[USRP-users] check send successfully

w xd wxd920502 at gmail.com
Mon Dec 1 20:48:58 EST 2014


Hi Marcus and Michael:
                                  Thank you for your kindly reply.I just
want to realize the function Michael said.I will try to test this
method.Thank you.
Best regards,
xdw

2014-12-02 3:05 GMT+08:00 Michael West <michael.west at ettus.com>:

> Hi xdw,
>
> Acknowledgement of the burst being sent comes back from the USRP to the
> host as an asynchronous message.  The tx_bursts example sends a burst and
> then waits for that acknowledgement by using the
> tx_stream->recv_async_msg() call and checking the metadata.  You can do the
> same in your application.  If you want to send several bursts before
> checking the asynchronous messages or have a separate thread handling the
> acknowledgements, you can correspond the message to the burst by examining
> the event code and timespec in the metadata.
>
> Regards,
> Michael E. West
>
> On Mon, Dec 1, 2014 at 8:23 AM, Marcus Müller <usrp-users at lists.ettus.com>
> wrote:
>
>> Hi xdw,
>>
>> Stopping the program will deconstruct the C++ USRP object, which might
>> in turn shut down parts of the USRP. You should not do that.
>>
>> Best regards,
>> Marcus Müller
>>
>> On 12/01/2014 04:37 PM, w xd wrote:
>> > Hi Marcus:
>> >                "You shouldn't have to care about whether the USRP sends
>> the
>> > signal" Do you means that although the program already stop,but the usrp
>> > can sends the signal in the right time?If this,I know.
>> > Best regards,
>> > xdw
>> >
>> > 2014-12-01 23:34 GMT+08:00 w xd <wxd920502 at gmail.com>:
>> >
>> >> Hi Marcus:
>> >>                 Thank you .
>> >>                 I know this,and i already said this before.Because  the
>> >> software isn't synchronized with the usrp.When the variable has been
>> become
>> >> 0,but the true signal haven't send.The program have already stop.I have
>> >> already tested it.
>> >> Best regards,
>> >> xdw
>> >>
>> >> 2014-12-01 23:27 GMT+08:00 Marcus Müller <marcus.mueller at ettus.com>:
>> >>
>> >>> Hello xdw,
>> >>>
>> >>>> I just want to send 10 bursts,then stop the program,for example.
>> >>> That is definitely a question of writing basic C++ code, which I
>> cannot
>> >>> really explain to you. There are plenty nice C++ guides out there!
>> >>> You could, for example, have a counter variable, that you reduce by 1
>> >>> every iteration (ie. every burst), and check for that variable being
>> > 0
>> >>> in the while() condition.
>> >>>
>> >>> Best regards,
>> >>> Marcus Müller
>> >>>
>> >>>
>> >>> On 12/01/2014 04:22 PM, w xd wrote:
>> >>>> Hi Marcus:
>> >>>>                 Thank you a lot.I just means that.
>> >>>>                 But in this case("they will be sent when as soon as
>> the
>> >>>> time comes),how to stop it in our design program?
>> >>>> For example,the tx_burst.cpp.I just want to send 10 bursts,then stop
>> the
>> >>>> program,for example.
>> >>>> Best regards,
>> >>>> xdw
>> >>>>
>> >>>> 2014-12-01 22:36 GMT+08:00 Marcus Müller <marcus.mueller at ettus.com>:
>> >>>>
>> >>>>> Hi xdw,
>> >>>>>
>> >>>>> I think I understood you correctly. If you successfully transmitted
>> the
>> >>>>> time-stamped samples to the USRP, they will be sent when as soon as
>> the
>> >>>>> time comes.
>> >>>>> You shouldn't have to care about whether the USRP sends the signal;
>> >>> it's
>> >>>>> designed to do that, and unless you do something drastic, that
>> *will*
>> >>>>> happen.
>> >>>>>
>> >>>>> Best regards,
>> >>>>> Marcus
>> >>>>>
>> >>>>> On 12/01/2014 03:34 PM, w xd wrote:
>> >>>>>> Hi Müller:
>> >>>>>>              Thank you.Maybe you haven't catch my problems.Because
>> the
>> >>>>>> signal i want to send will be send in the future.But the software
>> >>> will be
>> >>>>>> run when i run it.When the software has been run,but the signal
>> maybe
>> >>>>>> haven't send.It will be send in the future.So the software isn't
>> >>>>>> synchronized with the usrp.I can't know how to judge when the
>> signal
>> >>>>> send.
>> >>>>>>              Maybe my describe have a little confusing.Sorry for
>> that.
>> >>>>>> Best regards,
>> >>>>>> xdw
>> >>>>>>
>> >>>>>> 2014-12-01 22:26 GMT+08:00 Marcus Müller <marcus.mueller at ettus.com
>> >:
>> >>>>>>
>> >>>>>>> Hi w xd/xdw/xd/Xianda,
>> >>>>>>>
>> >>>>>>> If you were able to transmit the samples to your USRP, then they
>> >>> will be
>> >>>>>>> sent.
>> >>>>>>>
>> >>>>>>> I don't understand your second question; are you asking us how to
>> >>>>>>> terminate a C++ while loop?
>> >>>>>>>
>> >>>>>>> Best regards,
>> >>>>>>> Marcus Müller
>> >>>>>>>
>> >>>>>>>
>> >>>>>>> On 12/01/2014 03:17 PM, w xd wrote:
>> >>>>>>>> Hi Müller:
>> >>>>>>>>               Thank you.
>> >>>>>>>>               Because the signal  is send in the future.It has
>> been
>> >>>>>>>> described by metadata.It haven't send now.For example,I set a
>> >>> counter
>> >>>>> in
>> >>>>>>>> the loop.The counter have already reach some number.But the
>> signal
>> >>> now
>> >>>>>>>> haven't send.
>> >>>>>>>>              counter=1
>> >>>>>>>>              Do:
>> >>>>>>>>                      1.send a frame
>> >>>>>>>>                      2.counter++;
>> >>>>>>>>              while(...);
>> >>>>>>>> For example,when counter=15.But it haven't send 15 times.Because
>> it
>> >>>>> send
>> >>>>>>> in
>> >>>>>>>> the future.
>> >>>>>>>>             Another question:how to stop the loop?When i set
>> >>>>>>>> while(counter<=15),it stop fastly.And it haven't send.I have read
>> >>> the
>> >>>>>>>> example:tx_burst.cpp.And it set: while (not stop_signal_called
>> and
>> >>>>>>>> repeat);Now i just want the program send 15 times.
>> >>>>>>>> Best regards,
>> >>>>>>>> xdw
>> >>>>>>>>
>> >>>>>>>> 2014-12-01 22:04 GMT+08:00 Marcus Müller <
>> >>> usrp-users at lists.ettus.com>:
>> >>>>>>>>>  Hi xdw,
>> >>>>>>>>>
>> >>>>>>>>> what do you mean with "check send successfully"?
>> >>>>>>>>> usually, unless your send() call says something went wrong (ie.
>> the
>> >>>>>>> number
>> >>>>>>>>> it returns isn't the number of samples you wanted to send), it
>> >>> should
>> >>>>>>> have
>> >>>>>>>>> been successfull.
>> >>>>>>>>>
>> >>>>>>>>> Could you explain in a little more depth what you are trying to
>> do
>> >>> and
>> >>>>>>>>> which problems you encounter?
>> >>>>>>>>>
>> >>>>>>>>> Best regards,
>> >>>>>>>>> Marcus Müller
>> >>>>>>>>>
>> >>>>>>>>>
>> >>>>>>>>> On 12/01/2014 04:01 AM, w xd via USRP-users wrote:
>> >>>>>>>>>
>> >>>>>>>>> Hi all:
>> >>>>>>>>>          I just want to realize the following function,but i
>> don't
>> >>>>> know
>> >>>>>>> how
>> >>>>>>>>> to check the usrp send successfully.
>> >>>>>>>>>         Loop:
>> >>>>>>>>>                 1.send a frame--->sending a timed burst
>> >>>>>>>>>                 2.after send successfully--->tcpip accept and
>> >>> receive
>> >>>>>>>>>         The above repeated.Can someone have some ideas?
>> >>>>>>>>> Thank you.
>> >>>>>>>>> Best regards,
>> >>>>>>>>> xdw
>> >>>>>>>>>
>> >>>>>>>>>
>> >>>>>>>>>
>> >>>>>>>>>
>> >>>>>>>>> _______________________________________________
>> >>>>>>>>> USRP-users mailing listUSRP-users at lists.ettus.comhttp://
>> >>>>>>> lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>> >>>>>>>>> _______________________________________________
>> >>>>>>>>> USRP-users mailing list
>> >>>>>>>>> USRP-users at lists.ettus.com
>> >>>>>>>>>
>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>> >>>>>>>>>
>> >>>>>>>>>
>> >>>
>>
>>
>> _______________________________________________
>> USRP-users mailing list
>> USRP-users at lists.ettus.com
>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20141202/0c35ac94/attachment-0002.html>


More information about the USRP-users mailing list