Re: [USRP-users] use a usrp x310 as MIMO transmitter, daughterboard synchronization?
Hi Marcus, I have one more question, SBX and UBX support phase alignment (phase synchronization) on certain usrps. Does phase alignment means that after using timed commands, the phase offset between two front-ends will be always zero (or very close to zero) or just be a constant which doesn't vary when we re-tune. I observed that even if we don't use the timed commands, the phase offset will be fixed after tuning but the offset value is different for different runs, so I feel confused about the term 'phase alignment'. Thanks a lot! Best Yang On Wed, Jan 9, 2019 at 9:19 PM Marcus D. Leech wrote: > On 01/09/2019 08:54 PM, Yang Liu wrote: > > Hi Marcus, > > Thanks! > > I feel a little confused about the term 'frequency synchronization'. if > the two LOs on the two daughterboads are using the same frequency > reference, does it mean that after tuning, they will produce identical > CFOs to the same receiver (one receiving channel) or we can say that the > difference in CFOs are almost negligible? (lets for now ignore other > factors that can contribute to CFO) > I read through 'align LOs in the front-end' in the link > https://files.ettus.com/manual/page_sync.html. timed command makes sure > that different daughterboads can be tuned at the same time, so that it can > only help to mitigate the phase offset, but it doesn't say anything about > the potential frequency offset between the two daughterboards. > > Best, > Yang > > There will be no frequency offset between the two cards--they're > programmed identically, and use the same reference clock. So other than > phase-noise and phase-offset, they will be, for all intents, "identical". > > > > > On Wed, Jan 9, 2019 at 3:17 PM Marcus D Leech > wrote: > >> Yes they’ll use the same reference clock provided by the motherboard. >> >> There’ll be a residual phase offset which can be mitigated using timed >> tuning. >> >> Sent from my iPhone >> >> > On Jan 9, 2019, at 3:01 PM, Yang Liu via USRP-users < >> usrp-users@lists.ettus.com> wrote: >> > >> > Dear all, >> > >> > I have a usrp x310 with two UBXs installed. And I want to use it as a >> MIMO transmitter (two outgoing channels streaming at the same frequency). >> > >> > When we need to build a MIMO system by connecting different usrps, I >> know that we need to have either MIMO cable or external time and frequency >> reference. In the one usrp x310 case, since two daughterboards seems to be >> independent, do we need to synchronize them? or they are already >> synchronized by default (for example, two local oscillators are >> synchronized, and will be tuned to exactly same frequency?)? >> > >> > Thanks, >> > Yang >> > ___ >> > USRP-users mailing list >> > USRP-users@lists.ettus.com >> > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >> > > ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Re: [USRP-users] use a usrp x310 as MIMO transmitter, daughterboard synchronization?
Hi Marcus, Thanks! I feel a little confused about the term 'frequency synchronization'. if the two LOs on the two daughterboads are using the same frequency reference, does it mean that after tuning, they will produce identical CFOs to the same receiver (one receiving channel) or we can say that the difference in CFOs are almost negligible? (lets for now ignore other factors that can contribute to CFO) I read through 'align LOs in the front-end' in the link https://files.ettus.com/manual/page_sync.html. timed command makes sure that different daughterboads can be tuned at the same time, so that it can only help to mitigate the phase offset, but it doesn't say anything about the potential frequency offset between the two daughterboards. Best, Yang On Wed, Jan 9, 2019 at 3:17 PM Marcus D Leech wrote: > Yes they’ll use the same reference clock provided by the motherboard. > > There’ll be a residual phase offset which can be mitigated using timed > tuning. > > Sent from my iPhone > > > On Jan 9, 2019, at 3:01 PM, Yang Liu via USRP-users < > usrp-users@lists.ettus.com> wrote: > > > > Dear all, > > > > I have a usrp x310 with two UBXs installed. And I want to use it as a > MIMO transmitter (two outgoing channels streaming at the same frequency). > > > > When we need to build a MIMO system by connecting different usrps, I > know that we need to have either MIMO cable or external time and frequency > reference. In the one usrp x310 case, since two daughterboards seems to be > independent, do we need to synchronize them? or they are already > synchronized by default (for example, two local oscillators are > synchronized, and will be tuned to exactly same frequency?)? > > > > Thanks, > > Yang > > ___ > > USRP-users mailing list > > USRP-users@lists.ettus.com > > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
[USRP-users] use a usrp x310 as MIMO transmitter, daughterboard synchronization?
Dear all, I have a usrp x310 with two UBXs installed. And I want to use it as a MIMO transmitter (two outgoing channels streaming at the same frequency). When we need to build a MIMO system by connecting different usrps, I know that we need to have either MIMO cable or external time and frequency reference. In the one usrp x310 case, since two daughterboards seems to be independent, do we need to synchronize them? or they are already synchronized by default (for example, two local oscillators are synchronized, and will be tuned to exactly same frequency?)? Thanks, Yang ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Re: [USRP-users] how to calculate tx power
Hi Marcus, Thanks a lot! One more question: does "5dB" below max mean that "5dB" below max gain setting? Best, Yang On Fri, Jun 1, 2018 at 1:38 PM, Marcus D. Leech wrote: > On 06/01/2018 12:57 PM, Yang Liu wrote: > > Hi Marcus, > > The usrpX310 (UBX160) I am using has the max power constraint, the reason > I ask this is that I don't want to the signal power to be too large such > that it will hit the nonlinear range. I am wondering whether I can get the > right amplitude and tx_gain combination given the max power constraint > through the calculation. > > Thanks, > Yang > > That's a different, but related, question. > > Set your modulation amplitude to about 0.8 and power at about 5dB below > max and you should have a clean signal coming out near > maximum power. But this depends a bit on the PAPR of your signal, so, > you'll have to experiment a bit. > > > On Fri, Jun 1, 2018 at 12:42 PM, Marcus D. Leech via USRP-users < > usrp-users@lists.ettus.com> wrote: > >> On 06/01/2018 12:37 PM, Yang Liu via USRP-users wrote: >> >>> Dear all, >>> >>> I feel a little bit confused about how to get the actual tx power. >>> Suppose I am generating a PSK signal and send it to usrp x310, and signal >>> amplitude is a, and usrp tx gain is b, I am wondering if the output power >>> is calculated according to the following equation: >>> output_power (db) = 20*log10(a) + b. >>> >>> For example, if the amplitude is 0.4 and tx_gain is set to 20, then the >>> output power is 12db (or 42dbm)? >>> >>> Thanks a lot for your clarification in advance. >>> >>> Best, >>> Yang >>> >>> >>> The output power is whatever you *measure* at the antenna port. These >> devices are necessarily uncalibrated. You'll have to measure power under >> your >> operating conditions. >> >> >> >> >> ___ >> USRP-users mailing list >> USRP-users@lists.ettus.com >> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >> > > > ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Re: [USRP-users] how to calculate tx power
Hi Marcus, The usrpX310 (UBX160) I am using has the max power constraint, the reason I ask this is that I don't want to the signal power to be too large such that it will hit the nonlinear range. I am wondering whether I can get the right amplitude and tx_gain combination given the max power constraint through the calculation. Thanks, Yang On Fri, Jun 1, 2018 at 12:42 PM, Marcus D. Leech via USRP-users < usrp-users@lists.ettus.com> wrote: > On 06/01/2018 12:37 PM, Yang Liu via USRP-users wrote: > >> Dear all, >> >> I feel a little bit confused about how to get the actual tx power. >> Suppose I am generating a PSK signal and send it to usrp x310, and signal >> amplitude is a, and usrp tx gain is b, I am wondering if the output power >> is calculated according to the following equation: >> output_power (db) = 20*log10(a) + b. >> >> For example, if the amplitude is 0.4 and tx_gain is set to 20, then the >> output power is 12db (or 42dbm)? >> >> Thanks a lot for your clarification in advance. >> >> Best, >> Yang >> >> >> The output power is whatever you *measure* at the antenna port. These > devices are necessarily uncalibrated. You'll have to measure power under > your > operating conditions. > > > > > ___ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
[USRP-users] how to calculate tx power
Dear all, I feel a little bit confused about how to get the actual tx power. Suppose I am generating a PSK signal and send it to usrp x310, and signal amplitude is a, and usrp tx gain is b, I am wondering if the output power is calculated according to the following equation: output_power (db) = 20*log10(a) + b. For example, if the amplitude is 0.4 and tx_gain is set to 20, then the output power is 12db (or 42dbm)? Thanks a lot for your clarification in advance. Best, Yang ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Re: [USRP-users] sending sine wave, but created another unwanted frequency component
Hi Dmitry, I did more tests today, found that the these two frequency components are not symmetric, and the distance between these two is 5e6 (sampling rate at the transmitter side). If I set the tx sampling rate to be 10e6, and waveform frequency to be 4.8e6 (close to 10e6/2.0), then the distance between two signals are 10e6. If this problem is caused by not enough attenuation, why as I moved the sine wave to the center frequency, the second component disappeared? I feel confused about this. I am thinking that maybe it is caused by incorrect sampling rate at the transmitter side, if I want to let the sine wave move from -5e6/2.0 to 5e6/2.0, should I set the sampling rate at tx higher than 5e6? Thanks a lot for your help, Yang On Mon, Jan 22, 2018 at 3:48 AM, Дмитрий Михайличенко via USRP-users < usrp-users@lists.ettus.com> wrote: > > > 2018-01-21 20:55 GMT+03:00 Yang Liu via USRP-users < > usrp-users@lists.ettus.com>: > >> Dear all, >> >> In this application, I am trying to send a sine wave at a specific >> frequency to usrp x310: >> sine wave generator ---> usrpx310 >> >> For the sine wave generator, I use blocks.sig_source_c from gnuradio. The >> parameters at the transmitter are in the following: >> center frequency: 1e9 (usrp tuning frequency) >> sampling_rate: 5e6 >> waveform frequency: -2.4193125e6 (at the boundary of this frequency >> band). >> >> At the receiver side, I tuned the usrp to 1e9, and used 10e6 to sample >> the received data. According to the spectrum I observed, there are two >> frequency components, one is at -2.4193125e6, another one is around at >> -2.4193125e6 + 5e6 (not very sure if they are exactly symmetric). Actually, >> this happens when the sine wave is very close to the boundary (near -2.5e6 >> or 2.5e6). As I moved the waveform frequency to the center (1e9), the >> second frequency disappeared. Firstly, I thought that it is the power >> issue, however, after I decreased power level, the second component is >> still there. >> >> According to the function (blocks.sig_source_c), what it generates is a >> exp(j*2*pi*f_waveform/f_s). Therefore, there should not exist any second >> frequency component. I feel very confused about why this can happen. >> >> Any thoughts about this will be greatly appreciated! >> >> Best, >> Yang >> >> >> >> ___ >> USRP-users mailing list >> USRP-users@lists.ettus.com >> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >> >> > What you see is perhaps image of original frequency. It should be missed > in an ideal world but in real world there is imbalance of quadrature > channels as well as filters used during downsampling have limited stop band > attenuation. > > thanks, > Dmitry > > > ___ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > > ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
[USRP-users] sending sine wave, but created another unwanted frequency component
Dear all, In this application, I am trying to send a sine wave at a specific frequency to usrp x310: sine wave generator ---> usrpx310 For the sine wave generator, I use blocks.sig_source_c from gnuradio. The parameters at the transmitter are in the following: center frequency: 1e9 (usrp tuning frequency) sampling_rate: 5e6 waveform frequency: -2.4193125e6 (at the boundary of this frequency band). At the receiver side, I tuned the usrp to 1e9, and used 10e6 to sample the received data. According to the spectrum I observed, there are two frequency components, one is at -2.4193125e6, another one is around at -2.4193125e6 + 5e6 (not very sure if they are exactly symmetric). Actually, this happens when the sine wave is very close to the boundary (near -2.5e6 or 2.5e6). As I moved the waveform frequency to the center (1e9), the second frequency disappeared. Firstly, I thought that it is the power issue, however, after I decreased power level, the second component is still there. According to the function (blocks.sig_source_c), what it generates is a exp(j*2*pi*f_waveform/f_s). Therefore, there should not exist any second frequency component. I feel very confused about why this can happen. Any thoughts about this will be greatly appreciated! Best, Yang ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
[USRP-users] usrp x310: send asynchronous pkts to two channels.
Dear all, I ran into one problem while using usrp x310. In the experiment, I tried to let both daughterboard A and B send different messages while listening to signals from both daughterboards as well. I found I can't establish 4 usrp object, since in this way signals can't be sent to right daughterboard. Based on this observation, I created one usrp object, and it has two transmitting channels and two receiving channels. I am not sure whether my observation is correct, can we create 4 usrp instances at a time? Since the messages I tried to send to daughterboard A and B are uncorrelated, I am wondering whether I can send asynchronous messages to A and B respectively. For example, at time t0, I start to send messages to daughterboard A, and then at time t1 (>t0), I want to send messages to daughterboard B. Thanks a lot for your help in advance! Thanks, Yang ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
[USRP-users] doubts about burst tranmission using tag SOB and EOB
Dear all, I tried to achieve burst transmission on usrp X310, the brief tx flowgraph is in the below: message_burst_source -> signal processing blocks --> usrp_sink New message will be inserted into the message queue of message_burst_source. Since message_burst_source will attach SOB and EOB tag to the data stream, usrp will know the start and end of each burst. After reading 'Busrt Transmission' section in this link: https://gnuradio.org/doc/doxygen/classgr_1_1uhd_1_1usrp__sink.html, my understanding is that because of these two tags, even the two message arrives at usrp at a different time, usrp will not complain about underrun. However, according to results from my experiments, as the time difference between two contiguous message increases, underflow will show up and become more and more severe. For a real burst transmission, time gap between two message should be expected. Therefore, I feel confused about the using of SOB and EOB tags here. Can they help to remove "underrun" under burst transmission? If not, is there any other methods that can make burst transmission with usrp possible without triggering underrun? Thanks a lot for your help, Yang ___ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com