Optimization Issues Ericsson UMTS

8 Optimization issues

8.1 Pilot channel failure - high downlink interference

If inter-RAT mobility function is activated, i.e. FDD GSM handover support =
1, and works properly, this problem should not happen.
8.1.1 Symptoms
rom the drive test, followin! symptoms will be observed by usin! T"#$%
- Received "c&'o of the pilot channel is less than (1)d* and
- Received R$+, of the pilot channel is hi!h enou!h to maintain the
connection, e.!. - -1..d*m and
- /0 R$$I is very hi!h and
- The connection finally drops.
,lease read appendi1 +.
8.1.2 Reason 1 no dominant cell
There are many overlappin! cells at the problem area. The received si!nal
stren!ths of these pilots are almost the same, i.e. hi!h cch value. ,lease read
e2uation 1. for cch e2uation.
Figure 15. Example of many cells overlapping, the observation is from the
pilot scanning.
Solution
The most direct and effective way to solve this problem is to increase the pilot
channel power Primary CPICH power of the desired cell.
Figure 16. Result of increasing the pilot power of the desired cell e.g. pilot .
The drawbacks of this solution are%
- /ue to uneven pilot power settin!, some 3"s mi!ht no lon!er be connected
to the 4closest5 cell with respect to the pathloss. Then they transmit with hi!h
3" powers. As a result, the uplink interference level of the carrier is
conse2uently increased. It means uplink is not optimi6ed and this
phenomenon is called as uplink near-far problem. It is recommended that
(Reporting Range 1a: threshold or addition window&7 _
difference of the pilot power settin!s of two nei!hborin! cells _
Reporting Range 1a: threshold or addition window&7
It should be noted that the unit for the Reporting Range 1a: threshold or
addition window is ..8d*.
or e1ample, if the Reporting Range 1a: threshold or addition window is ),
i.e. 9d*, the mar!in for pilot power modification is e2ual to 4-9d* _ difference of
the pilot power settin!s of two nei!hborin! cells _ 9d*. ,lease read appendi1 *
for the restrictions to modify the pilot channel power.
- In case the pilot power of a cell is increased, the power of common channels
in that cell will simultaneously increase because their parameter settin!s are
relative to the pilot power value. At the same time, the re2uired power for the
downlink /,+:s in that cell also increase, please see e2uation );. inally,
the load of the cell becomes hi!h and then cell blockin! may happen.
- The downlink interference level of the carrier will be hi!her.
- The cell with hi!her pilot power will absorb more 3"s from its ad<acent
cells. Then the load of the cell will be hi!her.
- ,ilot power chan!es may lead to uplink covera!e and pilot covera!e
imbalance problems, please read chapter =.8 for details.
8.1.! Reason 2 dominant interferer
An undesired cell with very hi!h si!nal stren!th is found in the problem area.
Figure 1!. Example of the overshooting problem from the undesired cell, the
observation is from the pilot scanning.
Solution 1
The simplest solution to overcome this problem is to include the overshootin!
cell into the nei!hborin! cell list. This means the interferer now becomes a useful
radio link.
Figure 1". Result of adding the overshooting cell into the neighboring cell list.
The drawbacks of the solution 1 are%
- It creates more unnecessary handovers and e1cessive numbers of 3"s are in
soft handover.
- If the overshootin! cell is physically far way to the problem area, the
handover se2uence mi!ht be messed up after includin! it into the nei!hborin!
cell list.
- :andover event detection in the 3" becomes very slow if the list of the
nei!hborin! cells is very lon!.
- Additional radio link will cause out of hardware resources in both interferer
and interfered cell.
- Additional radio link will cause cell blockin! in both interferer and interfered
cell.
Solution !
An alternative solution is to chan!e the antenna confi!uration of the overshootin!
cell, e.!. tiltin! down the antenna, re-directin! the antenna orientation, reducin!
the antenna hei!ht.
>ith this solution, 30&/0 covera!e imbalance problem will not occur in the
interferer because both 30&/0 pathloss is modified simultaneously. #oreover,
the interferer probably will cover fewer 3"s, and transmit a lower total downlink
power. This means that its downlink interference contribution mi!ht be further
decreased, see i!ure 1;.
Figure 1#. $ownlin% interference is reduced by tilting down the antenna at the
interfering site.
The drawbacks of the solution 7 are%
- Time and cost consumption.
- The desired covera!e of the interferer is modified. +overa!e hole mi!ht
occur. ?erification of the covera!e should be done a!ain.
- 'ei!hborin! cells of the interferer will cover a lar!er area and will thus
absorb additional 3"s. The risk of hi!h blockin! rate therefore increases in
these cells. #oreover, due to transmit hi!h T1 power, they mi!ht become
interferers if their covera!es are not well confined.
Solution "
The third possible solution is to decrease the pilot power Primary CPICH
power of the overshootin! cell.
After decreasin! the pilot power, the total downlink power for the common
channels of the interferer decreases. >hen the pilot power is reduced, the power
of all other common channel decreases simultaneously because their parameter
settin!s are relative to the pilot power value.
#oreover, the total /0 /,+: T1 power of the interferer will probably decrease
because the interferer will covers fewer 3"s.
Figure &'. $ownlin% interference is reduced by reducing the pilot power
- It is not a solution for lon! term. The pilot power reduction does not affect
the physical pathloss. Therefore, when the load of the interferin! cell
!radually increases, the interference issues will occur a!ain at the same area
as previously.
- This solution is not suitable for a capacity limited interferer. The total
transmission power of the downlink control channels that is saved by
reducin! the pilot power, is used up by new /,+:s.
- 0ike the problem in increasin! pilot power, the uplink is not optimi6ed due to
uneven pilot power settin!. It is recommended that
(Reporting Range 1a: threshold or addition window&7 _
difference of the pilot power settin!s of two nei!hborin! cells _
Reporting Range 1a: threshold or addition window&7
It should be noted that the unit for the Reporting Range 1a: threshold or
addition window is ..8d*.
or e1ample, if the Reporting Range 1a: threshold or addition window is ),
i.e. 9d*, the mar!in for pilot power modification is e2ual to 4-9d* _ difference of
the pilot power settin!s of two nei!hborin! cells _ 9d*. ,lease read appendi1 *
for the restrictions to modify the pilot channel power.
- Reducin! the pilot power, the downlink channel estimation in the 3" is
affected. This influences the downlink 2uality. In the end, the 3" mi!ht
re2uest more power from base stations.
- >hen the pilot power is reduced, the ma1imum allowed /0 /+: power
decreases simultaneously because this parameter settin! is relative to the pilot
power value. Then, outa!e of the downlink /,+: will be hi!her if the pilot
power is reduced too much. ,lease read appendi1 * for the restrictions to
modify the pilot channel power.
- The desired covera!e of the interferer is modified. +overa!e hole mi!ht
occur. ?erification of the covera!e should be done a!ain.
- 'ei!hborin! cells of the interferer will cover a lar!er area and will thus
absorb additional 3"s. The risk of hi!h blockin! rate therefore increases in
these cells. #oreover, due to transmit hi!h T1 power, they mi!ht become
interferers if their covera!e are not well confined.
8.1." Reason ! low #est ser$ing PPilot%P&ot
The received "c&'o of the best servin! pilot channel is very low @near or less than
(1)d*A even thou!h there is no other cell. It means the pilot power settin! is not
lar!e enou!h to fulfill e1istin! downlink load.
Figure &1. Example of too low best serving ((ilot)(*ot, the observation is from
the pilot scanning.
Solution 1
The best solution is to add a new site with 4!ood covera!e control5 at the
problematic area.
It is a cost- and time-consumin! solution because installation work is needed.
After havin! a new site, the optimi6ation en!ineers have to verify the covera!e
once a!ain to ensure that the covera!e of the new site is well under control.
4Bood covera!e control5 is e1tremely important to >+/#A system. The
probability of interference problems can be minimi6ed with a suitable de!ree of
overlappin!.
'ote% In T/#A or /#A systems, the problems due to poor covera!e control
@e1cessive overlapA can be hidden by fre2uency plannin!.
Solution !
Another possible solution is to shift the downlink load of the problematic cell to
its ad<acent cells.
Solution "
It is an issue due to hi!h total downlink T1 power.
Solution #
The direct but ineffective solution is to increase the pilot channel power Primary
CPICH power of the problematic cell. >ith hi!h pilot power, the common
channel powers and the re2uired power for the downlink /,+:s will be
increased. Therefore, the total transmission downlink power of the cell will
conse2uently increase. At the end, the ratio of the ,,ilot&,Tot does not increase
much. ,lease read chapter =.1.7 for the drawbacks after the pilot power is
increased.
8.2 Pilot channel failure - out of pilot co$erage
If inter-RAT mobility function is activated, i.e. FDD GSM handover support =
1, and works properly, this problem should not happen.
8.2.1 Symptoms
- Received "c&'o of the pilot channel is less than (1)d* and
- Received R$+, of the pilot channel is very low, e.!. C-1..d*m and
- /0 R$$I is very low and
Figure &&. Example of out+of+pilot coverage, the observation is from the pilot
scanning
8.2.2 Reason 1 low pilot channel power
+overa!e hole appears because the power of the pilot channel is set too low.
Solution 1
The best solution is to add a new site at the problematic area. The drawbacks of
addin! new site are same these for the solution 1 in chapter =.1.D.
Solution !
The another solution is to increase the pilot channel power Primary CPICH
power. The drawbacks are same as these for the solution in chapter =.1.7.
8.2.! Reason 2 incorrect power measurement due to &'(
>hen tower mounted amplifier is employed, the power reference points for
uplink and downlink are near at the antenna ports, please see i!ure 79.
or the operators with usin! T#A, they have to input followin! information into
the 3TRA' in order to obtain accurate 30 and /0 power measurements, e.!.
uplink feeder attenuation, downlink feeder attenuation, uplink T#A !ain,
downlink T#A insertion loss, etc. Etherwise, all power measurements in the
3TRA' will be messed up.
Figure &. R,- reference points definitions.
If the sum of the attenuations of all parameters for downlink antenna path is less
than the sum of the real values, the pilot transmission power at the R*$ node is
then under-estimated and the pilot covera!e will become smaller. or instance,
*able 5. .n example to illustrate the incorrect power transmission for pilot
A ,lanned pilot power at the
reference power point
7Dd*m
* Real T#A insertion loss -1.d*
+ Real downlink feeder loss -9d*
/ = A-*-+ ,ilot power which the R*$
should transmit
9Fd*m
" ,arameter for T#A insertion
loss
-8d*
,arameter for downlink feeder
loss
-1d*
B = A-"- Incorrect pilot power which
the R*$ transmits
9.d*m
: = BG*G+ Actual pilot power at the
1Fd*
I = A-: ,ilot power difference Fd*
Solution
The uni2ue solution is to correct these parameters as the real ones. :owever, it is
a time consumin! solution and difficult to !et the accurate feeder loss and T#A
insertion loss values.
Further inormation
The ma1imum allowed /0 code power Ma$ D% Power is a relative parameter to
the pilot power settin! Primary CPICH power. Therefore, the covera!e of the
/0 /,+: will be shrunk with the same ratio like what for the pilot covera!e due
to improper parameter settin!s for antenna system when T#A is employed.
,lease see i!ure 7D.
Figure &/. 0ariation of the $1 $(23 coverage due to improper parameter
settings for antenna system when *4. is employed.
8.! Pilot channel failure - inter-R(& mo#ility pro#lem
8.!.1 Symptoms
The inter-RAT handover and cell chan!e functions are activated, i.e& FDD GSM
handover support = 1. :owever, from the drive test, it is found that the received
"c&'o of the pilot channel is less than (1)d* and the connection finally drops.
8.!.2 Reason 1 improper inter-R(& mo#ility )&R(* parameter
settings
- The 'sed re(uen)y threshold !d is less than or too close to (1)d*. As a
result, the 3" does not send event 7d measurement report to the 3TRA' to
re2uest B$# measurement before pilot channel failure.
- The *CDM+ R+, threshold is less than or too close to (1)d*. As a result,
the criteria for B$# handover or cell chan!e can be not fulfilled before pilot
channel failure.
Figure &5. 5mproper inter+R.* mobility 6*R.7 parameter settings
Solution
The direct solution is to increase both 'sed re(uen)y threshold !d and
*CDM+ R+, threshold so that they are far away the 4re!ion5 in where pilot
channel fails.
The chan!e of these parameters will lead more 3"s in compressed mode and
more inter-RAT handover or cell chan!es. It mi!ht cause the hi!h load in the
3TRA' cells because many users are in compressed mode and hi!h con!estion
in B$# network. urthermore, these parameters are the R'+ parameters. It
means chan!in! them will affect the entire cells in that R'+.
8.!.! Reason 2 compressed mode function failure
The 3" has sent the event 7d measurement report to the 3TRA' to re2uest B$#
measurement. :owever, due to compressed mode function failure, please read
chapter =.D9, B$# measurement cannot be set up and no event 9a setup attached
on measurement control is sent to the 3".
8.!." Reason ! improper inter-R(& mo#ility +S' parameter
settings
The GSM threshold "a is set too hi!h. As a result, the criteria for B$#
handover or cell chan!e can be not fulfilled before pilot channel failure.
Solution
The direct solution is to decrease the GSM threshold "a to the avera!e R$$I
value of the B$# carrier. The chan!e of this parameter will lead more inter-RAT
handover or cell chan!es and mi!ht cause hi!h con!estion in B$# network.
urthermore, this parameter is the R'+ parameter. It means chan!in! it will
affect the entire cells in that R'+.
8.!., Reason " -nter-R(& mo#ility e$ent detection failure
,lease read chapter =.DD.
8.!.. Reason , -nter-R(& function failure
,lease read chapters =.D8 and =.D).
8." Pilot pollution
8.".1 Symptoms
In idle or cellHA+: mode, phenomenon of the pilot pollution is that a 3"
cannot firmly camp on a cell at one location because of receivin! many pilot
channels with similar 2uality @or si!nal stren!thA, i.e., "c&'o @or R$+,A.
The condition to determine the area has pilot pollution in idle or cellHA+:
mode is that third pilot appears in the cell re-selection re!ion.
Figure &6. 2ell re+selection region
pilotHcountI1JyK means number of pilots appearin! within and includin! I1JyK
ran!e.
In cellH/+: mode, phenomenon of the pilot pollution is that a 3" at one location
fre2uently chan!es its active set cells @active set update rate is very hi!hA because
of receivin! many pilot channels with similar 2uality @or si!nal stren!thA, i.e.,
"c&'o @or R$+,A. It causes hi!h si!nalin! load in RR+ and Iub interfaces and the
capacity of R'+ is conse2uently reduced.
The condition to determine the area has pilot pollution in cellH/+: mode is the
number of pilots within the Reporting Range 1-: threshold or drop window
ran!e is lar!er than Ma$ +)tive Set.
'otes%
1. The unit for the Reporting Range 1-: threshold or drop window is ..8d*.
7. or the statistics result for the pilot pollution in pilot scannin!, it is
recommended that%
Per)entage o samples having pilot pollution in the
pilot s)anning should -e _ 1.L
9. The area havin! pilot pollution normally accompanies with hi!h downlink
interference. It is because much received common channel powers from the
overlappin! cells are accumulated in there.
D. In previous discussion, it is assumed that all overlappin! cells have handover
relationships with each other.
8.".2 Reason no dominant cell
The reason is too many overlappin! cells at an area due to poor cell plannin!, i.e.
hi!h cch value, please read e2uation 1..
It should be noted that below solutions are for the campin! and hi!h si!nalin!
load problems. or hi!h interference in /0, please read chapter =.1.7.
Solution 1
The direct solution is to remove the cells overlappin! by chan!in! the antenna
confi!urations or reducin! pilot powers of the unwanted cells. :owever, the
chan!e needs to involve many cellsJ otherwise, the result will not be obvious.
The drawbacks of chan!in! antenna confi!uration and reducin! pilot power are
respectively same as these for the solution 7 and 9 in chapter =.1.9.
Solution !
The most effective solution is to increase the pilot channel power Primary
CPICH power of the desired cell. The drawbacks are same as these for the
solution in chapter =.1.7.
Solution "
or the pilot pollution in idle or cellHA+: mode, . /set !sn 0CPIC RSCP1
parameter of the desired cell can be increased to create a dominated cell.
- It mi!ht cause different borders in idle&cellHA+: and cellH/+: modes. To
solve this problem, the same value should be !iven to the Cell Individual
/set and . /set !sn 0CPIC RSCP1.
- It mi!ht cause connection setup failure because 30 or /0 cannot transmit
enou!h initial power for the enlar!ed idle mode border.
8., )plink and pilot co$erage im#alance
8.,.1 Symptoms
- Received "c&'o of the pilot channel is lar!er than (1)d* and
- Transmitted 3" power reaches to ma1imum allowed value and
8.,.2 Reason 1 large pilot channel power
The uplink and pilot covera!e are imbalance because the pilot power of the cell is
set too lar!e.
Solution
>hen the uplink covera!e border @,RA+: or /,+:A cannot reach the soft
handover area location, the pilot covera!e is lar!er than the uplink covera!e. The
only way to solve this problem is to reduce the pilot power Primary CPICH
power. This modification will reduce the downlink covera!e and pull back the
soft handover area. 'othin! can be done on the uplink side since 3" T1 power is
restricted by terminal desi!n.
Figure &!. Reducing the pilot power can rectify uplin%+pilot coverage imbalance.
The drawbacks of reducin! pilot power are same as these for the solution 9 in
chapter =.1.9.
8.,.! Reason 2 high )/ &0 power restriction
Another possible reason is due to too hi!h 3" T1 power restriction, i.e. too low
'2 Ma$ 3ransmission Power. Thus, the 3" T1 power is limited by this
parameter settin!.
Solution
The '2 Ma$ 3ransmission Power should be set as the one used in
dimensionin! or cell plannin!. or e1ample, if the ma1imum 3" T1 power is
assumed to be 7Dd*m for all 3" classes in the dimensionin!, the '2 Ma$
3ransmission Power should be set to be 7Dd*m.
The '2 Ma$ 3ransmission Power parameter will affect the cell re-selection
procedures in idle mode. If it is set too hi!h, the 4,compensation5, which is e2ual
to ma1imum value between 4the '2 Ma$ 3ransmission Power - output power
of the 3" accordin! to its class5 and 6ero, becomes lar!e and the idle mode cell
covera!e for some 3" classes will then be shrunk.
8.,." Reason ! low restriction in )1 congestion control
The ,7.1 con!estion control can limit the received 30 R$$I. If the parameters are
set in too low restriction, i.e. too hi!h '% Cong reset level, the cell mi!ht be
overloaded in uplink and the uplink covera!e becomes smaller than the planned
one.
'ote% The cell bein! overloaded in uplink can be due to hi!h own cell
interference or hi!h other cells interference or both.
Solution
The '% Cong reset level should be set to be same as the one used in
dimensionin! or cell plannin!.
8.,., Reason " incorrect power measurement due to &'(
If the sum of all parameters for uplink antenna path is lar!er than the sum of the
real values, the uplink R$$I then will be under-estimated. The cell mi!ht have be
overloaded in uplinkJ however, the con!estion control does not determine this.
Therefore, the uplink covera!e will be shrunk than the planned one. or instance,
*able 6. .n example to illustrate the incorrect estimation for 61 R--5
A Received 30 R$$I at the
-=.d*m
* Real T#A !ain 1.d*
+ Real uplink feeder loss -9d*
/ = AG*G+ #easured 30 R$$I at the R*$
port
-F9d*m
" ,arameter for T#A !ain 7.d*
,arameter for uplink feeder
loss
-1d*
B = /-"- +alculated 30 R$$I sent from
R*$ to R'+
-;7d*m
Solution
The uni2ue way to solve this problem is to correct these parameters as the real
ones. :owever, it is a time consumin! solution and difficult to measure the
accurate feeder loss and T#A !ain value.
8.,.. Reason , )/ in compressed mode
>hen the 3" is in the uplink compressed mode and its T1 power reaches to the
ma1imum allowed value, it means the pilot power of the cell is set too lar!e. As a
result, the compressed mode uplink and pilot covera!e can be defined as
imbalance.
Solution
The only way to solve this problem is to reduce the pilot power Primary CPICH
power. The drawbacks of reducin! pilot power are same as these for the solution
9 in chapter =.1.9.
Figure &". Reducing the pilot power can rectify uplin% 8in compressed mode9+
pilot coverage imbalance.
8.. &ransmitting ma0 21 code power full co$erage
*elow considerations are only for the /0 /,+:s which are planned for full
covera!e.
8...1 Symptoms
rom the drive test, followin! symptoms will be observed by usin! T"#$ and
3"TR%
- Transmitted /0 code power reaches to the ma1imum allowed value and
- The /0 *0"R of the connection is hi!h and
- The connection finally drops or switches down to the ne1t lower data rate if
the radio bearer of the connection is ,$)D&9=D or ,$)D&17=.
rom the performance statistics, many down-switchin! events, i.e.
pmChSwitchP384P128 and pmChSwitchP128P64 are very hi!h, are found for
the /0 /,+:s which are ori!inally planned for full covera!e.
8...2 Reason low ma0imum allowed 21 code power
The ma1imum allowed /0 code power Ma$ D% Power is set too low, please see
i!ure 7;.
Figure &#. $1 $(23 and pilot coverage imbalance is solved by increasing the
maximum allowed code power.
Solution
The direct solution is to increase the ma1imum allowed /0 code power Ma$ D%
Power to match with the pilot covera!e. The drawbacks are%
- If the total /0 T1 power is already very hi!h, the chan!e mi!ht lead to
con!estion drops due to hi!h peak to avera!e #+,A output ratio.
- A sin!le 3" with hi!h ma1imum /0 code power at the cell border mi!ht
overload the cell and then the whole cell will be blocked, i.e. admission
control denies the new connections.
- Ma$ D% Power is a R* connection parameter. +han!in! it will affect all
connections with the same R* in a R'+.
'ote% In the inter-R'+ soft handover case, the ma1imum /0 power for the link
not in the $-R'+ is calculated accordin! to /0 power mappin!, not Ma$ D%
Power. If it is the case4 Intermediate relative power or ma$ power mapping,
Ma$ relative power or ma$ power mapping and Min relative power or ma$
power mapping should be simultaneously increased in order to increase the
ma1imum allowed /0 power. urthermore, these parameters for the power
mappin! are the cell parameters. It means chan!in! them will not affect the entire
cells in the R'+, unlike Ma$ D% Power. ,lease refer +,IMs capacity
mana!ement for the power mappin! calculation.
8.3 &ransmitting ma0 21 code power not full co$erage
Eperator may desi!n not full covera!e for ,$)D&9=D or ,$)D&17= radio bearer,
i.e. intended /0 /,+: and pilot covera!e imbalance. >hen the connection in
these types reaches to the covera!e border, the channel switchin! function then
switches it down to the ne1t lower bit rate radio bearer so as to maintain the
connection.
The al!orithm is like that the channel switchin! monitors the downlink
transmitted code power of all radio links in the active set. A down-switch, to the
ne1t lower bit rate radio bearer, is re2uested if all radio links in the active set use
a transmitted code power close to the ma1imum allowed code power.
8.3.1 Symptoms
3"TR%
- Transmitted /0 code power reaches to ma1imum allowed value and
- The /0 *0"R of the connection is hi!h and
8.3.2 Reason downlink power drift
>hen usin! downlink outer&inner loop power control durin! soft handover, the
downlink power of the involved radio links is assumed to be coordinated by
startin! the radio link at the correct power level and then by receivin! the same
T,+ commands from the 3". :owever, the received T,+ commands at different
R*$s will be affected with different errors and conse2uently the downlink output
power of the different radio links will start to drift and be uncoordinated between
the R*$s. This phenomenon is defined as the downlink power drift.
Therefore, if downlink power drift happens, it is possibly that not all radio links
reach to ma1imum allowed code power even thou!h the 3" has been at the
covera!e border. inally, the radio bearer cannot be switched down and the
connection drops. ,lease see i!ure 9..
Figure '. Example of downlin% power drift
Solution
The en!ineer should chan!e the method for downlink power control to
4balancin!5, i.e. D% power )ontrol method = 9, because the e1tent of downlink
power drift can be reduced throu!h the downlink power balancin! procedure.
R'+ allocates a reference power to each R*$, which involves in soft handover.
The downlink power used by each radio link is periodically ad<usted in si6e to the
offset between the power used by the /,++:&/,/+: and the reference power.
The result of this is that the R*$Ns power levels slowly conver!e. In this way, the
R*$ power drift is reduced.
The reference power for the R*$s is calculated in the R'+. It is as the avera!e of
the measured values for transmitted code power from all involved radio links,
e1cept the link is in uplink out of synchroni6ation.
'ote% The /0 power control method is a R'+ parameter. +han!in! it will affect
the whole R'+.
8.8 2ownlink 445s and pilot co$erage im#alance
8.8.1 Symptoms
rom the drive test, followin! symptom@sA will be observed by usin! T"#$ and
3"TR%
- 3" does not receive the pa!in! si!nal sent out from the 3TRA' or
- /urin! RR+ connection establishment, 3" does not receive RR+ connection
setup messa!e sent out from the 3TRA' or
- /urin! cell update, 3" does not receive cell update confirm messa!e sent out
from the 3TRA' or
- 3" does not receive any system information distribution or
- "tc.
8.8.2 Reasons low common channel powers
The powers of the /0 common channels are not automatically modified with the
variation of the network load. or that reason, the en!ineer has to carefully plan
enou!h common channel powers to fulfill the assumed downlink load. If mistake
is done durin! the power calculations or the real load of the downlink carrier is
lar!er than the assumed level, above symptoms may then occur.
The information of the power settin!s for the common channels can be obtained
in reference I=K.
Solution
/0 ++: and pilot covera!e imbalance problems can be solved by increasin! the
common channel power level. ,arameters for settin! the power on the common
channels are !iven in Table F. 'ote% All /0 ++: parameters are per cell
parameters
*able !. (arameters for common channels
Common )hannel Parameter or power setting
*+: 5CH power
$+: primaryS)hPower, se)ondaryS)hPower
A+: Ma$ F+CH power: First, Ma$ F+CH
power: Se)ond
,+: PCH power
,I+: PICH power
AI+: +ICH power oset
The drawbacks of increasin! the powers of the /0 common channels are to
introduce more /0 interference in the carrier and to increase the load of the cell.
8.6 -nsufficient recei$ed )1 2P45 power
8.6.1 Symptoms
3"TR%
- 3" T1 power does not reach the ma1imum allowed value and
- 30 $IR tar!et of the radio connection reaches to the ma1imum allowed $IR
tar!et and
- 30 *0"R of the radio connection increases and
8.6.2 Reason low ma0imum allowed )1 S-R target
The base station cannot receive sufficient power from the uplink dedicated
physical channel if the ma1imum allowed 30 $IR tar!et SIRma$: Ma$imum
allowed SIR target is set too low.
Figure 1. 4aximum allowed -5R target setting too low
Solution
In order to allow 3"s to transmit with hi!her power, the ma1imum allowed 30
$IR tar!et SIRma$: Ma$imum allowed SIR target should be set sufficiently
hi!h. The drawbacks are%
- The 3"s allow transmittin! with hi!her power. They cause hi!her uplink
interference to the ad<acent cells, i.e. other cells interference.
- The SIRma$: Ma$imum allowed SIR target is a R'+ parameter. +han!in!
it will affect all cells in the R'+.
8.17 Rapidly changing radio en$ironment
8.17.1 Symptoms
rom the drive test, the followin! symptoms will be observed by usin! T"#$
and 3"TR%
- The 3" rapidly moves into another radio environment, e.!. from indoor to
outdoor.
- 3" T1 power does not reach the ma1imum allowed value and
- 30 *0"R of the radio connection increases and
8.17.2 Reason )1 outer loop power control pro#lem
If a cell covers several diverse radio environments, e.!. outdoor, indoor, tunnels
etc., the outer loop power control may not properly be able to adapt the rapid
environment chan!es.
Solution 1
The en!ineer should check if the re!ulation type for the uplink outer loop power
control al!orithm is set to 4Oump5 re!ulator, i.e. '% /uter %oop Regulator = 1.
>ith the 4Oump5 re!ulator, the 30 $IR tar!et is chan!ed fre2uently to follow the
*0"R pro!ression more closely. The step to increase the $IR tar!et is lar!eJ on
the other hand, the step is small for decreasin! the $IR tar!et. This re!ulator is
!ood for rapid environment chan!es.
Figure &. :;ump< regulator, referred from 2(5 document
The drawbacks are%
- The 30 $IR tar!et will fluctuate much. To reduce the unnecessary
fluctuation, the step si6e of the 30 $IR tar!et '%SIRstep: Step si6e or SIR
target )hange should be taken very small.
- In order to speed up the $IR tar!et cover!ence, it is necessary that the initial
$IR tar!et is set below the assumed value instead of above it since if the
2uality is too !ood the *0"R estimation takes lon!er time than if the 2uality
is sli!htly too poor.
- '% /uter %oop Regulator is a R'+ parameter. +han!in! it will affect the
whole R'+.
Solution !
If the uplink interference of the network is not hi!h, the step si6e of the 30 $IR
tar!et '%SIRstep: Step si6e or SIR target )hange can be increased so that the
30 $IR tar!et can be chan!ed 2uickly to overcome the hi!h *0"R situation.
The drawbacks are%
- The avera!e 30 $IR tar!et will be hi!her and the 3"s will transmit in hi!h
power. Then the uplink interference in the carrier is conse2uently increased.
- '%SIRstep: Step si6e or SIR target )hange is a R'+ parameter. +han!in!
it will affect the whole R'+.
Solution "
The last possible solution is to split the cell into several cells. urthermore, if
necessary, settin! the Cell Individual /set to be positive value lets the 3"
connect to the cell, which may become !ood very 2uickly, in advance.
Figure . $iverse radio environments may be covered by dedicated cells
:owever, it is a cost- and time-consumin! solution because installation work is
needed.
8.11 Swapped feeders
8.11.1 Symptoms
$wapped feeders can cause many ma<or problems in the network, e.!. no
downlink covera!e, no uplink covera!e or hi!h 30&/0 interference. *elow are
some @not allA e1amples of swapped feeders%
Figure /. Example of swapped feeders
The monitorin! tools for swapped feeder problem are pilot scanner and T"#$.
The symptoms can be hi!h downlink interference, sli!ht hi!h 3" T1 power,
connection setup failure durin! random access or uplink /,+: synchroni6ation
procedures, no downlink covera!e, handover failure, wron! scramblin! code
covera!e, etc.
or e1ample, please refer i!ure 9D%
*able ". -ymptoms of swapped feeders
Solution
The direct solution is to check that feeders are not crossed and the scramblin!
codes are set correctly for the all cells in the site.
8.28 1ow data throughput
8.28.1 Symptoms
rom the drive test, followin! symptoms will be observed when downloadin!
files to&from the operatorMs server @or the known public serverA by T, and
simultaneously pin!in! that server. ,lease read chapter 8.D.
- Avera!e 30 or /0 throu!hput of the radio access bearer is much lower than
the data rate of the known source or
- Round trip time is very lon! or
- #any missin! packets.
8.28.2 Reason 1 poor radio link 8ualities
,oor radio links lead error bits in packet. In order to recover the packet, A# R0+
retransmits the problematic packets. :owever, too many retransmissions cause
lon!er RTT and the date throu!hput conse2uently decreases.
,lease read chapters =.). and =.)1.
8.28.! Reason 2 many down-switches due to admission control
*est effort cleanup mechanism allows more system accessibility to !uaranteed
and non-!uaranteed users. :owever, the throu!hput for non-!uaranteed users is
sli!htly reduced.
8.28." Reason ! many down-switches due to co$erage triggering
If ,$)D&9=D or ,$)D&17= radio bearer and pilot covera!e are imbalance, the
channel switchin! function then switches it down to the radio bearer in the ne1t
lower bit rate so as to maintain the connection when it reaches to the covera!e
border. :owever, the overall throu!hput of the connection becomes lower.
8.28., Reason " many down-switches in soft or softer hando$er
If an addition or replacement fails in soft or softer handover due to out of radio
resource in the tar!et cell, ,$)D&9=D or ,$)D&17= radio bearer may be switched
down to ,$)D&)D so as to maintain the mobility. :owever, the overall throu!hput
of the connection becomes lower.
8.28.. Reason , many down-switches due to congestion control
/ue to con!estion, the connection mi!ht be switched down to the common
channel. Therefore, the overall throu!hput of the connection becomes very low.
8.28.3 Reason . - high 9(45 usage
8.28.8 Reason 3 - high switching rates #etween dedicated and
common channels
8.26 (#normal disconnection in soft or softer hando$er
8.26.1 Symptoms
3"TR%
- 4RR+ connection release5 09 messa!e with unspecified reason is sent out
from the 3TRA' then
- 4RR+ connection release complete5 09 messa!e is sent from the 3" to the
3TRA' then
- +onnection drops.
These symptoms can be applied to the radio connection in the connected mode,
i.e. either cellH/+: or cellHA+:.
8.26.2 Reason 1 acti$e set update timer e0pires
In soft or softer handover, when the comin! cell has allocated resources for new
radio link, the 3TRA' sends an 4active set update5 messa!e to the 3" and starts
the active set update timer. If the 3TRA' receives an 4active set update
complete5 messa!e from the 3", the timer will be stopped. Etherwise, the
3TRA' waits for the timer e1piry. After the e1piry, the 3TRA' orders the 3"
to release the connection because the 3TRA' considers the connection is no
!ood.
It should be noted that%
- After the 3" sends the 4active set update complete5 messa!e to the 3TRA',
the 3" is still usin! the old nei!hborin! cell list but considers the handover
had been completed no matter if the 3TRA' receives that messa!e. If the
3TRA' does not receive the complete messa!e or the contents in the
messa!e cannot be decoded correctly, the 3TRA' will not send out the
4measurement control5 messa!e to update the nei!hborin! cell list for the
3".
- The active set update timer in ,7.. and ,7.1 are respectively 11 and 8
seconds.
Solution
It is possible that%
- The 3TRA' has sent the 4active set update5 to the 3"J however, the 3"
does not receive it. The reason is because of poor 2uality in the downlink.
,lease read chapter =.)1.
- The 3" sends the 4active set update complete5 to the 3TRA'J however, the
3TRA' does not receive it. It is due to poor uplink 2uality. ,lease read
chapter =.)..
- The 3" sends wron! contents in the complete messa!e and the 3TRA'
cannot decode it correctly. ,lease check what content the 3" sent in the
4active set update complete5 messa!e.
8.26.! Reason 2 measurement control failure
After the 3TRA' receives an 4active set update complete5 from the 3", the
3TRA' sends a 4measurement control5 messa!e to inform the 3" about the new
nei!hborin! cells. If the 3" cannot decode the 4measurement control5 messa!e
and sends a 4measurement control failure5 messa!e back to the 3TRA', the
3TRA' considers that the 3" will not measure on ordered nei!hborin! cells and
also not report on enterin! these cells. To avoid the 3" runs into a cell and starts
interferin!, the 3TRA' orders the 3" to release the connection.
Solution
The reason of havin! decodin! problem is due to poor downlink 2uality. ,lease
read chapter =.)1.
8.26." Reason ! monitored set cell #etter than the #est ser$ing cell
>hen addition or replacement of a radio link fails and the 2uality of this radio
link is better than that of the best cell in the active set, the 3TRA' orders to
release the connection in order to avoid e1cessive con!estion caused from that
3".
Figure 5. Example of monitored set cell better than the best serving cell
0ike what is described in chapter =.7;.7, the 3" mi!ht think the process of the
handover has been completed after sendin! the 4active set update complete5.
:owever, it mi!ht be not from the 3TRA' point of view. >hen the 3" sends the
4measurement report5 to the 3TRA' to report the 2ualities of the cells in the
active set, the 3TRA' will only consider this fake active set cell as a monitored
set cell. If the 2uality of this fake active set cell is better than that of the best cell
in active set, the 3TRA' will order to release the connection.
Solution
,lease read chapters =.9. and =.91.
8.!7 5ando$er e$ent detection failure
The definition of the handover event detection failure in this !uideline is that the
network side fails to receive the measurement report when the 3" enters @or
leavesA the desired @or undesiredA cell covera!e area.
8.!7.1 Symptoms
3"TR%
- The 3" sends 4measurement reports5 to the 3TRA'J however, the 3TRA'
does not receive them or
- The 3" does not !enerate 4measurement reports5 even if it enters the desired
cell covera!e area or
- The 3" does not !enerate 4measurement reports5 even if it leaves the
undesired cell covera!e area.
'ote% The covera!e area of the desired @or undesiredA cell can be verified by
usin! pilot scanner.
8.!7.2 Reason 1 poor uplink 8uality
If the 3" sends 4measurement reports5 to the 3TRA'J however, the 3TRA'
does not receive them, it implies the 2uality of the uplink may be no !ood.
Solution
,lease read chapter =.)..
8.!7.! Reason 2 incorrect hando$er series
The 3" mi!ht handover to wron! cells, i.e. incorrect cells in the active set at the
problematic area, and these active set cells do not have nei!hborin! relationship
with the desired cell. As a result, even thou!h the 3" enters the desired cell
covera!e area accordin! to the ori!inal plan, handover event is not detected.
:andover series could be messed up because%
- :andover function failure or
- Too many unnecessary handover relationships or
- 3nwanted cell overshootin!
Solution 1
If the problem is due to handover function failure, please read chapter =.91.
Solution !
If the problem is due to too many unnecessary handover relationships, the
en!ineer should carefully <ustify the usefulness of the handover relationships and
remove the unnecessary ones.
'ote% If the necessary relationship is mistakenly deleted, more connection drops
and hi!h downlink interference may happen due to missin! handover
relationship.
Solution "
If the problem is due to overshootin! of the unwanted cell, the en!ineer should
check why the cell is overshootin!.
- If the pilot power of the unwanted cell is too hi!h, the en!ineer should reduce
its pilot power back to a suitable level.
- If the pilot power of the unwanted cell is normal, please employ solution 7
and 9 in chapter =.1.9 to overcome the overshootin! problem.
8.!7." Reason ! - missing neigh#oring relationship
The possible reason why the handover event detection fails is because of missin!
nei!hborin! cell relationship. In the drive test, the en!ineer can monitor the
nei!hborin! cell window to check if the desired cell is included in the monitored
set.
Solution
The direct solution is to add the desired cell into the nei!hborin! cell lists of the
cells in the active set. :owever, it should be noted that too many nei!hborin! cell
relationships mi!ht slow down the search for the pilot channels in the 3".
8.!7., Reason " pilot pollution in dedicated mode
,ilot pollution can lead failure in the handover event detection. ,lease read
chapter =.D for the descriptions and solutions about pilot pollution.
8.!7.. Reason , - too long intra-fre8uency neigh#oring cells list
Accordin! to 9B,, standardi6ation, the ma1imum number of 3" measured cells
is 97 for the currently used fre2uency. Therefore, if the total number of cells in
active set and preliminary intra-fre2uency monitored subset is more than 97, the
preliminary intra-fre2uency monitored subset has to be reduced. /ue to this
reduction, the desired cell mi!ht be removed in the handover al!orithm and
handover event detection fails.
'ote%
- The cells in active set are the cells involved in soft or softer handover.
- The cells in the preliminary intra-fre2uency monitored subset are the cells
taken from the union of the currently used fre2uency cells in the nei!hborin!
cell lists of the active set cells and then deletin! the cells that are in the active
set. or e1ample @please see Table 11A, if there are three cells in the active set
( cell A, cell * and cell +, the preliminary intra-fre2uency monitored subset
will contain +ell A1, +ell A7, +ell A9, +ell AD, +ell *1, +ell *7, +ell *9,
+ell +1 and +ell +7.
*able 11. .n example to illustrate the preliminary intra+fre=uency monitored
subset
Active set cells Intra-fre2uency nei!hborin! cell lists
+ell A +ell *, +ell +, +ell A1, +ell A7, +ell A9, +ell
AD
+ell * +ell A, +ell +, +ell *1, +ell *7, +ell *9
+ell + +ell A, +ell *, +ell +1, +ell +7
- Re!ardin! the reduction in the preliminary intra-fre2uency monitored subset,
the handover al!orithm will select and then remove a number of cells from
the nei!hborin! cell lists in each active set cell respectively in order to be
sure that each active set cell can contribute its nei!hborin! cells for
monitorin!. The removed nei!hborin! cells are normally at the bottom of the
nei!hborin! cell lists.
Solution 1
If there are too many unnecessary handover relationships, the en!ineer should
carefully <ustify the usefulness of the handover relationships and remove the
unnecessary ones.
relationship.
Solution !
The en!ineer should put the nei!hborin! cells in a desired order so the hi!hest
ranked cells appear in the top of the list.
8.!7.3 Reason . improper cell indi$idual offset setting
If the cell individual offset Cell individual oset of the desired cell is too
ne!ative, handover event detection mi!ht fail.
Solution
The Cell individual oset should be chan!ed to not too lar!e ne!ative. If the
performance of load shiftin! becomes no !ood after the chan!e, new site must be
added instead of usin! Cell individual oset to shift the load.
8.!7.8 Reason 3 slow searching or fast )/
:andover events are probably overlooked because%
- There are many cells in the monitored set, the process for searchin! pilot
channel mi!ht become very slow.
- The 3" is in fast movin!.
Solution
The en!ineer should carefully <ustify the usefulness of the handover relationships
and remove the unnecessary ones.
relationship.
8.!7.6 Reason 8 small soft hando$er area
/ue to radio propa!ation, the soft handover area between two cells mi!ht be very
small, e.!. corner effect. Then, a handover event is probably overlooked.
Solution 1
Bivin! a positive value to the cell individual offset Cell individual oset of the
tar!et cell can enlar!e the soft handover area.
Solution !
The en!ineer can chan!e the antenna confi!uration or re-locate the sites so that
the cell overlappin! and soft handover area becomes lar!er. Ebviously, it is a cost
and time-consumin! solution.
8.!2 5igh acti$e set update rate
8.!2.1 Symptoms
rom the drive test, it can be observed that the number of cells in the active set is
chan!ed fre2uently.
rom the performance statistics, the avera!e rate in active set update, i.e.
of a cell is very hi!h.
It should be noted that event for handover replacement is e2ual to 4handover
addition5 event in a cell plus 4handover deletion5 event in another cell.
Therefore, the addition counter, i.e. pmNoTimesRlAddToActSet and deletion
counter, i.e. pmNoTimesRlDelFrActSet @in different cellsA will also been stepped
up when handover replacement. This is the reason why the replacement counter,
i.e. pmNoTimesRlRepInActSet is not included in the e2uation for active set
update rate.
8.!2.2 Reason 1 ping-pong caused due to improper hando$er
parameter settings
If the difference between Reporting Range 1a: threshold or addition window
and Reporting range 1-: 3hreshold or drop window is too small, a cell mi!ht
be removed immediately from an active set after it <ust is added into the active
set.
If Hysteresis 1) is too small, a cell mi!ht be replaced out from an active set after
it <ust replaces another cell.
Figure 55. 5mproper handover parameter settings
Solution
'ormally, the Reporting range 1-: 3hreshold or drop window is set D, i.e.
7d* lar!er than the Reporting Range 1a: threshold or addition window and
the Hysteresis 1) is set to be 7d*.
It should be noted these handover parameters are R'+ parameters. It means
chan!in! them will affect entire cells in the R'+.
8.!2.! Reason 2 pilot pollution in dedicated mode
,ilot pollution can lead hi!h update rate in the active set. ,lease read chapter =.D
for the descriptions and solutions about pilot pollution.
8.!2." Reason ! many unnecessary hando$er relationships
Too many unnecessary handover relationships can lead hi!h update rate in the
active set.
Solution
The en!ineer should carefully <ustify the usefulness of the handover relationships
and remove the unnecessary ones.
relationship.
8.!! *o suita#le cell
8.!!.1 Symptoms
rom the drive test, followin! symptoms will be observed%
- The 3" in the idle mode does not camp on any cell. The display of the 3"
shows 4no covera!e5 or
- After enterin! no suitable cell, the 3" in cellHA+: mode moves to idle
mode after min@3"78, 3"18A. It is same as connection drop for the 3" in the
cellHA+: mode. ,lease see i!ure 8).
Figure 56. 2onnection drop when the 6E is in cell>F.23 mode.
3"78 and 3"18 are waitin! times before the 3" moves to idle mode when the 3"
enters no suitable cell.
3"79, which is called 3"794 Periodi) )ell update timer in RA'E$, is timer for
periodic cell update. It is ad<ustable for the operator.
rom the pilot scannin!, it is observed%
- The 2(523>Ec)7o is e2ual or less than the .(ualMin: Minimum (uality
or sele)tion:resele)tion or
- The 2(523>R-2( is e2ual or less than the .r$levMin: Minimum level or
Sele)tion:Resele)tion ; ,compensation.
8.!!.2 Reason 1 high downlink interference
The possible reason causin! no suitable cell is same as that for the pilot channel
failure due to hi!h downlink interference. ,lease read chapter =.1.
'ote%
1. The 2(523>Ec)7o tri!!ered to determine the pilot channel failure is (1)d*.
:owever, the 2(523>Ec)7o tri!!ered to determine the no suitable cell is
.(ualMin: Minimum (uality or sele)tion:resele)tion.
7. If the system is allowed to have inter-RAT or inter-fre2uency cell re-selection
and the search parameters are set properly, above reason is not valid.
8.!!.! Reason 2 out of pilot co$erage due to 8uality
The possible reason causin! no suitable cell is same as that for the pilot channel
failure due to out of pilot covera!e. ,lease read chapter =.7.
'ote%
1. The 2(523>Ec)7o tri!!ered to determine the pilot channel failure is (1)d*.
:owever, the 2(523>Ec)7o tri!!ered to determine the no suitable cell is
.(ualMin: Minimum (uality or sele)tion:resele)tion.
7. If the system is allowed to have inter-RAT or inter-fre2uency cell re-selection
and the search parameters are set properly, above reason is not valid.
8.!!." Reason ! out of pilot co$erage due to signal strength
If the 3" moves far away the cell and the 2(523>R-2( is e2ual or less than the
@.r$levMin: Minimum level or Sele)tion:Resele)tion ; ,compensationA, no
suitable cell happens.
'ote%
1. ,compensation is e2ual to the ma1imum value between 4the '2 Ma$
3ransmission Power < output power of the 3" accordin! to its class5 and
6ero.
7. Inter-RAT and inter-fre2uency cell re-selection is tri!!ered by 2(523>Ec)7o
2uantity. Therefore, if the 2(523>Ec)7o is lar!er than the .(ualMin:
Minimum (uality or sele)tion:resele)tion G S R+3 Sear)h @or G S Inter
Sear)hA, inter-RAT @or inter-fre2uencyA cell re-selection will not be tri!!ered
even thou!h the 2(523>R-2( is e2ual or less than the @.r$levMin:
Minimum level or Sele)tion:Resele)tion ; ,compensationA.
Solution
The en!ineer should check the reason why out of pilot covera!e.
- If it is because of lack of outdoor covera!e, new outdoor site should be
added.
- If it is because of lack of indoor covera!e, new indoor site should be added.
- If it is because of blockin! the si!nal due to new buildin!, the en!ineer
should try to chan!e the antenna confi!uration or to move the site first. If
both of them do not work, new outdoor site should be added.
8.!!., Reason " improper inter-R(& search parameter setting
If the system is allowed to have inter-RAT cell re-selection but the inter-RAT
search parameter S R+3 Sear)h is set as 6ero or ne!ative value, no suitable cell
mi!ht happen when the 2(523>Ec)7o drops e2ual to or less than the .(ualMin:
Minimum (uality or sele)tion:resele)tion. ,lease see i!ure 8F.
Figure 5!. Example of improper S R+3 Sear)h setting
Solution
If the system is allowed to have inter-RAT cell re-selection, the S R+3 Sear)h
should be set to be positive value.
8.!!.. Reason , improper wrong inter-fre8uency search parameter
setting
If the system is allowed to have inter-fre2uency cell re-selection but the inter
fre2uency search parameter S Inter Sear)h is set as 6ero or ne!ative value, no
suitable cell mi!ht happen when the 2(523>Ec)7o drops e2ual to or less than
the .(ualMin: Minimum (uality or sele)tion:resele)tion. ,lease see i!ure
8=.
Figure 5". Example of improper S Inter Sear)h setting
Solution
If the system is allowed to have inter-fre2uency cell re-selection, the S Inter
Sear)h should be set to be positive value.
8.!!.3 Reason . high restriction on cell :re;-selection parameters
If the cell @reA-selection parameters are set to be very hi!h restrictions, no suitable
cell may happen even thou!h the actual 2uality and si!nal stren!th of the pilot are
!ood enou!h to provide covera!e.
The parameters for the cell @reA-selection are%
- .(ualMin: Minimum (uality or sele)tion:resele)tion
- .r$levMin: Minimum level or Sele)tion:Resele)tion
- '2 Ma$ 3ransmission Power
'ote% The '2 Ma$ 3ransmission Power should be set as the one used in
dimensionin! or cell plannin!. or e1ample, if the ma1imum 3" T1 power is
assumed to be 7Dd*m for all 3" classes in the dimensionin!, the '2 Ma$
3ransmission Power should be set to be 7Dd*m. If it is set too hi!h, the
4,compensation5, which is e2ual to ma1imum value between 4the '2 Ma$
3ransmission Power - output power of the 3" accordin! to its class5 and 6ero,
becomes lar!e and the idle mode cell covera!e for some 3" classes will then be
shrunk.
Solution
The en!ineer should correct these parameters to suitable values.
8.!" 4ell re-selection failure - )/ in idle or cell<9(45
mode
8.!".1 Symptoms
rom the drive test, followin! symptoms will be found that by usin! T"#$ and
pilot scanner%
- The 3" in cellHA+: mode does not send 4cell update5 messa!e to the
3TRA' even it has entered covera!e area of the desired cell or
- The 3" in idle mode camps on the wron! cell even it has entered covera!e
area of the desired cell.
8.!".2 Reason 1 - incorrect cell re-selection series
The 3" mi!ht have camped on to wron! cell and this cell does not have
nei!hborin! relationship with the desired cell.
Re-selection series could be messed up because%
- +ell update procedure failure @for the 3" in cellHA+: modeA or
- Too many unnecessary nei!hborin! relationships or
- 3nwanted cell overshootin!
Solution 1
If the problem is due to cell update function failure, please read chapter =.98.
Solution !
If the problem is due to too many unnecessary nei!hborin! relationships, the
en!ineer should carefully <ustify the usefulness of these relationships and remove
the unnecessary ones.
Solution "
If the problem is due to overshootin! of the unwanted cell, the en!ineer should
check why the cell is overshootin!.
- If the pilot power of the unwanted cell is too lar!e, the en!ineer should
decrease its pilot power back to a suitable level.
- If the pilot power of the unwanted cell is normal, please employ solution 7
and 9 in chapter =.1.9 to overcome the overshootin! problem.
8.!".! Reason 2 - missing neigh#oring relationship
The possible reason why the cell re-selection detection fails is because of missin!
nei!hborin! cell window to check if the desired cell is included in the
nei!hborin! cell list.
Solution
The direct solution is to add the desired cell into the nei!hborin! cell list.
:owever, it should be noted that too many nei!hborin! cell relationships mi!ht
slow down the search for the pilot channels in the 3".
8.!"." Reason ! pilot pollution in idle or cell<9(45 mode
,ilot pollution can lead failure in the cell re-selection event detection. ,lease read
chapter =.D for the descriptions and solutions about pilot pollution.
8.!"., Reason " improper cell re-selection offset setting
If the cell re-selection offset . /set 1sn 0CPIC RSCP1 or . /set !sn
0CPIC RSCP1 between the camped cell and the desired cell is too positive, the
rankin! in the cell re-selection procedure of the desired cell becomes very low.
Therefore, even thou!h the actual 2uality and si!nal stren!th of the pilot in the
desired cell are !ood enou!h to provide covera!e, the 3" does not camp on the
cell, i.e. cell re-selection fails.
Solution
The . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC RSCP1 should be
chan!ed to not too lar!e positive.
'ote% +han!in! cell re-selection offset may cause different borders in idle and
dedicated modes. To solve this problem, the same value should be !iven to the
individual offset Cell individual oset and cell offset . /set 1sn 0CPIC
RSCP1 or . /set !sn 0CPIC RSCP1.
8.!".. Reason , slow searching or fast )/
+ells mi!ht be overlooked because%
- There are many cells in the nei!hborin! cell list. The process for searchin!
pilot channel mi!ht become very slow.
- The 3" is in fast movin!.
Solution
The en!ineer should carefully <ustify the usefulness of the nei!hborin!
relationships and remove the unnecessary ones. If the necessary relationship is
mistakenly deleted, hi!h downlink interference may happen due to missin!
relationship.
8.!".3 Reason . small cell o$erlapping area
/ue to radio propa!ation, the cell overlappin! area between two cells mi!ht be
very small, e.!. corner effect. Then, cell re-selection is probably overlooked.
Solution
the cell overlappin! area becomes lar!er. Ebviously, it is a cost and timeconsumin!
solution.
'ote% 3nlike soft handover area, an overlappin! area between cells cannot be
enlar!ed by modifyin! system parameters. ,lease see i!ure 8;.
Figure 5#. 0ariations of soft handover area and idle 8cell>F.239 mode
coverage when changing the system parameters
In case 1, the parameter settin!s for both cells are normal. In case 7, a ne!ative
value is !iven to the cell re-selection offset in cell A. En the other hand, both cell
individual offset and cell re-selection offset in cell * are set as the same value but
in positive. It can be observed that even thou!h the soft handover area and idle
@cellHA+:A mode covera!e are chan!ed in case 7, the actual overlappin! area is
still the same as before.
8.!, 4ell update procedure failure
>hen a 3" in cellHA+: mode re-selects a new cell, cell update procedure is
tri!!ered and the 3" sends 4cell update5 messa!e to the 3TRA' to inform the
location. If the cell update procedure fails, the connection will be dropped from
cellHA+: mode to idle mode.
8.!,.1 Symptoms
3"TR%
- +ase 1% The 3" repeatedly transmits 4cell update5 messa!e to the 3TRA'J
however, the 3TRA' does not receive the 4cell update5 messa!e or
- +ase 7% The 3TRA' receives the 4cell update5 messa!eJ however, it does
not send the 4cell update confirm5 messa!e to the 3" or
- +ase 9% The 3TRA' sends the 4cell update confirm5 messa!e to the 3"J
however, the 3" does not receive it.
'ote%
1. 9B,, 78.991-?.9.F.., which the ,7.1 3TRA' follows, specifies that the
number of the retransmissions of the 4cell update5 should be ,"7! if the 3"
does not receive the 4cell update confirmation5 messa!e. After @,"7! G
1AP3"7!, the 3" moves to the idle mode, i.e. connection drops. ,lease see
i!ure )..
Figure 6'. Repetition of cell update messages 81 retransmissions9 from the 6E
point of view
7. ?9.7 is an internal counter in the 3".
9. Total possible number of re-transmittin! the 4cell update5 messa!e is ,"7!.
Therefore, total possible number of transmittin! the 4cell update5 messa!e is
,"7!G1.
D. In the ,7 3TRA', the hard-coded values for 3"7! and ,"7! are D seconds
and 9 times, respectively. Therefore, the 3" moves to the idle mode after 1)
seconds after sendin! the first 4cell update5 messa!e if the 3" does not
receive the 4cell update confirmation5 messa!e.
8. rom the 3TRA' point of view, the time of startin! to release the 3TRA'
resources for a 3" is *ait )ell update timer after receivin! the last 4cell
update5 messa!e from that 3". This timer is chan!eable for the operator and
the default value for the *ait )ell update timer is D8 minutes. :owever,
chan!in! it will affect the whole R'+. ,lease see i!ure )1.
Figure 61. 2ell update from the 6*R.7 point of view.
). If no suitable cell is detected, 3"18 3" timer is started. If the 3"18 3" timer
e1pires the 3" moves to idle mode. 3nlike ,1 3TRA', the 3"18 is hardcoded
to be 1=. seconds in the ,7 3TRA'. Therefore, no suitable cell never
happens durin! the repetition of the cell update, i.e. @,"7!G1AP3"7! seconds.
8.!,.2 Reason 1 random access procedure pro#lem
The reason for case 1 is because of problem in the random access procedure.
,lease read chapter =.9).
8.!,.! Reason 2 cell update function failure
The possible reasons for case 7 are due to some miscellaneous problems, like
below%
- >hen the /-R'+ or $-R'+ cannot allocate any R'TI values, the 3TRA'
orders to release the connection and sends a 4RR+ connection release5
messa!e to the 3" or
- >hen the R'+-Id and the +-Id are not correctly received by the $-R'+, the
3TRA' orders to release the connection and sends a 4RR+ connection
release5 messa!e to the 3" or
- >hen the R'+-Id and the +-Id are not correctly received by the /-R'+, the
cell update procedure will be stopped and all reserved resources at the /R'+
will be released or
- >hen the Iur common transport channels cannot be established, e.!. due to
time-out or no resources available, the si!nalin! connections will be released
or
- >hen the R'+ receives an "RRER I'/I+ATIE' messa!e, the R'+ will
i!nore the messa!e and continue with the procedure.
- "tc.
8.!,." Reason ! S-44P45 :9(45; failure
In the ,7 3TRA', power of the $-++,+: is not automatically modified with the
enou!h $-++,+: power to fulfill the assumed downlink load. If mistake is done
durin! the power calculation or the real load of the downlink carrier is lar!er than
the assumed level, $-++,+: mi!ht not cover the whole cell.
Solution
,lease read chapter =.=.
8.!. Random access procedure pro#lem
The random access procedure consists of the followin! steps%
1. The 3" transmits a preamble.
7. If the R*$ properly detects the preamble it sends an Acknowled!ement
Indicator @AIA on the Ac2uisition Indication +hannel @AI+:A.
9. If the 3" does not receive an AI, it transmits a new preamble with hi!her T1
power.
D. If the 3" receives the AI, the ,RA+: messa!e part is sent.
Figure 6&. 5llustration of the random access procedure
In order to be acknowled!ed, the received preamble power must be @-9) G
..8Ppream-le=thresholdA d* over the interference. In the ,7 3TRA', the
pream-le=threshold parameter is hard-coded to be 9D, i.e. -1;d*. Therefore, the
correct parameter settin!s for RA+: power rampin! should be set like below%
so that the ma1imum possible preamble power is lar!e enou!h for AI+:
acknowled!ement.
Accordin! to 30 open loop power control, the initial preamble power is%
If the estimated uplink pathloss is same as the actual uplink pathloss and the
measured uplink total interference is similar to the actual uplink total
interference, the e2uation F8 will become%
8.!..1 Symptoms
3"TR%
- +ase 1% R*$ does not send out the AI+: acknowled!ement to the 3" and the
transmission power of the 3" is not up the its ma1imum allowed T1 power or
- +ase 7% R*$ does not send out the AI+: acknowled!ement to the 3" and the
transmission power of the 3" is up the its ma1imum allowed T1 power or
- +ase 9% R*$ sends out the AI+: acknowled!ement to the 3"J however, the
3" does not receive it or
- +ase D% R*$ sends out the AI+: acknowled!ement to the 3" and the 3"
receives it and then sends out the RA+: messa!e.
8.!..2 Reason 1 improper parameter settings for power ramping on
R(45
The reason for case 1 in chapter =.9).1 is because of improper parameter settin!
for power rampin! on RA+:. The e2uality in the e2uation F) is broken.
Solution
The en!ineer should ad<ust the parameters for RA+: power rampin! so that the
Pream-le Retrans Ma$imum P Power ramp set G Constant >alue is !reater
than or e2ual to (1;d*.
It should be noted that these parameters are per cell parameters.
8.!..! Reason 2 PR(45 and pilot co$erage im#alance
The reason for case 7 in chapter =.9).1 is because of the imbalance between
,RA+: and pilot channel covera!e. The power rampin! is restricted by the
ma1imum allowed 3" T1 power.
Figure 6. Example of (R.23 and pilot coverage imbalance
The reasons of causin! the 4,RA+: and pilot covera!e imbalance5 are similar to
these for 4uplink and pilot covera!e imbalance5. They are%
- 0ar!e power in the pilot channel. ,lease read chapter =.8.7.
- :i!h restriction to the 3" T1 power. ,lease read chapter =.8.9.
- 0ow restriction in 30 con!estion control. ,lease read chapter =.8.D.
8.!.." Reason ! insufficient (-45 power
The reason for case 9 in chapter =.9).1 is because of insufficient AI+: power. In
the ,7 3TRA', power of the AI+: is not automatically modified with the
enou!h AI+: power to fulfill the assumed downlink load. If mistake is done
durin! the power calculation or the real load of the downlink carrier is lar!er than
the assumed level, AI+: mi!ht not cover the whole cell.
Solution
8.!.., Reason " insufficient power for R(45 message
The reason for case D in chapter =.9).1 is because of insufficient power for
RA+: messa!e.
>hen the preamble is acknowled!ed, it implies that%
:owever, in order to receive the RA+: messa!e, the 3" T1 power should be%
Therefore, the correct parameter settin!s should be%
If the e2uality of the e2uation F; is broken, the power for RA+: messa!e mi!ht
be insufficient.
Solution
The en!ineer should check if the parameter settin!s fulfill the e2uality of the
e2uation F;.
It should be noted that the pream-le=threshold is hardcoded to be (1;d* in the
system. It implies the operators are only allowed to ad<ust the Constant >alue
and Power /set Ppm to solve this problem. 'ote% They are per cell
parameters.
8.!... Reason , incorrect power measurement due to &'(
If the parameters are set incorrectly, the estimated uplink pathloss or measured
total uplink interference mi!ht be wron!. As a result, the e2uality of the e2uation
F) is broken.
Solution
The uni2ue way to solve this problem is to correct these parameters as the real
ones. :owever, it is a time consumin! solution and difficult to measure the
accurate feeder loss and T#A !ain value.
8.!3 'any intra-fre8uency cell re-selections
8.!3.1 Symptoms
rom the drive test, it can be observed that the 3" fre2uently re-selects the cells
in the current fre2uency.
8.!3.2 Reason 1 improper re-selection hysteresis setting
The possible reason is that re-selection hysteresis .hyst 1: Resele)tion
hysteresis or .hyst !: Resele)tion hysteresis is set too low, e.!. . d*.
Figure 6/. 5ntra+fre=uency cell re+selection in ping pong when re+selection
hysteresis is set to be ?ero.
Solution
The en!ineer should set the .hyst 1: Resele)tion hysteresis or .hyst !:
Resele)tion hysteresis to be lar!er than 6ero, e.!. 7d*
8.!3.! Reason 2 inconsistent cell re-selection offset setting
+ell re-selection offset . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC
RSCP1 is in the cell relation. If the offset from source cell to tar!et cell is a
positive value, the offset from the tar!et cell to the source cell must be same
value but in ne!ative. Etherwise, the 3" mi!ht repeatedly camp on them, i.e.
pin!-pon!.
Solution
The en!ineer should correct the cell re-selection offsets between two nei!hborin!
cells so as to make them consistent.
8.!3." Reason ! pilot pollution in the idle or cell<9(45 mode
,ilot pollution can lead many cell re-selections. ,lease read chapter =.D for the
descriptions and solutions about pilot pollution.
8.!3., Reason " many unnecessary cell re-selection relationships
Too many unnecessary cell re-selection relationships can lead many cell reselections.
Solution
The en!ineer should carefully <ustify the usefulness of the cell re-selection
relationships and remove the unnecessary ones.
'ote% If the necessary relationship is mistakenly deleted, cell re-selection mi!ht
fail.
8.!8 'any inter-fre8uency cell re-selections
8.!8.1 Symptoms
rom the drive test, it can be observed that the 3" fre2uently re-selects the cells
in different fre2uencies.
8.!8.2 Reason 1 high downlink interference
The possible reason causin! many inter-fre2uency cell re-selections is same as
that for the pilot channel failure due to hi!h downlink interference. ,lease read
chapter =.1.
'ote% The 2(523>Ec)7o tri!!ered to determine the pilot channel failure is (
1)d*. :owever, the 2(523>Ec)7o tri!!ered to determine for inter-fre2uency
cell re-selection is .(ualMin: Minimum (uality or sele)tion:resele)tion G S
Inter Sear)h.
8.!8.! Reason 2 out of pilot co$erage due to 8uality
The possible reason causin! many inter-fre2uency cell re-selections is same as
that for the pilot channel failure due to out of pilot covera!e. ,lease read chapter
=.7.
1)d*. :owever, the 2(523>Ec)7o tri!!ered to determine for inter-fre2uency
cell re-selection is .(ualMin: Minimum (uality or sele)tion:resele)tion G S
Inter Sear)h.
8.!8." Reason ! improper inter-fre8uency search parameter setting
If the inter-fre2uency search parameter S Inter Sear)h is set to too hi!h positive,
unnecessary inter-fre2uency cell re-selection may be caused.
Solution
The en!ineer should modify the S Inter Sear)h to a suitable value, e.!. Dd*.
8.!8., Reason " ping-pong in inter-fre8uency cell re-selections
If
1. The operator employs second carrier, which totally co-sites with the first
carrier and
7. The pilot power settin! of the second carrier is same as that of the first carrier
and
9. The .measure: Sele)tion Resele)tion .uality Measure is set to be 1, i.e.
2(523>R-2(,
The 3" may repeatedly re-select select two nei!hborin! cells in different
fre2uencies.
In i!ure )8, the 2(523>R-2(s of the cells in both carriers 1 and 7 are similar
because two carriers are co-located and the pilot power settin!s of them are the
same. If the loads in both carriers are similar, the inter-fre2uency handover
fre2uently happens between two co-locatin! cells.
Figure 65. 5llustration of the ping pong problem in the inter+fre=uency cell reselection
Solution
The possible solution is to set the .measure: Sele)tion Resele)tion .uality
Measure to be 7, i.e. 2(523>Ec)7o.
'ote%
1. :ierarchical cell structure feature can push the traffic from one carrier to
another carrier, even thou!h the 2uality of the pilot channel at that carrier is
not the best. :owever, this feature is not available in the ,7 3TRA'.
7. If the loads in the co-sitin! carriers are avera!ely distributed, e.!. employin!
load sharin! feature bein! in ,9 3TRA', the downlink interference levels of
the carriers at a location, i.e. /0 R$$I will also be similar. Therefore, this
solution is no lon!er work.
8.!6 'any outgoing inter-R(& cell re-selections
8.!6.1 Symptoms
rom the drive test, it can be observed that the 3" fre2uently re-selects to the
B$# cells.
8.!6.2 Reason 1 high downlink interference
The possible reason causin! many inter-RAT cell re-selections is same as that for
the pilot channel failure due to hi!h downlink interference. ,lease read chapter
=.1.
1)d*. :owever, the 2(523>Ec)7o tri!!ered to determine for inter-RAT cell reselection
is .(ualMin: Minimum (uality or sele)tion:resele)tion G S R+3
Sear)h.
8.!6.! Reason 2 out of pilot co$erage due to 8uality
The possible reason causin! many inter-RAT cell re-selections is same as that for
the pilot channel failure due to out of pilot covera!e. ,lease read chapter =.7.
1)d*. :owever, the 2(523>Ec)7o tri!!ered to determine for inter-RAT cell reselection
is .(ualMin: Minimum (uality or sele)tion:resele)tion G S R+3
Sear)h.
8.!6." Reason ! improper inter-R(& search parameter setting
If the inter-RAT search parameter S R+3 Sear)h is set to too hi!h positive,
unnecessary inter-RAT cell re-selection may be caused.
Solution
The en!ineer should modify the S R+3 Sear)h to a suitable value, e.!. Dd*.
8."1 'any location area or routing area updates
8."1.1 Symptoms
rom the drive test, many location area @0AA and routin! area @RAA updates, i.e.
, are found from observin! the 09 messa!e in
T"#$.
8."1.2 Reason 1 - short periodic 1( or R( updating timers
If the timer for periodic 0A or RA updatin!, i.e. 3"!1! or 3""1! respectively, is
too short, many location area or routin! area updates mi!ht be caused.
Solution
The en!ineer should ad<ust the 3"!1! and 3""1! timers to suitable values, e.!. 1
hour.
8."1.! Reason 2 - many location areas or routing areas
If there are too many location areas or routin! areas in a network, many location
area or routin! area updates mi!ht be caused.
Solution
The en!ineer should check if there are too many location areas or routin! areas in
a network, e.!. drivin! alon! the RA or 0A borders, and mer!es the unnecessary
location or routin! areas into other.
8."2 (#normal disconnection in inter-R(& hando$er or cell
change
8."2.1 Symptoms
- 4RR+ connection release5 09 messa!e with unspecified reason is sent out
from the 3TRA' then
- 4RR+ connection release complete5 09 messa!e is sent to the 3TRA' then
- +onnection drops.
These symptoms can be applied to the radio connection in the connected mode,
i.e. either cellH/+: or cellHA+:.
8."2.2 Reason 1 relocation o$erall :for hando$er; timer or cell
change timer e0pires
In inter-RAT handover, when the tar!et B$# cell has allocated resources for new
radio link, the 3TRA' sends an 4handover from 3TRA' command5 messa!e to
the 3" and starts the relocation overall timer. If the tar!et B$# cell reports the
3TRA' that the inter-RAT handover is completed, the timer will be stopped.
Etherwise, the 3TRA' waits for the timer e1piry. After the e1piry, the 3TRA'
orders the 3" to release the connection because the 3TRA' considers the
connection is no !ood.
In inter-RAT cell chan!e, when the 3TRA' decides to e1ecute inter-RAT cell
chan!e function, the 3TRA' sends a 4cell chan!e order from 3TRA'5 messa!e
to the 3" and starts the cell chan!e timer. If the core network initiates the Iu
release procedure, the timer will be stopped. Etherwise, the 3TRA' waits for the
timer e1piry. After the e1piry, the 3TRA' orders the 3" to release the
connection because the 3TRA' considers the connection is no !ood.
Solution
The possible reasons are%
- +onnection setup fails. >hen the 3" is campin! on the B$# network. The
en!ineer should optimi6e the B$# network.
- The 3" does not receive the 4handover from 3TRA' command5 or 4cell
chan!e order from 3TRA'5 even thou!h the 3TRA' sends it out. The
reason is because of poor 2uality in the downlink. ,lease read chapter =.)1.
8."2.! Reason 2 measurement control failure
In inter-RAT handover or cell chan!e, if the 3" sends a 4measurement control
failure5 messa!e to the 3TRA', the 3TRA' considers that the 3" will not
measure on ordered nei!hborin! cells and also not report on enterin! these cells.
To avoid the 3" runs into a cell and starts interferin!, the 3TRA' orders the 3"
to release the connection.
Solution
The reason is due to poor downlink 2uality. ,lease read chapter =.)1.
8."! 4ompressed mode function failure
8."!.1 Symptoms
3"TR%
- +ase 1% 3" sends event 7d measurement report to the 3TRA' to re2uest the
initiation of the compressed mode. :owever, the 3TRA' does not send
4,hysical channel reconfi!uration5 messa!e to the 3".
- +ase 7% The 3TRA' sends the 4,hysical channel reconfi!uration5 messa!e
to the 3"J however, it does not receive it.
- +ase 9% The 3" receives the 4,hysical channel reconfi!uration5 messa!eJ
however, it does not send out the 4,hysical channel reconfi!uration
complete5 messa!e to the 3TRA'.
- +ase D% The 3" sends out the 4,hysical channel reconfi!uration complete5 to
the 3TRA'J however, the 3TRA' does not receive it.
rom the performance statistics, followin! symptoms will be observed%
- +ase 8%
is very low.
- +ase )% hi!h number in pmCmStop.
8."!.2 Reason 1 compressed mode denied due to admission control
The possible reason for case 1 in the chapter =.D9.1 is because of lack of radio
resource. The admission control blocks the new access for compressed mode.
Solution
The admission policies for the new access in compressed mode are downlink T1
power, con!estion events and number of compressed mode user. The admission
control checks if
- The cell transmits too hi!h downlink T1 power.
- There are too many compressed mode users.
- There are too many con!estion events.
8."!.! Reason 2 poor downlink 8uality
The possible reason for case 7 in the chapter =.D9.1 is because of poor 2uality in
the downlink.
Solution
8."!." Reason ! compressed mode is not ready in the )/ side
The possible reason for case 9 in the chapter =.D9.1 is that the 3" is not ready to
prepare for activation of the compressed mode.
8."!., Reason " poor uplink 8uality
The reason for the case D in the chapter =.D9.1 is because of poor 2uality in the
uplink.
Solution
8."!.. Reason , failure in compressed mode attempt
>hen the 3TRA' sends out a 4,hysical channel reconfi!uration5 messa!e to the
3", the counter for number of compressed mode attempt is incremented one step.
>hen the 3TRA' receives a 4,hysical channel reconfi!uration complete5
messa!e from the 3", the counter for number of successful compressed mode is
incremented one step.
Therefore, the reasons for the case 8 in the chapter =.D9.1 are due to reasons 7, 9
and D. ,lease read chapters =.D9.9, =.D9.D and =.D9.8.
8."!.3 Reason . transmission gap pattern se8uences o$erlapping
The reason for the case ) in the chapter =.D9.1 is due to overlappin! of the
transmission !ap pattern se2uences.
>hen the 3" has received several compressed mode transmission !ap pattern
se2uences from the 3TRA' the confi!urations and several of these patterns are
simultaneously active, the 3" transmits 4,hysical channel reconfi!uration
failure5 messa!e to the 3TRA' if these se2uences are ille!al overlappin!. The
counter for number of compressed mode stops is incremented one step.
8."" Outgoing inter-R(& mo#ility e$ent detection failure
8."".1 Symptoms
3"TR%
- The 3" has started compressed mode to measure B$# cells and receives the
measurement control from the 3TRA' to indicate the B$# nei!hborin!
cells.
- Received "c&'o of the best servin! cell in active set is less than *CDM+
R+, threshold and
- A suitable B$# cell is found and received R$$I of that cell is lar!er than
GSM threshold "a and
- "vent 9a measurement report is not sent from the 3" to the 3TRA' or the
event 9a measurement report is sent to the 3TRA'J however, the 3TRA'
does not receive it.
'ote% The received R$$I of the suitable B$# cell can be measured by usin!
B$# carrier scanner or B$# T"#$.
8."".2 Reason 1 poor uplink 8uality
If the event 9a measurement report is sent to the 3TRA' but the 3TRA' does
not receive it, the possible reason for this case is because of poor 2uality in the
uplink.
Solution
8."".! Reason 2 missing +S' neigh#oring cell
The possible reason why the event detection fails is because of missin!
nei!hborin! cell window to check if the desired B$# cell is included in the inter-
RAT monitored set.
Solution
The direct solution is to add the desired cell into the nei!hborin! cell lists of the
cells in the active set. :owever, it should be noted that too many nei!hborin! cell
relationships mi!ht slow down the search for the B$# carriers.
8.""." Reason ! )&R(* co$erage rapidly fades
If the 3TRA' covera!e rapidly fades out, e.!. outdoor to indoor, the 3" mi!ht
not have enou!h time to move the B$# network. There are two possible cases%
- +ase 1% Avera!e B$# R$$I is hi!h.
Figure 66. 6*R.7 coverage rapidly fades and average @-4 R--5 is high.
- +ase 7% B$# R$$I rapidly increases.
Figure 6!. 6*R.7 coverage and @-4 R--5 rapidly fades and increases
respectively.
Solution 1
or both cases, the *CDM+ R+, threshold should be increased so that the
>+/#A RA' threshold is far away (1)d* and the event, i.e. event 9a, for the
inter-RAT handover or cell chan!e is tri!!ered early.
or case 7, the GSM threshold "a can be decreased or a positive value can be
!iven to the GSM )ell individual oset of the cell.
It should be noted that the *CDM+ R+, threshold and GSM threshold "a are
R'+ parameters. It means chan!in! them will affect whole R'+.
Solution !
the cell overlappin! area becomes lar!er. Ebviously, it is a cost and timeconsumin!
solution.
8.""., Reason " too long +S' neigh#oring list
Accordin! to 9B,, standardi6ation, the ma1imum number of 3" measured cells
is 97 for B$# fre2uencies. Therefore, if the total number of cells in the
preliminary B$# monitored subset are more than 97, the preliminary B$#
monitored subset has to be reduced. /ue to this reduction, the desired B$# cell
mi!ht be removed in the handover al!orithm and inter-RAT event detection mi!ht
be missed.
'ote%
- The cells in the preliminary B$# monitored subset are the cells taken from
the union of the B$# cells in the nei!hborin! B$# cell lists of the active set
cells. or e1ample @please see Table 17A, if there are three cells in the active
set ( cell A, cell * and cell +, the preliminary B$# monitored subset will
contain +ell !HA1, +ell !HA7, +ell !HA9, +ell !HAD, +ell !H*1, +ell !H*7,
+ell !H*9, +ell !H+1 and +ell !H+7.
*able 1&. .n example to illustrate the preliminary @-4 monitored subset
Active set cells B$# nei!hborin! cell lists
+ell A +ell !HA1, +ell !HA7, +ell !HA9, +ell !HAD
+ell * +ell !H*1, +ell !H*7, +ell !H*9
+ell + +ell !H+1, +ell !H+7
- Re!ardin! the reduction in the preliminary B$# monitored subset, the
handover al!orithm will select and then remove a number of cells from the
nei!hborin! B$# cell lists in each active set cell respectively in order to be
sure that each active set cell can contribute its nei!hborin! B$# cells for
monitorin!. The removed nei!hborin! B$# cells are normally at the bottom
of the nei!hborin! cell lists.
Solution 1
If there are too many unnecessary B$# handover relationships, the en!ineer
should carefully <ustify the usefulness of the B$# handover relationships and
remove the unnecessary ones.
and hi!h downlink interference may happen due to missin! B$# handover
relationship.
Solution !
The en!ineer should put the nei!hborin! B$# cells in a desired order so the
hi!hest ranked B$# cells appear in the top of the list.
8.", Outgoing inter-R(& hando$er function failure
8.",.1 Symptoms
3"TR%
- +ase 1% 4Relocation preparation failure5 RA'A, messa!e is sent to the $R'+
from the circuited switch core network.
- +ase 7% 4Relocation cancel5 RA'A, messa!e with cause value 4relocation
canceled @1.A5 is sent out to the circuited switch core network from the $R'+.
- +ase 9% 4Relocation cancel5 RA'A, messa!e with cause value 4TR"0E+prep
e1piry @9A5 is sent out to the circuited switch core network from the $-R'+.
- +ase D% The 3" sends 4:andover from 3TRA' failure5 messa!e to the
3TRA'.
- +ase 8% +onnection drops durin! the inter-RAT handover.
- +ase )% The successful ratio of inter-RAT handover to B$# per cell relation,
i.e. is very low.
- +ase F% The ratio of inter-RAT handover attempts to B$# where the 3"
returns to old channel per cell relation, i.e.
or
, is very hi!h.
- +ase =% The number of inter-RAT handover attempts to B$# where the
resource allocation in the B$# network fails per cell relation, i.e.
pmNo!tIr"t#oReso!rceAllocF"il, is very hi!h.
8.",.2 Reason 1 no resource a$aila#le in +S' network
The reason for case 1 in chapter =.D8.1 is because of no resource available in
B$# network.
Solution 1
The en!ineer should optimi6e the B$# network to lower the con!estion, e.!. add
more TRQs or re-dimension the B$# network. Ebviously, it is a cost- and timeconsumin!
solution because installation work is needed.
Solution !
The GSM amount proposal repeat can be increased so that the con!ested B$#
cell can be repeatedly attempted more. :owever, it will take lon!er inter-RAT
handover delay and increases the probability of havin! pilot channel failure.
It should be noted that the GSM amount proposal repeat is a R'+ parameter. It
means chan!in! it will affect the whole R'+.
8.",.! Reason 2 cannot fulfill the +S' re8uest
The reason for case 7 in chapter =.D8.1 is because the 3TRA' cannot fulfill the
re2uest from the B$# side.
The B$# tar!et system has allocated resources and sends a 4Relocation
command5 to the $-R'+. The command includes what RA*s that shall be
released. If the 3TRA' cannot fulfill the re2uest, a 4Relocation cancel5 RA'A,
messa!e with cause value 4relocation canceled @1.A5 is sent out to the circuited
switch core network from the $-R'+. It should be noted that the connection is
still kept.
8.",." Reason ! no response from core network
The reason for case 9 in chapter =.D8.1 is because of no response from the core
network.
After sendin! 4Relocation re2uired5 messa!e to core network to re2uest B$#
resource, the $-R'+ starts a timer called TR"0E+prep. >hen there is no response
from the core network and the timer e1pires, a 4Relocation cancel5 RA'A,
messa!e with cause value 4TR"0E+prep e1piry @9A5 is sent out to the core network
from the $-R'+. It should be noted that the connection is still kept.
8.",., Reason " failure to access +S' cell
The reason for cases D and F in chapter =.D8.1 is because of failure to access the
B$# cell. After the 3" receives 4:andover from 3TRA' command5, it tries to
access the tar!et B$# cell. If the attempt fails, the 3" sends a 4:andover from
3TRA' failure5 messa!e to the 3TRA' and returns to the old channel.
If the is very hi!h, the reason of
the failure is due to failure of connection setup in the B$# side.
Solution
The en!ineer should optimi6e the B$# network to make the B$# connection
setup successful.
8.",.. Reason , a#normal disconnection in inter-R(& hando$er
The reason for case 8 in chapter =.D8.1 is because of abnormal disconnection in
the inter-RAT handover. ,lease read chapter =.D7.
8.",.3 Reason . failure of outgoing inter-R(& hando$er attempt
>hen the 3TRA' sends out a 4:andover from 3TRA' command5 messa!e to
the 3", counter for number of inter-RAT handover attempts to B$# is
incremented one step. >hen the 3TRA' receives an 4Iu release command5
messa!e from the core network, counter for number of successful inter-RAT
handovers to B$# is incremented one step.
Therefore, the reasons for case ) in chapter =.D8.1 are same as the reasons for
cases D and 8 in chapter =.D8.1. ,lease read chapters =.D8.8 and =.D8.).
'ote%
is a ratio to reflect
number of inter-RAT handover attempts to B$# where 3" fails to connect to
B$# and cannot return to old channel.
8.",.8 Reason 3 no +S' resources or no response from core
network
The counter pmNo!tIr"t#oReso!rceAllocF"il is incremented one step when
the core network sends out a 4Relocation preparation failure5 or called TR"0E+prep
e1pires.
Therefore the reasons for case = in chapter =.D8.1 are same as the reasons for
cases 1 and 9 in chapter =.D8.1. ,lease read chapters =.D8.7 and =.D8.D.
8.". Outgoing inter-R(& cell change function failure
8."..1 Symptoms
3"TR%
- +ase 1% The 3" sends 4+ell chan!e order from 3TRA' failure5 messa!e to
the 3TRA'.
- +ase 7% +onnection drops durin! the inter-RAT cell chan!e.
- +ase 9% The ratio of inter-RAT cell chan!e attempts to B$# where the 3"
on dedicated channel returns to old channel per cell relation, i.e.
, is very hi!h.
8."..2 Reason 1 failure to camp on +S' cell
The reason for cases 1 and 9 in chapter =.D).1 is because of failure to camp on the
B$# cell. After the 3" receives 4+ell chan!e order from 3TRA'5, it tries to
camp on the tar!et B$# cell. If the campin! fails, the 3" sends a 4+ell chan!e
order from 3TRA' failure5 messa!e to the 3TRA' and returns to the old
channel. The reason of the failure is due to failure of connection setup in the
B$# side.
Solution
The en!ineer should optimi6e the B$# network to make the B$# connection
setup successful.
8."..! Reason 2 a#normal disconnection in inter-R(& cell change
The reason for case 7 in chapter =.D).1 is because of abnormal disconnection in
the inter-RAT cell chan!e. ,lease read chapter =.D7.
8."3 'any outgoing inter-R(& mo#ility - high downlink
interference
8."3.1 Symptoms
rom the performance statistics, many inter-RAT handover or cell chan!e, i.e.
hi!h number in pmNo!tIr"t#oS!ccess, pmNo!tIr"t#oAtt,
pmNo!tIr"tCcS!ccess and pmNo!tIr"tCcAtt are found.
The load in the B$# network increases due to too many inter-RAT handover and
cell chan!e.
- Received "c&'o of the pilot channel is less than *CDM+ R+, threshold
and
- Received R$+, of the pilot channel is hi!h enou!h to maintain the
connection, e.!. - -1..d*m and
- /0 R$$I is very hi!h and
- Inter-RAT handover or cell chan!e finally happens.
8."3.2 Reasons and solutions
The reasons and solutions in this case are same as these for the pilot channel
failure due to hi!h downlink interference. ,lease read chapter =.1.
1)d*. :owever, the 2(523>Ec)7o tri!!ered to determine the inter-RAT
mobility is *CDM+ R+, threshold.
8.,1 RR4 connection esta#lishment failure
A successful call establishment contains
- $uccessful establishment of an RR+ connection RR+ connection "stablishment provides the
ability to establish an RR+ connection, which is a lo!ical connection, between the 3" and
3TRA' at 09. A radio connection comprises the connection between a 3" and 3TRA'
includin! all the resources, i.e., 01, 07 and 09. The 3" makes the initial access to 3TRA' and
re2uests for a RR+ connection. Radio resources are allocated and a RR+ connection is
established between 3" and 3TRA'.
- $uccessful establishment of a si!nalin! connection
$i!nalin! connection establishment provides the ability to establish a si!nalin! connection
between the 3" and a core network, i.e. one si!nalin! connection towards the +$ core network
and one towards the ,$ core network. "stablishment of the first si!nalin! connection is initiated
by the 3" as soon as the RR+ connection is successfully established. An Iu control plane
connection is then established between 3TRA' and the core network. >hen a si!nalin!
connection is already established towards one core network, the 3" can initiate establishment of
a second si!nalin! connection to the other core network at any time. An Iu control plane
connection is then established between 3TRA' and that other core network.
- $uccessful establishment of a RA*
RA* establishment provides the ability to establish of a user plane data stream within 3TRA'.
In this chapter, the issues in RR+ connection establishment will be investi!ated.
8.,1.1 Symptoms
rom the drive test, followin! symptoms will be observed by usin! T"#$ and 3"TR%
- +ase 1% The 3" does not send out 4RR+ connection re2uest5 messa!e to the 3TRA'.
- +ase 7% The 3" receives 4RR+ connection setup5 messa!e and starts the transmission.
:owever, tar!et R*$ does not send out 4Radio link restore indicator5 to the R'+.
- +ase 9% The 3" receives 4RR+ connection setup5 messa!e and starts the transmission.
:owever, the 3" does not send out 4RR+ connection setup complete5 messa!e to the 3TRA'.
- +ase D% The 3" receives 4RR+ connection setup5 messa!e and starts the transmission. After a
while, the 3" sends out 4RR+ connection setup complete5 messa!e to the 3TRA'J however, the
RR+ connection establishment fails.
- +ase 8% The 3" receives 4RR+ connection re<ect5 messa!e with cause value 4con!estion5.
- +ase )% The 3" receives 4RR+ connection re<ect5 messa!e with cause value 4unspecified5.
- +ase F% The 3" repeatedly sends 4RR+ connection re2uest5 messa!es but the number of
transmission is less than ,"77 G 1 times.
- +ase =% The 3" does not receive any messa!e from the 3TRA'.
8.,1.2 Reason 1 failure in RR4 connection release
The reason for case 1 in chapter =.81.1 probably is because of failure in the RR+ connection
release. If a RR+ connection had been established, it is impossible to connect more than one RR+
connection unless the previous RR+ connection is released.
8.,1.! Reason 2 uplink dedicated radio link synchronization failure
or case 7 in chapter =.81.1, the 3" receives 4RR+ connection setup5 messa!e and starts the
transmission. :owever, tar!et R*$ does not send out 4Radio link restore indicator5 to the R'+.
It implies the 3" and 3TRA' are tryin! to synchroni6e each other but uplink is not
synchroni6ed.
Solution
,lease read chapter =.89.
8.,1." Reason ! downlink dedicated radio link synchronization failure
or case 9 in chapter =.81.1, the 3" receives 4RR+ connection setup5 messa!e and starts the
transmission. :owever, the 3" does not send out 4RR+ connection setup complete5 messa!e to
the 3TRA'. It implies the 3" and 3TRA' are tryin! to synchroni6e each other but downlink is
not synchroni6ed.
Solution
,lease read chapter =.8D.
8.,1., Reason " poor 8uality in uplink
or case D in chapter =.81.1, the 3" sends out 4RR+ connection setup complete5 messa!e to the
3TRA' via the established dedicated channelJ however, the RR+ connection establishment still
fails. It implies the 3TRA' does not receive the 4RR+ connection setup complete5 messa!e.
Therefore, the possible reason is because of poor 2uality in uplink.
Solution
8.,1.. Reason , connection re=ected #y load super$ision function
At each re2uest for establishment of a new RR+ connection, it is checked that this is accepted by
the processor load supervision function in the R'+. In case of re<ect from the load supervision
function, the $-R+: will send 4RR+ connection re<ect5 messa!e to the 3" and the procedure
ends. The cause value of the messa!e is 4con!estion5.
Solution
The en!ineer should re-dimension the capacity of the e1istin! R'+. If need be, more processors
are added.
8.,1.3 Reason . dedicated radio link setup failure
The reason for case ) in chapter =.81.1 is because of failure in the dedicated radio link setup.
Solution
8.,1.8 Reason 3 no suita#le cell
'ormally, the 3" will repeatedly transmit 4RR+ connection re2uest5 in or more than ,"77 G 1
times if the 3" does not receive any 4RR+ connection setup5 messa!e. Therefore, if the number
of transmissions is less than ,"77 G 1 times, it implies 4no suitable cell5 event happens durin!
RR+ connection establishment.
Solution
'ote%
1. 9B,, 78.991-?.9.F.., which the ,7.1 3TRA' follows, specifies that the number of the
retransmissions of the 4RR+ connection re2uest5 should be ,"77 if the 3" does not receive the
4RR+ connection setup5 messa!e.
7. The number of retransmissions can be !reater than ,"77 if the 3" has received 4RR+
connection re<ect5 messa!e or re-select another cell. In this case, the 3" counter ?9.. has been
reset.
8.,1.6 Reason 8 S-44P45 :9(45; failure
The 4RR+ connection re<ect5 or 4RR+ connection setup5 messa!e is transmitted via $-++,+:
@A+:A. If the 3" does not receive any messa!e from the 3TRA', the possible reason is
because of $-++,+: @A+:A failure.
In the ,7 3TRA', power of the $-++,+: is not automatically modified with the variation of the
network load. or that reason, the en!ineer has to carefully plan enou!h $-++,+: power to
fulfill the assumed downlink load. If mistake is done durin! the power calculation or the real load
of the downlink carrier is lar!er than the assumed level, $-++,+: mi!ht not cover the whole
cell.
Solution
8.,1.17 Reason 6 random access procedure pro#lem
*esides reasons 1 to =, the last possible reason is because of the problem in the random access
procedure. /ue to this issue, the 3TRA' does not receive the 4RR+ connection re2uest5
messa!e.
Solution
,lease read chapter =.9).
8.,2 2edicated radio link setup failure
8.,2.1 Symptoms
rom the drive test, followin! symptoms will be observed by usin! 3"TR%
- +ase 1% The R'+ does not send 4Radio link setup re2uest5 '*A, messa!e to the tar!et R*$.
- +ase 7% The R'+ sends 4Radio link setup re2uest5 to and receives 4Radio link setup response5
from the tar!et R*$, respectively.
- +ase 9% The R'+ receives 4Radio link setup failure5 '*A, messa!e from the tar!et R*$.
- +ase D% The R'+ does not receive any messa!e after sendin! the 4Radio link setup re2uest5.
8.,2.2 Reason 1 #locking due to admission control
The reason for case 1 in chapter =.87.1 is because of cell blockin! due to admission control.
8.,2.! Reason 2 ((12 esta#lishment failure
The reason for case 7 in chapter =.87.1 is because of failure in the AA07 establishment.
8.,2." Reason ! no R>S hardware resources
The reason for case 9 in chapter =.87.1 is because of lack of R*$ hardware.
Solution
8.,2., Reason " no R>S response
The reason for case D in chapter =.87.1 is because of no response from the tar!et R*$.
Solution
The en!ineer should check if%
1. The tar!et R*$ is blocked.
7. Transport network has problem.
8.,! )plink dedicated radio link synchronization failurecall
setup
8.,!.1 Symptoms
rom the drive test, it is found that the 3" receives 4RR+ connection setup5 messa!e and starts
the transmission. :owever, tar!et R*$ does not send out 4Radio link restore indicator5 to the
R'+.
- +ase 1% The transmission power of the 3" reaches to ma1imum allowed value.
- +ase 7% The transmission power of the 3" does not reach to the ma1imum allowed value.
8.,!.2 Reason 1 uplink and pilot co$erage im#alance
or case 1, please read chapter =.8.
8.,!.! Reason 2 improper cell re-selection offset setting
If the cell re-selection offset, i.e. . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC RSCP1 of
the camped cell is !iven too lar!e positive value, the new idle mode covera!e may be lar!er than
the ma1imum allowed 30 /,+: covera!e, i.e. the location where 3" transmits ma1imum
allowed 3" T1 power. ,lease see i!ure );. Therefore, synchroni6ation in uplink dedicated radio
link durin! connection setup fails.
Solution
The . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC RSCP1 should be chan!ed to not too
lar!e positive.
'ote% +han!in! cell re-selection offset may cause different borders in idle and dedicated modes.
To solve this problem, the same value should be !iven to the individual offset Cell individual
oset and cell offset . /set 1sn 0CPIC
RSCP1 or . /set !sn 0CPIC RSCP1.
8.,!." Reason ! incorrect initial power calculation for dedicated
channel
or case 7, it is probably that initial power for uplink /,++: is calculated incorrectly.
The initial uplink /,++: power is calculated in the 3" accordin! to%
'ote% The unit of the '% initial SIR target is in ..1d*.
Therefore,
- If the '% initial SIR target is set too low, the initial 30 /,++: power becomes not lar!e
enou!h for the establishment.
- If the '% DPCCH power oset is set too low, the initial 30 /,++: power becomes not lar!e
enou!h for the establishment.
- If the operator employs T#A and inputs wron! parameters for the antenna confi!uration, the
estimated downlink pathloss or the measured total uplink interference will be wron!. As a result,
the calculation for the initial 30 /,++: power becomes incorrect.
Solution 1
The en!ineer should ad<ust the '% initial SIR target and '% DPCCH power oset to suitable
values.
'ote%
- '% initial SIR target is a R* connection parameter. +han!in! it will affect all connections
with the same R* type in a R'+.
- '% DPCCH power oset is a R'+ parameter. +han!in! it will affect the whole R'+.
Solution !
The uni2ue way to solve T#A problem is to correct antenna confi!uration parameters as the real
ones. :owever, it is a time consumin! solution and difficult to measure the accurate feeder loss,
T#A !ain value, etc.
8.," 2ownlink dedicated radio link synchronization failurecall
setup
8.,".1 Symptoms
rom the drive test, it is found that the 3" receives 4RR+ connection setup5 messa!e and starts
the transmission. :owever, the 3" does not send out 4RR+ connection setup complete5 messa!e
to the 3TRA'.
8.,".2 Reason 1 intended 21 2P45 and pilot co$erage im#alance
The downlink covera!e of /,+: in some particular R* types may not be planned to balance
with the pilot covera!e. Therefore, synchroni6ation in downlink dedicated radio link durin!
connection setup will fail.
Solution
or 4intended /0 /,+: and pilot covera!e imbalance5, the en!ineer should consider if it is
necessary to provide full covera!e for that /0 /,+: type or not.
If need be, the ma1imum allowed /0 code power #a1 /0 ,ower should be increased.
'ote% The Ma$ D% Power is a parameter per R* connection. +han!in! it will affect all
connections in same R* in a R'+.
8.,".! Reason 2 not intended 21 2P45 and pilot co$erage im#alance
,lease read chapter =.).
8.,"." Reason ! improper cell re-selection offset setting
If the cell re-selection offset, i.e. . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC RSCP1 of
the camped cell is !iven too lar!e positive value, the new idle mode covera!e may be lar!er than
the ma1imum allowed /0 /,+: covera!e, i.e. the location where the R*$ transmits ma1imum
allowed /0 code T1 power.
,lease see i!ure );. Therefore, synchroni6ation in downlink dedicated radio link durin!
connection setup fails.
Solution
The . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC RSCP1 should be chan!ed to not too
lar!e positive.
'ote% +han!in! cell re-selection offset may cause different borders in idle and dedicated modes.
To solve this problem, the same value should be !iven to the individual offset Cell individual
oset and cell offset . /set 1sn 0CPIC RSCP1 or . /set !sn 0CPIC RSCP1
8.,"., Reason " too low initial downlink S-R target
If the initial $IR tar!et of the downlink channel Initial D% SIR target is set too low,
synchroni6ation in downlink dedicated radio link durin! connection setup will fail.
Solution
or 4too low initial downlink $IR tar!et5, the en!ineer should increase the Initial D% SIR target.
'ote% The Initial D% SIR target is a parameter per R* connection. +han!in! will affect all
connections in same R* in a R'+.
8.,".. Reason , improper #ackoff constant setting
The C-a)?o: Constant to -a)?o the open loop estimate is a constant to back off the Epen
0oop ,ower +ontrol estimate to a conservative startin! point.
:i!her C-a)?o: Constant to -a)?o the open loop estimate means the R*$ will transmit
hi!her initial downlink /,+: power. En the other hand, if it is set improperly, synchroni6ation in
downlink dedicated radio link durin! connection setup will fail because of not enou!h initial
downlink /,+: power.
Solution
The en!ineer can increase the C-a)?o: Constant to -a)?o the open loop estimate to allow
hi!her initial downlink /,+: power.
'ote% The C-a)?o: Constant to -a)?o the open loop estimate is a R'+ parameter.
+han!in! it will affect the whole R'+.
8.,, Signaling connection esta#lishment failure
8.,,.1 Symptoms
rom the drive test, followin! symptoms will be found by usin! T"#$ and
3"TR%
- +ase 1% The 3" sends out 4Initial direct transfer5 messa!e to the 3TRA'J however, the
3TRA' does not receive it.
- +ase 7% The 3TRA' sends out 4/ownlink direct transfer5 messa!e to the
3"J however, the 3" does not receive it.
8.,,.2 Reason 1 poor 8uality in uplink
The reason for case 1 in chapter =.88.1 is because of poor 2uality in uplink.
Solution
,leas read chapter =.)..
8.,,.! Reason 2 poor 8uality in downlink
The reason for case 7 in chapter =.88.1 is because of poor 2uality in downlink.
Solution
8.,3 Radio #earer esta#lishment failure SR> - 245 to
R(> - 245
8.,3.1 Symptoms
*y usin! T"#$ and 3"TR, followin! symptoms are found%
- +ase 1% The 3TRA' does not send 4Radio bearer setup5 messa!e to the 3".
- +ase 7% The 3TRA' sends out 4Radio bearer setup5 messa!e to the 3"J however, the 3" does
not receive it.
- +ase 9% The 3" receives 4Radio bearer setup5J however, it does not send out 4Radio bearer
setup complete5 to the 3TRA'.
- +ase D% The 3" sends out 4Radio bearer setup complete5 messa!e to the
3TRA'J however, the 3TRA' does not receive it.
8.,3.2 Reason 1 cell #locking
The cell does not have enou!h radio or hardware resources for the new radio bearer. Therefore,
the 3TRA' does not send 4Radio bearer setup5 messa!e to the 3".
Solution
8.,3.! Reason 2 poor 8uality in downlink
The reason for case 7 in chapter =.8F.1 is because of poor 2uality in downlink.
Solution
8.,3." Reason ! non-radio issues
The reasons for case 9 in chapter =.8F.1 are because of some non-radio issues%
- AA07 establishment between R'+ and the core network is not ER.
- The 3" has resource problems.
The reason for case D in chapter =.8F.1 is because of poor 2uality in uplink.
Solution
8.,8 Radio #earer esta#lishment failure SR> - 245 to
R(> - R(%9(
8.,8.1 Symptoms
- +ase 1% The 3TRA' does not send 4Radio bearer setup5 messa!e to the 3" on /+:.
- +ase 7% The 3TRA' sends out 4Radio bearer setup5 messa!e to the 3"J however, the 3" does
not receive it.
- +ase 9% The 3" receives 4Radio bearer setup5J however, it does not send out 4Radio bearer
setup complete5 to the 3TRA' on the RA+:.
- +ase D% The 3" sends out 4Radio bearer setup complete5 messa!e to the
3TRA' on the RA+:J however, the 3TRA' does not receive it.
- +ase 8% The 3TRA' receives the 4Radio bearer setup complete5 messa!eJ however, the radio
bearer establishment still fails.
8.,8.2 Reason 1 cell #locking
The cell does not have enou!h radio or hardware resources for the new radio bearer. Therefore,
the 3TRA' does not send 4Radio bearer setup5 messa!e to 3".
8.,8.! Reason 2 poor 8uality in downlink
The reason for case 7 in chapter =.8=.1 is because of poor 2uality in downlink.
Solution
8.,8." Reason ! cell update procedure failure
The reason for case 9 in chapter =.8=.1 is because of cell update procedure failure.
Solution
The reason for case D in chapter =.8=.1 is because of poor 2uality in uplink.
Solution
8.,8.. Reason , pro#lems occur during release of resources
The reasons for case 8 in chapter =.8=.1 are because problems occur durin! release of resources.
The resources could be%
- AA07
- :ardware
- Radio access bearer
- "tc.
8.,6 Radio #earer esta#lishment failure R(> - R(%9( to
R(> 245
8.,6.1 Symptoms
- +ase 1% The 3TRA' does not send 4Radio bearer release5 messa!e to the 3" on A+:.
- +ase 7% The 3TRA' sends 4Radio bearer release5 messa!e to the 3" on
A+:J however, the 3" does not receive it.
- +ase 9% The 3" receives 4Radio bearer release5 messa!e. En the other hand, the R*$ does not
send out 4Radio link restore indicator5 to the R'+.
- +ase D% The 3" receives 4Radio bearer release5 messa!eJ however, it does not send out 4Radio
bearer release complete5 messa!e on /+: to the 3TRA'.
- +ase 8% The 3" sends 4Radio bearer release complete5 messa!e on /+: to the 3TRA'J
however, the 3TRA' does not receive it.
8.,6.2 Reason 1 dedicated radio link setup failure
The reason for case 1 in chapter =.8;.1 is because of dedicated radio setup failure.
Solution
8.,6.! Reason 2 S-44P45 :9(45; failure
If the 3TRA' sends 4Radio bearer release5 messa!e to the 3" on A+:J however, the 3" does
not receive it, the possible reason is because of $-++,+: @A+:A failure.
In the ,7 3TRA', power of the $-++,+: is not automatically modified with the variation of the
network load. or that reason, the en!ineer has to carefully plan enou!h $-++,+: power to
fulfill the assumed downlink load. If mistake is done durin! the power calculation or the real load
of the downlink carrier is lar!er than the assumed level, $-++,+: mi!ht not cover the whole
cell.
Solution
8.,6." Reason ! uplink dedicated radio link synchronization failure
The reason for case 9 in chapter =.8;.1 is because of uplink dedicated radio link synchroni6ation
failure.
Solution
8.,6., Reason " downlink dedicated radio link synchronization failure
The reason for case D in chapter =.8;.1 is because of downlink dedicated radio link
synchroni6ation failure.
Solution
,lease read chapter =.8D.
8.,6.. Reason , poor 8uality in uplink
The reason for case D in chapter =.8;.1 is because of poor 2uality in uplink.
Solution
8..7 Poor 8uality in uplink
8..7.1 Symptoms
rom the drive test, it is found that hi!h *0"R occurs on the uplink. The measurement can be
obtained by usin! 3"TR.
8..7.2 Reason 1 pilot channel failure
If the received "c&'o of the best servin! pilot in active set is close to or less than
1)d*, it implies the pilot channel almost fails and the performance of channel estimation for the
downlink dedication channel becomes very poor.
/ue to poorer downlink 2ualities, the 3" will stop transmittin!, i.e. ,3" = ., and the 2uality on
uplink conse2uently becomes poor.
Solution
,lease read chapters =.1, =.7 and =.9 for the solutions.
8..7.! Reason 2 poor 8uality in downlink
If the 2uality on downlink is very poor, the 3" will stop transmittin! i.e. ,3" = ., and the 2uality
on uplink conse2uently becomes poor.
urthermore, poor 2uality in downlink mi!ht cause errors on the T,+. If the 3" follows the
wron! T,+ pattern to ad<ust its transmission power, the uplink 2uality becomes poor.
Solution
8..7." Reason ! - uplink and pilot co$erage im#alance
If the 3" has transmitted ma1imum allowed 3" transmission power, i.e. ,3" = ma1imum, the
reason of causin! poor 2uality in uplink is because of uplink and pilot channel imbalance.
Solution
8..7., Reason " insufficient recei$ed )1 2P45 power
The possible reason of causin! poor 2uality in uplink is because of insufficient received 30
/,+: power.
Solution
,lease read chapter =.;.
8..7.. Reason , rapidly changing radio en$ironment
The possible reason of causin! poor 2uality in uplink is because of rapidly chan!in! radio
environment.
Solution
,lease read chapter =.1..
8..1 Poor 8uality in downlink
8..1.1 Symptoms
rom the drive test, it is found that hi!h *0"R occurs on the downlink. The measurement can be
obtained by usin! 3"TR or T"#$.
8..1.2 Reason 1 pilot channel failure
If the received "c&'o of the best servin! pilot in active set is close to or less than (
1)d*, it implies the pilot channel almost fails and the performance of channel estimation for the
downlink dedication channel becomes very poor and the downlink 2uality conse2uently becomes
poor.
Solution
,lease read chapters =.1, =.7 and =.9 for solutions.
8..1.! Reason 2 transmitting ma0 21 code power
The reason of havin! poor downlink 2uality may be because the 3" has reached the downlink
dedicated channel border.
Solution
,lease read chapters =.) and =.F for solutions.
8..1." Reason ! poor 8uality in uplink
,oor 2uality in uplink mi!ht cause errors on the T,+. If the R*$ follows the wron! T,+ pattern
to ad<ust its transmission power, the downlink 2uality becomes poor.
Solution
8..1., Reason " rapidly changing radio en$ironment
The possible reason of causin! poor 2uality in downlink is because of rapidly chan!in! radio
environment.
Solution
,lease read chapter =.1..

Optimization Issues Ericsson UMTS

Uploaded by

Copyright:

Available Formats

Optimization Issues Ericsson UMTS

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Optimization Issues Ericsson UMTS

Uploaded by

Copyright:

Available Formats

8 Optimization issues

8.1 Pilot channel failure - high downlink interference

You might also like