Hi Don, First of all, I would use Time Scope without Unbuffer instead of Simulink Scope. It will be much faster.
Can you send a picture of the model(s) you use for this test? Do you monitor the underrun and overrun ports to make sure that the model runs in real-time? What is the sample rate (master clock rate and decimation/interpolation rate)? What is the frame size (samples per frame)? Regards, Ethem From: Kane, Don [US] (MS) [mailto:donald.k...@ngc.com] Sent: Thursday, January 25, 2018 2:17 PM To: Ethem Sozer <ethem.so...@mathworks.com> Cc: usrp-users@lists.ettus.com; Kane, Don [US] (MS) <donald.k...@ngc.com> Subject: RE: Matlab/Simulink to USRP in Burst Mode Ethem, Inserting zeroes seems to work as far as stopping transmission. To validate this, we put an Unbuffer block and Simulink Scope block on the SDRu Receiver Data port to see what the receiver is putting into our model. We tried several experiments with sending one frame, two frames, four frames or eight frames. Zeroes were inserted between each set of frames. The received signal shows the set of frames with gaps in between frames and it is inconsistent. Example: Eight frames comes in as a four frames grouped together and then gaps between the remaining four frames, six frames grouped together and then gaps between the last two frames, all eight frames with gaps between each frame, etc. There is no consistency in the received signal. We would like all the received frames to come out of the receiver as one continuous stream. We see inconsistency whether we send in groups of two frames, four frames or eight frames. Examples from our eight frame experiment is shown below and these are from the same run. [cid:image001.jpg@01D395E9.181E16A0] [cid:image002.jpg@01D395E9.181E16A0] Don From: Ethem Sozer [mailto:ethem.so...@mathworks.com] Sent: Thursday, January 25, 2018 7:02 AM To: Kane, Don [US] (MS) <donald.k...@ngc.com<mailto:donald.k...@ngc.com>> Cc: usrp-users@lists.ettus.com<mailto:usrp-users@lists.ettus.com> Subject: EXT :RE: Matlab/Simulink to USRP in Burst Mode Hi Don, The only supported way to stop transmissions from SDRu Transmitter block is to insert "zeros" to the signal stream when you do not want to transmit. You need to keep feeding the transmitter with data (zeros or real data). The burst mode is designed to enable the use of these blocks when the algorithm that supplies the transmitter (or the algorithm that consumes the output of the receiver block) cannot run in real-time. In this case, you can run your algorithm non-real-time, create your samples to be transmitted as a burst and pass to the block. Then the block will transmit these signals without any spaces between the frames. This will result in an RF stream that has properly transmitted bursts of frames separated by empty/un-defined spots. Regards, Ethem From: USRP-users [mailto:usrp-users-boun...@lists.ettus.com] On Behalf Of Kane, Don [US] (MS) via USRP-users Sent: Wednesday, January 24, 2018 1:43 PM To: usrp-users@lists.ettus.com<mailto:usrp-users@lists.ettus.com> Cc: Kane, Don [US] (MS) <donald.k...@ngc.com<mailto:donald.k...@ngc.com>> Subject: [USRP-users] Matlab/Simulink to USRP in Burst Mode We have our QAM waveform running successfully using Matlab/Simulink and the X310 USRP. We run with two laptops and two X310 USRPs and we can continuously sends Pings between the two laptops. We get no underruns and everything works fine as long as the data stream is continuous to the USRP (SDRu Transmitter block). Now we want to use the USRPs in burst mode where we only want the USRP to transmit when data is present. We have tried several combinations of setting the Enable Burst Mode checkbox in both the SDRu Transmitter and SDRu Receiver block. We have also tried putting an Enable on the SDRu Transmitter block. We have tried sending from 1-8 frames periodically to the SDRu Transmitter block. We put a scope on the receive side and never see a corresponding reception that shows only the frames we thought we sent. Question: How do we get the SDRu Transmitter to transmit only when frames are sent to it? Don Kane System Architect NORTHROP GRUMMAN Mission Systems Network Communications Division Resilient Network Systems 9326 Spectrum Center Blvd. San Diego, CA 92123 858-514-9355 donald.k...@ngc.com<mailto:donald.k...@ngc.com>
_______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
