Date of Award

Spring 5-10-2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Computer Science

First Advisor

WenZhan Song

Second Advisor

Xiaolin Hu

Third Advisor

Sushil Prasad

Fourth Advisor

Mark Keil


The efficient designs of Wireless Sensor Network protocols and intelligent Machine Learning algorithms, together have led to the advancements of various systems and applications for Smart Environments. By definition, Smart Environments are the typical physical worlds used in human daily life, those are seamlessly embedded with smart tiny devices equipped with sensors, actuators and computational elements. Since human user is a key component in Smart Environments, human motion activity patterns have key importance in building sensor network systems and applications for Smart Environments. Motivated by this, in this thesis my work is focused on human motion activity-aware sensor networks for Smart Environments. The main contributions of this thesis are in two important aspects: (i) Designing event activity context-aware sensor networks for efficient performance optimization as well as resource usage; and (ii) Using binary motion sensing sensor networks' collective data for device-free real-time tracking of multiple users. Firstly, I describe the design of our proposed event activity context-aware sensor network protocols and system design for Smart Environments. The main motivation behind this work is as follows. A sensor network, unlike a traditional communication network, provides high degree of visibility into the environmental physical processes. Therefore its operation is driven by the activities in the environment. In long-term operations, these activities usually show certain patterns which can be learned and effectively utilized to optimize network design. In this thesis I have designed several novel protocols: (i) ActSee for activity-aware radio duty-cycling, (ii) EAR for activity-aware and energy balanced routing, and (iii) ActiSen complete working system with protocol suites for activity-aware sensing/ duty-cycling/ routing. Secondly, I have proposed and designed FindingHuMo (Finding Human Motion), a Machine Learning based real-time user tracking algorithm for Smart Environments using Sensor Networks. This work has been motivated by increasing adoption of sensor network enabled Ubiquitous Computing in key Smart Environment applications, like Smart Healthcare. Our proposed FindingHuMo protocol and system can perform device-free tracking of multiple (unknown and variable number of) users in the hallway environments, just from non-invasive and anonymous binary motion sensor data.