu.in0000-0002-1201-3146
Keywords: IoT Health Monitoring, Multiple Linear Regression, Network Security, Direct Symmetry Test, Prim’s Algorithm, Trust-Based Mechanism, Hybrid Model.
Abstract
The healthcare monitoring systems, with the concept called ‘telehealth’, started early in the year 1948. In classical times, it was a challenge to connect the security and sustainability of the healthcare monitoring system. In the later years, introducing the Internet of Things (IoT) technology facilities into the world has helped the healthcare industry overcome the challenge. In IoT, the sensors are used to capture data in health monitoring systems. In this paper, a health monitoring system that combines the DHT11, MLX90614, AD8232, and MAX30100 sensors with an interpretable linear regression model to predict health risks is presented. The cross-validation is done for predictive performance, having R2 = 0.96. In Healthcare IoT, devices like wearable sensors and patient monitors are wirelessly connected. These networks often use mobile ad-hoc communication models. The health monitoring systems share vast quantities of data. Malicious nodes in the network can alter patient data and delay emergency messages. Here, the security and sustainability of the healthcare monitoring system are discussed, and the hybrid method is introduced, used for network security. The hybrid model is compared with existing models. The hybrid model with a large node number has better performance in PDR gain than LVT-CBGR by 6.04%. The QoS (Quality of Service) parameters for the ratio of packet delivery, latency, detection rate, overhead, and throughput are computed.