A Novel Modular Recurrent Wavelet Neural Network and Its Application to Nonlinear System Identification
作者： Haiquan Zhao , Xiangping Zeng
DOI: 10.1007/978-3-642-37829-4_11
关键词: Artificial intelligence 、 Modular design 、 Time delay neural network 、 Machine learning 、 Nonlinear system 、 Nonlinear system identification 、 Gradient descent 、 Algorithm 、 System identification 、 Recurrent neural network 、 Computer science 、 Computational complexity theory
摘要: To reduce the computational complexity and improve performance of recurrent wavelet neural network (RWNN), a novel modular based on pipelined architecture (PRWNN) with low is presented in this paper. Its modified adaptive real-time learning (RTRL) algorithm derived gradient descent approach. The PRWNN comprises number RWNN modules that are cascaded chained form inherits architectures (PRNN) proposed by Haykin Li. Since those can be performed simultaneously parallelism fashion, it would result significant improvement efficiency. And also further improved. Computer simulations have demonstrated provides considerably better compared to single model for nonlinear dynamic system identification.
springer.com
sci-hub.st 参考文章(20)
Fengyao Zhao, Liangming Hu, Zongkun Li, Nonlinear System Identification Based on Recurrent Wavelet Neural Network international symposium on neural networks. pp. 517- 525 ,(2009) , 10.1007/978-3-642-01216-7_54
S.S. Rao, B. Kumthekar, Recurrent wavelet networks world congress on computational intelligence. ,vol. 5, pp. 3143- 3147 ,(1994) , 10.1109/ICNN.1994.374736
Haiquan Zhao, Jiashu Zhang, Nonlinear dynamic system identification using pipelined functional link artificial recurrent neural network Neurocomputing. ,vol. 72, pp. 3046- 3054 ,(2009) , 10.1016/J.NEUCOM.2009.04.001
Haiquan Zhao, Jiashu Zhang, A Novel Adaptive Nonlinear Filter-Based Pipelined Feedforward Second-Order Volterra Architecture IEEE Transactions on Signal Processing. ,vol. 57, pp. 237- 246 ,(2009) , 10.1109/TSP.2008.2007105
Haiquan Zhao, Xiangping Zeng, Xiaoqiang Zhang, Zhengyou He, Tianrui Li, Weidong Zhao, Adaptive Extended Pipelined Second-Order Volterra Filter for Nonlinear Active Noise Controller IEEE Transactions on Audio, Speech, and Language Processing. ,vol. 20, pp. 1394- 1399 ,(2012) , 10.1109/TASL.2011.2175383
Haiquan Zhao, Xiangping Zeng, Jiashu Zhang, Tianrui Li, Nonlinear Adaptive Equalizer Using a Pipelined Decision Feedback Recurrent Neural Network in Communication Systems IEEE Transactions on Communications. ,vol. 58, pp. 2193- 2198 ,(2010) , 10.1109/TCOMM.2010.08.080612
C.-J. Lin, C.-C. Chin, Prediction and identification using wavelet-based recurrent fuzzy neural networks systems man and cybernetics. ,vol. 34, pp. 2144- 2154 ,(2004) , 10.1109/TSMCB.2004.833330
Wu Yilei, Song Qing, Liu Sheng, A Normalized Adaptive Training of Recurrent Neural Networks With Augmented Error Gradient IEEE Transactions on Neural Networks. ,vol. 19, pp. 351- 356 ,(2008) , 10.1109/TNN.2007.908647
Chi-Huang Lu, Design and Application of Stable Predictive Controller Using Recurrent Wavelet Neural Networks IEEE Transactions on Industrial Electronics. ,vol. 56, pp. 3733- 3742 ,(2009) , 10.1109/TIE.2009.2025714
J. Park, I. W. Sandberg, Universal approximation using radial-basis-function networks Neural Computation. ,vol. 3, pp. 246- 257 ,(1991) , 10.1162/NECO.1991.3.2.246