running srsENodeB on pluto

[fpgasdr]

Hi again,

just a wish for new year, it's possible?. how?

if not, how far is from been able to run o adalm pluto?.

I have tried but the only way to have connection with pluto board is via Soapypluto but no luck...

thanks

[LunaWuna]

I've tried on the Pluto plus board and this is what I get when running the built SRS eNB srsran/srsRAN_4G#1053

[ofontbach]

I've tried on the Pluto plus board and this is what I get when running the built SRS eNB srsran/srsRAN#1053

Dear @fpgasdr and @LunaWuna,

the same answer provided in #25 and #26 is valid for this. Please, give it a try. In case of need, please open zynq related issues in this repository, not in the main srsRAN one. Thanks!

[ofontbach]

@LunaWuna, we don't have a pluto plus and hence we can't guarantee it will work. On paper, the pluto plus should be quite similar to the AntSDR so as @fpgasdr says, the supported data-rates should not be constrained in the same way as when using usb 2.0.

[fpgasdr]

thanks @ofontbach for reply.

as you suggest I updated the repo and after compile and generate the new bitstream now the srsENodeB starts but as I don't have anything to test it the only output I see is this:

``julian@julian-Aspire-X3810:~/Documents/pluto/zynq_timestamping/app$ cd ./bin_app/srsenb; sudo nice -20 ./srsenb ./enb.conf --rf.device_name=Soapy
Active RF plugins: libsrsran_rf_uhd.so libsrsran_rf_soapy.so libsrsran_rf_zmq.so
Inactive RF plugins:
--- Software Radio Systems LTE eNodeB ---

Reading configuration file ./enb.conf...
WARNING: cpu0 scaling governor is not set to performance mode. Realtime processing could be compromised. Consider setting it to performance mode before running the application.

Built in Release mode using commit 254cc719a on branch HEAD.

Opening 1 channels in RF device=Soapy with args=default
Supported RF device list: UHD soapy zmq file
connect(): Connection refused
Failed to initiate S1 connection. Attempting reconnection in 10 seconds
Soapy has found device #0: device=PlutoSDR, driver=plutosdr, label=PlutoSDR #0 usb:1.15.5, uri=usb:1.15.5,
Soapy has found device #1: device=PlutoSDR, driver=plutosdr, label=PlutoSDR #0 ip:pluto.local, uri=ip:pluto.local,
Selecting Soapy device: 0
[INFO] Opening PlutoSDR #0 usb:1.15.5...
Setting up Rx stream with 1 channel(s)
[INFO] Using format CF32.
[INFO] Auto setting Buffer Size: 524288
[INFO] Set MTU Size: 524288
Setting up Tx stream with 1 channel(s)
[INFO] Using format CF32.
[INFO] Has direct TX copy: 1
[INFO] Auto setting Buffer Size: 32768
[INFO] Set MTU Size: 32768
Available device sensors:

• xadc_temp0
• xadc_voltage0
• xadc_voltage1
• xadc_voltage2
• xadc_voltage3
• xadc_voltage4
• xadc_voltage5
• xadc_voltage6
• xadc_voltage7
• xadc_voltage8
• adm1177_current0
• adm1177_voltage0
• ad9361-phy_temp0
• ad9361-phy_voltage2
  Available sensors for Rx channel 0:
  State of gain elements for Rx channel 0 (AGC supported):
• PGA: 71.00 dB
  State of gain elements for Tx channel 0 (AGC not supported):
• PGA: 80.00 dB
  Rx antenna set to A_BALANCED
  Tx antenna set to A

==== eNodeB started ===
Type to view trace
[INFO] Auto setting Buffer Size: 32768
[INFO] Set MTU Size: 32768
Setting frequency: DL=2680.0 Mhz, UL=2560.0 MHz for cc_idx=0 nof_prb=6
[INFO] Has direct RX copy: 1

I have another adalm pluto board where I can start the srsUE but this is other topic so maybe I can open in another thread.

but the output is strange:

``julian@julian-Aspire-X3810:~/Documents/pluto/zynq_timestamping/app$ sudo LD_LIBRARY_PATH=./bin_app nice -20 ./bin_app/srsue/srsue ./bin_app/srsue/ue.conf --gw.netns=ue1
Reading configuration file ./bin_app/srsue/ue.conf...
WARNING: cpu0 scaling governor is not set to performance mode. Realtime processing could be compromised. Consider setting it to performance mode before running the application.

Built in Release mode using commit 254cc719a on branch HEAD.

Opening 1 channels in RF device=iio with args=n_prb=6,context=ip:192.168.3.1
Active RF plugins: libsrsran_rf_iio.so
Inactive RF plugins:
Supported RF device list: iio
CH0 n_prb=6
CH0 context=ip:192.168.3.1
failed to create the rf_port with A_BALENCED
failed to create the rf_port with A
Waiting PHY to initialize ... done!
Attaching UE...
Setting manual TX/RX offset to 82 samples
..........................................................................

it keeps outputting dots randomly...what does it means? (I',m new to srsRAN sorry).

of course I launch the EPC in another terminal:

`julian@julian-Aspire-X3810:~/Documents/pluto/zynq_timestamping/app$ cd ./bin_app/srsepc; sudo ./srsepc ./epc.conf

Built in Release mode using commit 254cc719a on branch HEAD.

--- Software Radio Systems EPC ---

Reading configuration file ./epc.conf...
HSS Initialized.
MME S11 Initialized
MME GTP-C Initialized
MME Initialized. MCC: 0xf001, MNC: 0xff01
SPGW GTP-U Initialized.
SPGW S11 Initialized.
SP-GW Initialized.
`

[mbound]

I get to the same point with srsENB and the Pluto (standard Pluto). Checking with another SDR tuned on the DL frequency, GQRX shows a brief burst upon eNB initialization, but then nothing. No transmission from the eNB, seems like the samples don't actually get to the DAC, but no error is reported by SRS.

Does anyone know if there is some sort of debug trace function on the Pluto itself that could provide useful info?

[mbound]

I tried again with configuration suggested by @sauka to force the use of the rf_iio driver and I get this message repeating (still no transmission):

misaligned packet received from the DMA

[sauka]

Hey guys, we did a quick test running srsenb with antSDR as an RF frontend and the srsue could successfully attach; ping works well in both directions. In this test srsue from time to time reports underflow or overflow (I guess it has to do something with host PC performance) and as a result radio link failure occurs, but UE reattaches to the eNodeB and keeps working. Example:

I am sure performance can be improved by using more powerful PC and tuning better freq_offset and time_advance_nsamples parameters.

We didn't test it with Pluto+, however as Oriol said the USB2 interface can be a bottleneck. You can give it a try and share results here ;)

[fpgasdr]

Hi @sauka,

I have 2 adalm pluto and would like to test the same as you.

how you manage to make it?

just following the doc?

https://srsran.github.io/zynq_timestamping/app/antsdr.html

[sauka]

Hey @fpgasdr ,

Right, I was following that appnote, though I had to do a few minor changes. In my setup I have b200mini and antSDR. To use srsue with b200 I modified device_args in ue.conf and also I ran the command without setting LD_LIBRARY_PATH environment variable, so the command looks simply as:
sudo nice -20 ./srsue/srsue srsue/ue1.conf --gw.netns=ue1

Here are the RF related parameters from ue1.conf that worked well for me (device_args is just an example):

[rf]
freq_offset = 400
tx_gain = 80
rx_gain = 80
device_name = uhd
device_args = serial=318A396
time_adv_nsamples = 135

On the other hand to use srsenb with antSDR I've set LD_LIBRARY_PATH env (to make it using rf_iio driver from this repo), so I ran it like this:
pavel@fpga-master:~/antsdr_4g_test/zynq_timestamping/app/bin_app/srsenb$ sudo LD_LIBRARY_PATH=$(pwd)/.. nice -20 ./srsenb enb1.conf

Here are the RF related parameters from enb1.conf:

pavel@fpga-master:~/antsdr_4g_test/zynq_timestamping/app/bin_app/srsenb$ cat enb1.conf
[rf]
#dl_earfcn = 3350
tx_gain = 80
rx_gain = 80
device_name = iio
device_args = n_prb=6,context=ip:10.12.1.201

When I have a chance to try it with plutoSDR, I'll give it a go.

[woohuan]

In "End-to-End 4G testing with AntSDR", eNodeB is run on MicroPhase AntSDR, but there will be "Error reading rx ringbuffer. Invalid header" error message. Is there any solution, please?

[woohuan]

Additional information:
Before the eNodeB error is reported, the MicroPhase board outputs another error message:
AD9361 spi0.0: ad9361_calculate_rf_clock_chain: Failed to find suitable dividers: ADC clock below limit