MATLAB: What is the meaning of “scaled white noise” and “unscaled white noise” options in the Wavelet Analyzer

denoisewaveletWavelet Toolboxwdenoise

What is the meaning of "scaled white noise" and "unscaled white noise" options in the Wavelet Analyzer?

Best Answer

The meaning of "Unscaled white noise", "Scaled white noise", and "Non-white noise" correspond to the "one", "sln", and "mln" options in the legacy wden() function. You can see the help for WDEN:

>> doc wden

Please note that if denoising a signal is the goal, the waveletSignalDenoiser app is the recommended option over the legacy waveletAnalyzer UI.

Related Solutions

MATLAB: How to get the scales used with the “Dyadic default” option in the Wavelet Toolbox

The function "cwtft" returns a structure that among other things contains the scales used for the continuous wavelet transform. For this function the default wavelet type is Morlet which has a default value of 'scales' corresponding to the scales used with the "Dyadic default" option in the GUI. You can get these scales by passing "cwtft" any signal and reading the "scales" parameter of the structure. The following code loads a set of data shipped with MATLAB, passes it to "cwtft", and then stores the scales.

>> load kobe
>> out = cwtft(kobe);
>> dyadicDefault = out.scales

The documentation for "cwtft" also has the exact functions used for the "Default dyadic" scales under More About->Morlet Wavelet.

MATLAB: Should the “Receiver Thermal Noise” block affect the signal strength

The "Receiver Thermal Noise" block should not change the signal's strength at its peaks (i.e. dB). It does however add noise around the band of interest.

In this case, other filtering blocks are likely affecting the signal strength - it should not be related to the "Receiver Thermal Noise" block.

Consider the following when working with this type of high frequency model:

1. Plot multiple signals on one analyzer when possible to avoid mismatches with different sampling frequencies.

2. Set your simulation run time to something small (say <1s) for quicker debugging (especially for high frequency signals).

3. Debug component wise and break up your model to analyze the frequency response of each individual part.

4. Be conscience of real vs. complex sine waves and always check the spectra to see if it is the desired shape before any manipulations (sending to a product block, filtering, etc.)

Best Answer

Related Solutions

MATLAB: How to get the scales used with the “Dyadic default” option in the Wavelet Toolbox

MATLAB: Should the “Receiver Thermal Noise” block affect the signal strength

Related Question