USRP Software Radio

cancel
Showing results for 
Search instead for 
Did you mean: 

Alamouti BER trigger

Hi all,

 

I have the Alamouti BER model working consistently, with no errors and nice Combined RX constellation with even spacing, but I simply cannot get the BER to trigger correctly. The trigger light is never green and on the rare chance is does, its always 0.50 or so.

0 Kudos
Message 1 of 9
(5,676 Views)

CNIDude,

 

You may need to tweek some settings in order to get the triggering to work correctly.  Take a look at the comments on the community example you are using:

 

https://decibel.ni.com/content/docs/DOC-21751

 

It looks like some of the users there have had some triggering issues as well.  Some of their suggestions may help.

 

What models of USRP are you using?

0 Kudos
Message 2 of 9
(5,660 Views)

Hi Smoores,

 

Thanks for your reply!

 Unfortunately, all the "trigger" related discussion on the board you posted was related to the "Trim Waveform" block and its tendency to overly descimate the incoming packets at the front of the receiver. Thats caused by the algorithm incorrectly calculating the size of the packet header. I solved that issue a while ago.

What I am seeing is the packet will flow completely through the receiver and I am getting good output data on the back end of my receiver. My problem is that the BER comparitor block is not properly comparing the receiver system output data to the PN generated data. The BER block has a green LED that lights up whenever it senses a packet that comes in. This indicates that the BER "triggers" the BER comparitor which compares the RX output to the original PN generated signal. Mine worked for a while, the first few times I ran the model but the calculated BER was around 0.50 to 0.48 (which totally sucks). I was able to tweak the system in order to observe a consistent BER of 0.09 to 0.05 (which is much better, but not anywhere close to the theoretical values given a completely locked down matched filter system with forced PPS and Clock timing set and a distance of only one to two meters between antenna. I saved the program that night and the next day when I came back in the lab, I tried to use the saved system that I was previously working on 12 hours before and it completely stopped working. Now, I have the system working again, with clearly defined constellation groups at the receiver and no errors, but the system will not calculate a BER. Do you have any ideas, or do you know anyone who may be able to help me at this level?

 

Thanks again,

 

-J 

0 Kudos
Message 3 of 9
(5,639 Views)

J,

 

Thank you for the detailed response.  How are you generating your 10MHz clock and PPS?  Are you using a GPS device with length matched cables for the reference clock and PPS signal?  Please provide me with the make and model of any extra hardware you are using.  What are the models of your USRPs?

0 Kudos
Message 4 of 9
(5,622 Views)

Thanks again for your reply, SMoores,

 

I am using the following USRP's:

(1) USRP2 with r4/SBX

(1) USRP2 with r4/WBX, WBX+Simple GDB/WBX

(2) USRP N210 with Rev4

 

None of these units have a gps capability.

 

For timing, I have an Agilent 33250A for the 1Hz PPS, w/ 3.3-5 Vpp. This box is synched with an Agilent N5182A for the 10 MHz clock at -106.5 dBm. I tried to use the 33250A for both signals, but the 10MHz output waveform was putting out a bad looking waveform so I selected a more stable unit with a better square wave output. If I'm running a 2x2, what voltage output can I place on the Clock and PPS? 

 

Please let me know if you see any issues.

 

Thanks again!

 

-J

0 Kudos
Message 5 of 9
(5,601 Views)

I'm not sure what the expected voltage levels on the REF IN and PPS IN lines are for the Ettis USRPs.  I do know that for the NI branded USRPs the REF IN expects to see a peak-peak voltage between 0.632 Vp-p and 3.56 Vp-p and PPS IN accepts both 3.3V TTL signals and 5V TTL signals.

 

Do you know what daughter boards you have in your Ettis USRPs?

 

These  are the ones that have been tested with the NI driver:

 

  • WBX
  • SBX
  • CBX
  • XCVR2450
  • LFTX
  • LFRX
  • BasicTX
  • BasicRX
0 Kudos
Message 6 of 9
(5,587 Views)

Dear sir, 

 

I just readed your post, it seems that you are a expert on BER. I am new with USRP on BER, may I ask a quick favor?  Could you check my post and give me some help?  Thank you so much in advance.  I am just doing the basic psk + ber, that's it, no coding at all.

https://forums.ni.com/t5/USRP-Software-Radio/Basic-Bit-error-rate-calculation-Please-help-Thank-you-...

 

0 Kudos
Message 7 of 9
(4,101 Views)

anyone share with me a VI with Implementing 2x1 MISO system with LabVIEW and NI USRP 2921 using Alamouti code?? I need problem with the estimation channel in the receptor, i used the alamouti code with 2 antennas trnasmitter and one antenna receiver

0 Kudos
Message 8 of 9
(3,420 Views)

Hi all,

Does anybody know why this problem occurs or how to solve this issue? I have a proper synchronization and everything in the transmission diagram works, except for BER calculation.

 

Thanks!

0 Kudos
Message 9 of 9
(2,544 Views)