METHOD OF BROWN’S EXPONENTIAL FILTER ADAPTATION BY USING THE METHOD OF LEAST SQUARES

Authors

  • B. R. Boriak Poltava National Technical Yuri Kondratyuk University, Poltava
  • A. M. Silvestrov National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Kyiv
  • V. V. Lutsio Poltava National Technical Yuri Kondratyuk University, Poltava

DOI:

https://doi.org/10.18372/1990-5548.54.12306

Keywords:

Exponential smoothing, noise, forecast, tracking signal, low-pass filter, smoothing factor, least squares

Abstract

Proposed structure of the filtering algorithm gives an opportunity to avoid some of disadvantages of exponential smoothing. The main aim of proposed algorithm is to estimate filtration quality and get an ability to change smoothing factor during the work of the system. We used the method of least squares to estimate the difference between smoothed signal and signal that was built from filtered signal got by its approximation. This method might be used in the case if the trajectory of the tracking signal is not changing during the estimating process or it might be changed inconsiderably. This data processing algorithm can be used as filtering and forecasting system and integrated in systems with lags.

Author Biographies

B. R. Boriak, Poltava National Technical Yuri Kondratyuk University, Poltava

Post-graduate student

A. M. Silvestrov, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Kyiv

Doctor of Engineering Science. Professor

V. V. Lutsio, Poltava National Technical Yuri Kondratyuk University, Poltava

Post-graduate student

References

B. R. Boriak, and A. M. Silvestrov, “Filtering and forecasting signals algorithm based on exponential Brown’s filter,” Academic J. Control, Navigation and Communication Systems, vol. 4, no. 44, Poltava National Technical Yuri Kondratyuk University, Poltava, Ukraine, pp. 150–152, 2017.

R. G. Brown, “Exponential Smoothing for Predicting Demand,” Cambridge, Massachusetts, USA https://www.industrydocumentslibrary.ucsf.edu/tobacco/docs/#id=jzlc0130, 15 p.

NIST/SEMATECH e-Handbook of Statistical Methods, “Double Exponential Smoothing”, http://www.itl.nist.gov/div898/handbook/pmc/section4/pmc433.htm, 1 p.

Downloads

Issue

Section

THEORY AND METHODS OF SIGNAL PROCESSING