Modeling microphone in Pspice based on Neural Network

摘要：

Presently, the electret condenser microphone (ECM) is used in almost every consumer and communication audio application. In order to improve the efficiency of the circuit design, circuit simulation is necessary. In this article, we present a method based on Neural Network for modeling ECM, by which the fundamental characteristics of an ECM, including sensitivity, directivity, output impedance and frequency response, are modeled in Pspice. Firstly, each input-output characteristic is approximated with different Neural Network, after which the structures, weights and biases of the Neural Network depicting different characteristic of the ECM are acquired. Secondly, the structures are described in Pspice language to form sub-circuits respectively. Finally, these sub-circuits are integrated into a unitary sub-circuit based on the relationship between the inputs and output of the ECM to form the final model.

关键词： neural network pspice microphone model

作者: Li Hua Xiang Lian Ming Wang An Ning Yu

作者单位: Institute of Applied Electronic Technology, Northeast Normal University Chang Chun, China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 28-32

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）