Time delay Chebyshev functional link artificial neural network

Abstract In real applications, a time delay in the parameter update of the neural network is sometimes required. In this paper, motivated by the Chebyshev functional link artificial neural network (CFLANN), a new structure based on the time delay adaptation is developed for nonlinear system identification. Particularly, we present a new CFLANN-delayed recursive least square (CFLANN-DRLS), as an online learning algorithm for parameter adaptation in CFLANN. The CFLANN-DRLS algorithm exploits the time delayed error signal with the gain vector delayed by D cycles to form the weight increment term, which provides potential implementation in the filter with pipelined structure. However, it suffers from the instability problems under imperfect network delay estimate. To overcome this problem, we further propose a modified CFLANN-DRLS (CFLANN-MDRLS) algorithm by including a compensation term to the error signal. We analyze the stability and convergence of the proposed algorithm. Simulations in nonlinear system identification contexts reveal that the newly proposed CFLANN-MDRLS algorithm can effectively compensate the time delay of system and it is even superior to the algorithm without delay in some cases.

[1]  Alberto Carini,et al.  Fourier nonlinear filters , 2014, Signal Process..

[2]  Wen Yu,et al.  Multiple recurrent neural networks for stable adaptive control , 2006, Neurocomputing.

[3]  Markus Rupp,et al.  Analysis of LMS and NLMS algorithms with delayed coefficient update under the presence of spherically invariant processes , 1994, IEEE Trans. Signal Process..

[4]  Cheng-I Chen,et al.  Radial-Basis-Function-Based Neural Network for Harmonic Detection , 2010, IEEE Transactions on Industrial Electronics.

[5]  Jian Xu,et al.  Time delay identifiability and estimation for the delayed linear system with incomplete measurement , 2016 .

[6]  Lixin Tian,et al.  Chaotic characteristic identification for carbon price and an multi-layer perceptron network prediction model , 2015, Expert Syst. Appl..

[7]  M. Srinath,et al.  Conjugate gradient techniques for adaptive filtering , 1992 .

[8]  Giovanni L. Sicuranza,et al.  A study about Chebyshev nonlinear filters , 2016, Signal Process..

[9]  Jagdish C. Patra,et al.  A functional link artificial neural network for adaptive channel equalization , 1995, Signal Process..

[10]  Yoh-Han Pao,et al.  Adaptive pattern recognition and neural networks , 1989 .

[11]  Chin-Liang Wang.  Bit-serial VLSI implementation of delayed LMS adaptive FIR filters , 1994, IEEE Trans. Signal Process..

[12]  Danilo Comminiello,et al.  Improving nonlinear modeling capabilities of functional link adaptive filters , 2015, Neural Networks.

[13]  Lu Li,et al.  Adaptive neural network tracking control for a class of switched strict-feedback nonlinear systems with input delay , 2016, Neurocomputing.

[14]  Alex ChiChung Kot,et al.  Nonlinear dynamic system identification using Chebyshev functional link artificial neural networks , 2002, IEEE Trans. Syst. Man Cybern. Part B.

[15]  Badong Chen,et al.  Collaborative adaptive Volterra filters for nonlinear system identification in α-stable noise environments , 2016, J. Frankl. Inst..

[16]  D. P. Agrawal,et al.  A modular pipelined implementation of a delayed LMS transversal adaptive filter , 1990, IEEE International Symposium on Circuits and Systems.

[17]  R.D. Poltmann,et al.  Conversion of the delayed LMS algorithm into the LMS algorithm , 1995, IEEE Signal Processing Letters.

[18]  Geoffrey E. Hinton,et al.  Phoneme recognition using time-delay neural networks , 1989, IEEE Trans. Acoust. Speech Signal Process..

[19]  Badong Chen,et al.  Improved-Variable-Forgetting-Factor Recursive Algorithm Based on the Logarithmic Cost for Volterra System Identification , 2016, IEEE Transactions on Circuits and Systems II: Express Briefs.

[20]  Junfei Qiao,et al.  Self-organization of a recurrent RBF neural network using an information-oriented algorithm , 2017, Neurocomputing.

[21]  Nithin V. George,et al.  Nonlinear active noise control using spline adaptive filters , 2015 .

[22]  Andrew H. Jazwinski,et al.  Adaptive filtering , 1969, Autom..

[23]  Zongsheng Zheng,et al.  Memory improved proportionate M-estimate affine projection algorithm , 2015 .

[24]  Haiquan Zhao,et al.  Adaptive Volterra filter with continuous lp-norm using a logarithmic cost for nonlinear active noise control , 2016 .

[25]  Guoqing Huang,et al.  A semi-analytical formula for estimating peak wind load effects based on Hermite polynomial model , 2017 .

[27]  Badong Chen,et al.  Sparse normalized subband adaptive filter algorithm with l0-norm constraint , 2016, J. Frankl. Inst..

[28]  Amitava Chatterjee,et al.  A comparative study of adaptation algorithms for nonlinear system identification based on second order Volterra and bilinear polynomial filters , 2011 .

[29]  Lok-Kee Ting,et al.  Virtex FPGA implementation of a pipelined adaptive LMS predictor for electronic support measures receivers , 2005, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[30]  Khashayar Khorasani,et al.  Adaptive time delay neural network structures for nonlinear system identification , 2002, Neurocomputing.

[31]  Wan-De Weng,et al.  A channel equalizer using reduced decision feedback Chebyshev functional link artificial neural networks , 2007, Inf. Sci..

[32]  Li Xiangyang,et al.  A novel self-optimizing approach for knowledge acquisition , 2000 .

[33]  Chin-Liang Wang,et al.  Delayed least-mean-square algorithm , 1995 .

[34]  Wei-Der Chang.  Differential evolution-based nonlinear system modeling using a bilinear series model , 2012, Appl. Soft Comput..

[35]  Qing Song,et al.  A linear recurrent kernel online learning algorithm with sparse updates , 2014, Neural Networks.

[36]  Lennart Ljung,et al.  An improved phase method for time-delay estimation , 2009, Autom..

[37]  Zhigang Liu,et al.  A new detection approach of transient disturbances combining wavelet packet and Tsallis entropy , 2014, Neurocomputing.

[38]  Danilo Comminiello,et al.  Functional Link Adaptive Filters for Nonlinear Acoustic Echo Cancellation , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[39]  Ganapati Panda,et al.  Active mitigation of nonlinear noise Processes using a novel filtered-s LMS algorithm , 2004, IEEE Transactions on Speech and Audio Processing.

[40]  R. Seara,et al.  Stochastic analysis of the modified DNLMS algorithm for Gaussian data , 2006, 2006 International Telecommunications Symposium.

[41]  Lu Lu,et al.  Functional link artificial neural network filter based on the q-gradient for nonlinear active noise control , 2018, Journal of Sound and Vibration.

[42]  Fuyun Ling,et al.  Corrections to 'The LMS algorithm with delayed coefficient adaptation' , 1992, IEEE Trans. Signal Process..

[43]  Songwu Lu,et al.  Robust nonlinear system identification using neural-network models , 1998, IEEE Trans. Neural Networks.

[44]  Rui Seara,et al.  Leaky delayed LMS algorithm: stochastic analysis for Gaussian data and delay modeling error , 2004, IEEE Transactions on Signal Processing.

[46]  Jer-Nan Juang,et al.  Continuous-time bilinear system identification using single experiment with multiple pulses , 2012 .

[47]  V. J. Mathews,et al.  Polynomial Signal Processing , 2000 .

[48]  Pascal Frossard,et al.  Distributed Signal Processing via Chebyshev Polynomial Approximation , 2011, IEEE Transactions on Signal and Information Processing over Networks.

[49]  Pramod Kumar Meher,et al.  A high-speed FIR adaptive filter architecture using a modified delayed LMS algorithm , 2011, 2011 IEEE International Symposium of Circuits and Systems (ISCAS).

[50]  Giovanni L. Sicuranza,et al.  A new class of FLANN filters with application to nonlinear active noise control , 2012, 2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO).

[51]  Nithin V. George,et al.  Design of Adaptive Exponential Functional Link Network-Based Nonlinear Filters , 2016, IEEE Transactions on Circuits and Systems I: Regular Papers.

[52]  Fuyun Ling,et al.  The LMS algorithm with delayed coefficient adaptation , 1989, IEEE Trans. Acoust. Speech Signal Process..

[53]  Erasmo Carrera,et al.  Hierarchical theories of structures based on Legendre polynomial expansions with finite element applications , 2017 .

[54]  Mingyu Li,et al.  Complex-Chebyshev Functional Link Neural Network Behavioral Model for Broadband Wireless Power Amplifiers , 2012, IEEE Transactions on Microwave Theory and Techniques.

[55]  Mohammed Bennamoun,et al.  A Novel Adaptive Kernel for the RBF Neural Networks , 2016, Circuits, Systems, and Signal Processing.

[56]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[57]  Indra Narayan Kar,et al.  On-line system identification of complex systems using Chebyshev neural networks , 2007, Appl. Soft Comput..

[58]  David R. Selviah,et al.  Real-Time Nonlinear Parameter Estimation Using the Levenberg–Marquardt Algorithm on Field Programmable Gate Arrays , 2013, IEEE Transactions on Industrial Electronics.

[59]  Giovanni L. Sicuranza,et al.  Simplified volterra filters for acoustic echo cancellation in GSM receivers , 2000, 2000 10th European Signal Processing Conference.

[60]  Jean-Pierre Richard,et al.  Time-delay systems: an overview of some recent advances and open problems , 2003, Autom..

[61]  Lamine Mili,et al.  A Robust Generalized-Maximum Likelihood Unscented Kalman Filter for Power System Dynamic State Estimation , 2018, IEEE Journal of Selected Topics in Signal Processing.

[62]  Dharma P. Agrawal,et al.  Vectorization of the DLMS transversal adaptive filter , 1994, IEEE Trans. Signal Process..

[63]  Douglas L. Maskell,et al.  Chebyshev Functional Link Neural Network-based modeling and experimental verification for photovoltaic arrays , 2012, The 2012 International Joint Conference on Neural Networks (IJCNN).

[64]  Jiandong Duan,et al.  Chebyshev Functional Link Artificial Neural Network Based on Correntropy Induced Metric , 2018, Neural Processing Letters.

[65]  Ercan E. Kuruoglu,et al.  Nonlinear Least lp-Norm Filters for Nonlinear Autoregressive alpha-Stable Processes , 2002, Digit. Signal Process..

[66]  Weidong Luo,et al.  Robust on-line nonlinear systems identification using multilayer dynamic neural networks with two-time scales , 2013, Neurocomputing.

[67]  M. W Gardner,et al.  Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences , 1998 .