My goal - I'm interested in the white noise spectral region at higher frequencies, especially the phase information.
In this data acquisition instrument, digitised time data transfer is much faster for processing when "segmented", but with a twist that I am not sure about.
Example: Sampling rate of 60 MHz and ~1M samples acquired each segment
10 time-trace segments each with ~1 MSa (nearly 21.8 ms) but a "dead-time" between each segment of 10 us + small jitter up to 10 ns.
i.e. 10 segments of ~1 MSa with 10 us delay between each segment, total of 10 MSamples (~ 170.6 ms)
In post-processing I concatenate these 10 segments as one time trace of 10 MSa. But what is the effect of the 10 us delay between each segment in the frequency domain? I expect to see something along the lines of discontinuities... but what I'm be measuring is just white noise at high frequencies.
Alternatively: Just acquire one time trace of 10 MSa and FFT that.
Still learning Fourier transforms as I'm doing this and I'd like some advice please.