Anais De XXXVII Simpósio Brasileiro De Telecomunicações E Processamento De Sinais 2019
DOI: 10.14209/sbrt.2019.1570554422
|View full text |Cite
|
Sign up to set email alerts
|

On the Use of Vertex-Frequency Analysis for Anomaly Detection in Graph Signals

Abstract: Graph signals (GS) are widespread in many areas of data analysis, such as in social, genetics, and biomolecular networks as well as in several engineering applications. Detecting localized properties of GS using spectral tools while taking into account the underlying graph topology is still an active research topic called vertex-frequency analysis (VFA). This paper provides a brief and up-to-date overview on state-of-the-art VFA tools, namely windowed graph Fourier transform and spectral graph wavelet transfor… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
1
0
2

Year Published

2021
2021
2021
2021

Publication Types

Select...
3

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(3 citation statements)
references
References 20 publications
(36 reference statements)
0
1
0
2
Order By: Relevance
“…Instead, these behaviors are detected through interaction and comparison with their neighbors that manifest in the GFT domain as outliers. To our knowledge, this work is the initial adaptation of GSP techniques to the swarming domain, and the detection problems herein are challenging enough that multiple swarm measurements are needed for effective detection, demonstrating an anomaly detection problem where the signal is not only time-varying as in [23], but with a time varying graph, as well. This work is related to, but distinct from, the inference of dynamical parameters [24] and the identification of collective states [8,9,25], as this work more closely resembles clustering (i.e., unsupervised learning) of distinct behavior regimes within the swarm.…”
Section: Introductionmentioning
confidence: 99%
“…Instead, these behaviors are detected through interaction and comparison with their neighbors that manifest in the GFT domain as outliers. To our knowledge, this work is the initial adaptation of GSP techniques to the swarming domain, and the detection problems herein are challenging enough that multiple swarm measurements are needed for effective detection, demonstrating an anomaly detection problem where the signal is not only time-varying as in [23], but with a time varying graph, as well. This work is related to, but distinct from, the inference of dynamical parameters [24] and the identification of collective states [8,9,25], as this work more closely resembles clustering (i.e., unsupervised learning) of distinct behavior regimes within the swarm.…”
Section: Introductionmentioning
confidence: 99%
“…Além disso, se a distância entre os nós j e k é igual a dois, d k pnq também depende de u j pn´2q e assim por diante. Portanto, a topologia da rede desempenha um papel importante na forma como o sinal desejado d k pnq evolui em cada nó k. Isso torna a abordagem baseada em grafos adequada para problemas em que tanto o tempo como o espaço devem ser levados em consideração, como por exemplo, na meteorologia (NASSIF et al, 2018;HUA et al, 2018;SPELTA;MARTINS, 2018;LEWENFUS et al, 2019). Apesar das diferenças conceituais entre as duas abordagens, uma formulação matemática comum pode ser usada para descrevê-los até um certo ponto.…”
Section: Fonte: Autorunclassified
“…Além disso, nos últimos anos, foram propostos na literatura algoritmos adaptativos baseados em grafos com processamento distribuído (LUO et al, 2017;LI et al, 2018;BUI;RAVI;RAMAVAJJALA, 2017;LORENZO et al, 2016;LORENZO et al, 2018;NIGRIS et al, 2017). Tais soluções já despontam como ferramentas interessantes em uma série de aplicações em que a topologia da rede desempenha um papel importante na dinâmica dos sinais envolvidos, como meteorologia, smart grids, a internet das coisas, entre outras (NASSIF et al, 2018;HUA et al, 2018;LORENZO et al, 2017a;MOURA, 2013b;SHUMAN et al, 2013b;CHEN et al, 2015;ANIS;GADDE;ORTEGA, 2016;TSITSVERO;LORENZO et al, 2018;SPELTA;MARTINS, 2018;LEWENFUS et al, 2019).…”
Section: Conclusõesunclassified