Cellular-Assisted Device-to-Device Communications for Healthcare Monitoring Wireless Body Area Networks

We envisage each person will wear an individual wireless body area network (WBAN) for healthcare monitoring. Two WBANs in proximity may collide. Such WBAN collisions can significantly degrade communication quality of service (QoS), in terms of throughput and latency of healthcare applications. To avoid collisions, we propose to use cellular-assisted device-to-device (D2D) communications to connect on-body sensors to a hub within WBAN. These D2D transmissions are performed concurrently with cellular transmissions and thus, interference is imposed on each other. A radio resource allocation scheme has been developed to manage the interference. The scheme uses a combination of difference of convex optimization and Hungarian algorithm to control transmit power and to assign communication channel to both cellular and D2D transmissions. The objective is to maximize the sum of cellular throughput while avoiding collisions among co-located WBANs and satisfying QoS requirements of healthcare applications. Evaluation results confirm that WBAN collisions can be avoided at a crowded pedestrian walkway, even when there are as many as 6 pedestrian arrivals per second.

I. Introduction

Continuous monitoring of physiological signs, such as heart rate, blood pressure, body temperature, blood glucose level, etc. is an important aspect in a modern healthcare surveillance regime [1]. Through such healthcare monitoring, we can achieve precise application of medicine, ensure prompt response to an emergency, verify treatment compliance and perform accurate diagnosis of a long-term illness. For example, in clinical examinations, a blood pressure profile which is measured continuously over a time period is more useful than snapshots of measurement which are taken only at several moments. To facilitate continuous healthcare monitoring, a person may wear various sensors on the body [2], [3]. Through a wireless sensor network (WSN), these sensors are connected to an on-body communication hub, which is further connected to the remote medical control center through a commercially available wireless cellular network. In the literature, the on-body healthcare WSN is also called wireless personal area network (WPAN) or wireless body area network (WBAN). Hereafter in this paper, we use the term WBAN to include healthcare WSN and WPAN without any differentiation. In various existing works, WBANs are commonly implemented as standalone networks using diverse types of wireless communication technologies, such as Zigbee, Bluetooth, Wi-Fi, etc. These existing WBANs operate in a licence-free radio spectrum, typically at around 2.4 GHz.