I have a 2D graph that I have created of frequency vs. amplitude as a way to characterize a signal. I am trying to figure out how to write script that takes the WAV I have already imported into MATLAB, and figure out the peaks, the troughs, and the distance between each one. Can someone please help me with some general guidelines, not a complete code, for a script that will help me do this?
There are several submissions on the file exchange for this. For instance this one, seems applicable (I haven't tested it myself, but it looks good).
Try to search a bit on the file exchange and you will likely find something you can use.
You can also have a look at this question, and this one.
Related
I would like to ask you if there is any automated method for calculating the order of ARIMA(p,d,q) model for any type of a time series data, in MATLAB.
This will make the forecasting model more accurate and will also save me some time. I would appreciate if anybody could help me.
I'm not sure if that's what you are looking for, but maybe this post on Matlab File Exchange?
http://www.mathworks.com/matlabcentral/fileexchange/25611-arfima-simulations
I want to implement some kind of improvement of DBSCAN algorithm, where user do not need to enter input parameters (minPts and Eps). My idea is to use the K-distances plot, but what is the best method to compute the 'knee' of this plot? How to count when there are 2 or more knees on the plot?
Where I can find the source code for some DBSCAN improvement, like AUTODBSCAN, VDBSCAN, PDBSCAN or DBSCAN-DLP? Im looking for some basics, but nowhere I can find a good help. Maybe you've seen somewhere sample source codes?
DBSCAN has already been improved to death.
In Google Scholar, it has 5361 citations, and probably 1000+ of these "improve" DBSCAN. And probably a dozen of these use the k-distance plot. But none of these are used in practise.
If you want to continue this line of research, best get updated on what has been done since.
In particular, have a look at OPTICS which does away with the Epsilon parameter completely (except for performance reasons when using indexes).
Also have a look at HDBSCAN* by one of the original DBSCAN authors, Joerg Sander. That will likely be the most important DBSCAN extension besides his work on OPTICS and GDBSCAN.
I have a set of 3D points which numbers up around 1 million points. I am looking to visualise these with matlab.
I have tried the following functions:
plot3
scatter3
But they are both very sluggish. Is there a more efficient way to visualise this level of points in matlab? Maybe a way to mesh the points?
If not can anyone suggest a plug-in or even a different program for visualising 3D points?
You're going to run into efficiency issues no matter what plugin/program you use if you want all million+ points to show up in a plot. My suggestion would be to downsample. Use the plot3 or scatter3 function on every other point, or every nth point, until you get a figure that is not sluggish. As long as the variance in your data isn't astronomical, downsampling a little bit shouldn't affect the overall distribution of points (given that you have a million+ points). And any software that is able to display that much data without being sluggish is most likely downsampling/binning or using some interpolation technique to do so (so you might as well have control over it).
fscatter3 from the file exchange, does what you like.
Is there a specific reason to actually have it display that many points?
I Googled around a bit and found some people who have had similar issues (someone suggested Avizo as an alternate program but I've never used it):
http://www.mathworks.com/matlabcentral/newsreader/view_thread/308948
mathworks.com/matlabcentral/newsreader/view_thread/134022 (not clickable because I don't have enough rep to post more than two links)
An alternate solution would be to generate a histogram if you're more interested in the density of the data:
http://blogs.mathworks.com/videos/2010/01/22/advanced-making-a-2d-or-3d-histogram-to-visualize-data-density/
I you know beforehand roughly the coordinates of the feature you are looking for, try passing the cloud through a simple pass-through-filter, which essentially crops your point cloud. I.e. if you know that the feature is at a x-coordinate > 5, remove all points with x-coordinate < 5.
This filter could for the first coordinated be realized as
data = data(data(1,:) > 5,:);
Provided that your 3d data is stored in an n by 3 matrix.
This, together with downsampling, could help you out. If you still find the performance lagging, consider using something like the PCD viewer in PointCloudLibrary, check half way down the page at
http://pointclouds.org/documentation/overview/visualization.php
It is a stand alone app you could launch from matlab. I find it's performance far better than the sluggish matlab plotting tools.
For anyone who is interested I ended up finding a Point cloud visualiser called Cloud Compare. It is extremely fast and allows selection and segmentation as well as filtering on point clouds.
I'm looking for some functions in MATLAB in order to find out some parameters of sound,such az intensity,density,frequency,time and spectral identity.
i know how to use 'audiorecorder' as a function to record the sampled voice,and also 'getaudio', in order to plot it.But i need to realize the parametres of a sampled recorded voice,that i mentioned above.i'd be so thankful if anyone could help me.
This is a very vague question, you may want to narrow it down (at first) and to add as much contextual details as you can, it will certainly attract a lot more answers (also as mentionned by Ion, you could post it at http://dsp.stackexchange.com).
Sound intensity: microphones usually measures pressure, but you can get the intensity from that quite easily (see this question). Your main problem is that microphones are not usually calibrated, this means that you cannot associate an amplitude with a pressure. You can get sound density from sound intensity.
Frequency: you can get the spectrum of your sound by using the Fast Fourier Transform (see the Matlab function fft).
As for spectral or time identity, I believe these are psychoacoustics notions, which is not really my area of expertise.
I'm no expert but I have played with Matlab a little in the past.
One function I remember was wavread() to input a sound signal into Matlab, which if executed in this form [Y, FS, NBITS]=WAVREAD("AUDIO.WAV") would return something like:
AUDIO.WAV:
Fs = 100 kHz
Bits per sample = 10
Size = 100000
(numbers from the top of my head)
Now about the other things you ask, I'm not really sure. You can expect a better answer from somebody else. I think this question should be moved to Signal Processing SE btw.
One part of an app I'm currently working on will work as a tuner. I want to be able to use the iPhone to display the peak frequency of a signal given by the user. I have used the SCListener which worked very good on the iPhone simulator. However when I tried it on a real device it didn't.
Forums suggests that I use apple FFT and accelerate Framework to do this but it seems overly complicated. I would really appreciate if anyone that has programmed a tuner or similar could point me in a good direction!
Thanks!
There is a related post on dsp.stackexchange. It suggests that autocorrelation will work better than FFT at finding the fundamental, if the fundamental is lower in amplitude than the harmonics. Autocorrelation is slightly less tricky than FFT. The accelerate framework will come to your help there again for that. However this is not the case usually.
I don't know of any out of the box solutions which will do all the work for you. The vDSP Programming Guide has specific worked examples for real FFTs which you might want to look into, it takes some getting used to, but it's worth it really. FFT seems like the most logical first step in peak frequency extraction I'm afraid. Most sources seem also to suggest that applying a windowing function to the time domain signal before running the FFT is critical (or you will get high frequency artifacts because of discontinuities at the extremities).
Also you might want to check out this related SO post.
Peak frequency is often different from the pitch frequency that one would want a (music) tuner to estimate. Look up pitch estimation.
From previous experience doing this :
FFT isn't as always as accurate as you might think, and is computationally expensive
Autocorrelation gives pretty good results
If you have a strong fundamental, zero-crossing can be very accurate and is very computationally efficient (just count the number of times the signal crosses zero over a period of time, f = (2 x time period in seconds)/(number of zero crossings)\
Hope that helps.
Thanks for all the answers! I had missed a part in my code to make the SC listener work on the device as well but are now trying to change it for Apples own AVAudioRecorder since it is suppose to be a lot faster. The problem was that the cocos2d framework blocked the recording of sounds until you called for a method that allowed this. It works like a charm now! :)
Thanks again!