Hi Guys.
Do you know what the scaling factor represents in the 5g throughput calculation?
When and why it would not be 1?
Not sure, @sadanandk2.
From the description, it mentions band and combo.
It is also signaled (if not 1, which is what I saw as typical for TDD) for both TDD and FDD. The DL/DL ratio comes indirectly into the formulas, the way I see it.
Scaling factor is for higher CC to calculate thrput. Like for 1 CC, scaling factor 1 for 2CC with different band & MIMO layers.
3GPP TSG RAN WG1 Meeting #92 R1-1801352:
But what does it mean, and even more important, why would you scale down the max peak tput when doing CA?
As I understood, in LTE UE can achieve max thrput by MIMO layer/Modulation/ CC but no concept of SF.
So you can achive max thrput by different combo while in NR, it has mention SF to calculate how to downgrade MIMO layer/modulation to have same aggregated thrput value.
Thanks @ajaymsfather, i agree with You.
Also, I’ve checked the document (very good info) and they detail in examples.
Hi Experts.
About use of Scaling Factor in Throughput calculations in 5G:
How to decide among values 1, 0 .8, 0.75, 0.4?
Which have to choose when we go with 1cc and single layer data?
As far as I remember is for CA in NR in the idea that some bands only 2 layers are allowed.
So in 1CC it’s value should be 1?
Yes, I think so.
Scaling factor is signalled per band and per band combination.
So is about CA, layer and modulation order.
Source (and link with explanation): 5G UE Data Rate
Thanks, I got it.
Hi @RFSpecialist: Why UE bpc is 100 MHz, gNB configure 2CC 100 MHz for this UE?
Hi , I try to understand one thing:
- If phone supports 4 layers with 256qam for both TDD carrier N78, is that really a matter of the value of scaling factor?
Imagine B3 (4X4) + N78+N78 (4X4 each) and UE does not specify its scaling factor in UE capability.
and in that case if gnb assigns scaling factor as 1 then we should reach peak throughput?
Ty
It is an optional parameter that most of the times can just be ignored.
Refer to R1-1801352;
" The scaling factor is introduced to reflect the association of the maximum number of layers and the maximum modulation order with the band combination. If the maximum number of layers and the maximum number of modulation orders are reported per band and per band combination, the scaling factor is no more necessary."
So just ignore it. If you already have the modulation used for each layer/CC, there is no need to use the scaling factor. Its a mathematical thing they made to lower the average modulation order considering that in some cases it’s not possible to have the highest modulation order in every CC of a combo.