[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Burst Tagger block - What I'm I doing wrong?

From: Tom Rondeau
Subject: Re: [Discuss-gnuradio] Burst Tagger block - What I'm I doing wrong?
Date: Wed, 29 Apr 2015 10:35:52 -0400

On Thu, Apr 16, 2015 at 12:05 PM, Nicolas Cuervo Benavides <address@hidden> wrote:
Hi all!

Im trying to implement a system that should change it's normal functioning when another system with higher priority comes into play. For that, we have thought of two ways: changing the central frequency of our system of interest and/or stopping the sending process automaticaly.

Im in charge of the latter, and for that I've found that maybe the "burst tagger" block could be of use. As far as I have understood, that block adds SOB and EOB tags to the element that I'm about to send, and those are then understood in the UHD as metadata. Afterwards (and may be here is where I get a little bit confused or have the misconception of its functionality), 'if the UHD receives an EOB tag, it turns of the transmitter until a new SOB comes'.  (I would also like to know if the part in between the single quotation marks is also right. I've checked the code for UHD sink and I still dont have this part 100% clear).

Now, comes my main question and the reason why I think Im getting the concept wrong: I apply a signal (sinus) at the first port of the burst tagger and on the second I apply what would be my detection of the other system. In order to see if the tags come or not, I connect at the output of the block a "Tag debug", where I do can see that it is a 'False' tag when the trigger is <=0 and 'true' when it is >0. However, I still see the signal being sent through the USRP.

I dont know if the sinus signal is not suitable for this test experiment, or if actually the burst tagger does not do what I understood. If someone has a clearifying opinion or remark I would really appreciate it. I've checked some previous questions here in feeder and the example explanation was implemented as I did. Any comment will be highy welcomed.

Thank you in advance.



I think the issue is related to the fact that the gr-uhd and that burst_tagger weren't designed together. In fact, the burst tagger block came before the UHD blocks handled tags and there wasn't any coordination.

First, place a tag_debug block in the flowgraph after the burst_tagger. That will make sure that the burst_tagger is actually properly tagging bursts.

But the main problem is that the burst_tagger produces tags with the key "burst" and the value true or false. You'll need to edit this block to change it to using "tx_sob" for the start of burst and "tx_eob" for the end. That should get you closer to what you want.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]