Multilayer feedforward networks are universal approximators
Neural Networks
Neuro-Control Systems: Theory and Applications
Neuro-Control Systems: Theory and Applications
Wavelet neural networks for function learning
IEEE Transactions on Signal Processing
LMS learning algorithms: misconceptions and new results on converence
IEEE Transactions on Neural Networks
Choquet fuzzy integral based modeling of nonlinear system
Applied Soft Computing
New fuzzy wavelet neural networks for system identification and control
Applied Soft Computing
Choquet fuzzy integral based verification of handwritten signatures
Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
| Hi-index | 0.00 |
The present work demonstrates an application to approximate, control and denoise a continuous non-linear signal, using wavelet coefficients and neural network. Adaptive Least mean square (∞-LMS)algorithm is used for the parameter adjustment, and wavelet coefficients are used for making the system fast and denoising it. Neural networks have been established as a general approximation tool for fitting nonlinear models from input/output data. On the other hand, the recently introduced wavelet decomposition emerges as a new powerful tool for approximation. The procedure for the wavelet based adaptive control remains the same as for neural network only the concept of compression and denoising the reference signal is adopted. In the adaptive control, identification of plant is done offline and adjustments of controller parameters are performed on-line. The effectiveness of the proposed wavelet neural network architecture as applied to the Identification and control of unknown non-linear systems is discussed and extensive simulation results are presented.