Blind Detection Algorithm Based on Double Sigmoid Wavelet Chaotic Neural Network
Liu Hua · Video Engineering · 2015
For the defects of transient chaotic neural network( TCNN) in the blind detection environment,a new blind detection algorithm based on double sigmoid wavelet chaotic neural network( DSWCNN) is proposed,constructing the model and a new energy function and proving the stability of DSWCNN in asynchronous update mode and synchronous update mode separately. The design philosophy of the new network: adopting the activation function constituted by Mexican hat wavelet function and Sigmoid function,then adding a activation function for the each nerve cell to constitute double Sigmoid. The simulation shows that,because of the strong ability of approximation of the wavelet and the rapid convergence properties of double Sigmoid,the algorithm presented in this paper improves the global optimizing ability and optimizing precision.