The objective of this research is to develop new principles and techniques for adaptive operation in highly dynamic physical environments, using miniaturized, energy-constrained devices. The approach is to use holistic cross-layer solutions that simultaneously address all aspects of the system, from low-level hardware design to higher-level communication and data fusion algorithms to top-level applications. In particular, this work focuses on body area sensor networks as emerging cyber-physical systems. The intellectual merit includes producing new principles regarding how cyber systems must be designed in order to continually adapt and respond to rapidly changing physical environments, sensed data, and application contexts in an energy-efficient manner. New cross-layer technologies will be created that use a holistic bottom-up and top-down design -- from silicon to user and back again. A novel system-on-a-chip hardware platform will be designed and fabricated using three cutting-edge technologies to reduce the cost of communication and computation by several orders of magnitude. The broad impact of this project will enable the wide range of applications and societal benefits promised by body area networks, including improving the quality and reducing the costs of healthcare. The technology will have broad implications for any cyber physical system that uses energy constrained wireless devices. A new seminar series will bring together experts from many fields (including domain experts, such as physicians and healthcare professionals). The key aspects of this work that deal with healthcare have the potential to attract women and minorities to the computer field.

Project Report

Body sensor networks (BSN) are emerging cyber-physical systems that promise to improve the quality of life through improved health, augmented sensing and actuation for the disabled, independent living for the elderly, and reduced healthcare costs. However, the physical nature of BSNs introduces several new challenges. The human body, especially in the context of medical conditions, is a highly dynamic and unpredictable physical environment that creates constantly changing demands on sensing, actuation, and quality of service (QoS). At the same time, movement between indoor and outdoor environments and rapid physical movements of arms and legs constantly change the wireless channel characteristics. These dynamic application contexts can also have a dramatic impact on data and resource prioritization. Thus, BSNs must simultaneously deal with rapid changes to both top-down application requirements and bottom-up resource availability. This is made all the more challenging by the wearable nature of BSN devices, which necessitates a vanishingly small size and, therefore, extremely limited hardware resources and power budget. This work employs a clean slate approach, designing and implementing all the hardware and software from the ground up. At the hardware level a System on a Chip (SoC) was created that is an energy harvesting wearable sensor capable of continuous monitoring with a total power budget less than 30 μW. This requires extensive power management and a careful integration with the larger system beyond the SoC. To explore these interactions, our team modeled different MAC protocols and built features into the SoC to support the testing of a variety of protocols. The design of the SoC integrates over two dozen blocks including two microprocessors, four memories, four radios, and four analog front end channels. The boost converter on the SoC set several new world records. Designed for thermoelectric energy harvesting in 130nm CMOS, the boost converter reduces the achievable input voltage by 50% over the best prior art to 10mV, which allows wearable body sensors to continue operation with thermal gradients below 1oC. This sophisticated SoC is designed to support a suite of bioelectric physiological measurements. At the software level, a QoS framework has been developed to support coordination within and across multiple BSNs. Several new applications have been developed to exercise a multi-function BSN. The BSN includes a smartphone as an aggregator of data from multiple body sensors. The applications developed include fall detection, social interaction, and controlling the heart by music via a system calledMusicalHeart. MusicalHeart is a bio-feedback based, context aware, automated music recommendation system for smartphones. It includes a new wearable sensing platform which consists of a pair of sensor equipped earphones that communicate to the smartphone via the audio jack or by Bluretooth. MusicalHeart continuously monitors the heart rate and activity level of the user, while the user is listening to music. The physiological and contextual information are then sent to a remote server, which provides dynamic music suggestions to assist the user to maintain a target heart rate. The practicality of MusicalHeart is demonstrated by deploying it in two real world scenarios. The results show that MusicalHeart helps the user achieve a desired heart rate intensity with an average error of less than 12%, and its quality of recommendation improves over time.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Network Systems (CNS)
Type
Standard Grant (Standard)
Application #
1035771
Program Officer
David Corman
Project Start
Project End
Budget Start
2010-10-01
Budget End
2014-09-30
Support Year
Fiscal Year
2010
Total Cost
$800,000
Indirect Cost
Name
University of Virginia
Department
Type
DUNS #
City
Charlottesville
State
VA
Country
United States
Zip Code
22904