Lte Coverage and Capacity Dimensioning

Ministry of Higher Education National Telecommunication Institute Electronics and Communications Department

Long Term Evolution (LTE) Access Network Coverage and Capacity Dimensioning This thesis submitted in partial fulfilment of the requirements for the degree of high diploma in telecommunication and information engineering. Submitted by

● Amr Abdel-Magid Kassab ● Amr Mahmoud Morsy ● Mohammed Mahmoud Mohammed Saad ● Mohamed Mahmoud Mohamed Tantawy ● Mohamed Morsy Mohamed ● Hanaa Abdelmoety Kamel ● Walaa Abd-Elhamid Elawam Supervised By Dr.Hamed Abdel Fatah El Shenawy Cairo 2013

Acknowledgements First of all, we are grateful to ALLAHALMIGHTY, the most merciful, the most beneficent, who gave us strength, guidance and abilities to complete this thesis in a successful manner. We are thankful to our parents and our teachers that guided us throughout our career path especially in building up our base in education and enhance our knowledge. We are indebted to our supervisor Dr. Hamed Abd El Fattah ElShenawy for his supervision and his co-operation and support really helped us completing our project.

Abstract Long Term Evolution (LTE) is set of enhancement to the current cellular system in use. LTE is designed to have scalable channel bandwidth up to 20MHz, with low latency and packet optimized radio access technology. The peak data rate of LTE is 100 Mbps in downlink and 50 Mbps in the uplink. LTE support both FDD and TDD duplexing. LTE with OFDM technology in the down link, which provides higher spectral efficiency and more robustness against multipath fading LTE with SC-FDMA in the uplink LTE LTE with different MIMO configurations Dimensioning is initial phase of network planning. It provides estimate of the network elements count as well as the capacity of those elements. The purpose of our project to estimate the required number of eNodeBs needed to support users with certain traffic load with a desired level of quality of service (QOS) and cover the area of interest. This estimate fulfills coverage requirements and verified for capacity requirements . Coverage dimensioning occurs via radio link budget (RLB), maximum allowable propagation path loss (MAPL) is obtained. MAPL is converted into cell radius by using appropriate propagation models. The radius of the cell is used to calculate the number of sites required to cover the area of interest. The cell size and the site count are obtained. Capacity planning deals with the ability of the network to provide services to certain numbers of users with a desired level of quality of service (QOS). Capacity based site count is compared with coverage based site count. The greater one is selected as the final site count.

Project objectives  Overview of LTE system architecture and specifications  Dimensioning of LTE Network  Coverage dimensioning via radio link budget and propagation models  Capacity dimensioning  Numerical results using Visual Studio and basic language  Conclusions and suggestions for future work.

List of Contents Item

Page

1.0

Chapter One: Overview of LTE

1-1

1.1

Introduction

1-2

2.2

IMT-Advanced

1-2

1.3

LTE specifications

1-4

LTE Architecture

1-15

2.0

Chapter Two: LTE network dimensioning

2-1

2.1

Introduction

2-2

2.2

LTE network dimensioning

2-2

2.3

LTE network dimensioning inputs

2-6

2.4

Coverage planning inputs

2-7

2.5

Capacity planning inputs

2-8

2.6

LTE network dimensioning outputs

2-8

2.7

Comparison among dimensioning, planning, optimization

2-9

3.0

Chapter Three: Coverage dimensioning

3-1

3.1

Introduction

3-2

3.2

Concepts and Terminology

3-4

3.3

Link Budget Definition

3-5

3.4

Why we use Link Budget?

3-6

3.5

What are the types of Link Budget?

3-6

3.6

Up Link Budget (Up Link coverage)

3-7

3.7

Up Link Budget entries

3-7

3.8

Morphologies Classifications

3-28

3.9

Down Link Budget(Down Link coverage)

3-29

3.10

Down Link limited Link Budget

3-35

3.11

propagation models

3-37

3.12

Classifications of propagation models

3-39

i

3.13

Ericsson variant COST 231 Okomara-Hata wave propagation

3-42

model 3.14

Example of coverage dimensioning (Radio Link budget)

3-44

4.0

Chapter Four: Capacity dimensioning

4-1

4.1

Introduction

4-2

4.2

Uplink capacity

4-3

4.3

Downlink capacity

4-6

4.4

Application or service distribution model

4-13

5.0

Chapter Five: numerical results

5-1

5.1

Uplink budget

5-3

5.2

Effects on cell Radius (R)

5-17

5.3

Downlink capacity

5-21

6.0

Chapter Six: conclusion and suggestions for future work

6-1

6.1

Conclusion

6-2

6.2

Suggestions for future work

6-3

ii

List of figures Items

Page

Figure(1-1)

Overview of IMT advanced

1-2

Figure(1-2)

Resource element and resource block

1-14

Figure(1-3)

LTE architecture

1-15

Figure(1-4)

Evolved Packet System

1-15

Figure(2-1)

LTE network planning process

2-2

Figure(2-2)

Dimensioning basic steps

2-3

Figure(2-3)

LTE network dimensioning inputs

2-6

Figure(2-4)

LTE coverage planning

2-7

Figure(2-5)

LTE dimensioning outputs

2-9

Figure(2-6)

LTE optimization process stages

2-10

Figure(2-7)

LTE optimization process

2-11

Figure(2-8)

LTE optimization process

2-16

Figure(3-1)

LTE Dimensioning Process

3-4

Figure(3-2)

Resource Block Definition in Frequency Domain 3-11

Figure(3-3)

Downlink and Uplink User Scheduling in Time

3-12

and Frequency Domain. Figure (4.1)

channel bandwidth partitioning

4-22

Figure (4-2)

subscriber class deployment model

4-29

Figure(5-1)

flowchart of effective isotropic radiated power

5-3

Figure(5-2)

Effective Isotropic Radiated Power

5-3

Figure(5-3)

flowchart of sensitivity of eNodeB

5-5

Figure(5-4)

Sensitivity of Enhanced nodeB

5-5

Figure(5-5)

flowchart of Interference Margin

5-7

Figure(5-6)

flowchart of Log Normal Fading Margin

5-7

Figure(5-7)

flowchart of total margins

5-8

iii

Figure(5-8)

Total margin

5-8

Figure(5-9)

flowchart of total gains

5-10

Figure(5-10)

flowchart of total losses

5-10

Figure(5-11)

total gains and total losses

5-11

Figure(5-12)

flowchart of maximum allowable path loss

5-12

Figure(5-13)

Max. allowable path loss

5-13

Figure(5-14)

flowchart of cell radius using Ericson variant

5-14

Okumara -Hata Figure(5-15)

flowchart of site count

5-15

Figure(5-16)

cell radius and Site Count

5-15

Figure(5-17)

the effect of cell Loading Factor (Q) on the cell

5-17

Radius (R) Omni Figure(5-18)

the effect of cell Loading Factor (Q) on the cell

5-18

Radius (R) 3 sector Figure(5-19)

the effect of morphology on the cell Radius (R)

5-19

omni Figure(5-20)

the effect of morphology on the cell Radius (R) 3 5-20 sector

Figure(5-21)

downlink capacity

5-21

Figure (5-22)

Final Base site count

5-25

iv

List of tables

Item Table(1-1)

Page

Improvement in downlink spectral efficiency going 1-7 from 2G to 4G System

Table (1-2)

Targets for average spectrum efficiency

1-8

Table (3-1)

Bandwidths and number of physical resource 3-16 blocks

Table(3-2)

Channel models specifications 1

3-18

Table (3-3)

Channel models specifications 2

3-18

Table(3-4)

Channel propagation conditions

3-19

Table(3-5)

Maximum Doppler frequency for each channel 3-19 model

Table(3-6)

Semi –empirical parameters for uplink

3-21

Table(3-7)

Examples of F for varying tilt

3-23

Table(3-8)

Lognormal fading margins for varying standard 3-24 deviation of log normal fading

Table(3-9)

Values of penetration loss on different morphology 3-26 classes

Table(3-10)

Summarizes the features of different morphologies

3-28, 3-29

Table(3-11)

Examples of Fc at cell edge for varying tilt

3-33

Table(3-12)

Semi –empirical parameters for downlink

3-33

Table(3-13)

Fixed attenuation A in Ericsson variant COST 231 3-43 Okumara Hata propagation models

Table (3-14) Example of coverage dimensioning (Radio Link budget) Table(4-1)

3-44 3-45

SINR values corresponding to each modulation 4-4 coding scheme (MCS) v

Table(4-2)

semi- empirical parameters for up link

4-5

Table(4-3)

Semi- empirical parameters for downlink

4-11

Table (4.5)

applications or services distribution model

4-14

Table (4.6)

mobile service flows and QoS parameters

4-19

Table (4.7)

subscriber class distribution model

4-28

Table (4.8)

subscriber class traffic model

4-30

Table (5-1)

Default values of User Equipment Effective 5-4 Isotropic Radiated Power(EIRP)

Table(5-2)

Default values of Enhanced NodeB sensitivity

5-6

Table(5-3)

Default values of total margin

5-9

Table(5-4)

Default values of total Gain and losses

5-12

Table(5-5)

Default values of Maximum allowable path loss 5-14 (MAPL)

Table(5-6)

values of Cell Radius and Site count with 5-16 difference Base stations heights

Table(5-7)

The effect of cell Loading Factor (Q) on the cell 5-17 Radius (R) Omni

Table(5-8)

The effect of cell Loading Factor (Q) on the cell 5-18 Radius (R) 3 sector

Table(5-9)

the effect of eNodeB height on the cell Radius (R) 5-19 omni

Table(5-10)

the effect of eNodeB height on the cell Radius (R) 5-20 3 sector

vi

List of Acronyms and Abbreviations 16QAM: 16 point quadrature amplitude modulation 3GPP: Third Generation Partnership 64QAM: 64 point quadrature amplitude modulation 3G: third generation 4G: fourth generation

A ACK: Acknowledgement AGC: Automatic Gain Control AP: Access Point ARQ: Automatic Repeater Request AUC: Authentication center A/D: Analog to digital ADSL: Assymetric Digital Subscriber Line AMPS: Advanced Mobile Phone Services AWGN: Additive White Gaussian Noise

B BCH: Broadcast Channel BPSK: Binary Phase Shift Keying BSC: Base Station Controller BTS: Base Transceiver Station BW: Bandwidth BER: Bit Error Rate vii

C CDMA: Code Division Multiple Access CW: Continuous Wave CPL: Car Penetration Loss COST: Community Collaborative studies in the areas of science and technology

D DL: Downlink DSL: Digital Subscriber Line D/A: Digital to analog DU: Dense Urban

E EDGE: Enhanced Data Rate for GSM Evolution EIR: Equipment Identity Register EIRP: Effective Isotropic Radiated Power eNodeB: Enhanced NodeB (enhanced base station) EPA: extended pedestrian ETU: extended terrestrial EVA: extended vehicular EPC: Evolved Packet Core EPS: Evoved Packet System

F FDD: Frequency Division Duplex FDMA: Frequency Division Multiple Access FTT: Fast Fourier Transform FM: Frequency Modulation FWLL: Fixed Wireless Local Loop viii

FFM: Fast Fading Margin

G GGSN: Gateway GPRS Serving Node GMSC: Gateway Mobile Switching Center GMSK: Gaussian Minimum Shift Keying GSM: Global System for Mobile GPRS: General Packet Radio Service GUI: Graphical User Interface

H HARQ: Hybrid Automatic Repeater Request HLR: Home Location Register HSCSD: High Speed Circuit Switched Data HSDPA: High Speed Downlink Packet Access HSS: Home Subscriber Server HSUPA: High Speed Uplink Packet Access

I IMS: IP Multimedia Subsystem IM: Interference Margin IP: Internet Protocol

K KPI: Key Performance Indicator

L LTE: Long Term Evolution

M MBMS: Multimedia broadcast multicast services MB-SFN: Multicast/broadcast-single frequency network ix

MIMO: Multi Input Multi Output MME: Mobile Mobility Management Entity MRC: Maximal ratio combining MS: mobile Station MSC: Mobile Switching Center MAPL: Maximum Allowable Path Loss

O OFDM: Orthogonal Frequency Division Multiplexing OMC: Operation and Maintenance Center

P PAPR: Peak -to-average power ratio PCRF: Policy and Charging Rules Function PDCCH: Physical downlink control channel PDN: Public Data Network PLMN: Public land Mobile Network PRB: Physical Resource Block PSK: Phase Shift Keying PSTN: Public Switched Telephone Network P-GW: PDN Gateway PUCCH: Physical Uplink Control Channel PDCCH: Physical Downlink Control Channel

Q QAM: Quadrature Amplitude Modulation QPSK: Quadrature phase shift Keying QOS: Quality Of Service

x

R RFPA: Radio Frequency Power Amplifier RNC: Radio Network Controller RLB: Radio Link Budget

S SC-FDMA: Single Carrier-Frequency Division Multiple Access SGSN: Serving GPRS Support Node SIM: Subscriber Identity Module SINR: Signal Interference -to-noise ratio S-GW: Serving Gateway SRVCC: Single Radio Voice Call Continuity SMS: Short Message Service SU: Sub Urban

T TDD: Time Division Duplexing TDMA: Time Division Multiple Access TMA: Tower Mounted Amplifier

U UE: User Equipment UL: Uplink UMTS: Universal Mobile Telecommunication system UTRAN: UMTS Terrestrial Radio Access Network

V VLR: Visitor Location Register VOIP: Voice over IP

xi

W WCDMA: Wideband Code Division Multiple Access WIMAX: Worldwide Interoperability for Microwave Access

xii

Chapter One Overview of Long Term Evolution (LTE)

Chapter 1: Overview of Long Term Evolution (LTE)

Chapter one Overview of Long Term Evolution (LTE) 1.1. Introduction LTE is designed to meet users need for high speed data and media transport as well as high-capacity voice support .The LTE PHY employs some advanced Technologies that are new to mobile applications these include OFDMA -SC-FDMA –MIMO. The LTE PHY uses OFDMA in downlink and SC-FDMA on up link.

Figure (1-1) Overview of IMT Advanced

1.2. IMT-Advanced International Mobile Telecommunications Advanced (IMTAdvanced) is requirements issued by the ITU-R of the International Telecommunication Union (ITU) in 2008 for what is marketed as 4G mobile phone and Internet access service.

1-2

Chapter 1: Overview of Long Term Evolution (LTE) 1.2.1 IMT ADVANCED Requirements Specific requirements of the IMT-Advanced report included: 1- Based on an all-Internet Protocol (IP) packet switched network 2- Interoperability with existing wireless standards 3- A nominal data rate of 100 Mbit/s while the client physically moves at high speeds relative to the station,50 Mbit /s in the uplink and 1 Gbit/s while client and station are in relatively fixed positions. 4- Dynamically share and use the network resources to support more simultaneous users per cell. 5- Scalable channel bandwidth 1.4 MHz, 3 MHz, 5 MHz, 15 MHz and 20 MHz optionally up to 40 6- Peak link spectral efficiency of 15 bit/s/Hz in the downlink, and 6.25bit/s/Hz in the uplink (meaning that 1 Gbit/s in the downlink should be possible over less than 67 MHz bandwidth) 7- System spectral efficiency of up to 3 bit/s/Hz/cell in the downlink and 2.25 bit/s/Hz/cell for indoor usage 8- Seamless connectivity and global roaming across multiple networks with smooth handovers 9- Ability to offer high quality of service for multimedia support 10- support antenna configurations a- Downlink 4×2, 2×2, 1×2, 1×1 b- Uplink 1×2, 1×1 11- coverage a- full performance up to 5 km b-slight degradation 5 km-30 km c-operation up to 100 km should not be precluded by standard

1-3

Chapter 1: Overview of Long Term Evolution (LTE) 12- mobility a- optimized for low speed less than 15 km per hour b-high performance at speeds up to 120 km per hour c-maintain link at speeds up to 350 km per hour 13- LTE support efficient broadcast mode performance :multicast and broadcast 14- broadcast spectral efficiency 1bit /sec/Hz 15- LTE support paired and unpaired frequency band 16- It support FDD and TDD, half duplex TDD 17- Support adaptive modulation technique: High level and low level modulation 18- Support scalable FFT size 19- It support turbo code 20- It support low complexity low cost terminal 21- Support VOIP 60 session /Hz/cell 22- Support of cell sizes from tens of meters of radius (femto and Pico cells) up to over 100 km radius microcells 23- Simplified architecture: The network side of EUTRAN is composed only by the enodeBs. 24- Low data transfer latencies (sub-5ms latency for small IP packets in optimal conditions), lower latencies for handover and connection setup time. 1.3 LTE specifications 1.3.1 Peak Rates and Peak Spectral Efficiency For Data rate many services with lower data rates such as voice services are important and still occupy a large part of a mobile network’s overall capacity, but it is the higher data rate services that drive the design

1-4

Chapter 1: Overview of Long Term Evolution (LTE) of the radio interface. The ever increasing demand for higher data rates for web browsing, streaming and file transfer pushes the peak data rates for mobile systems from kbit/s for 2G, to Mbit/s for 3G and getting close to Gbit/s for 4G (Erik Dahlman, Stefan Parkvall, and Johan Sköld, 2011). For marketing purposes, the first parameter by which different radio access technologies are usually compared is the peak peruser data rate which can be achieved. This peak data rate generally scales according to the amount of spectrum used, and, for MIMO systems, according to the minimum of the number of transmit and receive antennas. The peak data rate can be defined as the maximum throughput per user assuming the whole bandwidth being allocated to a single user with the highest modulation and coding scheme and the maximum number of antennas supported. Typical radio interface overhead (control channels, pilot signals, guard intervals, etc.) is estimated and taken into account for a given operating point. For TDD systems, the peak data rate is generally calculated for the downlink and uplink periods separately. This makes it possible to obtain a single value independent of the uplink/downlink ratio and a fair system comparison that is agnostic of the duplex mode. The maximum spectral efficiency is then obtained simply by dividing the peak rate by the used spectrum allocation. The target peak data rates for downlink and uplink in LTE Release 8 were set at 100 Mbps and 50 Mbps respectively within a 20 MHz bandwidth, 7 corresponding to respective peak spectral efficiencies of 5 and 2.5 bps/Hz. The underlying assumption here is that the terminal has two receive antennas and one transmit antenna. The number of antennas used at the base station is more easily upgradeable by the network

1-5

Chapter 1: Overview of Long Term Evolution (LTE) operator, and the first version of the LTE specifications was therefore designed to support downlink MIMO operation with up to four transmit and receive antennas. When comparing the capabilities of different radio communication technologies, great emphasis is often placed on the peak data rate capabilities. While this is one indicator of how technologically advanced a system is and can be obtained by simple calculations, it may not be a key differentiator in the usage scenarios for a mobile communication system in practical deployment. Moreover, it is relatively easy to design a system that can provide very high peak data rates for users close to the base station, where interference from other cells is low and techniques such as MIMO can be used to their greatest extent. It is much more challenging to provide high data rates with good coverage and mobility, but it is exactly these latter aspects which contribute most strongly to user satisfaction. In typical deployments, individual users are located at varying distances from the base stations, the propagation conditions for radio signals to individual users are rarely ideal, and the available resources must be shared between many users. Consequently, although the claimed peak data rates of a system are genuinely achievable in the right conditions, it is rare for a single user to be able to experience the peak data rates for a sustained period, and the envisaged applications do not usually require this level of performance. A differentiator of the LTE system design compared to some other systems has been the recognition of these „typical deployment constraints‟ from the beginning. During the design process, emphasis was therefore placed not only on providing a competitive peak data rate for use when conditions allow, but also

1-6

Chapter 1: Overview of Long Term Evolution (LTE) importantly on system level performance, which was evaluated during several performance verification steps. System-level evaluations are based on simulations of multicell configurations where data transmission from/to a population of mobiles is considered in a typical deployment scenario. The sections below describe the main metrics used as requirements for system level performance. In order to make these metrics meaningful, parameters such as the deployment scenario, traffic models, channel models and system configuration need to be defined (Stefanie Sesia, Issam Toufik and Matthew Baker, 2011).

Table (1-1): Improvement in downlink spectral efficiency going from 2G to 4G System 1.3.2 Spectrum efficiency In this section, the target for peak spectrum efficiency, the average spectrum efficiency, and cell edge spectrum efficiency are defined. The target for average spectrum efficiency and the cell edge user throughput efficiency should be given a higher priority than the target for peak spectrum efficiency and VoIP capacity. The target for average spectrum

1-7

Chapter 1: Overview of Long Term Evolution (LTE) efficiency and the cell edge spectrum efficiency should be achieved simultaneously. The peak spectrum efficiency is the highest data rate normalized by overall cell bandwidth assuming error-free conditions, when all available radio resources for the corresponding link direction are assigned to a single UE.

The system target to support downlink peak spectrum

efficiency of 30 bps/Hz and uplink peak spectrum efficiency of 15 bps/Hz. Assumption of antenna configuration is (8x8) or less for DL and( 4x4) or less for UL Average spectrum efficiency is defined as the aggregate throughput of all users (the number of correctly received bits over a certain period of time) normalized by the overall cell bandwidth divided by the number of cells. The average spectrum efficiency is measured in b/s/Hz/cell. Advanced E-UTRA should target the average spectrum efficiency to be as high as possible, given a reasonable system complexity. The expectation at the end of the study item is that the values of all the targets (of the different configurations) will be made available, but currently the evaluation for the blanked out boxes in the table below, are a lower priority. Advanced E-UTRA should target the average spectrum efficiencies in different environments in Table (2-2).

Table (1-2): Targets for average spectrum efficiency

1-8

Chapter 1: Overview of Long Term Evolution (LTE) 1.3.3 Cell edge user throughput The cell edge user throughput is defined as the 5% point of CDF of the user throughput normalized with the overall cell bandwidth. Advanced E-UTRA should target the cell edge user throughput to be as high as possible, given a reasonable system complexity. A more homogeneous distribution of the user experience over the coverage area is highly desirable and therefore a special focus should be put on improving the cell edge performance. The expectation at the end of the study item is that the values of all the targets (of the different configurations) will be made available, but currently the evaluation for the blanked out boxes in the table below, are a lower priority. Advanced E- UTRA should target the cell edge user throughput below in different environments 1.3.4 Voice Capacity (VOIP) VoIP services convert your voice into a digital signal that travels over the Internet. If you are calling a regular phone number, the signal is converted to a regular telephone signal before it reaches the destination. VoIP can allow you to make a call directly from a computer, a special VoIP phone, or a traditional phone connected to a special adapter. In addition, wireless "hot spots" in locations such as airports, parks, and cafes allow you to connect to the Internet and may enable you to use VoIP service wirelessly. Some VoIP providers offer their services for free, normally only for calls to other subscribers to the service. Your VoIP provider may permit you to select an area code different from the area in which you live. It

1-9

Chapter 1: Overview of Long Term Evolution (LTE) also means that people who call you may incur long distance charges depending on their area code and service. Some VoIP providers charge for a long distance call to a number outside your calling area, similar to existing, traditional wire line telephone service. Other VoIP providers permit you to call anywhere at a flat rate for a fixed number of minutes. Depending upon your service, you might be limited only to other subscribers to the service, or you may be able to call anyone who has a telephone number - including local, long distance, mobile, and international numbers. If you are calling someone who has a regular analog phone, that person does not need any special equipment to talk to you. Some VoIP services may allow you to speak with more than one person at a time. Some VoIP services offer features and services that are not available with a traditional phone, or are available but only for an additional fee. You may also be able to avoid paying for both a broadband connection and a traditional telephone line.

If you're

considering replacing your traditional telephone service with VoIP, there are some possible differences. Some VoIP services don't work during power outages and the service provider may not offer backup power. Not all VoIP services connect directly to emergency services through 9-1-1. For additional information VoIP providers may or may not offer directory assistance/white page listings. 1.3.5 Mobility The system shall support mobility across the cellular network for various mobile speeds up to 350km/h (or perhaps even up to 500km/h depending on the frequency band).

1 - 10

System performance shall be

Chapter 1: Overview of Long Term Evolution (LTE) enhanced for 0 to 10km/h and preferably enhanced but at least no worse than E-UTRAand E-UTRAN for higher speeds. 1.3.6 Control Plane and User Plane Latency Control plane deals with signaling and control functions, while user plane deals with actual user data transmission. C-Plane latency is measured as the time required for the UE (User Equipment) to transit from idle state to active state. In idle state, the UE does not have an Reconnection. Once the RRC is setup, the UE transitions to connected state and then to the active state when it enters the dedicated mode. U-Plane latency is defined as one way state when it enters the dedicated mode. U-Plane latency is defined as one-way transmit time between a packet being available at the IP layer in the UE/E-UTRAN (Evolved UMTS Terrestrial Radio Access Network) edge node and the availability of this packet at the IP layer in the EUTRAN/ UE node. U-Plane latency is relevant for the performance of many applications. This tutorial presents in detail the delay budgets of C-Plane and U-Plane procedures that add to overall latency in state transition and packet transmission. Latency calculations are made for both FDD and TDD modes of operation. Technical details of C-Plane and U-Plane latency .This tutorial is organized as follows: Requirements and assumptions in Section This tutorial presents in detail the delay budgets of C- Plane and U-Plane procedures that add to overall latency in state transition and packet transmission. Latency calculations are made for both FDD and TDD modes of operation. Technical details of C-Plane and U-Plane latency are cited in This tutorial is organized as follows: Requirements

1 - 11

Chapter 1: Overview of Long Term Evolution (LTE) and assumptions in Section II, C- Plane latency analysis in Section III and U-Plane latency analysis in Section IV. The conclusions are summarized in Section V. All the values indicated in the tables are in mill seconds (ms). The method of calculating these latencies is illustrated in the appendix. Low latency where5 ms user plane latency for small IP packets (user equipment to radio access network [RAN] edge) .100 ms camped to active. 50 ms dormant to active. Scalable bandwidth where the 4G channel offers four times more bandwidth than current 3G systems and is scalable. So, while 20 MHz channels may not be available everywhere, 4G systems will offer channel sizes down to 5 MHz, in increments of 1.5 MHz. 1.3.7 Spectrum Allocation and Duplex Modes Transmission techniques exist  Simplex One party transmits data and the other party receives data.No simultaneous transmission is possible, the communication is one-way and only one frequency (channel) is used.  Half Duplex Each party can receive and transmit data, but not at the same time. The communication is two-way and only one frequency (channel) is used.  Full Duplex Each party can transmit and receive data simultaneously.

1 - 12

Chapter 1: Overview of Long Term Evolution (LTE) The communication is two-way and two frequencies. Full duplex main methods used are  Time Division Duplexing (TDD) The communication is done using one frequency, but the time for transmitting and receiving is different. This method emulates full duplex communication using a half-duplex link.  Frequency Division Duplexing (FDD) The communication is done using two frequencies and the transmitting and receiving of data is simultaneous. The advantages of TDD are typically observed in situations uplink and downlink data transmissions are not symmetrical. Transmitting and receiving is done using one frequency, the channel estimations for beam forming (and other smart antenna techniques) apply for both the uplink and the downlink. A typical disadvantage of TDD is the need to use guard periods between the downlink and uplink transmissions. The advantages of FDD are typically observed in situations where the uplink and downlink data transmissions are symmetrical (which is not usually the case when using wireless phones). More importantly, when using FDD, the interference between neighboring Radio Base Stations (RBSs) is lower than when using TDD. Also, the spectral efficiency (which is a function of how well a given spectrum is used by certain access technology) of FDD is greater than TDD.

1 - 13

Chapter 1: Overview of Long Term Evolution (LTE) Frequency band from 2600MHz to 2.6 GHz. Channel bandwidth up to 20 MHz Channel bandwidth on-demand (1.4 MHz, 3MHz, 5MHz, 10MHz, 15MHz, 20MHz). Charging / volume 1.3.8 Resource element and resource block A resource element is the smallest unit in the physical layer and occupies one OFDM or SC-FDMA symbol in the time domain and one subcarrier in the frequency domain as shown in figure (2-1) . Aresource block (RB) is the smallest unit that can be scheduled for transmission. An RB physically occupies 0.5 ms (1 slot) in the time domain and 180 KHz in the frequency domain .the number of subcarriers per RB and the number of symbols per RB vary as a function of the cyclic prefix length and subcarrier spacing.

Figure (1-2): Resource element and resource block

1 - 14

Chapter 1: Overview of Long Term Evolution (LTE) 1.4 LTE architecture

Figure (1-3) LTE architecture The combination of the EPC and the evolved RAN ( E-UTRAN) is the evolved packet system (EPS).

Figure (1-4) Evolved Packet System

1 - 15

Chapter 1: Overview of Long Term Evolution (LTE) 1.4.1 Access network E-UTRAN Consists only of enodeBs on the network side. The enodeB performs tasks similar to those performed by the nodeBs and RNC (radio network controller) together in UTRAN. The aim of this simplification is to reduce the latency of all radio interface operations. The eNBs are interconnected with each other by means of the X2 interface. The eNBs are connected by the S1 interface to the EPC (Evolved Packet Core). The eNB connects to the MME (Mobility Management Entity) by means of the S1-MME interface and to the Serving Gateway (S-GW) by means of the S1-U interface. The S1 interface supports a many-to-many relation between MMEs / Serving gateways and eNBs. eNodeB eNB interfaces with the UE and hosts the Physical (PHY), Medium Access Control (MAC), Radio Link Control (RLC), and Packet Data Control Protocol (PDCP) layers. It also hosts Radio Resource Control (RRC) functionality corresponding to the control plane. It performs many functions including radio resource management, admission control, scheduling, enforcement of negotiated UL QoS, cell information broadcast, ciphering/deciphering of user and control plane data, and compression/decompression of DL/UL user plane packet headers. Functions of eNodeB  Transmission & Reception

1 - 16

Chapter 1: Overview of Long Term Evolution (LTE)  Modulation & Demodulation  Radio resources allocation  Error Detection and Correction  Connectivity to the EPC  Header Compression & packet encryption  Scheduling and transmission of broadcast information 1.4.2 CORE NETWORK ( EPC ) The main logical nodes of the EPC are:  Mobility Management Entity (MME)  PDN Gateway (P-GW)  Policy and Charging Rules Function (PCRF)  Serving Gateway (S-GW).  Home Subscriber Server (HSS) 1- MME Mobility Management Entity is the control node that processes the signaling between the UE and the CN. Manages and stores UE context (for idle state: UE/user identities, UE mobility state, user security parameters). It generates temporary identities and allocates them to UEs.  Security Procedures (by interacting with the HSS).  Idle mode UE Tracking Area update & Paging  Handling QoS

1 - 17

Chapter 1: Overview of Long Term Evolution (LTE)  Choosing the SGW for a UE at the initial attach and at time of intra-LTE handover involving Core Network node relocation. 2-P-GW The PDN GW provides connectivity to the UE to external packet data networks by being the point of exit and entry of traffic for the UE The Packet data network gateway is responsible for:  IP address allocation for the UE  Charging (according to rules from the PCRF )  Filtering of downlink user IP packets into the different QoS based bearers  mobility anchor for interworking with non-3GPP technologies such as CDMA2000 and WiMAX networks 3-PCRF The Policy and Charging Rules Function is responsible for :  Real time Determination of policy & charging rules 

QoS handling.

4-S-GW The SGW routes and forwards user data packets, while also acting as the mobility anchor for the user plane during inter-eNB handovers and as the anchor for mobility between LTE and other 3GPP technologies (terminating S4 interface and relaying the traffic between 2G/3G systems and PDN GW).

1 - 18

Chapter 1: Overview of Long Term Evolution (LTE) The serving gateway is responsible for:  Routes and forwards user data packets  Mobility anchor for intra E-UTRAN mobility (when the UE moves between eNodeBs)  Mobility anchor with 2G/GSM and 3G/UMTS mobility.

5-HSS  Users subscription data  Information about the PDNs to which the user can connect  The identity of the MME to which the user is currently attached or registered  Authentication information

1 - 19

Chapter Two LTE Network Dimensioning

Chapter 2: LTE Network Dimensioning Chapter Two LTE Network Dimensioning 2.1 Introduction Dimensioning is a part of the whole planning process, which also includes detailed planning and optimization of the wireless cellular network as shown in figure: (2-1) Planning Dimensioning Requirements and strategy for coverage, capacity and quality

Coverage planning and site selection and acquisition Capacity requirement

Optimization Performance analysis in terms of quality, efficiency and availability

Parameter planning

Figure: (2-1) LTE network planning process

2.2 LTE network dimensioning It is the initial phase of network planning. It provides the first estimate of the network element count as well as the capacity of those elements. The purpose of dimensioning is to estimate the required number of eNodeBs needed to support a specified traffic load in an area. The aim of this whole exercise is to provide a method to design the wireless cellular network such that it meets the requirements set forth by the customer. This process can be modified to fit the needs of any wireless cellular network. This is a very important process in network deployment. Wireless cellular network dimensioning is directly related to the quality and effectiveness of the network. And can deeply affect its development. Wireless cellular network dimensioning follows basic steps shown in figure:

2-2

Chapter 2: LTE Network Dimensioning

Dimensioning steps

Data/Traffic Analysis

Coverage estimation

Capacity Evaluation

Transport Dimensioning

Figure (2-2): Dimensioning basic steps 2.2.1 Data and Traffic analysis This is the first step in LTE dimensioning. It involves gathering of required inputs and their analysis to prepare them for use in LTE dimensioning process. Operator data and requirements are analysed to determine the best system configuration. Wireless cellular dimensioning requires some fundamental data elements. These parameters include subscriber population, traffic distribution, geographical area to be covered, frequency band, allocated bandwidth, and coverage and capacity requirements. Propagation models according to the area and frequency band should be selected and modified if need. This is necessary for coverage estimation. System specific parameters like, transmit power of the antennas, their gains, estimate of system losses, type of antenna system used etc., must be known prior to the start of wireless cellular network dimensioning. Each wireless network has its own set of parameters. Traffic analysis gives an estimate of the traffic to be carried by the system. Different types of traffic that will be carried by the network are modulated. Traffic types may include voice calls, VOIP, PS or CS traffic. Overheads carried by each type of traffic are calculated and included in the model. Time and amount of traffic is also forecasted to evaluate the performance of the network and to determine whether the network can fulfil the requirements set forth.

2-3

Chapter 2: LTE Network Dimensioning 2.2.2 Coverage estimation It is used to determine the coverage area of each eNodeB. Coverage estimation calculates the area where eNodeB can be heard by the users (receivers). It gives the maximum area that can be covered by eNodeB. But it is not necessary that an acceptable connection (e.g a voice call) between eNodeB and receiver can be established in coverage area. However eNodeB can be detected by the receiver in coverage area. Coverage analysis fundamentally remains the most critical step in the design of LTE network as with 3G systems. RLB (Radio Link Budget) is at the heart of coverage planning which allows the testing of path loss model and the required peak data rates against the target coverage levels. The result is the effective cell range to work out the coverage-limited site count. This requires the selection of appropriate propagation model to calculate path loss. LTE RLB with the knowledge of cell size estimates and of the area to be covered is an estimate of the total number of sites is found. This estimate is based on coverage requirements and needs to be verified for the capacity requirements. Coverage planning includes radio link budget and coverage analysis RLB comprises of all the gains and losses in the path of the signal from transmitter to receiver. This includes transmitter and receiver gains as well as losses and the effect of the wireless medium between them. Free space propagation loss, fast fading and slow fading in taken into account. Additionally, parameters that are particular to some systems are also considered. Frequency hopping and antenna diversity margins are two examples. 2.2.3 Capacity evaluation Capacity planning deals with the ability of the network to provide services to the users with a desired level of quality. After the site coverage area is calculated using coverage estimation, capacity relates issues are analyzed. This involves selection of site and system configuration, e.g. channels used channel elements and sectors. These elements are different for each system. Configuration is selected such that it fulfils the traffic requirements. In some wireless cellular systems, coverage and capacity are interrelated, e.g. in WCDMA.

2-4

Chapter 2: LTE Network Dimensioning In this case, data pertaining to user distribution and forecast of subscriber’s growth is of almost importance. Dimensioning team must consider these values as the have direct impact on coverage and capacity, Capacity evaluation gives an estimate of the number of sites required to carry the anticipated traffic over the coverage area. Once the number of sites according to the traffic forecast is determined, the interfaces of the network are dimensioned. Number of interfaces can vary from a few in some systems to many in others. The objective of this step is to perform the allocation of traffic in such a way that no bottle neck is created in the wireless network. All the quality of service requirements are to be met and cost has to be minimized. Good interface dimensioning is very important for smooth performance of the network. With a rough estimate of the cell size and site count, verification of coverage analysis is carried out for the required capacity. It is verified whether with the given site density, the system can carry the specified load or new sites have to be added. In LTE, the main indicator of capacity is SINR distribution in the cell. This distribution is obtained by carrying out system levels simulations. SINR distribution can be directly mapped into system capacity (data rate). LTE cell capacity is impacted by several factors, for example, packet scheduler implementation, supported MCSs, antenna configurations and interference level. Therefore, many sets of simulation results are required for comprehensive analysis. Capacity based site count is then compared with coverage result and greater of the two numbers is selected as the final site count, as already mentioned in the previous section. 2.2.4 Transport Dimensioning Transport dimensioning deals with the dimensioning of interfaces between different network elements. In LTE, S1 (between eNodeB and a GW) and X2 (between two eNodeBs) are the two interfaces to be dimensioned. These interfaces were still in the process of being standardized at the time of this work. Therefore, transport dimensioning is not included in this thesis work.

2-5

Chapter 2: LTE Network Dimensioning An initial sketch of LTE network is obtained by following the above mentioned steps of dimensioning exercise. This initial assessment forms the basis of detailed planning phase. In this thesis, main emphasis is on steps two to four. First step is unnecessary because the data for the test cases is taken from WIMAX scenario, allowing its by pass. Coverage and capacity planning is dealt in detail and resulting site count is calculated to give an estimate of the dimensioned LTE network. Dimensioning of LTE will depend on the operator strategy and business case. The physical side of the task means to find the best possible solution of the network which meets operator requirements and expectations. In detail and resulting site count is calculated to give an estimate if the dimensioned LTE network. Dimensioning of LTE will depend on the operator strategy and business case. The physical side of the task means to find the best possible solution of the network which meets operator requirements and expectative. 2.3 LTE network dimensioning inputs LTE dimensioning inputs used in the development of methods and models for LTE dimensioning. LTE dimension inputs can be broadly divided into three categories; quality, coverage and capacity related inputs. LTE network dimensioning has three main processes shown in figure (2-3).

Dimensioning inputs

Quality inputs

Coverage planning inputs

Capacity planning inputs

Figure (2-3): LTE network dimensioning inputs

2-6

Chapter 2: LTE Network Dimensioning 2.3.1 Quality inputs Quality inputs include average cell throughput and blocking probability. These parameters are the customer requirements to provide a certain level of service to its users. These inputs directly translate into Qos parameters. Besides cell edge coverage probability is used in the dimensioning tool to determine the cell radius and thus the site count. Three methods are employed to determine the cell edge. These include user defined maximum throughput at the cell edge, maximum coverage with respect to lowest MCS (giving the minimum possible site count) and predefined cell radius. With a predefined cell radius, parameters can be varied to check the data rate achieved at this cell size. This option gives the flexibility to optimize transmitted power and determining a suitable data rate corresponding to this power. 2.4 Coverage planning inputs Required coverage probability plays a vital role in determination of call radius. Even a minor change in coverage probability causes a large variation in cell radius as shown in figure (2-4)

Radio Link budget (RLB)

MAPL Propagation model Cell size

Figure (2-4): LTE coverage planning LTE dimensioning inputs for coverage planning exercise are similar to the corresponding inputs for 3G UMTS networks. Radio link budget (RLB) is of central importance to coverage planning in LTE.

2-7

Chapter 2: LTE Network Dimensioning RLB inputs include transmitter and receiver antenna systems, number of antennas used, conventional system gains and losses, cell loading and propagation models. LTE can operate in both the conventional frequency bands of 900 and 1800 MHz as well as extended band of 2600 MHz. Models for all the three possible frequency bands are incorporated in this work. Additionally, channel types (pedestrian, Vehicular) and geographical information is needed to start the coverage dimensioning exercise. Geographical input information consists of area type information (Urban, Rural, etc.) and size of each area type to be covered. Furthermore, required coverage probability plays a vital role in determination of cell radius. Even a minor change in coverage probability causes a large variation in cell radius. 2.5 Capacity planning inputs Capacity planning inputs provides the requirements, to be met by LTE network dimensioning exercise. Capacity planning inputs gives the number of subscribers in the system, their demanded services and subscriber usage level. Available spectrum and channel bandwidth used by the LTE system are also very important for LTE capacity planning. Traffic analysis and data rate to support available services (Speech, Data) are used to determine the number of subscribers supported by a single cell and eventually the cell radius based on capacity evaluation. LTE system level simulation results and LTE link level simulation results are used to carry out capacity planning exercise along with other inputs. These results are obtained from Nokia's internal sources. Subscriber growth forecast is used in this work to predict the growth and cost of the network in years to come. This is a marketing specific input targeting the feasibility of the network over a longer period of time. Forecast data will be provided by the LTE operators. 2.6 LTE network dimensioning outputs Outputs or targets of LTE dimensioning process have already been discussed indirectly in the previous section. Outputs of the dimensioning phase are used to estimate the feasibility and cost of the network. These outputs are further used in detailed network planning. Dimensioning LTE network can help out LTE core network team to plan a suitable network design and to determine the number of backhaul links required in the starting phase of the network as shown in figure (2-5) Cell size is the main output of LTE dimensioning exercise.

2-8

Chapter 2: LTE Network Dimensioning Two values of cell radius are obtained, one from coverage evaluation and second from capacity evaluation. The larger of the number is taken as the final output. Cell radius is then used to determine the number of sites. Assuming a hexagonal cell shape, number of sites can be calculated by using simple geometry. This procedure is explained capacities of eNBs are obtained from capacity evaluation, along with the number of subscribers supported by each cell. Interface dimensioning is the last step in LTE access network dimensioning, which is out of scope of this thesis work. The reason is that LTE interfaces (S1 and S2) were still undergoing standardization.

Population statistics Number of subscribes Dimensioning Outputs

Area to be covered by the network Subscriber geographical spread Cell throughput Final site-count

Figure (3-5): LTE dimensioning outputs 2.7 Comparison among dimensioning, planning and optimization. Dimensioning is the initial phase of network planning. It provides the first estimate of the network element count as well as the capacity of these elements. The purpose of dimensioning is t estimate the required number of the radio base stations needed to support a specified traffic load in an area. The radio network planning process is designed to maximize the networks coverage, whilst at the same time providing the desired

2-9

Chapter 2: LTE Network Dimensioning capacity. In order to achieve this, there are number of stages that are typically performed, these include: Initial Planning, Detailed Planning and Optimization. Optimization is probably the most important stage when planning an LTE network. Typically it can be split into pre-launch optimization. There are however a number of different areas that may be optimized, these include.

Figure (2-6) optimization stages of LTE

2.7.1 Planning of LTE The radio network planning process is designed to maximize the networks coverage, whilst at the same time providing the desired capacity. In order to achieve this, there are a number of stages that are typically performed, these include:  Nominal or preliminary planning  Detailed planning  Optimization

2 - 10

Chapter 2: LTE Network Dimensioning

Figure (2.7) the cellular network planning processes 2.7.1.1 Nominal or preliminary cell planning A nominal or preliminary cell plan can be produced from the data compiled from coverage and traffic analysis. The nominal cell plains a graphical representation of the network and looks like a cell pattern on a map. During nominal cell planning, do not care about the position of the sites taking only in consideration the separation distance between sites. To simplify the network planning, hexagonal shaped cells are adopted although they are artificial or fictitious and do not exist in real world but it have become a widely promoted symbols for cellular structured system. Nominal cell plans are the first cell plans and forms the basis for further planning. In reality, each company has a planning tool which is a work station equipped with a software package based on link budget calculations and using certain propagation model to determine the cell radius and the results are displayed on the map using different colors. An up to date digital three dimensional map with high resolutions for the area where the network is to be planned is used to import the actual environment data that include the terrain fluctuations (height information), clutter distribution, dense degree of the area of interest. The area of interest is

2 - 11

Chapter 2: LTE Network Dimensioning divided into different sub regions according to different environment definitions. Each sub region has its own characteristics. The classification is based on the dense of buildings and their heights in the sub region. Each sub region is classified into one of the four categories: dense urban (DU), urban (UR), suburban (SU) and rural (RU).The planning tool determines the classification of each sub region. It is possible to import data from site survey files. Data can also be imported from field measurements files to tune the propagation model as will be explained in the following subsections. The area where the network is to be planned to be covered with cellular structured system is used. Two study cases are investigated:  Coverage oriented environment and rural environments.

represent

suburban

 Capacity oriented environment represent dense urban and urban environments. Using the software program developed by us the maximum allowable path loss (MAPL) is calculated using reverse link budget and forward link budget and the link balance was made and the least value was taken as an input to the propagation model. Thus, the cell radius was calculated using coverage criterion. The classification of sub regions according to their building density and heights is determined by us during site survey by observing the area features, landmarks and terrain in each sub region. 2.7.2 Detailed planning 2.7.2.1 Site surveys Once the nominal cell planning has been completed, site surveys can be performed for all the proposed site locations by the site survey team. The site survey includes: site search, candidate sites are chosen, the site survey team check the validity of each location of the sites, contact with the site owner, site location lease agreement, get permission of the new sites, and carry out the construction of the civil works, tower erection, transmission and interconnection between the network entities. Finally site acquisition. The following items must be checked for each site: The space for the equipment including: antennas, cable runs and power facilities. The exact site locations (with some shifts)are fed back to the network planning team to modify the network planning by shifting the

2 - 12

Chapter 2: LTE Network Dimensioning locations of the sites such that no dead zones were introduced and overlap between sites were reduced as much as possible. 2.7.2.2 Field measurements The purpose of the field measurements is to correct the propagation model to reflect the propagation status of wireless signal in the environment of the area of interest, thus making the model more practical meet the coverage requirement. To conduct field tests, the following steps have to be followed: You have to choose the frequency of the measurement. If there is interference on the frequency point to be used, choose a frequency point without interference. The transmission characteristics are almost the same when frequency difference is 10 MHz or so. Field measurements site choice: You have to choose the field measurements site. The field measurements site should not be too much higher than the surrounding buildings and 10 meters are suitable. To obtain as much data as possible for correcting various clutters, two or three field measurements sites with similar surrounding clutters (building heights, site height, and so on) can be chosen to carry out field measurements and data from several sites can be synthesized to execute the correction of the various clutters. Choose pertinent parameters of the field measurements site i.e. use omnidirectional antenna, choose proper transmission power, no obstruction surrounding the field measurements site, and clean the frequency point. The tools for field measurements includes: transmitter or CW transmitter, scanner or field strength meter and GPS handset. Before field measurements, you have to span antennas, install transmitter, and adjust output power and frequency point to proper values and transmitting signal. After field measurements, the field measurements data is put into a form acceptable for the planning tool load the field measurements file into the planning tool and correct the model. 2.7.2.3 System design (or final cell plan) The actual and the exact site locations are used to produce the final cell planning which is used for network installations, provided that no dead zones and overlap between sites is small as possible 2.5 System diagnosis The test team via the driving test and using test mobile system which is a testing tool. The testing tool includes mobile test units (MTUs) in cars and fixed test units geographically distributed. The testing tool consists of a MS with special software, a portable personal computer (PC) and a

2 - 13

Chapter 2: LTE Network Dimensioning global positioning system (GPS) receiver and mobile traffic recording (MTR) and cell traffic recording (CTR). The MS is used in active and idle mode. The PC is used for presentation, control and measurement data storage. The GPS receiver provides the exact position of the measurement site by utilizing satellites. When the satellite signals are shadowed, the GPS system switches to dead reckoning. Dead reckoning consists of a speed sensor and a gyro. This provides the position if satellite signals are lost temporarily. The measurement data can be imported to the planning tool and can be displayed on a map to compare the measured handoffs with the predicted cell boundaries for example to check the network performance, to evaluate the customer complaints, to verify that the final cell planning was implemented successfully. 2.7.2.4 System tuning After installation of the network, it is continuously monitored to determine how well it meets the coverage and capacity requirement using the measured data, parameters are changed. Other measurements can be taken if necessary. The parameters to be changed are such as eNodeB transmitted power, eNodeB antenna height, antenna down tilting angle, antenna type (gain, horizontal HPBW, and so on). Change handoff parameters, change, add or decrease channels. 2.7.2.5 System growth Cell planning is an ongoing process. If the network needs to be expanded to extend coverage due to increase in traffic of because or change in the environment Starting with a new capacity or traffic and coverage or power analysis. 2.7.2.6 eNodeB site choice When choosing eNodeB site, the following rules should be obeyed: 1) Antenna height should be higher to some degree than the surroundings. 2) Ensure that there is no obvious obstruction in surrounding environments. 3) Ensure that there is no obstruction surrounding the position of setting the global positioning (GPS) antenna. 4) Meet coverage goal requirement concerning the effective coverage of the eNodeB. 5) Predict traffic distribution in the coverage area and set the eNodeB sites on the places of real traffic need.

2 - 14

Chapter 2: LTE Network Dimensioning 6) Utilize existent sites such as telecom Egypt centrals in case of rural communication network and use other communication resources as possible such as towers, buildings. 7) Guarantee necessary space separation concerning the interference from other systems. 8) Avoid strong wireless transmitter, radar or other serious interference. 9) Choose places with convenient traffic, reliable electricity plant, if not available use generators or solar cell panels 10) Avoid being near the flammable or explosive buildings. 11) Avoid being near the industrial manufactories with poisonous gas or smoke and dust. 12) Avoid hospitals, educational buildings, military zones, church, mosques, and entertainment areas. 2.7.2.7 Antenna configuration and cell type choice The choice of eNodeB antenna should concern with the following factors: site type, dense degree of eNodeB and relative positions between them and dense degree of the area and so on. The following rules should be obeyed when choosing antennas: 1) In dense urban (DU) and urban (UR) areas i.e. in capacity oriented areas, sectorized cells or directional antennas with narrow power beam width (HPBW) angle can be chosen and large gain can be chosen to reduce the other cell interference and increase the capacity. 2) In suburban areas and rural areas with low capacity where user or population density is low i.e. In coverage oriented areas, Omni cells with omnidirectional antennas with high antenna height can be chosen. 3) In suburban areas and rural areas, when the capacity increases, directional antennas with wide half power beam width (HPBW) angle and large gain value can be chosen to increase coverage. 4) In highways, where there is no need to cover towns along the road, or at border area or at the coast, 2 sector configuration is the optimal solution with two directional antennas with narrower width and higher gain antennas. 5) Three sector cells is the optimum solution to meet both capacity and coverage in all morphologies. 6) Dual polarization is usually used in dense urban (DU) and urban (UR) areas and space diversity is usually used in suburban (SU) rural (RU) areas.

2 - 15

Chapter 2: LTE Network Dimensioning 2.7.3 LTE optimization Optimization is probably the most important stage when planning LTE network. Typically it can be split into pre-launch optimization and post-launch optimization. There is however a number of different areas that may be optimized these including:    

Capacity Coverage Configuration and parameters Interference

Prelaunching optimization It is done when the sites are on air but not available to users. It is done via drive test to determine gaps and holes for coverage and to ensure optimal operation for the network and to verify coverage, capacity and quality requirements.

Figure (2-8) LTE optimization process

2 - 16

Chapter Three Coverage Dimensioning

Chapter 3: Coverage Dimensioning

Chapter Three Coverage Dimensioning 3.1 Introduction The link budget calculations estimate the maximum allowed signal Attenuation, called path loss, between the mobile and the base station antenna. The maximum path loss allows the maximum cell range to be estimated with a suitable propagation model, such as Okumura–Hata. The cell range gives the number of base station sites required to cover the target geographical area. The link budget calculation can also be used to compare the relative coverage of the different systems . Network dimensioning requires determination the number or cells (number of sites) to cover a certain region and to determine the radius of each cell and the spacing between them either using traffic or coverage criteria. So, in this chapter we will discuss the coverage analysis using the link budget and certain propagation model. This chapter presents the outline and basic concepts required to dimension coverage in the Long Term Evolution (LTE) network with functions in the current release. The method presented in this document consists of concepts and mathematical calculations that are elements of a general dimensioning process. The detailed order and flow of calculations depends on the required output of and type of input for the specific dimensioning task. The method provides a specific dimensioning process example. By changing the prescribed inputs and outputs and the order of calculations, the dimensioning process can be adapted to other methods.

3-2


Input requirements for the capacity and coverage dimensioning process consist of a bit rate at the cell edge, one for downlink and one for uplink. The required output is site-to-site distance and cell capacity in the uplink and downlink. The method is developed for Frequency Division Duplex (FDD), but can also be used for Time Division Duplex (TDD) .  Limitations Limitations to the calculation method include the following:  Multiple Inputs Multiple Outputs (MIMO) is considered only for the downlink for a maximum of two antennas  Outer loop power control in the uplink is not modelled  The method is adapted and developed primarily as a mobile broadband service that can handle Voice over IP (VoIP) to a limited extent  Quality of Service (QoS) is not handled by the method  Assumptions Calculations for coverage and capacity are based on the following assumptions:  All user equipment is assumed to have two receiving antennas  All resource blocks are transmitted at the same power, including user data, as well as control channels and control signals  The coverage for control channels and control signals equals that of user data at the same power.  Layer 1 overhead for all control channels and control signals is included in the Signal-to-Interference-and-Noise Ratio (SINR) to bit rate relationships. 3-3


Figure (3-1) LTE Dimensioning Process 3.2 Concepts and Terminology The following terms are used in describing capacity and coverage dimensioning: Average user bit rate The bit rate achievable by a single user. When all resources in a cell are used, the average user bit rate can be the average throughput in one cell. It is a measure of average potential in a cell while all interfering cells are loaded to the dimensioned level. Cell edge The geographical location where the path loss between eNodeB and the antenna is at a specific maximum threshold value, as calculated using the quality requirement imposed on the network, guaranteeing the required quality with a probability of 95%, for example. Cell throughput Cell throughput is obtained in one cell when all cells are loaded to the dimensioned level, and the resource use is equal to system load, 3-4


interfering cells as well as interfered cells. It is the average throughput per cell as calculated across the entire network. Coverage (area) The percentage of cell area that can be served according to a defined quality requirement. With an assumed uniform subscriber density (often assumed in a dimensioning exercise), the percentage of served area equals the percentage of served users. Resource block It is the smallest unit in the physical layer and occupies one OFDM or SC-FDMA symbol in the time domain and one subcarrier in the frequency domain. A two-dimensional unit in the time-frequency plane, Consisting of a group of 12 carriers, each with 15 kHz bandwidth, and one slot of 0.5 ms. System load The extent of available air interface resource usage. The system load equals the ratio of used resource blocks as an average over the entire system. 3.3 link Budget Definition Illustrative example: you are planning a vacation .You estimate that you will need 1000 L.E to pay for the hotels, restaurant, food etc.. You start your vacation and watch the money get spent at each stop. When you get home, you pat yourself on the back for a job well done because you still have 50 L.E left in your wallet. We do something similar with communication links, called creating "a link budget" The traveller is the signal and instead of money it starts out with ”power". 3-5


It spends its power (or attenuates, in engineering terminology) as it travels wired or wireless. So you can use a credit card along the way for extra money infusion, the signal can get "margin" extra power infusion along the way from intermediate amplifiers such as microwave repeaters foe telephone links or from satellite transponders for satellite links. The designer hopes that the signal will complete its trip with just enough power to be decoded at the receiver with the desired signal quality. In our example, we started our trip with 1000 LE because we wanted a budget vacation. But what if our goal was a first-class vacation with stays at five stars hotels, best shows and travel by A1000LE budget would not be enough and possibly we will need instead $5000. The quality of the trip desired determines how much money we need to take along. Link budget means to catalog all losses and gains between the two ends of communication i.e. mobile station (MS) and eNodeB to yield the maximum allowable (or available or acceptable) loss in signal strength that can be tolerated between the transmitter and receiver. Link budget traces power expenditures along path from transmitter to receiver to identify or determine the maximum allowable path loss and to determine the maximum feasible cell radius using propagation model. Link budget is defined sometimes as the difference between transmitter effective isotropic radiated power (EIRP) and the minimum signal strength at the receiver i.e. the receiver sensitivity for acceptable quality .Link budget is specified in logarithmic units (decibels) .Link budget output is fed to propagation model to provide the greatest spatial 3-6


distance

between

transmitter

and

receiver

at

which

reliable

communication of the desired quality can still take place. 3.4 Why we use Link Budget? link budget is necessary to determine or calculate the maximum allowable or available, or accepted path loss (MAPL) where communication is achieved reliably or that will provides adequate signal strength at the cell boundary for acceptable voice quality over 90% of the coverage area if it is flat or 75% if it is hilly .Link budget is necessary to determine the radius of the cells, and finally to determine the locations of cell sites as well as the spacing between them to ensure reliable and uninterrupted communication as mobile stations (MSs) move through the coverage area of interest. 3.5 What are the types of link budget? Since communication in mobile cellular phone system between mobile stations (MSs) and eNodeB is bidirectional. Thus it depends on the quality of the both reverse link and forward link. There are two link budgets:  Reverse link budget (up link budget) i.e. as signal is transmitted from mobile station (MS) and received by eNodeB.  Forward link budget (down link budget) i.e. as signal is transmitted from eNodeB and received by mobile station (MS). The reverse link budget has to be considered in system design first then forward link budget and finally link balance will be made. But since coverage is usually reverse link limited, we will focus on reverse link budget (up link budget).

3-7


3.6 Up link Budget (uplink coverage) Most mobile telephony systems are frequently limited by the uplink, so it is useful to start link budget calculations with the uplink coverage requirements. The calculations are performed according to the following stages:  User equipment (UE) effective isotropic radiated or transmitted power per physical resource block (PRB)  The uplink required bit rate per physical resource Block (PRB) (Rrequired, PRB)  The uplink required SINR (  ) given the uplink required bit rate physical resource Block (PRB) (Rrequired, PRB).  ENodeB receiver sensitivity (SeNodeB)  Uplink noise rise or interference margin (IM) (BIUL)  Log normal fading margin (BLNF)  Uplink link budget maximum allowable path loss (MAPLUL) 3.7 Up Link Budget Entries: The following set of definitions is to be read in conjunction with the appended reverse link budget (uplink budget) spread sheet. 3.7.1 Maximum mobile station (MS) transmitted power per traffic channel: It is the power coming out of the radio / amplifier and into the antenna power; the power value is 23 dBm at cellular frequencies .These values are taken from minimum performance standards for a 200 milliwatt mobile station.

3-8


3.7.2 Mobile station (MS) transmitter antenna gain (dBi) : An antenna is a device used to transmit or receive radio frequency (RF). The radio produces an RF signal and the antenna is the transport medium used to direct that signal onto free space for its eventual reception by another antenna attached to a receiver. One of the most important aspects of an antenna is the antenna gain. Mobile station (MS) antenna gain is the measure of strength of the amplification effect of MS antenna directed signal with respect to signal loss. It is the output transmitted power from the mobile station, in a particular direction, compared to that produced in any direction by a perfect reference antenna (isotropic antenna or dipole antenna). An antenna can create an amplification effect depending on its construction. The amplification effect is the result of focusing the transmission signal into a tight beam. Antenna gain works by the same principle. Signal loss simply describes a decrease in signal strength. Gain and loss are very important to antenna and radio performance because they directly affect signal quality and the signal transmission and reception capabilities. Antenna gain has a direct effect on the total power radiated from an antenna. The value of the power transmitted into an antenna will not leave the antenna at the same value. It will be increased by the amount of gain of the antenna. Antenna gain can be expressed in dBi, decibels relative to an idea isotropic radiator or dBd, decibels relative to an idea dipole effective area. A half –wave dipole antenna has an isotropic gain of 2.15 dBi. This means that the dipole, in the direction of maximum radiation, is 2.15dB more intense than that of an isotropic radiator, based on the same input power. 3-9


The value taken for MS antenna gain ranges between 0 dB this is the gain of the mobile station (MS) antenna. At both cellular and personal communication system (PCS) frequencies, this is a dipole whose gain can be taken to be 2.2 dBi. Mobile station (MS) antenna gain is defined either absolute gain or relative gain.  The absolute gain: it is the ratio of maximum radiation intensity in (watts) per unit solid angle to the total input power over 4 .  The relative power antenna gain It is the power gain of the antenna concerned in certain direction to the power gain of a reference antenna assuming the input power is the same for each antenna. The reference antenna may be isotropic antenna or dipole antenna. The isotropic antenna radiates equally in all directions in all planes like point of source. It is fictions or hypothetical antenna and is used as a reference. 3.7.3 Head /Body Losses (dB) Head /body loss refers to the attenuation of the radio signal during both transmission and reception as the mobile station antenna is held to the ear of the mobile station (MS). At personal communication system (PCS) and cellular frequencies, this attenuation is mainly due to the head of the user while at lower frequencies (large wave lengths) the entire human body could distort the radiation pattern of the mobile station antenna. Head/body losses are the amount of power that is absorbed through the head and body of the human being from MS. Typically head /body loss values range from 2 to 5 dB (3dB).Values to be used in the link budget are typically provided by the wireless network operator based on field measurements or prior experience. It is 3 - 10


important to obtain these values from the operator since any design loss raises design cell count.  It worth to mention that in fixed wireless local loop (FWLL), the head /body loss is zero while in mobile cellular mobile has a value. 3.7.4 Physical resource block: A resource element is the smallest unit in the physical layer and occupies one OFDM or OFDMA symbol in the time domain and one subcarrier in the frequency domain. A transmitted OFDMA signal can be carried by a number of parallel subcarriers. Each LTE subcarrier is 15 kHz. Twelve subcarriers (180 kHz) are grouped into a resource block. Depending on the carrier bandwidth, LTE supports a varying number of resource blocks. The downlink has an unused central subcarrier. The following illustration shows resource block definition:

Figure (3-2) Resource Block Definition in Frequency Domain. A resource block is limited in both the frequency and time domains . One resource block is 12 subcarriers during one slot (0.5 ms).

3 - 11


In the downlink, the time-frequency plane of OFDMA structure is used to its full potential. The scheduler can allocate resource blocks anywhere, even non-contiguously. A variant used in the uplink requires the scheduled bandwidth to be contiguous and a single carrier. The method, called SC-FDMA, can be considered a separate multiple access method. A user is scheduled every Transmit Time Interval (TTI) of 1 ms, indicating a minimum of two consecutive resource blocks in time at every scheduling instance. The minimum scheduling in the frequency dimension is the width of one resource block. The scheduler is free to schedule users both in the frequency and time domain. The illustration in shows user scheduling in the time and frequency domain for downlink and uplink:

Figure (3-3) Downlink and Uplink User Scheduling in Time and Frequency Domain.

3 - 12


The bit rate requirement should be based on the service for which the system is dimensioned, and as a compromise between conflicting needs and trends, with the following considerations: ' With a small nPRB the required bit rate can be satisfied with a minimum

of resources. This leaves a maximum amount of space in the timefrequency resource plane for other users to maximize capacity. ' At a large nPRB , the transmitted blocks are spread over a frequency

interval, with less power used per physical resource block. A lower modulation scheme and/or a higher coding rate can be selected. The receiver is capable of decoding the transmissions at lower SINR, to give a higher path loss leading to an increased cell range. Additionally, the user ' equipment can reduce maximum output power when using large nPRB

according to the 3GPP document user equipment (UE) radio transmission and reception. The back-off allowed is not assumed to be used at cell edge. The impact from noise rise on the resulting coverage range when ' varying nPRB in the dimensioning plays a comparatively minor role, unless

the noise rise is very high. All physical resource blocks must be consecutive in the uplink. ' Large nPRB may be less probable if the scheduler operates efficiently.

Using a few different values of for calculating the link budget can be helpful. 3.7.5 User equipment effective isotropic radiated power (EIRP) per physical resource Block (PRB) All allocated resource blocks share the total user equipment output power. 3 - 13


Assuming that all resource blocks are allocated an equal amount of power, the power per physical resource block (PRB )is calculated in the following way:

Equation (3-1) represents power of user equipment per physical resource block

Equation (3-2) represents Effective Isotropic Radiated Power of user equipment Where: GUE is the user equipment transmitting antenna gain [dBi] LHBL is the head body loss [dB ] Gother is the gain due to using MIMO. LHBL is head/ body loss [dB] EIRP means effective or equivalent isotropic radiated power. This refers to the effective isotropically radiated power from the mobile station (MS) at the antenna connector or it is the power radiated within a given geographical. It is the effective input power to hypothetically isotropic antenna that achieves the maximum radiated intensity in any direction. It is a function of the MS transmitted power and the MS transmitter antenna gain and head/body losses. 3.7.6 eNodeB receiver thermal noise density No This simply refers to the thermal noise floor at absolute temperature. No is eNodeB thermal noise density and given by: No = 10log KT 3 - 14


Where: K is Boltzman constant = 1.3806488 × 10-23 Watt/Hertz/Kelvin or joule /Kelvin (J/K) T is temperature in kelvin degree =290 degree Kelvin or degree Celsius (K) =273+17 degree centigrade No =10 log (1.3806488 × 10-23 *290) / 1 mW = -174 dBm/Hz Equation (3-3) represents eNodeB receiver thermal noise density 3.7.7 eNodeB receiver noise figure (NF) (dB) The eNodeB receiver noise figure (NF) is a measure of the signal to noise ratio (SNR) degradation when signal enters receiver till SNR i reach the input of demodulator by the eNodeB front end RF amplifier and filter Noise figure is given by: NF=10 log (SNRi / SNRo) Equation (3-4) eNodeB receiver noise figure. Where: SNR: It is the input signal to noise ratio. SNRo is the output signal to noise ratio. 3.7.8 The uplink required bit rate per physical resource Block (PRB) (Rbrequired, PRB,UL) Dimensioning starts by defining the quality requirement. Quality is expressed as a certain bit rate that can be provided to one individual user at the cell edge with a certain probability. The required bit rate follows the service for which the system is dimensioned. All calculations are performed per physical resource block. Table (3-1) shows how to obtain the required bit rate per physical resource block; the ' required bit rate is divided by the number of physical resource blocks nPRB

3 - 15


that can be allocated to obtain that bit rate. The required bit rate per resource block is given by:

Equation (3-5) represents the required bit rate per physical resource block Where: : Required bit rate : Physical resource block ' In a real system, nPRB is selected by the scheduler on a 1 ms Time

Transmission Interval (TTI) level. In a dimensioning exercise, the ' number nPRB can be selected freely, guided by experience and

understanding of the system within the constraints of total deployed bandwidth, as shown in table (3-1).

Table (3-1) bandwidths and number of physical resource blocks (PRB) specified in 3GPP 3.7.9 The uplink required SINR (  ) given the uplink required bit rate Rrequired, PRB

3 - 16


Similar to High Speed Packet Access (HSPA) in WCDMA, LTE includes a variety of different transport formats with different modulation and coding schemes. Each format has a specified bit rate. The SINR requirement for decoding a particular transport format has been determined by a large set of simulations. The simulation results in a set of tables for different channel models and for different antenna arrangements. As an approximation, the simulation results have been fitted to a semi-empirical parameterized expression. The expression for the dependency between Rrequired,PRB and the SINR  is expressed along with the semi-empirical constants a0, a1, a2 and a3. Using the required bit rate Rrequired,PRB , a SINR  is obtained that represents the requirement on signal quality. For the transport formats in LTE, given the required bit rate per resource block, RPRB, the signal-to-interference-and-noise ratio (SINR), γ, is determined by a set of link simulations. The uplink cases simulated include the following:  Antenna techniques: 2-branch RX diversity  Modulation schemes: QPSK, 16-QAM  Channel models and Doppler frequency EPA 5 Hz, EVA 70 Hz, ETU 300Hz 

Performance analysis of multipath propagation channels: The multipath propagation condition consists of several parts:  A delay profile in the form of a ―tapped delay-line‖, characterized by a number of tapes at fixed positions on a sampling grid. The profile can be further characterized by the r.m.s delay spread and the maximum delay spanned by the taps. 3 - 17


 A combination of channel model parameters that include the Delay profile and Doppler spectrum that characterized by a classical spectrum shape and a maximum Doppler frequency. Both delay profiles and Doppler spectrum for various E-UTRA channel models were considered. The delay profiles are selected to be representative of low, medium and high delay spread environments. The resulting model parameters. Here the Excess tap delay and Relative power were analysed and the mobile radio channels such as Extended Pedestrian A, Extended Vehicular A, Extended typical urban Model and HSTC model performance were compared using the Table (3-2). Model

No. of channels taps

Max. Delay

7

410 ns

Extended Vehicular A(EVA)

9

2510ns

Extended typical urban(ETU)

9

5000ns

Extended Pedestrian A (EPA)

Table (3-2) Channel models specifications EPA Excess tab delay(ns)

EVA Relative power

Excess tab

Relative power (dB)

(dB)

delay(ns)

0

0

0

0

30

-1

-30

-1.5

70

-2

150

-1.4

90

-3

310

-3.6

110

-8

370

-0.6

190

-17.2

710

-9.1

410

-20.8

1090

-7

Table (3-3) Channel model specifications

3 - 18


Extended urban and HSTC model ETU

HSTC model

Excess tap delay (ns)




0

-1

0

-1

50

-1

900

-21

120

-1

1900

-35

200

0

2200

-39

230

0

2700

-39.1

500

0

6100

-43

1600

-3

7100

-21.2

2300

-5

10100

-35

5000

-7

-

-

Table (3-4) channel propagation conditions Table (3-4) shows channel propagation conditions that are used for the performance measurements in multi-path fading environment for low, medium and high Doppler frequencies. In this paper, the combination of channel models that include the Delay profile and the Doppler spectrum are considered for the simulation [5]. Model

Maximum Doppler frequency

EPA

5 Hz

EVA

70 Hz

ETU

300 Hz

HSTC

1340 Hz

Table (3-5) Maximum Doppler frequency channel model Table (4&5) shows multi-path delay profiles that are used for the performance measurements in multi-path fading environment. The Excess tap delay functions can be expressed in terms of Doppler spectrum as mentioned below. S (f) ∝1/((1-(f/fd)2)0.5

Equation (3-6) represents Doppler spectrum 3 - 19


Where: S (f): Doppler spectrum, f: Frequency, fd: It is the Doppler frequency which proportional inversely with Doppler spectrum. 1- Extended Pedestrian A (EPA)

Extended Pedestrian A. A propagation channel model based on the International Telecommunication Union (ITU) Pedestrian A model, extended to a wider bandwidth of 20 MHz. The pedestrian channel model represents a UE speed of 3 km/h. It described by Tau: is a vector of path delays, each specified in nano seconds. Tau: [0 30 70 90 110 190 410]/109 PDB (Power Delayed Bus): is a vector of relative path powers, in dB PDB = [0 -1 -2 -3 -8 -17.2 -20.8] 2- Extended Vehicular A (EVA) • Extended Vehicular A. A propagation channel model based on the International Telecommunication Union (ITU) Vehicular A model, extended to a wider bandwidth of 20 MHz. • The vehicular channel model represents UE speeds of 30, 120 km/h and higher. Tau= [0 30 150 310 370 710 1090 1730 2510]/ (109). PDB= [0 -1.5 -1.4 -3.6 -0.6 -9.1 -7 -12 -16.9]. 3-Extended Terrestrial Urban (ETU) A propagation channel model based on the GSM Typical Urban model, extended to a wider bandwidth of 20 MHz It models a scattering environment which is considered to be typical in a urban area. 3 - 20


Tau=

[0

50

120

200

230

500

1600

2300

5000]/(10^9)

PDB=[-1 -1 -1 0 0 0 -3 -5 -7] 3.7.10 the required signal-to-interference-and-noise ratio SINR  t arget given the required bit rate RPRB The results, including an implementation margin, have been fitted to a semi-empirical parameterized expression for the required signal-tointerference-and-noise ratio SINR  t arget given the required bit rate RPRB is written as follows:

Equation (3-7) represents the required signal-to-interference-and-noise ratio SINR. The semi-empirical parameters for uplink a0, a1,a2 and a3 are given in tables (3.6)

Table (3- 6) semi-empirical parameters for uplink 3.7.11 eNodeB receiver sensitivity (SeNodeB) eNodeB receiver sensitivity SeNodeB is the required signal power at the system reference point when there is no interference contribution from other user equipments. The following relation describes eNodeB receiver sensitivity per physical resource block (PRB): 3 - 21


Equation (3-8) represents eNodeB receiver sensitivity Where Nt is the thermal noise power density and is equal -174 dBm/Hz NfeNodeB is the noise figure of the eNodeB receiver [dB] WPRB is the bandwidth per physical resource block (PRB): 180 kHz  t arg et,UL is SINR requirement for the uplink traffic channel [dB] N PRB,UL is the thermal noise per physical resource block in uplink is

given by:

Equation (3-9) represents the thermal noise per physical resource block in uplink The eNodeB receiver can be assumed to have a noise figure of 2 dB with tower mounted amplifier (TMA) and 3 dB without. 3.7.12 Up link noise rise or interference margin (IM) BIUL : In LTE a user does not interfere with other users in the cell since they are separated in the frequency/time domain. The noise rise in the uplink depends only on interference from adjacent cells. In the link budget, an interference margin compensates for noise rise. The standard case of closed loop power control is shown as a linear ratio. The uplink interference margin is given by:

Equation (3-10) represents Up link noise rise or interference margin. Where: 3 - 22


 t arg et UL is the SINR target for the uplink open loop power control.

QUL is average uplink system load.

CLF is defined as the ratio of actual capacity to pole point capacity. Pole point capacity is defined as the capacity when all user equipment raise their power to infinity this is a hypothetical situation which is taken as a reference. F is the average ratio of path gains for interfering cells to those of the serving cell. F is defined and investigated thoroughly for WCDMA radio network dimensioning. Table (3.7) gives values for F at varying electric tilt with 30 meter antenna height and 3-sector sites. The values are based on system simulations.

Table (3.7) examples of F for varying tilt 3.7.13 Log normal fading margin: Fading is defined as the random variation (change or fluctuation) of the received signal. There are different types of fading: large scale or slow fading and small scale or fast fading. Fading is described using probability density functions: large scale or slow fading is log normal distributed while small scale or fast fading which is Rayleigh or Rican distributed. Rayleigh distribution describes 3 - 23


the received signal is due to only reflection and there is no line of sight (LOS). Log normal distribution describes signal changes due to abstraction in the path between eNodeB and mobile station (MS). Fading margin is an extra margin is included in the link budget. The lognormal fade margin is calculated based on the coverage objective, which is typically specified as a target coverage probability at cell edge. Typical numbers are 90% and 75% edge coverage probability. Achieving 90% edge coverage implies that at 90% of the locations at edge, a cell can be initiated and kept up. Using path loss models, one can relate area coverage probability to edge coverage probability and hence to fade margin requirement. 95% area coverage probability is mapped to 75% edge coverage. These values presume a completely noise limited receiver. The lognormal (or slow fading) margin models the required area coverage probability. By adding this margin, a probability is secured for setting up and maintaining a connection at a given quality. Table (3.8) shows fading margins in dB for varying standard deviation σ of the lognormal fading process and different coverage probabilities:

Table (3.8) log normal fading margins for varying standard deviation of lognormal Fading 3 - 24


Equation (3-11) represents Log normal fading margin Where: 

is the mean of log normal.



is the standard deviation of log normal.

P% is the coverage probability The standard components are given for link analysis in the radio interface. The standard margins for indoor, car penetration loss, body loss, feeder loss, jumper loss, and antenna gain are the same as any mobile network. A fading margin is required to guarantee a certain coverage probability. MAPLUL represents the maximum allowable path loss in uplink link budget, fed into the downlink link budget. 3.7.14 eNodeB receiver cable feeder, jumper and connector losses Feeder cable loss Feeder cable loss is the loss of electrical energy due to the inherent characteristics of the feeder cable. The eNodeB receiver feeder cable is dependent on the feeder type and length of feeder run. The receiver cable and connector losses are nominally taken in the range of 2 dB to 4 dB. When the cable length and diameter (and hence attenuation/feet) are known, the actual cable losses may be substituted in the link budget along with additional margin of 0.5 dB for connector (and duplexer) losses. Radio equipment should be placed as close as possible to the antennas in order to reduce the feeder cable loss. Typically feeder cable diameters used are 7/8" and 15/8" and corresponding attenuations are 6.15 dB/100 meters and 3.84 dB/meters. Jumper loss "Lj" 3 - 25


Jumper loss is the loss of electrical energy due to the connection of the tower top amplifier with the antenna using jumpers. A typical value of the jumper is 11.2 dB/100m.When the used jumper type and length is known, the total jumper loss can be calculated. Connector loss "Lc" It is the loss of electrical energy because of connectors that make the antennas tied with the top of the tower. A typical value of the connector loss is 1 dB. 3.7.15 Building / vehicle penetration loss: This refers to the attenuation of the signal as it passes through one or more walls of the building in the desired coverage area. When a mobile station (MS) is used inside the building and the eNodeB is situated outside, there is a loss when the signal penetrates the building. It is defined as the difference between the average signal strength outside the buildings and the average signal strength over the ground floor of the building. The value of penetration loss must be included when designing link budget. Table (3.9) shows the value of penetration loss on different morphology classes In building dense

In building

urban

suburban

20

18

In building rural

In car

12

9

Table (3-9) values of penetration loss on different morphology classes When the MS is situated in a car without external antenna, an extra margin has to be added to cope with the penetration loss of the car. This extra margin is typically 9 dB.

3 - 26


3.7.16 eNodeB receiver antenna gain (dBi) This refers to the gain of the receiving antenna at eNobeB .While the actual antennas used in the network may vary from site to site, a nominal, representative value is provided in the link budget based on the frequency of operation and sectorization. The nominal antenna gain values for personal communication systems (PCS) and cellular frequencies differ based on the cell Omni or sectorized .The gain units are dBi or gain with respect to an isotropic radiator. The value of antenna gain also can be varied depending to the manufacturer. Typically a value of eNondeB receiver antenna gain is typically 12 dBi for omni cell and 18 dBi for sectorized cell 3.7.17 Uplink link budget maximum allowable path loss (MAPLUL) Finally, the uplink link budget maximum allowable path loss (MAPLUL) Can be calculated as follows:

Equation (3-12) represents Uplink link budget maximum allowable path loss. Where: (MAPLUL) is the maximum allowable path loss due to propagation in the air [dB] BLNF

is the log-normal fading margin [dB]

BIUL

is the uplink interference margin [dB]

LCPL

is the car penetration loss [dB]

LBPL

is the building penetration loss [dB]

GeNodeB is the eNodeB receiver antenna gain [dBi] Gother

is the other gain [dBi]

Lf

is eNode B feeder loss [ dB ] 3 - 27


Lj

is the Jumper loss [dB]

LC

is connector loss [ dB ]

3.8 Morphologies classifications Dense urban (DU): Central business districts with skyscrapers or with buildings with having 10 to 20 stories and above, the building separation (S) less than 10 meters. Clutter height higher than 30 meters. Urban (UR): Residential , office area, hotels, hospitals etc with buildings having 5 to 10 stories and street width less than 5 meters and building separation (S) less than 10 meters. Clutter height higher from 15 to 30 meters. Suburban(SU): Mix of residential and business communications with 2 to 5 stories shops and offices. The building separation is (S) less than 20 meters. Villages or high ways scattered with trees and houses, some obstacles near the MS but not very congested. Rural area: Parks or fields with small trees with height less than 12 meters and 20% house density of residential area of

2 stories with wide roads, The

building separation is (S) less than 20 meters. Clutter height higher than 3o meters. Open areas: Clutter height higher than 3 meters open areas, parks, fields, paved areas. Morphology

Clutter height

class

(meters)

Dense urban

H>30

Building separation

Morphology definition

(meters) S<10

3 - 28

Building height more than 10 stories


Urban

15
S<10

Suburban

10
S<10

Building height between 5 to 10 stories and street width <5 meters Residential or office areas of 3-4 stories Residential areas of 2 stories with wide roads,

Rural

H<10

S<20

parks or fields with small trees<12meters and 20% house density

Open

H<3

S<20

Open areas, parks, fields, paved areas

Table (3-10) summarizes the features of different morphologies. 3.9 Downlink Budget The downlink link budget is calculated for the following purposes: • To determine the limiting link • To determine the bit rate that can be supported in the downlink at the uplink cell range limit. The calculations are performed according to the following steps: • Path loss from uplink • Bit rate requirement • Power per resource block • Downlink noise rise (interference margin) • Downlink link budget • Receiver sensitivity, UE • Bit rate at the cell edge • Concluding the link budge 3.9.1 Path loss from Uplink (MAPLUL) from the uplink link budget calculations is the starting point of the downlink calculations and is used to obtain a downlink noise rise estimate. At the end of the link budget calculation process, if the downlink (MAPLUL) is less than the uplink (MAPLUL), both the uplink 3 - 29


and downlink link budgets can be recalculated (including the noise rise) using the new (MAPLUL). 3.9.2 Bit Rate Requirement If the bit rate requirement is not expressed per resource block, it is divided by (Rreq) to obtain (Rreq). As with the uplink, the bit rate requirement is expressed per resource block in the calculations. However, unlike the uplink, the downlink scheduler can allocate resource blocks across the entire deployed bandwidth without requiring them to be consecutive. It can be shown that it is always favourable to spread the transmission across as many resource blocks as possible. Assuming this, the number of allocated 3.9.3 The down link required bit rate (Rb required, PRB,DL ) per physical resource block (PRB) If the down link bit rate requirement Rb, required,DL is not expressed per physical resource block (PRB), it is divided by

n PRB

to obtain

Rbrequired,PRB,DL . As with the uplink, the bit rate requirement is expressed per physical resource block in the calculations. However, unlike the uplink, the downlink scheduler can allocate physical resource blocks across the entire deployed bandwidth without requiring them to be consecutive. It can be shown that it is always favourable to spread the transmission across as many physical resource blocks (PRB) as possible. Assuming this, the number of allocated physical resource blocks n PRB in the downlink for dimensioning is set to the total number of physical resource blocks for the deployed bandwidth. In this process, the obtained bit rate requirement per physical resource block is not used directly to calculate power per physical resource block, 3 - 30


but to compare with the rate that can be obtained at the cell edge given by the uplink link budget. Alternatively, it can be used as a starting point for link budget calculations. 3.9.4 eNodeB radiated or transmitted power (EIRP) per physical resource block (PRB) The power in LTE is shared by all physical resource blocks. It is assumed that all physical resource blocks are allocated an equal amount of power. An individual physical resource block has no power control. Instead, users are scheduled with high rates every millisecond. The e Node B transmitted or radiated power per physical resource block is:

Equation (3-13) represents eNodeB transmitted power (EIRP) per physical resource block (PRB). Where: Pnorm,ref : is the sum of nominal power from all radio units in the cell at the reference point [W]. This means if MIMO is used with two radio units of 20 W each, is equal to 40W.It is expected that 20 W, 40 W and 60 W power classes will be common. The nominal power at the reference point can be reduced by loss in feeders. nRB

: is physical resource block.

3.9.5 Thermal noise per physical resource block in the downlink (N PRB,DL ) N PRB,DL is the thermal noise per physical resource block in the downlink, defined as follows:

3 - 31


Equation (3-13) represents Thermal noise per physical resource block in the downlink NfUE The assumed noise figure Nf for typical user equipment (UE) receiver is 7 dB. 3.9. 6 the down link noise rise or interference margin (IM) (BIDL) The down link noise rise on the cell edge is needed for the link budget and is calculated using the following expression where all quantities linear:

Equation (3-14) represents the down link noise rise or interference margin Where: Q DL : is the downlink system load.

FC : is the average ratio between the received power from other cells to that of own cell at cell edge locations. The load is modeled with QDL. The link budget must be true for a network with a given load. Normally, one design input is to determine the load for which the coverage is available. The cell plan quality is modeled with the factor FC . FC describes the ratio of received power from all other cells to that received from own cell at a location near the cell edge. Table (3-11) gives values at varying electric tilt with 30 meter antenna height, and 3-sector sites. The values are based on system simulations. For dimensioning other antenna heights, see appendix (A) 3 - 32


Table (3-11) examples of Fc at cell edge for varying tilt 3.9.7 The calculated down link SINR on the cell edge The downlink calculated SINR  on the edge of a cell with the size given by MAPLUL is given by the following equation:

Equation (3-15) represents down link SINR on the cell edge. 3.9.8 The down link calculated bit rate Rcalculated at cell edge The cell edge down link SINR estimate or calculated is transformed into a calculated bit rate per physical resource block, R bcalculated,PRB by the same type of semi-empirical relationship as for the uplink SINR requirement . For the downlink, the semi-empirical constants or parameters a0, a1,a2 and a3 are given in table (3-12) .

Table (3-12) Semi-empirical parameters for down link 3 - 33


The downlink cases simulated include the following:  Antenna techniques: SIMO 1x2, TX diversity 2x2, Open loop Spatial Multiplexing (OLSM) 2x2  Modulation schemes: QPSK, 16-QAM, 64-QAM  Channel models and Doppler frequency: extended pedestrian model A (EPA) 5 Hz, extended vehicular model A (EVA) 70 Hz, extended terrestrial urban model (ETU) 300 Hz  Number of OFDM symbols used for PDCCHs: 1 The uplink cases simulated include the following: • Antenna techniques: 2-branch RX diversity • Modulation schemes: QPSK, 16-QAM • Channel models and Doppler frequency EPA 5 Hz, EVA 70 Hz, ETU 300Hz The results, including an implementation margin, have been fitted to a semi-empirical parameterized expression for bit rate RPRB as follows:

Equation (3-16) represents the down link calculated bit rate Rcalculated at cell edge. Where: a0, a1,a2 and a3 are fitted parameters and the SINR  is expressed in dB. a0, a1,a2 and a3 are fitted parameters and the SINR  is expressed in dB. 3.9.9 Concluding link budget according to required and calculated bit rate The resulting or calculated bit rate is multiplied by the number of physical resource blocks (PRB) (nPRB) to obtain the maximum calculated bit rate (Rcalculated) expected on the cell edge. If the uplink is really the 3 - 34


limiting link, (Rcalculated) should be larger than the required bit rate (Rrequired). If the resulting or calculated bit rate (Rcalculated) is lower than the required bit rate (Rrequired), then the downlink is the limiting link 3.10 Downlink Limited Link Budget If the resulting or calculated bit rate (Rcalculated) is lower than the required bit rate (Rrequired), then the downlink is the limiting link. In that case, the true maximum cell range must be determined by back tracking the downlink link budget calculations. The downlink link budget calculations are performed according to the following steps: (1) R PRB, required is transformed into a required SINR  . (2)The required SINR is used to derive user equipment (UE) sensitivity (SUE) at the cell edge. (3)The user equipment (UE) sensitivity (SUE) is used in the link budget, initially with the same noise rise BIDL as before. 3.10.1 User equipment (UE) receiver sensitivity The user equipment sensitivity SUE is given by:

Equation (3-17) represents User equipment (UE) receiver sensitivity 3.11.2 Downlink budget maximum allowable path loss (MAPLDL) The down link budget maximum allowable path loss (MAPLDL )is described by the following equation:

3 - 35


Equation (3-18) represents Downlink budget maximum allowable path loss. Where: EIRPeNodeB,PRB is the effective isotropic radiated or transmitter power

per physical resource block at the system reference point [dBm] SUE

is the user equipment (UE) sensitivity [dBm]

(4)New signal attenuation for down link is derived with the following equation:

Equation (3-19) represents signal attenuation for down link. (5)The new down link signal attenuation L sa,max,DL is applied in to obtain a new BIDL (6)Equation in the above is iterated until L sa,max,DL and BIDL are constant. (7)The new L sa,max,DL converted to MAPLDL is now used to calculate the true cell range. MAPLDL is used as a measure of cell size. It is converted to geographical distance by a suitable wave propagation model. A down link limited system means that the uplink quality exceeds the requirement. If the bit rate on the cell edge for the uplink is needed, the uplink budget calculations also must be back tracked: (1)L sa,max,DL from the downlink is applied in to obtain the new down link MAPLDL and a new uplink noise rise (BIUL) is approximated with the following expression:

Equation (3-20) represents uplink noise rise. 3 - 36


Where: '' nPRB is the number of resource blocks allocated to the service responsible

for the interference. n’’PRB may or may not be equal to n’PRB, the number of resource blocks allocated to the service for which the link budget is calculated. H is the average attenuation factor, depends on the site geometry, antenna pattern, wave propagation exponent, and eNodeB antenna height. H is the standard average path loss factor used in coverage and capacity dimensioning value of 0.36 is recommended for dimensioning. (2)the equation of the uplink budget maximum allowable path loss (MAPLUL), is solved for the eNodeB sensitivity SeNodeB , and the downlink Lsa,max,DL is inserted.

(3)the equation of receiver sensitivity is solved for the uplink signal-tointerference-and- noise ratio (SINR) at the cell edge  .  is converted to a logarithmic value. (4)The corresponding calculated bit rate is Rb,calculteddetermined. 3.11 propagation models To make a design and plan of cellular mobile phone systems, accurate propagation characteristics of the environment should be known especially the path loss. The calculation of path loss is vital for the determination of RF cell coverage of eNodeB placement and in optimizing it. There are many prediction models that are used to predict path loss. Although these models differ in their methodologies, all have the distance between the transmitter and receiver as a parameter i.e. the path loss is heavily dependent on the distance between the transmitter and receiver. Other effects also come into play in addition to distance. In the 3 - 37


following subsections, the propagation model will be defined and why it is necessary. The different types of propagation predict models for terrestrial wireless communication systems will be presented briefly, and then an example of each type will be discussed in detail. The focus is placed on the following models: free space model, Cost 231 Okumara Hata model and Cost 231 Walfisch Ikegami model. The last two models are the most widely used software package for cellular system design. Definition of propagation model: Propagation model is a model used to determine the maximum range of the communication system which provides acceptable quality provided that the maximum allowable or permissible or accepted path loss (MAPL) is determined as accurately as possible via link budget. In cellular mobile phone system propagation model is used to calculate the maximum distance between the mobile station (MS) and the eNodeB at which reliable communication take place with the desired quality of service. and to determine the locations of cell site (CSs) and the spacing between the CSs in order to ensure reliable and uninterrupted communications as the MS moves through the required coverage area. The propagation models are necessary and essential because the various propagation effects and time varying, dynamic and difficult to predict. The signal traveling from the eNodeB to the mobile station follows many different paths before arriving at the receiving antenna of the MS. Each individual path affects the signal causing attenuation, delay and phase shift. In additional, the motion of the mobile station (MS) nearby scatters such as trucks and buses may cause Doppler frequency shifts in each received component. The received signal at the mobile station (MS) is therefore a result of direct rays, reflected rays and shadowing or any combinations of these 3 - 38


signals. The path loss can be obtained either by field measurements are time consuming and expensive while the models are simple and efficient to use. 3.12Classifications of propagation models Propagation models can be roughly divided into three types: the empirical, theoretical and semi-empirical models. 3.12.1Empirical propagation models Empirical models are usually set of equations, the model parameters are divided from extensive field measurements data. They are accurate for environments with the same characteristics as those where measurements were made. The input parameters for empirical models are usually qualitative and not very specific e.g. dense urban (DU), urban (UR), Suburban (SU) and rural (RU) areas and so on. One of the main drawbacks of empirical models is that they cannot be used for different environment without modifications. The output parameters are basically range specific. Empirical models examples are Okumara model and Hata model. 3.12.2 Theoretical propagation models They are derived physically assuming some ideal conditions for example over roof top diffractions model is derived using physical optics assuming uniform heights and spacing of buildings. Theoretical models examples are Walfisch and Bertoni model, Ikegami model and free space model. 3.12.3 Semi – empirical propagation models The parameters of the theoretical models are empirically to fit measurement data. Semi –empirical models examples are COST 231 – Walfisch Ikegami model and COST 231 Okumara Hata model. 3 - 39


Free space model The free space model is physical model because it describes how signal propagates. The free-space model is based on expanding spherical wave front as the signal radiates from a point source in space. The electromagnetic waves in free space diminish as a function of inverse square of the distance i.e. (1/d^2), where d is the distance between the transmitter and receiver and in our case the distance between the mobile station (MS) and eNodeB. It is mostly used in satellite communication systems where the signal travels through free space. Assume that the MS antenna and eNodeB antenna are arranged such that their directions of maximum gain are aligned i.e. the source and load impedances match the antenna impedances their polarization are matched and they are separated by a distance. Okumara model In 1968, Okumara model is an empirical developed by Yoshihisa Okumara based upon an extensive series of measurements of the field strength made in and around Tokyo city by Y. Okumara in VHF and UHF land mobile radio services at several frequencies in 100 MHz and 3 MHz. Okumara model is a graphics- based model using numerous of curves. Okumara model is applied for prediction of maximum allowable path loss over macro cell, built up areas. It is also successfully applied in other urban environment (outside Japan) taking urbanization factor, terrain type correction into account. Okumara model was limited from 1 Km to 100 Km distance. The frequencies range from 1 m to 10 meters. Okumara model’s drawback is the results are available in graghical form. 3 - 40


Okumara – Hata model In 1980, Hata model is an empirical formula derived from Okumara’s results. The measurements graphs results have been fitted to a mathematical model by M.Hata. The Okumara graphs have been approximated by Hata in a set of formulas. The Hata model is a formula- based for Okumara model and can be used more effectively. Okumara –Hata model is applied for prediction of maximum allowable path loss over macro cell, buit – up, quasi smooth areas but the equations were limited from 1 Km to 20 Km distance. The frequencies range from 150 to 1500 MHz. The mobile station antenna height should be between 1m to 10meters. The eNodeB antenna height ranges 30 to 200meters. Okumara –Hata model is easily computable. COST 231 Okumara Hata model The Cost 231 Okumara –Hata propagation model was and still is widely used for coverage calculation in microcellular network planning. In 1999, it was found by the European community collaborative studied in the areas of science and technology (COST) that Okumara Hata model underestimates path loss. Okumara Hata model for medium to small cities i.e. urban area has been extended and modified to correct the situation and to cover the frequency band from 1500 to 2000 MHz in the COST 231 project. Thus, COST 231 Okumara Hata model is considered semi empirical model after adjustment to cover the frequency band of 4G cellular systems for urban personal communication system (PCS) applications. The model include terrain information qualitatively by dividing the prediction area into a series of clutter and terrain categories namely dense urban, suburban and rural, open, quasi open… etc 3 - 41


environments. Okumara Hata model with related corrections is the most common model used in designing real systems. Okumara takes urban area as a reference and apply correction factors for conversion to the other classifications. In Okumara Hata model, the path loss is function of several parameters such as frequency, frequency range, height of MS antenna, height of eNodeB antenna, and building density. This model has been proven to be accurate and is used by computer simulation planning tools. For the parameters, there are only certain ranges in which the model is valid; that hb should only be between 30m to 200m, hm should be between 1m to 10m, d should be between 1 Km to 20Km 3.13 Ericsson variant of COST 231 Okumura–Hata Wave Propagation model This section describes the wave propagation characteristics. It is not expected that a channel wider than 5 MHz will have a significant difference in the ability to compensate for Rayleigh fading. The equation to calculate the cell radius R in kilometres is as follows:

R  10  Equation (3-21) represents cell radius Where:

A: is frequency-dependent fixed attenuation value, shown in table (3.14) hb: is base station or eNodeB antenna height [m] hm: is height of the user equipment (UE) antenna [m] a (hm) = (1.1 log F- 0.7) hm – (1.56 log F- 0.8) 3 - 42


Equation (3-22) represents a function of user equipment antenna in RU, UR, and SU a(hm): is the Mobile station Antenna height correction factor as described in the Hata Model for Urban Areas. a(hm)=3.2[log(11.75hm)]2 - 4.97 Equation (3-23) represents a function of user equipment antenna in DU areas Equation (3-24) represents maximum allowable pathloss as a function of cell radius.

Table (3-13) fixed attenuation A in Ericsson variant COST 231 OkumuraHata propagation model

3 - 43


3.14 Example of coverage dimensioning (Radio Link budget- Uplink) Coverage (UL)---- RL budget 1-UE parameters Item Unit P(UE) dBm P(UE) Watt Channel B.W MHz nPRB PRBs P(UE.PRB) Watt P(UE,PRB) dBm UE Tx gain dBi other UE gain dBi L(HBL) dB EIRP (UE,PRB) dBm 2-BS parameter Item Unit Boltzman constant J/K Ambient temperature K Thermal noise power density (Nt) dBm/Hz Noise figure of eNodeB (Nf) dB BW per resource block [W(PRB)] MHz a0 kbps a1 dB a2 dB a3 kbps Bit rate [R(PRB)] kbps SINR dB Sensitivity of e.NoodeB Item Gain of e.NodeB Other gain of e.NodeB Total Gain Building Pentration Loss [L(BPL)] Car Penteration Loss [L(CPL)] BS Feeder Specification e.NodeB Feeder Length e.NodeB Feeder Loss (Lf) e.NodeB Jumper Specification e.NodeB Jumper Length e.NodeB Jumper Loss (Lj)

dB 4-other e.NodeB parameter Unit Values dBi dBi dBi dBi

dBi

(dB/100m) meter dB dB/100m meter dB

3 - 44

Values 23 199.5262315 20 100 1.995262315 33 0 2 3 32 Values 1.38E-23 290 -174 2 0.18 536.6 20.76 13.28 0 64 -2.499734217 -181.9470092

18 4 22 15

9

3 30 1.05 2 5 0.1


e.NodeB Connector Loss (Lc) Total Loss Item

Mean of Log normal (µ) Standard Deviation (σ)

Area of Coverage edge Coverage Prob. F(p) Lognormal Fading Margin [B(LNF)] Cell Loading Factor [Q(UL)] F B(IUL) Fast Fading Margin [B(FFM)] Total Margin Item MAPL Item operating frequency (f) free space loss (A) e.NodeB antenna height (hb) MS antenna height (hm) parameters a(hm) Cell radius

dB dB 5-Margins Unit Values

1 26.15

---------

0

dB

3 Flat Area 90 3.844654697 0.64 0.7 1.26066082 2 7.105315517

% dB ----------------dB dB dB 6-Max allowable path loss Unit dBm Cell area Unit Values MHz dB meter 10 meter

Km

96.848651

Values 202.6916936

2600 151.1 20 30 5 RU 10.75499499 161.47407 223.61022

Table (3-14) Example of coverage dimensioning (Radio Link budget)

3 - 45

40

285.46807

Chapter Four Capacity Dimensioning

Chapter 4: Capacity Dimensioning

Chapter four Capacity Dimensioning 4.1 Introduction Capacity dimensioning obtains input information to the phases after radio interface dimensioning: transmission link dimensioning and eNodeB dimensioning. The method is specified for a certain system load. The dimensioning method finds the maximum capacity that the target cell can sustain momentarily, given the system load in the surrounding cells. It is improbable that all cells in a system are fully loaded at the same time, as observed in real networks of different technologies. The evaluation of capacity needs the following two tasks to be completed: Being able to estimate the cell throughput corresponding to the settings used to derive the cell radius Analyzing the traffic inputs provided by the operator to derive the traffic demand, which include the number of subscribers (U), the traffic mix and data about the geographical spread of subscribers in the deployment area The target of capacity planning exercise is to get an estimate of the site count based on the capacity requirements. Capacity requirements are set forth by the network operators based on their predicted traffic. Average cell throughput is needed to calculate the capacity-based site count. In LTE, the main indicator of capacity is SINR distribution in the cell. In this project, for the sake of simplicity, LTE access network is assumed to be limited in capacity by DL.

4-2


The purpose of this chapter is to describe the capacity dimensioning for the LTE network and to explain the methods used and factors impacting the capacity dimensioning process. This chapter includes several sections. The first section describes the cell throughput calculations, while the second part is about traffic demand estimation. Later sections concern with capacity based site count evaluation. Capacity Definition The number of connections that the wireless channel can support without unduly degrading the data services carried on the channel. 4.2 Uplink Capacity 4.2.1 IT is based on the following calculations:  Signal-to-Interference-and-Noise Ratio (SINR)  Cell throughput  Number of sites required 4.2.2Signal-to-interference-and-noise ratio The operating mode with power control assumes perfect power control (PPC) and infinite power dynamics. User equipment is received at the signal to interference plus noise ratio (SINR) identical to the bit rate per physical resource block (PRB) is identical to the bit rate corresponding to the SINR and the number of allocated physical resource blocks. By varying the load QUL , the average user throughput does not change However, the cell throughput and the cell range will change The most accurate evaluation of cell capacity (throughput under certain constraints) is given by running simulations. The best solution to derive cell throughput is direct mapping of SINR distribution obtained from a

4-3


simulator into MCS (thus, bit rate) or directly into throughput using appropriate link level results. Capacity dimensioning gives an estimate of the resources needed for supporting a specified offered traffic with a certain level of QoS (e.g. throughput or blocking probability). Theoretical capacity of the network is limited by the number of eNodeB’s installed in the network. Cell capacity in LTE is impacted by several factors, which includes interference level, packet scheduler implementation and supported modulation and coding schemes (MCSs). The SINR values to support each modulation coding scheme (MCS) are derived from look-up tables that are generated from link level simulations. As shown in table (4-1) for urban channel model and a fixed inter-site distance of 1732m in LTE network.

Table (4-1) SINR values corresponding to each modulation coding scheme (MCS)

4-4


The average signal-to-interference-and-noise ratio (SINR) yields a bit rate. The result is the bit rate per physical resource block. The average user bit rate is scaled proportionately with the number of Physical resource blocks corresponding to the deployed bandwidth. For the transport formats in LTE, the relationship between bit rate per resource block (RRB) and Signal-to-Interference-and-Noise Ratio (SINR), γ, is determined by a set of link simulations. The uplink simulations include the following:  Antenna configuration: 2-branch RX diversity  Modulation schemes: QPSK, 16-QAM  Channel models: EPA 5 Hz, EVA 70 Hz, ETU 300Hz The results, including an implementation margin, have been fitted to a semi-empirical parameterized expression as follows:

Equation (4-1) represents the required bit rate Where: a0, a1,a2 and a3 are fitted parameters and the SINR is expressed in dB. The semi-empirical parameter a0 represents the maximum obtainable bit rate in one resource block as shown in table (4-2)

Table (4-2) semi- empirical parameters for up link

4-5


In the uplink, one or more resource blocks are always allocated at each band edge to signalling for users in idle mode on the channel Physical Uplink Control Channel (PUCCH). For this reason, the number of physical resource blocks in uplink available for calculating capacity are always reduced by a number

value of 4 is recommended for

dimensioning. The resulting Uplink average user bit rate per cell is: Ravg,UL =RRB,UL (nRB - nPUCCH) Equation (4-2) represents Uplink average user bit rate per cell Average cell Throughput The Uplink average cell throughput is by the following equation: Tcell,UL = QUL Ravg,UL Equation (4-3) represents the uplink average throughput Where:  The nRB is different and larger than the number of resource blocks nRB used for uplink coverage dimensioning  QUL: is the uplink system load  Ravg,UL: Average UP Link data rate The site throughput: Where the site capacity is a multiple of the cell throughput, which depends on the number of cells per site (Not considering any hardware limitation) According to cell type. If omni cell then the site throughput is given by:  Tsite = Tcell

Equation (4-4)

If 3 sector cell, then the site throughput is given by:  Tsite = 3 × Tcell

Equation (4-5)

4-6


The total throughput or the overall data rate To determine the traffic demand estimation, the total throughput or the overall data rate is given by:  Ttotal = U × TU Equation (4-6) represents the total throughput or the overall data rate Where: U is the number of users in the network TU is the throughput per user or peak data rate The number of sites required Nsite = Equation (4-7) represents the number of sites required 4.3 Downlink Capacity The following downlink capacity calculations are performed:  Signal-to-Interference-and-Noise Ratio (SINR)  Cell throughput  The number of sites required. 4.3.1 The maximum signal attenuation Lsa max at the cell border The maximum allowable path loss from the uplink is used to find the maximum sustainable bit rate per physical resource block in the downlink. Lsa max is given by: Lsa,Max = MAPL + BLNF – (Gue + Gothers) + LPBL + LCPL+ Lf + Lc Equation (4-8) represents The maximum signal attenuation Lsa at the cell border

4-7

max


Where:  MAPLUL is the

uplink budget

maximum allowable path loss

(MAPL) from coverage calculation.  BLNF : log-normal fading margin  GUE

:

User equipment transmitting antenna gain [dBi]

Gothers : It is gains due to MIMO  LBPL : Building penetration loss  LCPL : Car penetration loss  LF : Feeder loss  LC : Connector loss 4.3.2 Thermal noise power density per physical resource block in downlink N PRB,DL = Nt + Nf + 10 Log (WPRB) Equation (4-9) represents Thermal noise power density per physical resource block in downlink. Where:  Nt: It is thermal noise power density = 10 log10 KT and is equal -174 dBm/Hz  Nf : noise figure of receiver = 7 d B  W(PRB) : bandwidth per physical resource block = 180 KHz  NPRB,DL : Down Link thermal noise per physical resource block  K : Boltzmann's constant and its value is 1.38* 10-23  T : Temperature and its value is 290 degree Kelvin

4-8


4.3.3 The eNodeB transmitted power per physical resource block

Equation (4-10) represents The eNodeB transmitted power per physical resource block Where:  Ptx

,eNodeB, PRB

is the eNode B transmitted power per physical

resource block at the system reference point  P (norm) : is the sum of normal power from all radio units in the cell at the reference point The average down link noise rise or interference margins is

Equation (4-11) represents the average down link noise rise or interference margins

Where: • BIDL : Interference margin (IM) • Ptx,eNodeB,PRB : is the eNodeB transmitted or radiated power per physical resource block. • QDL is the average downlink system load. • Fc is interference factor. • N PRB,DL :Noise thermal power density, down link , per PRB.

4-9


• nPRB is number of physical resource block. •

F is the cell plan quality factor. It describes the ratio of received power from all other cells to that received from own cell at a location near the cell edge locations.

• B(IDL) : Down Link interference Margin 4.3.4 Signal-to-Interference-and-Noise Ratio The downlink capacity is based on the Signal-to-Interference-andNoise Ratio (SINR) at the average location within a cell, denoted as a linear ratio. The average SINR is expressed in the average noise rise. This is similar to the interference margin, but the SINR is evaluated at an average location instead of at the cell edge. The resulting average downlink signal-to-interference-and-noise ratio (SINR), is given by the following equation:

Equation (4-12) represents Signal-to-Interference-and-Noise Ratio in downlink Where: • H is the average attenuation factor dependent on site geometry, antenna pattern, wave propagation exponent,and base station antenna height. • H is the standard average path loss factor used in coverage and capacity dimensioning And for dimensioning H is value of 0.36

4 - 10


• The down link bit rate per physical resource block • The average signal-to-interference-and-noise ratio (converted to logarithmic) • yields an average bit rate by way of and is the bit rate per physical resource block, • The bit rate per physical resource block, down link RPRB,DL Given SINR

Equation (4-13) represents the down link bit rate per physical resource block For the downlink, the semi-empirical parameters are given in table (4-3)

Table (4-3) Semi- empirical parameters for downlink 4.3.5The down link cell throughput The average down link user bit rate per cell is scaled proportionately with the number of physical resource blocks and is given by: Ravg,DL = RRB,DL (nRB,DL- nPDCCH)

4 - 11


Equation (4-14) represents the down link cell throughput 4.3.6 The down link cell throughput is given by: Tcell,DL = QDl × Ravg,DL Equation (4-15) represents the down link cell throughput Where: • Raverage,DL is average DL data rate • R PRB,DL is DL data rate per physical resource block. • n'PRB • nPDCCH

is the number of physical resource block is the number of Physical DL control channels

• QDL is the average down link system load 4.3.7 The total throughput Ttotal = U × TU

Equation (4-16)

Where:  Ttotal: The total throughput  Tsite: The site throughput 4.3.8 The site throughput  Tsite = Tcell 

(Omni cell )

 Tsite = 3 × Tcell ( 3 sector cell )

Equation (4-17) Equation (4-18)

Where: Tcell,DL is the DL cell throughput The number of sites required  Nsite =

Equation (4-19)

 Nsite: Number of sites required

4 - 12


4.4 Application or service distribution model A key element in network planning is to estimate the number of users that each BS may support. To have an idea about the maximum number of subscribers that a typical BS can serve the information of possible different traffic types and their parameters are essential. But On the other hand, mixed application packet data networks are notoriously difficult to treat with statistical methods for the general case. The traffic engineering for how the bandwidth is apportioned to the various active connections is typically left to operator configuration and is not included in the standard. In this project, different application classes are introduced and the desired parameters and usage percentage related to each of the applications are specified. There are five major classes’ services or applications as shown in table (1) that are:  Multiplayer interactive gaming  VoIP and Video Conference  Streaming Media  Web browsing and instant messeging  Media Content Downloading To fulfil the required QoS specifications of each application a number of important parameters must be met. These parameters are: bit error rate, jitter, latency and minimum throughput. The list above is sorted in a decreasing delay sensitivity order. The latency sensitivity gives an allocation priority to the suffering application. According to the service types, the first application group can be classified in the VBR services. Since the goal of this project is to decide the maximum capacity of a typical base station, will we focus on the

4 - 13


minimum reserved data rate of each VBR service and leave the maximum sustained data rate for more advanced scheduling procedures. The first application class i.e. Multiplayer Interactive Gaming needs a minimum reserved data rate of 50 kbps for each user. The second class belongs to the CBR service type with the average reserved data rate of 32 kbps for each user. The Streaming Media application group can be classified into VBR services with reserved data rate of 64 kbps. The last two application classes can be considered as best effort (BE) service type. The web browsing application group can be assigned the nominal data-rate of the user while the file transfer protocol (FTP) class is supported with the remaining capacity assigned to each Subscriber that is available after satisfying other guaranteed service types.

Table (4.5) applications or services distribution model However, the other important factor for capacity estimation of a typical base station is the user demands and the trend of each user type. In the

4 - 14


coming sections an application distribution scenario and two important scales to follow the market trends are presented. 4.4.1 Service Flows In the previous sections, we have examined the various factors that influence the overall channel bandwidth. What remains after accounting for the per-channel and per-packet overhead is the usable channel bandwidth. This channel size is the relevant quantity for determining the service capacity consistent with the QoS parameters. The traffic engineering for how the bandwidth is apportioned to the various active connections is typically left to operator configuration. In this section we will illustrate one way in which this could be accomplished. We begin by reviewing the three basic service types. In general service flows related to each application can be identified with two major traffic rate allocation types: (i)The Reserved Traffic Rate It is the committed information rate for the flow of the data rate that is unconditionally dedicated to the flow and therefore can be directly subtracted from the available user channel size to determine the remaining capacity. (ii)The Sustained Traffic rate It is the peak information rate that the system will permit. Traffic, submitted by a subscriber station at rates bounded by the minimum and maximum rates, is dealt with by the base station on a non-guaranteed basis.

4 - 15


Based on the above traffic rate allocation methods three service flows can be defined. These services are as follows: 4.4.2 Constant Bit Rate (CBR) Services System can support Constant Bit Rate (CBR) by configuring dedicated frequency-time channel grants to specific traffic flows. The dedicated resources correspond to a constant throughput rate. CBR service flows are suitable for applications with strict latency and throughput constraints and that generate a steady stream of fixed size packets such as VoIP. These service flows can be dynamically set up or torn down in response to detection by the system of changing traffic needs. On the downlink, the base station directly controls the scheduling of traffic and allocation of the frequency-time channel resources. Dedicating a portion of the channel bandwidth for CBR flows is therefore a matter of keeping track of the allocated resources and transporting any available packets from appropriately classified traffic. For the uplink the Unsolicited Grant Service (UGS) scheduling method is used. The base station dedicates a portion of the uplink channel bandwidth to a Subscriber Station corresponding to one or more service flows for the duration of the flow. The base station communicates this assignment to the Subscriber Station in the uplink channel usage maps that are periodically broadcast out to all stations. From a capacity standpoint, the key CBR QoS parameter is the unvarying Maximum Sustained Traffic Rate, which is the committed information rate for the flow. The maximum rate is unconditionally dedicated to the flow and therefore can be directly subtracted from the

4 - 16


available user channel size to determine the remaining capacity. The only overhead associated with CBR flows is the UGS grant overhead, which increases the size of the uplink channel usage map. Although the bandwidth is dedicated for a CBR service flow, the base station scheduler implementation could still elect to temporarily “borrow” the dedicated bandwidth on the downlink frame if there is no CBR traffic to send. The scheduler must however issue uplink grants according to the CBR service flow configuration whether or not the subscriber station has any traffic to send (the scheduler has no way of knowing in advance). CBR service has a maximum reserved traffic rate. This service is suitable for applications with strict latency and throughput constraints and those that generate a steady stream of fixed size packets such as VoIP.

4.4.3 Variable Bit Rate (VBR) Services For applications that have variable traffic throughput demands systems support Variable Bit Rate (VBR) services. VBR service flows are suitable for applications that generate fluctuating traffic loads including compressed streaming video and VoIP with silence suppression. On the down link (DL), the base station directly controls the scheduling of traffic and allocation of the frequency-time channel resources. Dedicating a portion of the channel bandwidth is therefore a matter of keeping track of the allocated resources and transporting any available packets from appropriately classified traffic. The base station

4 - 17


performs this scheduling successively for each TDMA frame that is sent out for example every 10 ms so that the time varying nature of the VBR traffic can be supported in real time. For the uplink (UL), there are several scheduling methods depending on the QoS requirements for the service flow. For flows with strict real time access constraints, periodic polling assures that the subscriber station will have guaranteed channel access up to a specified Minimum Reserved Traffic Rate. Real time Polling Service (rtPS) operates by having the base station poll individual subscriber stations periodically for example every frame to solicit bandwidth requests. Extended real time Polling Service (ertPS) operates more like UGS except that the committed maximum rate can be changed on the fly as controlled by subscriber station signalling. For flows with looser real time access constraints, non real time Polling Service (nrtPS) operates like rtPS except the polls can be directed at individual or groups of subscriber stations, and the latency of the base station response to bandwidth requests is not guaranteed. The subscriber stations can also use piggyback methods to request continuing channel access.

4 - 18


Table (4.6) mobile service flows and QoS parameters For capacity calculations, the two key VBR QoS parameters are the Minimum Reserved Traffic Rate and the Maximum Sustained Traffic Rate. For VBR, the minimum rate corresponds to the committed information rate. Since the minimum rate is guaranteed, it can be directly subtracted from the available user channel size to determine the remaining capacity. The maximum rate is the peak information rate that the system will permit. Traffic, submitted by a subscriber station at rates bounded by the minimum and maximum rates, is dealt with by the base station on a non-guaranteed basis. The overhead associated with VBR service comes from the polling method except for ertPS, which basically has the same overhead as UGS i.e. the size of the uplink channel usage maps is increased for each active flow.

4 - 19


If the polls are directed at a group of subscriber stations the responses must use a contention bandwidth request interval to respond since request collisions can occur. Although the bandwidth is dedicated for the Minimum Reserved portion of the VBR service flow, the base station scheduler implementation could still elect to temporarily “borrow” the dedicated bandwidth on the downlink frame if there is no traffic to send. The scheduler must however issue uplink grants for bandwidth requests according to the VBR service flow configuration for the Minimum Reserved QoS parameter whether or not the subscriber station has any traffic to send (the scheduler has no way of knowing in advance). VBR has a minimum reserved and a maximum sustained traffic rates. These types of service flows are suitable for applications that generate fluctuating traffic loads including compressed streaming video. 4.4.4 Best Effort (BE) Services Best effort (BE) services are intended for service flows with the loosest QoS requirements in terms of channel access latency and without guaranteed bandwidth. Best effort services are appropriate for applications such as web browsing and file transfers that can tolerate intermittent interruptions and reduced throughput without serious consequence. On the downlink, the base station directly controls the scheduling of traffic and allocation of the frequency-time channel resources. For best effort services, the affected traffic is sent as surplus capacity that is available after satisfying other guaranteed service types.

4 - 20


On the uplink, the base station should provide periodic contention intervals in order for subscriber stations with best effort flows to submit their bandwidth requests. The subscriber stations can also use piggyback methods to request continuing channel access. The overhead associated with best effort services comes from providing the contention intervals for bandwidth requests. BE are intended for service flows with the loosest QoS requirements in terms of channel access latency and without guaranteed bandwidth. Best effort services are appropriate for applications such as web browsing and file transfers that can tolerate intermittent interruptions and reduced throughput without serious consequence. For best effort services, the affected traffic is sent as surplus capacity that is available after satisfying other guaranteed service types. Figure (1) shows a schematic the available bandwidth that is partitioned Based on the presented bandwidth partitioning methodology, each of the desired applications can be assigned with the desired service flow based on its required quality of service (QoS) parameters. As mentioned before the realization procedure of this task is not included in the standard and each vendor must implement it utilizing appropriate traffic scheduling processes for time and frequency channel recourse allocations. The scheduling is directly controlled by each Base Station.

4 - 21


4.4.5 Sharing Non-Guaranteed Bandwidth In comparing best effort services against variable bit rate services an ambiguity becomes apparent. The system must by definition not admit more guaranteed bandwidth traffic onto the channel than it can supply. On the other hand, VBR and BE services can both have non-guaranteed traffic. For VBR it is the portion of traffic submitted at rates above the Minimum Reserved rate. For BE it is all of the submitted traffic. How should the scheduler deal with this situation in cases where there is insufficient remaining capacity to honor all requests? Shown graphically in figure (1), what should happen if regions C and D overlap? The answer is not specified by the 802.16 standard but is left to vendor implementation.

Figure (4.1) channel bandwidth partitioning Note that the figure illustrates the case where the scheduler actually has traffic to fill the guaranteed portion of the channel. If that were not the case then in theory the scheduler can temporarily borrow the guaranteed

4 - 22


bandwidth to satisfy non-guaranteed bandwidth requests. For capacity estimations we need to assume the worst case where the guaranteed bandwidth is in use. This should not come as a surprise; the base station scheduler design is similarly not described by the standard. The authors of the standard were trying to balance the conflicting requirements of creating a standard while allowing freedom where possible for product differentiation and innovation. One simple way to deal with the issue might be to implement a policy of fair-sharing the non-guaranteed bandwidth between VBR and BE. That is, equally divide any remaining bandwidth up between all requesting VBR and BE service flows. The problem with this approach is that is does not allow service providers much control to differentiate their services. The other problem is that, while VBR can specify a minimum information rate, BE services under severe congestion can be starved with throughput rates approaching zero. A better solution is to provide a method for prioritizing access to non-guaranteed bandwidth, which can be done by introducing the concept of service flow over-subscription. 4.4.6 Quality of service (QoS) Control modeling Dimensioning a network needs to keep in mind the user traffic demand and the applications it uses so that the density of Base Stations and backbone network dimensioning can fulfill the demand. Another important task in service provision is to support the QoS parameters of each connection over the demanded bandwidth. In our current algorithm we benefit two Over Subscription Ratio (OSR) and Contention Ratio

4 - 23


(CR) measures in order to apply quality of service (QoS) control over the expected traffic that will be explained in this section. 4.4.7 Contention Ratio (CR) As the customer base is growing, there must be a measure of the simultaneity of users requesting bit rate from the Base Stations because most users won’t demand data at the same time. In simplest terms it means that, the absolute peak demand on shared resources rarely occurs. This user simultaneity is defined by a parameter we call contention ratio. On the other hand, many of the connected subscribers will demand data whose packets can be delivered assuming some latency or jitter i.e. less priority. The available channel bandwidth can be allocated to the users in a guaranteed and non-guaranteed moods based on the applications. Generally, applying a contention ratio (CR) for the guaranteed bandwidth is a practice that operators should approach with caution since their customers naturally expect that their service agreements will be honored always. In our algorithm, no Contention Ratio is applied over the guaranteed partition of the channel bandwidth. However, in future developments assigning a CR over reserved bandwidths that correspond to the error or blocking probability of each application will result in a more accurate traffic modelling. According to the algorithm proceeded in this thesis, two contention ratios are defined for the non-guaranteed partition of the bandwidth. Typical values for contention ratios can be about 30 for residential users (less priority) up to 10 for business users (higher priority and throughput).In this case, if a Residential Class and a Business Class Subscribers have contracted a downlink BE service of the

4 - 24


rates 512 kbps and 1Mbps respectively, 512/30=17 kbps and 1000/10=100 kbps are the actual data-rates that must be considered in the system total capacity calculations. This is while the data rate of the services with guaranteed bandwidth (CBR,VBRMR) will remain untouched. Figure (2) illustrates the distribution of two different service classes traffic model. 4.4.8 Over Subscription Ratio (OSR) Over-subscription ratio, sometimes called over-booking ratio, in simplest terms means taking advantage of the fact that, for many systems, absolute peak demand on shared resources rarely occur. Examples are everywhere in daily life. Air lines aggressively over-subscribe their seat capacity. Public telephone networks over-subscribe their network switching capacity. The point of over-subscription is that system capacity requirements can be significantly reduced if the requirement to handle absolute worst-case scenarios is ignored. However, over-subscription comes at a price that is related to trading hard guarantees of service for soft statistical guarantees. Depending on the nature of shared resource usage i.e. the traffic, and how aggressively the resource is oversubscribed, there can be exceptional periods where there is more demand than can be served. The standard also includes the ability to specify a traffic priority QoS parameter for VBR and BE service flows. This allows basic grouping of priority between sets of service flows. However, it does not distinguish between guaranteed and non-guaranteed VBR traffic or allow division of priority beyond eight basic levels.

4 - 25


How mathematically rigorous the statistics of the guarantees are usually depends on how much is known about the offered traffic. One well-known example is the blocking probability associated with traditional voice Erlang statistics. On the other hand, mixed application packet data networks are notoriously difficult to treat with statistical methods for the general case. Often this results in resorting to empirical rules derived from traffic measurements of a given user population. In the case of mobile networks, operators can choose to oversubscribe the total network capacity in order to improve overall network utilization and cost per line business economics. There are two basic scenarios. An operator can choose to over-subscribe one or more service flow’s ‘guaranteed’ bandwidth, or they might choose to over-subscribe their non-guaranteed bandwidth. Generally over-subscription of guaranteed bandwidth is a practice that operators approach with caution since their customers naturally expect that their service agreements will be honored always. But the fine print of these agreements may also allow for hopefully rare periods when the network will not be able to support the guaranteed performance. One simple example could be that voice over internet protocol (VoIP) users are guaranteed that less than 1% of their call attempts will be blocked. This can be accomplished by using Erlang statistics to reserve an oversubscribed block of bandwidth sufficient to support a given number of voice lines. Over-subscription of non-guaranteed bandwidth is of course fair game but an operator must still balance their users’ service level expectations against the degree of over-subscription of the network

4 - 26


capacity. If users are told that they can expect “up to” some peak level of service but discover that during busy hours that they can only get one tenth of that service they will likely be dissatisfied with their service. Often this is handled by marketing a “typical” level of service associated with a given level of over-subscription (related to the total number of users) and an “up to” service rate limit. Returning to the issue of shared non-guaranteed bandwidth between VBR and BE service flows, one solution for prioritizing the access would be to associate a level of over-subscription to each service flow. For VBR flows there are two relevant independent levels of over-subscription, one for the guaranteed Minimum Reserved portion, and a second for the nonguaranteed portion corresponding to rates bounded by the Minimum Reserved and the Maximum Sustained limits. For BE flows there is just one level of over-subscription associated with the Maximum Sustained limit. If the system allows the service flows to be configured in this manner then the relative priority ranking of the non-guaranteed portions of the VBR and BE service flows can be accomplished. This in turn allows operators to calculate the total number of lines of service that can be provisioned for a given service scenario. The problem of allocating the aggregate system capacity to the various service flows must take into account the QoS requirements of those flows. Dedicated or guaranteed bandwidth must be dealt with first and what remains is shared by non-guaranteed services. OSR is the ratio of the total subscriber’s demand over the reference capacity of the base station when taking into account the adaptive

4 - 27


modulation. The reference capacity of the base station corresponds to the available bit rate of the lowest modulation scheme served with that BS. Subscriber classes distribution model Consider the two subscriber classes i.e. business and residential. Assume that the residential class occupies 58% of the users under cover of our base station while the business class users are confined to 42%. As shown in table (4.3). Subscriber class

Percentage or weight%

Business subscriber class (B)

58%

Residential subscriber class (R)

42%

Table (4.7) subscriber class distribution model The total subscriber’s demand capacity refers to the repartition of the subscribers based on their type of service. In this case the total capacity for OSR calculation would be: Ctot=N× (PR×BWR+ PR×BWB)

Equation (4-20)

Ctot = N × (58% x 512 + 42% x 1000) OSR = Ctot /Cref

Equation (4-21)

Where  N refers to the number of users that are connected to the base station (BS).  OSR is a measure of QoS in cell planning. A fair trade off between OSR and CRs of traffic model will provide us with a good measure of QoS control. This is because of the fact that the

4 - 28


CRs help us to have a realistic model of the in use traffic based on the modulation distribution of the subscribers within the coverage area, while the OSR gives us an idea about the traffic demand that the operator has committed. 4.4.9 Application or service Distribution and Market Trends Therefore, studying the traffic demand of existing service providers can give us an idea about the subscribers’ possible application distribution while using metropolitan broadband wireless services. As can be seen, the most significant usage belongs to HTTP web browsing applications. While the total percentage of the point to point (p2p) services is almost 60% of all traffic, due to applying bandwidth limitation over point to point (p2p) in October it drops to 14%. Streaming traffic increased from 1.24% 12.5% mainly because of submission of mobile TV. These values are used to model our application distribution. Table (4.3) summarizes this model which is the final distribution that will be taken in to consideration in our capacity calculation algorithm.

Figure (4-2) subscriber class deployment model

4 - 29


4.4.10 Subscribers’ traffic demand Now that all application distribution parameters are completely defined, the minimum bandwidth of the demanding traffic can be calculated. The phrase minimum demand here signifies that we are only relying on the minimum reserved data-rate required for the applications including guaranteed bandwidth.

Subscriber class Business subscriber class (BWB)

1Mbps

Residential subscriber class (BWR)

512 Kbps

Table (4.8) subscriber class traffic model This fact enables us to derive the maximum supportable capacity of a generic sector. In our algorithm, the traffic demand is categorized into 2 subscriber classes. Adding more classes is an easy task and won’t change The relations below conduct traffic demand calculation path for residential and business class subscribers and the Total Traffic Demand for DL. DRresrved=P1×DR1+P2×DR2+P3×DR3

Equation (4-22)

DRreserved = 25% x 50 + 10% x 32 + 12.5% x 64 DRshared-R=P4×DR4+P5 × (BWR-(DR1+DR2+DR3)

Equation (4-23)

DRshared-R = 32.5% x BWR + 20% x (BWR - (50+32+64) DRshared-B=P4×DR4+P5 × (BWB-(DR1+DR2+DR3)

Equation (4-24)

DRshared-B = 32.5% x BWB + 20% x (BWB - (50+32+64)) Traffic R = N x (%PR) x (DRreserved + (DRshared-R / CR R) Equation (4-25)

4 - 30


Traffic B = N x (%PB) x (DRreserved + (DRshared-B / CR B) Equation (4-26) Traffic Total = Traffic R + Traffic B The parameters are as follow: DRreserved: Minimum Reserved (Guaranteed) Data-rate for CBR/VBR Applications DRshared-R : Shared Data-rate for Residential Class users with BE Applications DRshared-B: Shared Data-rate for Business Class users with BE Applications BWR: Residential class subscribers data-rate based on user agreement BWB : Business class Subscribers data-rate based on user agreement N: Total number of the users connected to the sector %PR: Percentage of the residential class subscribers within the area under study CR R: Contention Ratio for residential class subscribers %PB: Percentage of the business class subscribers within the area under study CRB: Contention Ratio for business class subscribers

4 - 31

Chapter Five Numerical Results

Chapter 5: Numerical Results

Chapter Five Numerical Results

Start Preliminary study about LTE

Problem specific study and Review of the related works Theoretical Understanding (Input/output specification, etc) Basic Dimensioning Tool started Work on LTE Dimensioning and Tool Coverage Planning (Radio Link Budget, Number of sites needed based on Capacity) Capacity Evaluation

Review: Is the work complete? Proceed with documentation

End

Flow chart of Project Work

5-2


5.1 Up Link Budget 5.1.1 User Equipment effective Isotropic Radiated Power (EIRP):Transmitted power per PRB

Gain

EIRP

-H/B losses Figure (5-1) flowchart of effective isotropic radiated power

Figure (5-2) Calculation of EIRP

5-3


Inputs:n’PRbs UE antenna Gain P(UE) P(UE,PRB) Other Gains Head Body loss Output:EIRP(UE) Equation:P(UE,PRB) = EIRPUE,PRB = PUE,PRB + GUE + Gothers – LHBL Where:n’PRBs: Number of Physical Resource Blocks PUE: user equipment output Power P(UE,PRB): Power per Physical Resource Blocks GUE: User equipment transmitting antenna gain [dBi] LHBL : head body loss [dB] G others : It is gains due to MIMO EIRPUE,PRB: Effective Isotropic Radiated Power. Excel Results Coverage (UL)---- RL budget 1-UE parameters Item

Unit

Values

P(UE)

dBm

23

P(UE)

Watt

199.5262315

Channel B.W

MHz

20

nPRB

PRBs

100

P(UE.PRB)

Watt

1.995262315

P(UE,PRB)

dBm

33

UE Tx gain

dBi

0

other UE gain

dBi

2

L(HBL) in VOIP

dB

3

EIRP (UE,PRB)

dBm

32

Table (5-1) EIRP

5-4


5.1.2 Enhanced NodeB Sensitivity: -

Thermal noise

Noise figure

Log Wprb(B.W per PRB)

Sensitivity

SINR

Figure (5-3) flowchart of sensitivity of eNodeB

Figure (5-4) Sensitivity of Enhanced nodeB

5-5


Inputs:K T Nt Nf WPRB a0 a1 a2 a3

γtarget

Output:SeNodeB SINR N(PRB,UL) Equation:-

SeNodeB= Nt + Nf + logWPRB +γtarget,UL Where: Nt: It is thermal noise power density  Nf: noise figure of receiver  Wprb : bandwidth per physical resourse block  γtarget,UL: SINR requirement for uplink traffic channel Excel Results 2-BS parameter Item

Unit

Values

Boltzman constant

J/K

1.38E-23

Ambient temperature

K

290

Thermal noise power density (Nt)

dBm/Hz

-174

Noise figure of eNodeB (Nf)

dB

2

BW per resource block [W (PRB)]

MHz

0.18

a0

Kbps

536.6

a1

dB

20.76

a2

dB

13.28

a3

Kbps

0

Bit rate [R(PRB)]

Kbps

64

SINR

dB

-2.499734217

Sensitivity of e.NoodeB

dB

-181.9470092

Table (5-2) Default values of Enhanced eNodeB sensitivity

5-6


5.1.3 Interference Margin (IM)

Figure (5-5) flowchart of Interference Margin

5.1.4 Log Normal Fading Margin (BLNF)

Figure (5-6) flowchart of Log Normal Fading Margin

5-7


5.1.5 Total Margins:-

Figure (5-7) flowchart of total margins

Figure (5-8) Total margin Inputs:SINR (Gamma) CLF

5-8


Interference Factor (F) µ σ P% LNF BI (UL) FFM Outputs:(IM) BIUL BLNF Total Margins Equation:BIUL=

, , , BLM= norm inverse (P%,µ,σ) Total Margins= LNF+ BI(UL)+ FFM Where:γ target,Ul: Is the SINR target for the Uplink open loop Power Control QUE: Is the average Uplink System load F : It is the average ratio of Path gains for interfering cells to those of the serving cell. µ: is the mean of lognormal σ : is the standard deviation of lognormal P : is the coverage probability LNF: log normal fading margins BI(UL): Interference margin FFM: fast fading margin 5-Margins Item

Unit

Mean of Log normal (µ)

---------

Standard Deviation (σ)

dB

Area of Coverage

Values 0 3 Flat Area

edge Coverage Prob. F(p)

%

90

Lognormal Fading Margin [B(LNF)]

dB

3.844654697

Cell Loading Factor [Q(UL)]

---------

0.64

F

---------

0.7

B(IUL)

dB

1.26066082

Fast Fading Margin [B(FFM)]

dB

2

Total Margin

dB

Table (5-3) Default values Total margin

5-9


5.1.6 Total Gains:-

5.1.7 Total Losses

Figure (5-9) flowchart of total gains

Connector loss Connector length

Connector specifications Jumper loss

Total losses Jumper specifications

Jumper length Car penetration loss

Head/body loss

Building penetration loss

Figure (5-10) flowchart of total losses

5 - 10


Figure (5-11) total gains and total losses

Inputs:G1 G2 BPL CPL Jumper Length Jumper Loss Feeder length Feeder loss Contactor loss outputs:Gt: Total Gains Total Losses Equation:Total Gains = eNodeB antenna Gain + Other Gains Total Losses= BPL+ CPL+ Jumper Length+ Jumper Loss+ Feeder length+ Feeder loss+ Contactor loss Where:G1: eNodeB antenna Gain

5 - 11


G2: Other gains Gt : Total Gains BPL: Building Penetration Loss CPL: Car penetration loss 4-other eNodeB parameter Item

Unit

Values

Gain of e.NodeB

dBi

18

Other gain of e.NodeB

dBi

4

Total Gain

dBi

22

Building Pentration Loss [LBPL]

dBi

15

Car Penteration Loss [LCPL]

dBi

9

BS Feeder Specification

(dB/100m)

3

e.NodeB Feeder Length

Meter

30

e.NodeB Feeder Loss (Lf)

dB

1.05

e.NodeB Jumper Specification

dB/100m

2

e.NodeB Jumper Length

Meter

5

e.NodeB Jumper Loss (Lj)

dB

0.1

e.NodeB Connector Loss (Lc)

dB

1

Total Loss

dB

26.15

Table (5-4) Total Losses and Gain

5.1.8 Maximum Allowable Paths Loss (MAPL)

EIRP Gain -Losses

MAPL

-margins -Sensitivity of eNodeB Figure (5-12) flowchart of maximum allowable path loss

5 - 12


Figure (5-13) Max. Allowable path loss in using GUI in Matlab Inputs:EIRP(UE,PRB) SeNodeB Total losses (LBPL + LCPL + LeNodeB +Lj +LC ) Total Gains (GeNodeB + Gother) Total margins (BLNF + BIul) Output:MAPLUL Equation:MAPLUL=EIRPUL,PRB – SeNodeB – (BLNF + BIul) – (LBPL + LCPL + LeNodeB + Lj +LC ) + (GeNodeB + Gother) Where:BLNF: lognormal fading margin [dB] BIul: UL interference Margin [dB] LCPL: car penetration loss [dB] LBPL: Building Penetration Loss GeNodeB: eNodeB Reciever antenna gain Gother: is other Gain [dBi] Lf eNodeB: is eNodeB feeder loss [dB]

5 - 13


Lj: jumper loss LC: connector loss Excel Results 6-Max allowable path loss Item

Unit

Values

MAPL

dB

202.6916936

Table (5-5) Default values of Maximum allowable path loss (MAPL)

5.1.9 Cell Radius Using Ericson Variant Okumura-Hata

Figure (5-14) flowchart of cell radius using Ericson variant Okumara Hata

5 - 14


5.1.10 Site Count

Figure (5-15) flowchart of site count

Inputs:MAPL hb hm

Figure (5-16) cell radius and Site Count

5 - 15


a(hm) Frequency A Deployment Area Cell Radius Cell Area Outputs:R in kilometers Site Count Equation:R=10α α= Site Count = Where:Lo= A+13.82loghb+ a(hm) ϒ= 44.9 – 6.55loghb MAPL: maximum allowable paths loss hb: base station or eNodeB antenna height [m] hm: height of user equipment antenna [m] a(hm): inverse relationshipis written as follows MAPL=A – 13.8loghb – a(hm) + (44.9 – 6.55log hb)log R A: frequency-dependent fixed attenuation value Sc: site count =

Table (5-6) values of Cell Radius and Site count with difference Base stations heights

5 - 16


5.2 Effects on cell Radius (R) In the following subsections the effect of the following parameters will be investigated 1- Effect of cell types, we will considered omni cell and 3 sectors cell 2- Impact of different morphologies, we will considered Rural. Suburban and Urban. 3- Effect of cell loading factor 4- Effect of eNodeB antenna height 5.2.1 The effect of cell Loading Factor (Q) on the cell Radius (R) Omni CLF RU cell R UR SU DU

0.1 599.47 49 66.508 82 162.17 9 218.58

0.2 592.22 86 65.748 2 160.26 39 216.13 5

0.3 584.75 93 64.963 65 158.28 93 213.61 23

0.4 577.04 95 64.153 27 156.25 05 211.00 59

0.5 569.07 98 63.314 95 154.14 23 208.30 88

0.6 560.82 8 62.446 28 151.95 88 205.51 33

0.7 552.26 87 61.544 51 149.69 31 202.61 04

0.8 543.37 28 60.606 48 147.33 74 199.58 98

0.9 534.10 65 59.628 51 144.88 27 196.43 94

Table (5-7) the effect of cell Loading Factor (Q) on the cell Radius (R) Omni cell

Figure (5-17) the effect of cell Loading Factor (Q) on the cell Radius (R) Omni cell

5 - 17


We conclude that in case of omni cell as cell loading factor increase cell radius decreases for different types of morphologies For certain cell loading factor the cell radius increase as we go from urban to suburban to rural For example for cell loading factor 50% cell radius in urban = For cell radius in suburban = Km For cell radius for rural = Km 5.2.2 The effect of cell Loading Factor (Q) on the cell Radius (R) 3 Sector h(B) /meter RU cell R UR SU DU

10

20

30

40

50

60

70

80

90

100

253.1 066 56.07 967 96.84 865 215.7 335

444.5 555 90.77 23 161.4 741 375.6 464

636.3 798 123.3 535 223.6 102 534.7 792

832.8 866 155.2 647 285.4 681 697.0 233

1035. 47 187.0 305 347.8 312 863.6 646

1244. 719 218.9 044 411.0 675 1035. 261

1460. 917 251.0 28 475.3 765 1212. 091

1684. 21 283.4 883 540.8 769 1394. 3

1914. 677 316.3 421 607.6 433 1581. 973

2152. 359 349.6 283 675.7 252 1775. 158

Table (5-8) the effect of cell Loading Factor (Q) on the cell Radius (R) 3 sector

Figure (5-18) the effect of cell Loading Factor (Q) on the cell Radius (R) 3 sector

5 - 18


We conclude that for 3 sector cell the same result as omni cell but more over cell radius of 3 sector cell is larger than cell radius of omni cell on all morphologies For example for cell loading factor 50% cell radius in urban For cell radius in suburban For cell radius for rural 5.2.3 Effect of eNodeB antenna height on Cell Radius in omni cell h(B) /meter RU cell R

10

20

30

40

50

60

70

80

90

100

176.5 427

304.0 831

429.9 145

557.4 427

687.8 188

821.5 555

958.9 149

1100. 043

1245. 025

1393. 912

UR

39.11 576

62.08 971

83.33 301

103.9 171

124.2 365

144.4 842

164.7 695

185.1 606

205.7 025

226.4 265

SU

67.55 226

110.4 509

151.0 628

191.0 609

231.0 496

271.3 182

312.0 271

353.2 74

395.1 219

437.6 136

DU

150.4 748

256.9 482

361.2 769

466.5 107

573.6 96

683.3 069

795.5 906

910.6 878

1028. 683

1149. 629

Table (5-9) the effect of eNodeB antenna height on the cell Radius (R) omni cell

Figure (5-19) the effect of eNodeB antenna height on the cell Radius (R) omin cell

5 - 19


We conclude that the cell radius of Omni cell increases as eNodeB antenna height increase for different types of morphologies For certain eNodeB antenna height cell radius increases as we go from urban to suburban to rural for certain eNodeB antenna height for example H = 30 m cell radius in urban Cell radius in suburban Cell radius in rural

5.2.4 Effect of eNodeB antenna height types on Cell Radius in 3sector cell h(B) /meter

10

20

30

40

50

60

70

80

90

100

RU cell R

253.1 066

444.5 555

636.3 798

832.8 866

1035. 47

1244. 719

1460. 917

1684. 21

1914. 677

2152. 359

UR

56.07 967

90.77 23

123.3 535

155.2 647

187.0 305

218.9 044

251.0 28

283.4 883

316.3 421

349.6 283

SU

96.84 865

161.4 741

223.6 102

285.4 681

347.8 312

411.0 675

475.3 765

540.8 769

607.6 433

675.7 252

DU

215.7 335

375.6 464

534.7 792

697.0 233

863.6 646

1035. 261

1212. 091

1394. 3

1581. 973

1775. 158

Table (5-10) the effect of eNodeB antenna height on the cell Radius (R) 3 sector

Figure (5-20) the effect of eNodeB antenna height on the cell Radius (R) 3 sector

5 - 20


For 3 sector cell the same results as omni cell but more over cell radius of 3 sector cell is larger than cell radius of omni cell For 3 sector cell For certain eNodeB antenna height for example H = 30 m cell radius in urban Cell radius in suburban Cell radius in rural 5.3 Downlink capacity

Figure (5-21) downlink capacity

Inputs:MAPL LNF margin Total Gain Total Losses P (norm,ref) N (PRB) Q (CLF) H Channel Model Over Booking Factor

5 - 21


Subscriber Class Subscriber Data rate Code rate Number of user Application services Outputs:L (sa,max,DL) P (e Node B,PRB) R(PRB,DL) Interface Margine (BIDL) SINR Ravg,DL Total Through put Number of cell required (Nrequired) T cell,DL Equation:-

 Nrequired = Tt/Tsite 







5 - 22


 RAVG,DL = RRB,DL (nRB,DL - nPDCCH)   Tsite = Tcell,DL × number of active users (U)  Tsite = Tcell,DL  (Omni cell )  Tsite = 3 × Tcell (3 sector cell)  Ttotal = U × TU × (OBF)  Nsite =  Ctot=N× (PR×BWR+ PR×BWB)  C tot = N × (58% x 512 + 42% x 1000)  OBF = Ctot /Cref 

DRresrved=P1×DR1+P2×DR2+P3×DR3

 DRreserved = 25% x 50 + 10% x 32 + 12.5% x 64  DRshared-R=P4×DR4+P5 × (BWR-(DR1+DR2+DR3)  DRshared-R = 32.5% x BWR + 20% x (BWR - (50+32+64)  DRshared-B=P4×DR4+P5 × (BWB-(DR1+DR2+DR3)  DRshared-B = 32.5% x BWB + 20% x (BWB - (50+32+64))  Traffic R = N x (%PR) x (DRreserved + (DRshared-R / CR R)  Traffic B = N x (%PB) x (DRreserved + (DRshared-B / CR B)  Traffic Total = Traffic R + Traffic B = Tu × OBF • Where:n'PRBs : number of physical resourse block • RPRB,DL : Down Link data rate per physical resourse block • (nPUCCH) : number of physical DL control channels • QDL : The average downlink system load

5 - 23


• T(user) : Throughput for user • B(IDL) : Downlink Interference margin (IM) or noise rise • P(eNodeB,DL) : eNodeB transmitted or radiated power per physical resourse block • N(PRB,DL) : Down Link thermal noise per physical resourse block • L(sa,max) : Maximum Down Link signal attenuation • H : The average attenuation • R(avrege,DL) : Average Downl Link data rate • T(cell,DL) : The Downl Link data rate • T(total) : Total Throughput • T(site) : Throughput for site • SINR(DL,avg) : Average Downl Link signal to interference and noise ratio • DRreserved: Minimum Reserved (Guaranteed) Data-rate for CBR/VBR Applications • DRshared-R : Shared Data-rate for Residential Class users with BE Applications • DRshared-B: Shared Data-rate for Business Class users with BE Applications • BWR: Residential class subscribers data-rate based on user agreement

5 - 24


• BWB : Business class Subscribers data-rate based on user agreement • N: Total number of the users connected to the sector • %PR: Percentage of the residential class subscribers within the area under study • CR R: Contention Ratio for residential class subscribers • %PB: Percentage of the business class subscribers within the area under study • CR B: Contention Ratio for business class subscribers

Figure (5-22) Final Base site count

5 - 25

Chapter Six Conclusion and Suggestions for Future Work

Chapter 6: Conclusion and Suggestions for Future Work Chapter six Conclusion and Suggestions for Future Work 6.1 Conclusions In this project, we study LTE network coverage and capacity dimensioning. Dimensioning process is a part of planning process and provides the network elements count as well as the capacity of those elements. We considered only access network dimensioning .Thus; the output of the dimensioning is the number of eNodeB that fulfil coverage and capacity requirements. LTE coverage dimensioning is done via radio link budget (RLB) and suitable propagation models .The output of RLB is the MAPL. Then using a suitable propagation model, the cell radius is obtained. Cell radius is used to obtain site count. LTE capacity dimensioning is obtained, given the number of subscribers, their demanded services and subscriber usage level. The cell radius based on capacity is determined. Two values of cell radius are obtained: ▪ One from coverage dimensioning ▪ Second from capacity dimensioning the larger of the two numbers is taken as the final output In this thesis, we consider the effect of the following parameters: 1- Effect of cell types, we will considered omni cell and 3 sectors cell 2- Impact of different morphologies, we will considered Rural. Suburban and Urban. 3- Effect of cell loading factor 4- Effect of eNodeB antenna height The effect of cell Loading Factor (Q) on the cell Radius (R) Omni cell

We conclude that in case of Omni cell as cell loading factor increase cell radius decreases for different types of morphologies For certain cell loading factor the cell radius increase as we go from urban to suburban to rural

The effect of cell Loading Factor (Q) on the cell Radius (R) 3 Sector We conclude that for 3 sector cell the same result as omni cell but more over cell radius of 3 sector cell is larger than cell radius of omni cell on all morphologies

6-2

Chapter 6: Conclusion and Suggestions for Future Work Effect of eNodeB antenna height on Cell Radius in omni cell We conclude that the cell radius of omni cell increases as eNodeB antenna height increase for different types of morphologies For certain eNodeB antenna height cell radius increases as we go from urban to suburban to rural Effect of eNodeB antenna height on Cell Radius in 3 sector cell For 3 sector cell the same results as omni cell but more over cell radius of 3 sector cell is larger than cell radius of omni cell Finally, a dimension tool is developed. In this project, interference system based capacity dimensioning is studied. 6.2 Suggestions for future work  In this project we considered only LTE coverage and capacity dimensioning. Data analysis, Traffic analysis and Transport dimensioning can be studied in the future.  In this project, we considered only access network, LTE core network can be studied to determine core network nodes and the number of backhaul links required.  In this project, we considered VOIP only, other services such as web browsing, file transfer and multimedia can be studied individually, then developing traffic model for user including mixed services.  Detail LTE planning that include in addition to coverage and capacity dimensioning: frequency planning, neighbour planning and parameter planning, finally a planning tools is developed.  In addition to introducing digital three dimensional (3D) map which is imported in the planning tool as real prediction and simulations of the RF signal level in a real traffic distribution.  Other methods for capacity dimensioning such as cell ring based capacity method and modulation based capacity dimensioning method can be studied.  In this project, we consider FDD (Frequency Division Duplex), in the future we can use TDD (Time Division Duplex) or half duplex.

6-3

References 1) 3GPP Technical Report TR 25.813, “Radio Interface Protocol Aspects for Evolved UTRA”, version 7.0.0 2) “Long Term Evolution (LTE): an introduction,” Ericsson White paper, October 2007. 3) Dahlman, Parkvall, Skold and Beming, 3G Evolution: HSPA and LTE for Mobile Broadband, Academic Press, Oxford, UK, 2007. 4) Indoor radio planning : A practical guide for GSM, UMTS, HSPA, LTE , Second edition ,MortenTolstup, 2011 John Willey sons. 5) Wiley-VCH Verlag GmbH, Boschstrasse 12, D-69469 Weinheim, Germany 6) Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA 7) Abdul Basit Syed, Description of Models and Tool, Coverage and Capacity Estimation of 3GPP Long Term Evolution, February, 2009 .

Lte Coverage and Capacity Dimensioning

Recommend Documents