基于神经网络的自适应噪声抵消的研究-通信与信息系统专业论文.docxVIP

太原理工大学硕士研究生学位论文网络自适应滤波器是一种较为理想的自适应噪声抵消系统。 太原理工大学硕士研究生学位论文 网络自适应滤波器是一种较为理想的自适应噪声抵消系统。 (5)应用神经网络自适应噪声抵消系统来检测胎JL,tL,率，并进行 了计算机仿真。仿真结果表明，神经网络自适应噪声抵消系统对于微 弱信号的提取和处理具有重要的实用价值。 关键词： 自适应噪声抵消，神经网络，BP算法，胎儿心率，LMS算法， 语音增强 Ⅱ 太原理工大学硕士研究生学位论文T强E 太原理工大学硕士研究生学位论文 T强E S罩UpY OF ADAPTIVE NOISE CANCELLATlON BAS蓦D 0N艟U＆≮毛NETWORK ABSTRACT Noise cancellation is very important in signal processing，Adaptive noise cancellation technique can eliminate background noise in effect．It has become a major focus in the field ofsignal processing． The artificial neural network iS a theoretical mathematical model for the cerebral neural networks；it is嬲information-processing system that is based on stimulating the structure and function of the cerebral neural networks；it Can perform complicated operations and establish nonlinear relationships．The use of neural networks for designing a adaptive noise cancellation system is important theoretical significance and practical value。 The primary subject ofthis paper is to： (1)讹e Basic principle of adaptive filter and adaptive noise cancellation system is systematically introduced．Otherwise,some relative p越ameters implying the performance of noise cancellation are analyzed under introducing the algorithm ofLMS。 (2)Characters offeed-forward neural networks ofsingle-hide—layer and multi-hide-layers are mostly researched and analyzed under summarizing the design feature and leaning algorithm ofneural network． (3)Adaptive noise cancellation system based on four layers feed．forward neural network is established and its iteration fonimlation is III 太原理工大学硕士研究生学位论文deduced． 太原理工大学硕士研究生学位论文 deduced． (4)MATLAB simulations indicates that the new algorithm has good abilities of noise cancellation comparing wi像the simulation result of transversal adaptive FIR．In fact,A Neural nemork adaptive filter is a ideal noise cancellation system． (5)Detecting fetal heart rate by Neural network adaptive noise cancellation system and MATLAB simulation．Simulat