【摘要】 针对实际采集脉搏信号中各种噪声对信号质量的影响,提出一种循环平稳算法的脉搏信号噪声评估指标——质量因数,以评估脉搏信号受噪声影响程度。采用建模产生的脉搏信号、MIT-BIH数据库提供的脉搏信号与噪声信号叠加产生的试验数据进行验证。试验结果表明,质量因数可以准确地反映脉搏信号受噪声影响程度,脉搏信号信噪比越大,质量因数越大,则信号受噪声影响越小。该结果对正常、异常脉搏信号都适用。实际采集的较长脉搏信号,引入不同时宽的滑窗计算信号的时变质量因数,提取时变质量因数最大或较大的那段脉搏信号进行分析,减化了去除噪声等预处理过程,节省了计算量和时间。计算机完全可以自动计算质量因数,无需人工干预,为脉搏信号噪声的评估提供了新思路。更多还原

【Abstract】 In order to reduce the impact of various noises in actual pulse signal,a noise evaluation index(quality factor) was proposed to evaluate the influence of noise on pulse signals,based on cyclostationary algorithm.The test results of verification with the pulse signal generated by modeling and the experimental data generated by superposition of pulse signal and noise signal provided by MIT-BIH database showed that the quality factor could accurately reflect the extent to which the pulse signal was affected by noise.The higher the SNR of pulse signal was,the greater the quality factor was,indicating that the signal was less affected by noise,which was applicable to both normal and abnormal pulse signals.For the longer pulse signals actually collected,sliding Windows with different time width were introduced to calculate the time-varying quality factor of the signal,and the pulse signals with the largest or larger time-varying quality factor were extracted for analysis,by which the pretreatment process such as noise removal was simplified,and the computation amount and time were reduced.The quality factor could be calculated by the computer automatically without manual intervention,which provided a new idea for the evaluation of pulse signal noise.更多还原