TY - JOUR
T1 - A False Discovery Rate Oriented Approach to Parallel Sequential Change Detection Problems
AU - Chen, Jie
AU - Zhang, Wenyi
AU - Vincent Poor, H. V.
N1 - Funding Information:
Manuscript received July 15, 2019; revised January 21, 2020; accepted February 20, 2020. Date of publication March 2, 2020; date of current version March 20, 2020. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Remy Boyer. The work of Jie Chen and Wenyi Zhang was supported in part by the National Key Research and Development Program of China under Grant 2018YFA0701603, and in part by the National Natural Science Foundation of China under Grant 61722114. The work of H. Vincent Poor was supported by the U.S. National Science Foundation under Grant CCF-1908308. This article was presented in part at the 50th Annual Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, USA, 2016 [1], and in part at the 55th Annual Allerton Conference on Communications, Control, and Computing, Monticello, IL, USA, 2017 [2]. (Corresponding author: Wenyi Zhang.) Jie Chen and Wenyi Zhang are with the CAS Key Laboratory of Wireless-Optical Communications, and the Department of Electronic Engineering and Information Science, University of Science and Technology of China, Hefei 230027, China (e-mail: cj12@mail.ustc.edu.cn; wenyizha@ustc.edu.cn).
Publisher Copyright:
© 1991-2012 IEEE.
PY - 2020
Y1 - 2020
N2 - The problem of sequentially detecting changes in parallel data streams is formulated and investigated. Each data stream may have its own change point at which the underlying probability distribution of its data changes, and the decision maker needs to declare, sequentially, which data streams have passed their change points. With a large number of parallel data streams, the error metric is the false discovery rate (FDR), which is the expected ratio of the number of falsely declared data streams to the total number of declared data streams. A data stream is falsely declared if the detected change point is ahead of its actual change point. Decision procedures that are guaranteed to control the FDR level are developed, and it is also shown that the average decision delays (ADDs) of these decision procedures do not grow with the number of data streams. Numerical simulations and case studies are conducted to corroborate the analytical results, and to illustrate the utility of the decision procedures.
AB - The problem of sequentially detecting changes in parallel data streams is formulated and investigated. Each data stream may have its own change point at which the underlying probability distribution of its data changes, and the decision maker needs to declare, sequentially, which data streams have passed their change points. With a large number of parallel data streams, the error metric is the false discovery rate (FDR), which is the expected ratio of the number of falsely declared data streams to the total number of declared data streams. A data stream is falsely declared if the detected change point is ahead of its actual change point. Decision procedures that are guaranteed to control the FDR level are developed, and it is also shown that the average decision delays (ADDs) of these decision procedures do not grow with the number of data streams. Numerical simulations and case studies are conducted to corroborate the analytical results, and to illustrate the utility of the decision procedures.
KW - Average decision delay
KW - false discovery rate
KW - large-scale inference
KW - multiple change detection
KW - multiple hypothesis testing
UR - http://www.scopus.com/inward/record.url?scp=85081331410&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85081331410&partnerID=8YFLogxK
U2 - 10.1109/TSP.2020.2977466
DO - 10.1109/TSP.2020.2977466
M3 - Article
AN - SCOPUS:85081331410
SN - 1053-587X
VL - 68
SP - 1823
EP - 1836
JO - IEEE Transactions on Signal Processing
JF - IEEE Transactions on Signal Processing
M1 - 9020141
ER -