Drive Test Analysis

1 Nokia Siemens Networks Presentation / Author / Date
For internal use

3G RANOP 1
Module 4 Drive Test Analysis
For internal use
Note for teacher
A set of example traces (Nemo) with comments can be used to
demonstrate failures.
Quickplace link:
http://qp2.connectin.nokia.com/Quick!lace/npcommunit"qp/!ae#ibra
r"$22%&'()**+$,A(-.nsf/h./ndex/2'0$'-,0,1&)2&',$22%02$A**1
*-,1&/34pen-ocument56esort-escendin78+
For internal use
3G RAN Optimization
RF
Optimisation
and Neighbour
Planning
9 R F optim isation
9 N e !ite "ntegration
9 N eighbour plan
optim isation
!ignalling
Flos
9 R R# $stablishm ent
9 R A % $stablishm ent
9!& O
9"!& O
Drive Test
Analysis
9 D rive !urvey A nalysis
9 !ystem Per'orm an(e
( ) RR# and RA% phases
#luster
Preparatio
n
9 # luster health (he()s
9 Param eter (onsisten(y
(he()
9 N eighbour list
veri'i(ation
9 * plin) inter'eren(e as a
problem indi(ator
- "nter !ystem
+or)ing and
Optimisation
9 < > 3G ,G # ellresele(tion
9 N eighbour Planning
9 & andover Pro(ess and
(om pressed m ode
9 3G "!& O servi(e analysis
( ) AMR and P!
9 G !M "!& O O ptim isation
Parameter
Optimisati
on
9 * se o' Param eters to
optim ise net or)
per'orm an(e
&!DPA
Optimisati
on
9 & !D PA A ir "nter'a(e
# apa(ity
9 & !D PA D rive Test - P"s and
Param eter O ptim isation
For internal use
Ob.e(tives
:- A'ter this module the delegate shall be able to
9 *nderstand the di''erent elements re/uired 'or an
e''e(tive drive test program
9 *nderstand ho all aspe(ts o' drive data (an be used to
obtain a vie o' netor) per'orman(e
9 ( ) - ( ) Di''erentiate beteen genuine RF and non genuine system
'ailures
9 *nderstand the -P"s that (an be obtained
9
For internal use
Drive Test rocess
#luster
Preparation
9 De'ine #luster
9 De'ine Drive Route
Data Analysis
9 0evels o'
Reporting
9 Failure Type
%rea)don
9 Post Pro(essing
'un(tionality
9 -P" Gap Analysis
Data Preparation
and #olle(tion
9 !ervi(es to test
9 -P" de'initions
9$/uipment
9$1(lusions
9 #all Patterns
9 O!! Alarms
9 Netor) !tats
Trouble
!hooting
9 Root #ause
Analysis
9 #orre(tive
A(tion
9 Further data
( ) logging maybe
9 Ne !ite "ntegration
9 RF Tuning
9 Netor) Optimisation
9 ( ) Netor) %en(hmar)ing Golden Routes
+hy Drive the netor)2
For internal use
Drive Test Analysis Test !"uipment
Nemo 4utdoor :ulti with ;2<s= >canner and ?!>
0and * nit
N 3 M P
FTP
!erver
9 "m portant to drive ith # alllogging
e/uipm ent and 3G and ,G s(anner in the
sam e vehi(le
For internal use
Drive Test Analysis Test !"uipment#
$canner
@he purpose of usin the 6' scanner is to be able to scan and measure
all used carriers/cells and their correspondin -# scramblin codes.
-ifferent >cramblin codes

$!$/A 6>$! Balue (d(m)
$!/$A 2cNo Balue (d()
For internal use
Drive Test Analysis Test !"uipment# %!
@he >canner measures all >$s= whereas the ;2 onl" measures >$ sinals from
the cells that the s"stem has informed/ordered the ;2 throuh the ($A
(neihbour list) or Bia the Cmeasurement controlD messae.
Nokia Siemens Networks Presentation / Author / Date
For internal use
Drive Test Analysis &all atterns
AMR #all
( ) Mobile Originated #alls MO#
9 , min (alls
9 34 se( idle
9 ( / ) *$ in Dual mode ,G 3G
( ) Mobile Terminated #alls MT#
9 , min (alls
9 34 se( idle
9 ( / ) *$ in dual mode ,G 3G
9 . $nough (all samples have to be made to ma)e the measurement statisti(ally valid
9 - % "n a 54 (all sample one dropped (all ill (ause a (hange in per'orman(e o' ,
9 - . % "n a 544 (all sample one dropped (all ill (ause a (hange in per'orman(e o' 4 ,
9 #all length should be de'ined at the beginning
9 +e (an use di''erent (all testing patterns 'or di''erent optimisation te(hni/ues
9 ( ) !hort #alls 'or #alls setup per'orman(e and delay
9 ( ) 0ong (alls 'or Drop (all per'orman(e and !&O per'orman(e
$ &all
?!6> Attach=
!-! $ontext ActiBation
'@! -ownload (8:( file)/'@! ;pload (%** E(
file)
!-! $ontext -eactiBation
?!6> -etach
Alternate download and upload with 1* sec
idle time
>ession is upload or download
;2 in -ual mode (2?/1?)
1! Nokia Siemens Networks Presentation / Author / Date
For internal use
Drive Test Analysis Definin' ()s
AMR Statistics
Threshold KPIs
"#$ Setu% time 5 se&
"#$ $$' (
RAW END USER
Event Count Ratio Count Ratio
$all Attem%ts 132 122
$all Setu% Su&&ess 'ate 1!8 81)8( 1!8 88)5(
$all Setu% Failure 'ate 24 18)2( 14 11)5(
Failures *ue to +ool ,+SF- 1! 41)7(
Failures *ue to $ore Pro.lem 1! 41)7( 1! 71)4(
Failure AS/ ,S0n&- Pro.lem 2 8)3( 2 14)3(
Failure *ue to 1ow $o2era3e 1e2els 2 8)3( 2 14)3(
$all Setu% Su&&ess an* Setu% +ime 4 5s 7 6)5( 7 6)5(
1on3 Setu% *ue to slow &ell resele&tion !)!( ! !)!(
1on3 Setu% *ue to &lash with 5nter'A+ resele&tion !)!( ! !)!(
1on3 Setu% *ue to /nknown ,sus%e&t /6- !)!( ! !)!(
1on3 Setu% *ue to /nknown !)!( ! !)!(
Average Call Setup Time 3.66 3.66
$all $om%letetion 'ate 1!5 7)2( 1!5 7)2(
$all Dro% 'ate 3 2)8( 3 2)8(
$all Dro% Poor 37 $o2era3e 1 33)3( 1 33)3(
$all Dro% on 7S" *ue to 5nter8eren&e 2 66)7( 2 66)7(
#2erall $all $om%letion 'ate 1!5 7)5( 1!5 86)1(
5S9# Attem%t 14 14
5S9# Su&&ess 14 1!!)!( 14 1!!)!(
5S9# Faile* ! !)!( ! !)!(
5S9# Faile* &ause %h0si&al &hannel 8ailure ! N/A ! N/A

PS Data :P5 'e%ort
KPI Thresholds Value Unit
Atta&h +ime 4 se&
PDP A&ti2ation +ime 2 se&
F+P Downloa* +hrou3h%ut 11! k.it/s
F+P /%loa* +hrou3h%ut 55 k.its
Signalling Statistics Count Success rate
Atta&h Attem%t 155
Atta&h Su&&ess 14! !)32(
Atta&h +ime more than threshol* 5 3)57(
Atta&h +ime less than threshol* 135 6)43(
Atta&h Faile* 15 )68(
Average Attach Setup Time 1.36
A&ti2ate PDP $onte;t Attem%t 124
A&ti2ate PDP $onte;t Su&&ess 124 1!!)!!(
A&ti2ation +ime more than threshol* 2 1)6!(
A&ti2ation +ime less than threshol* 123 8)4!(
A&ti2ate PDP $onte;t Faile* ! !)!!(
Average PDP Context Activation Time 0.96
F+P Downloa* Attem%ts 51
F+P Downloa* Su&&ess 48 4)12(
F+P Downloa* throu3h%ut more than threshol* 25 52)!8(
F+P Downloa* throu3h%ut less than threshol* 23 47)2(
Average FTP Downloa Throughput 10!.0"
F+P /%loa* Attem%ts 32
F+P /%loa* Su&&ess 3! 3)75(
F+P /%loa* throu3h%ut more than threshol* 2! 66)67(
F+P /%loa* throu3h%ut less than threshol* 1! 33)33(
Average FTP #ploa Throughput $$.$3
Data Transfer Cut-off Ratio 6.02%
PDP $onte;t Dro%%e* 4 3)23(
Dea&ti2ate PDP $onte;t 'e<uest 121
Dea&ti2ate PDP $onte;t A&&e%t 121 1!!)!!(
Dea&ti2ate PDP $onte;t Failure ! !)!!(
Deta&h 'e<uest 281
Deta&h A&&e%t 12 45)1(
Overall Data Session Completion Rate 78 62.90%
RAU Statistics Count Success rate
'A/ Attem%t 22
'A/ Su&&ess 13 5)!(
'A/ Faile* 4!)1(
Cell Reselection Statistics Count Success rate
$ell 'esele&tion 8rom /"+S to 7S" Attem%ts 2
$ell 'esele&tion 8rom /"+S to 7S" Su&&ess 1 5!)!!(
Cell %e&election 'rom #(TS to )S( Dela* +"+ 1$."!
Cell %e&election 'rom #(TS to )S( Dela* Signalling ,-A
$ell 'esele&tion 8rom 7S" to /"+S Attem%ts 1
$ell 'esele&tion 8rom 7S" to /"+S Su&&ess ! !)!!(
Cell %e&election 'rom )S( to #(TS Dela* +"+ ,-A
Tie in S!ste Seconds Ratio
7S" 171 1)!2(
/"+S 1655 8)8(
9 N eed to agree
9 + hat ra 'igures ill(ontain
9 + hat $nd * ser ill(ontain
9 O ther (uts o' D ata
For internal use
:P5=!8 PS PDP
&onte;t
a&ti2ation
>Same as 1!
>35! o&&urren&es

>PS PDP a&ti2ation tri33er %oint?
>/6 not PS atta&he*
>/6 sen*s 1
st
@''$ $onne&tion 'e<uestA
>PS PDP a&ti2ation &om%letion tri33er %oint?
>/6 re&ei2es @''$? Downlink Dire&t +rans8er ,S"? A&ti2ate PDP
$onte;t A&&e%t-A
>Su&&ess8ul &om%letion i8?
>/6 re&ei2es @''$? Downlink Dire&t +rans8er ,S"? A&ti2ate PDP
$onte;t A&&e%t-A
>$lusterBPDPB$onte;tBA&ti2ationBS'B( C PDP
$onte;t A&ti2ation $om%letion / PDP $onte;t
A&ti2ation Attem%ts D1!!
8(
:P5=! PS
throu3h%ut
*ownlink

>'e%eate* F+P &allsE 1 ".0te 8ile
*ownloa*E 1 /e
>384 k.%s
>A2era3e throu3h%ut
>35! &alls

>Data Session Set=u% tri33er %oint?
>/6 re&ei2es 1
st
D1 %a&ket
>Data Session $om%letion tri33er %oint?
>/6 re&ei2es last %a&ket
>Session out%ut?
>A2era3e *ownlink throu3h%ut
>$lusterBPSBD1B+hrou3h%ut C A2era3e
+hrou3h%ut #2er Data Sessions
2!! :.%s

:P5 5D $riteria +est Des&ri%tion +ri33er Statisti&s Su&&ess
'ate
:P5=!1 Foi&e 12)2
k.%s A"'
$$S'
>1 "#$G 1 "+$ in the same 2anE
2 /es
>1 &all ,%art0 A to %art0 HE %art0 A
terminates the &all-? 15 se&on*s
&all G 15 se&on*s i*le) Part0 A to
.e lo33e*
>35! &alls

>$all set=u% tri33er %oint?
>Part0 AAs /6 sen*s 1
st
@''$ $onne&tion 'e<uestA
>$all &om%letion tri33er %oint?
>Part0 AAs /6 re&ei2es @''$ $onne&tion 'elease ,&ause normal-A
>15 se& &all *uration ,8rom ''$ $onn) 'e<uest to ''$ $onn)
'elease-
>Part0 AAs /6 re&ei2es @''$ $onne&tion 'elease ,&ause normal-A
>$lusterBA"'B#2erallB$$S'B( C $all
$om%letion / $all Attem%ts D1!!
8(
:P5=!2 Foi&e D$' >1 "#$ to PS+NE 1 /e
>1 &all? ! se&on*s &all G 15
se&on*s i*le
>35! &alls

>$all Set=u% tri33er %oint?
>37 /6 re&ei2es @''$? Downlink Dire&t +rans8er ,Alertin3 8or NIE
$onne&t A&knowle*3e 8or A/-A
>$all &om%letion tri33er %oint?
>37 /6 re&ei2es @''$ $onne&tion 'elease ,&ause normal-A
>! se& &all *uration ,8rom ''$ $onn) 'e<uest to ''$ $onn)
'elease-
>37 /6 re&ei2es @''$ $onne&tion 'elease ,&ause normal-A
>$lusterBA"'B#2erallBD$'B( C 1 = $all
$om%letion / Set u% $alls D1!!
J 2(
:P5=!3 Foi&e 37=
27
9an*o2er
,inter= K
intra="S$-
>1 "#$ to PS+NE 1 /e in *ual
mo*e
>1 &all? &ontinuous &all until 3727
&o2era3e .or*er is %asse*
>5! o&&urren&es as a minimum

>37 to 27 9# start %oint?
>/6 re&ei2es @''$? 9an*o2er From /+'AN $omman*A
>37 to 27 9# &om%letion tri33er %oint?
>/6 sen*s @9an*o2er $om%leteA to 27 HSS
>/6 sen*s @9an*o2er $om%leteA to 27 HSS

>$lusterBFoi&eB37to27B9#B#2erallBS'B( C 9#
$om%letion / 9# Attem%ts D1!!
8(

Drive Test Analysis Test &ase definition
!*amples
!hort (alls to test #all !etup !u((ess

0ong (alls to test Drop (all ratio
For internal use
Need to Define the () measurement +from
Drive test,
> $all >etup >uccess F $>>6 (Boice= circuit switched data)
> >uccessful call setup means that C-#/;# -irect @ransfer ($$: Alertin)D messae
is receiBed b" ;2.
6$all >etup @ime (Boice= circuit switched)
6$all setup dela" is measured from #1 messaes= startin from C66$ $onnection
>etupD messae to C-# -irect @ransfer ($$: Alertin)D messae.
6$all -rop (Boice= circuit switched)
6A dropped call occurs. @he call will be dropped in case 66$ connection release
(not normal release) messae has been send from 6N$ to ;2.
> >ession >etup >uccess (packet switched)

> @his is related to !-! context actiBation. >uccessfull" actiBated !-! context
means that actiBate !-! context accept messae has been sent from 6N$ to
;2 (66$: downlink direct transfer (>::actiBate !-! context Accept)).
6>ession >etup @ime (packet switched)
6@he Csession setup timeD is the dela" between the time the ;2 sends the data
session actiBation request until ?!6> attach and !-! context actiBation has
been successfull" completed.
6>ession -rop (packet switched)
6>ession drop rate can be defined as the number of successful !-! deactiBations
aainst number of successful !-! actiBations.
For internal use
Definition of &all $et-%p $uccess Rate +&$$R,
*$
: RR# #onne(tion Re/uest
RR# #onne(tion !etup phase
, , Resour(e Reservation in RN# %T!
Transmission
: RR# RR# #onne(tion Re/uest !etup
RR# #onne(tion A((ess phase
RN# aits reply 'rom *$
: RR# RR# #onne(tion #ompleted
: RR# "nitial Dire(t Trans'er (m servi(e re/uest
: RANAP "nitial *$ Message
( D"R$#T TRAN!F$R #all
) Pro(eeding
: RANAP RA% Assignment Re/uest
: - RR# Radio %earer !et up
: RR# Radio %earer !etup
#omplete
RA% #onne(tion !etup phase
, , Resour(e Reservation in RN# %T!
Transmission
: RANAP RA% Assignment
Response
( ) D"R$#T TRAN!F$R Alerting
%T!
RN# #N
( D"R$#T TRAN!F$R #onne(t)
( D"R$#T TRAN!F$R #onne(t
) A()noledge
RA% #onne(tion A((ess phase
RN# aits reply 'rom *$
- #all !et
up !u((ess
Rate
- #all !et
up Time
!u((ess'ul / ( : ) (all setup means that D0 *0 Dire(t Trans'er ## Alerting message is re(eived by *$
For internal use
Definition of the &all &ompletion Rate +&&R,
*$
RN# MG+
Node %
#all $stablished
"u Release #ommand
"u Release #omplete
RR# #onne(tion Release
RR# #onne(tion Release #omplete
Radio 0in) Deletion Re/uest
Radio 0in) Deletion Response
: A0#AP Release Re/uest
: A0#AP Release Response
: A0#AP Release Re/uest
: A0#AP Release Response
#all Released
( ) Dire(t Trans'er Dis(onne(t
( ) Dire(t Trans'er Release
( ) Dire(t Trans'er Release #omplete
( ) D"R$#T TRAN!F$R Alerting
( D"R$#T TRAN!F$R #onne(t)
( ) D"R$#T TRAN!F$R #onne(t A()noledge
#all Drop
Rate
#all
Duration
For internal use
Definition of $ession $etup Time - $
*$
#N
RN# +%T!
"N"T"A0 D"R$#T TRAN!F$R ( Atta(h Re/uest)
( & Donlin) Dire(t Trans'er Authenti(ation
) #iphering Re/uest
*$ already has an RR# (onne(tion
: !##P #onne(tion
Re/uest
: !##P #onne(tion
#on'irm
: & RANAP Authenti(ation #iphering
Re/uest
!e(urity Mode
#ommand
( & *plin) Dire(t Trans'er Authenti(ation
) #iphering Response
: & RANAP Authenti(ation #iphering
Response
: RANAP !e(urity Mode
#ommand
!e(urity Mode
#ommand
: RANAP !e(urity Mode
#ommand
: RANAP #ommon
"D
: Donlin) Dire(t Trans'er "dentity
Re/uest
: RANAP "dentity
Re/uest
: *plin) Dire(t Trans'er
"dentity Response
: RANAP "dentity
Response
: Donlin) Dire(t Trans'er
Atta(h A((ept
: Dire(t Trans'er Atta(h
A((ept
: *plin) Dire(t Trans'er
Atta(h #omplete
: Dire(t Trans'er Atta(h
#omplete
( "N"T"A0 D"R$#T TRAN!F$R A(tive PDP
) #onte1t Re/uest
( D"R$#T TRAN!F$R A(tive PDP #onte1t
) Re/uest
( D"R$#T TRAN!F$R A(tive PDP (onte1t A((ept)
: RANAP RA% A!!"GNM$NT R$3*$!T
: RR# Radio %earer
- !et up
: - RR# Radio %earer !et up
#omplete
: RANAP RA% A!!"GNM$NT R$!PON!$
: RR# Measurement #ontrol
( "N"T"A0 *$ M$!!AG$ GPR!
) Atta(h
!ession
- !et up
Time
For internal use
NonFenuine failures
:easurement s"stem fault
($ollection @ool or Anal"sis)
?enuine failures
6' issue ($oBerae /
/nterference / !oor
dominance)
:issin neihbour
>"stem issue G(@>
>"stem issue 6N$
$ore network issue
>"stem
(;nknown )
+ % T
!
R N # # ore N +
* $
"ub
"u
"t is bene'i(ialto
(ategorise (all'ailures
during the analysis
and reporting
.ailure /rea0do1n
For internal use
@he E!/ measurement conditions should be used to define exclusion
areas durin driBe test for acceptance of a cluster.
All failures happenin in those area= that do not respect the coBerae
requirements should be discarded.

Si3nal
a.o2e
'S$P
threshol*

Si3nal
.elow
'S$P
threshol*

Si3nal a.o2e
6&/No
thresho l*

Si3nal
.elow
6&/No
threshol*

F 88*d(m
F F,%d(m
F )*d(m
@ime
6>$!
&all
O(
&all
O(
&all
O(
&all
NO(
&all
NO(
F 82d(
F %d(
F 2*d(
2c/No
Threshold
2xample of $all >uccess $riteria
-iscard sample

7
&all - +A,
&all - +/,
F 88*d(m
F
F )*d(m
@ime
6>$!
&all
O(
&all
O(
&all
O(
&all
NO(
&all
NO(
F
F %d(
F 2*d(
2c/No
Threshold
-iscard sample
7
7
&all - +A,
F 88*d(m
F
F )*d(m
@ime
6>$!
&all
O(
&all
O(
&all
O(
&all
NO(
&all
NO(
F
F %d(
F 2*d(
2c/No
Threshold
-iscard sample
7
7 F 88*d(m
F
F )*d(m
@ime
6>$!
&all
O(
&all
O(
&all
O(
&all
NO(
&all
NO(
F
F %d(
F 2*d(
2c/No
Threshold
-iscard sample
7
7
&all - +A,
&all - +/,
.ailure /rea0do1n
For internal use
.ailure /rea0do1n
Call Set"u# $ailures %rea&do'n " %enchar& Drive
5(
2!(
!(
5(
!(
!(
1!(
5(
35(
15(
!(
5(
6<ui%ment
'e3istration $lash
$ore Network
Fail in 27
'AN ,8i;e* in ne;t release-
Site maintenan&e ,+"/:-
'AN ,#ther-
"issin3 nei3h.ours
37 'F
slow &ell 'esele&tion
site &on8i3uration K maintenan&e
,Nokia-
#ther
Call Set"u# $ailures Call Set"u# E(ce#tions
$ailure T!#e total )
6<ui%ment ! !)!
'e3istration $lash ! !)!
$ore Network ! !)!
Fail in 27 6 1)4
'AN ,8i;e* in ne;t release- ! !)!
Site maintenan&e ,+"/:- 16 51)6
'AN ,#ther- ! !)!
"issin3 nei3h.ours ! !)!
37 'F 3 )7
Slow &ell resele&tion ! !)!
site &on8i3uration K maintenan&e ,Nokia- ! !)!
#ther 6 1)4
Call Setu# $ailure %rea&do'n " $inal Drive
!(
!(
!(
1(
!(
!(
!(
1!(
!(
!(
1(
52(
6<ui%ment
'e3istration $lash
$ore Network
Fail in 27
'AN ,8i;e* in ne;t release-
Site maintenan&e ,+"/:-
'AN ,#ther-
"issin3 nei3h.ours
37 'F
Slow &ell resele&tion
site &on8i3uration K maintenan&e
,Nokia-
#ther
Call Set"u# $ailures Call Set"u# E(ce#tions
$ailure T!#e total )
6<ui%ment 1 5)!
'e3istration $lash 4 2!)!
$ore Network ! !)!
Fail in 27 1 5)!
'AN ,8i;e* in ne;t release- ! !)!
Site maintenan&e ,+"/:- ! !)!
'AN ,#ther- 2 1!)!
"issin3 nei3h.ours 1 5)!
37 'F 7 35)!
slow &ell 'esele&tion 3 15)!
site &on8i3uration K maintenan&e ,Nokia- ! !)!
#ther 1 5)!
A M R # all!etup Failures
For internal use
Drive $urvey Analysis rocess $ummary
Dia'ram
For internal use
Non Genuine &all $etup .ailure $cenarios
:easurement s"stems are often not perfect and ma" introduce errors
in data collection or anal"sis
2xamples of nonFenuine failures seen:

For internal use
Non Genuine &all $etup .ailures
:easurement s"stem failures b" driBe test tool
9
For internal use
:easurement s"stem failures
b" drive test tool
$all attempt durin #ocation
Area update (#A update
clash)
0A * pdate R e/uest
# allattem pt
R R# Release
For internal use
:easurement s"stem failures b" Actix workbook
- R R # # onne(tion 'or "nter R AT (ellresele(tion

0A * pdate Re/uest
For internal use
Non Genuine Drop &all $cenarios
:easurement s"stem failure examples:

-riBe test tool
;ser initiated C;# $$ -isconnectD
Anal"sis Gorkbook
> /nterF6at cell reselection from 2? to 1? interpreted as drop call
> $omplete (e. ,* seconds) call on either 1? or 2?
> No drop in the lo file / same drop listed twice
9
For internal use
Genuine &all $etup .ailure $cenarios
6' issue
/nterference / -ominance / $oBerae
:issin neihbour
>"stem /ssue F (@>
> No response to C66$ $onnection 6equestD
C66$ $onnection 6eHectD to C66$ $onnection 6equestD
>"stem issue F 6N$
> C$$ -isconnectD after C$all !roceedinD due to C-# 66$ $onnection
6eleaseD
$ore NG
> C$: >erBice AbortD after C$: >erBice 6equestD
>"stem issue (test number)
> C$$ -isconnectD after C$$ !roressD
For internal use
Genuine Drop &all scenarios
6' issue
/nterference / -ominance / $oBerae
>"stem issue (@>
>udden C$$ -isconnectD due to C-# 66$ $onnection 6eleaseD
>udden drop to idle= no disconnect messain
>"stem issue 6N$
>udden C$$ -isconnectD due to C-# 66$ $onnection 6eleaseD

For internal use
.ailure 2ocation
9 A nalyse the
signalling
'lo to 'ind
the lo(ation
o' 'ailure and
potential
(ause
9 * $ log m ay
only (apture
som e o' the
m essages
For internal use
#all !etup Failure Analysis
For internal use
DRN# *$ #N !RN# +%T!
. - , RR# (onne(tion set up
. 8 !ervi(e $stablished
. / & 9 %ran(h addition deletion A(tive set update
. , 1 #ell sear(h %##&
& de(oding RA#&
a((ess
. <--> - 3 *$ #! #N signalling
. - 4 RA% set up
. : !ervi(e Released
. <--> - 5 *$ #! #N !ignalling
AMR &$ &all hases
For internal use
!tart
%est
server s
> R!#P
-14,d%m
%est
server s
/ > $( No
-1,d%
#overage
Optimi;ation
Dominan(e
Optimi;ation
( ) A"#& A#-
re(eived2
<es
Missing
Neighbo
ur 2
Neighbour list
Optimi;ation
No
No
No
<es
& *0 (overage RA#&
. parameter Optimi;ation
( ) (hanging serving (ell
RR#
#onne(tion
!etup
re(eived2
( ) D#& RR#
#onne(tion setup
#ompleted sent
'rom *$2
Radio
%earer setup
'ailure Re(ei
ved2
& Report Finish
#he() 'ailure (ause
( / Not radio problem (ell
) update
& Report Finish
( ) #he() 'ailure (ause
( A# optimi;ation (he()
Pr1Noise
& ) inter'erer around %T!
& Report Finish
RR# !etup
Re.e(t
re(eived2
<es
No
<es
<es
<es
& Report Finish
( : Reason o' problem 01 syn(
) 'ail
& Report Finish
( ) #he() 'ailure (ause
#all !etup
Failure
Analysis
Pro(ess
A
%
#
D
$
<es
<es
No
No
No
No
For internal use
&all setup failures R. issue
6' issue3 $oBerae / /nterference / -ominance
9
9
!ee the e1ample in Module 3 RF Optimisation
For internal use
&all setup failures Missin' Nei'h3our
:issin neihbour anal"sis oBer the whole route (1?F1?= 1?F2?)
>earch for failures due to missin 1?F1? neihbours
>earch for failures due to missin 1? I2? neihbours
/t is suested to place 2? scanner to the test Behicle

For internal use
&all $etup .ailure Analysis- /loc0 / -
@he purpose of this actiBit" is to check the 6andom Access !rocess is
workin adequatel" b" inBestiatin whether A/ (Acquisition
/ndicator) has been receiBed throuh -# A/$A
/f A/$A was not receiBed b" ;2= the cause of the problem can be
classified into:
9
@he (asic theor" for 6A$A setup procedure and plannin parameters
can be found in :odule & I !arameter 4ptimisation
9
For internal use
&all $etup .ailure Analysis- /loc0 / -
*$
+%T! RN#
/ Preamble RA#&
/ A(/uisition "ndi(ator A"#&
: / RR# RR# #onne(tion Re/uest PRA#&
: N%AP RAD"O 0"N- !$T*P R$3*$!T
: / RR# RR# #ONN$#T"ON !$T*P FA#&
01 !yn(hronisation
: N%AP !<N#&RON"!AT"ON "ND"#ATOR
: / RR# RR# #ONN$#T"ON !$T*P #OMP0$T$ D#&
*$ in #$00=D#& state
: N%AP RAD"O 0"N- !$T*P R$!PON!$
For internal use
RA#& Pro(ess

Donlin)
%!
/ 01 A#- A"#&
*plin)
M!
Preamble
1
Not dete(ted
Message part
Preamble
,
PRA#&=preamble=retrans
# PRA#& preambles transmitted during one PRA#& (y(le ithout re(eiving A"#& response
*$t1PoerMa1PRA#&

RA#&=t1=Ma1
# preamble poer ramping (y(les that (an be done be'ore RA#& transmission 'ailure is reported
PoerRamp!tepPRA#&preamble
PoerO''set0astPreamblePRA#&message
: "nitial preample poer
9 = - ( ) + ( ) + Pt1 #P"#&transmissionPoer R!#P #P"#& R!!" %! PRA#&Re/uiredRe(eived#"
For internal use
- - #all !etup Failure Analysis %lo() %
>olutions for 6A$A optimisation
M a1 * $ T1
po er hit
the
( * $=P=M A > ,
) 4d% m 2
To in(rease PRA#&=Preamble=retrans
Or PoerRamp!tepPRA#&Preamble
To in(rease PRA#&=Preamble=retrans
Or PoerRamp!tepPRA#&Preamble
No
<es
"s *0
"nter'eren(e
abnormally
&"G&2
<es
No
Report there might be an inter'ering
sour(e Nearby the serving (ell
Report there might be an inter'ering
sour(e Nearby the serving (ell
#hange the !erving (ell to (over the problem Area
=> *$ is too 'ar to rea(h the serving (ell
#hange the !erving (ell to (over the problem Area
=> *$ is too 'ar to rea(h the serving (ell
For internal use
&all $etup .ailure Analysis- /loc0 /
Open loop Poer #ontrol parameters 'rom RA#& "n'o message
For internal use
&all setup failures $ystem issue /T$
No response to C66$ $onnection 6equestD

For internal use
&all setup failures $ystem issue /T$
C66$ $onnection 6eHectD after C66$ $onnection 6equestD
For internal use
&all $etup .ailure Analysis
;2 has the appropriate -#/;# coBerae but if 6N$ does not allow to set up the
66$ connection of the requested 6A( (6adio Access (earer)= $all setup will
fail.
Admission $ontrol (A$) is inBolBed in 66$ connection setup. A$ can reHect 66$
reHect 66$ connection >etup due the -# #oad= ;# load or -# >preadin codes
/f measured ;# (!rx@otal) or -# (!tx@otal) load exceeds taret

thresholds (PrxTarget and PtxTarget) A$ can still admit new
6A( to the cell if a new nonFcontrollable load keeps below
taret thresholds (in practice this means that A$ can admit
onl" new controllable load 6A(s i.e. N6@ 6A(s)
/f measured ;# (!rx@otal) or -# (!tx@otal) load exceeds
oBerload thresholds (PrxTarget J PrxOffset and PtxTarget J
PtxOffset) then A$ canKt admit more 6A(s to the cell
For internal use
9 - During the pre optimi;ation phase it is unli)ely that A# ill stop an RR#
(onne(tion setup during the drive testing be(ause there are normally
. ( ) very 'e *$s in the netor) Tra''i( loading is trivial
9 , &oever it should be (he()ed that measured Pt1Total and Pr1Total are
( . . ) ( . . , % ) less than Pt1Target e g 44d%m and Pr1Target e g 4d% 84 loading
. respe(tively
9 "' D0 A# , # does not allo RR# setup (he() the T1 poer o' +%T!
, . o' (hannels transmitted !ignaling messages
9 "' *0 A# : does not allo RR# setup #he() out i' there is an
. inter'ering sour(e nearby the serving (ell
For internal use
@o check if #a"er 8 >"nchroniLation (slot/frame s"nc) has failed

/f C66$ $onnection >etupD was receiBed b" ;2 but ;2 does not send C66$
$onnection >etup $ompletedD= we will report C#8 s"nchroniLation failureD and
haBe to check #8 s"stem messaes.
For internal use
9 ## Dis(onne(t a'ter
#all Pro(eeding
9 Good RF (onditions
9Failures in RAB setup occur
between the RAB
Assignment Request being
received from Core Networ
and the RAN sending out
Radio Bearer !etup" #herefore
the failure is between B#!
and . #ore Netor)
&all setup failures $ystem issue RN&
For internal use
9 ## Dis(onne(t a'ter #all
( .) Pro(eeding (ont
9 ( An e1ample site shos high
values on (ounter
RA%=!TP=FA"0=#!=?O"#$=%T!
during the drive test
9 "n the re(ent (he() the
. (ounter shoed no 'ailures
9
&all setup failures $ystem issue RN&
For internal use
&all setup failures &ore N4
>ecurit" :ode $ommand
$ommon /-
;2 6N$ :?G Node (
66$ $onnection 2stablishment
/nitial -irect @ransfer ($: >erBice 6equest) >$$!: $onnection 6equest
>$$!: $onnection $onfirm
#ocation6eportin $ontrol
9 RRC: Initial Direct Transfermessae is sent usin acknowleded mode 6#$ to the $> core
domain. 6outin is to be based upon the local !F@:>/
9 @he NA> messae is not read b" the 6N$ but is forwarded to the multimedia atewa". @he NA>
messae includes the /:>/ as a ;2 identit"
9 @he SCCP: Connection Requestmessae establishes the connection orientated sinallin link i n
the same wa" as it was for the 66$ connection phase.@his does not reserBe an" resources for the
A:6 call itself.
> +he Connection Con'irmmessa3e i*enti8ies the 'N$ with a *estination lo&al re8eren&e whi&h is the
same as the sour&e re8eren&e within the Connection %e.ue&t messa3e
> +he Connection Con'irmmessa3e i*enti8ies the $S &ore with a sour&e lo&al re8eren&e
> +he $S &ore sen*s a %A,AP/ 0ocation %eporting Control messa3e to the 'N$ re<uestin3
in8ormation re3ar*in3 the lo&ation o8 a %arti&ular /6
> +he %A,AP/ Common 1Dmessa3e s%e&i8ies the 5"S5 .elon3in3 to the /6
9 The Security Mode Commandmessae triers the start or stop of cipherin and interit"
protection.
C$: >erBice AbortD after
C$: >erBice 6equestD
?ood 6' conditions
C>ecurit" :ode
$ommandDFmessae
not receiBed b" ;2=
thus the failure is
belieBed to be at $ore
Network.
For internal use
&all setup failures $ystem )ssue +test
num3er,
C$$ -isconnectD after C$$
!roressD
> $ause: recoBer" on timer expir"
> @he call oes Bia /N >$! to a
recordin.
> A static test was done b" Nokia
$ustomer $are and in few
instances the call dropped after 1*
seconds of recordin passed.
Aence the problem is associated
with the test number not the 6AN
34
se(
: # ause re(overy on tim er e1piry
For internal use
Drop #all Analysis
For internal use
!tart
%est
server s
> R!#P
-14,d%m
%est
server s
/ > $( No
-1,d%
#overage
Optimi;ation
Dominan(e
Optimi;ation
<es
<es
Neighbour list
Optimi;ation
Missin
g
Neighb
our
<es
&all Drop .ailure Analysis rocess
!&O
Failed
"nvestigate possible %T! or RN# problem
No
"!&O
Failed
No
"!&O Failure Analysis
%
<es
!&O Failure Analysis
A
No
No
No
For internal use
&all Drop .ailure Analysis rocess +$5O
Analysis,
D0 A!*
re(eiv
ed
<es
!#
#lash
*$ T1
Poer
Ma1
Fi1 !#
#lash
#P"#&
Optimisation
*plin)
"nter'
eren(e
/ 0oad Optimisation $1ternal "nter'erer
0in)
*nbalan
(ed
<es
<es
<es
<es
No
No
No
<es
"nter
RN# &O
#he() "ur
<es
#ongest
ion on
target
(ell
0oad Optimisation
No
<es
No
#he() neighbour de'inition
parameters
#he() RF
0evels
No
D0 T1
Poer
Ma1
<es
!tart
D
#
For internal use
Drop call failures R. issue
9 RF drops mostly due to poor dominan(e
or inter'eren(e
9 , Poor (overage (ould lead to "!&O
although poor dominan(e or
. inter'eren(e (an (ause "!&O to 'ail
9 Rapid 'ield drop (an (ause drop due to
(overage
9 Poor dominan(e or inter'eren(e (an
( ) (ause #ompressed Mode #M to start
. even i' R!#P is still good
9 "n #M *$ transmits ith higher poer
( ) more inter'eren(e and spends less
( time on 3G less a((urate measurement
) reporting
9 Poor dominan(e or inter'eren(e (an
lead to A(tive !et update 'ailures and
. eventually to drop (all
Poor dominan(e (auses A(tive !et update 'ailures
For internal use
DL synchronisation is lost -> UE has stopped transmitting
#rChAgg and $% $&CC' B(R high
For internal use
#ransport Channel B(R" Btw )(*+,RNC -.AC la/er0
!ometimes $&CC' B(R -btw )(*+,1B#!0 , (an be a better indi(ator o' hat@s happening to the dedi(ated (hannel than the #P"#& $(No in parti(ular in the (ase that poer (ontrol may not be tra()ing ell
Fairl/ good C&2C' &ilot (cNo
For internal use
Drop call failures $ystem issue /T$
Drop to "D0$
>udden drop to idle= no disconnect messain
For internal use
Drop call failures $ystem issue RN&
!udden RR# #onne(tion Release
DP##& %$R
9CC $isconnect due to $%
RRC Connection Release
9 No response to *0
Measurement Reports
9 "n the e1ample site had
, no alarms & good RF %$R
9 Not able to add !#,85 to
, A(tive !et ne1t (all on
=> the same (ell no
. 'ailure
9 Di''i(ult to
troubleshoot i' the
'ailure does not happen
=> systemati(ally
'ollo up in the ne1t
/ ee)s drive do a
separate drive test in
the area
For internal use
Drop call failures +$& conflict,
9 !udden drop to idle mode
( no dis(onne(t
) messaging
9Cause of the failure3
overshooting site and !C
reuse
9!hort term solution to add
overshooting neighbour in
A$4! definitions
Cell ABC5 !C678
% Transport (hannel %0$R 144
For internal use
Drop &all - %plin0 )nterference
*0 inter'eren(e 'rom the !"%9 message
For internal use
Drop #all 0in) %alan(e
9 & *0 D0 Poer #ontrol (ommands (an help
. indi(ating problems in lin) balan(e
9 , P# 're/uen(y is 1544 &; thus ideally the sum
o' P# (ommands to in(rease or de(rease
poer is 1544
9 . . < , $ g i' the sum o' *0 P# (ommands is 1544
this ould indi(ate *$ is starting to loose
syn(hroni;ation
9 , in #ompressed Mode there is less P# (ommands
*$ spends time on ,G
UE RX power control message: DL reception weak -> UE is ordering WBT to increase power!
"m o# UL $% commands & '())* UE not recei+ing all the $% commands!
For internal use
Drop call failures $ystem issue RN& or /T$ 6
C$$ -isconnectD due to C-# 66$ $onnection 6eleaseD is Hust a
consequence of failure which can be due to different reasons
6ule out (@> failures

$heck the site performance from $ounters (/ub= >erBice leBel= cell resources
>A4= etc) and that site is carr"in traffic
!rxNoise= receiBe link parameters= alarms
>$Ireuse
;2 performance 3
/dentified causes for ActiBe >et ;pdate failure
C-eafD sites (!rxNoise)
'ault" AG
>$Freuse
For internal use
Drive Test Analysis Reporting
For internal use
Drive Test Analysis Reportin' 2evels
Optimiser
9 ?ery &igh 0evel -P"s that give a
users per(eption o' the netor)
( ) Netor) -P"
9 -P"s that provide an $ngineering
. vie o' netor) per'orman(e e g
, , , . #!!R #DR O#!R !&O per'orman(e
( -P"s re/uired 'or ea(h
) optimisers area
9 &ighly detailed -P"s that give a
detailed pi(ture o' netor)
per'orman(e at an $ngineering
level and allo root (ause
analysis
Regional Optimisation
Manager
!enior
Management
9 Pro(essing Drive Data to provide the in'ormation re/uired
at the bottom level means that the higher level
in'ormation (an be easily e1tra(ted
9 The di''erent reporting levels may ant to see -P"s based
( . . ) on di''erent (uts o' the data e g ra or end user
For internal use
() reportin'
NonFenuine failures to be remoBed from the raw E!/<s
9
Call Co#letion
$all Set/% Su&&ess 'ate * ) 975 * ) 979
$all Dro% 'ate * ) 13 * ) 03
#2erall $all Su&&ess 'ate * ) 962 * ) 976
$all $onne&tion time JC s 8 * ) 976 * ) 976
No o8 $all Attem%ts ,within &o2era3e- 634 625
No o8 $all Setu% Failures 16 13
No o8 Su&&ess8ul $all Setu%s 618 612
No o8 $all Dro%s 8 2
No o8 $om%lete* $alls 610 610
Call Co#letion
$all Set/% Su&&ess 'ate * ) 975 * ) 979
$all Dro% 'ate * ) 31 * ) 21
Call Co#letion
$all Set/% Su&&ess 'ate * ) 981 * ) 986
$all Dro% 'ate * ) 13 * ) 03
No o8 $all Attem%ts ,within &o2era3e- 630 621
No o8 $all Setu% Failures 12 9
No o8 Su&&ess8ul $all Setu%s 618 612
No o8 $all Dro%s 8 2
No o8 $om%lete* $alls 610 610
- N on genuine (allsetup 'ailures rem oved
- N on genuine drops rem oved
Final
- P"s
Final- P"s
For internal use
() reportin'
Geekl" E!/ trends (nonFenuine failures should be excluded)
All Routes
( 0
( 10
( 20
( 30
( 40
( 50
( 60
( 70
( 80
( 90
( 100
w
k
4
1
w
k
4
2
w
k
4
3
w
k
4
4
w
k
4
5
w
k
4
6
w
k
4
7
w
k
4
8
w
k
4
9
w
k
5
0
w
k
5
1
w
k
5
2
w
k
0
2
0
100
200
300
400
500
600
700
800
$all Set/% Su&&ess 'ate
$all Dro% 'ate
#2erall $all Su&&ess 'ate
+ime on 7 3
+ime on 7 2
$all Attem%ts
For internal use
() reportin'
(@> failure chart (call setup failure 5 drops)

$umulatiBe number of failures that occurred per site oBer time
/f the ;2 is spendin onl" a small percentae of time on 1? problems ma" not
be identified.
# $00=#
# ell=A
# ell=D
# $00=%
For internal use
/T$ failure e*amples7 &!228A
#on histor" of failures (oBer weeks +&= +0= *2)
$all >etup failure scenarios:
D
1
rd
sector showin low aBerae !rxNoise I8*) d(m $ommissionin
data (feeder loss) was found incorrect.
After this site was still failin= not carr"in traffic.

For internal use
/T$ failure e*amples7 &!228/
'ailures onl" on week +,
No alarms
At G(@>: :AA parameters ok
At 6N$: :AA7*= cable loss 7 1 d( (-!$$A init pwr)
!rxNoise checked 4E= 4:$ statistics showed the site carried traffic
durin the driBe.
No failures in the followin weeks driBes
For internal use
/T$ failure e*amples7 &!228&
'ailures oBer weeks ++= +%= +,= %*
No response to C66$ $onnection 6equestD most frequent failure= also
one case of sudden drop to idle.
@est calls were made= the counters were not incremented durin the
test. !rotocol anal"ser proBed no actiBit" in /ub. @he counters were
incremented onl" after site reset.
Alarm CG>:A 66Fbus errorD
@he site had fault" G@6= incorrect feeder loss in the commissionin
file.
9
For internal use
/T$ failure e*amples7 &!228D
'ailures on weeks +0 and +)
/ncorrect feeder loss in commissionin data.

For internal use
!ummary
9 !ervi(es to be measured and -P"s to be (al(ulated need to
be (learly de'ined
9 ?arious (all patterns e1ist to e1er(ise the -P"s
9 Reporting 0evels should be understood
9 De'ine and agree e1(eptions in advan(e
9
For internal use
9ersion control
?ersion Date !tatus Oner
RAN44 ,448 %ase version , , Mi)e Ro(he !teve &unt Gareth
, Davies Pe))a Ranta
. RA!451=v1 4 . . ,4 44 ,449 $1ample (ase available
Minor (hanges (he()ed 'or RA!451
level
-irsi TerAvAinen

Drive Test Analysis

Uploaded by

Copyright:

Available Formats

Drive Test Analysis

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

Drive Test Analysis

Uploaded by

Copyright:

Available Formats

1 Nokia Siemens Networks Presentation / Author / Date

For internal use

-ifferent >cramblin codes

!hort (alls to test #all !etup !u((ess

> >ession >etup >uccess (packet switched)

2xamples of nonFenuine failures seen:

21 Nokia Siemens Networks Presentation / Author / Date

:easurement s"stem failures b" Actix workbook

- R R # # onne(tion 'or "nter R AT (ellresele(tion

:easurement s"stem failure examples:

27 Nokia Siemens Networks Presentation / Author / Date

33 Nokia Siemens Networks Presentation / Author / Date

3 Nokia Siemens Networks Presentation / Author / Date

/f measured ;# (!rx@otal) or -# (!tx@otal) load exceeds taret

@o check if #a"er 8 >"nchroniLation (slot/frame s"nc) has failed

6ule out (@> failures

(@> failure chart (call setup failure 5 drops)

65 Nokia Siemens Networks Presentation / Author / Date

/ncorrect feeder loss in commissionin data.

You might also like