I am studying a fast lightning transient signal using HIFREQ and FFTSES. In order to get a good resolution of the lightning signal in the time domain, I have to use a large number of time samples (2^11) and a short time window (300 microseconds). As a result, the program recommends computation frequencies up to 4 MHz. Is it necessary to run all these frequencies?
Generally, you do not have to run HIFREQ at all frequencies recommended by FFTSES. You only have to run HIFREQ at zero Hertz, as well as any other frequencies you believe are appropriate. For instance, you can choose to ignore frequencies recommended by FFTSES that are higher than a certain cut-off frequency if you realize that the input signal is too small past that frequency or if you expect the response function for the GPR or electromagnetic fields of interest to be negligible at those frequencies. The actual value of the cut-off frequency depends on the case being studied. It could be 1 MHz or even as low as 500 kHz. The criterion is that the modulated spectrum (product of the response and the input signal) should be negligeably small for frequencies larger than the cut-off.
No Related Articles Available.
No Attachments Available.
No Related Links Available.
No user comments available for this article.