Author's School

School of Engineering & Applied Science

Author's Department/Program

Computer Science and Engineering

Language

English (en)

Date of Award

Summer 9-1-2014

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Chenyang Lu

Abstract

Recent years have witnessed growing interest in deploying wireless sensing applications in real-world environments. For example, home automation systems provide fine-grained metering and control of home appliances in residential settings. Similarly, assisted living applications employ wireless sensors to provide continuous health and wellness monitoring in homes. However, real deployments of Wireless Sensor Networks (WSNs) pose significant challenges due to their low-power radios and uncontrolled ambient environments. Our empirical study in over 15 real-world apartments shows that low-power WSNs based on the IEEE 802.15.4 standard are highly susceptible to external interference beyond user control, such as Wi-Fi access points, Bluetooth peripherals, cordless phones, and numerous other devices prevalent in residential environments that share the unlicensed 2.4 GHz ISM band with IEEE 802.15.4 radios.

To address these real-world challenges, we developed two practical wireless network protocols including the Adaptive and Robust Channel Hopping (ARCH) protocol and the Adaptive Energy Detection Protocol (AEDP). ARCH enhances network reliability through opportunistically changing radio's frequency to avoid interference and environmental noise and AEDP reduces false wakeups in noisy wireless environments by dynamically adjusting the wakeup threshold of low-power radios.

Another major trend in WSNs is the convergence with smart phones. To deal with the dynamic wireless conditions and varying application requirements of mobile users, we developed the Self-Adapting MAC Layer (SAML) to support adaptive communication between smart phones and wireless sensors. SAML dynamically selects and switches Medium Access Control protocols to accommodate changes in ambient conditions and application requirements.

Compared with the residential and personal wireless systems, industrial applications pose unique challenges due to their critical demands on reliability and real-time performance. We developed an experimental testbed by realizing key network mechanisms of industrial Wireless Sensor and Actuator Networks (WSANs) and conducted an empirical study that revealed the limitations and potential enhancements of those mechanisms. Our study shows that graph routing is more resilient to interference and its backup routes may be heavily used in noisy environments, which demonstrate the necessity of path diversity for reliable WSANs. Our study also suggests that combining channel diversity with retransmission may effectively reduce the burstiness of transmission failures and judicious allocation of multiple transmissions in a shared slot can effectively improve network capacity without significantly impacting reliability.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7445JJJ

Download

Share

COinS