This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).
Obtaining physiological/behavioral data from human subjects in their natural environments is essential to conducting ecologically valid social and behavioral research. While several body area wireless sensor network (BAWSN) systems exist today for physiological data collection, their use has been restricted to controlled settings (laboratories, driving/flying scenarios, etc.); significant noise, motion artifacts, and existence of other uncontrollable confounding factors are the often cited reasons for not using physiological measurements from natural environments. In order to provide scientifically valid data from natural environments, a BAWSN system must meet several unique requirements (1) Stringent data quality without sensing redundancy, (2) Personalization to account for wide between person differences in physiological measurements, and (3) Real-time inferencing to allow for subject confirmation and timely intervention.
Intellectual Merit: In this project, a multidisciplinary team of researchers spanning various computing disciplines and behavioral sciences are developing a general purpose framework called FieldStream that will make it possible for BAWSN systems to provide long term unattended collection of objective, continuous, and reliable physiological/behavioral data from natural environments that can be used for conducting population based scientific studies. FieldStream is being incorporated in two real-life projects ? NIH sponsored AutoSense at Memphis and NSF sponsored Urban Sensing at UCLA, to help validate the assumptions, establish the feasibility of developed solutions, and to uncover new requirements.
Broader Impact: By making it possible to obtain scientifically valid objective data from the field, FieldStream promises to help solve several behavioral problems of critical importance to human society that have remained unanswered for lack of such data.
