Despite many energy efficient communication techniques developed in the past few years, energy supply of sensor nodes remains the main challenge in the design of wireless and body sensor networks. Harvesting energy from ambient sources such as light, wind or vibration is the most attractive solution to this problem. Due to the stochastic nature of these ambient sources of energy, the amount of harvested energy can no longer be described by simple and deterministic models, rather sophisticated stochastic models are necessary that accurately describe the variation and correlation of the harvested energy in time and space. Such models are not available today.

This project develops detailed stochastic energy models based on long term empirical measurements of three different ambient sources of energy, namely human motion for body sensor networks and solar and wind for wireless sensor networks. For human motion source, the measurements take place at multiple locations on the subject?s body using wearable accelerometers, for the duration of two days. For wind and solar sources the measurements are performed using multiple pyranometers and wind sensors, for the duration of one year.

Analogous to the channel and traffic models widely in use today, the energy models resulting from this research will provide a basis upon which many research projects concerning design and performance analysis of harvesting-aware communication techniques can be built. These models will be disseminated through publications as well as code provided freely on the Internet.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Network Systems (CNS)
Type
Standard Grant (Standard)
Application #
1321403
Program Officer
Thyagarajan Nandagopal
Project Start
Project End
Budget Start
2012-08-01
Budget End
2014-08-31
Support Year
Fiscal Year
2013
Total Cost
$113,173
Indirect Cost
Name
The University of Central Florida Board of Trustees
Department
Type
DUNS #
City
Orlando
State
FL
Country
United States
Zip Code
32816