Idle Mode Cell Selection not working with newly added EUTRA band

[Leonard Fisser]

I am using the LTE module of the simulator to look at some basic performance metrics.

I added Band 31 to the lte-spectrum-value-helper by adding the following entry to the EutraChannelNumbers struct:  

`{ 31, 462.5, 9870, 9870, 9919, 452.5, 27760, 27760, 27809}`

While methods like propagation loss models, REM, and SNIR calculation work flawlessly, the idle mode cell selection in an EPC-scenario does not work correctly.

No PSS measurements seem to arrive at the RLC and as such, the UE is stuck in cell searching without getting any measurement values.

If I switch back to Earfcn 100/18100 my scenario works fine, only the newly added channel seems to cause problems.

Is adding channels not as simple as I thought, did I miss parameters?

I would be happy about any direction, as the PHY model can be quite complicated at times.

[Zoraze Ali]

Hi Leonard,

I had a quick look. Check this line: https://gitlab.com/nsnam/ns-3-dev/-/blob/master/src/lte/model/lte-spectrum-value-helper.cc#L105. Could you extend this "if" to take in the consideration the DL EARFCN of the new band?

Kind regards,

Zoraze

[Biljana Bojović]

Did you try to enable logs for LteUePhy and LteSpectrumValueHelper just to make sure that there everything is in place? E.g. I am interested in log of function void
LteUePhy::DoStartCellSearch (uint32_t dlEarfcn).


[Leonard Fisser]

Hi Zoraze.

thanks for you quick response.
Yes I modified that if statement to make sure FDD is correctly used. Logging the UL and DL frequencies in manual attachment scenarios looks fine to me as well.

Kind regards,
Leonard

P.S.: Sorry for posting on the tracker earlier :)

[Leonard Fisser]

Hi Biljana,

thanks for the quick reply, I only logged LteUePhy and LteUeRlc during my debug, so I just looked into LteSpectrumValueHelper as well. As far as I can tell the DoStartCellSearch is called correctly and PSD are being computed (Find a log snippet in the attached .txt)

Best regards,

Leonard

[Biljana Bojović]

Hi Leonard,

can you please confirm that you are configuring the following attribute of UE device according to earfcn that you use in your configuration?

 .AddAttribute ("DlEarfcn",
                   "Downlink E-UTRA Absolute Radio Frequency Channel Number (EARFCN) "
                   "as per 3GPP 36.101 Section 5.7.3. ",
                   UintegerValue (100),
                   MakeUintegerAccessor (&LteUeNetDevice::SetDlEarfcn,
                                         &LteUeNetDevice::GetDlEarfcn),

I understand that the problem is when you are NOT using earfcn 100, right?

BR,
Biljana

[Leonard Fisser]

Hi Biljana,

I feel dumb, I forgot to set the UeNetDevice parameter and only set the EnbNetDevice EARFCN values.. hence the mismatch and no cell identification..

Config::SetDefault ("ns3::LteEnbNetDevice::DlEarfcn", UintegerValue(9895));

Config::SetDefault ("ns3::LteEnbNetDevice::UlEarfcn", UintegerValue(27785));

Config::SetDefault ("ns3::LteUeNetDevice::DlEarfcn", UintegerValue(9895)); // Forgot this parameter

Thank you for taking the time to help. Much appreciated.

Best regards,
Leonard Fisser