PERFORMANCE EVALUATION OF DYNAMICCALL ADMISSION CONTROL ALGORITHM
Ebere Omeje
ANANA, IKPONGAKARASE JAMES
PERFORMANCE EVALUATION OF DYNAMICCALL ADMISSION CONTROL ALGORITHM FOR WCDMA
BASED 3G NETWORKS
Ebere Omeje Digitally Signed by: Content manager’s Name
DN : CN = Webmaster’s name
O= University of Nigeria, Nsukka
OU = Innovation Centre
FACULTY OF ENGINEERING
DEPARTMENT OF ELECTRONIC ENGINEERING
ANANA, IKPONGAKARASE JAMESPG/MENGR/14/68105
i
PERFORMANCE EVALUATION OF DYNAMIC PRIORITY FOR WCDMA
Digitally Signed by: Content manager’s Name
DN : CN = Webmaster’s name
O= University of Nigeria, Nsukka
FACULTY OF ENGINEERING
DEPARTMENT OF ELECTRONIC
ANANA, IKPONGAKARASE JAMES
ii
PERFORMANCE EVALUATION OF DYNAMIC
PRIORITY CALL ADMISSION CONTROL ALGORITHM FOR WCDMA BASED 3G NETWORKS
BY
ANANA, IKPONGAKARASE JAMES PG/MENGR/14/68105
DEPARTMENT OF ELECTRONIC ENGINEERING FACULTY OF ENGINEERING
UNIVERSITY OF NIGERIA, NSUKKA
AUGUST, 2016
iii
APPROVAL PAGE
PERFORMANCE EVALUATION OF DYNAMIC PRIORITY CALL ADMISSION CONTROL
ALGORITHM FOR WCDMA BASED 3G NETWORKS
BY
ANANA, IKPONGAKARASE JAMES
(PG/M.ENG/14/68105)
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
AWARD OF MASTER OF ELECTRONIC ENGINEERING (TELECOMMUNICATION
OPTION) IN THE DEPARTMENT OF ELECTRONIC ENGINEERING, UNIVERSITY OF
NIGERIA, NSUKKA.
ANANA, IKPONGAKARASE JAMES SIGNATURE____________DATE__________
(STUDENT)
PROF. C. I. ANI SIGNATURE____________DATE__________
(SUPEVISOR)
EXTERNAL EXAMINER SIGNATURE____________DATE__________
DR, M. A. AHANAEKU SIGNATURE____________DATE__________
(HEAD OF DEPARTMENT)
PROF. E. S. OBE SIGNATURE____________DATE__________
(CHAIRMAN, FACULTY
POSTGRADUATE COMMITTEE)
iv
CERTIFICATION
This is to certify that ANANA, IKPONGAKARASE JAMES, a postgraduate student in the
department of Electronic Engineering with registration number PG/M.ENG/14/68105 has
satisfactorily completed the requirement of the course and research work for the degree of Master
in Engineering in Electronic Engineering.
_______________________ _____ __________________________
PROF. C. I. ANI DR. M. A. AHANAEKU
(SUPERVISOR) (HEAD OF DEPARTMENT)
_____________________________________
PROF. E. S. OBE
(CHAIRMAN, FACULTY POSTGRADUATE COMMITTEE)
v
DECLARATION
I ANANA, IKPONGAKARASE JAMES a postgraduate student in the Department of Electronic
Engineering with Registration number PG/MENGR/14/68105 declare that the work contained in
this report is original and has not been submitted in part or in whole for any other degree of this or
any other institution.
_____________________________ __________________
ANANA, IKPONGAKARASE JAMES DATE
vi
DEDICATION
This research work is dedicated to the Almighty God the giver of life and wisdom and my Father
Elder Engr. J. U. Anana the first Communication Engineer known to me.
vii
ACKNOWLEDGEMENT
I acknowledge the Almighty God for the successful completion of this research work. I acknowledge my supervisor Prof C. I. Ani who is the head of department, for his meticulous supervision to ensure this work emerges with outstanding excellence. I also acknowledge the lecturers and entire staff of Electronic Engineering department for their contribution to success of this work. To my parents, very dear sisters and brothers especially Mrs. Inemesit Okoro, who contributed and gave their unending support to ensure that this work is completed timely I acknowledge you. To you my good friends who contributed by way of sourcing materials and giving a helping hand even through prayers to make this work a reality God bless you real good.
viii
ABSTRACT
The wideband code division multiple access (WCDMA) based 3G cellular mobile wireless networks is expected to provide diverse range of multimedia services to mobile users with guaranteed quality of service (QoS). In order to provide the diverse quality of service required by the users of these networks, an effective radio resource management (RRM) is necessary. Radio resource management is responsible for the efficient and optimal utilization of network resources while providing QoS guarantees to various applications. Call admission control is a form of radio resource allocation scheme used for QoS provisioning in a network, which restricts access to the network based on resource availability, in order to prevent network congestion and consequent service degradation. This research focuses on how to maintain service continuity with quality of service guarantees and provide service differentiation to mobile user’s traffic profile by efficiently utilizing system resources. The services are divided into four traffic classes’ handoff real-time, handoff non-real time, new call real-time and new call non-real-time respectively, giving higher priority to handoff traffic classes. It uses an algorithm referred to as dynamic prioritized uplink call admission control (DP-CAC), an efficient tool that provides better performance for WCDMA based 3G network. Beyond system utilization, revenue and grade of service as the key performance indicators, this research work also considers the queuing delay and the call blocking/dropping probability of each traffic class. From the simulation results and analysis it is discovered that the new call non-real-time traffic class experiences greater queuing delay of 1.42E-11 at increasing traffic intensity compared to other traffic classes in the system. It is also discovered that at peak traffic intensity of 3.60E+03 handoff RT has a probability of 1.59E-02, handoff NRT a probability of 1.69E-02, new call RT a probability of 2.00E-02 and new call NRT a probability of 2.10E-02 showing that call blocking/dropping probability of handoff and new calls at high traffic condition is minimized. This is achieved because the model dynamically switches handoff traffic to its reserved channel, and allows new calls to go through the general server thereby providing service continuity to handoff traffic and fairness to new call traffic classes respectively.
ix
TABLE OF CONTENT
Title Page i
Approval Page ii
Declaration iii
Dedication iv
Acknowledgement v
Abstract vi
Table of Content vii
List of Figures x
List of Tables xii
Acronyms xiii
CHAPTER ONE - INTRODUCTION
1.0 Background of Study 1
1.1 Problem Statement 6
1.2 Objectives of Study 7
1.3 Scope of Study 8
1.4 Methodology 8
1.5 Significance of the Study 9
1.6 Dissertation Outline 9
CHAPTER TWO - LITERATURE REVIEW
2.0 Evolution of Cellular Network 10
2.1 First Generation (1G) Networks 10
2.1.1 Physical Architecture 11
2.1.2 Technology 12
2.1.3 Modulation 12
x
2.1.4 Protocol 13
2.2 Second Generation (2G) Network 14
2.2.1 Frequency of Operation 14
2.2.2 Technology 15
2.2.3 Modulation 16
2.2.4 GSM System Physical Architecture 18
2.2.5 GSM Protocol Architecture 24
2.3 High Speed Circuit Switched Data 28
2.4 Packet Digital Cellular Systems 2.5G 29
2.4.1 GPRS Architecture 31
2.4.2 GPRS Protocol Architecture 33
2.5 Enhanced Data Rates for GSM Evolution (EDGE) 35
2.6 Third Generation Cellular Network (3G) 36
2.6.1 UMTS Radio Interface 38
2.6.2 UMTS Architecture 39
2.6.3 Universal Terrestrial Radio Access Network (UTRAN) 44
2.6.3.1 Node B 46
2.6.3.2 The Radio Network Controller 46
2.6.4 UMTS Core Network 50
2.6.5 UMTS Interfaces 53
2.6.6 UMTS Radio Interface Protocol Architecture 54
2.6.6.1 Layer 1 55
2.6.6.2 Layer 2 56
2.6.6.3 Layer 3 59
2.7 WCDMA Concepts 60
2.7.1 Power Control 63
2.7.2 Handoff 65
2.7.3 Channelization Codes 69
2.7.4 Scrambling Codes 69
2.7.5 Code Allocation 70
2.8 Radio Resource Management 71
xi
2.8.1 Resource Allocation 74
2.8.1.1 Methods of Resource Allocation 75
2.8.2 Radio Resources 76
2.8.2.1 Types of Radio Channels 77
2.9 Call Admission Control 82
2.9.1 CAC Design Considerations 83
2.9.2 Multiple Service Types 84
2.10 Related Works 85
CHAPTER THREE – RESEARCH METHODOLOGY
3.0 Adopted Network 88
3.1 Adopted Network Architecture 90
3.2 Physical Model 92
3.3 DP-CAC Algorithm 94
CHAPTER FOUR – SIMULATION RESULT
4.0 Simulation Model 98
4.1 Results and Discussion 105
CHAPTER FIVE – CONCLUSION AND RECCOMENDATION
5.0 Conclusion 118
5.1 Contribution to Knowledge 119
5.2 Recommendation 119
References
xii
LIST OF FIGURES
Figure 2.1: Functional architecture of a GSM system 18
Figure 2.2: GSM interfaces 23
Figure 2.3: Protocol architecture for signaling in GSM 24
Figure 2.4: GPRS architecture reference model 32
Figure 2.5: GPRS transmission plane protocol reference model 34
Figure 2.6: The UMTS physical architecture 40
Figure 2.7: UMTS network domain 41
Figure 2.8: UTRAN architecture 45
Figure 2.9: logical role of RNC 50
Figure 2.10: UMTS core network architecture 51
Figure 2.11: UMTS radio interface protocol architecture 54
Figure 2.12: Hard handoff procedure 67
Figure 2.13: Soft handoff procedure 68
Figure 2.14: Spreading and scrambling 70
Figure 3.1: Multimedia Services 89
Figure 3.2: WCDMA Network Architecture 91
Figure 3.3: DP-CAC Physical Model 93
Figure 3.4: Node B System Model 94
Figure 3.5: Flow-chart for DP-CAC Algorithm 97
Figure 4.1: Real Time Traffic Source (Voice) 99
Figure 4.2: Non-Real Time Traffic Source (Data) 100
Figure 4.3: Real Time Traffic Source (Video) 100
Figure 4.4: Calls Arriving from Respective Sources 101
xiii
Figure 4.5: Flow of Traffic to DP-CAC Switch 102
Figure 4.6: Computational Module 103
Figure 4.7: Decision Making Model 103
Figure 4.8: Complete Simulation Model 104
Figure 4.9: Graph of system capacity utilization (revenue) against offered traffic 106
Figure 4.10: Comparison between system capacity utilization (revenue) with DP-CAC Algorithm and without DP-CAC Algorithm 108
Figure 4.11: Graph of grade of service against offered traffic 109
Figure 4.12: Comparison between grade of service Performance with DP-CAC Algorithm and without DP-CAC Algorithm 110
Figure 4.13: Graph showing queuing delay against traffic intensity for each traffic class 111
Figure 4.14: Graph of call blocking and dropping probability against traffic Intensity for handoff and new calls respectively 112
Figure 4.15: Graph of call blocking and dropping probability against Traffic Intensity
for the respective call classes 113
Figure 4.16: Graph of call blocking and dropping probability at a server capacity of 24 channels 115
Figure 4.17: Graph of call blocking and dropping probability at a server capacity of 12 Channels 116 Figure 4.18: Graph of call blocking and dropping probability at a server capacity of
6 channels 117
xiv
LIST OF TABLES
Table 2.1: Differences between WCDMA and GSM air interfaces 62
Table 3.1: Service Priority Classes 92
Table 3.2: Computation Parameters 95
Table 3.3: Traffic Model 95
Table 3.4: Performance Measures 96
xv
ACRONYMS
AMPS Advanced Mobile Phone System
BCCH Broadcast Channel
BCH Broadcast Channel
BER Bit Error Rate
BMC Broadcast/Multicast Control Protocol
BoD Bandwidth on Demand
BPSK Binary Phase Shift Keying
BS Base Station
BSS Base Station Subsystem
BSC Base Station Controller
CAC Call Admission Control
CB Cell Broadcast
CBC Cell Broadcast Center
CBS Cell Broadcast Service
CCCH Common Control Channel
CCH Common Transport Channel
CDMA Code Division Multiple Access
CDPD Cellular Digital Packet Data
CM Connection Management
CN Core Network
CPCH Common Packet Channel
CPICH Common Pilot Channel
CRC Cyclic Redundancy Check
CRNC Controlling RNC
xvi
CS Circuit Switched
CTCH Common Traffic Channel
DCA Dynamic channel allocation
DCCH Dedicated Control Channel
DCH Dedicated Channel
DECT Digital Enhanced Cordless Telephone
DL Downlink
DNS Domain Name Service
DP-CAC Dynamic Priority Call Admission Control
DPCCH Dedicated Physical Control Channel
DPDCH Dedicated Physical Data Channel
DRNC Drift RNC
DS-CDMA Direct Spread Code Division Multiple Access
DSCH Downlink Shared Channel
DSL Digital Subscriber Line
DTCH Dedicated Traffic Channel
EDGE Enhanced Data Rates For GSM Evolution
EFR Enhance Full Rate
ETSI European Telecommunications Standards Institute
FACH Forward Access Channel
FDD Frequency Division Duplex
FDMA Frequency Division Multiple Access
FTP File Transfer Protocol
GERAN GSM/EDGE Radio Access Network
GGSN Gateway GPRS Support Node
xvii
GMSC Gateway MSC
GPRS General Packet Radio Service
GPS Global Positioning System
GSM Global System for Mobile Communication
HARQ Hybrid Automatic Repeat Request
HLR Home Location Register
HSDPA High Speed Downlink Packet Access
HSUPA High Speed Uplink Packet Access
IM Interference Margin
IMS IP Multimedia Sub-System
IMSI International Mobile Subscriber Identity
IP Internet Protocol
ISDN Integrated Services Digital Network
ITU International Telecommunications Union
Iu BC Iu Broadcast
JTACS Japanese Total Access Communication Systems
LAI Location Area Identity
LAN Local Area Network
LM Load Margin
MAC Medium Access Control
MAI Multiple Access Interference
ME Mobile Equipment
MGW Media Gateway
MIMO Multiple Input Multiple Output
MM Mobility Management
xviii
MMS Multimedia Message
MS Mobile Station
MSC/VLR Mobile Services Switching Centre/Visitor Location Register
NAS Non Access Stratum
NBAP Node B Application Part
NMT Nordic Mobile Telephone
NRT Non-Real Time
O&M Operation and Maintenance
OSS Operations Support System
OVSF Orthogonal Variable Spreading Factor
PC Power Control
PCCCH Physical Common Control Channel
PCCPCH Primary Common Control Physical Channel
PCH Paging Channel
PCPCH Physical Common Packet Channel
PCS Persona Communication Systems
PDC Personal Digital Cellular
PDCP Packet Data Convergence Protocol
PDP Packet Data Protocol
PDSCH Physical Downlink Shared Channel
PDU Protocol Data Unit
PHY Physical Layer
PI Page Indicator
PICH Paging Indicator Channel
PLMN Public Land Mobile Network
xix
PRACH Physical Random Access Channel
PS Packet Switched
PSTN Public Switched Telephone Network
QAM Quadrature Amplitude Modulation
QoS Quality of Service
QPSK Quadrature Phase Shift Keying
RAB Radio Access Bearer
RACH Random Access Channel
RAI Routing Area Identity
RAN Radio Access Network
RANAP RAN Application Part
RB Radio Bearer
RF Radio Frequency
RLC Radio Link Control
RNC Radio Network Controller
RNS Radio Network Sub-System
RNSAP RNS Application Part
RRC Radio Resource Control
RRM Radio Resource Management
RT Real Time
SAP Service Access Point
SCCP Signalling Connection Control Part
SCH Synchronisation Channel
SDU Service Data Unit
SEQ Sequence
xx
SF Spreading Factor
SGSN Serving GPRS Support Node
SHO Soft Handover
SIP Session Initiation Protocol
SIR Signal to Interference Ratio
SM Session Management
SMS Short Message Service
SN Sequence Number
SNR Signal to Noise Ratio
SRB Signalling Radio Bearer
SRNC Serving RNC
SRNS Serving RNS
SS7 Signalling System #7
TACS Total Access Communication System
TCH Traffic Channel
TCP Transport Control Protocol
TD/CDMA Time Division CDMA, Combined TDMA and CDMA
TDMA Time Division Multiple Access
TDD Time Division Duplex
TE Terminal Equipment
TF Transport Format
TFCI Transport Format Combination Indicator
TFCS Transport Format Combination Set
TFI Transport Format Indicator
TMSI Temporary Mobile Subscriber Identity
xxi
UDP User Datagram Protocol
UE User Equipment
UL Uplink
UM Unacknowledged Mode
UMTS Universal Mobile Telecommunication Services
URA UTRAN Registration Area
URL Universal Resource Locator
USIM UMTS Subscriber Identity Module
UTRA Universal Terrestrial Radio Access (3GPP)
UTRAN UMTS Terrestrial Radio Access Network
VoIP Voice over IP
VPN Virtual Private Network
WAP Wireless Application Protocol
WCDMA Wideband Code Division Multiple Access
WCDMA Wideband Code division multiple access
WWW World Wide Web
1
CHAPTER ONE
INTRODUCTION
1.0 Background of Study
For some years now, cellular telephony systems have been experiencing a level of unprecedented
growth in the world of telecommunications. When the first cellular technologies were brought into
service, at the beginning of the 1980s, there was a rather slow take-off in the number of
subscribers, hardly presaging the subsequent spectacular growth [1, 2]. The slow take-off of
subscribers was however a result of incompatibility between the systems, and major differences in
the use of the radio segment [2, 3]. Unfortunately, travelers who go to countries where the
technology offered by their operator is not represented find themselves suddenly deprived of their
communication tool because the subscriber management is not at all the same on the different
systems [1, 2, 4, 5]. It was therefore imperative to have a unified standard which will address these
issues and this led to the evolution of the first generation analogue system and then to fourth
generation system referred to as the long term evolution system (LTE). The different evolutions of
the cellular network have their respective frequency of operation, modulation scheme, protocol of
operation, access mode technology, and physical architecture, but one common feature is the
signaling standard.
Signaling refers to the exchange of control information between components of a network
(telephones, switches) in order to establish, manage, and disconnect calls [2, 3, 5]. The purpose of
network signaling is to set up a circuit between the calling and called parties so that user traffic
(voice, fax, and analog dial-up modem, for example) can be transported bi-directionally. When a
circuit is reserved between both parties, the destination local switch places a ringing signal to alert
the called party about the incoming call. This signal is classified as subscriber signaling because it
2
travels between a switch (the called party's local switch) and a subscriber (the called party). A
ringing indication tone is sent to the calling party telephone to signal that the telephone is ringing.
If the called party wishes to engage the call, the subscriber lifts the handset into the off-hook
condition. This moves the call from the set-up phase to the call phase [2, 3, 4, 5].
Signaling between mobile stations and the network require radio resources. Since large amount of
signaling traffic is being exchanged between the networks before communication can be
established, large amount of bandwidth and radio channels are also required [2, 3]. This high
demand for wireless communication services requires increased system capacities, the simplest
solution would be to allocate more bandwidth to these services, but the electromagnetic spectrum
is a limited resource, which is increasingly scarce [6, 9]. The radio resources such as radio
(frequency) spectrum and transmit powers are generally limited due to the physical and regulatory
restrictions, as well as the interference limited nature of wireless cellular networks [4, 6, 9]. If
these resources are not properly managed there will be increased interference in the systems which
will result to poor quality of service. Therefore, to provide communication services with high
capacity and good QoS, it is imperative to employ efficient and effective methods for sharing the
radio spectrum [8, 9, 12, 24].
Spectrum sharing methods are called multiple access techniques. Multiple access technique
involves radio channel allocation to users of the system. The objective of multiple access
techniques is to provide communication services with sufficient bandwidth when the radio
spectrum is shared with many simultaneous users. The most common multiple access techniques
are frequency division multiple access (FDMA), time division multiple access (TDMA), and code
division multiple access (CDMA) [4, 6, 7, 9]. FDMA was used in the first generation (1G) of the
cellular systems such as advanced mobile phone service (AMPS) systems which were basically
3
analog systems. TDMA enhances FDMA by further dividing the bandwidth into channels by the
time domain as well; TDMA is used as the access technology for global system for mobile
communications (GSM), which is representative of the second generation (2G) of cellular systems
[1, 2, 4, 9]. The digital transmission techniques of the 2G mobile radio networks have already
improved upon the capacity and voice quality attained by the analog mobile radio systems of the
first generation, however, more efficient techniques allowing multiple users to share the available
frequencies are necessary. Unlike FDMA and TDMA, CDMA transmission does not work by
allocating channels for each user, Instead, CDMA utilize the entire bandwidth for transmission of
each user [2, 4, 7, 9]. Therefore CDMA’s access method is achieved by assigning each user a
distinguished spreading code called chip code. This chip code is used to transform a user’s
narrowband signal to a much wider spectrum prior to transmission in a manner known as a spread
spectrum transmission. The enhanced CDMA access method which is known as the wideband
code division multiple access (WCDMA) is employed in universal mobile telecommunication
systems (UMTS) which is representative of third generation (3G) of cellular systems.
3G mobile communication systems evolved as a response to the challenge of developing systems
that increased the capacity of the existing 2G systems [4, 5, 10]. This required that the
infrastructure be designed so that it can evolve as technology changes, without compromising the
existing services on the existing networks. Separation of access technology, transport technology,
service technology and user application from each other make this demanding requirement
possible [4, 5, 15]. UMTS was developed as the migration of the European Telecommunications
Standards Institute (ETSI) 2G/2.5G systems GSM/GPRS (general packet radio services); with the
aim of facilitating as much as possible the extension of the existing networks of these worldwide
systems as well as the interoperability of the new UMTS system with the previous networks [4, 8,
4
11]. The limits of the GSM radio interface technology are some of the main reasons why the
decision was made to reconsider the definition of a new radio technology. Ultimately, within
UMTS, this decision led to the definition of the radio interface technology that we now know as
WCDMA, while keeping the core network similar to that existing in GSM/GPRS systems [8, 9,
11].
WCDMA introduces a significant degree of complexity in the design and operation of the radio
interface, as it supports spectral efficiency, general quality of service parameters, multimedia
services and bit rate up to 2Mbps which are its distinct characteristics when compared to existing
radio interface technologies [8, 11, 12]. Spreading is the process fundamental to the operation of
the WCDMA interface particularly the direct sequence CDMA, in direct sequence spreading; the
information signal is multiplied by a high frequency signature sequence, also known as a spreading
code or spreading sequence which increases the narrow bandwidth of the user to a wider
bandwidth [8, 10, 11]. In WCDMA, all users transmit in the same frequency band in an
uncoordinated fashion, referred to as an asynchronous transmission scenario which imposes time
offsets that result in multiple access interference in the uplink, while multipath interference is due
to the different arrival times of the same signal via the different paths at the receiver and is present
in both uplink and downlink [8, 11, 12]. As the number of users’ increases, the multiple access
interference increases too, thus, the capacity of WCDMA is known to be interference limited as it
is capable of accommodating additional users at the expense of a gradual degradation in
performance of the system in a fixed bandwidth [4, 8, 11].
The performance of WCDMA based cellular network depends largely on radio resource
management. Radio Resource Management (RRM) is a set of algorithms that control the usage of
radio resources which is located in the mobile terminal and the nodeB, in order to maximize the
5
overall system capacity, satisfy some predefined quality of service (QOS) requirement level to
different users according to their traffic profiles and provide optimum utilization of the system in
the cellular network [8, 9, 11]. RRM functions are realized through what is known as resource
allocation (RA), which determines how resources should be assigned by a nodeB to a mobile
subscriber. There are however factors that are affecting efficient resource allocation mechanisms
some of which are but not limited to error prone wireless channel, limited bandwidth and mobility
of mobile subscribers [4, 12]. Therefore, in order to study effective resource management
algorithms, it is necessary to understand and define the conditions that limit the cellular capacity;
these conditions are related to the service characteristics (voice, video, or data), the propagation
channel variations, the power control operation, and the user mobility patterns. The basic RRM
algorithms can be classified as follows: Handoff and mobility management algorithms, call
admission control (CAC) algorithms, and power control algorithms [9, 10, 11]. The call admission
control mechanism is an important component of RRM as it affects the resource allocation
efficiency and quality of service guarantees provided to users [9, 11]. The focus of this research is
on the performance evaluation of dynamic priority call admission control algorithm for WCDMA
based 3G networks.
6
Limitation of Cellular Radio Systems
Wireless cellular networks are relatively complex systems compared to the wired networks, hence
there are several factors which make it difficult to provide quality of service guarantees. These
include but are not limited to the following;
� Resources in cellular networks are very limited due to the limited radio frequency
spectrum.
� Limit to the maximum number of channels.
� Restriction to number of available channels that can be assigned to each cell.
� Wireless channels are inherently unreliable and prone to bursty errors due to noise,
multipath fading and interferences.
� Users tend to move around during a communication session causing handoffs between
adjacent cells, reduction in cell size to accommodate more users in a given area makes it
difficult to deal with the mobility related problems [4, 9, 12].
1.1 Problem Statement
The universal mobile telecommunication system (UMTS) is required to support a wide range of
applications (multimedia traffic) each with its own specific QoS requirement. Each traffic class has
its own application level QoS requirements in terms of delay, jitter, bit-error-rate (BER),
throughput and burstiness as well as call level requirements in terms of call blocking probability
for new calls and call dropping probability for handoff calls. This requires the media access control
(MAC) protocols and call admission control (CAC) mechanism to respectively enable application
level and call level performance guarantees for the traffic classes. This study focuses on call
admission control (CAC) with call level QoS guarantees.
7
1.2 Aim and Objectives
The general goal of emerging wireless cellular networks is to enable communication with a person,
at any time, at any place, and in any form. However, due to the distinct characteristics of this
network and the aforementioned limitations, the quality of service requirements, network
throughput and performance are often compromised. This study focuses on how to maintain
service continuity with quality of service guarantees and provide service differentiation to mobile
user’s traffic profile by efficiently utilizing system resources. It uses an algorithm referred to as
dynamic prioritized uplink call admission control (DP-CAC), an efficient tool that provides better
performance for WCDMA based 3G network and overcomes the shortcomings of complete
partitioning based algorithm. The objective of DP-CAC is outlined thus:
• At low and moderate traffic intensity, it ensures the optimum system utilization while QoS
is satisfied.
• At high traffic intensity, it ensures the fairness of resource usage amongst different traffic
class while maintaining QoS requirements.
• Support preferential treatment to higher priority calls by serving its queue first.
• To ensure best system utilization and revenue while satisfying the required QoS and
fairness.
• To minimize call dropping/blocking probability of handoff and new calls.
8
1.3 Scope of Study
This research work presents in detail the WCDMA 3G radio interface, its physical and protocol
architecture. It explains radio resource management and various resource allocation schemes, with
its focus on dynamic resource allocation strategies. The design considerations of call admission
control schemes and quality of service parameters are explored of which more attention is given to
dynamic priority call admission control scheme. The system model and simulation results are also
analyzed and presented.
1.4 Methodology
The methodology below was adopted to efficiently realize the objective of this research work;
A. Review of the Evolution of cellular networks to the third generation.
B. Detailed review of WCDMA radio interface technology for UMTS.
C. Review of Radio Resource Management and resource allocation schemes.
D. Review of existing call admission control schemes for WCDMA radio interface uplink.
E. Propose a call admission control scheme for WCDMA uplink radio interface following the
best admission schemes in review.
F. Develop an algorithm and a simulation model for the proposed admission scheme.
G. Simulate the model and obtain data using MATLAB.
H. Analyze Data using defined key performance indicators (KPI’s).
I. Computation of QoS parameter and other performance measures of the network using a
computational model.
J. Compare Performance of the proposed scheme with other existing schemes.
9
1.5 Significance of the Study
The significance of this study is based on the results from the simulation which indicate the
superiority of dynamic priority call admission control as it is able to achieve a better balance
between optimum system utilization, revenue, quality of service provisioning and fairness to all
traffic classes.
1.6 Dissertation Outline
The remaining part of this research work is organized thus: chapter two is a detailed review of the
evolution of cellular networks and radio resource management strategies, chapter three explains
the proposed scheme as well as it describes the system model, traffic model and performance
measures while chapter four discusses the results obtained from the simulation process and finally
the research work is concluded in chapter five.
10
CHAPTER TWO
LITERATURE REVIEW
2.0 Evolution of Cellular Network
In early networks, the emphasis was to provide radio coverage with little consideration for the
number of calls to be carried. As the subscriber base grew, the need to provide greater traffic
capacity had to be addressed. The cellular wireless generation (G) generally refers to a change in
the fundamental nature of the service, non-backwards compatible transmission technology, and
new frequency bands [1, 3]. New generations have appeared in every ten years, since the first
move from 1981-An analog (1G) to digital (2G) network. After that there was (3G) multimedia
support, spread spectrum transmission and 2011 all –IP Switched networks (4G) emerged. The last
few years have witnessed a phenomenal growth in the wireless industry, both in terms of mobile
technology and its subscribers [2]. There has been a clear shift from fixed to mobile cellular
telephony and new mobile generations do not pretend to improve the voice communication
experience. It tries to give the user access to a new global communication reality, whose aim is to
reach communication ubiquity (every time, everywhere) and to provide users with a new set of
services [3, 5, 2].
2.1 First Generation (1G) Networks
The first generation of cellular networks consisted of analog transmission systems for both voice
and data. A set of wireless standards were developed in the 1980's, these different standards were
used in various countries. Advanced Mobile Phone System (AMPS) AMPS, also known as IS-54,
on the 800MHz band, involves some 832 channels per carrier, and originated in the United States
[1, 4, 5]. Total Access Communication System (TACS) TACS operates in the 900MHz band,
offers 1,000 channels, and originated in the United Kingdom, Japanese Total Access
11
Communication Systems (JTACS) JTACS works in the 800MHz to 900MHz band, and it comes
from Japan. Nordic Mobile Telephone (NMT) the original variation of NMT was 450MHz,
offering some 220 channels, had a very large coverage area, thanks to its operation at 450MHz, but
the power levels are so intense that mobile sets were incredibly heavy [1, 2, 5]. NMT originated in
Denmark, Finland, Norway, and Sweden. Other 1G standards used in Europe include Germany and
Austria's C-Netz, Sweden's Comvik, NMT-F (France's version of NMT900), and France's
Radiocom 2000 (RC2000).
2.1.1 Physical Architecture
The analog cellular architecture comprises three components namely, the mobile unit, transceiver
station and mobile telephone switching office. The base transceiver station tower transmits signal
to and from the mobile unit and is connected to the mobile telephone switching office (MTSO)
through a microwave link or wire line [1, 2]. The MTSO interfaces into the terrestrial local
exchange to complete calls over the public switched telephone network (PSTN). When a mobile
unit is on, it emits two numbers consistently: the electronic identification number and the actual
phone number of the handset, which are picked up by the transceiver stations, and depending on
the signal level, they can determine whether the mobile unit is well within the cell or transitioning
out of that cell [1, 3, 5]. If the unit's power levels start to weaken and it appears that the unit is
leaving the cell, an alert is raised that queries the surrounding base transceiver stations to see
which one is picking up a strong signal coming in. As the unit crosses the cell perimeter, it is
handed over to an adjacent frequency in that incoming cell this process is known as handover. The
mobile unit cannot stay on the same frequency in between adjacent cells because that would create
co-channel interference (i.e., interference between cells) [1, 5].
12
2.1.2 Technology
The access mode technology used in the first generation analog system is the Frequency Division
Multiple Access (FDMA). FDMA is a process of allowing mobile stations to share radio frequency
allocation by dividing up that allocation into separate radio channels where each radio device can
communicate on a single radio channel during communication [1, 5]. A frequency band can be
divided into several communication channels using frequency division multiplexing (FDM). When
a device is communicating on an FDM system using a frequency carrier signal, its carrier channel
is completely occupied by the transmission of the device. For some FDM systems, after it has
stopped transmitting, other transceivers may be assigned to that carrier channel frequency [5].
When this process of assigning channels is organized, it is called frequency division multiple
access (FDMA). Transceivers in an FDM system typically have the ability to tune to several
different carrier channel frequencies [1, 2, 5]. 1G systems are circuit switched networks.
2.1.3 Modulation
Modulation is the process of changing the amplitude, frequency, or phase of a radio frequency
carrier signal (a carrier) to change with the information signal (such as voice or data). Mobile
systems use analog or digital modulation, but the first generation systems use analog modulation.
Analog modulation is a process where the amplitude, frequency or phase of a carrier signal is
varied directly in proportion or in direct relationship to the information signal [1, 3]. A voice call
gets modulated to a higher frequency of about 150MHz and up as it is transmitted between radio
towers.
13
2.1.4 Protocol
Cellular digital packet data (CDPD) is the protocol that was used 1G system. It is defined as a
connectionless, multiprotocol network service that provides peer network wireless extension to the
Internet [3, 5]. CDPD is a packet data protocol designed to work over AMPS. It was envisioned as
a wide area mobile data network that could be deployed as an overlay to existing analog systems.
And also a common standard that will take advantage of unused bandwidth in the cellular airlink,
that is, the wireless connection between the service provider and the mobile subscriber. Unused
bandwidth is a result of silence in conversations as well as the moments in time when a call is
undergoing handoff between cells. These periods of no activity can be used to carry data packets
and therefore take advantage of the unused bandwidth [1, 5].
CDPD provides mobile packet data connectivity to existing data networks and other cellular
systems without any additional bandwidth requirements. However, CDPD does not use the MSC
for traffic routing. The active users are connected through the mobile data base stations (MDBS) to
the Internet via intermediate systems (MD-IS) which act as servers and routers for the data user
[22].
Problems of 1G
� Roaming not supported
� Different standards, frequencies and frequency spacing.
� Security – eavesdropping was common place
� Difficult to expand
� Limited capacity
� Analogue systems – larger than required amount of the frequency had to be allocated to
each call
14
� Extremely slow data rates – only during silent period [1, 5].
2.2 Second Generation (2G) Network
The second generation (2G) of cellular technology was marked by a shift from analogue to digital
systems. Shifting to digital networks had many advantages; firstly, transmission in the digital
format aided clarity since the digital signal was less likely to be affected by electrical noise.
Secondly, transmitting data over digital network is much easier, data could also be compressed,
saving a lot of time, and finally with the development of new multiplexing access techniques, the
capacity of the cellular network could be increased [3, 4, 5].
The second generation standard known as Global system for Mobile communication (GSM) was
developed and published to provide a unified standard for cellular communication which will
address the incompatibility and roaming problems of the analogue systems [1, 3, 5]. The GSM
standard introduced the subscriber identity module (SIM) card, which held information about the
user and provided memory to store phone numbers and text messages. It could be shifted from one
handset to another, allowing users to choose handsets according to their fancy without having to
bother about the cellular service provider [1, 4, 5].
2.2.1 Frequency of operation
GSM systems operate in the 900MHz and 1800 GHz bands throughout the world with the
exception of the Americas where they operate in the 1900 GHz band [1, 5]. Cellular systems allow
reuse of the same channel frequencies many times within a geographic coverage area. The
technique called frequency reuse, which makes it possible for a system to provide service to more
customers called system capacity, by reusing the channels that are available in a geographic area
[1, 5, 7]. Two frequency bands 45 MHz apart have been reserved for GSM operation 890–915MHz
15
for transmission from the mobile station to the base station, that is, uplink, and 935–960MHz for
transmission from the base station, that is, downlink [5, 7]. Each of these bands of 25 MHz width
is divided into 124 single carrier channels of 200 kHz width. In each of the uplink/downlink bands
there remains a guard band of 200 kHz, of which each Radio Frequency Channel (RFCH) is
uniquely numbered, and a pair of channel with the same number forms a duplex channel with a
duplex distance of 45 MHz [1, 5].
2.2.2 Technology
The second generation network being digital systems, combine the Frequency Division Multiple
Access (FDMA) and Time Division Multiple Access (TDMA) technology. Time division multiple
access (TDMA) is a process of sharing a single radio channel by dividing the channel into time
slots that are shared between simultaneous users of the radio channel [3, 7]. When a mobile radio
communicates with a TDMA system, it is assigned a specific time position on the radio channel.
By allowing several users to use different time positions (time slots) on a single radio channel,
TDMA systems increase their ability to serve multiple users with a limited number of radio
channels. For example if we have a given bandwidth, FDMA could divide this bandwidth into 200
carrier frequency bands, TDMA will further divide each frequency band into eight (8) time slots,
allowing eight subscribers to utilize a single frequency band using their respective time slots
simultaneously, thereby increasing the number of subscribers to 1600 for the entire allocation. 2G
still utilizes the circuit switched technology [1, 3, 5, 7].
While GSM technology was developed in Europe, Code Division Multiple Access (CDMA)
technology was developed in North America. CDMA uses spread spectrum technology to break up
speech into small, digitized segments and encodes them to identify each call as well as
16
distinguishes between multiple transmissions carried simultaneously on a single wireless signal. It
carries the transmissions on that signal, freeing network room for the wireless carrier and
providing interference-free calls for the user [5, 7].
2.2.3 Modulation
The modulation technique used in second generation systems is digital modulation. Digital
modulation is a process where the amplitude, frequency or phase of a carrier signal is varied by the
discrete states (on and off) of a digital signal. Amplitude Shift Keying modulation turns the carrier
signal on and off with the digital signal [2, 4, 5]. Frequency Shift Keying modulation shifts the
frequency of the carrier signal according to the on and off levels of the digital information signal.
The phase shift modulator changes the phase of the carrier signal in accordance with the digital
information signal. The standard modulation scheme used in 2G cellular network is Gaussian
Minimum Shift Keying (GMSK). Gaussian Minimum Shift Keying (GMSK) has advantages of
being able to carry digital modulation while still using the spectrum efficiently. One of the
problems with other forms of phase shift keying is that the sidebands extend outwards from the
main carrier and these can cause interference to other radio communications systems using nearby
channels [2, 6]. In view of the efficient use of the spectrum in this way, GMSK modulation has
been used in a number of radio communications applications and is possibly the most widely used
is the GSM cellular technology.
17
Protocol
CDPD protocol was also in use in the second generation network as data transmission method did
not change in the digital shift [1, 5, 22].
2G Services – text messages, picture messages, Fax
Improvements
� Supports roaming through its unified standard.
� Allows for much greater penetration intensity.
� Higher spectrum efficiency than analog systems.
� It holds sufficient security for both the sender and the receiver.
� All text messages are digitally encrypted. This digital encryption allows for the transfer of
data in such a way that only the intended receiver can receive and read it
� Provides voice clarity and reduces noise in the line.
� Digital signals are considered environment friendly.
� Digital encryption has provided secrecy and safety to the data and voice calls.
� Greater network capacity.
� Fewer dropped calls [2, 3, 5, 7].
Limitation
2G systems are still circuit switched networks like the 1G systems and have slow data rate of 9.6
kbps which is not suitable for web browsing and multimedia applications [2, 3, 5].
2.2.4 GSM System Physical Architecture
GSM 2G cellular networks has a hierarchical, complex system architecture comprising many
entities, interfaces, and acronyms which co
(RSS), the network and switching subsystem (NSS), and the operation subsystem (OSS)
Each subsystem will be discussed in more detail in the following sections. Generally, a GSM
customer only notices a very small fraction of the whole network
some antenna masts of the base transceiver stations (BTS
Figure 2.1: Functional architecture of a GSM system
Radio subsystem: The radio subsystem (RSS) comprises all radio specific entities, i.e., the mobile
stations (MS) and the base station s
GSM System Physical Architecture
GSM 2G cellular networks has a hierarchical, complex system architecture comprising many
entities, interfaces, and acronyms which consists of three main subsystems. The
itching subsystem (NSS), and the operation subsystem (OSS)
Each subsystem will be discussed in more detail in the following sections. Generally, a GSM
customer only notices a very small fraction of the whole network – the mobile stations (MS) and
some antenna masts of the base transceiver stations (BTS).
Functional architecture of a GSM system [17]
The radio subsystem (RSS) comprises all radio specific entities, i.e., the mobile
stations (MS) and the base station subsystem (BSS). The RSS and the NSS are connected via the A
18
GSM 2G cellular networks has a hierarchical, complex system architecture comprising many
nsists of three main subsystems. The radio sub system
itching subsystem (NSS), and the operation subsystem (OSS) [6, 17].
Each subsystem will be discussed in more detail in the following sections. Generally, a GSM
the mobile stations (MS) and
The radio subsystem (RSS) comprises all radio specific entities, i.e., the mobile
ubsystem (BSS). The RSS and the NSS are connected via the A
19
interface and the connection to the OSS via the O interface. The A interface is typically based on
circuit-switched Pulse Code Modulation ( PCM) systems, whereas the O interface uses the
Signaling System No. 7 (SS7) based on X.25 carrying management data to/from the RSS [17].
Mobile station (MS): The MS comprises all user equipment and software needed for
communication with a GSM network [3, 4, 17]. An MS consists of user independent hard and
software and of the subscriber identity module (SIM), which stores all user-specific data that is
relevant to GSM. While an MS can be identified via the international mobile equipment identity
(IMEI), a user can personalize any MS using his or her SIM, of which, user-specific mechanisms
like charging and authentication are based on the SIM, not on the device itself. Device-specific
mechanisms, e.g., theft protection, use the device specific IMEI [3, 4, 17]. Without the SIM, only
emergency calls are possible. It also contains many identifiers and tables, such as card-type, serial
number, a list of subscribed services, a personal identity number (PIN), a PIN unblocking key
(PUK), an authentication key, and the international mobile subscriber identity (IMSI). The PIN is
used to unlock the MS while using the wrong PIN three times will lock the SIM, where in such
cases; the PUK is needed to unlock the SIM. The MS stores dynamic information while logged
onto the GSM system, such as, the cipher key and the location information consisting of a
temporary mobile subscriber identity (TMSI) and the location area identification (LAI) [3, 6, 7,
17]. Apart from the telephone interface, an MS can also offer other types of interfaces to users with
display, loudspeaker, microphone, and programmable soft keys, further interfaces comprise
computer modems, Bluetooth etc.
Base station subsystem (BSS): A GSM network comprises many BSSs, each controlled by a base
station controller (BSC). The BSS performs all functions necessary to maintain radio connections
20
to a mobile station (MS), coding/decoding of voice, and rate adaptation to/from the wireless
network part. Besides a BSC, the BSS contains several base transceiver stations (BTS) [3, 7, 17].
Base transceiver station (BTS): A BTS comprises all radio equipment, which are antennas, signal
processing transceivers, and amplifiers necessary for radio transmission. A BTS can form a radio
cell or, using sectorized antennas, several cells and is connected to mobile station via the Um
interface (ISDN U interface for mobile use), and to the BSC via the Abis interface. The Um
interface contains all the mechanisms necessary for wireless transmission (TDMA, FDMA), while
the Abis interface consists of 16 or 64 kbit/s connections. A GSM cell can measure between some
100 m and 35 km depending on the environment but also expected traffic [7, 17].
Base station controller (BSC): The BSC basically manages the BTSs. It reserves radio
frequencies, handles the handoffs from one BTS to another within the BSS, and performs paging
of the MS. The BSC also multiplexes the radio channels onto the fixed network connections at the
A interface [4, 6, 17].
Network and switching subsystem: The core of the GSM system is formed by the network and
switching subsystem (NSS). The NSS connects the wireless network with standard public
networks, performs handovers between different BSSs. It comprises functions for worldwide
localization of users and supports charging, accounting, and roaming of users between different
providers in different countries [3, 17]. The NSS consists of the following switches and databases:
Mobile services switching center (MSC): MSCs are high-performance digital integrated
switched digital network (ISDN) switches, that set up connections to other MSCs and to the BSCs
via the A interface. They form the fixed backbone network of a GSM system, and also manage
several BSCs in a geographical region [3, 6, 17]. An MSC handles all signaling needed for
21
connection setup, connection release and handover of connections to other MSCs. The standard
signaling system No. 7 (SS7) is used for this purpose. SS7 covers all aspects of control signaling
for digital networks, reliable routing and delivery of control messages, establishing and monitoring
of calls. An MSC also performs all functions needed for supplementary services such as call
forwarding, multi-party calls, reverse charging etc [3, 17].
A gateway MSC (GMSC) has additional connections to other fixed networks, such as PSTN and
ISDN. Using additional interworking functions (IWF), it can also connect to public data networks
(PDN) such as X.25 [4, 17].
Home location register (HLR): The HLR is the most important database in a GSM system as it
stores all user-relevant information and these user-specific information elements only exist once
for each user in a single HLR, which also supports charging and accounting. This comprises static
information, such as the mobile subscriber ISDN number (MSISDN), subscribed services (e.g.,
call forwarding, roaming restrictions, GPRS), and the international mobile subscriber identity
(IMSI) [3, 4, 17]. Dynamic information is also needed in the HLR, for instance the current location
area (LA) of the MS, the mobile subscriber roaming number (MSRN), the current VLR and MSC,
such that as soon as an MS leaves its current LA. The information in the HLR is updated, this
information is necessary to localize a user in the worldwide GSM network [6, 17]. HLRs can
manage data for several million customers and contain highly specialized data bases which must
fulfill certain real-time requirements to answer requests within certain time-bounds.
Visitor location register (VLR): The VLR associated to each MSC is a dynamic database which
stores all important information needed for the MS users currently in the LA that is associated to
the MSC (e.g., IMSI, MSISDN, HLR address). If a new MS comes into an LA the VLR is
responsible for, it copies all relevant information for this user from the HLR. This hierarchy of
22
VLR and HLR avoids frequent HLR updates and long-distance signaling of user information,
some VLRs in existence, are capable of managing up to one million customers [3, 17].
Operation subsystem: The third part of a GSM system, the operation subsystem (OSS), contains
the necessary functions for network operation and maintenance. The OSS possesses network
entities of its own and accesses other entities via SS7 signaling [4, 17]. The following entities have
been defined:
Operation and maintenance center (OMC): The OMC monitors and controls all other network
entities via the O interface (SS7 with X.25). Typical OMC management functions are traffic
monitoring, status reports of network entities, subscriber and security management, and accounting
and billing. OMCs use the concept of telecommunication management network (TMN) as
standardized by the ITU-T [3, 6, 17].
Authentication centre (AuC): As the radio interface and mobile stations are particularly
vulnerable, a separate AuC has been defined to protect user identity and data transmission. The
AuC contains the algorithms for authentication as well as the keys for encryption and generates the
values needed for user authentication in the HLR. The AuC may be situated in a special protected
part of the HLR [3, 4, 17].
Equipment identity register (EIR): The EIR is a database for all IMEIs, i.e., it stores all device
identifications registered for this network. It has a blacklist of stolen (or locked) devices, contains a
list of valid IMEIs (white list), and a list of malfunctioning devices (gray list) [17, 44]. As MSs are
mobile, they can be easily stolen, and with a valid SIM anyone could use the stolen MS, so
theoretically an MS is useless as soon as the owner has reported a theft, but unfortunately, the
23
blacklists of different providers are not usually synchronized and the illegal use of a device in
another operator’s network is possible [3, 17, 44].
GSM Interfaces
The communication relationships between the GSM network components are formally described
by a number of standardized interfaces.
Figure 2.2: GSM interfaces [44]
The A interface between BSS and MSC is used for the transfer of data for BSS management, for
connection control and for mobility management.
Within the BSS, the Abis interface between BTS and BSC and the air interface Um have been
defined.
The B interface is used by an MSC which needs to obtain data about an MS staying in its
administrative area, to request the data from the VLR responsible for this area. Conversely, the
MSC forwards to this VLR any data generated at location updates by MSs [4, 6, 44].
24
If the subscriber re-configures special service features or activates supplementary services, the
VLR is also informed first, which then updates the HLR. This updating of the HLR occurs through
the D interface. The D interface is used for the exchange of location-dependent subscriber data and
for subscriber management. The VLR informs the HLR about the current location of the mobile
subscriber and reports the current MSRN [4, 6]. The HLR transfers all of the subscriber data to the
VLR that is needed to give the subscriber their usual customized service access. The HLR is also
responsible for giving a cancellation request for the subscriber data to the old VLR once the
acknowledgement for the location update arrives from the new VLR. If, during location updating,
the new VLR needs data from the old VLR, it is directly requested over the G interface.
Furthermore, the identity of subscriber or equipment can be verified during a location update; for
requesting and checking the equipment identity, the MSC has an interface F to the EIR [6, 17, 44].
2.2.5 GSM Protocol Architecture
Figure 2.3: Protocol architecture for signaling in GSM [17]
25
The figure above shows the protocol architecture of GSM network with signaling protocols,
interfaces, as well as the entities of the physical architecture, but the main interest lies in the Um
interface, as the other interfaces occur between entities in a fixed network.
Layer 1, the physical layer: This handles all radio-specific functions. This includes the creation
of bursts according to the required formats, multiplexing of bursts into a TDMA frame,
synchronization with the BTS, detection of idle channels, and measurement of the channel quality
on the downlink [6, 7, 17]. The physical layer at Um uses Gaussian Minimun Shift Keying
(GMSK) for digital modulation and performs encryption/decryption of data, which means
encryption is not performed end-to-end, but only between MS and BSS over the air interface.
Synchronization also includes the correction of the individual path delay between an MS and the
BTS, as all MSs within a cell use the same BTS and thus must be synchronized to this BTS, so the
BTS generates the time-structure of frames and slots, but a problematic aspect in this context is the
different round trip times (RTT) [7, 17]. This occurs between an MS close to the BTS which has a
very short RTT, whereas an MS about 35km away from the BTS already exhibits about 40% of the
total RTT available for each slot, this will result in large guard spaces, therefore the BTS sends the
current RTT to the MS, which then adjusts its access time so that all bursts reach the BTS within
their limits [6, 7, 17].
The main tasks of the physical layer comprise channel coding and error detection/correction,
which is directly combined with the coding mechanisms. Channel coding makes extensive use of
different forward error correction (FEC) schemes, which adds redundancy to user data, allowing
for the detection and correction of selected errors, and the power of an FEC scheme depends on the
26
amount of redundancy, coding algorithm and further interleaving of data to minimize the effects of
burst errors [4, 7]. The FEC is also the reason why error detection and correction occurs in layer
one and not in layer two as in the ISO/OSI reference model, the GSM physical layer tries to correct
errors, but it does not deliver erroneous data to the higher layer [7, 17].
Layer 2, LAPDm: The link access procedure D-channel modified (LAPDm) protocol has been
defined at the Um interface of layer two, it offers reliable data transfer over connections, re-
sequencing of data frames, and flow control. As there is no buffering between layer one and two,
LAPDm has to obey the frame structures and recurrence patterns defined for the Um interface,
further services provided by include segmentation and reassembly of data and
acknowledged/unacknowledged data transfer which are basic signaling messages [3, 6, 7, 17].
Layer three, network layer: Comprise three sub-layers; that is the radio resource management
(RR), Mobility management (MM), Call management (CM) and only a part of this layer is
implemented in the BTS, the remainder is situated in the BSC, which supports its functions via the
BTS management (BTSM) [6, 7, 8, 17].
Radio Resource Management (RR): The main tasks of RR are setup, maintenance, and release of
radio channels and dedicated connections; it also directly accesses the physical layer for radio
information and offers a reliable connection to the next higher layer and the following functions:
- Channel allocation
- Handover
- Timing advance
- Power control
- Frequency hopping
27
Mobility management (MM): This contains functions for registration, authentication,
identification, location updating, and the provision of a temporary mobile subscriber identity
(TMSI) that replaces the international mobile subscriber identity (IMSI) and which hides the real
identity of an MS user over the air interface. While the IMSI identifies a user, the TMSI is valid
only in the current location area of a VLR. MM offers a reliable connection to the next higher layer
[7, 17].
Call management (CM): This layer contains three entities; call control (CC), short message service
(SMS), and supplementary service (SS). SMS allows for message transfer using the control
channel certain logical channels, while SS offers user identification, call redirection, or forwarding
of ongoing calls, features such as closed user groups and multiparty communication may also be
available [6, 8, 17, 44]. Closed user groups are of special interest to companies because they allow,
for example, a company specific GSM sub-network, to which only members of the group have
access. CC provides a point-to-point connection between two terminals and is used by higher
layers for call establishment, call clearing and change of call parameters [8]. This layer also
provides functions to send in-band tone, called dual tone multiple frequencies (DTMF), over the
GSM network which are used for the remote control of answering machines or the entry of PINs in
electronic banking and are, also used for dialing in traditional analog telephone systems. These
tones cannot be sent directly over the voice codec of a GSM MS, as the codec would distort the
tones, they are transferred as signals and then converted into tones in the fixed network part of the
GSM system [6, 8, 17].
Signaling system No. 7 (SS7) is used for signaling between an MSC and a BSC. This protocol also
transfers all management information between MSCs, HLR, VLRs, AuC, EIR, and OMC. An
MSC can also control a BSS via a BSS application part (BSSAP) [4, 17].
28
2.3 High Speed Circuit Switched Data
The first phase of GSM specifications provided only basic transmission capabilities for the support
of data services, with the maximum data rate in these early networks being limited to 9.6 kbps on
one timeslot [3, 7]. HSCSD was the first improvement of 2+G that clearly increased the achievable
data rates in the GSM system; the maximum radio interface bit rate of an HSCSD configuration
with 14.4-kbps channel coding is 115.2 kbps, which is up to eight times the bit rate on the single-
slot full-rate traffic channel (TCH/F) [3, 7]. Practically, the maximum data rate is limited to 64
kbps owing to core network and A-interface limitations. The main benefit of the HSCSD feature
compared to other data enhancements introduced later is that it is an inexpensive way to implement
higher data rates in GSM networks owing to relatively small incremental modifications needed for
the network equipment. Terminals, however, need to be upgraded to support multi-slot capabilities
[7, 17]. The basic HSCSD terminals with relatively simple implementation can receive up to four
and transmit up to two timeslots and thus support data rates above 50 kbps.
Two types of HSCSD configurations exist at the radio interface which include symmetric and
asymmetric, for both types of configurations, the channels may be allocated on either consecutive
or non-consecutive timeslots, taking into account the restrictions defined by the mobile station’s
multi-slot classes [4, 6, 7]. A symmetric HSCSD configuration consists of a co-allocated bi-
directional TCH/F channel while an asymmetric HSCSD configuration consists of a co-allocated
unidirectional or bi-directional TCH/F channel. A bi-directional channel is a channel on which the
data are transferred in both uplink and downlink directions. On unidirectional channels for
HSCSD, the data is transferred in downlink direction only [7]. The same frequency-hopping
sequence and training sequence is used for all the channels in the HSCSD configuration. In
29
symmetric HSCSD configuration, individual signal level and quality reporting for each HSCSD
channel is applied. For an asymmetric HSCSD configuration, individual signal level and quality
reporting is used for those channels [6, 7, 17].
The quality measurements reported on the main channel are based on the worst quality measured
among the main and the unidirectional downlink timeslots used. In both symmetric and
asymmetric HSCSD configuration, the neighboring cell measurement reports are copied on every
uplink channel used. For n channels, HSCSD requires n times signaling during handover,
connection setup and release, and each channel is treated separately [7]. The probability of
blocking or service degradation increases during handover, as in this case a BSC has to check
resources for n channels, not just one. All in all, HSCSD has been an attractive interim solution for
higher bandwidth and rather constant traffic, example, file download. However, it does not make
much sense for bursty internet traffic as long as a user is charged for each channel allocated for
communication [3, 6, 7]. HSCSD exhibits some major disadvantages because it still uses the
connection-oriented mechanisms of GSM, these are not at all efficient for computer data traffic,
which is typically bursty and asymmetrical; while downloading a larger file may require all
channels reserved, typical web browsing would leave the channels idle most of the time, and
allocating channels is reflected directly in the service costs, as, once the channels have been
reserved, other users cannot use them [7, 17].
2.4 Packet Digital Cellular Systems 2.5G
The circuit-switched bearer services were not particularly well suited for certain types of
applications with a bursty nature because circuit-switched connection has a long access time to the
network, and the call charging is based on the connection time [4, 7]. In packet-switched networks,
30
the connections do not reserve resources permanently, but make use of the on demand allocation,
which is highly efficient, particularly for applications with a bursty nature. Therefore there was
need for an upgrade to a more flexible and powerful data transmission that avoids the problem of
HSCSD, which led to the introduction of general packet radio service (GPRS) standard and
wireless application protocol (WAP) [1, 7]. Wireless Application Protocol (WAP) defines how
Web pages and similar data can be passed over limited bandwidth wireless channels to small
screens being built into new mobile telephones. GPRS defines how to add IP support to the
existing GSM infrastructure as well as provides both a means to aggregate radio channels for
higher data bandwidth and the additional servers required to off-load packet traffic from existing
GSM circuits [1, 7].
The general packet radio service (GPRS) provides packet mode transfer for applications that
exhibit traffic patterns such as frequent transmission of small volumes example, web request or
infrequent transmissions of small or medium volumes like, typical web responses according to the
requirement specification [6, 7, 17]. Compared to existing data transfer services, GPRS uses the
existing network resources more efficiently for packet mode applications, and provides a selection
of QoS parameters for the service requesters; it also allows for broadcast, multicast, and unicast
services. The overall goal in this context is the provision of a more efficient and, thus, cheaper
packet transfer service for internet applications that usually rely solely on packet transfer [7, 17].
Network providers support this model by charging on volume and not on connection time as is
usual for traditional GSM data services and for HSCSD. The main benefit for users of GPRS is the
‘always on’ characteristic – no connection has to be set up prior to data transfer, clearly, GPRS
was driven by the tremendous success of the packet-oriented internet, and by the new traffic
models and applications [1, 7]. For the new GPRS radio channels, the GSM system can allocate
31
between one and eight time slots within a TDMA frame, each time slots are not allocated in a
fixed, pre-determined manner but on demand. All time slots can be shared by the active users; up-
and downlink are allocated separately and also allocation of the slots is based on current load and
operator preferences [7, 17].
Users of GPRS can specify a QoS-profile which determines the service precedence (high, normal,
low), reliability class and delay class of the transmission, and user data throughput, so it adaptively
allocates radio resources to fulfill these user specifications.
2.4.1 GPRS architecture
The GPRS architecture introduces two new network elements, which are called GPRS support
nodes (GSN) and are in fact routers. All GSNs are integrated into the standard GSM architecture,
and many new interfaces have been defined [3, 6, 17].
The gateway GPRS support node (GGSN): This is the interworking unit between the GPRS
network and external packet data networks (PDN). This node contains routing information for
GPRS users, performs address conversion, and tunnels data to a user via encapsulation. The GGSN
is connected to external networks IP or X.25 via the Gi interface and transfers packets to the
serving GSN via an IP-based GPRS backbone network Gn interface [6, 7, 17].
Figure 2.4
Serving GPRS Support Node (SGSN):
(SGSN) which supports the MS via the Gb interface. The SGSN provides a number of functions
within the UMTS network architecture
• Mobility management: When a UE attaches to the Packet Switched domain of the UMTS
Core Network, the SGSN generates MM information based on the mobile's current
location.
• Session management: The SGSN manages the data sessions providing the required quality
of service and also managing what are termed the PDP (Packet data Protocol) contexts, i.e.
the pipes over which the data is sent.
• Interaction with other areas of the network:
within the network only by communicating with other
other circuit switched areas.
• Billing: The SGSN is also responsible billing. It achieves this by monitoring the flow of
user data across the GPRS network. CDRs (Call Detail Records) are generated by the
2.4: GPRS architecture reference model [17]
Serving GPRS Support Node (SGSN): The other new element is the serving GPRS support node
(SGSN) which supports the MS via the Gb interface. The SGSN provides a number of functions
within the UMTS network architecture [6, 7, 17].
When a UE attaches to the Packet Switched domain of the UMTS
Core Network, the SGSN generates MM information based on the mobile's current
The SGSN manages the data sessions providing the required quality
d also managing what are termed the PDP (Packet data Protocol) contexts, i.e.
the pipes over which the data is sent.
Interaction with other areas of the network: The SGSN is able to manage its elements
within the network only by communicating with other areas of the network, e.g. MSC and
other circuit switched areas.
The SGSN is also responsible billing. It achieves this by monitoring the flow of
user data across the GPRS network. CDRs (Call Detail Records) are generated by the
32
The other new element is the serving GPRS support node
(SGSN) which supports the MS via the Gb interface. The SGSN provides a number of functions
When a UE attaches to the Packet Switched domain of the UMTS
Core Network, the SGSN generates MM information based on the mobile's current
The SGSN manages the data sessions providing the required quality
d also managing what are termed the PDP (Packet data Protocol) contexts, i.e.
The SGSN is able to manage its elements
areas of the network, e.g. MSC and
The SGSN is also responsible billing. It achieves this by monitoring the flow of
user data across the GPRS network. CDRs (Call Detail Records) are generated by the
33
SGSN before being transferred to the charging entities (Charging Gateway Function, CGF)
[6, 7, 17].
The SGSN is connected to a BSC via frame relay and is basically on the same hierarchy level as an
MSC. The GR, which is typically a part of the HLR, stores all GPRS-relevant data. GGSNs and
SGSNs can be compared with home and foreign agents, respectively, in a mobile IP network.
Packet data is transmitted from a PDN, via the GGSN and SGSN directly to the BSS and finally to
the MS, before sending any data over the GPRS network, an MS must attach to it, following the
procedures of the mobility management. The attachment procedure includes assigning a temporal
identifier, called a temporary logical link identity (TLLI), and a ciphering key sequence number
(CKSN) for data encryption [7, 8, 17]. For each MS, a GPRS context is set up and stored in the
MS and in the corresponding SGSN, this context comprises the status of the MS (which can be
ready, idle, or standby; the CKSN, a flag indicating if compression is used, and routing data which
includes theTLLI, the routing area RA, a cell identifier, and a packet data channel, PDCH,
identifier. Besides attaching and detaching, mobility management also comprises functions for
authentication, location management, and ciphering which lies between MS and SGSN [8, 17]. In
idle mode an MS is not reachable and all contexts are deleted, while in the standby state only
movement across routing areas is updated to the SGSN but not changes of the cell because
permanent updating would waste battery power and no updating would require system-wide
paging. The update procedure in standby mode is a compromise. Only in the ready state every
movement of the MS is indicated to the SGSN [7, 8, 17].
2.4.2 GPRS Protocol Architecture
The protocol architecture of GPRS introduces new protocols on the transmission plane which were
not available in the protocol architecture of the GSM network. All data within the GPRS backbone
that is between the GSNs is transferred using the GPRS
use two different transport protocols, either the reliable transmission control protocol (TCP)
needed for reliable transfer of X.25 packets or the non
for IP packets [7, 8]. The network protocol for the GPRS back
To adapt to the different characteristics of the underlying networks, the sub
convergence protocol (SNDCP) is used between an SGSN and the MS. On top of SNDCP and
GTP, user packet data is tunneled from t
reliability of packet transfer between SGSN and MS, a special
which comprises address request (
Figure 2.5: GPRS tr
A base station subsystem GPRS protocol (BSSGP) is used to convey routing and QoS
information between the BSS and SGSN; BSSGP does not perform error correction and works on
top of a frame relay (FR) network
transfer data over the Um interface; the radio link protocol (RLC) provides a reliable link, while
is transferred using the GPRS tunneling protocol (GTP)
use two different transport protocols, either the reliable transmission control protocol (TCP)
needed for reliable transfer of X.25 packets or the non-reliable user datagram protocol (UDP) used
. The network protocol for the GPRS backbone is IP using any lower layers.
To adapt to the different characteristics of the underlying networks, the sub-network dependent
convergence protocol (SNDCP) is used between an SGSN and the MS. On top of SNDCP and
GTP, user packet data is tunneled from the MS to the GGSN and vice versa. To achieve a high
reliability of packet transfer between SGSN and MS, a special logical link control
address request (ARQ) and FEC mechanisms for PTP services
GPRS transmission plane protocol reference model [
A base station subsystem GPRS protocol (BSSGP) is used to convey routing and QoS
information between the BSS and SGSN; BSSGP does not perform error correction and works on
twork [3,6, 7, 17]. Radio link dependent protocols are needed to
transfer data over the Um interface; the radio link protocol (RLC) provides a reliable link, while
34
protocol (GTP) [17]. GTP can
use two different transport protocols, either the reliable transmission control protocol (TCP)
reliable user datagram protocol (UDP) used
bone is IP using any lower layers.
network dependent
convergence protocol (SNDCP) is used between an SGSN and the MS. On top of SNDCP and
he MS to the GGSN and vice versa. To achieve a high
logical link control (LLC) is used,
[8, 17].
ansmission plane protocol reference model [17]
A base station subsystem GPRS protocol (BSSGP) is used to convey routing and QoS-related
information between the BSS and SGSN; BSSGP does not perform error correction and works on
. Radio link dependent protocols are needed to
transfer data over the Um interface; the radio link protocol (RLC) provides a reliable link, while
35
the MAC controls access with signaling procedures for the radio channel and the mapping of LLC
frames onto the GSM physical channels [3, 6, 8]. The radio interface at Um needed for GPRS does
not require fundamental changes compared to standard GSM, however, several new logical
channels and their mapping onto physical resources have been defined, for example, one MS can
be allocated up to eight packet data traffic channels (PDTCHs) [17]. Capacity can be allocated on
demand and shared between circuit-switched channels and GPRS, and is done dynamically with
load supervision or alternatively, capacity can be pre-allocated. A very important factor for any
application working end-to-end is that it does not notice any details from the GSM/GPRS-related
infrastructure, the application uses TCP on top of IP, and IP packets are tunneled to the GGSN,
which forwards them into the PDN. All PDNs forward their packets for a GPRS user to the GGSN,
the GGSN asks the current SGSN for tunnel parameters, and forwards the packets via SGSN to the
MS [3, 6, 7, 8, 17]. All MSs are assigned private IP addresses which are then translated into global
addresses at the GGSN; the advantage of this approach is the inherent protection of MSs from
attacks where the subscriber typically has to pay for traffic even if it originates from an attack [7,
17].
2.5 Enhanced data rates for GSM evolution (EDGE)
Enhanced data rates for GSM evolution (EDGE) is a major enhancement to the GSM data rates.
GSM networks have already offered advanced data services, like circuit-switched 9.6-kbpsdata
service and SMS, for some time [7, 6]. High-speed circuit-switched data(HSCSD), with multi-slot
capability and the simultaneous introduction of 14.4-kbps per timeslot data, and GPRS are both
major improvements, increasing the available data rates from 9.6 kbps up to 64 kbps (HSCSD) and
160 kbps (GPRS). EDGE is specified in a way that will enhance the throughput per timeslot for
36
both HSCSD and GPRS [3, 7]. The enhancement of HSCSD is called ECSD (enhanced circuit
switched data), whereas the enhancement of GPRS is called EGPRS (enhanced general packet
radio service). In ECSD, the maximum data rate will not increase from 64 kbps because of the
restrictions in the A-interface, but the data rate per timeslot will triple. Similarly, in EGPRS, the
data rate per timeslot will triple and the peak throughput, with all eight timeslots in the radio
interface, will reach 473 kbps [7, 17].
The enhancement behind tripling the data rates is the introduction of the 8-PSK (octagonal phase
shift keying) modulation in addition to the existing Gaussian minimum shift keying. An 8-PSK
signal is able to carry 3 bits per modulated symbol over the radio path, while a GMSK signal
carries only 1 bit per symbol [6, 7, 17]. The carrier symbol rate of standard GSM is kept the same
for 8-PSK, and the same pulse shape as used in GMSK is applied to 8-PSK. The increase in data
throughput does not come for free, the price being paid in the decreased sensitivity of the 8-PSK
signal. This affects the radio network planning, and the highest data rates can only be provided
with limited coverage. The GMSK spectrum mask was the starting point for the spectrum mask of
the 8-PSK signal, but along the standardization process, the 8-PSK spectrum mask was relaxed
with a few dB in the 400 kHz offset from the centre frequency. This was found to be a good
compromise between the linearity requirements of the 8-PSK signal and the overall radio network
performance [3, 7, 17].
2.6 Third Generation Cellular Network (3G)
3G mobile communications systems arose as a response to the challenge of developing systems
that increased the capacity of the existing 2G systems [4, 5, 10]. This required that the
infrastructure be designed so that it can evolve as technology changes, without compromising the
37
existing services on the existing networks. Separation of access technology, transport technology,
service technology and user application from each other make this demanding requirement
possible [4, 5, 15].
The decision to base 3G specifications on GSM was motivated by widespread deployment of
networks based on GSM standards, the need to preserve some backward compatibility, and the
desire to utilize the large investments made in the GSM networks [10, 14, 15], as a result, despite
its many added capabilities, the 3G core network bears significant resemblance to the GSM
network. 3G is designed to raise the data rate to 2 megabits per second (2 Mbps) – a much higher
rate than 2G and 2.5G, specifically, 3G systems offer between 144 Kbps to 384 Kbps for high-
mobility and high coverage, and 2 Mbps for low-mobility and low coverage applications [1, 3, 10,
15]. In other words, 3G systems mandate data rates of 144 Kbps at driving speeds, 384 Kbps for
outside stationary use or walking speeds, and 2 Mbps indoors which supports wireless web-based
access, E-mail, video teleconferencing and multimedia services consisting of mixed voice and data
streams, and high speed internet access over very wide geographical areas, the frequency allocated
to 3G networks are 1885-2025 MHz for the first band and 2110-2200 MHz for the second band [1,
2, 15]. However, the indoor rate of 2 Mbps from 3G competes with high-speed 802.11 wireless
LANs that offer data rates of 11 to 54 Mbps [4, 15].
The best known example of 3G is the Universal Mobile Telecommunications System (UMTS) – an
acronym used to describe a 3G system that originated in Europe with the overall idea that its users
will be able to use 3G technologies all over the world under different banners. This roaming ability
to use devices on different networks is made possible by satellite and land based networks [2, 10,
15]. UMTS provides a consistent service environment even when roaming via “Virtual Home
Environment” (VHE), a person roaming from his network to other UMTS operators experiences a
38
consistent set of services, independent of the location or access mode. 3G networks use a
connectionless packet-switched communications mechanism where data is split into packets to
which an address uniquely identifying the destination is appended [14, 15]. This mode of
transmission, in which communication is broken into packets, allows the same data path to be
shared among many users in the network, by breaking data into smaller packets that travel in
parallel on different channels, the data rate can be increased significantly [13, 15].
2.6.1 UMTS Radio Interface
The major objectives and requirements of the UMTS network which includes support of general
quality of service, support of multimedia services and support of 2 Mb/s, made the reuse of the
existing technology of the 2G network in the context of 3G networks very difficult [8, 12]. Hence
the need for the development of an entirely new radio interface for the UMTS networks which led
to the proposal and adoption of wideband code division multiple access (WCDMA) by the third
generation partnership project (3GPP) [4, 8, 12]. WCDMA is a wideband direct sequence code
multiple access (DS-CDMA) technology, proposed as the multiple access technology in the FDD
mode of the UMTS terrestrial radio access network (UTRAN) system. In comparison with the
general DS-CDMA systems that have been deployed in the second generation systems, WCDMA
is characterized by a wide bandwidth of 5 MHz and a constant high chip rate of 3.84 Mcps and the
modulation scheme adopted is the quadrature phase shift keying (QPSK) modulation [8, 21]. The
wideband frequency is chosen because it can provide a high data rate required for 3G networks in
good conditions as well as it provides better handoff mechanisms, such as soft handoff for circuit-
switched bearer channels, while the wide bandwidth of the spread spectrum system resolves more
multipath problems and thus improves the system performance [2, 8, 12, 21].
39
2.6.2 UMTS Architecture
The basic structure of the UMTS system is split into three main components: the core network
(CN); the UMTS terrestrial radio access network (UTRAN); and the user equipment (UE), which
is further separated into the access stratum (AS) and non access stratum (NAS). The access stratum
carries all of the signaling and user data messages that relate to the access technology used across a
specific interface in that part of the system [8, 12, 21]. Across the radio interface, the access
stratum protocols are the lower level protocols between the UE and the UTRAN, and between the
UTRAN and the CN, examples of the types of signaling messages that are carried via the access
stratum are messages that control the power control loops in the system, that control the handover
procedures or that allocate channels to a user for use, for instance in a speech call [4, 8, 12]. The
non access stratum carries the signaling messages and user data messages that are independent of
the underlying access mechanism, these signaling and user data are passed between the UE and the
CN, an example of an NAS signaling message is one associated with a call setup request and
management functions, where the call setup messages are independent of the underlying access
mechanism [8, 12, 21].
Figure
The UE consists of two logical entities: the Mobile Equipment (ME) is the actu
used for radio communication over the Uu interface, whereas the UMTS Subscriber Identity
Module (USIM) is a smartcard that contains subscriber identity information and performs
authentication algorithms. The interface between the USIM and
The UTRAN consists of two logical entities and interfaces between them
stations, which are also called Node Bs, convert the data stream from the Uu interface to th
interface. In the first release of
task of the Node B was the inner loop power control, but some of the latest features of WCDMA
have introduced several new functionalities for the Node B to handle. These include for example
packet scheduling, resource allocation, congestion control and retransmission handling in some
cases [8, 17, 21]. The Radio Network Controller (RNC) is a network element that owns and
controls radio resources in its domain, i.e. the
main elements are Home Location Register (HLR), Mobile
Location Register (MSC/VLR), Gateway MSC (GMSC), Serving GPRS Support Node (SGSN)
and Gateway GPRS Support Node
2.6: The UMTS physical architecture [8]
The UE consists of two logical entities: the Mobile Equipment (ME) is the actu
used for radio communication over the Uu interface, whereas the UMTS Subscriber Identity
Module (USIM) is a smartcard that contains subscriber identity information and performs
authentication algorithms. The interface between the USIM and ME is called the Cu interface.
The UTRAN consists of two logical entities and interfaces between them [12
which are also called Node Bs, convert the data stream from the Uu interface to th
interface. In the first release of the UMTS standard, the only radio resource management related
task of the Node B was the inner loop power control, but some of the latest features of WCDMA
have introduced several new functionalities for the Node B to handle. These include for example
et scheduling, resource allocation, congestion control and retransmission handling in some
. The Radio Network Controller (RNC) is a network element that owns and
resources in its domain, i.e. the base stations connected to it. In the core network, the
Location Register (HLR), Mobile Services Switching Centre / Visitor
Location Register (MSC/VLR), Gateway MSC (GMSC), Serving GPRS Support Node (SGSN)
and Gateway GPRS Support Node (GGSN) [8, 12, 21].
40
The UE consists of two logical entities: the Mobile Equipment (ME) is the actual radio terminal
used for radio communication over the Uu interface, whereas the UMTS Subscriber Identity
Module (USIM) is a smartcard that contains subscriber identity information and performs
ME is called the Cu interface.
[12, 21]. The base
which are also called Node Bs, convert the data stream from the Uu interface to the Iub
the only radio resource management related
task of the Node B was the inner loop power control, but some of the latest features of WCDMA
have introduced several new functionalities for the Node B to handle. These include for example
et scheduling, resource allocation, congestion control and retransmission handling in some
. The Radio Network Controller (RNC) is a network element that owns and
In the core network, the
Services Switching Centre / Visitor
Location Register (MSC/VLR), Gateway MSC (GMSC), Serving GPRS Support Node (SGSN)
UMTS Network Domain
Fig
User Equipment Domain
The User Equipment domain consists of the terminal that allows the user access to the mobile
services through the radio interface, which is further split into two sub domain;
Mobile Equipment (ME) domain:
is sub-divided into the Mobile Termination (MT) entity, which performs the radio transmission
and reception, and the Terminal Equipment (TE), which contains the ap
two entities may be physically located at the same hardware device depending
application, for example, in the case of a handset used for a speech application, both MT and TE
are usually located in the handset, whil
application, the handset will contain the MT and
contains the web browser [11, 17, 21
Universal Subscriber Identity Module (USIM) domain:
containing the USIM is a removable smart card. The USIM contains the identi
Figure 2.7: UMTS network domain [17]
The User Equipment domain consists of the terminal that allows the user access to the mobile
ces through the radio interface, which is further split into two sub domain;
bile Equipment (ME) domain: This represents the physical entity being a handset
divided into the Mobile Termination (MT) entity, which performs the radio transmission
and reception, and the Terminal Equipment (TE), which contains the applications
two entities may be physically located at the same hardware device depending
or example, in the case of a handset used for a speech application, both MT and TE
are usually located in the handset, while if the same handset is being used for a web browsing
application, the handset will contain the MT and the TE can reside in an external device
, 17, 21].
r Identity Module (USIM) domain: The physical har
containing the USIM is a removable smart card. The USIM contains the identi
41
The User Equipment domain consists of the terminal that allows the user access to the mobile
sents the physical entity being a handset that in turn
divided into the Mobile Termination (MT) entity, which performs the radio transmission
plications [11, 17]. These
two entities may be physically located at the same hardware device depending on the specific
or example, in the case of a handset used for a speech application, both MT and TE
e if the same handset is being used for a web browsing
the TE can reside in an external device that
he physical hardware device
containing the USIM is a removable smart card. The USIM contains the identification of the
42
profile of a given user, including his identity in the network as well as information about the
services that this user is allowed to access depending on the contractual relationship with the
mobile network operator. So, the USIM is specific for each user and allows him to access the
contracted services in a secure way by means of authentication and encryption procedures
regardless of the ME that is used [8, 11, 12 , 21]. The USIM card contains all the data relating to
the subscriber, including the following:
• The International Mobile Station Identifier (IMSI);
• The Mobile Station International ISDNN umber (MSISDN);
• The preferred language, used for broadcast information and for terminal menu options;
• The encryption and integrity keys for the circuit switched and packet switched domains.
• The list of forbidden networks;
• The user's temporary identities vis-a-vis the circuit switched and packet switched domains;
• The identities of the current location area and routing area of the mobile for the circuit
switched and packet switched domains respectively [2, 12, 17].
Infrastructure Domain
The infrastructure domain in the UMTS architecture contains the physical nodes that terminate the
radio interface allowing the provision of the end-to-end service to the UE. In order to separate the
functionalities that are dependent on the radio access technology being used from those that are
independent, the infrastructure domain is in turn split into two domains, namely the Access
Network and the Core Network domains separated by the Iu reference point [11, 17]. This allows
there to be a generic UMTS architecture that enables the combination of different approaches for
the radio access technology as well as different approaches for the core network. With respect to
the core network, and in order to take into account different scenarios in which the user
43
communicates with users in other types of networks, that is, other mobile networks, fixed
networks, Internet, etc., three different sub-domains are defined:
Home Network domain: This corresponds to the network to which the user is subscribed, so it
belongs to the operator that has the contractual relationship with the user. The user service profile
as well as the user secure identification parameters are kept in the home network and should be
coordinated with those included in the USIM at the UE [11, 17].
Serving Network domain: This represents the network containing the access network to which
the user is connected in a given moment and it is responsible for transporting the user data from
the source to the destination [11, 12, 17]. Physically, it can be either the same home network or a
different network in the case where the user is roaming with another network operator. The serving
network is then connected to the access network through the Iu reference point and to the home
network through the Zu reference point. The interconnection with the home network is necessary
in order to retrieve specific information about the user service abilities and for billing purposes [11,
17].
Transit Network domain: This is the core network part located on the communication path,
between the serving network and the remote party, and it is connected to the serving network
through the Yu reference point [4, 11]. Where the remote party belongs to the same network to
which the user is connected, the serving network and the transit network are physically the same
network, and sin general, the transit network may not be a UMTS network, for example, in the
case of a connection with a fixed network or when accessing the Internet [11, 17].
44
2.6.3 Universal Terrestrial Radio Access Network (UTRAN)
The UTRAN is composed of Radio Network Subsystems (RNSs) that are connected to the Core
Network through the Iu interface that coincides with the Iu reference point of the overall UMTS
architecture. Each RNS is responsible for the transmission and reception over a set of UMTS cells
where the connection between the RNS and the UE is done through the Uu or radio interface [8,
11, 12, 17]. The RNSs comprises a number of Nodes B and one Radio Network Controller (RNC),
connected through Iub interfaces and RNCs belonging to different RNSs are interconnected by
means of the Iur interface.
Main Requirements for UTRAN:
• The major impact on the design of UTRAN has been the requirement to support soft
handover one terminal connected to the network via two or more active cells and the
WCDMA-specific Radio Resource Management algorithms.
• The maximization of the commonalities in the handling of packet-switched and circuit-
switched data, with a unique air interface protocol stack and with the use of the same
interface for the connection from UTRAN to both the PS and CS domains of the core
network.
• The maximization of the commonalities with GSM networks, when possible.
• Use of the ATM transport as the main transport mechanism in UTRAN [7, 17, 21]
45
Figure 2.8: UTRAN architecture [11]
UTRAN Frequency Division Duplex mode: In this mode, the uplink and downlink transmit with
different carrier frequencies, thus requiring the allocation of paired bands. The access technique
being used is WCDMA, which means that several transmissions in the same frequency and time
are supported and can be distinguished by using different code sequences [4, 7, 11].
UTRAN Time Division Duplex mode: The uplink and downlink operate with the same carrier
frequency in this mode but in different time instants, thus they are able to use unpaired bands. The
access technique being used is a combination of TDMA and DS-CDMA, which means that
simultaneous transmissions are distinguished by different code sequences (DS-CDMA component)
and that a frame structure is defined to allocate different transmission instants (time slots) to the
different users (TDMA component) [11].
46
2.6.3.1 Node B
The UTRAN Node B is equivalent to the BTS in GSM networks. Its main role is to provide radio
reception and transmission for one or more of the UTRAN cells. The technical implementation and
the internal architecture of the Node B are left to the manufacturer and thus, one can conceive of
Node Bs made up of one or several cells, using omni-directional or sectorial antennae [2, 7, 11, 17,
22]. A node B is the termination point between the air interface and the network and it is composed
of one or several cells or sectors, a cell stands as the smallest radio network entity that has its own
identification number, denoted as Cell ID. Conceptually, a cell is regarded as a UTRAN Access
Point through which radio links with the UEs are established, from a functional point of view, the
cell executes the physical transmission and reception procedures over the radio interface [11, 17,
22]. Node B controls the data flow between the Uu and lub interfaces, it performs the air interface
Layer1 processing such as, channel coding and interleaving, rate adaptation, spreading, it extracts
the MAC protocol data units, and transports them across the lub interface to the RNC and also
participates in radio resource management operations such as the inner loop power control [12,
22].
2.6.3.2 The Radio Network Controller
The Radio Network Controller (RNC) is the network element responsible for the control of the
radio resources of UTRAN where the UMTS Radio Resource Management (RRM) algorithms are
executed. On the network side, the RNC interoperates with the core network through the Iu
interface and establishes, maintains and releases the connections with the core network elements
that the UEs under its control require in order to receive the UMTS services, it also terminates the
Radio Resource Control (RRC) protocol that defines the messages and procedures between the
mobile and UTRAN. It logically corresponds to the GSM BSC [11, 12, 22].
47
Functions of the RNC
● Call admission control: It is very important for WCDMA systems to keep the interference below
a certain level. The RNC calculates the traffic within each cell and decides, if additional
transmissions are acceptable or not.
● Congestion control: During packet-oriented data transmission, several stations share the
available radio resources. The RNC allocates bandwidth to each station in a cyclic fashion and
must consider the QoS requirements [12, 17].
● Encryption/decryption: The RNC encrypts all data arriving from the fixed network before
transmission over the wireless link and vice versa.
● ATM switching and multiplexing, protocol conversion: Typically, the connections between
RNCs, node Bs, and the core network are based on ATM. An RNC has to switch the connections
to multiplex different data streams.
● Radio resource control: The RNC controls all radio resources of the cells connected to it via a
node B. This task includes interference and load measurements. The priorities of different
connections have to be obeyed [11, 12, 17].
● Radio bearer setup and release: An RNC has to set-up, maintain, and release a logical data
connection to a UE (the so-called UMTS radio bearer).
● Code allocation: The WCDMA codes used by a UE are selected by the RNC. These codes may
vary during a transmission.
● Power control: The RNC only performs a relatively loose power control of the outer loop. This
means that the RNC influences transmission power based on interference values from other cells
or even other RNCs. But this is not the tight and fast power control performed 1,500 times per
48
second. This is carried out by a node B. This outer loop of power control helps to minimize
interference between neighboring cells or controls the size of a cell [11, 12, 17].
● Handover control and RNS relocation: Depending on the signal strengths received by UEs and
node Bs, an RNC can decide if another cell would be better suited for a certain connection. If the
RNC decides for handover it informs the new cell and the UE as explained in subsection 4.4.6. If a
UE moves further out of the range of one RNC, a new RNC responsible for the UE has to be
chosen. This is called RNS relocation.
● Management: The network operator needs a lot of information regarding the current load,
current traffic, error states etc. to manage its network. The RNC provides interfaces for this task as
well [12, 17].
Logical Role of the RNC
In case one mobile to UTRAN connection uses resources from more than one RNS the RNCs
involved have two separate logical roles but one RNC normally contains all functionality.
Controlling RNC: This is the role with respect to the Node B, the RNC controlling one Node B is
indicated as the Controlling RNC (CRNC) of the Node B which is responsible for the load and
congestion control of its own cells, and also executes the admission control and code allocation for
new radio links to be established in those cells [4, 11, 12].
Serving RNC: This role is taken with respect to the UE, the SRNC is the RNC that holds the
connection of a given UE with the CN through the Iu interface. It can be regarded as the RNC that
controls the RNS to which the mobile is connected at a given moment [8, 11, 12, 17]. When the
UE moves across the network and executes handover between the different cells, it may require a
SRNS that is, the RNS having the SRNC relocation procedure when the new cell belongs to a
different RNC. This procedure requires the communication between the SRNC and the new RNC
49
through the Iur interface in order for the new RNC to establish a new connection with the CN over
its Iu interface [8, 11, 17]. The SRNC also terminates the Radio Resource Control Signaling, that
is, the signaling protocol between the UE and UTRAN; it performs the Layer2 processing of the
data to and from the radio interface. Basic Radio Resource Management operations, such as the
mapping of Radio Access Bearer (RAB) parameters into air interface transport channel parameters,
the handover decision, and outer loop power control, are executed in the SRNC. The SRNC may
also but not always be the CRNC of some Node B used by the mobile for connection with UTRAN
[12, 22].
Drift RNC: This role is also taken with respect to the UE and is a consequence of a specific type
of handover that exists with WCDMA systems, denoted as soft handover. In this case, a UE can be
simultaneously connected to several cells, that is, having radio links with several cells, then, when
the UE moves in the border between RNSs, it is possible that it establishes new radio links with
cells belonging to a new RNC while at the same time keeping the radio link with some cells of the
SRNC [8, 11, 12, 22]. The new RNC takes the role of DRNC, and the connectivity with the core
network is not done through the Iu of the DRNC but still through the Iu of the SRNC, thus
requiring it to establish resources for the UE in the Iur interface between SRNC and DRNC. Only
when all the radio links of the old RNC are released and the UE is connected only to the new RNC,
will the SRNS relocation procedure is executed [11, 12,].
50
Figure 2.9: Logical role of RNC [2]
2.6.4 UMTS Core Network
While the UMTS radio interface, WCDMA, represent a bigger step in the radio access evolution
from GSM networks, the UMTS core network did not experience major changes, both UTRAN
and GPRS/EDGE (GERAN) based radio access network connect to the same core network [2, 8,
12]. UMTS core network has two domains: Circuit Switched (CS) domain and Packet Switched
(PS) domain, to cover the need for different traffic types, the division comes from the different
requirements for the data, depending on whether it is real time (circuit switched) or non-real time
(packet data). However, it should be understood that several functionalities can be implemented in
a single physical entity and all entities do not necessarily exist as separate physical units in real
network [2, 12].
The Core Network is the part of the mobile network infrastructure that covers all the functionalities
that are not directly related with the radio access technology, thus it is possible to combine
different core network architectures with different radio access networks. Examples of these
functionalities are the connection and session management, which includes establishment,
51
maintenance and release of the connections and sessions for circuit switched and packet switched
services, as well as mobility management which includes keeping track of the area where each UE
can be found in order to route calls to it [2, 8, 11, 12]. The initial implementation of UMTS was
seen simply as an extension of the GSM/GPRS networks because they maintained the existing core
network for GSM/GPRS with small modifications in order to make it compatible with the new
UMTS access network but is now trending to an all IP core network [8, 11].
Figure 2.10: UMTS core network architecture [2]
Circuit Switched domain: The circuit switched domain supports the traffic composed by
connections that require dedicated network resources, and allows the interconnection with external
CS networks like the Public Switched Telephone Network (PSTN) or the Integrated Services
Digital Network (ISDN), the Iu reference point between core and access networks in this interface
52
is denoted as Iu_CS [2, 11, 12]. The circuit switched domain is composed of three specific entities,
namely the MSC, the GMSC and the VLR. The MSC interacts with the radio access network by
means of the Iu_CS interface and executes the necessary operations to handle circuit switched
services. This includes routing the calls towards the corresponding transit network and establishing
the corresponding circuits in the path [2, 4, 11].
The MSC is the same as that which is used in the GSM network. The only difference being that a
specific interworking function (IWF) is required between the MSC and the access network in
UMTS. The reason is that in GSM the speech traffic delivered to the core network by the access
network uses 64kb/s circuits while in UMTS the speech uses adaptive multi-rate technique (AMR)
with bit rates between 4.75kb/s and 12.2kb/s. These are transported in the access network with
Asynchronous Transfer Mode (ATM) technology [2, 8, 11]. This is why the term 3G MSC is
sometimes used to differentiate between the MSC from GSM system and the MSC from UMTS
networks.
The VLR is a database associated with a MSC that contains specific information like identifier,
location information, etc. about the users that are currently in the area of this MSC. This allows the
performing of certain operations without the need to interact with the HLR. The information
contained in the VLR and the HLR must be coordinated [3, 7, 11]. The GMSC is a specific MSC
that interfaces with the external circuit switched networks and is responsible of routing calls to and
from the external network, to this end. It interacts with the HLR to determine the MSC through
which the call should be routed. In WCDMA, the communication between the entities of the
circuit switched domain is done by means of 64kb/s circuits and uses Signaling System No. 7
(SS7) for signaling purposes [2, 3, 8, 11].
53
Packet Switched domain: The PS domain supports a traffic composed of packets, which are
groups of bits that are autonomously transmitted and independently routed. No dedicated resources
are required throughout the connection time, since the resources are allocated on a packet basis
only when needed [2, 11, 22]. This allows a group of packet flow to share the network resources
based on traffic multiplexing and also allows the interconnection of external PS networks, like the
Internet. The Iu reference point between core and access networks in this interface is denoted as
Iu_PS. The PS domain is composed of two specific entities, namely the SGSN and GGSN, which
perform the necessary functions to handle packet transmission to and from the UEs. The SGSN is
the node that serves the UE and establishes a mobility management context including security and
mobility information. It interacts with the UTRAN by means of the Iu_PS interface. The GGSN, in
turn, interfaces with the external data networks and contains routing information of the attached
users. IP tunnels between the GGSN and the SGSN are used to transmit the data packets of the
different users [2, 8, 11, 12, 22].
2.6.5 UMTS Interfaces
Cu interface: This is the electrical interface between the USIM smartcard and the ME. The
interface follows a standard format for smartcards.
Uu interface: This is the WCDMA radio interface, through which the UE accesses the fixed part
of the system, and is therefore probably the most important open interface in UMTS.
Iu interface: This connects UTRAN to the core network, similarly to the corresponding interfaces
in GSM.
Iur interface: The open Iur interface allows soft handover between RNCs from different
manufacturers, and therefore complements the open Iu interface.
54
Iub interface: The Iub connects a Node B and an RNC [8, 12, 22].
2.6.6 UMTS Radio Interface Protocol Architecture
The protocol architecture of the UMTS network that exists across the Uu interface between the UE
and the radio access network is separated into a control plane on the left. It is responsible for the
transmission of the control signaling messages and the user plane on the right is responsible for the
transmission of the user data messages such as speech and packet data [4, 8, 11].
The radio interface is composed of Layers 1, 2 and 3. The lowest layer, Layer 1, is the physical
layer, which is based on WCDMA technology, Layer 2, is split into four sub-layers: Medium
Access Control (MAC), Radio Link Control (RLC), Packet Data Convergence Protocol (PDCP)
and Broadcast/Multicast Control (BMC). Furthermore, Layer 3 is divided into control plane and
user plane and, together with Layer 2 [8, 11, 12].
Figure 2.11: UMTS radio interface protocol architecture [12]
55
2.6.6.1 Layer 1
In the protocol model this layer takes care of the actual transmission of data across the radio path,
which is also the case in the well-known OSI (Open Systems Interconnection) reference model.
The physical layer is the lowest data transmission layer and it only provides the means of
transmitting raw bits over the physical data link. It also includes tasks like forward error correction
(channel coding), interleaving, error detection (CRC), closed loop power control and
synchronization [2, 8, 11, 12, 17 40].
The physical layer interfaces the MAC sub-layer of Layer 2 (the data link layer) and offers so
called transport channels (TrCH) as a service to MAC. A transport channel is characterized by how
the information is transferred over the radio interface; the MAC layer offers logical channels as a
service to the RLC sub-layer of Layer 2. A logical channel is characterized by the type of
information transferred [2, 12, 40]. It also interfaces the Radio Resource Control (RRC) layer of
Layer 3 (the network layer), which can be used for controlling the physical layer. The RRC
terminates in the UTRAN and this protocol contains all procedures to control, modify and release
Radio Bearers (RB), its messages use radio bearer services offered by Layer 2 for transport [2, 8,
12, 40].
Specific functions of this layer include; RF processing aspects, chip rate processing, symbol rate
processing and transport channel combination [8, 12, 40]. In the transmit direction, the physical
layer receives blocks of data from the higher layers. It transports blocks via transport channels
from the MAC layer and multiplexes them onto a physical channel. In the receive direction, the
physical layer receives the physical channels, extracts and processes the multiplexed data and
delivers it up to the MAC. Within the WCDMA system, the physical channels are constructed
using special codes referred to as channelization codes and scrambling codes [2, 8, 12, 40].
56
2.6.6.2 Layer 2
This Layer comprises the four protocols namely MAC protocol, RLC protocol, PDCP protocol and
BMC protocols which are discussed in detail below.
The Medium Access Control (MAC) Protocol
The MAC provides some important functions within the overall radio interface architecture. It is
responsible for the dynamic resource allocation under the control of the RRC layer. Part of the
resource allocation requires the MAC to use relative priorities between services to control the
access to the radio interface transmission resources [2, 8, 12]. These functions comprise the
mapping between the logical and the transport channels, the transport format selection and priority
handling of data flow. It is also responsible for UE identification management in order to facilitate
transactions such as random access attempts and the use of downlink common channels [2, 8, 12].
When the RLC is operating in transparent mode, that is when the data are passing through the RLC
layer without any header information and when ciphering is enabled, it is the function of the MAC
actually to perform the ciphering task.
The MAC is also responsible for traffic volume measurements across the radio interface. To
achieve this, it monitors the buffer levels for the different RLC instances that are delivering data to
it [4, 8, 11]. There are multiple channels entering the MAC referred to as logical channels and
there are multiple channels leaving the MAC referred to as transport channels. The number of
logical channels coming in and the number of transport channels leaving are not necessarily the
same, but it provides a multiplexing function that result in different logical channels being mapped
onto the same transport channels [2, 8, 11, 12].
57
The Radio Link Control (RLC) Protocol
The RLC provides a number of different types of transport service, the transparent, the
unacknowledged, or the acknowledged mode of data transfer. Each mode has a different set of
services that define the use of that mode to the higher layers [8, 11, 12, 17]. Services provided by
the RLC include segmentation and reassembly. This allows the RLC to segment large protocol
data unit (PDUs) into smaller PDUs. A concatenation service is also provided to allow a number of
PDUs to be concatenated. The acknowledged mode data transfer service provides a very reliable
mechanism for transferring data between two peer RLC entities [8, 11, 12]. It also provides flow
control and in-sequence delivery of PDUs. Error correction is provided by an automatic repeat
request (ARQ) system, where PDUs identified as being in error can be requested to be
retransmitted. Flow control is the procedure by which the transfer of PDUs across the radio
interface can be governed to prevent buffer overload. For instance at the receiving end, the ARQ
system that is used could result in PDUs arriving out of sequence, the sequence number in the
PDU can be used by the RLC to ensure that all PDUs arrive in the correct order [8, 11, 12, 40].
Packet Data Convergence Protocol (PDCP)
The PDCP layer is defined for use with the PS domain only, at the inputs to the PDCP layer are the
PDCP service access points (SAPs). The PDCP layer for WCDMA provides header compression
(HC) functions and support for lossless SRNS relocation. Lossless SRNS relocation is used when
the SRNC is being changed, and it is required that no data are lost; data that are not acknowledged
as being correctly received are retransmitted once the new SRNC is active [8, 11].
PDCP Functions
Compression of redundant protocol control information (TCP/IP and RTP/UDP/IP headers) at the
transmitting entity, and decompression at the receiving entity. The header compression method is
58
specific to the particular network layer, transport layer or upper layer protocol combinations, for
example TCP/IP and RTP/UDP/IP [8, 12, 22].
Transfer of user data: This means that the PDCP receives a PDCP SDU from the non-access
stratum and forwards it to the appropriate RLC entity and vice versa.
Support for lossless SRNS relocation: In practice this means that those PDCP entities which are
configured to support lossless SRNS relocation have PDU sequence numbers, which, together with
unconfirmed PDCP packets are forwarded to the new SRNC during relocation. Only applicable
when PDCP is using acknowledged mode RLC with in-sequence delivery [2, 8, 12, 22].
Broadcast and Multicast Control (BMC) Protocol
The BMC provides support for the cell broadcast SMS, its messages are received on a common
physical channel. The messages are periodic with a periodicity defined by parameters that are
broadcast to the UE in system information broadcast (SIB) messages. The UE is able to select and
filter the broadcast messages according to settings defined by the user [4, 8, 11].
Storage: The BMC in RNC stores the Cell Broadcast messages received over the cell broadcast
centre (CBC)–RNC interface for scheduled transmission.
Traffic volume monitoring and radio resource request for CBS: On the UTRAN side, the BMC
calculates the required transmission rate for the Cell Broadcast Service based on the messages
received over the CBC–RNC interface, and requests appropriate CTCH/ FACH resources from
RRC [8, 12, 17, 22].
Scheduling: BMC protocol receives scheduling information together with each Cell Broadcast
message over the CBC –RNC interface. Based on this scheduling information, on the UTRAN side
the BMC generates schedule messages and schedules BMC message sequences accordingly. On
the UE side, the BMC evaluates the schedule messages and indicates scheduling parameters to
59
RRC, which are used by RRC to configure the lower layers for CBS discontinuous reception.
Transmission of BMC messages to UE. This function transmits the BMC messages (Scheduling
and Cell Broadcast messages) according to the schedule. Delivery of Cell Broadcast messages to
the upper layer. This UE function delivers the received non-corrupted Cell Broadcast messages to
the upper layer [8, 11, 12, 22].
2.6.6.3 Layer 3
The Radio Resource Control (RRC) protocol
The RRC protocol resides in the layer3 of the architecture and is responsible for the establishment,
modification and release of radio connections between the UE and the UTRAN. The radio
connections are commonly referred to as the RRC connection, which is used to transfer RRC
signaling messages. It also provides transportation services for the higher layer protocols that use
the connections created by the protocol [8, 11, 22]. In the UE there is a single instance of the RRC
protocol while in the UTRAN there are multiple instances of the protocol, one per UE, this RRC
entity in the UE receives its configuration information from the RRC entity in the UTRAN. In
addition to establishing an RRC connection that is used by the various sources of signaling, the
RRC protocol is also responsible for the creation of user plane connections, referred to as radio
access bearers (RABs), these RABs are created to transport user plane information such as speech
or packet data across the radio interface from the UE to the core network [8, 11]. The RRC
protocol provides radio mobility functions including elements such as the control of soft-handover
to same-frequency UMTS cells, hard-handover to other UMTS cells and hard-handover to other
radio access technology (RAT) cells such as GSM. Cell updates and UTRAN registration area
(URA) updates are procedures that are used to allow the UTRAN to track the location of the UE
within the UTRAN [8, 11, 12, 17].
60
2.7 WCDMA Concepts
WCDMA is a wideband Direct-Sequence Code Division Multiple Access (DS-CDMA) system,
where user information bits are spread over a wide bandwidth by multiplying the user data with
quasi-random bits (called chips) derived from CDMA spreading codes. In order to support very
high bit rates up to 2 Mbps, the use of a variable spreading factor and multi-code connections is
supported [8, 12, 23]. The physical aspects of the WCDMA air interface, is characterized by the
flow of information at 3.84 Mega chips per second (Mcps) which is be divided into 10 ms radio
frames, each further divided into 15 slots of 2560 chips. The notion of chips is introduced instead
of the more typical bits, because chips are the basic information units in WCDMA, where bits
from the different channels are coded by representing each bit by a variable number of chips and
what each chip represents depends on the channel [8, 23]. The fundamental concept in WCDMA
includes channelization and scrambling, channel coding, power control, and handover.
WCDMA physical layer and air interface
When comparing different cellular systems with each other, the physical layer of the radio
interface typically contains most of the differences and is therefore the most interesting part of the
study. In the OSI reference model, the physical layer is the lowest layer and it includes the
transmission of signals and the activation and deactivation of physical connections. The physical
layer has a major impact on equipment complexity with respect to the required baseband
processing power in the terminal and in the base station equipment [8, 12, 23].
WCDMA technology also introduces new challenges to the implementation of the physical layer,
as third generation systems are wideband from the service point of view as well, the physical layer
needs to be designed to support various different services. More flexibility is also needed for future
service introduction [23, 41]
61
The physical layer offers data transport services to higher layers and it is designed to support
variable bit rate transport channels, to offer so-called bandwidth-on-demand services and to be
able to multiplex several services within the same Radio Resource Control (RRC) connection [8,
23, 43].
The basic idea in WCDMA is that the signal to be transferred over the radio path is formed by
multiplying the original baseband digital signal with another signal, which has a much greater bit
rate. This operation is called channelization and the number of chips per data symbol is called the
Spreading Factor (SF) [12, 23]. We need to make a clear separation between the different kinds of
bits in WCDMA, one bit of baseband digital signal, the actual information, is called a symbol. On
the other hand, one bit of code signal used for signal multiplying is called a chip. The code signal
bit rate, i.e. the chip rate, is fixed in WCDMA being 3.84 million chips per second (3.84 Mcps)
[23, 41, 43]. The symbol rate indicates how many data symbols are transferred over the radio path
and it is expressed as kilosymbols per seconds (ks/s).
Differences between WCDMA and Second Generation Air Interfaces
The second generation systems were built mainly to provide speech services in macro cells [4, 12].
To understand the background to the differences between second and third generation systems, we
will look at the new requirements of the third generation systems which are listed below:
• Bit rates up to 2 Mbps;
• Variable bit rate to offer bandwidth on demand;
• Multiplexing of services with different quality requirements on a single connection, e.g.
speech, video and packet data;
• Delay requirements from delay-sensitive real time traffic to flexible best-effort packet data;
• Quality requirements from 10 % frame error rate to 10-6 bit error rate;
62
• Co-existence of second and third generation systems and inter-system handovers for
coverage enhancements and load balancing;
• Support of asymmetric uplink and downlink traffic, e.g. web browsing causes more loading
to downlink than to uplink;
• High spectrum efficiency;
• Co-existence of FDD and TDD modes [12, 23, 41, 43].
Table 2.1: Differences between WCDMA and GSM air interfaces [12] WCDMA GSM Carrier spacing 5 MHz 200 kHz
Frequency reuse factor 1 1–18
Power control frequency 1500 Hz 2 Hz or lower
Quality control Radio resource management algorithms
Network planning (frequency planning)
Frequency diversity 5 MHz bandwidth gives multipath diversity with Rake receiver
Frequency hopping
Packet data Load-based packet scheduling
Time slot based scheduling with GPRS
Downlink transmit diversity Supported for improving downlink capacity
Not supported by the standard, but can be applied
WCDMA Service Capability
WCDMA does not use the same principle as GSM with terminal class mark. The terminals tell the
network, upon connection set-up, a set of parameters indicating the radio access capabilities of the
particular terminals. This determines the maximum user data rate supported in a particular radio
configuration, given independently for the uplink and downlink directions [12, 23, 41].
63
• 32 kbps class. This is intended to provide a basic speech service, including AMR speech as
well as some limited data rate capabilities up to 32 kbps.
• 64 kbps class. This is intended to provide a speech and data service, with simultaneous data
and AMR speech capability.
• 144 kbps class. This class has the air interface capability to provide, for example, video
telephony or various other data services [12, 23].
• 384 kbps class is being further enhanced from 144 kbps and has, for example, multicode
capability, which points toward support of advanced packet data methods provided in
WCDMA.
• 768 kbps class has been defined as an intermediate step between 384 kbps and 2 Mbps
class.
• 2 Mbps class. This is the state-of-the-art class and has been defined for the downlink
direction only but also possible for uplink [8, 12, 23].
2.7.1 Power Control
In WCDMA technology, power control is critical because it ensures that just enough power is used
to close the links, either downlink, from the Node B to the mobile device, or uplink, from the
mobile to the Node B. Of the two links, the uplink is more critical because it ensures that all
instances of UE are detected at the same power by the cell; thus each UE contributes equally to the
overall interference and no single UE will overpower and consequently desensitize the receiver [8,
12, 23]. Without power control, a single UE transmitting at full power close to the Node B would
be the only one detected while the others would be drowned out by the strong signal of the close
user who creates a disproportionate amount of interference. On the downlink, power control serves
64
a slightly different purpose, because the Node B’s power must be shared among common channels
and the dedicated channels for all active users [12, 23, 41]. Similarly, all channels are orthogonal
to each other with the exception of the synchronization channel; thus the signal, or power, from
any channel is not seen as interference. Ideally, the other channels do not affect the sensitivity;
however, power control is still required to ensure that a given channel is using only the power that
it needs which increases the power available for other users, effectively increasing the capacity of
the system. Conceptually, two steps are required for power control:
• Estimate the minimum acceptable quality.
• Ensure that minimum power is used to maintain this quality.
Outer loop power control handles the first step while inner loop handles the second [8, 12, 23].
Ideally, the outer loop should monitor the Block Error Rate (BLER) of any established channel and
compare it to the selected target. If they differ, the quality target, estimated in terms of Signal-to-
Interference Ratio (SIR), is adjusted. The closed loop power control can then compare, on a slot-
by-slot basis, the measured and target SIR, and send power-up or power-down commands. Power
control processes run independently in the uplink and downlink, each signaling to the other the
required adjustment by means of Transmit Power Control (TPC) bits [8, 23]. The outer loop, on
the other hand, is not as strictly controlled by the standard and is thus implementation-dependent:
neither its rate nor the step sizes are signaled to the other end. Moreover, although the purpose of
the closed loop is to ensure that the BLER target is met, the implementation may be based on other
measurements such as SIR, or passing or failing the Cyclic Redundancy Check (CRC) [8, 12, 23,
43].
65
2.7.2 Handoff
Handoff basically involves change of radio resources from one cell to another adjacent cell. From
a handoff perspective, it is important that a free channel is available in a new cell whenever
handoff occurs so that undisrupted service is available [12, 23, 24]. It takes place either within the
same node B, inter node B within the same RNC, inter RNC within the same MSC or between
different MSCs and there are different reasons for the handover to become necessary [6, 8, 12].
Handoff is as important for UMTS as any other form of cellular telecommunications system and it
is essential that UMTS handoff is performed seamlessly so that the user is not aware of any
change. Any failures within the UMTS handoff procedure will lead to dropped calls which will in
turn result in user dissatisfaction and ultimately it may lead to users changing networks, thereby
increasing the churn rate [8, 24]. A RAKE receiver is a form of radio receiver that has been made
feasible in many areas by the use of digital signal processing, which supports handoff in UMTS
network. It is often used to overcome the effects of multipath propagation and achieves this by
using several sub-receivers known as "fingers" which are given a particular multipath component.
Each finger then processes its component and decodes it [6, 12]. The resultant outputs from the
fingers are then combined to provide the maximum contribution from each path. In this way rake
receivers and multipath propagation can be used to improve the signal to noise performance.
Hard Handoff: The name hard handoff indicates that there is a "hard" change during the handoff
process usually a “break before make”. For hard handoff the radio links are broken and then re-
established. Although hard handoff should appear seamless to the user, there is always the
possibility that a short break in the connection may be noticed by the user [6, 11, 17, 24].
The basic methodology behind a hard handoff is relatively straightforward. There are a number of
basic stages of a hard handoff:
66
• The network decides a handoff is required dependent upon the signal strengths of the
existing link, and the strengths of broadcast channels of adjacent cells.
• The link between the existing node B and the UE is broken.
• A new link is established between the new node B and the UE.
Although this is a simplification of the process, it is basically what happens. The major problem is
that any difficulties in re-establishing the link will cause the handoff to fail and the call or
connection to be dropped [6, 11, 24].
UMTS hard handoffs may be used in a number of instances:
• When moving from one cell to an adjacent cell that may be on a different frequency.
• When implementing a mode change, e.g. from FDD to TDD mode.
• When moving from one cell to another where there is no capacity on the existing channel
and a change to a new frequency is required [11, 24].
One of the issues facing UMTS hard handoffs as also experienced in GSM is that, when usage
levels are high the capacity of a particular cell that a UE is trying to enter may be insufficient to
support a new user. To overcome this, it may be necessary to reserve some capacity for new users.
This may be achieved by spreading the loading wherever possible - for example UEs that can
receive a sufficiently strong signal from a neighboring cell may be transferred out as the original
cell nears its capacity level [6, 11, 24].
67
Figure 2.12: Hard handoff procedure [24]
Soft Handoff: Soft handoff is a form of handoff that was enabled by the introduction of CDMA
which occurs when a UE is in the overlapping coverage area of two cells. Links to the two base
stations can be established simultaneously and in this way the UE can communicate with two base
stations, by having more than one link active during the handoff process. This provides a more
reliable and seamless way in which to perform handoff [8, 12, 24].
In view of the fact that soft handover uses several simultaneous links, it means that the adjacent
cells must be operating on the same frequency or channel. Since the UEs do not have multiple
transmitters and receivers that would be necessary, if they were operating on different frequencies.
When the UE and node B undertake a soft handoff, the UE receives signals from the two node B’s
and combines them using the RAKE receiver capability available in the signal processing of the
UE [12, 24].
In the uplink the situation is more complicated as the signal combining cannot be accomplished in
the node B as more than one node B is involved. Instead, combining is accomplished on a frame
by frame basis; the best frames are selected after each interleaving period. The selection is
accomplished by using the outer loop power control algorithm which measures the signal to noise
68
ratio (SNR) of the received uplink signals. This information is then used to select the best quality
frame. Once the soft handoff has been completed, the links to the old node B are dropped and the
UE continues to communicate with the new node B [8, 12].
As can be imagined, soft handoff uses a higher degree of the network resources than a normal link,
or even a hard handover. However this is compensated by capacity maximization, improved
reliability and performance of the handoff process [8, 11, 23].
Figure 2.13: Soft handoff procedure [24]
Softer Handoff: A form of handoff referred to as softer handoff is really a special form of soft
handoff. It occurs when the new radio links that are added are from the same node B. These may
occur when several sectors are served from the same node B, thereby simplifying the combining as
it can be achieved within the node B and not require linking further back into the network [8, 12].
UMTS softer handoff is only possible when a UE can hear the signals from two sectors served by
the same node B. This occurs as a result of the sectors overlapping, or more commonly as a result
of multipath propagation resulting from reflections from buildings, etc [8, 12].
In the uplink, the signals received by the node B, and the signals from the two sectors can be
routed to the same RAKE receiver and then combined to provide an enhanced signal.
69
In the downlink, it is a little more complicated because the different sectors of the node B use
different scrambling codes. To overcome this, different fingers of the RAKE receiver apply the
appropriate de-spreading or de-scrambling codes to the received signals. Once this has been done,
they can be combined as before [12].
In view of the fact that a single transmitter is used within the UE, only one power control loop is
active. This may not be optimal for all instances but it simplifies the hardware and general
operation [8, 12, 23].
2.7.3 Channelization Codes
The channelization codes are sequences of chips that are applied to the data to be transmitted to
produce a stream of chips that have the data superimposed upon them. The channelization codes
are of relatively short length that can vary depending upon the desired transmitted data rate, and
are made from something referred to as an orthogonal function or waveform [8, 12, 23]. It is the
properties of orthogonality that are particularly important for the channelization code. It comprises
a sequence of 1s and 0s and the duration of these 1s and 0s is known as the chip period, and the
number of chips per second is the chip rate which is 3.84 Mcps. In the uplink, the channelization
code is used to control the data rate that a user transmits to the cell. In the downlink, it is used to
separate users within a cell and also to control the data rate for that user [8, 23].
2.7.4 Scrambling codes
In scrambling operation, a scrambling code is applied to the signal, which makes signals from
different sources separable from each other. Scrambling is used on top of spreading and it
separates terminals or base stations from each other. The symbol rate is not affected by the
scrambling operation [23, 41]. In contrast to channelization codes, scrambling codes are quite long
and are created from streams that are generally referred to as pseudo-noise sequences. In the
70
uplink the scrambling code is used for two main reasons; the first is to separate the users on the
uplink with each user being assigned a scrambling code that is unique within the UTRAN. The
second reason is to provide a mechanism to control the effects of interference both from within the
cell, intra-cell interference as well as from other adjacent cells inter-cell interference, while on the
downlink is used to control the effects of interference [8, 23].
The use of channelization codes and scrambling codes is different on the uplink and the downlink.
This means that the radio interface is asymmetrical, with different functions provided by the codes
in the different directions of the links [8, 23, 41].
Fig 2.14: Spreading and scrambling [41]
2.7.5 Code allocation
For the uplink, the channelization code is selected by the UE based on the amount of data required
for transmission. The specifications define a relationship between required data rate and code
selection. The scrambling code on the uplink is selected and assigned by the UTRAN when the
physical channel is established or possibly when it is re-configured. For the downlink, the
channelization code is allocated by the UTRAN from the ones that are available in that cell at that
point when an allocation is required. The scrambling code on the downlink is used within that cell
and by more than one UE in the cell. The scrambling code is likely to be assigned as part of the
71
radio planning function, but a cell receives the allocation from the operations and maintenance
entity [8, 12, 23].
2.8 Radio Resource Management
Radio resource management (RRM) is the system level control of co-channel interference and
radio transmission characteristics in cellular network. It is a set of algorithms that control the usage
of radio resources. Its management techniques are used to improve the utilization of radio
resources in order to provide maximum system capacity of the cellular network [9, 11, 24]. A
Radio Resource Unit (RRU) is defined as a set of basic physical transmission parameters necessary
to support a signal waveform transporting end user information corresponding to a reference
service [9, 11]. Particularly:
In Frequency Division Multiple Access (FDMA), a radio resource unit is equivalent to a certain
bandwidth within a given carrier frequency, for example, in Total Access Communication System
(TACS), a radio resource unit is a 25 KHz portion in the 900 MHz band.
In Time Division Multiple Access (TDMA), a radio resource unit is equivalent to a pair consisting
of a carrier frequency and a time slot. For example, in GSM a radio resource unit is a 0.577 ms
time slot period every 4.615 ms on a 200 KHz carrier in the 900 MHz, 1800 MHz or 1900 MHz
bands [11, 24].
In Wideband Code Division Multiple Access (WCDMA), a radio resource unit is defined by a
carrier frequency, a code sequence and a power level. The main difference arising here with
respect to other techniques is that, the required power level necessary to support a user connection
is not fixed, but depends on the interference level which makes the capacity of the WCDMA
network interference limited [9, 11].
72
Radio Resource and QoS management functionalities are very important in the framework of
WCDMA based systems because the system relies on them to guarantee a certain target QoS,
maintain the planned coverage area and offer a high capacity. Objectives which tend to be
contradictory for instance, capacity may be increased at the expense of a coverage reduction;
capacity may be increased at the expense of a QoS reduction, etc. Radio network planning
provides a thick tuning of these elements, while RRM will provide the fine tuning mechanisms that
allow a final matching [9, 11]. In WCDMA, users transmit at the same time and frequency by
means of different spreading sequences, which in most of the cases are not perfectly orthogonal.
Consequently, there is a natural coupling among the different users that makes the performance of
a given connection much more dependent on the behavior of the rest of the users sharing the radio
interface compared with other multiple access techniques like FDMA or TDMA [9, 11, 12]. In this
context, RRM functions are crucial in WCDMA because there is not a constant value for the
maximum available capacity, since it is tightly coupled to the amount of interference in the air
interface. Although an efficient management of radio resources may not involve an important
benefit for relatively low loads, when the number of users in the system increases to a critical
number, good radio resources management will be absolutely necessary [9, 11, 12]. RRM
functions can be implemented in many different algorithms, and these impacts on the overall
system efficiency and on the operator infrastructure cost, so RRM strategies play an important role
in WCDMA UMTS scenario.
In general terms, real time services have more stringent QoS requirements compared to non real
time applications and, consequently, the former will require more investment by the network
operator than the latter. Nevertheless, if the amount of available radio resources is too low, non
real time users may experience a non-satisfactory connection, usually in terms of an excessive
73
delay. Then, it will be necessary for the network operator to set some target QoS values for non
real time applications as well [11, 23].
Objectives of RRM
• Maximize performance of all users with coverage capacity
• Guarantee the quality of service for different applications
• Maintain planned coverage
• Maximize system capacity
RRM Algorithms
The basic RRM algorithms can be classified as follows:
• Handoff and mobility management algorithm,
• Call admission control (CAC) algorithm, and
• Power control algorithm.
If a new call is admitted to access the network, then the CAC algorithm will make a decision to
accept or reject it according to the amount of available resources versus users QoS requirements,
and the effect on QoS of exiting calls that may occur as a result of the new call [9, 23]. If the call is
accepted the following has to be decided: transmission (bit) rate, node B, and channel assignment
and transmission power. Most of these resources have to be dynamically controlled during the
transmission. For example, the node B assignment has to be changed as the UE moves further
away from the node B. The handoff algorithm takes care of the re-assignment of node Bs. When
moving closer to the node B, the same received signal strength (RSS) can be upheld for a lower
transmitted power [9, 11, 12]. Thus, efficient power control algorithm is needed to reduce the
transmission power and to keep the interference levels at a minimum in the system, in order to
74
provide the required QoS and to increase the system capacity. Since available bandwidth (radio
resource spectrum) in cellular communication is limited, it is important to utilize it efficiently. For
this reason, frequencies are reused in different cells in the system. This can be done as long as
different users are sufficiently spaced apart, to ensure that interference caused by transmission by
other users will be negligible [9, 11, 12, 23].
2.8.1 Resource Allocation
As the number of users is increasingly growing, the wireless network should serve as many users
as possible, given the limited resources, an example being the bandwidth. On the other hand, as
various service types, such as voice, video, and data, are being offered, quality of service (QoS) is
highly demanded [9, 17, 21]. Obviously, these two requirements are competing against each other,
and to strike a good balance between these two competing requirements, resource allocation plays
a key role. Essentially, resource allocation is responsible for efficient utilization of network
resources while providing QoS guarantees to various applications. However, this goal is not easy
to achieve, since resource allocation is confronted with more difficulty such as error-prone
wireless channel, limited bandwidth, and mobility in mobile wireless networks than in wired
networks [9, 17, 21, 24].
Traffic channel allocation in a cellular system is important from the performance point of view,
which usually covers how a node B should assign traffic channels to the UE’s. As the channels are
managed only by the node B of a cell, a user attempting to make a new call needs to submit a
request for a channel, and the node B can grant such an access to the UE provided that a channel is
readily available for use by the node B [17, 21, 24]. If this is possible, most of the time the
probability that a new call will be blocked or the blocking probability for a call originated in a cell
75
can be minimized. One way to ascertain such a radio resource to be free is to increase the number
of channels per cell, if this is done, then every cell would expect to have a larger number of
channels. However, because a limited frequency band is allocated for wireless cellular networks,
there is a limit to the maximum number of channels. Therefore, there is restriction to the number of
available traffic channels that can be assigned to each cell, especially because of the interference
limited nature of WCDMA based systems [17, 21, 24]. Channel allocation implies that a given
radio spectrum is to be divided into a set of disjoint channels, which can be used simultaneously by
different UEs, while interference in adjacent traffic channels could be minimized by having good
separation between traffic channels.
2.8.1.1 Methods of Resource Allocation
There are basically three methods of resource allocation the includes;
• Fixed Channel Allocation (FCA)
• Dynamic Channel Allocation (DCA) and
• Hybrid Channel Allocation (HCA)
Fixed Channel Allocation
In FCA schemes, a set of traffic channels is permanently allocated to each cell of the system. If the
total number of available channels in the system is divided into sets, the minimum number of
channel sets required to serve the entire coverage area is related to the frequency reuse distance.
One approach to address increased traffic of originating and handoff calls in a cell is to temporarily
borrow free traffic channels from neighboring cells [21, 24]. There are many possible channel-
borrowing schemes, from simple to complex, and they can be selected based on employed
controller software and the feasibility of borrowing under given conditions.
76
Dynamic Channel Allocation
DCA implies that traffic channels are allocated dynamically as new calls arrive in the system. It is
achieved by keeping all free channels in a central pool, which means that when a call is completed,
the channel currently being used is returned to the central pool [21, 24]. In this way, it is fairly
straightforward to select the most appropriate channel for any new call with the aim of minimizing
the interference. Since the allocation of different traffic channels for a current traffic is known,
then, a DCA scheme overcomes the problem of an FCA scheme. A free channel can be allocated to
any cell, as long as interference constraints in that cell can be satisfied [21, 24]. The selection of a
channel could be very simple or could involve one or more considerations, including future
blocking probability in the vicinity of the cell, reuse distance, usage frequency of the candidate
channel, average blocking probability of the overall system, and instantaneous channel occupancy
distribution [21, 24].
Hybrid Channel Allocation
HCA schemes are a combination of FCA and DCA schemes, with the traffic channels divided into
fixed and dynamic sets. This means that each cell is given a fixed number of channels that is
exclusively used by the cell. A request for a channel from the dynamic set is initiated only when a
cell has exhausted using all channels in the fixed set. A channel from the dynamic set can be
selected by employing any of the DCA schemes [9, 21, 24].
2.8.2 Radio Resources
Radio resources in wireless cellular networks such as radio frequency spectrum (bandwidth),
transmit powers, transmission bit rate and base stations are generally limited due to the physical
and regulatory restrictions and also the interference-limited nature of wireless cellular networks [9,
77
12, 24]. Thus, to provide communication services with high capacity and good quality of QoS, it is
imperative to employ efficient and effective methods for sharing the radio spectrum. Spectrum
sharing methods are called multiple access techniques. Multiple access technique involves radio
channel allocation to users of the system [9, 12, 24]. The objective of multiple access techniques is
to provide communication services with sufficient bandwidth when the radio spectrum is shared
with many simultaneous users. A channel can be thought of as a portion of the radio spectrum that
is temporarily allocated for a specific purpose, such as user’s phone call [8, 9, 24].
2.8.2.1Types of Radio Channels
There are different types of radio channel which enables a flexible architecture that allows the
provision of services by making use of different configuration of the radio interface. Thus it
becomes possible to accommodate different degrees of quality of service [8, 11, 12, 23]. They are
the logical, transport and physical channels. By using these channels it is possible to carry the data
for the control and payload in a structured manner and provide efficient effective communications.
Therefore the 3G UMTS channels are thus an essential element of the overall system.
Logical channels: These channels allow communication between the RLC and MAC layers, and
they are characterized by the type of information that is being transferred across these layers. As a
result, there are logical channels for the transfer of user traffic, and also logical channels for the
transfer of control information, which can be either dedicated to specific users or common to a set
or to all of them [8, 11, 12, 17].
Transport channels: These are defined between MAC and PHY layers and they specify how the
information from logical channels should be adapted to get access to the radio transmission
medium. Therefore, they define the format used for the transmission in terms of, channel coding,
interleaving or bit rate [8, 11, 12]. Different transport channels are defined, mainly distinguishing
78
between transport channels operating in dedicated mode i.e. allocated to a specific user and in
common mode i.e. users should contend for the access to such channels whenever they have some
information to be transmitted [11, 17].
Physical channels: They are defined in the physical layer and specify the nature of the signals that
are transmitted either in the uplink or in the downlink direction. These include code, time and
frequency multiplexed with the signals coming from other users and nodes B. Physical channels
include also physical signals, which serve as a support for the transmission on the physical
channels e.g. supporting the random access procedures but do not contain information from upper
layers [11, 12, 17].
UMTS Logical Channels
The 3G logical channels include:
• Broadcast Control Channel (BCCH) (downlink): This channel broadcasts information
to UEs relevant to the cell, such as radio channels of neighboring cells, etc.
• Paging Control Channel (PCCH) (downlink): This channel is associated with the PICH
and is used for paging messages and notification information [11, 12].
• Dedicated Control Channel (DCCH) (up and downlinks): This channel is used to carry
dedicated control information in both directions.
• Common Control Channel (CCCH) (up and downlinks): This bi-directional channel is
used to transfer control information.
• Shared Channel Control Channel (SHCCH) (bi-directional): This channel is bi-
directional and only found in the TDD form of WCDMA / UMTS, where it is used to
transport shared channel control information.
79
• Dedicated Traffic Channel (DTCH) (up and downlinks): This is a bidirectional channel
used to carry user data or traffic.
• Common Traffic Channel (CTCH) (downlink): A unidirectional channel used to
transfer dedicated user information to a group of UEs [11, 12].
UMTS Transport Channels
The 3G transport channels include:
• Dedicated Transport Channel (DCH) (up and downlink). This is used to transfer data to
a particular UE. Each UE has its own DCH in each direction.
• Broadcast Channel (BCH) (downlink). This channel broadcasts information to the UEs
in the cell to enable them to identify the network and the cell.
• Forward Access Channel (FACH) (down link). This is channel carries data or
information to the UEs that are registered on the system. There may be more than one
FACH per cell as they may carry packet data [8, 11, 12].
• Paging Channel (PCH) (downlink). This channel carries messages that alert the UE to
incoming calls, SMS messages, data sessions or required maintenance such as re-
registration.
• Random Access Channel (RACH) (uplink). This channel carries requests for service
from UEs trying to access the system
• Uplink Common Packet Channel (CPCH) (uplink). This channel provides additional
capability beyond that of the RACH and for fast power control.
• Downlink Shared Channel (DSCH) (downlink).This channel can be shared by several
users and is used for data that is "bursty" in nature such as that obtained from web
browsing etc [8, 11, 12].
80
UMTS Physical Channels
The 3G UMTS physical channels include:
• Primary Common Control Physical Channel (PCCPCH) (downlink). This channel
continuously broadcasts system identification and access control information.
• Secondary Common Control Physical Channel (SCCPCH) (downlink) This channel
carries the Forward Access Channel (FACH) providing control information, and the Paging
Channel (PACH) with messages for UEs that are registered on the network.
• Physical Random Access Channel (PRACH) (uplink). This channel enables the UE to
transmit random access bursts in an attempt to access a network.
• Dedicated Physical Data Channel (DPDCH) (up and downlink). This channel is used to
transfer user data [8, 11, 12].
• Dedicated Physical Control Channel (DPCCH) (up and downlink): This channel carries
control information to and from the UE. In both directions the channel carries pilot bits and
the Transport Format Combination Identifier (TFCI). The downlink channel also includes
the Transmit Power Control and FeedBack Information (FBI) bits.
• Physical Downlink Shared Channel (PDSCH) (downlink): This channel shares control
information to UEs within the coverage area of the node B.
• Physical Common Packet Channel (PCPCH): This channel is specifically intended to
carry packet data. In operation the UE monitors the system to check if it is busy, and if not
it then transmits a brief access burst. This is retransmitted if no acknowledgement is gained
with a slight increase in power each time. Once the node B acknowledges the request, the
data is transmitted on the channel [8, 11, 12].
81
• Synchronization Channel (SCH): This channel is used in allowing UEs to synchronize
with the network.
• Common Pilot Channel (CPICH): This channel is transmitted by every node B so that the
UEs are able estimate the timing for signal demodulation. Additionally they can be used as
a beacon for the UE to determine the best cell with which to communicate.
• Acquisition Indicator Channel (AICH) : The AICH is used to inform a UE about the Data
Channel (DCH) it can use to communicate with the node B. This channel assignment
occurs as a result of a successful random access service request from the UE.
• Paging Indication Channel (PICH): This channel provides the information to the UE to
be able to operate its sleep mode to conserve its battery when listening on the Paging
Channel (PCH). As the UE needs to know when to monitor the PCH, data is provided on
the PICH to assign a UE a paging repetition ratio to enable it to determine how often it
needs to 'wake up' and listen to the PCH [8, 11, 12].
• CPCH Status Indication Channel (CSICH): This channel, which only appears in the
downlink carries the status of the CPCH and may also be used to carry some intermittent or
"bursty" data. It works in a similar fashion to PICH.
• Collision Detection/Channel Assignment Indication Channel (CD/CA-ICH): This
channel, present in the downlink is used to indicate whether the channel assignment is
active or inactive to the UE.
82
2.9 Call Admission Control
The CAC is an algorithm that manages radio resources in order to adapt to traffic variations. CAC
is always performed when a mobile initiate’s communication in a new cell either through a new
call or handoff, furthermore, admission control is performed when a new service is added during
an active call. CAC makes a decision to accept or reject a new call according to the amount of
available resources versus user QoS requirements, and the effect on the QoS of existing calls that
may occur as a result of the new call [9, 11, 45].
A connection is accepted if resources are available and the requested QoS can be met, and if other
existing connections and their agreed upon QoS will not be adversely affected. Moreover, the
admission control algorithm ensures that the interference created after adding a new call does not
exceed a pre-specified threshold. The purpose of an admission control algorithm is to regulate
admission of new users into the system, while controlling the signal quality of the already serviced
users without leading to call dropping [9, 11, 45]. The admission control algorithm will then
balance between high capacity and interference. Another goal of admission control is to optimize
the network revenue. This can, for example, be done by maximizing the instantaneous reward
achievable when a new service request arrives. The reward associated with each QoS level is
assumed to increase with the amount of resources required for the service [9, 11, 45].
In a WCDMA scenario, where there is no hard limit on the system capacity, admission control
must operate dynamically depending on the amount of interference that each radio access bearer
adds to the rest of the existing connections.
From the performance point of view, there are different indicators to evaluate and compare
admission control algorithms. Typically, the admission probability that is the probability that a
new connection is accepted or equivalently the blocking probability that is, the probability that the
83
new connection is rejected is used as a measurement of the accessibility to the system provided by
a certain algorithm [9, 11, 45]. Therefore, admission control algorithms must take into
consideration that the amount of radio resources needed for each connection request will vary;
similarly, the QoS requirements in terms of real time or non real time transmission should also be
considered in an efficient admission control algorithm. Clearly, admission conditions for non real
time traffic can be more relaxed on the assumption that the additional radio resource management
mechanisms complementing admission control will be able to limit non real time transmissions
when the air interface load is excessive [9, 11, 45].
2.9.1 CAC Design Considerations
When designing a call admission control (CAC) scheme, several issues are taken into
consideration. First, handoff connection requests needs to be given higher priority than new
connection requests. As it is well known, a handoff request occurs when a user engaged in a call
connection moves from one cell to another. To keep the QoS contract agreed during the connection
setup stage, the network should provide uninterrupted service to the previously established
connection [9, 11, 45]. However, if the new cell does not have enough resources, the ongoing
connection will be forced to terminate before normal completion. Since mobile users are more
sensitive to the termination of an ongoing connection than the blocking of a new call connection,
handoff call connections are usually given higher priority over new call connections.
Second, since the various services offered by the network have inherently different traffic
characteristics, their QoS requirements may differ in terms of bandwidth, delay, and connection
dropping probabilities. It is the network’s responsibility to assign different priorities to these
services in accordance with their QoS demands and traffic characteristics [9, 11, 45]. Finally, when
84
there are multiple types of services coexisting in the network, it is critical that the network can
provide fairness among those services in addition to satisfying their specific QoS requirements.
Thus, the network needs to fairly allocate network resources among different users such that
differentiated QoS requirements can be satisfied for each type of service independent of the others
[9, 11, 45]. Accordingly, the interest lies in two connection-level QoS metrics, namely, the new
call blocking probability (CBP) which is the system capacity measurement and the handoff call
dropping probability (CDP) which is the system quality measurement. Blocking occurs when a
new user is denied access to the system, while call dropping means that a call of an existing user is
terminated, call dropping is considered to be more costly than blocking. In addition, the system
utilization is also considered.
2.9.2 Multiple Service Types
A variety of applications, such as voice, video, and data, are to be supported with QoS guarantee in
WCDMA networks. Due to their different traffic characteristics, they may have different QoS
requirements in terms of delay, delay jitter, bit error rate (BER) [9, 11, 24, 45]. In UMTS, four
traffic classes are supported:
Conversational: This class of traffic has stringent requirements on delay and delay jitter, although
they are not very sensitive to BER. Typical applications include voice and video telephony.
Streaming: Real-time streaming video belongs to this class. Usually, it is less sensitive to delay or
delay jitter than the conversational class.
Interactive: Applications belonging to this class include web browsing and database retrieval.
Typically, the response time they require should be within a certain range and the BER should be
very low since the payload content should be preserved.
85
Background: It may include data services like email or file transfer. For these services, the
destination can tolerate delays ranging from seconds to minutes. However, data to be transferred
has to be received error-free [9, 11, 24].
2.10 Related works
The existing resource sharing schemes are based on two main categories, complete sharing and
complete partitioning. In complete sharing a user is always offered access to the network provided
that there are sufficient resources at the time of request, and all traffic classes share the resources
indiscriminately. In complete partitioning, the available channels or resources are partitioned such
that for each call class, only a fixed section of the resource is available. Therefore the calls are
accepted whenever there are available resources in their corresponding partition otherwise they are
blocked or queued [9, 21, 24, 41].
Several uplink CACs designed for 3G WCDMA have been proposed in the literature [25-38].
These CACs can be classified based on the admission criterion into the following four categories:
power-based CAC, through-put based CAC, interference-based CAC and signal-to-interference
(SIR) CAC. For power-based CAC algorithms the total received power is monitored, while
throughput-based CACs monitor the system load. Interference-based CAC algorithms monitor the
total received interference, and SIR-based CACs monitor the SIR figure experienced by every
user. A reserved capacity for WCDMA is defined as a fraction of cell capacity in terms of the total
interference referred to as interference margin (IM) or in terms of total load referred to as the load
margin (LM).
A CAC algorithm using multiple power-based thresholds for multiple services was proposed in
[31, 32]. By setting a higher priority for voice traffic, the voice traffic is given a higher priority
86
compared to data traffic. An interference based admission control strategy with multiple
interference margin (IM) was analyzed where only two classes of traffic were considered in [33,
34]. A throughput-based admission control with multiple load margin (LM) was proposed where
four classes of traffic were considered in [35]. Recently, dynamic-threshold schemes have been
discussed in the literature to improve the QoS guarantees for higher priority calls [36, 38]. A
throughput-based algorithm that allows different adaptive LM for newly originating and handoff
calls was proposed in [36], and the LM value is adapted using the arrival rates and the estimation
of the blocking probability. The IM needed for high priority calls is estimated by using the signal-
to-noise interference ratio and the (SNR) and distance information of mobile users in neighbouring
cells as seen in [37]. Radio resource management (RRM) in each node B estimates the amount of
IM by considering traffic load in its current cell as well as traffic conditions in neighbouring cells
[38]. All these schemes introduce a large communication and processing overhead in order to keep
up-to-date information about the neighbouring cells, moreover the queuing techniques were not
used.
The above CAC techniques have several disadvantages; the focus is mainly on prioritization using
different fixed or dynamic threshold values for interference margin or load margin without using
buffering techniques. The major limitation of fixed threshold schemes is that the reserved capacity
for higher priority classes may remain unutilized while lower priority classes are being blocked.
Most of the dynamic schemes rely on changing the threshold value based on periodic estimation in
order to decrease the failure of higher priority handoff calls at the expense of lower priority new
calls, this introduces large communication and processing overhead in order to keep up-to-date
information about the state of neighboring cells and therefore limits the scalability. In addition this
will increase the threshold when the estimates indicate high handoff traffic loads without giving a
87
more balanced performance between new calls and handoff calls. Finally, these schemes do not
provide detailed classification of calls based traffic type that is, real time and non-real time and
request type that is originating and handoff call, and no attempt is made to employ queuing for all
classes of calls.
The performance of complete sharing based CAC and complete partitioning based CAC was
compared with the performance of dynamic prioritized uplink call admission control by calculating
total current usage load occupied by each connected call class and also by calculating the dynamic
priority value for the respective call classes present in the queue [39]. It also presented the
utilization and grade of service for real-time and non-real time calls in the system. Maximum
Shannon capacity is estimated, alongside the outage probability and signal to interference ratio
which are employed in single cell and multi-cell admission control scheme [46]. This does not
specifically address handoff traffic class.
This research work focuses on DP-CAC for handoff and new calls which are divided into four
traffic classes’ handoff real-time, handoff non-real time, new call real-time and new call non-real-
time respectively, giving higher priority to handoff traffic classes. DP-CAC uses FIFO queues for
the different traffic classes in order to minimize losses and also uses channel reservation for
handoff in order to reduce its dropping probability. The simulation model is built using MATLAB,
at high traffic condition the model switches handoff traffic to its reserved channel, and allows new
calls to go through the general server which provides fairness to new call traffic class and reduces
its blocking probability. Besides the evaluation of Qos metrics, system utilization, revenue and
grade of service, this research work also considers the queuing delay and the call
blocking/dropping probability of handoff and new calls in evaluating the performance of Dynamic
Priority Call admission control algorithm.
88
CHAPTER THREE
RESEARCH METHODOLOGY
3.0 ADOPTED NETWORK
The network adopted for this research is a typical WCDMA network, which is designed to meet
the objectives and requirements for 3G system. These objectives include support of general quality
of service (QoS), multimedia services and 2Mbps.
Support of general QoS
QoS in general is defined in terms of three quantities namely: data-rate, delay and error
characteristics. Data-rate is usually defined in terms of average data rate and peak data-rate (both
measured over some defined time period). It is the objective of the 3G network to offer a service to
a user that lies within the requested data rate. If the service were a constant data rate service such
as a 57.6 kb/s modem access, then the peak and the average data-rates would tend to be the same.
If, on the other hand, the service is an Internet access data service, then the peak and average data
rates could be quite different.
Delay: The second element of QoS is some measure of delay. In general, delay comprises two
parts. First, there is the type of delay, and second there is some measure of the magnitude of the
delay. The type of delay defines the time requirements of the service, such as whether it is a real-
time service (such as voice communications) or a non-real-time service (such as e-mail delivery).
The magnitude of the delay defines how much delay can be tolerated by the service. Bi-direction
services such as voice communications in general require low delay, while uni-directional services
such as e-mail delivery can accept higher delays, measured in terms of seconds. The objective for
the delay component of QoS, therefore, is to match the user’s service requirements for the delay to
the delay that can be delivered by the network. In a system operating correctly, the delay of the
data should correspond to the dela
Error characteristics: The final component of a typical QoS de
the end to-end link (by end-to-end it is assumed that the QoS is de
communicating users). The error characteristic de
frame error rate (FER). This is a measure of how many errors can be introduced across the link
before the service degrades below a level de
variable and dependent upon the service. Services such
be quite tolerant to errors. Other services, such as packet data for Web access, are very sensitive to
errors.
Support of multimedia services
The second requirement of the WCDMA network is to provide multimedia se
Multimedia is simply a collection of dat
application. The data streams that comprise this multimedia connection will, in general have
differing QoS characteristics.
Figure
the delay that can be delivered by the network. In a system operating correctly, the delay of the
data should correspond to the delay specified within the QoS.
component of a typical QoS definition is the error characteristics of
end it is assumed that the QoS is defined for the link between two
communicating users). The error characteristic defines items such as the bit error rate (B
frame error rate (FER). This is a measure of how many errors can be introduced across the link
before the service degrades below a level defined to be acceptable. The error characteristic is
variable and dependent upon the service. Services such as speech, for example, have been found to
be quite tolerant to errors. Other services, such as packet data for Web access, are very sensitive to
Support of multimedia services
The second requirement of the WCDMA network is to provide multimedia services to the users.
Multimedia is simply a collection of data streams between the user and some other end user’s
application. The data streams that comprise this multimedia connection will, in general have
Figure 3.1multimedia services [8]
89
the delay that can be delivered by the network. In a system operating correctly, the delay of the
finition is the error characteristics of
fined for the link between two
fines items such as the bit error rate (BER) or the
frame error rate (FER). This is a measure of how many errors can be introduced across the link
fined to be acceptable. The error characteristic is
as speech, for example, have been found to
be quite tolerant to errors. Other services, such as packet data for Web access, are very sensitive to
rvices to the users.
some other end user’s
application. The data streams that comprise this multimedia connection will, in general have
90
It is the objective of the network to allow the user to have such a multimedia connection, which
comprises a number of such data streams with different QoS characteristics, but all of them
multiplexed onto the same physical radio interface connection.
Support of 2Mbps
The final objective of the WCDMA network is the ability to provide data-rate up to 2Mbps. This
high data rate is required for certain types of application such as high quality video transmission
and high speed internet access.
3.1 ADOPTED NETWORK ARCHITECTURE
The WCDMA network architecture adopted for this research work comprises three major sections,
the mobile station, the radio access network and the core network. The mobile station comprises
the mobile equipment and the universal subscriber identity module (USIM). The Mobile
Equipment (ME) is the radio terminal used for radio communication over the Uu interface. The
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. The RAN generally is responsible for functions that relate to access, such as the radio
access, radio mobility and radio resource utilization. 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) owns and controls the radio resources in its domain (the Node Bs connected to
it). RNC is the service access point for all services provided by the core network, for example,
management of connections to the UE. The core network is responsible for the higher layer
functions such as user mobility, call control, session management and other network centric
functions such as billing, security control.
91
Figure 3.2WCDMA network architecture [42]
This research work is considering only one node in the WCDMA network architecture and that is
the node B. It is the first access point that communicates directly with the mobile station and the
RAN. Several activities take place in this node ranging from signalling, scheduling, load and
overload control, QoS provision and dynamic resource allocation. These activities constitute the
focus of this research work which makes the node B the central focus of the DP-CAC algorithm.
The Objectives of Dynamic Priority CAC Algorithm
• Ensure best system utilization and revenue while satisfying the required QoS and fairness.
At low and moderate traffic load, it ensures the best system utilization while QoS is
satisfied. At high load, it ensures the fairness of resource usage amongst different class
• Provide a scalable and easy to implement RRM procedure.
• Eliminate the requirement for traffic estimation and communication with neighbouring
cells.
• Support preferential treatment to higher priority calls by serving its queue first.
92
3.2 Physical Model
The physical model adopted for a WCDMA cellular network supporting heterogeneous traffic
assumes two types of services; real-time service (RT) such as conversational and streaming traffic
class and non-real-time service (NRT) such as interactive and background traffic. The priority
classes of incoming call requests are divided into four types, these types are: RT service handoff
request; NRT service handoff requests; newly originating RT calls; and newly originating NRT
calls.
Table 3.1 Service priority classes [12]
Class Traffic type Symbol Call class description 1 RT λh1 Conversational and Streaming – handoff
call 2 NRT λh2 Interactive and Background – handoff
call 3 RT λn1 Conversational and Streaming – new call
4 NRT λn2 Interactive and Background – new call
The capacity of WCDMA cell is defined in terms of the cell load where the load factor, ɳ, is the
instantaneous resource utilization while ɳmax is the maximum cell capacity. By using the concepts
of threshold and queuing techniques, each call class has its own FIFO queues with finite capacities.
A call class request is placed in its corresponding queue if it cannot be serviced upon its arrival and
assigned a resource when available based on its calculated priority.
93
Figure 3.3: DP-CAC physical model
The DP-CAC model employs queuing, prioritization, and channel reservation. Queuing is
necessary when the power level received by the node B in the current cell reaches a certain
threshold, namely the handoff threshold; a call is placed in the queue from the neighbour cell for
providing service which remains in the queue until either an available channel in the new cell is
found or the power by the node B in the current cell drops below a second threshold, called the
receiver threshold. Since handoff requests arrival process is Poisson, that is, calls arrive in an
exponential distribution, queuing is effective and also when traffic is high especially in a densely
populated area. Queuing is very beneficial in macro-cells with cell radius exceeding 35 km since
the UE can wait for handoff before signal quality drops to an unacceptable level.
Prioritization involves channel assignment strategies that allocate channels to handoff requests
more readily than new calls [45]. While in channel reservation a number of channels are reserved
FIFO Queue
ɳ4
ɳ2
ɳ3
ɳ1
Server ɳmax
λn2
λn1
λh2
λh1
DP-CAC
94
exclusively for handoff calls in a cell, the remaining channels are used among the new and handoff
calls as can be seen in the node B system model below (figure 3.3). This method not only
minimizes dropping of handoff calls, but also increases total carried traffic as well as it provides
optimal resource utilization.
Figure 3.4 node B system model
3.3 DP-CAC Algorithm
Resource utilization and individual QoS requirement can be improved by using the Dynamic
Priority Call Admission Control (DP-CAC) algorithm. The resource allocation to a traffic class can
be dynamically adjusted according to the traffic load variations and QoS requirements, while most
of the free capacity from the under-loaded traffic classes can be utilized by the over-loaded traffic
classes. Dynamic priority is implemented to protect from resource starvation.
The DP-CAC algorithm attempts to manage resource allocations amongst the different call classes,
and to efficiently utilize the resources while satisfying the QoS requirements. Similar to [31, 34,
36], only the uplink direction is considered in this research work where it is assumed that
whenever the uplink channel is assigned the downlink is established. To implement the admission
control for the WCDMA cellular networks, first an estimate of the total cell load must be
95
computed and then employed in the decision process of accepting or rejecting new connections,
the analysis also assumes perfect power control operation where the UE and its home node B use
only the minimum needed power in order to achieve the required performance. This algorithm is
used as a tool to maintain service continuity with QoS guarantees and also to provide service
differentiation to mobile users.
Table 3.2 Computation parameters [39] Parameter Equation Description 1 Load factor increment ∆ɳi =
������� /�� Gi = processing gain
ei = bit energy-to-noise density
2 Processing gain Gi = W/Ri W = chip rate, 3.84Mcps R = bit-rate of service class
3 Bit energy-to-noise density ei = Eb/No Eb = bit energy No = output noise
4 Current total load factor nc = (1+f) �
������ /ei���
���
α – activity factor ei – traffic intensity Bi – number of already connected class i call f – interference from adjacent cell
Table 3.3 Traffic Model
Arrival processes Poisson Arrival rates λh1, λh2, λn1, and λn2
Total arrival λ= λh1+ λh2+λn1+λn2
Channel holding time 1/μ
Traffic Intensity ρ = ��
Eb/No 5 and 2 for RT and NRT service respectively
R (Bit-rate) 12.2kbps and 256kbps for RT and NRT service
respectively.
96
Table 3.4 Performance measures [39]
KPI Equation Description 1 System Utilization U = � �ni ∗ �bi�
��� ni – average number of connections in each traffic class that the system can accept bi – bandwidth of connection = ∆ɳi
2 System Revenue r = � �ni ∗ �ri = bi� ���
3 Grade of service GoS = 1-PB PB– call blocking probability 4 Blocking Probability PB = ��� !
"! $ e%�& n = arrival at a time λ = no of arrival a unit time interval t = time in seconds
98
CHAPTER FOUR
SIMULATION RESULT
4.0 Simulation Model
Real Time Traffic Source (Voice): This module was realized by modelling Real Time traffic
class as Markov-Modulated Poisson traffic to cater for the active and silent behaviour of voice
using MATLAB. This module as shown in fig. 4.1 comprises Time Based Entity Generator which
generates calls at a transmission rate of 2.048Mbps using an intergeneration time of a statistical
distribution (Poisson). The generated call passes through an Enable Gate which is regulated by a
function call subsystem that is being controlled by an Entity Departure Event To Function Call
Block and whose basic function is to allow the carrier signal generated by a Time Based Entity
Generator that serves as an envelope for the cell generated thus providing a silent and active form
of traffic pattern for the voice traffic.
99
Figure 4.1: Real Time Traffic Source (Voice)
Non-Real Time Traffic Source (Data): This module was realized by modelling Non-Real time
traffic class also as Markov-Modulated Poisson traffic to cater for the ON-OFF behaviour of data
traffic using MATLAB. This module as shown in fig. 4.2 comprises Time Based Entity Generator
which generates traffic (containing the data to be transmitted) at a transmission rate of 10Mbps
using an intergeneration time of a statistical distribution (Poisson with an exponential mean
varying in this case). The generated traffic passes through an Enable Gate which is regulated by a
function call subsystem that is being controlled by an Entity Departure Event To Function Call
Block and whose basic function is to allow the carrier signal generated by a Time Based Entity
Generator that serves as an envelope for the traffic generated thus providing a ON-OFF form of
traffic Pattern for the data traffic.
101
The flowchart model is developed into a simulation model using MATLAB/simulink. The
simulation model comprises four bursty sources of handoff calls and new calls λh1, λh2, λn1, and λn2
which represent the four different traffic classes’ classified as real time and non-real time traffic
respectively. The different traffic sources are combined together through a path combiner via a
switch which separates handoff calls from new calls and carries them through separate channels,
that is, handoff calls through another switch. This is combined with new calls in a path combiner to
the node B where they access radio resources. Once the maximum traffic load threshold is attained
for the general server in the node B the switch that outputs handoff calls to the path combiner is
alerted in order to block that output and switch to the channel reserved for handoff calls only. This
reduces the calls in the general server and allows new calls to be served even during high traffic
intensity which is where the fairness comes in.
Step-by-Step Conversion of Flowchart Modules
Start Initialize ɳMax Wait for call arrivals Call arrivals?
The start in the flow chart implies that in the simulation model the system is run, and the maximum
server capacity ɳMax of the node B is being initialized awaiting calls that will be arriving from the
different traffic sources. These calls arriving are combined in a path combiner with four inputs and
a single output as shown in fig 4.4.
Figure 4.4 Calls arriving from respective sources
102
If there is no call arrival the system returns to the initialization stage.
Handoff and New call Compute ɳc + ∆ɳi ɳc + ∆ɳi ≤ ɳmax
The single output of combined calls (handoff and new calls) from the path combiner is passed through an
output switch of single input and double output, which is used to separate the handoff calls from the new
calls. The separated calls are passed through their respective get attribute block, which outputs the value of
number of call arrivals (λh1&h2, and λn1&n2) that have departed from this block since the start of the
simulation. Since handoff calls have higher priority, it passes through a DP-CAC switch while the new calls
proceed directly to the path combiner with double input and single output as shown in fig 4.5.
Figure 4.5: Flow of traffic to DP-CAC switch
This model employs resource sharing by handoff and new calls in the node B, prioritization, and
channel reservation for handoff calls. Hence the DP-CAC switch has two outputs. One output
routes handoff calls to resources in the general server used by both traffic classes, while DP-CAC
switch monitors the p signal’s value throughout the simulation and reacts to changes by selecting
the corresponding entity output port to switch handoff calls to the reserved channel. Before these
calls can access resources in the node B, the system computes the load factor increment ∆ɳi for a
103
new call request i and the current system load, ɳc of calls in the general server. This is computed
using a computational module in the system, which uses the number of calls, arrived in the general
server at the node B and predefined simulation parameters in tables 3.2 and 3.3 respectively.
Figure 4.6 computational module
At the start of the simulation, the ɳc + ∆ɳi ≤ ɳmax is less than the maximum channel capacity of the
general server therefore calls arriving from handoff and new call request are allowed access to the general
server.
Figure 4.7 decision making model
Figure 4.7 above, is the decision making module of the DP-CAC algorithm. From the flowchart
model it is shown in fig 4.7 that handoff and new calls arrive in the system through the path
104
combine. The current total load factor ɳc in the general server and the load factor increment is
computed. The total value of the two factors is compared with the maximum load margin ɳmax
which is the maximum capacity of the general server. If the value is less than ɳmax, then all calls are
admitted in the general server, but if the value is greater than or equal to ɳmax , then the handoff calls
are separated from the new calls and switched to the channels reserved for handoff calls only. This
process is triggered by the auto-system functional block to the DP-CAC switch to ensure service
continuity to handoff calls which will in turn reduce the call dropping probability. The new calls
on the other hand are placed in their respective queue with finite capacity, while waiting for a free
channel in the general server. Queuing the new calls also help to minimize the call blocking
probability, as soon as there are free channels in the general server the new calls are served first,
thereby providing fairness to all traffic classes.
Figure 4.8 complete simulation model
105
4.1 Simulation Results and Discussion
The DP-CAC algorithm simulation is run using the complete simulation model and the traffic
model was employed to observe the rate of call arrival from the different traffic classes. The
channel holding time of the system and the departure rate of calls from the system after being
served were also recorded. The parameters obtained during the simulation of the model were
computed using a computational tool and the equations in the tables in chapter three were also
used where applicable.
System Utilization
The system utilization is computed using its equation and the results are plotted against the total
offered traffic. From figure 4.9 graph of system utilization (revenue), the results show that for DP-
CAC, at low and moderate offered traffic the system is utilized by all traffic classes even though it
is not the optimum capacity that is being utilized, which is expected for all systems at low and
moderate traffic conditions. At high offered traffic situation, DP-CAC dynamically controls the
priority level of queued calls and thus preventing new call traffic class from being adversely
affected by handoff call traffic class. The DP-CAC switch in the simulation model automatically
switches handoff calls to its reserved channel and admits the new calls into the general server
whenever there is an available channel. The system utilization increases as the offered traffic
increases. From figure 4.9 the result show clearly that at an offered traffic of 320 users per time the
capacity of the system is only 30% utilized, and as the offered traffic increases to 800 users, the
capacity of the system also increases to 80% utilization. As the offered traffic approaches its peak
known as congestion period at an offered traffic of 1520 users, the capacity utilization of the
system also attains it optimum at 95% and remains constant with further admission of users into
the system.
Figure 4.9: Graph of system capacity utilization (r
Revenue
The revenue is also computed using appropriate equation, and it is discovered that it has the same
curve as system utilization. It therefore follows that the higher and more efficient the system
capacity utilization, the higher the revenue generated
the traffic offered the higher the revenue generated.
0
10
20
30
40
50
60
70
80
90
100
0 200 400
Uti
liza
tio
n
Offered Traffic (calls/second)
of system capacity utilization (revenue) against offered traffic
The revenue is also computed using appropriate equation, and it is discovered that it has the same
curve as system utilization. It therefore follows that the higher and more efficient the system
capacity utilization, the higher the revenue generated by the system, indicating also that the higher
the traffic offered the higher the revenue generated.
600 800 1000 1200 1400 1600 1800
Offered Traffic (calls/second)
106
against offered traffic
The revenue is also computed using appropriate equation, and it is discovered that it has the same
curve as system utilization. It therefore follows that the higher and more efficient the system
, indicating also that the higher
1800
Revenue
107
Figure 4.10 shows the comparison between the results obtained for system capacity utilization
when DP-CAC algorithm is implemented and when it is not implemented in the system. Without
DP-CAC the different traffic classes are assigned separate channels of fixed capacity. Since traffic
conditions for handoff calls and new calls vary at every instance, the unutilized loading limits of
handoff calls cannot be used by the new call class and vice versa, therefore capacity is wasted
which results in inefficient total system utilization and loss of revenue. But when different traffic
classes are allowed to share radio resources, at low traffic condition the total cell capacity is
available for all arrived call classes to enhance the resource utilization while at high traffic
condition the dynamic priority is used to differentiate between handoff and new calls in order to
ensure fairness amongst traffic classes.
The graph of comparison between system capacity utilization with DP-CAC algorithm and without
DP-CAC algorithm figure 4.10 show that DP-CAC increases the utilization of the system. For
instance, at an offered traffic of 800 users, the system capacity utilization without DP-CAC
algorithm is still below 50%, precisely it is at 48%. The reason being that the different traffic
classes are served using separate fixed channel capacity and the different traffic classes have
varying load intensity at every instance, if the new calls have higher load intensity than the handoff
calls at a particular time, the new call blocking probability is increased while there are still
available channels that are unutilized by the handoff call traffic class that cannot be used
overloaded new call traffic class. This results in inefficient and under utilization of the total system
capacity and loss of revenue as shown in figure 4.5 below.
Figure 4.10: Comparison between system
Grade of Service
The system grade of service was computed using its equation and the results plotted against
offered traffic as seen in figure 4.11
graph it shows that at low and moderate traffic, the system offers higher grade of service which
decreases as the offered traffic increases, but
instance, at traffic within 500 users per time the grade of service is still optimum
offered traffic increases the grade of service decreases. At an offered traffic of 1200 users, the
grade service reduces to a minimum of 0.695 and remains constant with increasing traffic. This is
regulated by the DP-CAC admission control
0
10
20
30
40
50
60
70
80
90
100
0 200 400 600
Uti
liza
tio
n
Offered Traffic (calls /second)
Comparison between system capacity utilization (revenue) with DPand without DP-CAC algorithm.
The system grade of service was computed using its equation and the results plotted against
fic as seen in figure 4.11 graph of grade of service against offered traffic. From the
t low and moderate traffic, the system offers higher grade of service which
decreases as the offered traffic increases, but not below an average acceptable threshold.
within 500 users per time the grade of service is still optimum
offered traffic increases the grade of service decreases. At an offered traffic of 1200 users, the
grade service reduces to a minimum of 0.695 and remains constant with increasing traffic. This is
CAC admission control algorithm to maintain the desired QoS guarantee.
600 800 1000 1200 1400 1600 1800
Offered Traffic (calls /second)
108
with DP-CAC algorithm
The system grade of service was computed using its equation and the results plotted against
graph of grade of service against offered traffic. From the
t low and moderate traffic, the system offers higher grade of service which
not below an average acceptable threshold. For
within 500 users per time the grade of service is still optimum at 0.936, as the
offered traffic increases the grade of service decreases. At an offered traffic of 1200 users, the
grade service reduces to a minimum of 0.695 and remains constant with increasing traffic. This is
algorithm to maintain the desired QoS guarantee.
Rev(No DP-CAC)
Rev(DP-CAC)
Figure 4.11 Graph
Figure 4.12 below shows the comparison between the system grade of service with DP
algorithm and without DP-CAC. The results show
below acceptable and agreed upon service requirement (threshold).
that at an offered traffic of 1000 users the grade of service falls to 0.600, and even crashes further
to 0.500 at an offered traffic of 2000 users, this is because DP
to regulate and maintain the desired QoS guarantee.
0.00E+00
1.00E-01
2.00E-01
3.00E-01
4.00E-01
5.00E-01
6.00E-01
7.00E-01
8.00E-01
9.00E-01
1.00E+00
0.00E+00 5.00E+02 1.00E+03
Gra
de
of
Se
rvic
e
Offered Traffic (calls/second)
Graph of grade of service against offered traffic
shows the comparison between the system grade of service with DP
CAC. The results show that without DP-CAC the grade of service falls
below acceptable and agreed upon service requirement (threshold). It can be seen from figure 4.12
that at an offered traffic of 1000 users the grade of service falls to 0.600, and even crashes further
offered traffic of 2000 users, this is because DP-CAC algorithm is not implemented
to regulate and maintain the desired QoS guarantee.
1.00E+03 1.50E+03 2.00E+03 2.50E+03 3.00E+03
Offered Traffic (calls/second)
109
shows the comparison between the system grade of service with DP-CAC
CAC the grade of service falls
It can be seen from figure 4.12
that at an offered traffic of 1000 users the grade of service falls to 0.600, and even crashes further
CAC algorithm is not implemented
GoS(DP-CAC)
Figure 4.12 Comparison between
Queuing Delay
The average delay experienced by the different
value when the system has attained a maximum traffic load condition at increasing traffic intensity
Particularly, the new call non-real time traf
other traffic classes while the handoff traffic classes experience very minimal queuing delay
result from figure 4.13 graph showing queuing
illustrates this. The graph shows that
at low traffic intensity and as the traffic intensity increases it has a delay
NRT experiences a delay of 2.40E
intensity. The new call RT experiences a delay of
increasing traffic intensity. The new call NRT experiences a delay of 1.39E
intensity and 1.42E-11 at increasing traffic intensity.
0.00E+00
1.00E-01
2.00E-01
3.00E-01
4.00E-01
5.00E-01
6.00E-01
7.00E-01
8.00E-01
9.00E-01
1.00E+00
0.00E+00 5.00E+02 1.00E+03
Gra
de
of
Se
rvic
e
Offered Traffic (calls/second)
Comparison between grade of service performance with DP-CACwithout DP-CAC algorithm.
verage delay experienced by the different traffic classes in the network lies within a constant
value when the system has attained a maximum traffic load condition at increasing traffic intensity
real time traffic class experiences greater queuing delay compared to
while the handoff traffic classes experience very minimal queuing delay
graph showing queuing delay against traffic intensity for each traffic class
The graph shows that the handoff RT traffic class experiences a delay of 1.08E
at low traffic intensity and as the traffic intensity increases it has a delay of 1.15E
NRT experiences a delay of 2.40E-12 at low traffic intensity and 2.50E-12 at increasing traffic
intensity. The new call RT experiences a delay of 7.60E-12 at low traffic intensity and 7.80E
increasing traffic intensity. The new call NRT experiences a delay of 1.39E
11 at increasing traffic intensity.
1.00E+03 1.50E+03 2.00E+03 2.50E+03 3.00E+03
Offered Traffic (calls/second)
GoS(DP
GoS(No DP
110
CAC algorithm and
lies within a constant
value when the system has attained a maximum traffic load condition at increasing traffic intensity.
greater queuing delay compared to
while the handoff traffic classes experience very minimal queuing delay. The
for each traffic class
the handoff RT traffic class experiences a delay of 1.08E-12
1.15E-12. The handoff
12 at increasing traffic
12 at low traffic intensity and 7.80E-12 at
increasing traffic intensity. The new call NRT experiences a delay of 1.39E-11 at low traffic
GoS(DP-CAC)
GoS(No DP-CAC)
Figure 4.13 Graph showing q
Call Blocking and Dropping Probability
The call blocking and dropping probabilities were computed using its equation and th
plotted against traffic intensity as seen in figure 4.14
probability against traffic intensity for handoff and new calls respectively.
both the call dropping and blocking probabiliti
intensity, but as the traffic intensity increases there is a rise in both
probability. For instance, at a traffic intensity of 2.25E+03 the handoff
starts to rise with a probability of 9.69E
probability of 1.59E-02 at 3.60E+03. The new call blocking probability follows the same trend
a traffic intensity of 2.25E+03;
with increasing traffic intensity to a probability of 2.00E
the blocking probability of new calls is higher than th
0.00E+00
2.00E-12
4.00E-12
6.00E-12
8.00E-12
1.00E-11
1.20E-11
1.40E-11
1.60E-11
0.00E+00 2.00E+06
De
lay
Traffic Intensity (calls/second)
Graph showing queuing delay against traffic intensity for each traffic class
Blocking and Dropping Probability
The call blocking and dropping probabilities were computed using its equation and th
intensity as seen in figure 4.14 below graph of call blocking
probability against traffic intensity for handoff and new calls respectively. The result shows that
both the call dropping and blocking probabilities are relatively constant at low and moderate traffic
ut as the traffic intensity increases there is a rise in both the dropping and blocking
probability. For instance, at a traffic intensity of 2.25E+03 the handoff call dropping probability
tarts to rise with a probability of 9.69E-04 and steadily rises with increasing traffic intensity to a
02 at 3.60E+03. The new call blocking probability follows the same trend
it starts to rise with a probability of 1.80E-03 and steadily rises
with increasing traffic intensity to a probability of 2.00E-02 at 3.60E+03. It is
the blocking probability of new calls is higher than the dropping probability of handoff calls. This
2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07
Traffic Intensity (calls/second)
111
for each traffic class.
The call blocking and dropping probabilities were computed using its equation and the results
blocking and dropping
The result shows that
es are relatively constant at low and moderate traffic
dropping and blocking
call dropping probability
04 and steadily rises with increasing traffic intensity to a
02 at 3.60E+03. The new call blocking probability follows the same trend, at
03 and steadily rises
also observed that
e dropping probability of handoff calls. This
Newcall NRT
NewCall RT
Handoff NRT
Handoff RT
112
is because handoff calls have a higher priority than the new calls, and the DP-CAC algorithm is
used as a tool to maintain service continuity for handoff calls while still ensuring fairness for lower
priority traffic class in the system.
Figure 4.14 Graph of call blocking and dropping probability against Traffic Intensity for handoff
and new calls respectively.
Figure 4.15 below shows the call blocking and dropping probability for the four different traffic
classes. The results show that the handoff RT and NRT traffic classes have lower call dropping
probability while the new call RT and NRT traffic classes have higher blocking probability all of
which have been sufficiently minimized with the use of DP-CAC algorithm. From the graph it is
observed that during congestion period the different traffic classes have the following probabilities,
at peak traffic intensity of 3.60E+03 handoff RT has a probability of 1.59E-02, handoff NRT a
probability of 1.69E-02, new call RT a probability of 2.00E-02 and new call NRT a probability of
-4.51E-17
5.00E-03
1.00E-02
1.50E-02
2.00E-02
2.50E-02
0.00E+005.00E+021.00E+031.50E+032.00E+032.50E+033.00E+033.50E+034.00E+03
Blo
ckin
g P
rob
ab
ilit
y
Traffic Intensity (calls/second)
New calls
Handoff calls
2.10E-02. This again shows the priority level
highest priority while new call NRT t
Figure 4.15: Graph of call blocking and dropping probability
Call Blocking and Dropping Probabilities with varying Server Capacity
For further performance evaluation of DP
delivery for WCDMA based 3G
in order to investigate its effect on the call dropping and blocking probability of both handoff and
0.00E+00
5.00E-03
1.00E-02
1.50E-02
2.00E-02
2.50E-02
0.00E+005.00E+021.00E+03
Blo
ckin
g P
rob
ab
ilit
y
Traffic Intensity (calls/second)
. This again shows the priority level of each traffic class; handoff RT traffic class has
highest priority while new call NRT traffic class has least priority
blocking and dropping probability against Traffic Intensityrespective traffic classes.
Blocking and Dropping Probabilities with varying Server Capacity
For further performance evaluation of DP-CAC algorithm on QoS requirements and service
delivery for WCDMA based 3G networks. There is need to vary the general capacity of the node B
in order to investigate its effect on the call dropping and blocking probability of both handoff and
1.00E+031.50E+032.00E+032.50E+033.00E+033.50E+034.00E+03
Traffic Intensity (calls/second)
113
each traffic class; handoff RT traffic class has
Traffic Intensity for the
CAC algorithm on QoS requirements and service
is need to vary the general capacity of the node B
in order to investigate its effect on the call dropping and blocking probability of both handoff and
New calls NRT
New calls RT
Handoff NRT
Handoff RT
114
new calls. The three graphs following, that is, figure 4.16, figure 4.17, and figure 4.18 present the
behaviour of the system and the resultant effect on the call dropping and blocking probability.
Figure 4.16 the graph of call blocking and dropping probability at a server capacity of 24 channels
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
24 channels. The result shows that there is very small difference in both probabilities at low and
moderate traffic, the difference being 1.7E-04 at a traffic intensity of 1.0E+03 because all traffic
classes are admitted into system at low and moderate traffic. At a traffic intensity of 2.25E+03 the
difference between the dropping probability of handoff calls and the blocking probability of new
calls increases to 8.61E-04 and at a traffic intensity of 3.15E+03 the difference between both
probabilities further increases to 3.0E-03. From the analysis of this result, it is observed that the
difference between call blocking and dropping probability increases as the traffic intensity
increases, which again buttresses that the system gives higher priority to handoff calls to ensure
service continuity of users.
Figure 4.16: Graph of call blocking
Figure 4.17 graph of call blocking
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
12 channels. The result shows that there is a significant difference in both probabilities at low and
moderate traffic, the difference being 7.40E
intensity of 2.25E+03 the difference between the dropping probability of handoff calls and the
blocking probability of new calls increases to 1.11
difference between both probabilities further increases to 3.0E
that there is a significant difference between the call dropping and blocking probability of handoff
and new calls at the same traffic
of the reduction in the general channel capacity of the system.
-4.51E-17
5.00E-03
1.00E-02
1.50E-02
2.00E-02
2.50E-02
0.00E+005.00E+021.00E+03
Blo
ckin
g P
rob
ab
ilit
y
locking and dropping probability at a server capacity of 24 channels
locking and dropping probability at a server capacity of 12 channels
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
12 channels. The result shows that there is a significant difference in both probabilities at low and
c, the difference being 7.40E-04 at a traffic intensity of 1.0E+03. At a traffic
intensity of 2.25E+03 the difference between the dropping probability of handoff calls and the
ty of new calls increases to 1.11E-03 and at a traffic intensity of 3.15E+03 the
difference between both probabilities further increases to 3.0E-03. From this analysis it is observed
that there is a significant difference between the call dropping and blocking probability of handoff
and new calls at the same traffic intensity compared to the result in figure 4.16. This is as a result
of the reduction in the general channel capacity of the system.
1.00E+031.50E+032.00E+032.50E+033.00E+033.50E+034.00E+03
Traffic Intensity calls/second
115
server capacity of 24 channels.
server capacity of 12 channels
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
12 channels. The result shows that there is a significant difference in both probabilities at low and
04 at a traffic intensity of 1.0E+03. At a traffic
intensity of 2.25E+03 the difference between the dropping probability of handoff calls and the
ty of 3.15E+03 the
From this analysis it is observed
that there is a significant difference between the call dropping and blocking probability of handoff
. This is as a result
4.00E+03
New calls
Handoff calls
Figure 4.17: Graph of call blocking
Figure 4.18 graph of call blocking
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
6 channels. The result shows that there is a dynamic difference in both probabil
moderate traffic, the difference being 1.64E
intensity of 2.25E+03 the difference between the dropping probability of handoff calls and the
blocking probability of new calls increases to 3
difference between both probabilities further increases to 0.4E
observed that there is a dynamic
0.00E+00
5.00E-03
1.00E-02
1.50E-02
2.00E-02
2.50E-02
0.00E+005.00E+021.00E+03
Blo
ckin
g P
rob
ab
ilit
y
Traffic Intensity calls/second
locking and dropping probability at a server capacity of
locking and dropping probability at a server capacity of 6
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
6 channels. The result shows that there is a dynamic difference in both probabil
moderate traffic, the difference being 1.64E-03 at a traffic intensity of 1.0E+03. At a traffic
intensity of 2.25E+03 the difference between the dropping probability of handoff calls and the
blocking probability of new calls increases to 3.0E-03 and at a traffic intensity of 3.15E+03 the
difference between both probabilities further increases to 0.4E-02. From this analysis it is
dynamic difference between the call dropping and blocking probability of
1.00E+031.50E+032.00E+032.50E+033.00E+033.50E+034.00E+03
Traffic Intensity calls/second
116
server capacity of 12 channels.
probability at a server capacity of 6 channels
presents the call blocking and dropping probability of handoff and new calls at a server capacity of
6 channels. The result shows that there is a dynamic difference in both probabilities at low and
03 at a traffic intensity of 1.0E+03. At a traffic
intensity of 2.25E+03 the difference between the dropping probability of handoff calls and the
03 and at a traffic intensity of 3.15E+03 the
02. From this analysis it is also
difference between the call dropping and blocking probability of
New calls
Handoff calls
handoff and new calls at the same traffic intensity compared to the result in
4.16. This is as a result of the further
Figure 4.18: Graph of call blocking It can be seen from the trend in figure 4.16, figure 4.17 and figure 4.18
capacity, both call dropping and blocking probability
capacity. From the above investigation, it
increases significantly when the channel capacity is reduced, while the dropping probability of
handoff calls increases relatively because it has a higher priority level than the
class.
0.00E+00
5.00E-03
1.00E-02
1.50E-02
2.00E-02
2.50E-02
0.00E+005.00E+021.00E+03
Blo
ckin
g P
rob
ab
ilit
y
d new calls at the same traffic intensity compared to the result in figure 4.17
further reduction in the general channel capacity of the system.
locking and dropping probability at a server capacity of
from the trend in figure 4.16, figure 4.17 and figure 4.18 that by
both call dropping and blocking probability increases generally with decreasing channel
ove investigation, it is observed that the blocking probability of new calls
increases significantly when the channel capacity is reduced, while the dropping probability of
handoff calls increases relatively because it has a higher priority level than the
1.00E+031.50E+032.00E+032.50E+033.00E+033.50E+034.00E+03
Traffic Intensity calls/second
117
figure 4.17 and figure
reduction in the general channel capacity of the system.
server capacity of 6 channels.
that by varying the server
increases generally with decreasing channel
observed that the blocking probability of new calls
increases significantly when the channel capacity is reduced, while the dropping probability of
handoff calls increases relatively because it has a higher priority level than the new call traffic
4.00E+03
New calls
Handoff calls
118
CHAPTER FIVE
CONCLUSION AND RECOMMENDATION
5.0 Conclusion
The objective of this research is on how to maintain service continuity with quality of service
guarantees and to provide service differentiation to mobile user’s traffic profile by efficiently
utilizing system resources. In order to achieve this, DP-CAC algorithm is used and a simulation
model is developed whose aim is to ensure optimal system capacity utilization, maintain agreed
upon grade of service and QoS requirement and also to minimize call dropping and blocking
probability of handoff and new calls respectively.
This work followed the designed methodology step by step and has arrived its ending by
presenting the results and investigations from the research. It is based on these results and
investigations that the following conclusion is stated.
This research work concludes by stating that the design of QoS-aware CAC is a critical issue for
WCDMA based cellular networks supporting heterogeneous traffic. As shown by the results, DP-
CAC algorithm provides an acceptable QoS for each traffic class and prevents the traffic class with
higher priority from overwhelming the traffic class with lower priority in order to enhance fairness
at high traffic conditions. DP-CAC also provides service continuity with QoS guarantees to users
by ensuring that handoff calls are serviced even during congestion period by dynamically
switching the calls to its reserved channel. Further results from investigation show that the call
dropping and blocking probability of handoff and new calls are minimized by DP-CAC algorithm
which generally improves the throughput of the network. Results also show that the call dropping
and blocking probability increases with reduced channel capacity, therefore base station
119
configuration and upgrade for WCDMA based 3G networks require standard channel capacity to
be able to provide QoS requirements and optimum service delivery.
5.1 Contribution to Knowledge
DP-CAC algorithm is often used to evaluate the performance of a system based on the capacity
utilization and grade of service for real time and non real time services even though they are often
classified into handoff and new call traffic classes respectively.
This research work being concerned on how to maintain service continuity with quality of service
guarantees and to provide service differentiation to mobile user’s traffic profile by efficiently
utilizing system resources. It has been able to investigate the behavior of the call dropping and
blocking probability of handoff and new calls with DP-CAC algorithm and with varying general
server capacity. It has also been able to investigate the delay experienced by the different traffic
classes in the system as a contribution to knowledge in the aspect of call admission control for
WCDMA networks. These are areas that are usually mentioned in passing as reviewed in most
literature and papers on call admission control, but this work has been able consider them in detail.
5.2 Recommendation
This research work recommends through its results and findings that future planning or upgrade of
WCDMA based 3G networks should be carried out with standard base station configuration using
improved node B cabinets. This will have a minimum capacity of 24 radio transceiver channels for
optimum quality of service guarantees. Implementing DP-CAC algorithm for its optimal capacity
utilization will generate optimum revenue for mobile network operators. Further research work
may be done on overlay network deployments required for 4G data services. Having these base
stations installed and operated by mobile operators will ensure the right equipment form factor for
the right situation to meet the ever-growing need for greater capacity.
120
REFERENCES
1. Harte Lawrence; “Introduction to Mobile Telephone Systems 1G, 2G, 2.5G, and 3G Wireless
Technologies and Services”. United States : ALTHOS Publishing Inc, (2006).
2. Pierre Lescuyer; “Computer Communications and Networks; UMTS: Origins, Architecture and the Standard”. London : Springer-Verlag, (2004).
3. Gunnar Heine; “GSM Networks:Protocols,Terminology, and Implementation”. London : Artech House, INC., (1999).
4. Gottfried Punz; “Evolution of 3G Networks; The Concept, Architecture and Realization of Mobile Networks Beyond UMTS”. Germany : Springer-Verlag/Wien, (2010).
5. Sharma Pankaj; “Evolution of Mobile Wireless Communication Networks-1G to 5G as well as Future Prospective of Next Generation Communication Network”. International Journal of Computer Science and Mobile Computing (IJCSMC), August 2013 Issue. 8 Vol. 2,, pp. pg.47 – 53.
6. Eberspächer Jörg, et al. “GSM – Architecture, Protocols and Services” Third Edition. United Kingdom : John Wiley & Sons Ltd, (2009).
7. Halonen T., Romero J. & Melero J.; “GSM, GPRS and EDGE Performance; Evolution Towards 3G/UMTS”. England : John Wiley & Sons Ltd, (2003).
8. Andrew Richardson; “WCDMA Design Handbook”. United Kingdom : Cambridge University Press , (2005).
9. Azzedine Boukerche; “Handbook of Algorithms for Wireless Networking and Mobile Computing” Chapman & Hall/CRC Taylor & Francis Group, Boca Raton, (2006).
10. Hanzo L., Blogh J .S., & Dr. Ni S.; “3G, HSPA and FDD versus TDD Networking”, (2008).
11. Perez-Romero J., Oriol S. & Diaz-Guerra M. A.; “Radio Resource Management Strategies in UMTS”, John Wiley and Sons, Inc. (2005).
12. Holma H. & Toskala A.; “WCDMA for UMTS: Radio Access for Third Generation Mobile Communications”, John Wiley and Sons, Ltd, England, (2004).
13. Mohamed H., “Call Admission Control in Wireless Networks: a Comprehensive Survey”, IEEE communications Surveys & Tutorials, , pp. 50-69, First Quarter ( 2005).
14. Liers F., & Mitschele-Thiel A.; “UMTS data capacity improvements employing dynamic RRC timeouts”. In PIMRC, (2005).
121
15. Minoru Etoh, “Next Generation Mobile Systems 3G and Beyond”, John Wiley & Sons Ltd England (2005).
16. A Survey Report on “Generations of Network: 1G, 2G, 3G, 4G, 5G”
17. Joschen Schiller; “Mobile communications” second edition Addison Wesley Pearson Education Limited (2003).
18. Parameswaran R.; “Dynamic resource allocation schemes during handoff for mobile multimedia wireless networks”, IEEE Journal on selected areas in communications, July (1999).
19. Paschos G. S., Poltis I. D. & Kotsopoulos S. A.; “A Quality of Service Negotiation-based Admission Control Scheme for WCDMA Mobile Wireless Multiclass Services”, IEEE Transaction On Vehicular Technology., vol. 54, no. 5, sept. (2005).
20. Gerber Z. M., Mao S. S., & Spatscheck O.; “Tail Optimization Protocol for Cellular Radio Resource Allocation”, F. Qian, Z. Wang, A. In ICNP, (2010).
21. Jianxin Y., Lei X., Chun N. , David C. W. & Yong H. C.; “Resource allocation for end-to-end QoS provisioning in a hybrid wireless WCDMA and wireline IP-based DiffServ network”; European Journal of Operational Research, July (2007) pp 1139–1160.
22. Mooi C. C. & Qinqing Z.; “Design and Performance of 3G Wireless Networks and Wireless LAN”, Springer Science-l-Business Media, Inc (2006).
23. Christophe Chevallier et al.; “WCDMA (UMTS) Deployment Handbook”: Planning and Optimization Aspects John Wiley & Sons Ltd, England (2006).
24. Agrawal D. P., Qing-An Z.; “Introduction to Wireless and Mobile Systems”, 3rd Edition Cengage Learning USA (2006).
25. Mohamed H.; “Call admission control in wireless networks”: a comprehensive survey, IEEE communications Surveys and Tutorials, first quarter (2005).
26. Leong C. W., Zhuang W., Cheng Y., & Wang L.; “Optimal resource allocation and adaptive call admission control for voice/data integrated cellular networks”, IEEE Transactions on Vehicular Technology 55 (2) (2006) 654-669.
27. Ma X. , Liu Y. & Trivedi K. S.; “Modeling performance analysis for soft handoff schemes in CDMA cellular systems”, IEEE Transactions on Vehicular Technology 55 (2) (2006) 670-680.
28. Ghaderi M., & Bautaba R.; “Call admission control for voice/data integration in broadband wireless networks”, IEEE Transactions on Mobile Computing 5 (3) (2006) 193-207.
122
29. Shin S. M., Cho C., & Sung D. K.; ´Interference-based channel assignment for DS-CDMA cellular systems”, IEEE Transactions on Vehicular Technology (1999) 233-239.
30. Liu Z. & Zarki M. E.; “SIR-based call admission control for DS-CDMA cellular systems”, IEEE journal on Selected Area in communications 12 (4) (1994) 638-644.
31. Kuri J. & Mermelstein P.; ´call admission on the uplink of a CDMA system based on the total received power”, in: proceedings of ICC, vol. 3, Vancouver, BC, Canada, (1999) pp. 1431-1436.
32. Fantacci R., Mennuti G. & Tarchi D., “A priority based admission control strategy for WCDMA systems”, IEEE International Conference on Communications (2005).
33. Kuenyoung K. & Youngnam H.; “A call admission control with the thresholds for multi-rate traffic in CDMA systems” in: Vehicular Technology Conference Proceedings, IEEE 51 (2) (2000) pp 830-834.
34. Wang Y., Weidong W., Zhang J. & Ping Z.; “Admission control for multimedia traffic in CDMA systems”, Communication Technology Proceedings, International Conference,9-11 (2) (2003) pp 799-802.
35. Yu O., Saric E., Li A.; adaptive prioritized admission over CDMA, IEEE Wireless Communications and Networking Conference 2 (2005) pp 1260-1265.
36. Yu O., Saric E., Anfei L.; “Fairly adjusted multimode dynamic guard bandwidth admission control over CDMA systems”, IEEE Journal on Selected Areas in communications 24 (3) (2006) pp 579-592.
37. Huan Chen, Kumar S. , Jay C. C.; “Dynamic call admission control scheme for QoS priority handoff in multimedia cellular systems”, IEEE Wireless Communications and Networking Conference (WCNC2002) 1 (2002) 114-118.
38. Huan C., Sunil K. & Jay C. C.; “QoS-aware radio resource management scheme for CDMA cellular networks based on dynamic interference margin (IGM)”, Computer Networks, 46 (2002) 867-879.
39. Salman, A. A. & Mahmoud, A.S.; “Dynamic radio resource allocation for 3G and beyond mobile wireless networks”, Journal of Computer Communications (2006).
40. 3GPP (2007), “Technical Specification TS 25.201, Physical layer – general description”, version 7.3.0 Release 7. Retrieved August 1st, 2007, from http://www.3gpp.org/ftp/Specs/html-info/25201.htm.
41. Holma, H & Toskala, A; “WCDMA for UMTS, Radio Access for Third Generation Mobile Communications”, 4th edition, John Wiley & Sons Ltd, Chichester. (2007) ISBN: 978-0-470-31933-8.
123
42. Holma, H & Toskala, A; “HSDPA / HSUPA for UMTS”, John Wiley & Sons Ltd, Chichester.
(2006), ISBN: 978-0-470-01884-2.
43. Soldiani D., Li M. & Cuny R; “QoS and QoE Management in UMTS Cellular Systems”, John Wiley & Sons Ltd, Chichester. (2006) ISBN: 978-0-470-01639.
44. Li-Hsing Yen; “Global System for Mobile Communication (GSM)”, Dept. of CSIE, Chung Hua University.
45. Sanchita G. and Amit K; “Call Admission Control in Mobile Cellular Networks”, Springer Heidelberg New York (2013).
46. Ashraf T., Sufian Y., & Sattar A; “Performance Evaluation of Quality Metrics for Single and Multi Cell Admission Control with Heterogeneous Traffic in WCDMA Networks”, International Journal of Engineering and Technology Volume 4 No. 1, January, (2014).
47. Ferdinand A. and Jeyakumar M.K.; “Vertical Handoff and Admission Control Strategy in 4G Wireless Network Using Centrality Graph Theory”, Research Journal of Applied Sciences, Engineering and Technology 7(22): May (2014).
48. Barry Stern; “Next-Generation Wireless Network Bandwidth and Capacity Enabled by Heterogeneous and Distributed Networks” © 2012, 2013 Freescale Semiconductor, Inc.
49. Romano F, Giada M & Daniele T, “A Priority Based Admission Control Strategy for WCDMA Systems” IEEE explore June (2005).