An Improved Signal Segmentation Using Moving Average and Savitzky-Golay Filter

Analysis of long-term EEG signals needs that it be segmented into pseudo stationary epochs. That work is done by regarding to statistical characteristics of a signal such as amplitude and frequency. Time series measured in real world is frequently non-stationary and to extract important information from the measured time series it is significant to utilize a filter or smoother as a pre-processing step. In the proposed approach, the signal is initially filtered by Moving Average (MA) or Savitzky-Golay filter to attenuate its short-term variations. Then, changes of the amplitude or frequency of the signal is calculated by Modified Varri method which is an acceptable algorithm for segmenting a signal. By using synthetic and real EEG data, the proposed methods are compared with original approach (simple Modified Varri). The simulation results indicate the absolute advantage of the proposed methods.

[1]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[2]  J. Steinier,et al.  Smoothing and differentiation of data by simplified least square procedure. , 1972, Analytical chemistry.

[3]  Jean Gotman,et al.  Adaptive segmentation of electroencephalographic data using a nonlinear energy operator , 1999, ISCAS'99. Proceedings of the 1999 IEEE International Symposium on Circuits and Systems VLSI (Cat. No.99CH36349).

[4]  G. Ivanova,et al.  Fractal dimension as a feature for adaptive electroencephalogram segmentation in epilepsy , 2001, 2001 Conference Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[5]  Lenka Lhotská,et al.  Classification of Long-Term EEG Recordings , 2004, ISBMDA.

[6]  Mostefa Mesbah,et al.  Time-Frequency Feature Extraction of Newborn EEG Seizure Using SVD-Based Techniques , 2004, EURASIP J. Adv. Signal Process..

[7]  Mostefa Mesbah,et al.  Time-frequency based newborn EEG seizure detection using low and high frequency signatures. , 2004, Physiological measurement.

[8]  Jianwen Luo,et al.  Savitzky-Golay smoothing and differentiation filter for even number data , 2005, Signal Process..

[9]  L. Wong,et al.  Time-frequency evaluation of segmentation methods for neonatal EEG signals , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[10]  Tong Li,et al.  Web-Based Application for Traffic Anomaly Detection Algorithm , 2007, Second International Conference on Internet and Web Applications and Services (ICIW'07).

[11]  H. Hassanpour,et al.  Adaptive Segmentation Using Wavelet Transform , 2007, 2007 International Conference on Electrical Engineering.

[12]  D. Wang,et al.  Automatic audio segmentation using the Generalized Likelihood Ratio , 2008, 2008 2nd International Conference on Signal Processing and Communication Systems.

[13]  Hamid Hassanpour.  A time-frequency approach for noise reduction , 2008, Digit. Signal Process..

[14]  H Hasanpour,et al.  ADAPTIVE SEGMENTATION WITH OPTIMAL WINDOW LENGTH SCHEME USING FRACTAL DIMENSION AND WAVELET TRANSFORM , 2009 .

[15]  Jing Hu,et al.  Denoising Nonlinear Time Series by Adaptive Filtering and Wavelet Shrinkage: A Comparison , 2010, IEEE Signal Processing Letters.

[16]  Hamid Hassanpour,et al.  An Improved Signal Segmentation Method using Genetic Algorithm , 2011 .