The Wavelet Neural Network Blind Equalization Algorithm Based On Fuzzy Neural Network Controlling

At present, the chief obstacle of high speed underwater sound correspondence is the severe inter-symbol interference, caused by multi-path effects and channel distortion of sound propagation, degrades the relialility and decreases the rate of the propagation. So, it has the important meaning to analysis and research the theory and the algorithm of blind equalize in view of the underwater sound channel.Using Multi-wavelet theory, Neural network, Fuzzy theory and making full use of their superiority, and aiming at the disadvantage of CMA (Constant Modulus Algorithm), the wavelet neural network blind equalization algorithm based on Fuzzy neural network controlling are studied deep in this dissertation. The main contributions are as follows:1. Analying variable step-sizes blind equalization algorithms based on Fuzzy Neural Network control(1) The article has conducted the thorough research to the development of the Fuzzy theory and neural network, variable step-sizes blind equalization algorithms based on Fuzzy Neural Network control is proposed. via using the excellent features of them. The computer simulations proved the efficiency of the algorithm:a relatively low mean square error, fast convergence rate and good noise immunity have been achieved.2. Hybrid Wavelet Neural Network blind equalization algorithm based on Fuzzy Neural Network controllingTaking the Fuzzy Neural Network as the controller of step-sizes, structuring the fuzzy rules via the knowledge of human experience, and controlling the step-sizes via fetching intormation from the Mean Square Error, It adopts big step-size to improve the algorithm convergence rate in the beginning, and adopts small step-size to improve the algorithm convergence precision when the algorithm is coverged. It can solve the contradiction between the convergence rate and convergence precision.3. Dynamic Fuzzy Wavelet Neural Network Feedback blind equalization algorithmIn wireless communications systems, the process of blind equalization is a dynamic process because of the time-varying and uncertain characteristic of the channel. So, It necessary to study the Dynamic Fuzzy Wavelet Neural Network Feedback blind equalization algorithm(1) Dynamic Wavelet Neural Network Feedback blind equalization algorithm based on Fuzzy ControlThe proposed algorithm has the memory function via taking the output of wavelet neural network as the part of the input in the next moment, so, the dynamic function of the network is shown.(2) Dynamic Recurrent Fuzzy Wavelet Neural Network Blind Equalization AlgorithmIn fact, The proposed algorithm adjusts the parameters of the fuzzy wavelet neural network, then dynamically adjusts the entire network, and the equalixing effect is attained.