Behavior learning of a memristor-based chaotic circuit by extreme learning machines

As the behavior of a chaotic Chua s circuit is nonstationary and inherently noisy, it is regarded as one of the most challenging applications. One of the fundamental problems in the prediction of the behavior of a chaotic Chua s circuit is to model the circuit with high accuracy. The current paper presents a novel method based on multiple extreme learning machine (ELM) models to learn the chaotic behavior of the four elements canonical Chua s circuit containing a memristor instead of a nonlinear resistor only by using the state variables as the input. In the proposed method four ELM models are used to estimate the state variables of the circuit. ELMs are first trained by using the data spoilt by noise obtained from MATLAB models of a memristor and Chua s circuit. A multistep-ahead prediction is then carried out by the trained ELMs in the autonomous mode. All attractors of the circuit are finally reconstructed by the outputs of the models. The results of the four ELMs are compared to those of multiple linear regressors (MLRs) and support vector machines (SVMs) in terms of scatter plots, power spectral density, training time, prediction time, and some statistical error measures. Extensive numerical simulations results show that the proposed system exhibits a highly accurate multistep iterated prediction consisting of 1104 steps of the chaotic circuit. Consequently, the proposed model can be considered a promising and powerful tool for modeling and predicting the behavior of Chua s circuit with excellent performance, reducing training time, testing time, and practically realization probability.

PDF

___

[1] Chua LO. Memristor - the missing circuit element. IEEE T Circuit Theor 1971; 18: 507519.
[2] Strukov DB, Snider GS, Stewart DR, Williams RS. The missing memristor found. Nature 2008; 453: 8083.
[3] Muthuswamy B, Kokate PP. Memristor-based chaotic circuits. IETE Tech Rev 2009; 26: 417429.
[4] Muthuswamy B. Memristor based chaotic circuits. Technical Report. University of California, No: UCB / EECS 2009 - 6, Berkeley, November 2009.
[5] Biolek Z, Biolek D, Biolkova V. SPICE model of memristor with nonlinear dopant drift. Radioengineering 2009; 18: 210214.
[6] Joglekar YN, Wolf SJ. The elusive memristor: properties of basic electrical circuits. Eur J of Phys 2009; 30: 661675.
[7] Muthuswamy B, Chua LO. Simplest chaotic circuits. Int J Bifurcat Chaos 2010; 20: 15671580.
[8] Prodromakis T, Peh BP, Papavassiliou C, Toumazou C. A versatile memristor model with nonlinear dopant kinetics. IEEE T Electron Dev 2011; 58: 30993105.
[9] Yakopcic C, Taha T, Subramanyam G, Pino R, Rogers S. A memristor device model. IEEE T Electron Dev L 2011; 32: 14361438.
[10] Rak A, Cserey G. Macro modeling of the memristor in SPICE. IEEE T Comput Aid D 2010; 29: 632636.
[11] Itoh, M, Chua LO. Memristor cellular automata and memristor discrete-time cellular neural networks. Int J Bifurcat Chaos 2009; 19: 36053656.
[12] Wu A, Zeng Z, Zhu X, Zhang J. Exponential synchronization of memristor-based recurrent neural networks with time delays. Neurocomputing 2011; 74: 30433050.
[13] Pershin Y, Ventra MD. Practical approach to programmable analog circuits with memristors. IEEE T Circuits-I 2010; 57:18571864.
[14] Wey TA, Jemison WD. An automatic gain control circuit with TiO2 memristor variable gain amplifier. Analog Integr Circ S 2012; 73: 663672.
[15] Witrisal K. Memristor-based stored-reference receiverthe UWB solution? Electr L 2009; 45: 713714.
[16] Merrikh-Bayat F, Bagheri-Shouraki S. Mixed analog-digital crossbar-based hardware implementation of sign-sign LMS adaptive filter. Analog Integr Circ S 2011; 66: 4148.
[17] Talukdar A, Radwan A, Salama K. Generalized model for memristor-based wien family oscillators. Microelectr J 2011; 42: 10321038.
[18] Snider GS, Williams RS. Nano/cmos architectures using a field-programmable nanowire interconnect. Nanotechnology 2007; 18: 111.
[19] Wang W, Jing T, Butcher B. Fpga based on integration of memristors and cmos devices. In: 2010 IEEE International Symposium on Circuits and Systems; 30 May2 June 2010; Paris, France. pp. 19631966.
[20] Cong J, Xiao B. Mrfpga: a novel fpga architecture with memristor-based reconfiguration. In: 2011 IEEE/ACM International Symposium on Nanoscale Architectures; 89 June 2011; San Diego, CA, USA. pp. 18.
[21] Xia Q, Robinett W, Cumbie MW, Banerjee N, Cardinali TJ, Yang JJ, Wu W, Li X, Tong WM, Strukov DB. Memristor-cmos hybrid integrated circuits for reconfigurable logic. Nano L 2009; 9: 36403645.
[22] Ebong IE. Training memristors for reliable computing. PhD, University of Michigan, USA, 2013.
[23] Driscoll T, Pershin Y, Basov D, Ventra MD. Chaotic memristor. Appl Phys A -Mater 2011; 102: 885889.
[24] Itoh M, Chua LO. Memristor oscillators. Int J Bifurcat Chaos 2008; 18: 31833226.
[25] Lin ZH, Wang HX. Efficient image encryption using a chaos-based PWL memristor. IETE Tech Rev 2010; 27: 318325.
[26] Koksal M, Hamamci SE. Analysis of four element Chua circuit containing new passive component memristor. In: 3rd Conference on Nonlinear Science and Complexity, Ankara, Turkey, 2831 July 2010; pp.16.
[27] Brillinger DR. Time Series, Data Analysis, and Theory. New York, NY, USA: McGraw-Hill, 2001.
[28] Manabe Y, Chakraborty B. A novel approach for estimation of optimal embedding parameters of nonlinear time series by structural learning of neural network. Neurocomputing, 2007; 70: 13601371.
[29] Leung H, Lo T, Wang S. Prediction of noisy chaotic time series using an optimal radial basis function neural network. IEEE T Neural Netw 2001; 12: 11631172.
[30] Sapankevych NL, Sankar N. Time series prediction using support vector machines: a survey. IEEE Comput Intell Mag 2009; 4: 2438.
[31] Farsa MA, Zolfaghari S. Chaotic time series prediction with residual analysis method using hybrid ElmanNARX neural networks. Neurocomputing 2010; 73: 25402553.
[32] Pan F, Zhang H, Xia M. A hybrid time-series forecasting model using extreme learning machines. In: Second International Conference on Intelligent Computation Technology and Automation; 1011 Oct 2009; China. pp.933936.
[33] Haykin S. Neural Networks and Learning Machines. 3rd ed. Upper Saddle River, NJ, USA: Prentice Hall, 2008.
[34] Huang GB, Zhu QY, Siew CK. Real-time learning capability of neural networks. IEEE T Neural Networ 2006; 17: 863878.
[35] Huang GB, Zhu QY, Siew CK. Extreme learning machine: theory and applications. Neurocomputing 2006; 70: 489501.
[36] Huang GB, Zhou H, Ding X, Zhang R. Extreme learning machine for regression and multiclass classification. IEEE T Syst Man Cy B 2012; 42: 513529.
[37] Decherchi S, Gastaldo P, Leoncini A, Zunino R. Efficient digital implementation of extreme learning machines for classification. IEEE T Circuits-II 2012; 59: 496500.
[38] Suykens JAK, Vandewalle J. Learning a simple recurrent neural state space model to behave like Chuas double scroll. IEEE T Circuits-I 1995; 42: 499502.
[39] Suykens JAK, Vandewalle J. Recurrent least squares support vector machines. IEEE T Circuits-I 2000; 47: 11091114.
[40] Hanbay H, T¨urko˘glu I, Demir Y. An expert system based on wavelet decomposition and neural network for modeling Chuas circuit. Expert Syst Appl 2008; 34: 22782283.
[41] T¨urk M, O˘gra¸s H. Recognition of multi-scroll chaotic attractors using wavelet-based neural network and performance comparison of wavelet families. Expert Systems with Applications 2010; 37: 86678672.
[42] Wang XD, Ye MY. Identification of memristor-based chaotic systems using support vector machine regression. In: Luo Q. Editor. Information and Automation. Springer, 2011; 86: 365371.
[43] Sun JC. Chaotic behavior learning of Chuas circuit. Chin Phys B 2012; 21: 100502-1100502-7.
[44] Chang CC, Lin CJ. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol 2011; 2: 127.
[45] U¸car A, Demir Y, G¨uzeli¸s C. A penalty function method for designing efficient robust classifiers with inputspace optimal separating surfaces. Turk J Elec Eng & Comp Sci 2014; 22: 16641685.
[46] Cherkassky C, Ma Y. Practical selection of SVM parameters and noise estimation for SVM regression. Neural Networks 2004; 17: 113126.