Nakamura M, Nishida S, Shibasaki H
Department of Electrical Engineering, Saga University, Japan.
IEEE Trans Biomed Eng. 1991 Mar;38(3):309-12. doi: 10.1109/10.133216.
In order to improve the signal-to-noise ratio of evoked potentials, the responses to repetitive stimuli are usually recorded, and the averaging method is applied to them. If each starting point of the averaging varies with respect to the response because of asynchronous sampling of the measurement data, the average waveform of the evoked potential signal is distorted by the averaging. We analyzed the asynchronous averaging in the frequency domain and derived a relationship between the fluctuating time and the degree of the deterioration. By using this relationship, a required sampling rate was derived to ensure the accuracy of the averaging. Furthermore, we also deduced a compensation procedure for the asynchronous averaging in case of coarse sampling.
为了提高诱发电位的信噪比，通常会记录对重复刺激的反应，并对其应用平均法。如果由于测量数据的异步采样，平均的每个起始点相对于反应发生变化，诱发电位信号的平均波形就会因平均而失真。我们在频域中分析了异步平均，并推导出波动时间与恶化程度之间的关系。利用这种关系，得出了确保平均准确性所需的采样率。此外，我们还推导了粗采样情况下异步平均的补偿程序。
