Draganova R
Institute of Physiology, Bulgarian Academy of Sciences, Sofia.
Med Eng Phys. 1999 Jan;21(1):9-14. doi: 10.1016/s1350-4533(99)00017-x.
The method of linear transformation (LT) in the modification known as method of Buys Ballot was examined in its ability to reveal polyharmonic periodicities in EEG signals. Applying this method to simulated signals-additive mixtures of sinusoids, periodic squared waves and Gaussian noise some rules were found, which can be used in analyzing of EEG signals. The results were compared to that of Discrete Fourier analysis. The conclusion was made that in contrast to the method of Fourier giving the frequency contents, LT method is able to reveal not only the mono- but also polyharmonic components with their waveshape (the reconstruction) even if their frequency bands are overlapping.
在被称为布伊·巴洛特方法的一种改进方法中,研究了线性变换(LT)方法揭示脑电图(EEG)信号中多谐波周期性的能力。将该方法应用于模拟信号——正弦波、周期性方波和高斯噪声的叠加混合物时,发现了一些可用于分析EEG信号的规则。将结果与离散傅里叶分析的结果进行了比较。得出的结论是,与给出频率成分的傅里叶方法不同,即使其频段重叠,LT方法不仅能够揭示单谐波成分,还能揭示具有波形(重构)的多谐波成分。