How to Test the trained model using IQ dataset

6 visualizaciones (últimos 30 días)
Med Future
Med Future el 24 de Ag. de 2023
I trust you're doing great. I've completed the initial segment of the provided MATLAB Example at the link you shared (https://www.mathworks.com/help/radar/ug/radar-and-communications-waveform-classification-using-deep-learning.html).
Since the example generates IQ-based input data and saves it as images using the Wigner-Ville distribution, I now possess Test Data in IQ format (Test.mat file is attached for Testing IQ data). My aim is to test the Model.
Could you kindly advise on steps to convert the IQ format data into Wigner-Ville distribution Images and proceed with model testing, following the example's guidance?
The following is the code for generating IQ dataset for Training. Please modified the below code for test on IQ dataset
function helperGenerateTFDfiles(parentDir,dataDir,wav,truth,Fs)
% This function is only intended to support
% ModClassificationOfRadarAndCommSignalsExample. It may change or be
% removed in a future release.
[~,~,~] = mkdir(fullfile(parentDir,dataDir));
modTypes = unique(truth);
for idxM = 1:length(modTypes)
modType = modTypes(idxM);
[~,~,~] = mkdir(fullfile(parentDir,dataDir,char(modType)));
end
for idxW = 1:length(truth)
sig = wav{idxW};
TFD = wvd(sig,Fs,'smoothedPseudo',kaiser(101,20),kaiser(101,20),'NumFrequencyPoints',500,'NumTimePoints',500);
TFD = imresize(TFD,[227 227]);
TFD = rescale(TFD);
modType = truth(idxW);
imwrite(TFD,fullfile(parentDir,dataDir,char(modType),sprintf('%d.png',idxW)))
end
end

Respuestas (0)

Productos


Versión

R2022a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by