Hi Michael,
Here is the public repo with the custom block source code: https://st...@code.oceanit.com/stran/gr-per.git . The block is called Ansect Per ('Ansect' is the name of my own project I am using it for and 'per' stands for packet error rate). The source code is under /lib/ansect_per_impl.cc. Warning: I am not a programmer or GNU Radio savvy person, so I apologize in advance if my code is weird :). I kind of hacked something together that worked, but I hope it is useful to someone else! The README explains more. I updated UHD to 3.11 and now I don't get underruns and it doesn't freeze, but whatever is being sent through the radio isn't correct still. Essentially the tx LED comes on and the frequency spectrum looks good from the tx side, but the rx side gets garbled data (detects a lot of invalid packets I am using an X310 with two UBX-160 daughterboards, I also tried the WBX daughterboards which gave the same output. My version of GNU Radio is 3.7.11. ). I use the same computer to run the same code but using an N210 with a UBX-40 daughterboard, and I get the correct packet data (ofdm frequency spectrum looks great and good packets) on the rx side (using another computer + N210 to run rx_ofdm.py). This is the output from when I run tx_ofdm.py: ./tx_ofdm.py Gtk-Message: Failed to load module "overlay-scrollbar" [INFO] [UHD] linux; GNU C++ version 5.4.0 20160609; Boost_105800; UHD_3.11.0.1-3-g8bb15ee1 [INFO] [X300] X300 initialization sequence... [INFO] [X300] Determining maximum frame size... [INFO] [X300] Maximum frame size: 1472 bytes. [INFO] [X300] Setup basic communication... [INFO] [X300] Loading values from EEPROM... [INFO] [X300] Setup RF frontend clocking... [INFO] [X300] Radio 1x clock:200 [INFO] [RFNOC DMA FIFO] Running BIST for FIFO 0... [INFO] [RFNOC DMA FIFO] BIST passed (Throughput: 1318 MB/s) [INFO] [RFNOC DMA FIFO] Running BIST for FIFO 1... [INFO] [RFNOC DMA FIFO] BIST passed (Throughput: 1309 MB/s) [WARNING] [RFNOC] [0/Radio_0] defines 2 input buffer sizes, but 1 input ports [INFO] [RFNOC RADIO] Register loopback test passed [INFO] [RFNOC RADIO] Register loopback test passed [WARNING] [RFNOC] [0/Radio_1] defines 2 input buffer sizes, but 1 input ports [INFO] [RFNOC RADIO] Register loopback test passed [INFO] [RFNOC RADIO] Register loopback test passed [INFO] [CORES] Performing timer loopback test... [INFO] [CORES] Timer loopback test passed [INFO] [CORES] Performing timer loopback test... [INFO] [CORES] Timer loopback test passed I use the tx_ofdm code and take out the channel model, OFDM receiver, and Tag debug blocks. Instead I replace it with a UHD Sink block that has a samp rate of 1e6, center freq of 1.412e9, gain of 30 (absolute), tx/rx antenna, and bandwidth of 250kHz. When I hook up a frequency spectrum analyzer to the X310, I don't get the frequency response of the ofdm that I expect, but the same code will work for the N210. If you need more information to replicate the problem, please let me know! Thanks for your time, Sarah From: Michael Dickens <michael.dick...@ettus.com> Sent: Friday, March 16, 2018 2:54 AM To: Sarah Tran <st...@oceanit.com>; usrp-users@lists.ettus.com Subject: Re: [USRP-users] OFDM TX/RX BER Calculator and ofdm_tx underruns with x310 Hi Sarah - You're welcome & the PER block sounds cool ... do you have a public repo where we can view the source? Well done! As for your other question: Have you tried updating to the latest UHD release (3.11)? That might fix the issue "out of the box" ... and if not, try building from source via the latest GIT master branch. Even though 3.10 isn't really that old, it's old enough that updating might fix the issue. If not, then please provide the startup info from UHD to show its version & build, along with some code / specifics on how to replicate your issue & we'll definitely take a look & see what we can do. Cheers! - MLD On Thu, Mar 15, 2018, at 6:19 PM, Sarah Tran wrote: I was able to make a PER block and it turns out I was dropping some packets, so when I took into account those dropped packets when I performed the BER, I was in a much better place 10e-2. Thank you for the explanation and help! I had another question. I was previously using the N210 for this, and have since upgraded to the X310. However I get severe underruns (I tried the WBX and UBX-160 daughtercards). No matter what samp rate I use (from 400kHz to 10 MHz), I get lots of U's printed on my console and then the X310 stops transmitting and freezes. I saw a previous thread that mentioned it was a problem with the UHD 3.10 code (which I am running), but didn't say how it was fixed. (See http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/2017-March/023924.html). Do you have any idea what the problem could be? I also put a null source instead of file source and changed the uhd sink to a probe+message debug, and it looks like my rates are good 2e7. The N210 works great and the X310 will transmit other data at the same rates, but not the ofdm_tx code.
_______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
