MIMO utilization decreased after TM4 enabled

Hello everyone, we have enabled TM4 on a small cluster and previously there was TM3 enabled.
In post BM throughput has increased, traffic volume has increased, CQI has also improved and 64QAM modulation usage also improved as expected.
But MIMO utilization decreased significantly.
Any reaons why?

TM3 &TM4 are both MIMO (Open and closed loops resp.).
MIMO is still being used, but it could be rank1 Utilization has increased over rank 2.

I have exactly faced the same.
Rank 1 Utilization increased over Rank 2.
Rank 2 Utilization we say MIMO Utilization.
The question is how to correlate this change?
Duo to better feedback mechanism after enabling CL MIMO (TM4)?

Please check if your CQI report samples increased?
Or Average CQI improved?

Yes Average CQI has improved, just Rank 2 degraded.

Do you think Rank 1 is MIMO?

If our CQI improved than our UE should reply on Rank 2 rather than Rank 1?
Just want to highlight in TM4, we have 2 counters for Rank 1, i.e. Rank 1 Spatial multiplexing and Rank 1 Transmit diversity, unlike in TM3 where we have just one counter for Rank 1.

Yes, switching happen between single and dual codeword by cqi, ri threshold for tm4 mode.
For tm3 mode switching between open loop and tx diversity on basis of cqi, ri threshold.

Hi All, let me try to explain:

TM3 and TM4, Rank 1and rank 2 are all MIMO since the signal is transmitted over multiple Antennas.
Rank 2 is that 2 different code-works are de-multiplexed and then transmitted over 2 different Antennas (2*2 let’s say (2 on the left)).
If CQI is good, radio conditions are good, then conditions are me for rank 2 (2 different code-words).
If radio conditions (cell edge for example), CQI is bad, then 1 code-word is duplicated (not demux in this case) and then sent the same over the 2 different antennas, meaning here that rank 1 is used and usage increases over rank 1.
On cell level, if rank 1 increases after the change, check if the cell is overshooting, Out of Coverage %. Try to Tune cell edge users…then rank 2 increases further and also DL Throughput eventually increases.


Thanks but generally rank 1 linked with tx diversity TM2 ie MIMO not working as on both layer same codeword transmitted.
Generally in good network rank 2 % /> rank 1 % for that overlapping, overshooting to be optimised.

Correct. Rank 1 is not labled as MIMO generally.

In rare cases we have seen IBS deployments with single port antenna adding to the Rank1 proportion in a market, but cell wise this can be identified easily.

IBS use TM1 single Tx mode.
For TM2 as well rank 1 seen.

I think my explanation was not clear.

MIMO refers to having multiple data streams transmitted on the same time/frequency resource (which is rank).
Having duplicated stream transmitted on different antennas is TM2 (which is rank1).
But to optimize usage of rank2 is what mentioned above.
Also you can/may improve Rank2 by keeping nbofRxAntennas/nbOfTxAntennas parameter set to 2 (not 0 nor 1) ensuring that both antenna ports are enabled to transmit signals.

It’s clear. Only point, you mentioned rank 1 as MIMO, rest all look good.

But rank 1 can also be MIMO, diversity MIMO.

What I meant above by MIMO, the concept itself: mutiple input mutiple output, TM2 where we have 2 transmitting antennas is MIMO, but called transmit diversity.

1 Like

Still I a’int getting your point.

Formula for MIMO utilization is:
( TM3 RI2 + TM4 RI2 )
( TM 3 RI1 + TM3 RI2 + TM4 RI 1 TD + TM4 RI 1 SM + TM4 RI 2).

Now the concern is if TD and SM are MIMO or not, if not than why instead of improving my CQI, UE respond with RI 1?
Why not RI 2?

Which vendor nbofRxAntennas/nbOfTxAntennas ?

these 2 parameters are for E/// vendor