TY - GEN
T1 - Quickest Detection of Growing Dynamic Anomalies in Networks
AU - Rovatsos, Georgias
AU - Veeravalli, Venugopal V.
AU - Towsley, Don
AU - Swami, Ananthram
N1 - Funding Information:
This work was sponsored in part by the Army Research Laboratory under Cooperative Agreement W911NF-17-2-0196 (IoBT CRA), and in part by the National Science Foundation (NSF) under grant CCF 16-18658.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/5
Y1 - 2020/5
N2 - The problem of quickest growing dynamic anomaly detection in sensor networks is studied. Initially, the observations at the sensors, which are sampled sequentially by the decision maker, are generated according to a pre-change distribution. At some unknown but deterministic time instant, a dynamic anomaly emerges in the network, affecting different sets of sensors as time progresses. The observations of the affected sensors are generated from a post-change distribution. It is assumed that the number of affected sensors increases with time, and that only the initial and the final size of the anomaly are known to the decision maker. The goal is to detect the emergence of the anomaly as quickly as possible while guaranteeing a sufficiently low frequency of false alarm (FA) events. This detection problem is posed as a stochastic optimization problem by using a delay metric that is based on the worst possible path of the anomaly. A detection rule is proposed that is asymptotically optimal as the mean time to false alarm goes to infinity. Finally, numerical results are provided to validate our theoretical analysis.
AB - The problem of quickest growing dynamic anomaly detection in sensor networks is studied. Initially, the observations at the sensors, which are sampled sequentially by the decision maker, are generated according to a pre-change distribution. At some unknown but deterministic time instant, a dynamic anomaly emerges in the network, affecting different sets of sensors as time progresses. The observations of the affected sensors are generated from a post-change distribution. It is assumed that the number of affected sensors increases with time, and that only the initial and the final size of the anomaly are known to the decision maker. The goal is to detect the emergence of the anomaly as quickly as possible while guaranteeing a sufficiently low frequency of false alarm (FA) events. This detection problem is posed as a stochastic optimization problem by using a delay metric that is based on the worst possible path of the anomaly. A detection rule is proposed that is asymptotically optimal as the mean time to false alarm goes to infinity. Finally, numerical results are provided to validate our theoretical analysis.
KW - Dynamic anomaly
KW - quickest change detection
KW - transient dynamics
KW - worst-path approach
UR - http://www.scopus.com/inward/record.url?scp=85089196965&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85089196965&partnerID=8YFLogxK
U2 - 10.1109/ICASSP40776.2020.9053019
DO - 10.1109/ICASSP40776.2020.9053019
M3 - Conference contribution
AN - SCOPUS:85089196965
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 8926
EP - 8930
BT - 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020
Y2 - 4 May 2020 through 8 May 2020
ER -