Monday, 14 September 2015

X2 Handover

Problem Description
A LTE mobile network operator in country H has found that the number of X2-based outgoing handover attempt as indicated by the “L.HHO.X2.IntraFreq.
PrepAttOut” counter in the E-UTRAN (eNodeB) and the number of X2-based handover attempt as indicated
by the “intra-MME X2-based handover request times” counter in the EPC are always not consistent with each other after the deployment of the X2 interface.
As a result, the X2-related KPIs in the performance report generated by the E-UTRAN and that generated by the EPC are not consistent as well.
eNodeB Verion:  Version: eNodeB V100R005C01SPC200
MME Version: USN9810 PS9.1
Handling Process


Referring to the above signalling flow, the E-UTRAN increments the counter “L.HHO.X2.IntraFreq.PrepAttOut” that is used to record the number of X2-based
handover at measurement point A where the source eNodeB has decided to perform the X2-based handover. For the MME,
it increments the “intra-MME X2-based handover request times” at measurement point D whenever the MME receives the “Path Switch Request” message from the target eNodeB. 
However, the E-UTRAN will trigger a “target” eNodeB to initiate the transmission of the “Path Switch Request” message also in the scenarios in which in (i) an
inter-eNodeB RRC connection re-establishment due to handover failure and (ii) an inter-eNodeB RRC connection re-establishment
due to other (i.e. radio link failure) (this is a proprietary implementation of Huawei) in addition to the a normal X2-based inter-eNodeB handover.
As a consequence, the “L.HHO.X2.IntraFreq.PrepAttOut” counter in the eNodeB and the “intra-MME X2-based handover request times” counter at the MME may show a different value.
Root Cause
N/A
Solution
1.Compare the statistics of the “L.HHO.X2.IntraFreq.PrepAttOut” counter in the E-UTRAN and the statistics of the “intra-MME X2-based handover request times” in the EPC to
see if the value of the “L.HHO.X2.IntraFreq.PrepAttOut” counter is always greater that of the “intra-MME X2-based handover request times” counter or vice versa.
2.Analyze the mechanism of the X2-based handover, the corresponding signalling flow,
the measurement point of the “L.HHO.X2.IntraFreq.PrepAttOut” counter and the scenario under which the counter is incremented in the E-UTRAN.
3.Analyze the mechanism of the path switch request process, the corresponding signalling flow,
the measurement point of the “intra-MME X2-based handover request times” counter and the scenarios under wich the counter in incremented in the EPC.
Suggestions and Summary
According to the above problem analysis, the “intra-MME X2-based handover request times” counter implemented in the EPC does not really reflect the number of X2-based handover attempt only,
but also the number RRC connection re-establishment due to handover failure or other failure (i.e. radio link failure). For this reason,
the “L.HHO.X2.IntraFreq.PrepAttOut” counter in the E-UTRAN (eNodeB) should always be referred to for the exact
number of X2-based handover performed in the E-URAN or for the calculation of the X2-based handover KPIs.

Budi Prasetyo

About Budi Prasetyo

All About LTE

Subscribe to this Blog via Email :