วีดีโอ

The easiest way to do it is to capture the I/Q stream just use a file sink. The I/Q file is just a binary floating point data with the real value followed by the imaginary binary value. Read that and convert it into any file type that you please.

The colons are missing because the TH-cam editor was doing strange things with them - It may be fixed now - I will check to see if I can put the colons back in.

I just took a look at the GNSS Firehose RF - it sounds pretty impressive. I played with the GPS a little with the limeSDR mini finding a good antenna was the main problem. The SDRs (RSP1, RSP2, RSPDuo, LimeMini 2.0, ANTSDR e200, Cariboulite 6G, HackRF One, RTL_SDR, bladeRF 2.0 micro XA5, RFSPACE NetSDR+, AirSpy HF+) that I have are mostly the same in sensitively. Although, a few standout in various frequency ranges as both good and bad.

It can be change to use just about any SDR that transmits - What SDR do you have ?

​@@daleranta1012 A4 version

The main takeaway from this video is that the 240 cables are as good as the 400 cable (in this frequency range) - so do not waste the extra money on the 400 cable. The longer the cable the more the loss.

As For antennas in the 460 MHZ range , a ground plane works very well and it is about 11 inches high with 6 inch arms. A 4 element beam is is 15 inches long and 13 inches wide and gives about 10 times the power of the ground plane, but it is directional. The slim jim is supposed to be slightly better than the ground plane - I tired building one, but I could never get it to work.

Most cable providers now encrypted their signals - in the old days you could plug the SDR into the TV-cable and decode the channels directly. The cable boxes are supposed to have a USB out on them - I think that that is still a legal requirement, but try a get one from a cable company.

It uses 3 programs gnuradio, iqSDR, and ffplay. You can sometimes get gnuradio to install and work on windows in the last 5 tries I got it going once. I could make iqSDR work on windows. I spent some time looking for a ffmpeg and ffplay for windows, but I unexpectedly did not turn anything up. I expected a simple .exe install, but I could not (in a few hours) find one ?

Thanks. If you look part way down the page at github.com/righthalfplane/SdrGlut You will find a list of many projects that I found interesting.

Every big city still has some broadcast TV

​@@daleranta1012Some? With multiplexing, there are more channels than ever.

Lets see now, I use several versions of MacOS, several version of Window, several version of Linux, and several versions of the Raspberry pi and they all act a little different. Remembering, who is and which is which it impossible for me. I send forty years writing programs LANL. I worked on the Crestone Project which ran one of the largest super computer calculations ever done and It used the I/O routines that I had written.

Yes, you may be right. I do notice less funny stuff - if I click on the title bar.

I think Matlab may read I/Q files directly - the I/Q file is just the raw floating point data ( real1,imag1, real2, imag2, ...). Each complex number is just the real part followed by the imag part - usually 8 bytes.

🤦‍♀

oo-boon-too

You wrong. I'm sure the right pronunciation is Ubuntu as in Ubuntu not in Ubuntu

I must have been confused ive been calling it ubuntu as in ubuntu and not ubuntu all these years 😂

Had me crying laughing

jübunta

Thanks !

ew buon twu

I was just guessing from the spelling - I have no idea what is correct.

not really, it's more like uu-buun-too (/ʊˈbʊntuː/)

​@@daleranta1012what a wild guess to make lmao

I posted a photo at - groups.google.com/g/sdrglut-users/c/ky3enNvZdZc

No, additional FPGA functions were used. If it can handle a 8 MSPS bandwidth in real time, that is all it needs - I suspect aBladeRF 2.0 xA4 can do that easily. If there is a problem, you can clear many of them up by adjusting the rx-gain. I will be putting out a new video soon - I found a better way to do things.

Not an issue. So nice of you to ping. It was a false alarm as I had installed antsdr_uhd not overwriting original UHD driver and that has some issue. It appears. Now I removed original UHD driver and installed antsdr_uhd in default /usr/local locations. And now it is getting detected. Thanks a lot once again.

You are missing the xquartz routines. You can get the one that you need from - www.xquartz.org/releases/

If you cannot run it, usually you will see something highlighted in red - that is where the problem. If it starts, but errors out then you should get a message telling what the problem is.

@@daleranta1012 thanks but as I said I am very new in GNURadio and I cant really understand the errors well. Maybe I cant run the flowgraph because I dont have any Hardware like SDR right ? You use Osmocom sink / source and I think it refers to use a hardware. Or can I run the flowgraph without having any hardware ?

@@alirezamohammadi4536 The example as given works using the BladeRF SDR and with a little work you can use practically any SDR. You do need a SDR transmitter and receiver for the example to work.

@@daleranta1012 I dont have any regarding hardware. Can I change the Flowgraph so it works without any Hardware? How is it possible or wich components shall be replaced to what ? Thank you again in advanced

I think I have found how to resolve this by replacing the osmocom sink source by a chanel model right ? but to wich components shall I link the in-output of the both chanel model ?

You make the bandwidth 8 MHZ centered at the centered of the frequency range of the channel. I put a Channel Frequency listing at - github.com/righthalfplane/rfspace in the png files TV-1.png and TV-2.png

The WiFi networks use the OFDM to transfer data, but that is only part of what's going on. BPSK and QPSK are two of the methods used by OFDM to move the data. BPSK can handle higher levels of noise than QPSK before errors start creeping in.

I used one of those cheep 300 to 75 ohm adapters - It seemed to work Ok.

At that time, I was using 32 bit Bullseye. Now, I have switched over to 64 bit Bookworm.

@@daleranta1012 Thanks for your response. Also what version of Python were you using..?

@@bemshimapeter78 I do not remember - it worked with the default version (2.7 ?, 3.9 ?). Since then, I have had lots of problems with python versions. Versions just before 3.10 were fairly compatible. Versions 3.11+ make a break and do not support some important things used by gnuradio and other programs. So, if you are getting python build errors back off to at least 3.10.

@@daleranta1012 Thanks for your response. I have been having the same issue. it's more like the ARM build isn't available for versions above 3.7 I guess I'll just work with what is available. Thanks for your time once again. 🤝

reply if you solve this, same problem here

What do you want to use the timestamps for ? A few years ago, I was looking at the timestamps provided by my SDRs and the BladeRF was the only one that actually had a timestamp that was more for less correct.

I'm tring to calculate the distance between two USRP B210.I need to add theTX and RX timestamps.@@daleranta1012

@@daleranta1012 I'm trying to use the timestamps to calculate the distance between the two USRP B210.I need the TX and RX timestamp to calculate the time delay.

@@user-et9uu4kz7t That was what I thought. If you have the "GPS Locked TCXO Reference", that gives a foot accuracy. With just the B210, you only have 1000 foot accuracy. In the Soapy call to "readStream", timeNs is the timestamp in nanoseconds.

I ran the simulation with one suggested set of the upper and lower hats and the results were poorer. So, I did not try adding them. The simulation may be wrong - you really never know for sure until you try it. I do not like the shield - you have to rotate it then - the hats may have been for a shield version.

Thanks for the information Dale. @@daleranta1012

The hats are for VHF-hi band.

​@@davidanderson8643 Yes, I know and the simulation says that the performance is poorer on the VHF bands with the hats. I got a VNA a few weeks ago - the measured performance is pretty good from 198 MHZ to 900 MHZ. Now, may be I will try the hats and see what they really do.

The first thing to check is that the amplitude (A*sin(w*t)) of the input signal to the transmitter is less than 1.0 - I usually force it to be 0.95. Then, start with the transmit gain low and turn it up until you start to see receiver distortion - that is how you get the SNR down.

Yes, That it true and it used to test reception under different conditions. Some of the receivers can pull the signal out of the noise better than others. Some times you can clearly hear a signal that shows on neither the spectrum or waterfall displays - that is really spooky.

Yes, you are correct - I corrected it in the comments

Yes - that could be part of the difference - although, the difference at lower frequencies is not as great ?

Not having VOLK just slows,things down a little bit and is not really needed - I see a warning not a an error - Where is your error ? If you find a "volk_profile" around just run it - github.com/gnuradio/volk shows how to install it on some systems. I have never installed SoapyVOLKConverters.

That is a vague statement - I do not know what you mean. I try to give the people all the information that they need and I include it with the video - that seems the best practice to me.

My hand held has no problem crossing the 26 miles from Cuyamaca peak to my house - it is perfect line of sight. Going between a hand held and a base station more than doubles the range.

The main point of the video is - at 18 feet the barJan looses 2/3 of your power - so for anything that is connect to a distant antenna you had better be using some thing else. The barJan is lighter but less flexible than the Times.

@@daleranta1012 - Thank you, makes sense. At the 3-30 MHz range (opposed to 460 MHz) this loss diminishes greatly, so I was interested in the physical properties for my 3-30 MHz use.

@@johnlagreca6288 At 30 MHZ it does not matter much - I tried the 400 for a 30 foot run and the loss in the one extra connector required canelled the gain - so I return the 400 and used a smaller size.

Are You Having Some Problem ? Normally, you would need to switch the devices and antennas to match your setup. It would be best to start with the frequency in the example - some of these methods have problems at the higher frequencies. I used a rtl-stick to receive - I would have had to switch to my HackRF One to try it at 5Ghz and that would have required fiddling to get the frequencies correct. What was your setup ?

Yes, it is noisy here most of the day, but sometimes the noise drops to a low level

Using the same trick that was used in this video - Sending Two Or More Data Streams With gnuradio-companion th-cam.com/video/g0wN2VcWnA4/w-d-xo.html to receive all of the wifi channels at once - you could see all of the active channels and it should not be too difficult to figure out which channels that your desktop is using.

Here is an example using gnu radio-companion Receiving Two Or More Data Streams With gnuradio-companion. th-cam.com/video/7G7D5hoC7dI/w-d-xo.html

My HackRF One is 20 ppm the BladeRF xA5 is 0.5 ppm - so you likely want a HackRF One with a TCXO

@@daleranta1012the person u replied to deleted their message, but i believe ur trying to say ppb not ppm. if so, i have a hackrf, a bladerf xa4 and a gspdo. would i be better buying another bladerf for full dupled, or a hackrf? ive experienced weird issues with bladerf trying to run full duplex, so was looking to buy another sdr and phase lock them with the gapdo

Installing on a Mac M1 is easy once you have HomeBrew installed - /opt/homebrew/bin/brew install git cmake libsndfile itpp portaudio /opt/homebrew/bin/brew install --HEAD mbelib git clone git@github.com:LouisErigHerve/dsd.git cd dsd mkdir build cd build /opt/homebrew/bin/cmake ../ make -j 8 sudo make install I usually field questions on my google group - groups.google.com/g/sdrglut-users

Yes, that is possible, but the slower notes would change so seldom it is not clear how much that it would add.

@@daleranta1012 I like to use USER font for 98 character to put Pixel Art inside.