摘要
在分析模糊神经网络的模型、系统结构和学习算法的基础上,提出了基于自适应神经模糊推理系统(ANFIS)的自适应噪声抵消算法,并对算法进行了仿真和分析.仿真结果表明,在合理选取隶属度函数类型及其数目的条件下,ANFIS能够根据训练样本对隶属度函数参数等系统参数进行优化设计,从而大大提高模糊滤波输出的信噪比.
On basis of analysing the model,system structure and learning algorithms of fuzzy neural network,the paper puts forward an adaptive noise canceling algorithm based on adaptive neuro-fuzzy inference system(ANFIS),and performs algorithmic simulation and analysis.Simulation results show,in condition of reasonably selecting the type and number of the membership functions,ANFIS can obtain the optimal design for the system parameters such as the membership function parameters according to the training samples,thus greatly improve the output signal to noise ratio(SNR) of the fuzzy filter.
出处
《赣南师范学院学报》
2011年第3期35-39,共5页
Journal of Gannan Teachers' College(Social Science(2))
关键词
自适应神经模糊推理系统
非线性逼近
噪声对消
adaptive neuro-fuzzy inference system
nonlinear approximation
noise canceling