Designing Transmission Strategies for Enhancing Communications in Medical IoT Using Markov Decision Process

Abstract

The introduction of medical Internet of Things (IoT) for biomedical applications has brought about the era of proactive healthcare. Such advanced medical supervision lies on the foundation of a network of energy-constrained wearable or implantable sensors (or things). These miniaturized battery-powered biosensor nodes are placed in, on, or around the human body to measure vital signals to be reported to the sink. This network configuration deployed on a human body is known as the Wireless Body Area Network (WBAN). Strategies are required to restrict energy expenditure of the nodes without degrading performance of WBAN to make medical IoT a green (energy-efficient) and effective paradigm. Direct communication from a node to sink in WBAN may often lead to rapid energy depletion of nodes as well as growing thermal effects on the human body. Hence, multi-hop communication from sources to sink in WBAN is often preferred instead of direct communication with high transmission power. Existing research focuses on designing multi-hop protocols addressing the issues in WBAN routing. However, the ideal conditions for multi-hop routing in preference to single-hop direct delivery is rarely investigated. Accordingly, in this paper an optimal transmission policy for WBAN is developed using Markov Decision Process (MDP) subject to various input conditions such as battery level, event occurrence, packet transmission rate and link quality. Thereafter, a multi-hop routing protocol is designed where routing decisions are made following a pre-computed strategy. The algorithm is simulated, and performance is compared with existing multi-hop protocol for WBAN to demonstrate the viability of the proposed scheme.