Module spectrum_analyzer::scaling

This module contains convenient public transform functions that you can use as parameters in crate::samples_fft_to_spectrum for scaling the frequency value (the FFT result). They act as “idea/inspiration”. Feel free to either compose them or create your own derivation from them.

Structs

• SpectrumDataStats
  Helper struct for SpectrumScalingFunction that is passed into the scaling function together with the current frequency value. This structure can be used to scale each value. All properties reference the current data of a crate::spectrum::FrequencySpectrum.

Functions

• combined
  Combines several scaling functions into a new single one.
• divide_by_N
  Divides each value by N. Several resources recommend that the FFT result should be divided by the length of samples, so that values of different samples lengths are comparable.
• divide_by_N_sqrt
  Like divide_by_N but divides each value by sqrt(N). This is the recommended scaling in the rustfft documentation (but is generally applicable). See https://docs.rs/rustfft/latest/rustfft/#normalization
• scale_20_times_log10
  Calculates the base 10 logarithm of each frequency magnitude and multiplies it with 20. This scaling is quite common, you can find more information for example here: https://www.sjsu.edu/people/burford.furman/docs/me120/FFT_tutorial_NI.pdf
• scale_to_zero_to_one
  Scales each frequency value/amplitude in the spectrum to interval [0.0; 1.0]. Function is of type SpectrumScalingFunction. Expects that SpectrumDataStats::min is not negative.

Type Aliases

• SpectrumScalingFunction
  Describes the type for a function that scales/normalizes the data inside crate::FrequencySpectrum. The scaling only affects the value/amplitude of the frequency, but not the frequency itself. It is applied to every single element.