An Intelligent Framework for Energy Efficient Health Monitoring System Using Edge-Body Area Networks

Abstract

Body Area Networks (BAN) consists of sensors, microcontrollers interfaced with the wireless transceivers. BAN sensors are implanted or placed on the body's surface which allows for continuous monitoring of patients' health parameters. According to recent studies, BAN is a viable option for an effective transmission of detected parameters to the nearby health care centers. This transmission helps in energy consumption for further better diagnosis. With the advent of machine learning and Internet of Things (IoT), BAN has taken the dimension in achieving the better performance with limited threshold. Although, BANs are light weight implanted nodes, the problem in improving the performance still remains demur for researchers. This paper proposes the edge based BAN which integrates the powerful Bi layered feed forward (BLFF) learning models for efficient data transmission with lower consumption of energy. The proposed model works on the adaptive distance principle of Extreme Learning Machines (ELM) which detects the cluster head BAN network. The extensive experimentation has been carried out to find the consumption of energy in the network. Additionally, the performance of the proposed ELM-BLFF learning model has been compared with the other machine learning models which are integrated in BAN-IoT frameworks. An experimental result demonstrates that the proposed ELM-BLFF model outperforms the traditional learning model with 30% lesser in terms of energy consumption.