Can same NR cell be used as NSA cell and SA cell simultaneously?

BWP switching comes from DCI, isn’t it?
Max 4 BWP configured but any of it can be used, but only one at a time.

Yes is it feature in network.
I mean some switch or there is some threshold criteria.

Yes, and UE can be switched to monitor csi rs rather than ssb in this active bwp.
Going idle ue will switch to default bwp.
Muliple default bwp can be used if the bandwidth division is based on service type - ex urllc.

Not only dci, there is some other way also to switch bwp.

Wait: mobility is based on SSB measuremnts while scheduling is based on csi-rs measurements.
So UE has to read time to time SSB.

Mobility can be csi-rs based also.

We have bwp - inactivity timer as well.

Yes agree on basis of csi-rs beam ie refined beams.

I meant mobility between cells, not between beams.

Under report config to add mod list there is IE rs type which can be ssb or csi-rs.

Need to check between cells can we use csi-rs or not.

In that case, it is like an inter freq measurement for UE - if the active bwp of current cell and ssb containing bwp of neighbor cell is different bandwidth.

You’re right @RFSpecialist, for cell mobility it’s ssb rsrp not csi-rs.
Csi-rs used for beam switching only.

And for scheduling.
csi-rs measurements are also used for scheduling I think, because cqi RI and PMI are based on csi-rs.

Yes that is csi-rs RE case not beam.
We discussed last year same in detail for csi-rs, csi-rs beam.
If I am not wrong here: Csi-rs beams (beamformed) versus csi-rs reference signals (from one single antenna port)

Seems csi-rs signal used as well for cell level.

Ok, so there can be mobility between cells based on csi-rs but how many networks are like this?
All that I have seen are based on ssb measurements.
I mean for reportconfigNR rs type is always 0 (SSB).

Even I didn’t seen as well.
Let me try to find if get it.

Nokia, Samsung do not support mobility based on CSI-RS.
It is an addition overhead to transmit CSI-RS along with SSB.

Right, but coming back to original query: do we have more SSB blocks in frequency or just one in 20 PRBs related to pointA?

Yes it possible to have more tan 1 SSB in freq domain.

Having separate channel and synchronisation rasters means that the poisiton of the SS/PBCH Block within the channel bandwidths is not fixed. With the exception of the smaller channel bandwidths there is a choice of SS/PBCH Block positions within the channel bandwidth. Cells using larger channel bandwidths can transmit multiple SS/PBCH Blocks distributed across the channel bandwidth.

Handover decision will be based on SSB.
But if you activate Beam Refinement or PMI beam, CSI-RS will be reported for refined beams.

Yes , it should be - when going idle , UE default bwp can be changed - therefore in this case ssb need to be there in that bwp.