[srslte-users] Clock Rate Definitions and Overclocking

BALTACI, Ayguen ayguen.baltaci at airbus.com
Tue Jun 26 15:02:18 UTC 2018


Hi Xavier,

I just tried the benchmark test, my eNB and UE PCs are able to handle up to ~33 Msps.

Unfortunately, using otw format sc12 did not help me in this issue. I have tested it multiple times enabling it at UE, eNB and both, but the outcome was always the same. I reduced the sampling rate around ~33 Msps not to overflow the USRPs but the results did not change. I am wondering whether I change the sampling rate correctly in the source code.

The image “changingSamplingRate” is where I modify the sampling rate in phy_common.cc and “changingClockRate” is where I modify only the clock rate without changing the sampling rate (txrx.cc at eNB and phch_recv.cc at UE). When I change the sampling rate, UE keeps searching for eNB and when I change the clock rate, then UE keeps looping at RACH. Are these correct way to modify them?

By the way, I can run 75 PRB fine but I have problem running 100 PRB. Then, I changed the device_args to otw sc12 in 100 PRB but it did not get any better.

Thanks for your time in helping!


Best Regards

Aygun Baltaci | Master Thesis Student
Wireless Comm.| Airbus
+49-89-60736588<tel:%28904%29%20662-6564>
Willy-Messerschmitt-Straße 1,
82024 Taufkirchen



From: Xavier Arteaga [mailto:xavier at softwareradiosystems.com]
Sent: Dienstag, 26. Juni 2018 10:28
To: BALTACI, Ayguen
Cc: srslte-users
Subject: Re: [srslte-users] Clock Rate Definitions and Overclocking

Hi Aygun,
Have you tried to run this<https://kb.ettus.com/Verifying_the_Operation_of_the_USRP_Using_UHD_and_GNU_Radio#Benchmarking_your_system>?

I have just run a radio benchmark I wrote with srsLTE Radio API. I have experienced that otw format sc12 format is not only required for 2x2 MIMO at 20MHz cell, but also for sampling a single channel at 46.08MHz.

So, I am pretty sure you need to run with "device_args = type=b200,otw_format=sc12" both ue and enodeb.

Also, have you tried 75 PRB?

I would be happy to know how it goes :)

Regards,
Xavier

On 26 June 2018 at 09:31, BALTACI, Ayguen <ayguen.baltaci at airbus.com<mailto:ayguen.baltaci at airbus.com>> wrote:
Hi Xavier,

Thank you very much for the explanation, it makes more sense now.

I did not try the example applications, I was running the full emulator.

50 PRB cell works fine, so I don’t think it is related to USB3 throughput. I thought that I needed some RF calibration on SDR when I double the sampling rate using 25 PRB. I am not sure, though. I am not clear how doubling the sampling rate influences the system.

Best Regards
Aygun

From: Xavier Arteaga [mailto:xavier at softwareradiosystems.com<mailto:xavier at softwareradiosystems.com>]
Sent: Montag, 25. Juni 2018 17:33
To: BALTACI, Ayguen; srslte-users
Subject: Re: [srslte-users] Clock Rate Definitions and Overclocking

Hi Aygun,
The first clock rate is for cell search. The initial cell search and selection procedure is done at a sampling rate of 1.92MHz. Once the cell is found and selected, the sampling rate is set according to the cell bandwidth. Since the sampling rate during cell search stage is too low (<10MHz), the clock is set to a higher multiple.

You could try to run the lib/examples/pdsch_enodeb and the lib/examples/pdsch_ue examples with double sampling rate. Do they work ok?

Also, does a 50PRB cell work at double rate? You could have encountered some USB3 throughput issues if you are using the default configuration. By default, the over the wire format is in 16bit width integers while the ADC has only 12bits (in B210). you could add to device_args (in ue.conf) the parameter otw_format=sc12 for having a more compact format over the usb3.

Regards,
Xavier




On 25 June 2018 at 15:06, BALTACI, Ayguen <ayguen.baltaci at airbus.com<mailto:ayguen.baltaci at airbus.com>> wrote:
Hi mates,

I am new to the forum and srsLTE. I am running srsepc + srsenb + srsue. I am trying to understand why USRP board is set to two different clock rates during initialization (see the attached image). Is 30.72 MHz clock rate overwritten by 23.04 MHz in the end?

I want to overclock the system to measure the performance changes. Could you direct me how to do so? I modified the source code so that I could increase the clock rate from “23.04” MHz to “46.08” MHz (2nd one in the image) both at eNB and at UE, but then UE could not find the eNB.

I look forward to your suggestions,

Best Regards
Aygun

The information in this e-mail is confidential. The contents may not be disclosed or used by anyone other than the addressee. Access to this e-mail by anyone else is unauthorised.
If you are not the intended recipient, please notify Airbus immediately and delete this e-mail.
Airbus cannot accept any responsibility for the accuracy or completeness of this e-mail as it has been sent over public networks. If you have any concerns over the content of this message or its Accuracy or Integrity, please contact Airbus immediately.
All outgoing e-mails from Airbus are checked using regularly updated virus scanning software but you should take whatever measures you deem to be appropriate to ensure that this message and any attachments are virus free.

_______________________________________________
srslte-users mailing list
srslte-users at lists.softwareradiosystems.com<mailto:srslte-users at lists.softwareradiosystems.com>
http://www.softwareradiosystems.com/mailman/listinfo/srslte-users

The information in this e-mail is confidential. The contents may not be disclosed or used by anyone other than the addressee. Access to this e-mail by anyone else is unauthorised.
If you are not the intended recipient, please notify Airbus immediately and delete this e-mail.
Airbus cannot accept any responsibility for the accuracy or completeness of this e-mail as it has been sent over public networks. If you have any concerns over the content of this message or its Accuracy or Integrity, please contact Airbus immediately.
All outgoing e-mails from Airbus are checked using regularly updated virus scanning software but you should take whatever measures you deem to be appropriate to ensure that this message and any attachments are virus free.

The information in this e-mail is confidential. The contents may not be disclosed or used by anyone other than the addressee. Access to this e-mail by anyone else is unauthorised.
If you are not the intended recipient, please notify Airbus immediately and delete this e-mail.
Airbus cannot accept any responsibility for the accuracy or completeness of this e-mail as it has been sent over public networks. If you have any concerns over the content of this message or its Accuracy or Integrity, please contact Airbus immediately.
All outgoing e-mails from Airbus are checked using regularly updated virus scanning software but you should take whatever measures you deem to be appropriate to ensure that this message and any attachments are virus free.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20180626/7af048a3/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: changingClockRate.png
Type: image/png
Size: 19035 bytes
Desc: changingClockRate.png
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20180626/7af048a3/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: changingSamplingRate.png
Type: image/png
Size: 13108 bytes
Desc: changingSamplingRate.png
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20180626/7af048a3/attachment-0001.png>


More information about the srsran-users mailing list