Question

RRC reestablishment happens due to

-> Radio Link Failure.
-> Handover Failure.
-> Reconfiguration Failure.
-> Integrity Failure.

1Q)Does UE goes for RACH of all these cases or only for RLF.

2Q)Actually what makes UE get RLF.

Handover / (Failure) case:

3Q)Is there any possibility during handover UE might fail to do RACH with target.
Actually during HO ,target provides rach dedicated config.So if RACH fails ,then will it send reestablishment to target or source or other.
What was the behaviour ?

4Q)Will ue has to do reestablishment with only source,any possibility of sending to other cell(target or other).

5Q)Are there any messages between eNB's to forward during reestablishment rejection cases (which version>10.5/message in 36331 specifies).

My doubt :

1) Except RLF ,for other cases still UE was in connected mode only,so we can derive crnti & pcell from ue context &
forward the reestablishment request.

2) If we need to go for RACH of all the cases ,to which eNB will UE sends RACH & reestablishment request.
How it derives crnti & pcell to include it onto reestablishment request.

3) what if during HO , TS1RELOCOverall_timer_id & TX2RELOCoverall_id expires before Source (/any cell) receives reestablishment request.

3) What is the significance of mobility during these situations.

P.S :- Please correct me ,if any of my question or assumption was wrong.

Answer 1

In any case of re-establishment, ue will perform contention based to enb.

Rlf is detected by lower layer, N310 timer is running to detect out of sync at lower layer. During this timer if lower layer is not able to decode any data in DL then if inform to l3 as rlf.

HO failure - if ue is not able to do successful Rach at target after getting mobility parameters before expiry of t304 timer then ho failure will be considered by ue.

Rach failure during ho - it may be possible that after getting mobility parameters, ue is not in target cell coverage so Rach is not getting successful.

Re-establishment can be performed to any cell. When ue perform re-establishment then ue select the cell .so whichever cell is getting selected with that ue will perform re-establishment.

During re-establishment, ue gets new crnti and if re-establishment successful then ue continue with new crnti.

If ue performs re-establishment after t relocoverall timer expiry then re-establishment will not be successful as enb will release the ue context after t relocoverall timer expiry
Hope this will help.

Comments

@Veer Pal Singh Yadav Thanks for the reply. Kindly clarify a little more of below.

So, during reestablishment UE performs rach and gets new crnti and sends
request upon to any cell(source or target or maybe other) which it has chosen
and performed rach. At this, it may send the reestablishment to a source enodeb with new crnti,  then how source enodeb identifies UE in its context with this new crnti because source enodeb has UE context with old generated crnti before the reestablishment. How UE gets validated, is it with short mac id (which UE fills from its context).
Even if UE sent reestablishment to target following rach,  it won't be having any context related to the UE, then target may trigger RLF indication (is it correct from 36.423) towards neighboring cell by including crnti optional short mac id.
How will the neighboring cell or source acts, and how UE gets successful reestablishment parameters.

---

Enb validates ue context by old crnti, PCI and shortmac-i which ue sends in rrc re-establishment request.
If ue sends rrc re-establishment request to target (where ue context created during ho preparation) then enb will validate the ue context and will send rrc re-establishment to ue .if target enb is not able to validate the ue or does not have ue context then it will send rrc re-establishment reject to ue.
If ue sends rrc re-establishment request to target/any other cell except source cell and if these cells have source cell as neighbor then this cell sends rlf indication to source cell