Abstract
In this work we investigate the use of a fuzzy inspired ap- proach for anomaly detection in different electromagnetic time series datasets. The method proposed uses simple methods in a serialized way to achieve anomaly detection. Each method is a smaller component of the system. Each of them adds an element towards the anomaly detection: A smoothing filter removes any unwanted noise, an automated peak finding with Fast Fourier Transformation and correlation reduces the dimension- ality of the signal, a fuzzy inference system encodes the signal before the final comparison and respective output. The proposed method is evalu- ated in 5 benchmark datasets with promising results and the F-Score is used to demonstrate its performance. The method is also evaluated in real datasets gathered from the SWARM satellites for the detection of possible anomalies prior and post to a seismic event. The preliminary experimental results prove to be promising for the proposed method for the detection of anomalies in electromagnetic time series datasets.
Original language | English |
---|---|
Title of host publication | Unknown Host Publication |
Publisher | Springer |
Pages | 575-587 |
Number of pages | 12 |
DOIs | |
Publication status | Published (in print/issue) - Oct 2015 |
Event | 8th International Conference on Knowledge Science, Engineering and Management - Chongqing, China Duration: 1 Oct 2015 → … |
Conference
Conference | 8th International Conference on Knowledge Science, Engineering and Management |
---|---|
Period | 1/10/15 → … |
Keywords
- Anomaly Detection
- Fuzzy Inference System
- Time Series
- Fast Fourier Transformation
- Electromagnetic Signal
- SWARM satellites
Fingerprint
Dive into the research topics of 'A Fuzzy Inspired Approach to Seismic Anomaly Detection'. Together they form a unique fingerprint.Student theses
-
Development and Application of Collective Anomaly Detection methods to Electromagnetic Satellite Data
Christodoulou, V. (Author), Wilkie, G. (Supervisor) & Bi, Y. (Supervisor), Sept 2020Student thesis: Doctoral Thesis
File