Over the past decade, many wireless sensor network systems have been deployed in various real-life environments with the goal of remotely collecting data from the physical world. Such low-power wireless systems, when combined with low-power medical sensors, provide the potential to alleviate many of the manual routine tasks in the healthcare domain as well. In turn, such automation techniques in the patient monitoring process can help increase the quality of patient care at many of the busy clinical environments and disaster scenes.
In this dissertation, we present MEDiSN, a wireless sensor network system for monitoring the physiological data of patients in clinical environments or victims in disaster sites. MEDiSN consists of mobile patient monitoring devices that collect physiological data and connect to a self-configuring wireless backbone network to relay their data to a gateway. As a preparation step for our deployments, we experimentally analyze the wireless channel environment at our target clinical environment and finally, evaluate the performance of MEDiSN in a real clinical environment using multiple pilot studies. Furthermore, based on the development and deployment experiences of MEDiSN, this dissertation identifies several factors that can assist wireless sensing systems to expand their horizons to new applications.
As an effort to address these factors, and moreover, expand the application coverage of the MEDiSN system, this dissertation introduces three major improvements made to MEDiSN. First, this dissertation introduces Egs, a more flexible and powerful mote platform. Egs is an ARM Cortex-M3-based mote platform that provides application developers with a powerful, yet, energy efficient sensing platform to serve in a variety of applications. Second, this dissertation discusses about the design and implementation experiences of Internet-standards compliant networking stacks for sensor networks. Specifically, we present the evaluation of the IETF RPL routing protocol in various configurations and also present MoMoRo, a supporting layer designed with the goal of assisting data collection protocols, such as RPL, to provide mobile devices in MEDiSN with a high packet reception performance while minimizing the packet overhead. The final part of the dissertation presents two adaptive transmission power control techniques for mobile nodes in low-power wireless networks designed with the goal of minimizing energy consumption and maximizing bandwidth utilization.
JeongGil Ko received the Bachelors in Engineering degree in computer science and engineering from Korea University in 2007 where he also worked as an undergraduate research student with Dr. Hyogon Kim at the Wireless Data Communications Laboratory from 2005 to 2007. He received the Master of Science in Engineering degree at the Department of Computer Science, Johns Hopkins University in 2009. At Johns Hopkins, JeongGil Ko has been a member of the Hopkins interNetworking Research Group (HiNRG) led by Dr. Andreas Terzis. In 2010, he was at the Stanford Information Networking Group (SING) with Dr. Philip Levis at Stanford University as a visiting researcher. He is a recipient of the Abel Wolman Fellowship awarded by the Whiting School of Engineering at the Johns Hopkins University in 2007 and his research interests include wireless healthcare sensing systems, low-power embedded network system and protocol design, and the deployment of such embedded sensing systems to real environments.
Upon completing his Ph.D. in 2012 at Johns Hopkins, JeongGil Ko will continue his research at the Electronics and Telecommunications Research Institute (ETRI) in Daejeon, Korea as a research engineer.