| 摘要 |
By analyzing especially repetitive signals, e.g. signals which represent heart pulse or respiration of animals or humans, a time varying signal is measured, which after being weighted in a part of the signal are transformed to the frequency domain, e.g. by a Fourier transformation. The transformed signal, which besides useful signal components can contain noise, is then signal processed, in a matter suitable for identifying the fundamental frequencies of the signal. The signal processing is carried out by sampling the frequency transformed signal by a number of whole multiples of the analyses frequency F to X points, and multiplying the value with each other in order to create a value T. For each of the X point, Y points are sampled, where the difference in frequency between X and Y is ½ F. By multiplying the value form the Y points, N is formed. A relation is created between T an N, e.g. in the form of T/N, which can e.g. be displayed in a co-ordinate system. The process is repeated with a number of new analysis frequency values, giving the result that the fundamental frequencies of an examined signal are clearly visualized and thereby can be identified. |
