Agreed, if the 300ms doesn’t represent a drawback on the operator’s perspective, it is fine 
A drawback it is, surely, but if compensates the goal of redirecting traffic.
This is the risky solution:
Set up an sevice based A2 event in TDD for redirection. Keep A2 value highest like -43 dBm.
By the time ue report MR, all SIP communication is already done.
Once eNB recives A2, it will immediately redirect UE.
Hence no HO preparation delay.
Risk is as below:
- If ue do not get good coverage of redirected FDD ,call is screwed uo. In HO, we check RF xondition of target.
- If redirected FDD has reach to 100% CAC, rrc connection request may not get accept at all
- If you are using any server based aplication for measuring muting, that aplication cannot measure mute time caused by redirection.
But why would you set up A2 at -43dBm?
You would need the UE to be at the towers feet! LOL!
And for this scenario, no matter if the HO is delayed, the UE will continue to receive the QCI1 and QCI5 packets with no hassle (just tsome measurement gaps).
I dont like SBHO to be honest, but, if you must do it, you should use reasonable A2 values :).
And maybe consider SPID as a possible solution.
It’s less than -43 dBm. Not better than
True that, I read A4!
So you want to use regular A2 to push traffic out? Not SVS Based HO?
SVS based of course.
Then no A2, it is QCI1 establishment -> RRC Reconfiguration with A4.
Do you see implementations with A2?
Yes i did that.
Serving to measure here, just for triggering redirection.
No measuremeng of target here.
Thats how we shift more volte traffic.
Are you talking about service based HO or Redirection?
A2 for Service based HO.
And then redirection.
So, you trigger the UE to another layer by measuring the serving cell not the target cell?
HO followed by redirection?
Ok, now you lost me.
Dont get confused, no HO happening here.
A2 just doesnt make HO.
Just to trigger redirection we make it onve MR received for A2 SVS.
Is it ok now?
Ok, forget HO.
Once the UE establishes QCI1 it is reconfigured with an A2 at -45 dBm to measure the serving cell, then once it reports, the eNB triggers a redirection to a predefined target EARFCN?
You got it 
.
Very interesting strategy, very agressive!
But risk is there. 
Which vendor is this?
I want to make a trial if possible!
Got me really curious! 
Samsung.
But I think everyone may have it.
Ericsson surely have.
A2 used for redirection as well, parameter threshold 4.
A2 used for inter freq / inter rat measurements trigger.
For VoLTEe we can control redirection using switch actredirectqci1 by keeping true or false.
Do we need any specific feature other than SErvice based redirection or Ho?
How we can apply SPID here for smooth transition?