Name
Abstract | ||
|---|---|---|
The problem of sequential fault detection and isolation in multiple data streams is considered. In this work, it is assumed that many independent parallel data streams, each of which has a (possibly infinite) change point, are sequentially observed with a maximum sampling constraint. The pre-change data follow a known distribution, and the post-change data follow one of J possible distributions. A sequential procedure is proposed to detect the changes for all data streams, and to isolate the types of changes upon their detection. The sequential procedure is shown to control the false discovery rate. An asymptotic upper bound on the average detection delay over the parallel data streams is also derived. A simulation study is presented to illustrate the proposed procedure and to corroborate the analysis. |
Year | Venue | Field |
|---|---|---|
2017 | 2017 55TH ANNUAL ALLERTON CONFERENCE ON COMMUNICATION, CONTROL, AND COMPUTING (ALLERTON) | Multiple data,False discovery rate,Data stream mining,Upper and lower bounds,Computer science,Fault detection and isolation,Algorithm,Sampling (statistics),STREAMS,Distributed computing |
DocType | ISSN | Citations |
Conference | 2474-0195 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jie Chen | 1 | 91 | 38.15 |
| Wenyi Zhang | 2 | 705 | 62.34 |
| H. V. Poor | 3 | 25411 | 1951.66 |