Post-processing of Drive Test Data

&
Analysis of
W-CDMA Networks
(Internship Report)

SUBMITTED BY:

YOUR NAME
Graduate Student
INTRODUCTION:
First Generation and Second Generation Wireless Systems have enabled voice
communications to go wireless in many of the leading markets. But customers are
increasingly finding value in other services such as Text Messaging, Video and access to
Data Networks. Thus there is a huge demand for networks that support these services.
Third generation systems are specifically designed for enabling multimedia
communications.
Multimedia communication includes:
• Person to person communication with enhanced high quality images and video.
• Access to information on public and private networks.
• High data rate access and new flexible communication capabilities

With such a high demand for 3rd Generation wireless communication networks, all
the telecom vendors are vying for a piece of the 3G Cellular Networks Market. Ericsson
is one of the vendors that is ahead of the race and is all set to launch Cingular’s
WCDMA based 3G Wireless Cellular Communication System.
During the course of internship at Ericsson the job description was to support the
Launch of Cingular`s WCDMA Network for the various markets.
Responsibilities included:
1) Processing Drive Test Data using Ericsson In-house Tools.
2) Ensuring that the generated KPI`s were accurate.
3) Analyzing Drop Calls and Call Setup Failures
4) Analyzing the ThroughPut for Packet Calls.
5) Running Macros on the Results to ensure proper results
6) Coordinating with Market personnel to ensure proper availability of Resources.
7) Support the development of Omega Tool by giving feedback on the tool to the
developers.
Introduction to 3G WideBand-CDMA Technology:
Worked on UMTS (Universal Mobile Telecommunication Systems) .UMTS is
based on WCDMA technology. WCDMA is called UTRA (Universal Terrestrial Radio
Access) FDD (Frequency Division Duplex) and TDD(Time Division Duplex).

SUMMARY of the Main Parameters of WCDMA:

WCDMA is a wideband Direct-Sequence Code Division Multiple Access(DS-
CDMA) system. The Duplexing method used is Frequency division duplexing/Time
division duplex ( the user information are spread over a wide bandwidth by multiplying
the user data with quasi-random bits).Thus in order to support very high bit rates upto
2Mbps, the use of a variable spreading factor and multicode connections are supported.
The Chip rate of 3.84Mcps is used which leads to a carrier bandwidth of approximately 5
MHz . WCDMA supports two basic modes of operation: Frequency Division
Duplex(FDD) and Time Division Duplex(TDD).
FDD Mode: A separate 5MHz carrier frequencies are used for the uplink and
downlink respectively.
TDD Mode: Only one 5MHz is time shared between the uplink and downlink.
Uplink is the connection from the mobile to the base station, and downlink is that from
the base station to the mobile.
• WCDMA supports the operation of asynchronous base stations, so that there is no
need for a global time reference such as GPS. Deployment of indoor and micro
base stations is easier when no GPS signal needs to be received.
• WCDMA employs coherent detection on uplink and downlink based on the use of
pilot symbols or common pilot.
• WCDMA air interface has been crafted in such a way that multi-user detection
and smart adaptive antennas can be deployed by the network operator as a system
option to increase capacity and/or coverage.
• WCDMA is designed to be deployed in conjuction with GSM. Therefore
handovers between GSM and WCDMA are supported in order to be able to
leverage the GSM coverage for the introduction of WCDMA.
1.2 Generic Principles of WCDMA operations:
1.2.1. Spreading and De-spreading:
User data is assumed to be a BPSK-modulated bit sequence of rate R, the user
data assuming the values of +1or -1.The spreading operation, is the multiplication of each
user data bit with a sequence of 8 code bits, called chips. Thus the resulting spread data is
at the rate of 8*R and has the same random appearance as the spreading code. Here the
spreading factor is considered to be 8. This wideband signal would then be transmitted
across a wireless channel to the receiving end.
During de-spreading we multiply the spread user data/chip sequence, bit duration
by the bit duration, with the very same 8 code chips as we used the spreading of these
bits. Thus the original user bit sequence has been recovered perfectly, provided we have
the perfect synchronization between the spread user signal and the de-spreading code.

Fig 1: Spreading and De-spreading

1.2.2. Multipath Radio Channels:

Radio propagation in the land mobile channel is characterized by the multipath
reflections, diffractions and attenuation of the signal energy. These are caused by natural
obstacles such as buildings, hills etc which results in so-called multipath propagation.
There are two effects resulting from multipath propagation .
1) The signal energy may arrive at the receiver across clearly distinguishable
instants. The arriving energy is smeared into a certain multipath delay profile.
2) Also for a certain time delay position there are usually many paths nearly equal in
length along which the radio signal travels. This may result into signal
cancellation which is called fast fading. Thus the signal power may drop
considerably 20-30 dB when phase cancellation of multipath reflection occurs.

τ2
τ3
τ0
τ1

Fig 2: Multipath Propogation

1.2.3. Rake Receiver Principle:

The delay dispersive energy is combined by utilizing multiple RAKE fingers
allocated to those delay positions on which significant energy arrives.
 Each RAKE finger tracks a different multipath component
– Also used to track other cells during soft handover
 Sliding correlator used to obtain a correlation peak for each multipath
component
 – 1 to 2μs delay between multipaths in urban/suburban
– 20μs delay in rural/hilly
– Most RAKE receivers can handle up to 30μs delay
 Searcher finger is used to measure other cells
– Used to determine when to perform handovers

1.2.4. Different type of Handovers:

• Hard Handover:
Hard handover means that all the old radio links in the UE are removed before the
new radio links are established. Hard handover can be seamless or non-seamless.
Seamless hard handover means that the handover is not perceptible to the user. In
practice a handover that requires a change of the carrier frequency (inter-
frequency handover) is always performed as hard handover
.
• Soft Handover:
Soft handover means that the radio links are added and removed in a way that the
UE always keeps at least one radio link to the UTRAN. Soft handover is
performed by means of macro diversity, which refers to the condition that several
radio links are active at the same time. Normally soft handover can be used when
cells operated on the same frequency are changed.

• Softer handover:
Softer handover is a special case of soft handover where the radio links that are
added and removed belong to the same Node B (i.e. the site of co-located base
stations from which several sector-cells are served. In softer handover, macro
diversity with maximum ratio combining can be performed in the Node B,
whereas generally in soft handover on the downlink, macro diversity with
selection combining is applied.
2. UMTS Radio Access Network Architecture:
2.1. System Architecture:
The System architecture consists of logical network elements and the interfaces.
The network elements are grouped into Radio Access Network(RAN,UMTS Terestrial
RAN=UTRAN) that handles all radio-related functionality and the Core Network which
is responsible for switching and routing calls and data connections to user and radio
interfaces.
UMTS network elements consists of sub-networks and it is possible to have
several network elements of the same type so UMTS system is called modular. Thus
having sub-networks , are distinguished from each other with unique identities. Such a
network is called UMTS PLMN(Public Land Mobile Network) ..Typically one PLMN is
operated by a single operator and is connected to other PLMNs as well as to other types
of network, such as ISDN, PSTN, the internet .
The network elements in a PLMN are shown in the figure as follows:

Node B
RNC MSC/VLR GMSC PLMN, PSTNetc
USIM
Node B
HLR External Networks

ME Node B
RNC SGSN GGSN Internet
UE Node B
UTRAN CN

Fig 3: Network elements in a PLMN

The brief description of the network elements is described as follows:

2.1.1. UE: The UE consists of two parts
• The Mobile Equipment (ME) is the radio terminal used for radio communication
over the Uu interface.
 • The UMTS Subscriber Identity Module (USIM) is a smartcard that holds the
subscriber identity, performs authentication algorithms, and stores authentication
and encryption keys and some subscription information that is needed at the
terminal.

2.2.2. UTRAN: The UTRAN consists of two distinct elements:

• The Node B converts the data flow between the Iub and Uu interfaces. It also
participates in radio resource management.
• The Radio Network Controller (RNC) is the service access point for all services
UTRAN provides the CN.

2.1.3. HLR: (Home Location Register) is a database located in the user’s home system
that stores the master copy of the user’s service profile. The service profile consists of,
for example, information on allowed services, forbidden roaming areas, and
supplementary service information such as status of call forwarding and the call
forwarding number. It is created when a new user subscribes to the system, and remains
stored as long as the subscription is active.

2.1.4. MSC/VLR(Mobile Services Switching Centre/Visiting Location Register) is

the switch (MSC) and database(VLR) that serves the UE in its current location for circuit
Switched(CS) services. The MSC function is used to switch the CS transactions, and the
VLR function holds a copy of the visiting user’s service profile as well as more precise
information on the UE’s location within the serving system.

2.1.5. GMSC(Gateway MSC) is the switch at the point where UMTS PLMN is
connected to external CS networks. All incoming and outgoing CS connections go
through GMSC.

2.1.6. SGSC(Serving GPRS(General Packet Radio Service Support Node)

functionality is similar to that of MSC/VLR but is typically used for packet switched(PS)
services. The part of the network that is accessed via the SGSN is often referred to as the
PS domain.

2.1.7. GGSN (Gateway GPRS Support Node) functionality is close to that of GMSC
but is in relation to PS services.

The external networks can be divided into two groups:

2.1.8. CS networks: These provide circuit-switched connections, like the existing
telephony service. ISDN and PSTN are examples of CS networks.

2.1.9. PS networks: These provide connections for packet data services. The internet is
one example of a PS network.

Interfaces in UMTS system:

2.1.10. Cu interface: This is the electrical interface between the USIM smartcard and
ME.

2.1.11. Uu interface: This is WCDMA radio interface.The Uu is the interface which the
UE accesses the fixed part of the system.

2.1.12. Iu interface: This connects UTRAN to the CN.

2.1.13. Iur interface: The open lur interface allows soft handover between RNCs from
different manufacturers

2.1.14. Iub interface: The Iub connects a Node B and an RNC.

2.2 UTRAN Architecture

UTRAN consists of one or more Radio Network Sub-systems (RNS). An RNS is
a sub-network within UTRAN and consists of one Radio Network Controller (RNC) and
one or more Node Bs. RNCs may be connected to each other via an Iur interface. RNCs
and Node Bs are connected with an Iub interface.

The radio Network Controller: The RNC is the network responsible for the control of
radio resources of the UTRAN. It interfaces the CN (normally to one MSC and one
SGSN) and also terminates the RRC (Radio Resource Control) protocol that defines the
messages and procedures between the mobile and UTRAN.

2.3 Channels:
2.3.1.Downlink Logical Channels(L3:
a) Control Logical Channels:
• BCCH(Broadcast Control Channel): Broadcasts cellsite and system
information to all UE
• PCCH(Paging Control Channel): Transmits paging information to a UE when
the UE’S location is unknown.
• CCCH(Common Control Channel): Transmits control information to a UE
when there is no RRC Connection
• DCCH( Dedicated Control Channel): Transmits control information to a UE
when there is a RRC connection.
b) Traffic Logical Channels:
• CTCH(Common Traffic Channel): Traffic channel for sending traffic to a
group of UEs
• DTCH(Dedicated traffic Channel): Traffic Channel dedicated to one UE.

2.3.2. Downlink Transport Channels (L2):

a) Common Transport Channels:
• BCH(Broadcast Channel): Continous transmission of system and cell
information
• PCH (Paging Channel): Carries control information to UE when location is
unknown.
 • PICH (paging indication channel): Pending activity indicated
• FACH (Forward Access Channel): Used for transmission of idle-mode control
information to a UE.
b) Dedicated Transport Channels:
• DCH(Dedicated Channel): Carries dedicated traffic and control data to one
UE .Used for BLER measurements.

2.3.3. Downlink Physical Channels (L1)

a) Common Physical Channels:
• P-CCPCH Common Control Physical Channel (Primary): Broadcasts cell site
information. Timing reference for all DL.
• SCH (Synchronization Channel): Fast Synch. Codes 1 and 2: time-multipleded
with P-CCPCH.
• S-CCPCH Common Control Physical Channel(Secondary): Transmits idle-
mode signaling and control information to UEs.
• CPICH: Common Pilot Channel.

b) Dedicated Physical Channels:

• DPDCH: Dedicated Downlink Physical Data Channel
• DPCCH: Dedicated Downlink Physical Control Channel: Transmits connection-
mode signaling and control to UEs.

c) Indicator Physical Channels:

• AICH: (Acquisition Indicator Channel): Acknowledges that BS has
acquired a UE Random Access attempt.
• AP-AICH(Access Preamble Indicator Channel): Acknowledges that BS
has acquired a UE Packet access attempt
• CD/CA-ICH: Confirms that there is no ambiguity between UE in a
Packet Access attempt. Optionally provides available Packet Channel assignments
 • CSICH: Broadcasts status information regarding packet channel
availability.

3. Description of the software used at Ericsson:

Responsible for Austin Market for the post processing of Scanner data and UE
information which were being used for the initial tuning of the Cingular UMTS network
and is due to be launched in December 2005.
The Austin market basically used to give me Clusters in the form of log files.
These log files contained the drive test information collected by the Drive test engineers.
There were two types of drive one was Voice Drive (AMR loaded) and Packet Drive (PS
loaded). In Voice Drive there were long calls and short calls. Short calls were made for
30 seconds approximately and long calls were calculated from 30 minutes to 1 hour.
Voice Drive consists of information regarding call setup, call set-up timing,
dropped call, time for the call, end messages, layer3 messages, event data information,
scanner data, BLER, EC/Io, Ec, RSCP.
These log files were exported into FMT files with the help of TEMS Investigation
Version 6.0 Software. The details what the TEMS investigation Ver 6.0 Software is as
below,

3.1. TEMS Investigation 7.0 version :

It is a leading air-interface test tool for UMTS networks.
The tool allows operators to monitor radio parameters, speech quality and decode air
interface messages efficiently and easily.

TEMS Investigation tool is used to

1) Track poorly defined neighbor relations
2) detect a wide range of radio network configuration errors.
3) Simulate and verify network changes
 4) Decode Layer 3 messages and display network data on map line charts or status
monitor.
5) Identify coverage, quality and capacity measurements
6) Perform system testing
7) Perform fault-tracing and troubleshooting
8) Verify perdiction
9) Tuning
10) Active Set Distribution

This tool automatically collects correlated data from follow phones, scanners and
GPS. It measures and analyze parameters such as throughput , call setup success, drop
call rate, pilot pollution and missing neighbors. Captures and displays handoff.

3.2. Omega tool:

The Omega Tool is an in house tool used to post-process drive test data and
produce outputs useful in Initial tuning and Report generation.
Omega Tool is used for post-processing the TEMS log files from a cluster drive to derive
the KPIs from the drive and to extract meaningful data for analysis
The Omega Tool performed various functions like
1) Input Consolidation
2) Post-Processing
3) Data analysis
4) Report Generation
5) Data export
6) Input Adaptation
The processed data was stored in an MS access database. A drive test consists of
several individual files. The first task of omega tool is to extract data from individual text
and put in a single database. All the data from one drive test instance will be stored in
that database.
The post-processing, analysis, export and Report Generation consists of following
calculations
 1) UE Call Sequence:
Call Sequence were needed for the evaluation of drop call rate and call success rate.
The Layer3 messages gave the information regarding the call setup and dropped call.
The sequence followed was as follows:
a) RRC Connection Request
b) RRC Connection Setup complete
c) RRC Connection Reject
d) Alerting
e) Connect
f) Disconnect
g) Dropped Call
h) Activate PDP Context Accept
i) Session Start
j) Session End
k) Session Error
A Call sequence has to start and end normally. A call sequence always starts with
an RRC Connection Request message. The end sequence is signaled by the message
Disconnect or event Dropped call.
For Packet calls, the end of the sequence is indicated by session end or session
error messages. It is possible not to receive the end of Sequence indications. This can
happen because of UE or drive test software bugs. All such sequence will be marked as
incomplete. Incomplete sequences will not be used for calculations.
 CS Voice Call Sequence

RRC Connection RRC

Request Connection
Paging Setup
Type 1 Disconnect or
Complete Alerting
Dropped Call

Fig 4: Voice Call sequence

PS Call Sequence

RRC Connection RRC

Request Connection Activate
Paging Setup
Type 1 PDP Session
Complete Context Start
Accept Session
End OR
Session
Error

Fig 5: Packet sequence

3.3. LAYER-3 Messages in TEMS Investigation 7.0 Software
 Fig 5: Layer 3 messages on TEMS
3.4 . UE KPI(Keys Process interface) Evaluation:
KPIs will be compiled from the call sequences as shown in the figure. Incomplete
call sequences will not be considered during the KPI evaluation.

3.4.1. AMR voice Dropped Call Rate

a) Discrete Calls: Discrete calls last for a fixed duration and are either dropped or ended
normally

Voice Dropped Call Rate= Total number of Dropped Call events from complete
sequences/Total number of alerting messages

b) Long Calls: A long call is maintained until the time the call is either dropped or is
normally ended at the time of the completion of the route. To calculate the drop rate of
 a long call drive test, the mean call duration must be given. The call sequences have to
divided into two groups.
Sequence that result in a dropped call-Sequence A
Sequence that are terminated normally with a Disconnect message-Sequence B

Dropped Call Rate= Number of sequence a calls/ Total effective number of calls.

3.4.2. AMR Voice Call Success Rate

For voice call success rate , the following need to be calculated.

Total Attempts= Total number of the last RRC connection Request message in the
RRC Connection procedure with cause code for voice call origination.
Voice call success rate= 1- Total Failures/Total Attempts

3.4.3. Bler Calculation= Errored/ (Errored blocks+ Good blocks)

3.4.4. PS Dropped Call Rate = number of session errors/ number of Activate PDP
Context Accept messages.

3.4.5. PS Call Success Rate= 1- total failures/total attempts

3.4.6. Transport channel throughput= (1-BLER)* RLC throughput

3.4.7. Scanner Data Evaluation:

UMTS scanner data is used for examining the RF environment. The scanner
measures CPICH, Ec/Io and Ec. This information is used for the further improvement of
the radio environment. For every scanner sample point, the scanner samples points are
arranged in the descending order i.e the strongest server will have the best Ec. Each
scrambling code has the information regarding Ec/Io, Ec, cell name and rank. The rank
determines the position of the scrambling code. Scanner sample points are assigned to
bins and a list of scrambling codes present in the bin is made and Median Ec, Median
Ec/Io, average rank and number of samples reported. If the number of samples for the
scrambling code is <X% of the total samples in the bin that server is discarded. X% is the
settable parameter.
For every bin total samples in the bin, scrambling code of the server, cell name of
the server, Median Ec of the server, median Ec/Io of the server, average rank of the
server, number of samples of the server is exported into Mapinfo files.

3.4.8. Pilot Pollution Analysis for the scanner:

Pilot pollution will be analysed to determine which areas have the pilot pollution
and which cell causes pollution. The first “active set size” scrambling code are assigned
to be active, the remaining scrambling are assigned to be polluters if their median Ec is
within the pilot pollution threshold dB. If more than maximum number of polluters are
found, then only first maximum number of polluters are retained.

3.4.9. Neighbor Analysis:

The purpose of neighbor analysis is to determine missing UTRAN neighbor from

the scanner data For every scanner sample point, the scanner samples are arranged in the
order of the strongest server (highest Ec) to the weakest server (lowest Ec).

Each scrambling code has the information about the Ec, Ec/Io, Cell name and the
rank. The rank determines if the position of the scrambling code (SC) in the sorted list
indicated above. Scanner samples points are assigned to bins.
 A list of all SCs present in the bin is determined. For every SC in the bin, the
following values are calculated: Median Ec, Median Ec/Io, Average rank, Number of
samples reported. For every bin, all the SCs within “active set threshold” of each other
are separated. The process is repeated for all the bins. A list of all the neighbors from all
the bins is compiled. This list is called the Feasible Neighbor List. Feasible neighbor list
is checked against the neighbor list in TEMS Investigation cell file. Neighbor relations
are checked in both directions. A list of all the neighbors present in the feasible neighbor
list but missing tin the TEMS Investigation cell file is compiled. This list is called the
Missing Neighbor List.A list of all the neighbors absent in the feasible neighbor list but
present in the TEMS Investigation cell file is compiled. This list is called the Additional
Neighbor List.

4. Conclusion:
It was a great learning experience working with Ericsson in the UMTS team for
Post-Processing of Drive Test Data and help in the Launch of the Cingular network at
Austin .Learnt about the existing and the new recent technologies used in wireless
cellular systems. Looking Forward to learn more and as Ericsson’s LOGO says “Taking
you Forward”. I believe that it has really taken me forward and I am looking forward to
move more forward.

5. References:
• WCDMA for UMTS Radio access for 3G Mobile Communication
Third Edition by Harri Holma and Antti Toskala.
• Reference material provided by Ericsson.