5.The others three factors mension above have some effects on model. The best preprocessing methods of UTS, YS and ELO model are [0.1, 0.8], [0.01, 0.99] and [0.01, 0.99], and best hidden layers are single hiddenlayer with 10, 11 and 12 neurons.
6.The CGA is applied to optimize the hiddenlayer structure of the RBF networks, at the same time a fitness function is designed, then the CGA-RBFN model is proposed.
7.The analysis shows that: by the sacrifice of compression ratio, the quality of the recovered image can be improved with the increase of the number of the neurons in the hiddenlayer.
8.The idea of back propagation(BP) neural network was adopted in the proposed algorithm. The on-line learning strategy of EBPA with two inputs, single hiddenlayer and two outputs was applied in this scheme. The input layer included given signal and feedback digital position;
9.Back propagation network (BP network)is a non-linearity artificial neural network, and Robert Hecht Nielson has proved that any continuous function in enclosed interval can be approached by performing a BP network with a hiddenlayer.
10.The structure of Higher-order CMAC neural network and the address calculation method of active basis functions in hiddenlayer are studied. In this paper several calculation methods of Higher-order Basis Function are presented.
