GEOPHYSICAL RESEARCH, 2016, vol.17, no.1, pp.37-45.
Abstract References Full text (in Russian) Full text (in English)
UDC 550.334
ENTROPY MEASURE OF OUTLIERS IN GPS TIME SERIES
©2016 P.V. Yakovlev
Ordzhonikidze Russian State Geological Prospecting University, Moscow, Russia
Abstract. The method of detection of significant outliers in GPS time series is proposed. The approach consists in measuring deviation at each time point relative to both left and right parts of the original time series of such statistics as standard deviation. The standard deviation is chosen because of its sensitivity to small value changes in time series. Normalized entropy is used to define to what extend outliers affect a signal.
Examples of 30-minute GPS signal analysis before and after the mega-earthquake in Japan (March 11, 2011) and also maps of normalized entropy that identify anomalous zones are presented in the article. It is shown that the epicenter is characterized by low entropy of outliers both before and after the seismic catastrophe. While the low entropy of outliers after the event is easily explained by post-seismic and aftershocks effects, revealing the anomaly of low entropy of outliers that appeared before the earthquake is more important result of the analysis carried out.
Keywords: GPS signals, time series analysis, earthquake forecast, outliers detection.
References
Aggarwal C.C., and Yu P.S. Outlier detection for high dimensional data, Proc. 2000 ACM SIGMOD International Conference on Management of Data, Santa Barbara, California, USA, 2001
Bay S.D. and Schwabacher M. Mining distance-based outliers in near linear time with randomization and a simple pruning rule, Proc. ACMSIGKDD Conference on Knowledge Discovery and Data Mining, Washington, DC, USA, 2003
Chernetskiy V.I. Matematicheskoe modelirovanoe stokasticheskikh system (Mathematic modelling of stochastic systems), Izd. of Petrozavodsk State University, 1994 (in Russ).
Leung C.K.-S., Thulasiram R.K., and Bondarenko D.A. An efficient system for detecting outliers from financial time series, Lecture Notes in Computer Science, 2006, vol. 4042, pp. 190-198. doi: 10.1007/11788911_16
Lyubushin A.A. Geophysical and Ecological Monitoring Systems Data Analysis. Moscow: Nauka, 2007
Lyubushin A.A. Synchronization Trends and Rhythms of Multifractal Parameters of the Field of Low-Frequency Microseisms, Izv. Phys. Solid Earth, 2009, vol. 45, no. 5, pp. 381–394
Lyubushin A.A. Prognostic properties of random fluctuations of geophysical characteristics, Biosphere, 2014, vol. 6, no. 4, pp. 313–332
Lyubushin A.A., Yakovlev P.V., and Rodionov E.A. Multiple analysis of GPS signal fluctuation parameters before and after the megaearthquake in Japan on 11 March 2011, Geophys. Research, 2015, vol. 16, no. 1, pp. 14-23
Lyubushin A.A., Kopilova G.N., Kasimova V.A., and Taranova L.N. About field properties of low-frequency noise registered in Kamchatka network of broadband seismic stations, Vestnik KRAUNC, 2015, vol. 2, no. 26, pp. 20-36
Lyubushin A.A. and Yakovlev P.V. Entropy measure of stepwise component in GPS time series, Izv Phys. Solid Earth, 2016, vol. 52, no. 1, pp. 98-107
Lyubushin, A. Prognostic properties of low-frequency seismic noise, Natural Science, 2015, vol. 4, pp. 659-666. doi: 10.4236/ns.2012.428087
Lyubushin A. and Yakovlev P. Properties of GPS noise at Japan islands before and after Tohoku mega-earthquake, SpringerPlus, 2014, 3:364 doi:10.1186/2193-1801-3-364, http://www.springerplus.com/content/3/1/364
Novitskiy P.V. and Zograf I.A. Estimation of measurement results errors, Energoatomizdat, 1991
Osorio I., Lyubushin A., and Didier S. Automated seizure detection: Unrecognized challenges, unexpected insights, Epilepsy & Behavior, 2011, vol. 22, Supplement 1, pp. S7-S17. The Future of Automated Seizure Detection and Prediction, DOI: 10.1016/j.yebeh.2011.09.011
Sobolev G.A. and Lyubushin A.A. Microseismic Impulses as Earthquake Precursors, Izv Phys. Solid Earth, 2006, vol. 42, no. 9, pp. 5-17
Sobolev G.A., Lyubushin A.A., and Zakrzhevskaya N.A. Asymmetrical pulses, the periodicity and synchronization of low frequency microseisms, Izv Phys. Solid Earth, 2008, vol. 2, no. 2, pp. 135-152
Solovyev A.A., Agayan S.M., Gvishiani A.D., Bogoutdinov Sh.R., and Shulliat A. Recognition of disturbances with specified morphology in time series. Part II. Spikes on 1-s magnetograms, Izv Phys. Solid Earth, 2012, vol. 48, no. 5, pp. 37-52
Sun P., Chawla S., and Arunasalam B. Outlier detection in sequential databases, Proc. SIAM International Conference on Data Mining, 2006.