Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
Do You want to recognise the signal i.e. estimate the frequencies, phases and number of tones or just detect the sinusoids from noise?
If You want to detect, then You should try to predict a signal with linear prediction and measure the error
Sinusoidal signal should predict well, so the error...
You have second order section filter coefficients. Try to use this
https://www.mathworks.com/help/signal/ref/sos2tf.html
to convert to Transfer function and then
just filter the signal
https://www.mathworks.com/help/matlab/ref/filter.html
Don't forget about G matrix.
Probably You have to construct 2 matrices. You should have several examples (of Your leaves) to learn Your network, so those examples will be placed in rows. Each example will have features that You extracted earlier. Features will be placed in columns. As a result You should have:
input =...
Try to use resample method:
https://www.mathworks.com/help/ident/ref/resample.html?searchHighlight=resample
But first you have to do the LP filtration to ensure the Nyquist requirement with new sample rate.
Thanks a lot for the answer.
I think that is the key of the kernel method, because the transformation/mapping that You mention above can be done by a simple basis functions (e.g polynomial, sigmoidal etc.)
I just didn't get the difference between those two methods.
Thanks a lot once again...
Hello all, I have several basic questions about Machine Learning to better solidify my knowledge.
I hope to find the answers here.
1. SVM
I have a one dimensional feature vector (x) of length N and the target vector that haves labels of two classes (e.g. 0 and 1). The case is not linearly...
Probably it should be the indexing multiplication and division. Try to change '*' to '.*' and '/' to './'
i.e.
d = round(d0 * L / d0(15));
to
d = round(d0 .* L ./ d0(15));
Basically, decimation algorithms reduce number of samples. The length of the vector is smaller, but the time vector has greater dt.
I'm not sure what result do You want to get.
While You decimate the signal, You will get the same shape of the curve, with the smaller number of samples.
If you...
The MIT-BIH data base is written in 212 format coding.
**broken link removed**
You can use the special toolbox (wfdb) in Matlab to read every file from physionet
**broken link removed**
Basically, the number of feature that can be extracted from fft is unlimited. You can try to calculate the spectral flatness, energy at Your region of interest (or energy ration from different part of regions), or try to find continous peaks along time domain in STFT (using linear prediction or...
Still, there is no information what do You want to detect. Speech recognition? i.e. detect what word You say? Or just appearance of voice in the background noise?
Maybe do You want the select the voiced or unvoiced phonemes?
There is no universal algorithm for feature extraction. The best way is to observe the signal (FFT in Your case) and select the features that are the most correlated with your classes.
If You want to get some advice, You should write more information about Your problem.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.