0% found this document useful (0 votes)
79 views20 pages

HOSR Problem Snap Shot: Jabodetabek HO Attmp Distribution

1. The document analyzes handover success rates (HOSR) across different regions in a mobile network. HOSR is lowest for handovers due to uplink quality and highest for handovers to better cells. 2. It describes the key steps in the handover process, including measurement collection, reporting, preprocessing, threshold comparison, and target cell selection. 3. Several parameters and timers influence handover signaling and execution, and their settings impact whether handovers are fast and reliable or cause unnecessary attempts.

Uploaded by

Laurie Richard
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLS, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
79 views20 pages

HOSR Problem Snap Shot: Jabodetabek HO Attmp Distribution

1. The document analyzes handover success rates (HOSR) across different regions in a mobile network. HOSR is lowest for handovers due to uplink quality and highest for handovers to better cells. 2. It describes the key steps in the handover process, including measurement collection, reporting, preprocessing, threshold comparison, and target cell selection. 3. Several parameters and timers influence handover signaling and execution, and their settings impact whether handovers are fast and reliable or cause unnecessary attempts.

Uploaded by

Laurie Richard
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLS, PDF, TXT or read online on Scribd
You are on page 1/ 20

HOSR Problem Snap Shot

1 week average measurements (W16)


HOSR HO Fail HOSR_UQHOSR_DQ HOSR_UL HOSR_DL HOSR_dist HOSR_BC HOSR_DR
Jabodetabek 87.01% 94206 65.66% 80.14% 81.65% 83.54% 0.00% 95.34% 90.95%
West Java 79.53% 45250 64.68% 75.38% 76.00% 76.73% 32.50% 91.86% 94.56%
Central Java 82.47% 48664 63.65% 79.14% 79.63% 79.86% 60.78% 90.89% 83.42%
East Java 82.83% 49904 67.08% 79.98% 79.24% 78.64% 60.00% 92.88% 76.00%
HOSR start from BWHCI
Cluster
1.21%
4.84%
40.45%
17.77%
0.0001%
35.69%
0.04%
Jabodetabek HO Attmp Distribution
% HO_Attmp_UQ % HO_Attmp_DQ % HO_Attmp_UL % HO_Attmp_DL
% HO_Attmp_dist % HO_Attmp_BC % HO_Attmp_DR
1.92%
3.65%
53.21%
15.97%
0.0022%
25.20%
0.05%
West Java HO Attmp Distribution
% HO_Attmp_UQ % HO_Attmp_DQ % HO_Attmp_UL % HO_Attmp_DL
% HO_Attmp_dist % HO_Attmp_BC % HO_Attmp_DR
2.43%
2.17%
43.15%
21.73%
0.0019%
30.45%
0.07%
Central Java HO Attmp Distribution
65.66%
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
HOSR_UQ
64.68%
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
HOSR_UQ
63.65%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Uplink Level, Better Cell , and Downlink Level become the most highest reasons for handover attempts in Hutch network. For
has the highest success rate performance, while handover due to uplink quality has the lowest success rate. HOSR for uplink and downlink level are also not quite good.
Bad HOSR performance in
Hutch network
0.0019%
% HO_Attmp_UQ % HO_Attmp_DQ % HO_Attmp_UL % HO_Attmp_DL
% HO_Attmp_dist % HO_Attmp_BC % HO_Attmp_DR
1.68%
2.65%
47.70%
18.06%
0.0003%
29.77%
0.14%
East Java HO Attmp Distribution
% HO_Attmp_UQ % HO_Attmp_DQ % HO_Attmp_UL % HO_Attmp_DL
% HO_Attmp_dist % HO_Attmp_BC % HO_Attmp_DR
0.00%
10.00%
20.00%
30.00%
40.00%
HOSR_UQ
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
HOSR_UQ
% HO_Attmp_UQ % HO_Attmp_DQ % HO_Attmp_UL % HO_Attmp_DL % HO_Attmp_dist % HO_Attmp_BC % HO_Attmp_DR
1.21% 4.84% 40.45% 17.77% 0.0001% 35.69% 0.04%
1.92% 3.65% 53.21% 15.97% 0.0022% 25.20% 0.05%
2.43% 2.17% 43.15% 21.73% 0.0019% 30.45% 0.07%
1.68% 2.65% 47.70% 18.06% 0.0003% 29.77% 0.14%
HOSR start from BWHCI
65.66%
80.14%
81.65%
83.54%
0.00%
95.34%
90.95%
HOSR_UQ HOSR_DQ HOSR_UL HOSR_DL HOSR_dist HOSR_BC HOSR_DR
Jabodetabek HOSR Distribution
64.68%
75.38% 76.00% 76.73%
32.50%
91.86%
94.56%
HOSR_UQ HOSR_DQ HOSR_UL HOSR_DL HOSR_dist HOSR_BC HOSR_DR
West Java HOSR Distribution
63.65%
79.14% 79.63% 79.86%
60.78%
90.89%
83.42%
Central Java HOSR Distribution
in Hutch network. For Success Rate Distribution, handover due to better cell
has the lowest success rate. HOSR for uplink and downlink level are also not quite good.
HOSR_UQ HOSR_DQ HOSR_UL HOSR_DL HOSR_dist HOSR_BC HOSR_DR
67.08%
79.98% 79.24% 78.64%
60.00%
92.88%
76.00%
HOSR_UQ HOSR_DQ HOSR_UL HOSR_DL HOSR_dist HOSR_BC HOSR_DR
East Java HOSR Distribution
Handover Process
What may cause HO Fail ?
HOSR = SINTHINT [al]
ATINHIRC [all]
1. Target suffer TCH blocking ; no TCH available in target cell
2. Loss of HO ACCESS message until T3124 expiry ; loss may be caused by radio interface problems
3. Loss of HO CMD message until T8 expiry ; loss may be caused by radio interface problems or unnecessary T8 & TTRAU setting.
Rule : T8 < TTRAU
4. No SABM after PHYS INFO message transmission until T3105 expiry for NY1+1 times ; loss of
message may be caused by radio interface problems or unnecessary T3105 and NY1 setting.
Rule : (NY1+1) * T3105]<<T8<TTRAU
What may cause HO Fail ?
HOSR = SINTHINT [al]
ATINHIRC [all]
1. Target suffer TCH blocking ; no TCH available in target cell
2. Loss of HO ACCESS message until T3124 expiry ; loss may be caused by radio interface problems
3. Loss of HO CMD message until T8 expiry ; loss may be caused by radio interface problems or unnecessary T8 & TTRAU setting.
What steps are occurred within Handover process ?
Step 1 : Measurement values for HO
Values which are measured and calculated within each SACCH multiframe (480ms) :
Step 2 : Measurement reporting & Neighbour Cell Book Keeping
Step 3 : Measurement Pre-Processing
Step 1
Measurements :
DL/UL lev
DL/UL qual
TA
Traffic load
Step 2
Measurement reporting
&
Neighbour Cell Book Keeping
Step 3
Measurement
Pre-Processing
Step 6
Target Cell List Evaluation
Intra BSC
Inter BSC
Step 7
Selection of New Channel
Step 8
Handover Execution
- Rx Qual UL&DL
- Rx Lev UL&DL
- Rx Lev_NCELL (n)
- MS_BS_DIST
-
BCCH_FREQ_NCELL
- MS need BCCH_ARFCN_NCELL(n)
- The MS reports to the BTS the level measured on a certain ARFCN(n) together with the relative BCCH_FREQ_NCELL_(n)
- The MS decodes the Base Station Identity Code BSIC(n)
- BCCH_FREQ_NCELL(n) and BSIC(n) CI_NCELL(n)uniquely
- Measurements report by MS reporting of the strongest cells with known and allowed BSIC maximum: 6 cells on each SACCH multiframe
- The measured (and reported) data per SACCH multiframe are preprocessed within the BTSE using a gliding average window
- The size of the window can be set separately for RXQUAL, RXLEV, DIST and PBGT. The measured RXLEV_FULL/SUB or RXQUAL_FULL/SUB
gliding window with a multiplicity (weight) given by the parameter W_LEV_HO or W_QUAL_HO, respectively

















- Parameter related for measurement pre-processing : HOAVQUAL, HOAVELEV, HOAVDIST, HOAVPWRB
- The adjustment of the averaging window size mainly depends on the change of the radio propagation conditions
- The setting of the averaging window size has to be a compromise between a fast decision and a reliable decision.
- Therefore it is recommended to use a larger window size for the optional handover (better cell) to do not cause a lot of unnecessary handovers and a smaller window size for the
mandatory handover causes (quality, level, distance) to be able to react quickly on a sudden decrease of link quality



Step 4 : Threshold Comparison & Network Detection
Step 5 : Target Cell Lists Generation
Step 6 : Target Cell Lists Evaluation
- Parameter related for HO decision : HOLTHQUDL/UL, HOLTHLVDL/UL, HOTMSRM (for DIST HO), HOTDLINT / HOTULINT (for INTRACELL HO)
mandatory HO), HOM (for BC HO)
- When in condition more than one HO causes are fulfilled, therefore HO will be performed with rank of evaluation : 1. Quality Intercell HO, 2. Level HO, 3. Distance HO, 4. BC HO, 5. Quality Intracell
HO,
6. Traffic HO, 7. Forced HO
- Generally, HO causes can be classified into 2 categories : 1. radio criteria (imperative) : quality, level, distance, BC ; and 2. network criteria : congestion
- In specific cases, (power budget) back HO should be avoided :
1. If an imperative handover occurs, then a back handover should be avoided if the reason for this second handover is the power budget
Relative parameter : NOBAKHO, TINHBAKHO
2. In case of a forced handover (e.g. after a directed retry), a back handover should be avoided only if the reason for this second handover is the power budget
Relative parameter : NOBAKHO, TIMERFHO
- After consecutive handover failures further handovers towards the same cell should be avoided. For this reason the BSC informs the BTS after a number of subsequent HO failures to exclude
the cell from the target cell list.
Relative parameter : NOFREPHO, TINHFAIHO, MAXFAILHO
Rule should be regarded : TINHFAIHO (BTS timer) > THORQST (BTS timer) > T7 (BSC timer)
- In a situation, where for a connection in an interfered cell consecutive intracell handovers are performed without improving the quality of the connection (especially if frequency hopping is active),
further intracell handovers should be limited.
Relative parameter : ELIMITCH, TINOIERCHO, MAIRACHO
Rule should be regarded : TINOIERCHO (BTS timer) > THORQST (BTS timer) > T7 (BSC timer)
- The maximum number of cells to be included in the target cell list is given by the parameter N_CELL (Parameter NCELL in object HAND, Range: 0 ... 15)
- Neighbor cells which are included in the target cell list have to fulfill the following conditions :
1. for quality, level, and distance intercell HO (minimum condition) :
RXLEV_NCELL(n) > RXLEV_MIN(n) + MAX(0, MS_TXPWR_MAX(n) - P)
2. for power budget HO :
RXLEV_NCELL(n) > RXLEV_MIN(n) + MAX(0, MS_TXPWR_MAX(n) - P) & PBGT(n) - HO_MARGIN(n) > 0
- The ranking of the target cells in the target cell list is performed on the basis of the power budget minus handover margin value ("better cell value") :
PRIO_NCELL(n) = PBGT(n) - HO_MARGIN(n)

- Intra BSC HO
If the first cell within the target cell list is within its BSS area, the BSC
selects a channel at the corresponding BTS. If no channel is available at that
BTS, the next cell within the target cell is tried
- Inter BSC HO
If the first target cell (or the ones tried in further steps) does not belong to the
own BSS area, a Handover Required message is sent to the MSC. This
message contains a reduced target cell list (without the cells tried internally).
Handover Signalling and Timer ?
Parameter & Timer for HO Signalling
Object DB Name Unit
BTS T3105 MS10=10 ms
BTS T3124
HAND THORQST 2*SACCH multiframe
BSC T7 HLFSEC = 0.5 sec
MS100 = 100 msec
SEC5 = 5 sec
BSC T8 HLFSEC = 0.5 sec
SEC5 = 5 sec
BTS NY1 -
Directed Retry HO ?
SDCCH to SDCCH HO ?
- Directed retry is the transition (handover) from a SDCCH in one cell to a TCH in another cell during call setup because of unavailability of an empty TCH within the first cell
- Relative Parameter : ENFORCHO (Object : BSC; to enable/disable the sending of Forced HO Request messages for running SDCCH co
- SDCCH-SDCCH HO provides the handover between two SDCCH channels due to radio conditions. If a MS is using a SDCCH, the BTS che
TCH allocation. If the measurements indicate that a SDCCH HO is required, an Intercell/Intracell HO Condition Indication message ( incl. a target cell list ) is sent from BTS to BSC
- SDCCH-SDCCH handover is only triggered by the handover causes classified as radio criteria:
1. Quality Intercell
2. Quality Intracell
3. Level Intercell
4. Distance Intercell
5. Power Budget Intercell
- Flowchart SDCCH to SDCCH HO :










Rule : (NY1+1)* T3105 << T8 < TTRAU
Traffic HO ?
Fast Uplink HO ?
Level HO Margin ?










- Relative Parameter : IERCHOSDCCH (Object HAND ; Intra-BSC intercell SDCCH-to-SDCCH handover), IRACHOSDCCH (Object HAND ; Intra
SDCCH
HO), EISDCCHHO (Object BSC ; Inter-BSC SDCCH HO)
- Traffic HO allows to move a connection for traffic reasons from one cell to an adjacent cell, even in case that a HO due to radio conditions is not required
- Traffic HO increase network capacity
- Traffic HO procedures :
1. The BSC starts a timer (TRFCT)
2. When this timer (TRFCT) expires, if the feature is enabled in the cell the BSC calculates the traffic level (in percentage)
3. The cell traffic level value is compared with the predefined percentage of high traffic level threshold (TRFHITH)
4. If the calculated value is higher or equal to TRFHITH parameter the cell is regarded as a high traffic level cell, otherwise the cell is regarded as a low traffic level cell
5. If the cell is classified as a high traffic level cell, the BSC sends a message to the BTS to enable the handover for traffic reason
6. If the cell is classified as a low traffic level cell, the BSC sends a message to the BTS to disable the handover for traffic reason
7. The neighboring cells included in the candidate list are ranked. When the candidate list is ready, the BTS sends to the BSC an inter cell handover condition message that
includes the preferred list of candidates and the cause traffic for the handover request
8. When the BSC receives the request, the list is analyzed and the cells with traffic higher than TRAFFICIN are discarded
9. In the target cell a back handover for traffic reasons and power budget is inhibited for the TPENTR duration time
- Relative Parameter : BHOFOT, TRFHOM, TRFHOE, TRFHITH, TRFLTH, TRFMS, TRFMMA


- Fast Uplink Handover is used as high-speed handover to prevent from rapid uplink level loss and is applied to save connections in special places with drop call problems
- When the power level of a mobile decreases rapidly, a handover can be performed to a predefined cell ( cell that have attribute fastULHoCell set to TRUE) to save the connection
- The Fast Uplink Handover is performed after power control has reached its max allowed values
- The handover detection is performed for the uplink path and is based on a comparison of the uplink measurement receive level with a threshold, this HO is detected when the uplink
measurement receive level is lower than this threshold
- As candidate for target cell lists, minimum condition applied for neighbor cells are :
RXLEV_NCELL(n) > RXLEVMIN(n) + max(0; Pa) + FULRXLVMOFF

- "Level Handover Margin" Parameter provides a new target cell criteria that takes the different requirements for level handover and quality handover into consideration
- The creation of the target cell list is based on these criteria :
RXLEV_NCELL(n) > RXLEV_MIN(n) + max (0,MS_TXPWR_MAX-P) AND PBGT(n) > LEVHOM(n)
- Relative parameter : ELEVHOM (Object HAND , flag to enable/disable Level HO Margin), LEVHOM (Object ADJC)

Step 4
Threshold Comparison
&
Network Detection
Step 5
Target Cell List
Generation
Target Cell List Evaluation
me
The size of the window can be set separately for RXQUAL, RXLEV, DIST and PBGT. The measured RXLEV_FULL/SUB or RXQUAL_FULL/SUB values are put into the
essary handovers and a smaller window size for the
Parameter related for HO decision : HOLTHQUDL/UL, HOLTHLVDL/UL, HOTMSRM (for DIST HO), HOTDLINT / HOTULINT (for INTRACELL HO), RXLEVMIN (for BC HO, and target cell lists for
ntercell HO, 2. Level HO, 3. Distance HO, 4. BC HO, 5. Quality Intracell
2. network criteria : congestion
the power budget
r this second handover is the power budget
the BTS after a number of subsequent HO failures to exclude
he quality of the connection (especially if frequency hopping is active),
HAND, Range: 0 ... 15)
lue ("better cell value") :
If the first cell within the target cell list is within its BSS area, the BSC
selects a channel at the corresponding BTS. If no channel is available at that
BTS, the next cell within the target cell is tried
If the first target cell (or the ones tried in further steps) does not belong to the
own BSS area, a Handover Required message is sent to the MSC. This
message contains a reduced target cell list (without the cells tried internally).
Remarks
period for repetition of PHYSICAL INFORMATION
Start: sending of PHYSICAL INFORMATION by BTS
Stop: reception of correctly decoded signaling or TCH frame on new channel
from MS at BTS
Expiry: repetition of PHYSICAL INFORMATION until NY1+1 times
Not adjustable by DB parameter
min. time for HO COND. IND. messages for the same connection
Start: sending of HANDOVER CONDITION INDICATION by BTS
Stop: HANDOVER COMMAND received , lost communication with MS, call cleared
Expiry: repetition of HANDOVER CONDITION INDICATION
Waiting time for a HANDOVER COMMAND from the MSC
Start: sending of HANDOVER REQUIRED by BSC
Stop: HANDOVER COMMAND received, lost communication with MS, call cleared
determines the time to receive the HANDOVER COMPLETE message
purpose: keep the old channels available for a sufficient time in order to allow
the MS to return to the old channel return to it if the handover is not successful and
to release the old channel if the MS is lost.
start: transmission of a HANDOVER COMMAND from the BSC to the MS
stop: a) intra-BSC handover: receipt of a HANDOVER COMPLETE or a HANDOVER
FAILURE from the MS
b) inter-BSC handover: receipt of a CLEAR COMMAND from the MSC or
HANDOVER FAILURE from the MS
maximum number of repetitions of the physical information by the BTS
vailability of an empty TCH within the first cell
Relative Parameter : ENFORCHO (Object : BSC; to enable/disable the sending of Forced HO Request messages for running SDCCH connections; to enable/disable DR HO)
SDCCH HO provides the handover between two SDCCH channels due to radio conditions. If a MS is using a SDCCH, the BTS checks the radio conditions as in the case of
ge ( incl. a target cell list ) is sent from BTS to BSC
SDCCH handover), IRACHOSDCCH (Object HAND ; Intracell HO due to quality for SDCCH to
dio conditions is not required
herwise the cell is regarded as a low traffic level cell

he BSC an inter cell handover condition message that
in special places with drop call problems
e fastULHoCell set to TRUE) to save the connection
ith a threshold, this HO is detected when the uplink
and quality handover into consideration
Jabodetabek Cluster => proposed for setting trial
Current Audit Proposed
HOAVQUAL HAND 6-2 6-2
HOAVELEV HAND 8-2 6-2
HOAVDIST HAND 8 8
HOAVPWRB HAND 8 8
HOLTHQUDL HAND 5 5
HOLTHQUUL HAND 5 5
HOLTHLVDL HAND 12 10
HOLTHLVUL HAND 6 6
HOTMSRM HAND 34 ; 35 35
HOTDLINT HAND 35 35
HOTULINT HAND 31 ; 35 31
RXLEVMIN ADJC 12 14
HOM ADJC 67 67
NOBAKHO HAND TRUE TRUE
TINHBAKHO ADJC 30 30
TIMERFHO ADJC 12 12
NOFREPHO HAND TRUE TRUE
TINHFAIHO ADJC 5 8
MAXFAILHO HAND 2 2
THORQST HAND 8 8
T7 BSC HLFSEC-7 HLFSEC-7
ELIMITCH HAND TRUE TRUE
TINOIERCHO HAND 10 ; 60 10
MAIRACHO HAND 2 ; 4 4
NCELL HAND 6 ; 8 6
T3105 BTS MS10-10 MS10-10
T8 BSC HLFSEC-12 HLFSEC-14
NY1 BTS 50 50
ENFORCHO BSC ENABLED ENABLED
IERCHOSDCCHHAND TRUE ; FALSETRUE
IRACHOSDCCHHAND FALSE FALSE
EISDCCHHO BSC DISABLED DISABLED
ELEVHOM HAND FALSE TRUE
LEVHOM ADJC 67 ; 69 67
Value
DB Parameter Object
Smaller window size for mandatory HO to
be able to react quickly on sudden
decrease of link quality => advantage :
prevent TCH drop & loss connection
Increase RXLEVMIN for guarantee target cell
lists have good Rx Level;
also must follow rule : RXLEVMIN >
HOLTHLVDL to prevent ping-pong HO
Must follow the rule : TINHFAIHO > THORQST(HAND) > T7 (SET BSC [TIMER])
Give more opportunities to the cell to perform certain number of consecutive successful
intracell handovers due to quality that are permitted in the same BTS for a single
connection => advantages : keep good quality within single connection
Give sufficient time for BTS to receive HANDOVER COMPLETE message ; give
more time for BTS to keep the old channels available for a sufficient time in order to
allow the MS to return to the old channel return to it if the handover is not successful
and to release the old channel if the MS is lost
This setting provide new target cell criteria for HO due to level, that is PBGT(n) >
LEVHOM(n). This criteria / margin will more guarantee target in better level & quality
condition, which will increase successful probability for HO performing. When we detect
most HO attempts due to level are occured in network, then we should activate this feature.
Must follow the rule : TINHFAIHO > THORQST(HAND) > T7 (SET BSC [TIMER])
Give more opportunities to the cell to perform certain number of consecutive successful
intracell handovers due to quality that are permitted in the same BTS for a single
Give sufficient time for BTS to receive HANDOVER COMPLETE message ; give
more time for BTS to keep the old channels available for a sufficient time in order to
allow the MS to return to the old channel return to it if the handover is not successful
This setting provide new target cell criteria for HO due to level, that is PBGT(n) >
LEVHOM(n). This criteria / margin will more guarantee target in better level & quality
condition, which will increase successful probability for HO performing. When we detect
most HO attempts due to level are occured in network, then we should activate this feature.

You might also like