DOI QR코드

DOI QR Code

Efficient Anomaly Detection Through Confidence Interval Estimation Based on Time Series Analysis

시계열 분석 기반 신뢰구간 추정을 통한 효율적인 이상감지

  • Kim, Yeong-Ju (Mokpo National University Dept. of Computer Engineering) ;
  • Heo, You-Kyung (Mokpo National University Department of Computer Engineering) ;
  • Park, Jin-Gwan (Mokpo National University Department of Computer Engineering) ;
  • Jeong, Min-A (Mokpo National University Dept. of Computer Engineering)
  • Received : 2014.05.08
  • Accepted : 2014.08.07
  • Published : 2014.08.31

Abstract

In this paper, we suggest a method of realtime confidence interval estimation to detect abnormal states of sensor data. For realtime confidence interval estimation, the mean square errors of the exponential smoothing method and moving average method, two of the time series analysis method, where compared, and the moving average method with less errors was applied. When the sensor data passes the bounds of the confidence interval estimation, the administrator is notified through alarming. As the suggested method is for realtime anomaly detection in a ship, an Android terminal was adopted for better communication between the wireless sensor network and users. For safe navigation, an administrator can make decisions promptly and accurately upon emergency situation in a ship by referring to the anomaly detection information through realtime confidence interval estimation.

본 논문은 센서 데이터의 이상을 감지하기 위하여 실시간 신뢰구간을 추정하였다. 실시간 신뢰구간 추정은 시계열분석 방법인 지수평활법과 이동평균법의 평균제곱오차를 비교하여 오차가 적은 이동평균법을 적용하였다. 이와 같이 추정된 신뢰구간을 측정된 센서 데이터가 이탈하게 되면 이상감지 경보를 통해 관리자에게 알려준다. 제안한 방법은 선박 내부의 실시간 이상감지를 위한 것으로 무선센서네트워크(WSN)와 사용자의 접근성을 높이기 위해 안드로이드 단말기를 사용하였다. 관리자는 실시간 신뢰구간에 따른 이상감지 정보를 활용하여 선박 내부에서 발생한 위급한 상황에서 신속하고 정확하게 의사결정을 함으로써 안전운항을 할 수 있다.

Keywords

References

  1. J. H. Park, B. T. Jang, and D. S. Lim, "Safe operation of the shipyard and ship building digital technology developments supported," Korean Inst. Inf. Sci. Eng.(KIISE), vol. 31, no. 1, pp. 55-63, Jan. 2003.
  2. A. C. Harvey, Time Series Models, 2nd Ed., MIT Press, (308), 1993.
  3. H. Zou and Y. H. Yang, "Combining Time Series Model for Forecasting," Int. J. Forecasting, vol. 20, no. 1, pp. 69-84, 2004 https://doi.org/10.1016/S0169-2070(03)00004-9
  4. K. H. Cho and D. H. Lee, "A Study on Traffic Anomaly Detection Scheme Based Time Series Model," J. KICS, vol. 33, no. 5, pp. 304-309, 2008.
  5. H. G. No, SPSS / Excel by time series analysis, HYOSAN, (323), 2008.
  6. C. Lim, M. Michael, "Time Series Forecasts of International Travel Demand for Australia," Tourism Management, vol. 23, no. 4, pp. 389-396, Aug. 2002. https://doi.org/10.1016/S0261-5177(01)00098-X
  7. Y. H. Kim, Time Series Prediction, HSPN, (448), 2002.
  8. P. Buonadonna, D. Gay, J. M. Hellerstein, W. Hong, and S. Madden, "Task: Sensor network in a box," in Proc. 2nd European Workshop on Wirel. Sensor Netw., pp. 133-144, Istanbul, Turkey, Feb. 2005.
  9. G. Werner-Allen, K. Lorincz, J. Johnson, J. Lees, and M. Welsh, "Fidelity and yield in a volcano monitoring sensor network," in Proc. 7th USENIX Symp. Operating System Design and Implementation, pp. 381-396, Berkeley, USA, Nov. 2006.
  10. K. Ni, N. Ramanathan, M. N. H. Chehade, L. Bal zano, S. Nair, S. Zahedi, E. Kohler, G. Pottie, M. Hansen, and M. Srivastava, "Sensor Network Data Fault Types," J. ACM Trans. Sensor Netw., vol. 5, no. 3, pp. 1-29, Aug. 2009.
  11. E. Elnahrawy and B. Nath, "Cleaning and Querying Noisy Sensors," in Proc. Int. Workshop Wirel. Sensor Netw. Appl., pp. 78-87, New York, USA, Sept. 2003.
  12. S. R. Jeffery, G. Alonso, M. J. Franklin, W. Hong, and J. Widom. "Declarative support for sensor data cleaning," in Proc. Int. Conf. Pervasive Computing, Lecture Notes in Comput. Sci., vol. 3968, pp. 83-100, Dublin, Ireland, May 2006.
  13. G. Shmueli, N. R. Patel, and P. C. Bruce, TData Mining for Business Intelligence, E&B, (460), 2006.
  14. S. D. Lee and U. R. Lee, Time series data analysis using SAS, TAMJI, (319), 2006.