How it works:
1. Most of the FastPTC program is relatively straight forward - the only bit that is not is the
sweeping of the noise. This is acheived by the following method:
2. Generate several seconds of a low-pass noise of half the desired bandwidth
3. On each masker presentation take a random chunk of this low-pass noise
4. Mutiply this chunk of low-pass noise in the time domain by a sinusoid of the chosen
frequency (i.e. desired centre frequency of the noise band). Remember, multiplication in the
time domain is equivalent to convolution in the frequency domain
5. Apply appropriate ramps to the chunk of (now band-pass) noise
6. Present the noise (and tone)
7. Repeat the process with a new random chunk of the noise and a sinusoid of
corresponding to the next centre frequency of the masker
8. In other words, it is not a true swept noise, but a series of noise presentations at discrete
steps
9. The user can choose the range over which the noise sweeps, the rate of the sweep and
the size of the amplitude steps over which the noise is tracked.
10. Because the noise is presented in discrete bursts rather than continuously, the target
tone is presented in the middle of noise burst, and by default is amplitude modulated to make
it easier to distingish from the noise.