Band Analysis
Refers to how we can [see page 9, alter] the time/frequency resolution to better analyse what we're interested in, for example when running a fourier-transform.
See the graphical example [see page 10, here].
Narrowband analysis
We have a larger number of samples (increase the analysis frame \(N\)).
This:
- Decreases the spacing between spectral components.
- Reduces the ability to respond to changes in the signal.
However it gives:
| Affect | Meaning |
|---|---|
| Good spectral analysis | Gives a good overview of important sections of the spectrogram. |
| Poor time resolution | Harmonics can move so fast that they don't appear at expected spacing. |
Wideband Analysis
We have a less samples (decrease the analysis frame \(N\)).
This:
- Increases the spacing between the spectral components.
- Increases the ability to respond to changes in the signal.
Poor spectral resolution, high temporal resolution.
| Affect | Meaning |
|---|---|
| Good Time Resolution | Gives a lot of fine detail in how the frequency changes over time. |
| Poor Spectral Resolution | Blurs frequency along the x-axis making it hard to get an overview high level change. |