Normalized Least Mean Square with Fuzzy Variable Step Size for Time Series Prediction with Chaotic Dynamic Based on Autoregressive Moving Average Model

Resumen: In this paper, the Autoregressive Moving Average (ARMA) model is used for the time series prediction with chaotic dynamic. The estimation of the weights vector of ARMA model can be obtained through an adaptive algorithm based on stochastic gradient descent, such that the prediction performance of a chaotic time series performed through ARMA model is influenced by the value of the step size. For this, in this paper, a new version of Normalized Least Mean Square (NLMS) algorithm is proposed with the step size adapted by a Mamdani Fuzzy Inference System (MFIS) used for estimation of the weights vector of ARMA model, for the time series prediction with chaotic dynamic.

¿Cómo citar?

Rodrigo Possidonio Noronha.Normalized Least Mean Square with Fuzzy Variable Step Size for Time Series Prediction with Chaotic Dynamic Based on Autoregressive Moving Average Model. Memorias del Congreso Nacional de Control Automático, pp. 142-147, 2021.

Palabras clave

ARMA Model, Chaotic Systems, Fuzzy Systems, NLMS, Step Size, Time Series

• Ahmadi, M.H.E., Royaee, S.J., Tayyebi, S., and Boozarjomehry, R.B. (2020). A new insight into implementing mamdani fuzzy inference system for dynamic process modeling: Application on flash separator fuzzy dynamic modeling. Engineering Applications of Artificial Intelligence, 90, 103485.
• Ansari, K., Park, K.D., and Kubo, N. (2019). Linear time-series modeling of the gnss based tec variations over southwest japan during 2011–2018 and comparison against arma and gim models. Acta Astronautica, 165, 248–258.
• Aslam, M.S., Shi, P., and Lim, C.C. (2021). Self-adapting variable step size strategies for active noise control systems with acoustic feedback. Automatica, 123, 109354.
• Casco-Sanchez, F., Lopez-Guerrero, M., Javier-Alvarez, S., and Medina-Ramirez, R.C. (2019). A variablestep size nlms algorithm based on the cross-correlation between the squared output error and the near-end input signal. IEEJ Transactions on Electrical and Electronic Engineering, 14(8), 1197–1202.
• Dong, R. and Pedrycz, W. (2008). A granular time series approach to long-term forecasting and trend forecasting. Physica A: Statistical Mechanics and its Applications, 387(13), 3253–3270.
• Farmer, J.D. and Sidorowich, J.J. (1987). Predicting chaotic time series. Physical review letters, 59(8), 845.
• Garroppo, R.G. and Callegari, C. (2020). Prediction of mobile networks traffic: enhancement of the nmls technique. In 2020 IEEE 25th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks (CAMAD), 1–6. IEEE.
• Gharehbaghi, A. and Lind´en, M. (2017). A deep machine learning method for classifying cyclic time series of biological signals using time-growing neural network. IEEE transactions on neural networks and learning systems, 29(9), 4102–4115.
• Golshan, M.H. and Samet, H. (2009). Updating stochastic model coefficients for prediction of arc furnace reactive power. Electric Power Systems Research, 79(7), 1114– 1120.
• Iancu, I. (2012). A Mamdani type fuzzy logic controller. InTech Romania.
• Lee, J., Zhang, P., Gan, L.K., Howey, D.A., Osborne, M.A., Tosi, A., and Duncan, S. (2018). Optimal operation of an energy management system using model predictive control and gaussian process time-series modeling. IEEE Journal of Emerging and Selected Topics in Power Electronics, 6(4), 1783–1795.
• Lorenz, M. and Brunke, M. (2021). Trends of nutrients and metals in precipitation in northern germany: the role of emissions and meteorology. Environmental Monitoring and Assessment, 193(6), 1–20.
• Mackey, M.C. and Glass, L. (1977). Oscillation and chaos in physiological control systems. Science, 197(4300), 287–289.
• Makridakis, S. and Hibon, M. (1997). Arma models and the box–jenkins methodology. Journal of forecasting, 16(3), 147–163.
• Ng, Y., Mohamad, H., and Chuah, T. (2009). Block-based fuzzy step size nlms algorithms for subband adaptive channel equalisation. IET Signal Processing, 3(1), 23– 32.
• Nonejad, N. (2021). An overview of dynamic model averaging techniques in time-series econometrics. Journal of Economic Surveys, 35(2), 566–614.
• Orozco-Tupacyupanqui, W., Nakano-Miyatake, M., and Perez-Meana, H. (2015). A novel neural-fuzzy method to search the optimal step size for nlms beamforming. IEEE Latin America Transactions, 13(2), 402–408.
• Peng, Y., Zhang, S., Zhang, J., and Huang, F. (2020). Combined-step-size proportionate decorrelation nlms algorithm for adaptive echo cancellation. In 2020 39th Chinese Control Conference (CCC), 2957–2962. IEEE.
• Prasad, S.R. and Patil, S.A. (2016). Implementation of lms algorithm for system identification. In 2016 International Conference on Signal and Information Processing (IConSIP), 1–5. IEEE.
• Rahman, S., Rashid, M.M., and Alam, M.Z. (2019). A unified analysis of proposed wavelet transform domain lms-algorithm for arma process. In 2019 5th International Conference on Advances in Electrical Engineering (ICAEE), 195–200. IEEE.
• Samet, H., Mojallal, A., and Ghanbari, T. (2013). Employing grey system model for prediction of electric arc furnace reactive power to improve compensator performance. Przeglad Elektrotechniczny, 89(12), 110– 115.
• Shin, H.C., Sayed, A.H., and Song, W.J. (2004). Variable step-size nlms and affine projection algorithms. IEEE signal processing letters, 11(2), 132–135.
• Strutz, T. (2019). Estimation of measurement-noise variance for variable-step-size nlms filters. In 2019 27th European Signal Processing Conference (EUSIPCO), 1–5. IEEE.
• Wang, L.X. (1999). A course in fuzzy systems. PrenticeHall press, USA.