CQI is always 0 when disabling MIMO feedback in cttc-nr-mimo-demo example

[Maria Eduarda Veras Martins]

Hello everyone,

I am running the simulation with periodic channel fading updates enabled by setting

updatePeriodMs=5.
When I run the simulation with MIMO feedback enabled (the default setting, --enableMimoFeedback=true), the CQI value varies over time as expected, following the channel's fading dynamics. Everything seems correct in this scenario, and I can plot the CQI variation over time.

However, when I disable the MIMO feedback by passing the --enableMimoFeedback=false command-line argument, I observe that the reported CQI is always 0 for the entire duration of the simulation.

Is this the expected behavior? It seems a bit strange to me. I would expect the UE to still report a valid CQI based on the channel conditions for a non-MIMO transmission (e.g., SISO), which should vary with the fading, rather than staying fixed at 0.

Thank you for your time and help!

[Gabriel Ferreira]

Seems like a oopsie. Would you mind retesting with
Config::SetDefault("ns3::NrHelper::CsiFeedbackFlags", UintegerValue(CQI_PDSCH_SISO));

It is currently being set to CQI_PDSCH_MIMO, which is kind of misdealing because this is part of MIMO feedback =x

[Gabriel Ferreira]

*misleading

[Gabriel Ferreira]

Opened issue to track this https://gitlab.com/cttc-lena/nr/-/issues/255

[Gabriel Ferreira]

Issue fixed upstream.