DSPcorner
DSPcorner
  • Видео 21
  • Просмотров 98 752
#HowTo - IIR-Filters : Basics in GNU Octave
In this video, I am showing you quick&dirty the basics of IIR-filters. IIR stands for infinite impulse response filters (remember: the second type of digital filters are the FIR filters - finite impulse response filters). IIR filters generally have a feedback loop e.g. to produce a new output sample, we will need a previous output sample. I will show you some basic math that we need in order to be able to describe an IIR filter and to implement it in GNU Octave (spoiler: we need the filter coefficients). We will start with the difference equation. We will then transform this difference equation to the frequency domain using the Z-transform (ya it's the frequency domain when we are replaci...
Просмотров: 3 423

Видео

#HowTo - Plot A Probability Density Function Into A Histogram In GNU Octave
Просмотров 5 тыс.3 года назад
In this video, I am showing you quick&dirty how to plot a probability density function (PDF) into a histogram plot in GNU Octave. Remember: For random variables, we have the PDF and the CDF (cumulative distribution function). Here I will show you how to generate random numbers according to a normal distribution, calculate the histogram and plot the PDF of a normal distribution into the same plo...
#HowTo - Amplitude Modulation in GNU Octave
Просмотров 3,7 тыс.4 года назад
In this video, I am showing you how to perform amplitude modulation (AM) in GNU Octave. AM means that we are modulationg the amplitude of our carrier signal (e.g. the signal we need for transmission) using our message signal. The message signal is the information we want to transmit e.g. music, voice or whatever. AM is old but gold! Ok a lots of drawbacks but its a nice starting point to the fi...
#HowTo - Discrete Fourier Transform Matrix in GNU Octave
Просмотров 2,6 тыс.4 года назад
In this video, I am showing you how to perform the Discrete Fourier Transform (DFT) using matrix notation. Therefore we have to create a Discrete Fourier Transform matrix that we apply to our signal et voila we get the DFT of that signal. By the way, the Fast Fourier Transform (FFT) and DFT are doing the same from a math point of view but the FFT is a very efficient implementation of the DFT. W...
#HowTo - Polynomial Regression in GNU Octave
Просмотров 1,9 тыс.4 года назад
In this video, I am showing you 'quick&dirty' how to perform polynomial regression on a noisy signal. Using the polynomial regression we estimate the unknown coefficients. A non-linear relationship between x and y is given BUT the estimation problem itself remains linear! Because the regression function is linear in the unknown parameters (the unknown polynomial coefficients! the math in the li...
#HowTo - Matched Filtering in GNU Octave
Просмотров 3,9 тыс.4 года назад
In this video i am showing you 'quick&dirty' how to perform the matched filter operation on a noisy signal. The matched filter output e.g. then will tell us if our desired signal is inside the noisy signal. In general the matched filter is an optimal filter maximizing the signal-to-noise ratio on its output when additive stochastic noise is present. SOmetime there is maybe that much noise that ...
DSPcorner - HowTo Z-Score Data in GNU Octave
Просмотров 4324 года назад
In this video, I am going to show "quick&dirty" how to Z-score (or also known as standard score) data in GNU Octave. Data means that we have e.g. sample values from a certain normal distribution. A normal distribution is defined by two values: mean and standard deviation. If you want to compare sample values from different normal distributions, then the Z-score comes into the game. Z-score the ...
DSPcorner - HowTo Moving Average Filter in GNU Octave
Просмотров 3,2 тыс.4 года назад
In this video, I am going to show "quick&dirty" how to implement a moving average filter in GNU Octave. The moving average filter is a simple way to perform signal smoothing. Despite the "manual" implementation, you can also use the filter-command where you have to describe the filter according to the z-transform (not part of the video but for proper digital signal processing you have to unders...
DSPcorner - HowTo FFT-Based Interpolating a Signal in Time Domain in GNU Octave
Просмотров 5574 года назад
In this video, I am going to show "quick&dirty" how to interpolate a periodic signal in the time domain (in that case a sine-wave) using the FFT-command in GNU Octave. You all know the zero-padding, to increase the FFT resolution. Here we go the other way around, but we do not insert the zeros at the end of calculated FFT. Instead, we insert the zeros in the middle of the calculated FFT! This i...
HowTo Perform System Identification #1 in GNU Octave
Просмотров 2,1 тыс.4 года назад
In this video, I am going to show "quick&dirty" how to estimate the impulse response of an unknown system in GNU Octave using the least mean square algorithm (LMS). It is part one of a series of algorithms that I am going to present here :) The LMS algorithm is a basic adaptive algorithm that estimates the system coefficients in each iteration using the known input and output signals of the unk...
HowTo Resample A Signal #1 (Interpolation) in GNU Octave
Просмотров 4,4 тыс.4 года назад
In this video i am going to show "quick&dirty" how to resample a signal in GNU Octave. Resampling means that we change the sampling frequency. If we decrease the sampling frequency than we call it decimation or down-conversion. If we increase the sampling frequency we call it interpolation or up-conversion. In this video i will show you how to interpolate a signal. This involves two steps: inse...
HowTo Short Time Fourier Analysis (spectrogram) in GNU Octave
Просмотров 3,5 тыс.4 года назад
In this video i am going to show "quick&dirty" how to use the short-time Fourier analysis (STFT) to display the frequency domain vs. time domain. In contrast to the Fourier transform (fft in digital) where the result is the overall frequency spectrum, the STFT displays what is going on in the frequency domain at a certain time. So you are able to see a frequency change over time. Powerful tool!...
HowTo Perform Linear Regression in GNU Octave
Просмотров 10 тыс.4 года назад
In this video i am going to show "quick&dirty" (one take!) how to calculate the linear regression of a noisy dataset. So we try to fit a linear model into our noisy data! This means estimating the model parameters. # The code from the video is available here: textuploader.com/1qpft # Some background about the linear regression: en.wikipedia.org/wiki/Linear_regression # Have also a look at the l...
HowTo Cross-correlate in GNU Octave
Просмотров 3,3 тыс.4 года назад
In this video i am going to show "quick&dirty" (one take!) how to cross-correlate two signals in GNU Octave. So you could e.g. determine a shift between these two signals (time difference). # The code from the video is available here: textuploader.com/1lhe8 # Some background about the cross-correlation: en.wikipedia.org/wiki/Cross-correlation # Have also a look at the link below to see the diff...
HowTo Filter a Signal in GNU Octave
Просмотров 16 тыс.4 года назад
In this video i am going to show "quick&dirty" how to filter a signal using a low-pass FIR filter in GNU Octave. # The code from the video is available here: textuploader.com/1l5qr # Some background about the finite impulse response filter FIR (the other type are infinite impulse response filters IIR): en.wikipedia.org/wiki/Finite_impulse_response # Some background to the low-pass filter in gen...
HowTo Calculate the Fast Convolution in GNU Octave
Просмотров 1,9 тыс.4 года назад
HowTo Calculate the Fast Convolution in GNU Octave
HowTo Plot A Histogram in GNU Octave
Просмотров 4,6 тыс.4 года назад
HowTo Plot A Histogram in GNU Octave
HowTo Calculate The Hilbert Transform in GNU Octave
Просмотров 2,5 тыс.4 года назад
HowTo Calculate The Hilbert Transform in GNU Octave
HowTo Plot a Chirp in GNU Octave
Просмотров 8914 года назад
HowTo Plot a Chirp in GNU Octave
HowTo Plot a Sine Wave in GNU Octave
Просмотров 10 тыс.4 года назад
HowTo Plot a Sine Wave in GNU Octave
HowTo Plot a FFT in GNU Octave
Просмотров 15 тыс.4 года назад
HowTo Plot a FFT in GNU Octave

Комментарии

  • @jeongwookhur
    @jeongwookhur 4 месяца назад

    Now 2024, I see your video. Great!!!

  • @fredstratford5104
    @fredstratford5104 8 месяцев назад

    Next video, ditch the music or lower the volume, spoils good content

  • @CoriMusic97
    @CoriMusic97 11 месяцев назад

    Great video! So if i have a 1 .wav file sample for the entire keybed.....its better to have a higher sample rate for clean pitchshifting ?

  • @vilsondasrapadura7342
    @vilsondasrapadura7342 Год назад

    Can you repost your pdf? The link has expired and its such a good work! Thanks a lot for the kindness of making this video!

  • @decoderX2
    @decoderX2 Год назад

    can you please reupload the code?

  • @hasanborakaygn
    @hasanborakaygn Год назад

    Many thanks! You were extremely helpful.

  • @WadoNeil
    @WadoNeil Год назад

    Echt nuetzlich! Thanks a lot

  • @brandonlee9528
    @brandonlee9528 Год назад

    GERMANY?

  • @josueruiz8166
    @josueruiz8166 Год назад

    rifado padrino me ayudaste un montón.

  • @aidinha7095
    @aidinha7095 Год назад

    Thank you for the video. The address for the code from the video doesn't exist. Could you please share the code? Thank you.

  • @michaelsteadman7973
    @michaelsteadman7973 2 года назад

    Please turn off music in any new videos.... VERY distracting

  • @agahgarnadi3047
    @agahgarnadi3047 2 года назад

    Sadly, the link to the code is broken at the time of writing.

  • @slightlychaotic8988
    @slightlychaotic8988 2 года назад

    Thank you for another helpful, well-explained video.

  • @eie_for_you
    @eie_for_you 2 года назад

    Thanks!

  • @pewolo
    @pewolo 2 года назад

    There is actually a simple way of fitting line to data in Octave using polyfit(x,y,1) which will give you directly the coefficient and slope. The "1" indicate the degree of the polynomial function.

  • @slightlychaotic8988
    @slightlychaotic8988 2 года назад

    Thanks, I enjoyed all the aspects you explored in this one video - real/complex, decibel, and especially the fftshift, of which I was completely unaware.

  • @thepruh1151
    @thepruh1151 2 года назад

    Heyo, question. I want to make a histogram with ticks that are on the edges of the bins, rather than the centres, is there a way of doing that? Another question: how do I align the ticks with what's on the graph? Because when I plot my histograms, the ticks and grids are always unaligned? Thanks in advance and thank you for sharing the video.

  • @timzajic2999
    @timzajic2999 2 года назад

    Thank you for the video. I would like to say that I believe the normalized frequency requires dividing by (Fs/2) and not (Fs) as you do. With the example you are doing with the frequencies so far apart it is not noted that an error was made in using Fs rather than (Fs/2). If anyone does not agree, please indicate so.

    • @ggendron1
      @ggendron1 2 года назад

      I agree. It would be great to get a comment from the author to make sure we are right.

    • @slightlychaotic8988
      @slightlychaotic8988 2 года назад

      I agree, too

    • @shanmukeshr1696
      @shanmukeshr1696 Год назад

      Ig it has to do with nyquist rate as the signal is being sampled here

  • @nathanfonseca284
    @nathanfonseca284 2 года назад

    Por favor

  • @nathanfonseca284
    @nathanfonseca284 2 года назад

    Preciso fazer um trabalho para amanhã, poste o video sobre FIR AGORAAAAAAAAAA

    • @vilsondasrapadura7342
      @vilsondasrapadura7342 Год назад

      Se tu tá em um vídeo em inglês é porque tu entende inglês, mas ele não entende português qual a lógica disso?

    • @nathanfonseca284
      @nathanfonseca284 Год назад

      @@vilsondasrapadura7342 好吧,伙计,非常好的评论。

  • @barroilh
    @barroilh 2 года назад

    Very good but I cannot see what you are typing. Just too small. Instead of having a video insert of yourself it would be more useful to have a video insert with a zoomed take of what you are typing.

  • @ismetozturk947
    @ismetozturk947 2 года назад

    very good

  • @adityaprabhu6
    @adityaprabhu6 2 года назад

    Your screen is a little small, hard to read. Otherwise, helpful video.

  • @JJCotek
    @JJCotek 2 года назад

    code isn't aveilable, add it next time in discription or attach it in the comments

    • @JJCotek
      @JJCotek 2 года назад

      oh, and thx for the video

  • @michaelhill4111
    @michaelhill4111 2 года назад

    May you give an more complex description of what you are doing, I would like to be able to explain it myself?

  • @jameshopkins3541
    @jameshopkins3541 2 года назад

    I know continuos convolution and discrete convolution but no Fast Convolution

  • @Batmans_cousin
    @Batmans_cousin 2 года назад

    More videos!!!!

  • @radhey04ec
    @radhey04ec 2 года назад

    Nice tutorial, very helpful... Greetings from India 🇮🇳 Pls sir make video on How EDA explore Data Analysis for ML Dataset....

  • @radhey04ec
    @radhey04ec 2 года назад

    Really sir very useful....

  • @radhey04ec
    @radhey04ec 2 года назад

    Beautiful Tutorial👨‍🏫📓

  • @dareemmanuel6079
    @dareemmanuel6079 2 года назад

    how did you arrange the octave window into 3 compartments especially the right part please

  • @jaafarwamwere6282
    @jaafarwamwere6282 2 года назад

    Thanks

  • @electroniquegenerale
    @electroniquegenerale 2 года назад

    Very nice video, very usefull. Now i will check how to make FFT from Wav file, may be in the other video, let's see

  • @danhellgren5671
    @danhellgren5671 2 года назад

    Is it possible in Octave to read in external data from e.g USB and do the filtering continuously in real time?

  • @yashchincholikar2322
    @yashchincholikar2322 2 года назад

    Please can you upload the code again here

  • @xuhakan4089
    @xuhakan4089 2 года назад

    so, is this matched filter a corss-correlation operation? thanks.

  • @andrewwarr7683
    @andrewwarr7683 2 года назад

    Very neat and tidy, thanks. Your videos are so clear, more effective than the overly worked Matlab referenced stuff that relies on "fvtool" and other design tools that hide DSP in apps not available outside the commercial product. I would love it if could you do a video on sampling an audio file, something like a chord on a piano like a C7 that has 4 notes (C,E,G,A#), FFT and check the accuracy of recorded sound against known frequencies, doing some filtering to change it to a CMaj (C,E,G) then reconstruct the filtered frequencies "IFFT " into a .wav file and compare that to a recorded CMaj off the web or recorded yourself.

  • @jaydeepshah2026
    @jaydeepshah2026 2 года назад

    Very useful video sir....

  • @skyline2022
    @skyline2022 2 года назад

    Brilliant. Awesome presentation on FIR, octave scripting as well and emphasize on functions used in the script. Bitte !

  • @skyline2022
    @skyline2022 2 года назад

    Thanks for your time and efforts. Brilliant presentation. I am learning and enjoying a lot from your presentation. Fantastic learning medium for signal processing enthusiasts

  • @skyline2022
    @skyline2022 2 года назад

    Excellent delivery of matched filter concept. Thanks for your presentation.

  • @Exkobar13
    @Exkobar13 2 года назад

    Great video! So you know, the link for the code is down. Chears!

  • @navjotsingh2251
    @navjotsingh2251 3 года назад

    I liked this video, please do more cool statistical videos in octave! I do know some R, but would like to learn other tools like octave/MATLAB. This is an interesting video and a nice refresher on polynomial regression.

  • @hareshchandwani9864
    @hareshchandwani9864 3 года назад

    Beautifully explained..

  • @JesseMcCall
    @JesseMcCall 3 года назад

    Wonderful tutorial!

  • @donkelly36
    @donkelly36 3 года назад

    Nice video, thx. I also suggest removing background music, as it detracts from the voice audio.

  • @SpechtiGER
    @SpechtiGER 3 года назад

    Thanks a lot! Great Tutorial

  • @jakelozanta11
    @jakelozanta11 3 года назад

    The link in code cannot be seen..please share it again..thank you.

  • @rolanportento9704
    @rolanportento9704 3 года назад

    thank you sir, you were a huge help, greetings from the Philippines <3

  • @franciscomiranda3239
    @franciscomiranda3239 3 года назад

    No file for downloading??? Where is that??