Computer Science and Engineering, Department of


Date of this Version

Spring 3-9-2010


A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Doctor of Philosophy, Major: Engineering (Computer Engineering), Under the Supervision of Professors Byrav Ramamurthy and Mehmet Can Vuran.
Lincoln, Nebraska: May, 2010
Copyright 2010 Yuyan Xue.


Wireless Sensor Networks (WSNs) consist of low-power embedded devices with integrated sensing, computing and wireless communication capabilities. These devices, called sensor nodes or motes, are often battery-powered and deployed in a distributed manner to provide observations on the physical world. Reliably and promptly collecting the sensing data to convey the features of a surveillance area, especially the events of interest, to the sink is one of the most critical requirements of WSN design. However, dynamic wireless channel conditions and the constrained energy resources make it a challenging task to provide the end-to-end performance guarantees in multi-hop WSNs.

The objective of this research is to develop new communication protocols that provide soft Quality of Service (QoS) guarantees for event-based WSNs in terms of latency, reliability and service-differentiation capability. By examining the application-specific end-to-end communication requirements and the fundamental resource limitations of WSNs, cross-layer solutions are developed in this work, to support Service-Differentiated Real-time Communication through an integrated MAC and network layer protocol, SDRCS, and to support Loss-Tolerant Reliable Event Sensing through a transport layer protocol, LTRES. An analytical framework based on a realistic log-normal channel model is also developed to quantitatively analyze how the end-to-end latency and energy efficiency can be improved by tuning the MAC and network layer protocol parameters. Besides the theoretical research, the design, implementation, and deployment details of a state-wide real-time groundwater monitoring network in Nebraska are provided at the end, to demonstrate the advantages of wireless communication and networking technologies in improving the accuracy, coverage, and cost efficiency of real world environmental monitoring applications.