Towards Scalable and Efficient Embedded Sensing Networks

Yin Chen, Johns Hopkins University

Over the past decade, many embedded sensing networks have been deployed to pervasively monitor the physical world at previously infeasible or impractical spatial and temporal resolutions. The result of this novel sensing ability is deeper understanding and higher degree of control of the physical world. The decreasing cost and size of the sensor nodes will enable even larger scales of embedded sensing systems and realize many more applications. However, this continued evolution poses numerous research challenges that have not been adequately addressed before. This dissertation proposes and evaluates solutions to some of those challenges, and aims to bring the vision of pervasive inter-connectivity among every physical objects in our lives a step closer to reality. The first part of this dissertation develops methods to mitigate the unreliable nature of low power radio communications, including a lightweight calibration mechanism for IEEE 802.15.4 radios, a framework for leveraging radio spatial diversity, and a distributed algorithm that is robust to packet losses. The second part of this dissertation designs an ultra low-power time synchronization module and develops an FM fingerprinting based indoor localization system, to provide time and location services for embedded sensing systems.

Speaker Biography

Yin Chen received the Bachelors in Engineering degree in Automation from Tsinghua University in 2007. In Fall 2007, he enrolled in the Ph.D. program in the Department of Computer Science at the Johns Hopkins University. At Johns Hopkins, Yin Chen was a member of the Hopkins interNetworking Research Group (HiNRG) led by Dr. Andreas Terzis. His research interests include application development, protocol design, information processing, and time synchronization in networked embedded sensing systems, as well as indoor localization, tracking, and human-centric sensing with mobile devices.