On 12/02/2014 01:35 AM, WuHaoyang via       USRP-users wrote:
 
 Hello, thanks for your comments!
   
   1. I'm very sure I've saved the correct complex data in a .txt       file. the complex data use "float32" format, each complex data has       64bits. I did a lot of similar experiment with single antenna       mode, so I'm sure the file data is correct.
   
   2. How can I specify the subdev spec accordingly (A:0 B:0), just       add "(A:0 B:0)" following the "Mb0: Subdev Spec"?
      Just use:
 
 A:0 B:0
 
 In the subdev spec.
 
 
   3. What does "L" & "S" mean exactly?
      L = Late packet.   The packet arrived with a time-spec on it, but     that time had already come-and-gone when the packet arrived at the     USRP
 S = Sequence error.  usually because the network interface is     dropping packets
 
 Are you doing this on a VM, rather than a naked hardware machine?
 
 
 
            ------------------
     Wu           Haoyang
     
   
    
            
     
     
     
     ------------------ Original           ------------------
                From:  "usrp-users";<usrp-users@lists.ettus.com>;
       Date:  Thu, Nov 27, 2014 06:46 PM
       To:  "usrp-users"<usrp-users@lists.ettus.com>;             
       Subject:  Re: [USRP-users]回复： Can USRP X300             support 2*2 MIMO
     
     
     
     A couple of comments:
     
     - Are you truly saving complex data in a .txt file? Usually .txt
     indicates something else. The same file->throttle->Freq         Sink will show
     the correct spectrum, right?
     
     On 11/27/2014 04:05 AM, WuHaoyang via USRP-users wrote:

     - For X300 in particular, we have fixed a lot of things since         3.7.0. I
     recommend an update.
     

     To reiterate, please make sure the file data is correct.
     

     - You shouldn't need to, but you might want to specify the         subdev spec
     accordingly (A:0 B:0).
     

     - Red means 'TX', so according to your pic, you *are* tx'ing on         both
     channels.
     

     These are fine.
     

     These are not good. I wonder if the MIMO setup is too fast... Is         there a
     chance you could try running GNU Radio master branch?
     
     Cheers,
     M
     
     
     _______________________________________________
     USRP-users mailing list
     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

‍ I'm doing this on a naked hardware machine, not VM. In single antenna mode (using gigabit ethernet port), we can set the sample rate up to 20MHz. but when we try to use MIMO, we can see the "L" & "S" even when we set the sample rate under 1MHz. ------------------ Original ------------------ From: "usrp-users";<usrp-users@lists.ettus.com>; Date: Tue, Dec 2, 2014 02:42 PM To: "usrp-users"<usrp-users@lists.ettus.com>; Subject: Re: [USRP-users] Can USRP X300 support 2*2 MIMO On 12/02/2014 01:35 AM, WuHaoyang via USRP-users wrote: Hello, thanks for your comments! 1. I'm very sure I've saved the correct complex data in a .txt file. the complex data use "float32" format, each complex data has 64bits. I did a lot of similar experiment with single antenna mode, so I'm sure the file data is correct. 2. How can I specify the subdev spec accordingly (A:0 B:0), just add "(A:0 B:0)" following the "Mb0: Subdev Spec"? Just use: A:0 B:0 In the subdev spec. 3. What does "L" & "S" mean exactly? L = Late packet. The packet arrived with a time-spec on it, but that time had already come-and-gone when the packet arrived at the USRP S = Sequence error. usually because the network interface is dropping packets Are you doing this on a VM, rather than a naked hardware machine? ------------------ Wu Haoyang ------------------ Original ------------------ From: "usrp-users";<usrp-users@lists.ettus.com>; Date: Thu, Nov 27, 2014 06:46 PM To: "usrp-users"<usrp-users@lists.ettus.com>; Subject: Re: [USRP-users]回复： Can USRP X300 support 2*2 MIMO A couple of comments: - Are you truly saving complex data in a .txt file? Usually .txt indicates something else. The same file->throttle->Freq Sink will show the correct spectrum, right? On 11/27/2014 04:05 AM, WuHaoyang via USRP-users wrote: > version information: > UHD version: 003.007.000 - For X300 in particular, we have fixed a lot of things since 3.7.0. I recommend an update. > This is our flowgraph, Two "File Sources" include different time domain > waveforms that generated by ourselves. We try to send different signal > with two TX antennas simultaneously. To reiterate, please make sure the file data is correct. > Following picture is the configuration of USRP Sink module. - You shouldn't need to, but you might want to specify the subdev spec accordingly (A:0 B:0). > when this test case is executed, we can not see the correct wave in RX > end. It seems that only the signal in Channel0 has been send out.(we use > single RX antenna, not MIMO mode, to receive and save the TX wave, and > print the waveform using MATLAB). > > The following picture shows the front panel of X300, > > We can see the LED behave like this: > TX/RX: Red - Red means 'TX', so according to your pic, you *are* tx'ing on both channels. > REF: Green > PPS: Green, twinkle > LINK: sometimes Red, sometimes Green These are fine. > Also we can see a lot of characters "L" and "S" in GNU Radio control panel. These are not good. I wonder if the MIMO setup is too fast... Is there a chance you could try running GNU Radio master branch? Cheers, M _______________________________________________ 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

On 12/02/2014 02:03 AM, WuHaoyang wrote: > ‍ I'm doing this on a naked hardware machine, not VM. In single > antenna mode (using gigabit ethernet port), we can set the sample rate > up to 20MHz. but when we try to use MIMO, we can see the "L" & "S" > even when we set the sample rate under 1MHz. > That absolute maximum aggregate sample-rate that can be supported over a single 1GiGe link is 25Msps, so if you're trying to stream two channels at those rates, it won't work. This is just straight 1GiGe link occupancy. > > > ------------------ Original ------------------ > *From: * "usrp-users";<usrp-users@lists.ettus.com>; > *Date: * Tue, Dec 2, 2014 02:42 PM > *To: * "usrp-users"<usrp-users@lists.ettus.com>; > *Subject: * Re: [USRP-users] Can USRP X300 support 2*2 MIMO > > On 12/02/2014 01:35 AM, WuHaoyang via USRP-users wrote: >> Hello, thanks for your comments! >> >> 1. I'm very sure I've saved the correct complex data in a .txt file. >> the complex data use "float32" format, each complex data has 64bits. >> I did a lot of similar experiment with single antenna mode, so I'm >> sure the file data is correct. >> >> 2. How can I specify the subdev spec accordingly (A:0 B:0), just add >> "(A:0 B:0)" following the "Mb0: Subdev Spec"? > Just use: > > A:0 B:0 > > In the subdev spec. > >> >> 3. What does "L" & "S" mean exactly? > L = Late packet. The packet arrived with a time-spec on it, but that > time had already come-and-gone when the packet arrived at the USRP > S = Sequence error. usually because the network interface is dropping > packets > > Are you doing this on a VM, rather than a naked hardware machine? > > >> >> ------------------ >> Wu Haoyang >> >> >> ------------------ Original ------------------ >> *From: * "usrp-users";<usrp-users@lists.ettus.com>; >> *Date: * Thu, Nov 27, 2014 06:46 PM >> *To: * "usrp-users"<usrp-users@lists.ettus.com>; >> *Subject: * Re: [USRP-users]回复： Can USRP X300 support 2*2 MIMO >> >> A couple of comments: >> >> - Are you truly saving complex data in a .txt file? Usually .txt >> indicates something else. The same file->throttle->Freq Sink will show >> the correct spectrum, right? >> >> On 11/27/2014 04:05 AM, WuHaoyang via USRP-users wrote: >> > version information: >> > UHD version: 003.007.000 >> >> - For X300 in particular, we have fixed a lot of things since 3.7.0. I >> recommend an update. >> >> > This is our flowgraph, Two "File Sources" include different time domain >> > waveforms that generated by ourselves. We try to send different signal >> > with two TX antennas simultaneously. >> >> To reiterate, please make sure the file data is correct. >> >> > Following picture is the configuration of USRP Sink module. >> >> - You shouldn't need to, but you might want to specify the subdev spec >> accordingly (A:0 B:0). >> >> > when this test case is executed, we can not see the correct wave in RX >> > end. It seems that only the signal in Channel0 has been send >> out.(we use >> > single RX antenna, not MIMO mode, to receive and save the TX wave, and >> > print the waveform using MATLAB). >> > >> > The following picture shows the front panel of X300, >> > >> > We can see the LED behave like this: >> > TX/RX: Red >> >> - Red means 'TX', so according to your pic, you *are* tx'ing on both >> channels. >> >> > REF: Green >> > PPS: Green, twinkle >> > LINK: sometimes Red, sometimes Green >> >> These are fine. >> >> > Also we can see a lot of characters "L" and "S" in GNU Radio >> control panel. >> >> These are not good. I wonder if the MIMO setup is too fast... Is there a >> chance you could try running GNU Radio master branch? >> >> Cheers, >> M >> >> >> _______________________________________________ >> 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 > >