TY - GEN
T1 - Multi-resident identification using device-free IR and RF fingerprinting
AU - Schafermeyer, Erich R.
AU - Wan, Eric A.
AU - Samin, Shadman
AU - Zentzis, Noah
AU - Preiser, Nicholas
AU - Condon, John
AU - Folsom, Jon
AU - Jacobs, Peter G.
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2015/11/4
Y1 - 2015/11/4
N2 - Remote monitoring of health and mobility is critical in the support of aging-in-place for seniors. However, it is challenging to passively monitor individuals in multi-resident homes. In this paper we present a new method for the identification of individuals using simple wall-mounted radio frequency (RF) transceivers and IR sensors with fingerprinting techniques. The approach is passive or device-free in that it does not require the person being identified to wear any transmitting device Classification is achieved using features derived from measuring the disruption of RF received signal strength (RSS) among 4 transceivers positioned across either a hallway or doorframe. Three IR sensors provide timing information. Results are given for 3 test subjects (1 female, 2 males). The approach achieves over 98% classification accuracy in distinguishing the female from the male subjects and over 83% in distinguishing between the males using a Gaussian Mixture Model for classification. More than 2300 labeled examples per subject were used for training. When the training data is reduced to less than 140 examples per subject, 96% and 82% classification accuracy is still achieved respectively.
AB - Remote monitoring of health and mobility is critical in the support of aging-in-place for seniors. However, it is challenging to passively monitor individuals in multi-resident homes. In this paper we present a new method for the identification of individuals using simple wall-mounted radio frequency (RF) transceivers and IR sensors with fingerprinting techniques. The approach is passive or device-free in that it does not require the person being identified to wear any transmitting device Classification is achieved using features derived from measuring the disruption of RF received signal strength (RSS) among 4 transceivers positioned across either a hallway or doorframe. Three IR sensors provide timing information. Results are given for 3 test subjects (1 female, 2 males). The approach achieves over 98% classification accuracy in distinguishing the female from the male subjects and over 83% in distinguishing between the males using a Gaussian Mixture Model for classification. More than 2300 labeled examples per subject were used for training. When the training data is reduced to less than 140 examples per subject, 96% and 82% classification accuracy is still achieved respectively.
UR - http://www.scopus.com/inward/record.url?scp=84953325291&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84953325291&partnerID=8YFLogxK
U2 - 10.1109/EMBC.2015.7319632
DO - 10.1109/EMBC.2015.7319632
M3 - Conference contribution
C2 - 26737532
AN - SCOPUS:84953325291
T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
SP - 5481
EP - 5484
BT - 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2015
Y2 - 25 August 2015 through 29 August 2015
ER -