Akıllı Telefonda Derin Öğrenme ile Deprem Erken Uyarı SistemiTasarımı

Ülkemiz gibi deprem kuşağında olan bir coğrafya için deprem araştırmaları ve olası erken uyarı sistemlerine dair olan yeni yaklaşımlar, son zamanlarda meydana gelen depremleri de göz önünde bulunduracak olursak (ör. İzmir, 2020), artan bir önem ve ihtiyaç teşkiletmektedir. Özellikle uyku halinde iken yakalanılan depremler, bilindiği üzere, çok daha vahim sonuçlar doğurmaktadır. Bu çalışmada,mevcut çalışmalardan farklı olarak, ilk tasarımı yapılan deprem erken uyarı sistemi yaklaşımı, uyku halinde iken, içinde bulunansensörler aracılığı ile ivmeölçer’e dönüştürülen akıllı telefonlar sayesinde, ReQuakenition ismini verdiğimiz bir telefon uygulamasıarayüzü ile acil durumlarda olası bir depremin haber verilmesi amaçlandı. Afet ve Acil Durum Yönetimi Başkanlığı (AFAD) web sayfasından indirilen gerçek deprem verilerinden yararlanarak Uzun kısa süreli belleğe sahip (Long-Short Term Memory: LSTM)tekrarlayan sinir ağı mimarisi (Recurrent Neural Network: RNN) derin öğrenme algoritmaları ile eğitilen verilerden elde edilen sonuçlarda %82’nin üzerinde duyarlılık gözlemlendi. Elde edilen bu ilk sonuçlar, son derece yaygın olarak kullanılan akıllı telefonların,deprem erken uyarı sistemlerinde kullanılmak üzere, jeodezik ve sismik ağların yanı sıra çok daha yoğun ve homojen bir ivmeölçer ağıgibi çalışabilmesi adına ümit vericidir.

A Preliminary Design of Smartphone-Based Earthquake Early Warning System via Deep Learning

For an earthquake-prone geography like our country, earthquake studies and new approaches to possible early warning systems are ofincreasing importance and need, considering especially the recent earthquakes (e.g. Izmir, 2020). As it is known, earthquakes that areoccurring especially while in sleep have much more serious consequences. In this study, unlike the current studies, a preliminaryearthquake early warning system approach has been designed for the first time to be used while in sleep. It aims to notify a possibleearthquake thanks to smartphones that are converted into accelerometers with the sensors inside. A smartphone application interfacecalled ReQuakenition has been also designed for providing a user-friendly tool for emergency cases. By making use of real earthquakedata downloaded from the Disaster and Emergency Management Presidency (AFAD) website, 82 % sensitivity was observed in theresults obtained from the data trained with the Recurrent Neural Network architecture (RNN) Long-Short Term Memory (LSTM) deep learning algorithms. These initial results are promising for the widely used smartphones to work as a much denser and homogeneousaccelerometer network as well as geodetic and seismic networks for use in earthquake early warning systems.

PDF

___

Allen R.M., Kong, Q., Martin-Short, R. (2020). The MyShake Platform: A Global Vision for Earthquake Early Warning. Pure Appl Geophys. doi: 10.1007/s00024-019-02337-7
Allen, R. M., Gasparini, P., Kamigaichi, O., & Bose, M. (2009). The status of earthquake early warning around the world: An introductory overview, Seismological Research Letters, 80(5), 682-693.
Chen, Z., Zou, H., Jiang, H., vd. (2015). Fusion of WiFi, smartphone sensors and landmarks using the kalman filter for indoor localization, Sensors (Switzerland). doi: 10.3390/s150100715
del Rosario, MB., Redmond, SJ., Lovell, NH. (2015). Tracking the evolution of smartphone sensing for monitoring human movement, Sensors (Switzerland).
Gang, H-S., Pyun, J-Y. (2019). A Smartphone Indoor Positioning System Using Hybrid Localization Technology, Energies, 12(19), 3702.
Hassan, MM., Uddin, M.Z., Mohamed, A., Almogren, A. (2018). A robust human activity recognition system using smartphone sensors and deep learning, Futur Gener Comput Syst. doi: 10.1016/j.future.2017.11.029.
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
Horvath, Z., Jenak, I., Wu, T., Xuan, C. (2016). Sensitivity of sensors built in Smartphones. In: Harmony Search Algorithm, Springer, 305–313.
Kong, Q., Allen, R. M., Schreier, L., & Kwon, Y. W. (2016a). MyShake: A smartphone seismic network for earthquake early warning and beyond. Science advances, 2(2), e1501055.
Kong, Q., Allen, R. M., & Schreier, L. (2016b). MyShake: Initial observations from a global smartphone seismic network. Geophysical Research Letters, 43(18), 9588-9594.
Lee, S., Suh, J., Choi, Y. (2018). Review of smartphone applications for geoscience: current status, limitations, and future perspectives, Earth Sci. Informatics.
Lima, W.S., Souto, E., El-Khatib. K., vd. (2019). Human activity recognition using inertial sensors in a smartphone: An overview, Sensors (Switzerland). doi: 10.3390/s19143213.
Lipton, Z.C., Berkowitz, J., Elkan, C. (2015). A critical review of recurrent neural networks for sequence learning, arXiv preprint arXiv:1506.00019.
Liu, J., Chen, R., Chen, Y., vd. (2012a). iParking: An intelligent indoor location-based smartphone parking service, Sensors (Switzerland). doi: 10.3390/s121114612.
Liu, J., Chen, R., Pei, L., vd. (2012b). A hybrid smartphone indoor positioning solution for mobile LBS, Sensors (Switzerland). doi: 10.3390/s121217208.
Majumder, S., Deen, M.J. (2019). Smartphone sensors for health monitoring and diagnosis, Sensors (Switzerland).
Manos, A., Klein, I., Hazan, T. (2019). Gravity-based methods for heading computation in pedestrian dead reckoning, Sensors (Switzerland). doi: 10.3390/s19051170.
Real Ehrlich, C., Blankenbach, J. (2019). Indoor localization for pedestrians with real-time capability using multi-sensor smartphones, Geo-Spatial Inf Sci. doi: 10.1080/10095020.2019.1613778
Retscher, G. (2019). Indoor Altitude Determination Using MEMS-based Sensors in Smartphones, In Proceedings of the ION 2019 Pacific PNT Meeting (pp. 615-627).
Sherstinsky, A. (2020). Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network. Physica D: Nonlinear Phenomena, 404, 132306.
Voicu, R.A., Dobre, C., Bajenaru, L., Ciobanu RI (2019) Human physical activity recognition using smartphone sensors. Sensors (Switzerland). doi: 10.3390/s19030458
Zhuang, Y., Yang, J., Li, Y., vd. (2016). Smartphone-based indoor localization with bluetooth low energy beacons, Sensors (Switzerland). doi: 10.3390/s16050596.