 Hello, I'm Didier Stavens. I'm going to show you some more things you can do with the GNU radio companion and without any dedicated hardware like an SDR. Here we are going to look at some simple filters and we are going to do that with audio frequencies so that we can also listen in. So a new schema. I'm going to change this to the WX GUI widgets and here the sample rate I'm going to put this at 48 kHz because we are going to connect this to the speaker of my computer audio sync. So as a source here I'm going to generate some noise with a noise source. Double click here. You can see it's blue so the output is complex. I'm going to change that to floating point numbers like this. Then I'm taking an audio sync and we connect the two save filters and now I can run this and now when I run this you will hear the noise. Here is the noise. Let me stop this. So this is random noise that is generated here and it sounds like static on a radio that is not tuned to any radio station. We can also visualize this with instrumentation widgets here WX and here for example I can use the fast Fourier transformations sync so that you can see the spectrum of frequencies. This here. It's blue so complex. I want it to float. I can connect this and then I can run this and now here you can see the frequencies and the power for the different frequencies going from 0 Hz all the way up to 24000 Hz or 24 kHz. Let me stop this and I'm going to add another display. A waterfall sync. It's similar to this output but here you also get a history. Let me change this to float like this. So what you see here again from 0 to 24 and the color gives the intensity. So here we are between minus 30 and minus 40 dB and you can see here that between 30 and 40 we are in the green and that's what you see here. And this is the time parameter, the scale. So here we are already running this for 14 seconds and you can see this moving on like this. I can stop this. Okay so now I'm going to introduce a filter and the first thing that I'm going to do is introduce a low pass filter like this. So this is blue. I'm going to change this to floating point. And then what the filter wants is a cutoff frequency and a transition width. It's also a frequency. So these are in red. These are required. So I'm going to put this let's say at 10000 Hz and the transition width I'm going to put this at 100 like this. And then let me remove all these connectors and then reroute this. So from the noise generator to the filter and then from the filter to the sync and the FFT and the waterfall like this. And when we run this you see now that here at 10 kHz we no longer have a signal and the signal drops down and we have this. So that's what a low pass filter does. It lets all the frequencies until 10 kHz pass and the rest it blocks. Now filters like this are actually implementations of analog filters and the cutoff frequency is not precise at 10 kHz itself. It's also the transition width that makes how steep this cutoff is. So let me stop this. So let me change this to a much wider transition width. So instead of having 100 Hz let's put in 5 kHz or 5000 like this. And if I run this you see here that the cutoff here is not so steep. That it slowly goes down until here 14. So let's stop this. So that was a low pass filter. Now let's take a high pass filter here. So float, float, 10100 and as you can guess now it's the higher frequencies that will pass and the lower frequencies that will be blocked like this. And here you see the opposite. And that's also what you see here in the waterfall plot. Here we have some noise while here in the dark blue this is filtered out. You can also combine a high pass filter and a low pass filter to make a band pass filter. And this is actually also a component that exists here, a band pass filter. So float, float. And now you have to give in two frequencies. And what I'm going to do is from 300 Hz to 3000 Hz and let's say transition width of 10. And I'm choosing these numbers because from 300 Hz to 3000 Hz that's about the frequency range that you would hear on the telephone landline. A plain old telephone system like this. And here you can see the pass here is a bit of blue. And it also sounds different. It sounds like this is static coming out of a handset. And then the last one that I want to show is a notch filter. So the opposite of a band pass filter here, a band reject filter. So float, float. I'm going to say 3000 this time to 6000 and a transition width of 100. And let's connect all these things. And we can run this. And here you can see now that frequencies up till 3000 Hz stay pass. Then the rest is blocked until 6000 Hz here. And then we resume. And that's what you can also see here in the waterfall plot. That was an illustration of the different filters that you can use in GNU radio companion. And as you can see of course there are much more filters than the one that I showed you. So you have some more things to explore.