What is the best strategy for shorter CSFB CST?

Hello All.

What is the best strategy for shorter CSFB CST?

LTE Huawei

What is the best strategy for shorter CSFB CST

1 Like

To make unwanted nbr and frequency delete, make colocated nbr high if this strategy or frequency base priority, blind and redirection always on and in 3G or 2G, reselection parameter and Frequency defined to LTE.

Fast reselection.

In Nokia we use frequency based priority with equal priority (using IMLB). Very good CST during CSFB.

But Huawei with nbr defintion the CST is much longer even when setting equal priorities.

1 Like

Make colocated sector priority high and delete unwanted nbr and frequency.

1 Like

We did collocated cells high priority, but no impact on CST.

Meanwhile CSFB SR degraded.

Flash CSFB.

1 Like

Implement very early assignment is lower layer is GSM, else implement few procedural improvments at UMTS side.

Rest for LTE side do following things:

  1. Add all necessary nbrs and delete unwanted nbrs.
  2. Make collocated site higher priority and rest lower priority.
  3. Make sure RIM switch enabled along with Flash CSFB.
  4. If RIM is controlled by CORE end, check DSP GERANRIMINFO or UTRANRIMINFO command and check the RIM associations. If there is any missing link, add it.
  5. If RIM is controlled by eNodeB, then check the relevant switch and implement whole feature and also check eNACC switch at GSM end as well.
  6. Chexk the CSFB priority with respect to the lower technology.
  7. Also check the delay is at which end, CORE end or Radio end, split the analysis and involve CORE end if needed.

Most important thing, check CSFB should be blind base set, it shouldn’t be measurement based.

4 Likes

Can you share the configuration of very early assignment

Try to apply flash CSFB or (RIM)
4nbrs or 8nbrs configured will sent in RRC Rlease with their sibs
The ue don’t need to read sibs for 3g nbrs again as it sent in rrc release when applying Flash CSFB
It reduces call setup time

Majority of the pointers have been mentioned ,
Rim activation, neighbour optimization, blind ho to utran strategy.

However having a similar challenge were all the above are activated but cst >12 s for some calls. Any ideas how to improve ?