2013
DOI: 10.1109/jstsp.2012.2237381
|View full text |Cite
|
Sign up to set email alerts
|

Space-Time Signal Processing for Distributed Pattern Detection in Sensor Networks

Abstract: A theory and algorithm for detecting and classifying weak, distributed patterns in network data is presented. The patterns we consider are anomalous temporal correlations between signals recorded at sensor nodes in a network. We use robust matrix completion and second order analysis to detect distributed patterns that are not discernible at the level of individual sensors. When viewed independently, the data at each node cannot provide a definitive determination of the underlying pattern, but when fused with d… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
39
0

Year Published

2015
2015
2024
2024

Publication Types

Select...
4
2
1

Relationship

2
5

Authors

Journals

citations
Cited by 42 publications
(40 citation statements)
references
References 22 publications
1
39
0
Order By: Relevance
“…We focus on computer networks as a motivating application area; however, we do not make any assumptions or heuristic arguments that are specific to computer networks. The results and exposition in this paper extend the results reported in two conference papers presented at the 2012 SPIE meeting [16], [17].…”
Section: Introductionsupporting
confidence: 85%
See 1 more Smart Citation
“…We focus on computer networks as a motivating application area; however, we do not make any assumptions or heuristic arguments that are specific to computer networks. The results and exposition in this paper extend the results reported in two conference papers presented at the 2012 SPIE meeting [16], [17].…”
Section: Introductionsupporting
confidence: 85%
“…The decomposition model in (16) states that Y is a linear combination of mutually uncorrelated time traces that represent the core contributing sources to each of the measured time series. Our approach is to simultaneously determine those nodes whose behavior is well-explained by the behavior of all their peers, as well as those nodes that appear to be simultaneously affected by an unusual underlying process that is outside the mainstream.…”
Section: F Latent Signal Modelsmentioning
confidence: 99%
“…For example, one can imagine generalizing Isomap to the case where both geodesics and smoothing splines are not a good approximation of long manifold distances. In such a case, one can attempt to treat the long manifold distances as unknown, and employ matrix completion techniques like [43], [44] on distance matrices where some entries are not observed. Institute where his focus is on the WPI Data Science Program.…”
Section: Resultsmentioning
confidence: 99%
“…kl ] 1≤k,l≤n for i = j. The translation invariance properties (1) and (2) imply that C (i,i) and C (i,j) are Toeplitz matrices. Therefore c (i,i) kl and c (i,j) kl depend on k and l only through the quantity k−l.…”
Section: Definitionsmentioning
confidence: 99%
“…Correlation analysis of multivariate time series is important in many applications such as wireless sensor networks, computer networks, neuroimaging, and finance [1,2,3,4,5]. This chapter focuses on the problem of detecting hub time series, ones that have a high degree of interaction with other time series as measured by correlation or partial correlation.…”
Section: Introductionmentioning
confidence: 99%