I have done some downtilting as well, pushed more traffic to 3G.
I have two U2100 layers + 1 900.
But your RRC signalling attempts have been increased, but you already have done downtilt.
So what is erab success rate now?
RRC SR have been stable and very good trend.
Only E-RAB is declining, with main failures being No Radio Resources (namely low RRC Connected Users (FDD) license).
I think I know what you’re referring to, why RRC is not impacted, right?
Due to increase in attempts of low inactivity timer, signalling has increased.
And RRC setup complete is connected state.
Srb1 start so your RRC rate is good.
I got you. Thanks for the explaination!
This is the trend of RRC.
RRC Setup Sucess Rate (%)
And E-RAB for the same period.
E-RAB Setup Sucess Rate (%)
Ok Cell C and Cell A eRAB KPI is reversed of 14 and 15.
Even Cell B is degraded, as compared to 14.
Yep, may that’s returned us to my original question: since the failures are mostly caused by lack of RRC User License (actual values exceeds the allocated ones) and since the license is distributed on eNodeB level, not cell level (or that’s how I understands), why certain cells get impacted rather than others?
Is it related to scheduling for example?
Btw, these are measurements of L.E-RAB.FailEst.NoRadioRes.RRCUserLic for the three cells
Hourly, for today.
So it reversed.
It’s lot of failures after 13 h.
License is the thing, the solution.
Or new site as you told will be coming.
So any TNL or packet loss issue apart from that.
And about erab drops, is it fine?
There are some TNL failures due to packet loss (fiber link issues) but they should get resolved within some days.
TNL, if improved, will help your KPI, definitely.
Check ping test.
If value is high, then it’s their problem.
And license too is killing.
Actually, number of failure due to license for this site contribute about 80%.
Rest are mostly TNL.
Yes, but TNL, if solved, will help somewhat more.
Sorry, my mistake.
URA PCH would be related to 3G.
But maybe some of below itens could help also:
-
Decrease RxLevMIn in order to decrease the number of users, maybe neighbor sites could handle this traffic. It can help if you have some tilt optimization issues/limitations;
-
In case of 3G as a traffic overflow solution, optimize threshX_Lo and S_nonIntraSearch as well as if other 4G frequency cells have more resoucers available*
-
Optimize cellReselectionPriority*
*followed by Handover parameters optimization maybe using A4 event.
Thanks @Pirm, I did reduce the 4G cell radius using E-tilt and worked on pushing more some of the traffic to 3G, will consider A4 event as well.
And thanks all the others too, for example about UE inactivity reduction.
Thanks for the contribution!
1 Like
Hello dears, Update:
As menitoned, I had a degradation in E-RAB Success Rate due to L.E-RAB.NoRadioRes.
After some actions (E-RAB improved from avg 40% to 70-80%), I decreased UE Inactivity timer from 5s to 2s.
Good results have been achieved as shown below:
Using MOD QCIPARA, I decreased all QCIs to 2 s except for QCI 69, which I made it only 3s
I decided to make QCI 69 to be 3 s.
Other QCIs values are 2/3 of QCI 69.
Thanks guys for the help!
But user experience is affected. How about no of counter erab setup failure?
Any idea how to check amount of signaling in LTE?
Is RRC Requests a reliable measure?
Yes, it is.
But it’s ok as long as you don’t get RRC failures.