基于一种自适应折中平—滑阈值的脉搏信号小波去噪  

Pulse Signal Wavelet De-noising Based on an Adaptive Compromising-Smooth Threshold

作　　者：张浩[1] 陈宏铭[1] 程玉华[1] 

机构地区：[1]上海北京大学微电子研究院

出　　处：《中国集成电路》2013年第8期13-17,共5页China lntegrated Circuit

摘　　要：人体脉搏信号在采集过程中会掺杂着各种噪声信号,而传统所采用的小波变换阈值去噪法在处理脉搏信号时又存在着缺陷,本文采用的折中-平滑阈值函数,并基于斯坦恩无偏估计(SURE)优化算法去寻找最优阈值门限,与在原有阈值基础上的小波自适应阈值去噪做对比,通过利用Matlab对脉搏信号进行去噪仿真。结果表明,依据信噪比(SNR)和信号平滑度[1]的评判标准,该方法去噪效果更好,结果令人满意。The pulse signals may be mixed with all kinds of noises in the collection process. There exist many defects in the traditional wavelet threshold de-noising method when used in the treatment of pulse signals. In this paper, a new Compromising-Smooth threshold function based on the Stein＇ s Unbiased Risk Estimation （ SURE ） optimizing algorithm is adopted for the optimal threshold. Through the simulation of the pulse signals de-noising based on Matlab and compared with wavelet adaptive threshold based on the traditional threshold de-noising, we get that the new method can reach a better result, according to the signal to noise ratio （ SNR ） and signal smoothness . The result is satisfying.

关 键 词：脉搏信号 小波阈值去噪 SURE 自适应阈值 

分 类 号：TN911.4[电子电信—通信与信息系统]