Optimal reliable design of energy-efficient Wireless Body Area Networks

A Wireless Body Area Network (WBAN) is a network of wireless devices that are used to monitor the health of individuals or provide them with information pertinent to their personal health. A WBAN typically consists of multiple sensors and a central sink and it enables the tracking of multiple body signals without the need for cables or other wired connections, allowing for a more comfortable user experience. The used sensors are extremely small and lightweight, making them well-suited for long-term medical monitoring. In this paper, we consider the joint problem of device positioning and data routing in WBANs leveraging relay-based multi-hop communication to address network energy consumption and reliable data transmission, which are two fundamental challenges in the WBAN design. The deployment of a WBAN is the most important factor that impacts both the network lifetime and reliability. On the other side, routing in multi-hop WBANs is of paramount importance for efficient communications between sensors and the centralized hub. For this reason, we define and assign to each relay a concave and non-increasing function (called reliability function) which permits to decrease consistently congestion occurrence in the network by balancing relays loads. In this work, we consider the problem of designing an energy-aware routing on WBANs while ensuring that all data are delivered to the sink with a sufficiently large probability. This problem is formulated as a non-convex mixed-integer non-linear programming problem and proven to be NP-hard. We first present an efficient linearization technique for the problem and then propose lower- and upper-bounding schemes for the problem, which can be applied within a cutting plane algorithm-based heuristic to solve the problem. Finally, we evaluate the proposed model with a real topology scenario and compare its performance with the most notable methods presented in the literature. Numerical results show that our model, independently of patient body movement, can transfer data through reliable paths from biosensors to the sink while further offering these options: (1) the probability that each biosensor successfully transfers its data to the sink is adjustable according to the desired application; (2) congestion is limited through a careful balancing of incoming traffic to relays, and (3) a small number of relays is installed, while ensuring an energy-efficient routing of data