Early Detection of Seismic Signal Anomalies Using Raspberry Pi 5 and Lightweight Machine Learning Models
DOI:
https://doi.org/10.21009/03.1401.FA14Abstract
Data integrity is crucial for seismic monitoring systems, but is often compromised by anthropogenic or instrumental anomalies. This paper proposes a lightweight edge computing framework using Raspberry Pi 5 for real-time anomaly detection. MiniSEED data from the high-noise TOJI station were processed through segmentation, statistical or spectral feature extraction, and unsupervised models (isolation forest and autoencoder). The results show a detection latency of 78-113 ms with minimal resource consumption (<35% CPU, <200 MB RAM) and 82% correlation with ground-truth anomalies. This framework can be used on networked seismographs with limited resources such as those of the BMKG.
References
[1] U. Ashraf et al., “Application of unconventional seismic attributes and unsupervised machine learning for the identification of fault and fracture network,” Applied Sciences, vol. 10, no. 11, p. 3864, Jun. 2020.
[2] J. Cai et al., “Research on U-Net seismic signal denoising combined with residual dense blocks,” Measurement Science and Technology, vol. 35, no. 5, p. 056104, May 2024.
[3] M. Carratù, E. De Benedictis, G. D’Elia, and M. D. Falco, “A novel methodology for unsupervised anomaly detection in industrial electrical systems,” IEEE Transactions on Instrumentation and Measurement, vol. 72, pp. 1–12, 2023.
[4] P. P. Fastykovsky, S. V. Fastykovsky, and D. V. Fastykovsky, “Autonomous compact seismic device for detection and recognition of moving person and vehicles,” in Proc. 2020 IEEE 15th Int. Conf. Advanced Trends in Radioelectronics, Telecommunications and Computer Engineering (TCSET), Lviv-Slavske, Ukraine, 2020, pp. 969–972.
[5] F. Finazzi, “Earthquake detection at the edge: IoT crowdsensing network,” Future Internet, vol. 14, no. 5, p. 143, May 2022.
[6] J. Gu, S. Yu, and D. P. G. Trad, “Unsupervised deep learning for off-the-grid seismic data regularization and denoising,” Geophysics, vol. 89, no. 3, pp. 1–15, 2024.
[7] S. Han, D. Kim, and H. Song, “AI-based anomaly detection integrating autoencoders and isolation forests,” Medium, Jan. 23, 2024. [Online]. Available: https://medium.com/@han.s.ml/ai-based-anomaly-detection-integrating-autoencoders-and-isolation-forests-e9b995f214b5
[8] B. Housner, “AI on the edge: IoT and edge computing redefine data architectures,” Database Trends and Applications (DBTA), Jan. 19, 2024. [Online]. Available: https://www.dbta.com/Editorial/Trends-and-Applications/AI-on-the-Edge-IoT-and-Edge-Computing-Redefine-Data-Architectures-162310.aspx
[9] A. Carvalho, C. Machado, and F. Moraes, “Raspberry Pi performance analysis in real-time applications with the RT-Preempt patch,” in Proc. 2019 Latin American Robotics Symposium (LARS), Rio Grande, Brazil, 2019, pp. 162–167.
[10] F. T. Liu, K. M. Ting, and Z.-H. Zhou, “Isolation Forest,” in Proc. 2008 IEEE Int. Conf. Data Mining, Pisa, Italy, 2008, pp. 413–422.
[11] H. Ma, Y. Liu, and H. Chen, “Noise reduction for desert seismic data using spectral kurtosis adaptive bandpass filter,” Acta Geophysica, vol. 67, pp. 123–131, 2019.
[12] D. E. McNamara and R. P. Buland, “Ambient noise levels in the continental United States,” Bulletin of the Seismological Society of America, vol. 94, no. 4, pp. 1517–1527, Aug. 2004.
[13] S. M. Mousavi and G. C. Beroza, “A machine-learning approach for seismic analysis of industrial facilities,” Journal of Seismology, vol. 26, pp. 1–15, 2022.
[14] N. Nakata, L. Gualtieri, and A. Fichtner, Seismic Ambient Noise. Cambridge, UK: Cambridge University Press, 2019.
[15] A. T. Ringler et al., “A review of timing accuracy across the Global Seismographic Network,” Seismological Research Letters, vol. 93, no. 1, pp. 317–331, Jan. 2022.
[16] M. Sakurada and T. Yairi, “Anomaly detection using autoencoders with nonlinear dimensionality reduction,” in Proc. MLSDA 2014 – 2nd Workshop on Machine Learning for Sensory Data Analysis, 2014, pp. 4–11.
[17] L. Seydoux, R. Balestriero, and P. Poli, “Clustering of continuous seismic data using deep scattering networks,” Geophysical Journal International, vol. 223, no. 3, pp. 1733–1748, Dec. 2020.
[18] A. Tolea, A. Tiganescu, and C. Ionescu, “Analysis of ambient seismic noise at Bucovina Seismic Array (BURAR), Romania,” Romanian Reports in Physics, vol. 75, p. 705, 2023.
[19] P. H. Tran, C. Heuchenne, and S. Thomassey, “An anomaly detection approach based on the combination of LSTM autoencoder and isolation forest for multivariate time series data,” in Developments of Artificial Intelligence Technologies in Computation and Robotics. World Scientific, 2020, pp. 485–492.
[20] L. Utkin, S. Zatenko, and A. Meldo, “Improved anomaly detection by using the attention-based Isolation Forest,” Algorithms, vol. 16, no. 1, p. 19, Jan. 2023.
