Browse Theses

In recent years, Wireless Sensor Networks (WSN) have attracted great research attention because of their broad application in military surveillance, precise agriculture, preventative monitoring, wildlife tracking, smart housing and assisted living. The currently deployed sensor networks have exhibited WSN's ability and efficiency for information gathering by serving as ears, eyes, mouths and noses for humans to sense the physical world. However, even though sensor networks are powerful, individual sensor nodes are tiny and resource limited in terms of computation, communication, storage and power supply. Hence, it is extremely challenging to network all these resource limited sensor devices into powerful systems.

In this dissertation, we deal with four major challenges for sensor networking: radio irregularity, radio interference, limited bandwidth, and lack of Quality of Service (QoS). Our contributions in taming these four aspects can be summarized as: (1) For radio irregularity, a new energy model is created to regenerate the irregular radio patterns, and a set of solutions are developed to conquer radio irregularity. (2) For radio interference, a runtime scheme is developed to detect the interference relations among neighboring nodes, and the results are used for better TDMA designs. (3) To improve bandwidth, a multi-frequency MAC design is developed that considers two sensor network limits: single half-duplex radio transceiver, and small application packet sizes. (4) A QoS framework is designed and implemented for body sensor networks. This BodyQoS is asymmetric and radio-agnostic to meet sensor networking challenges. Effective bandwidth is also provided for high priority streams.

The overall result of the dissertation is significant progress towards designing and developing more efficient and powerful sensor networks that can operate in the real world. The research results have benefited simulation studies as well as real system design and implementation in wireless sensor network research.