Adaptive signal processing
Adaptive filter theory
Structure identification of fuzzy model
Fuzzy Sets and Systems
International Journal of Approximate Reasoning
Neuro-fuzzy and soft computing: a computational approach to learning and machine intelligence
Neuro-fuzzy and soft computing: a computational approach to learning and machine intelligence
Neural Fuzzy Control Systems with Structure and Parameter Learning
Neural Fuzzy Control Systems with Structure and Parameter Learning
An adaptive neural fuzzy filter and its applications
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Dynamic fuzzy neural networks-a novel approach to functionapproximation
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
A recurrent fuzzy-neural model for dynamic system identification
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
An online self-constructing neural fuzzy inference network and its applications
IEEE Transactions on Fuzzy Systems
Identification and control of dynamic systems using recurrent fuzzy neural networks
IEEE Transactions on Fuzzy Systems
Noisy speech processing by recurrently adaptive fuzzy filters
IEEE Transactions on Fuzzy Systems
A fast approach for automatic generation of fuzzy rules by generalized dynamic fuzzy neural networks
IEEE Transactions on Fuzzy Systems
IEEE Transactions on Neural Networks
A recurrent self-organizing neural fuzzy inference network
IEEE Transactions on Neural Networks
Thresholding neural network for adaptive noise reduction
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
In this paper, an online self-improved fuzzy filter (OSFF) is proposed. It is based on radial-basis-function networks (RBFN) and implements the TSK fuzzy systems functionally. As a prominent feature of OSFF, the system is hierarchically constructed and self-improved in the training process with a novel online clustering strategy for structure identification. Moreover, the filter is adaptively tuned to be an optimal status by a hybrid sequential algorithm for parameters determination. In detail, the proposed OSFF system has the following features: (1) Hierarchical structure self-construction. There is no predetermination initially for OSFF, i.e., it is not necessary to determine the initial number of fuzzy rules and input data space clustering in advance. The fuzzy rules, i.e., the RBF neurons are generated automatically in training process under a proposed criterion, minimum firing strength (MFS). (2) Online clustering. Instead of selecting the centers and widths of membership functions arbitrarily, an online clustering method is applied to ensure the reasonable representation of input terms of an input variable. It not only ensures the proper feature representation, but also optimizes the structure of the filter by reducing the number of fuzzy rules. (3) All free parameters in the premise and consequence parts are online determined by a hybrid sequential algorithm without repeated computation to make real-time applications possible. The centers and widths of membership functions of an input variable are allocated initially in the scheme of structure identification and optimized in the scheme of parameters determination. The parameters in the consequent parts of OSFF are updated in each iteration by a sequential recursive algorithm. Due to the hybrid learning algorithm, low computation load and less memory requirements are achieved. Simulation results, compared with other similar approaches for some benchmark problems, show that the proposed OSFF system can tackle these problems with fewer fuzzy rules and obtain better or same accuracy with lower system resource requirements.