Page 1 of 1

2.4GHz TX Range/Power

Posted: Mon Dec 25, 2017 3:16 am
by ThisChessPlayer
I do not come from a RF background, so please bear with me. Has anyone been able to transmit and receive effectively between two BladeRFs at a distance of say 30-50 meters on 2.4GHz bands? I am beginning to wonder whether there is something seriously wrong about my ideas/setup considering the range I am getting is very limited (often under 5 meters). Does it sound possible to do this with a BladeRF connected to say a Raspberry Pi 3 or cheap single-board computer?

The setup uses 1.5MHz bandwidth, 1.5MHz sampling rate, and OFDM transmission scheme with packet validation using GNURadio as a method of transferring packets of information between two BladeRFs. There does not appear to be a lot of noise in the testing environment, although upon inspection of the signal through a spectrum analyzer, it appears that the produced signal is very weak and disappears at about 5 meters. Have tried using an antenna that should work with 2.4GHz, but that did not work out. This setup also currently works on the lower ISM frequencies (433MHz, 915MHz), just not 2.4GHz.

Can anyone point me in the right direction? I would greatly appreciate it.


Re: 2.4GHz TX Range/Power

Posted: Mon Dec 25, 2017 3:51 am
by RadioGuy

I'm a new to radio as well so this is just a few thoughts.

I have some software I made that does digital transmission at various frequencies as well. Like you, things would would great at lower frequencies like 915MHz, but not so well at 2.4GHz. I found that running a calibration from the command line(a.k.a bladeRF-cli -e calibrate lms) at every power up seems to help lots. Just running a calibrate was enough to make things work decent at 2.4GHz.

As for the transmission power of these things, it's very low and I think that's somewhat intentional. Since it's meant for experimentation and it's very easy to transmit outside the ISM range, you don't want to get in trouble with the law etc.. Nuand does sell a amplifier board that can boost the transmission power up to 1 watt, but it appears to be out of stock until January. You can also use an external amplifier with the bladeRF as well.

I did do a range test with the family radio demo that runs in the lower frequency range to a walkie-talkie and I was only getting a range of a street or two over in a suburban area. Like I said, I think this is intentional.

Finally, I don't think the RPI 3 is going to be powerful enough for your purposes. I have a similar setup with 1.5MHz bandwidth and 1.5MHz sampling rate and the software can very easily peg out a 1.8GHz quad proc Intel. You may want to run top on the RPI 3 to see what kind of CPU utilization you are getting. I suspect you may be surprised that may be part of the data loss issues.

Re: 2.4GHz TX Range/Power

Posted: Mon Dec 25, 2017 5:13 am
by ThisChessPlayer
Thank you for the very prompt reply and suggestions!

I will definitely give the calibration a shot and edit with results.

As for the software side of things, I believe I might be able to get things running reasonably on a pi. I understand that SDR is very intensive, but it is looking good if the issue is a simple calibration since I did try OFDM with a BladeRF + pi setup a while back and it communicated at least one way short-range with BladeRF + laptop.

Also, am currently amplifying receiver, so that is giving a bit more range.

I would like to know what some of you more experienced BladeRF and radio guys think about this if possible, since you guys definitely know a lot of things we newbies don't :)

Re: 2.4GHz TX Range/Power

Posted: Thu Jan 11, 2018 4:58 pm
by ThisChessPlayer

Looks like one of my boards is having trouble calibrating on multiple frequencies. From reading the page below on calibrating the BladeRF, it appears that the error I have is very high. Whether I am performing the calibration on 433MHz, 915MHz, or 2.45GHz, the error is always above 10, usually more in the area of 30-70. There is some 2.45GHz interference, but should not be as much on other frequencies in the area I am in. Additionally, my second board does not have this issue, reporting errors much closer to 0. ... Correction

Screenshot of my calibration procedure and results:

So, based on the above, is there anything I can try in addition to the manual corrections?