Main differences between proportional fair scheduler in NR and LTE

I am looking for a detailed description about main differences between NR and LTE proportional fair scheduler, but I can not find a doc or …
If any body know about it, please share with me.
Thanks

Very probably such thing as scheduling is a proprietary solution. As far as I know, there is no direct and full description on how scheduling has to work either for 4G nor 5G in 3GPP specs. So, that’s why there is no informative docs or presentation in the Internet.

So how we can learn about them?
and differences? (specifically in NR and LTE)

You can search the credit based shaping schedule algorithm. Then mapping it to the QoS definition and related parameters in LTE and NR, i.e., (I am not sure exact terminologies, like Buffer Size, Maximal Rate, Maximal aggregated Rate etc.) LTE and NR have different specifications for each data bearer’s QoS, which is exactly the differences of LTE and NR for proportional fair scheduling.

1 Like

Hi 5G/LTE Experts.

When round robin scheduling is better than PF (proportional fairness) scheduling?

How do we measure fairness depend on what?

Round robin is like 100 PRB divided to 5 users each will get 20 RBs regardless of RF cobditions.

PF and EPF is much better.

Irrespective of either user need such RB’s or not, proportional fair maintain fairness among user as per its requirements of RB’s.

Thanks!

Any other suggestion for KPI test scenario in order to check performance of PF compared to RR (Round Robin scheduler)?

Like to measure throughput/fairness, any other criteria suggestion to check the performance of my PF scheduler compared to RR scheduler?

From KPI’s view, PRB utilization is the foremost KPI which may be impacted.
Because in RR scenario, UE has to wait in queue for scheduling as one by one each user will be scheduled. However, in EPF scenario, fairness is maintained by means of requirement of RB’s by UE and by virtue of this our resource grid will be utilized smartly and efficiently.

Means in round robin cell edge and cell mid user are being treated same albeit in EPF (enhanced proportional fair) cell edge users and middle users treated differently.

This will help in throughput of cell edge users.

Got you!

Any other suggested criteria to check the KPI performance of PF compared to RR other than cell throughput, fairness?

I mean by criteria. Suggested KPI test scenario in order to check the performance of OF compared to RR.

In case of RR , what could be possible disadvantage for cell - center (middle ) or nearer users if we compare it with PF or EPF? If any.

Means you are asking for field KPI?
Not getting your point.

So many. Check this doc where these queries already explained:

Comparing Different LTE Scheduling Schemes

2 Likes

I’m asking about any suggested test scenario KPIs for validating / checking my PF performance compared to RR.

Same as saying In case of RR, what could be possible disadvantage / advantage if we compare it with PF or EPF? If any.

Hope this document will address all of your queries :point_down:

1 Like