TY - GEN
T1 - Being SMART about failures
T2 - 14th International Conference on Ubiquitous Computing, UbiComp 2012
AU - Kapitanova, Krasimira
AU - Hoque, Enamul
AU - Stankovic, John A.
AU - Whitehouse, Kamin
AU - Son, Sang H.
PY - 2012
Y1 - 2012
N2 - Inexpensive wireless sensing products are dramatically reducing the cost of in-home sensing. However, these sensors have been found to fail often and prohibitive maintenance costs may negate the cost benefits of inexpensive hardware and do-it-yourself installation. In this paper, we describe a new technique called SMART that uses applicationlevel semantics to detect, assess, and adapt to sensor failures. SMART detects sensor failures at run-time by analyzing the relative behavior of multiple classifier instances trained to recognize the same set of activities based on different subsets of sensors. Once a failure is detected, SMART assesses its importance and adapts the classifier ensemble in attempt to avoid maintenance dispatch. Evaluation on three homes from two public datasets shows that SMART decreases the number of maintenance dispatches by 55% on average, identifies non-fail-stop failures at run-time with more than 85% accuracy, and improves the activity recognition accuracy under sensor failures by 15% on average.
AB - Inexpensive wireless sensing products are dramatically reducing the cost of in-home sensing. However, these sensors have been found to fail often and prohibitive maintenance costs may negate the cost benefits of inexpensive hardware and do-it-yourself installation. In this paper, we describe a new technique called SMART that uses applicationlevel semantics to detect, assess, and adapt to sensor failures. SMART detects sensor failures at run-time by analyzing the relative behavior of multiple classifier instances trained to recognize the same set of activities based on different subsets of sensors. Once a failure is detected, SMART assesses its importance and adapts the classifier ensemble in attempt to avoid maintenance dispatch. Evaluation on three homes from two public datasets shows that SMART decreases the number of maintenance dispatches by 55% on average, identifies non-fail-stop failures at run-time with more than 85% accuracy, and improves the activity recognition accuracy under sensor failures by 15% on average.
KW - Activity recognition
KW - Failure detection
KW - Failure severity
KW - Machine learning
KW - Wireless sensor networks
UR - http://www.scopus.com/inward/record.url?scp=84867461210&partnerID=8YFLogxK
U2 - 10.1145/2370216.2370225
DO - 10.1145/2370216.2370225
M3 - Conference contribution
AN - SCOPUS:84867461210
SN - 9781450312240
T3 - UbiComp'12 - Proceedings of the 2012 ACM Conference on Ubiquitous Computing
SP - 51
EP - 60
BT - UbiComp'12 - Proceedings of the 2012 ACM Conference on Ubiquitous Computing
Y2 - 5 September 2012 through 8 September 2012
ER -