LOW NOISE AMPLIFIER PERFORMANCE STUDY FOR WIRELESS MAN
BASED ON IEEE 802.16A STANDARD
MAZHANIZA BINTI MAZUMIL
A project report submitted in partial fulfillment of the
requirements for the award of the degree of
Master of Engineering (Electrical – Electronics and Telecommunications)
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
MAY 2007
PSZ 19:16 (Pind. 1/97)
UNIVERSITI TEKNOLOGI MALAYSIA
BORANG PENGESAHAN STATUS TESIS υ
JUDUL:
LOW NOISE AMPLIFIER PERFORMANCE STUDY FOR
WIRELESS MAN BASED ON IEEE 802.16A STANDARD
SESI PENGAJIAN: SEMESTER II 2006/2007
Saya
MAZHANIZA BINTI MAZUMIL
(HURUF BESAR)
mengaku membenarkan tesis (PSM/Sarjana/Doktor Falsafah )* ini disimpan di Perpustakaan
Universiti Teknologi Malaysia dengan syarat-syarat kegunaan seperti berikut :
1.
2.
3.
4.
Tesis ini adalah hakmilik Universiti Teknologi Malaysia.
Perpustakaan Universiti Teknologi Malaysia dibenarkan membuat salinan untuk tujuan
pengajian sahaja
Perpustakaan dibenarkan membuat salinan tesis ini sebagai bahan pertukaran antara
institusi pengajian tinggi.
**Sila tandakan ()
√
SULIT
(Mengandungi maklumat yang berdarjah keselamatan
atau kepentingan Malaysia seperti yang termaktub di dalam
AKTA RAHSIA RASMI 1972).
TERHAD
(Mengandungi maklumat TERHAD yang telah ditentukan
oleh organisasi/badan di mana penyelidikan dijalankan).
TIDAK TERHAD
Disahkan oleh
(TANDATANGAN PENULIS)
(TANDATANGAN PENYELIA)
Alamat Tetap :
KAMPUNG SELEMAK
TANAH DATAR, 71300 REMBAU
NEGERI SEMBILAN
Tarikh :14 MEI 2007
CATATAN :
*
**
PROF. DR. THAREK BIN ABD RAHMAN
Tarikh : 14 MEI 2007
Potong yang tidak berkenaan
Jika tesis ini SULIT atau TERHAD, sila lampirkan surat daripada pihak berkuasa/organisasi
berkenaan dengan menyatakan sekali sebab dan tempoh tesis ini perlu dikelaskan sebagai
SULIT atau TERHAD.
♦ Tesis dimaksudkan sebagai tesis bagi Ijazah Doktor Falsafah dan Sarjana secara penyelidikan
atau disertasi bagi pengajian secara kerja kursus dan penyelidikan atau Laporan Projek
Sarjana Muda (PSM).
“I hereby declare that I have read this project report and in my opinion this project
report is sufficient in terms of scope and quality for the award of the degree of
Master of Engineering (Electrical – Electronics and Telecommunications)”.
Signature
:
Name of Supervisor
: Prof. Dr. Tharek bin Abd Rahman
Date
: 14 May 2007
ii
“I declare that this thesis entitled “Low Noise Amplifier Performance Study for
Wireless Man Based on IEEE 802.16a Standard” is the result of my own research
except as cited in references.” The thesis has not been accepted for any degree and is
not concurrently submitted in candidature of any other degree.
Signature
:
Name of Candidate
:
Mazhaniza binti Mazumil
Date
:
14 May 2007
iii
“To my beloved family and friends, thanks for being there, throughout this journey”
iv
ACKNOWLEDGEMENT
In the name of Allah, Most Merciful, Most Compassionate. It is by God
willing; I was able to complete this project within the time given. Here, I would like
to take this opportunity to express my sincere gratitude to my project supervisor,
Professor Dr. Tharek Abdul Rahman for his invaluable guidance, advice and support
towards completing this thesis succesfully.
Besides, I also thank to laboratory technician, Mr. Mohamed Abu Bakar,
postgraduates, undergraduates and the staffs of Wireless Communication Centre in
Universiti Teknologi Malaysia who have directly or indirectly assisted me in this
project. My deepest gratitude also goes out to my examiners, P.M. Dr. Jafri bin Din
and Dr. Nor Hisham bin Hj. Khamis for their constructive comments and suggestions
in evaluating my project
A special thank to research Master student, Mr. Waeil for his guidance and
support in this project.
Finally, I would like to express my warmest gratitude to my parents for their
support and encouragement.
v
ABSTRACT
In a rapid expanding worldwide wireless communications industry today, the
demand for Wireless Metropolitan Area Network (WMAN) systems is growing very
fast as well. New WMAN system based on IEEE 802.16a standard delivers high
data rate with the optional bandwidth, better spectral efficiency, improved
performance under multipath fading conditions and less interference in low-mobility
wireless conditions than earlier systems. To support high data in the systems multicarrier modulation, Orthogonal Frequency Division Multiplex (OFDM) is used. In
practice, component such as low noise amplifier (LNA) should be chosen based on
low cost. On the other hand, the effect of non-linear distortion must be considered
very carefully because the OFDM system is very sensitive to it. So, the WMAN
system must be tested and verified by using measurements so as to see the the system
meets the requirements of the IEEE standard. For RF receiver tests in particular, bit
error rate (BER) and packet error rate (PER) are required. There are several possible
approaches to test the complete WMAN system. However, in this project, a special
method that combines test equipment and simulation software has been developed to
verify a low noise amplifier (LNA) prototype by measuring the bit error rate (BER)
and packet error rate (PER) performance. Comparison has been made between the
simulated and tested performances of LNA. The results show that the LNA is within
the specification and standard.
vi
ABSTRAK
Saban hari, permintaan terhadap sistem Wireless Metropolitan Area Network
(WMAN) meningkat dengan drastik selaras dengan industri komunikasi wayarles
yang pesat membangun. Sistem WMAN yang berasaskan piawaian IEEE 802.16a
mampu menghantar data pada kadar yang tinggi dengan pilihan berbagai kadar jalur
lebar,
kecekapan
spektra
yang
lebih
baik,
persembahan
yang
lebih
memberangsangkan dan kurang interferen berbanding dengan sistem sebelum ini.
Untuk membekalkan kadar data yang tinggi, satu teknik dinamakan Orthogonal
Frequency Division Multiplex (OFDM) digunakan.
Secara praktik, komponen
seperti low noise amplifier (LNA) perlu dipilih berdasarkan kos yang rendah. Pada
masa yang sama, kesan seperti non-linear distortion harus diambilkira dengan teliti
kerana sistem OFDM sangat sensitif terhadap kesan seperti ini. Oleh itu, sistem
WMAN harus diuji dan ditentusahkan untuk memastikan sistem berada pada taraf
piawaian IEEE yang diingini. Bagi penerima frekuensi radio, persembahan seperti
bit error rate (BER) dan packet error rate (PER) perlu diperoleh. Terdapat pelbagai
cara untuk menguji satu sistem WMAN. Walau bagaimanapun, dalam projek sarjana
ini, satu kaedah unik yang menggabungkan peralatan dan perisian simulasi
dibangunkan untuk menentusahkan LNA dengan mengukur persembahan BER dan
PERnya. Perbandingan dibuat di antara keputusan persembahan yang diperoleh dari
simulasi dan dari pengujian terhadap LNA ini. Keputusan menunjukkan LNA yang
diuji berada pada taraf piawaian dan spesifikasi yang diperlukan.
vii
TABLE OF CONTENT
CHAPTER
1
TITLE
PAGE
TITLE
i
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENT
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENT
vii
LIST OF TABLES
x
LIST OF FIGURES
xi
LIST OF SYMBOLS
xiii
LIST OF ABBREVIATION
xiv
LIST OF APPENDICES
xviii
INTRODUCTION
1
1.1 Introduction
1
1.2 Project Background
2
1.3 Problem Statement
4
1.4 Objective
4
1.5 Project Scope
5
1.6 Project Contribution
6
viii
2
LITERATURE REVIEW
7
2.1 Project Overview
7
2.2 RF Receiver Architecture
9
2.2.1 Superheterodyne Receiver
2.3 Amplifier
3
11
2.3.1 Low Noise Amplifier
13
2.3.2 IF Amplifier
14
2.4 Mixer
14
2.5 Oscillator
16
2.6 Filters
18
2.6.1 Band-Select (BS) Filter
19
2.6.2 Image-Reject (IR) Filter
19
2.6.3 Channel Select (CS) Filter
20
METHODOLOGY
21
3.1 Introduction
21
3.2 Instrument and Tools
22
3.2.1 ADS 2002C
22
3.2.2 89600S Vector Signal Analyzer
23
3.2.3 E4438C ESG Vector Signal Generator
24
3.3 Procedures
3.3.1 Concept Diagram
4
10
RF RECEIVER PERFORMANCE
24
25
30
MEASUREMENTS
4.1 Introduction
30
4.2 BER Definition
31
4.3 E b/No Definition
32
4.4 PER Definition
36
ix
5
SIMULATION AND MEASUREMANT SETUP
38
5.1 Introduction
38
5.2 Setup on Hardware and Software
39
5.2.1 Setup on E4438C Vector Signal Generator
39
5.2.2 Setup on 89600S Vector Signal Analyzer
40
5.2.3 Setup on Agilent 89600 VSA software
40
5.2.4 Setup on ADS 2002C
42
5.3 Simulation Setup
6
7
42
5.3.1 Simulation 1
43
5.3.2 Simulation 2
45
5.3.3 Simulation 3
50
5.3.4 Simulation 4
52
5.3.5 Simulation 5
52
5.4 Device Under Test (DUT) Setup
55
RESULT AND ANALYSIS
56
6.1 Introduction
56
6.2 Label of Graph
56
6.3 Analysis of Result
58
CONCLUSION
65
7.1 Future Work
67
REFERENCES
68
APPENDIX A
70
APPENDIX B
72
APPENDIX C
73
APPENDIX D
74
APPENDIX E
75
x
LIST OF TABLES
TABLE NO
TITLE
PAGE
6.1
Explanation on types of performance
57
6.2
Explanation on simulations
57
6.3
Labeling of traces according to the type of performance
57
and simulation
xi
LIST OF FIGURES
FIGURE NO
TITLE
PAGE
1.1
Superheterodyne Receiver
5
2.1
Verification of specification for the RF front-end
10
subsystem
2.2
Amplifier with power gain G
12
2.3
A Generalized mixer model
15
2.4
Example of a mixer downconverter system
15
2.5
Example of a mixer upconverter system
16
2.6
The spectrum (fundamental and harmonics) of a square
18
wave with phase noise sidebands. The inset shows the
effect of the sidebands in the time domain: jitter
3.1
89600 VSA
23
3.2
Agilent E4438C ESG Vector Signal Generator
24
3.3
Simulation on RF Receiver for BER and PER
25
Performance
3.4
Simulation on RF Receiver for BER and PER
25
Performance using Captured Signal (sdf file format)
3.5
Simulation on RF Receiver (except LNA is real) for
27
BER and PER Performance
3.6
Simulation on the LNA only for BER and PER
28
performance
3.7
BER and PER measurement for LNA
28
4.1
Typical RF Communication System Receiver Block
32
Diagram
5.1
Signal from 89600S VSA is captured back to
44
xii
simulation platform and saved as SDF file
5.2
Signal from source is downloaded to E4438C ESG
48
5.3
Signal from 89600S VSA is captured back to
48
simulation platform and saved as SDF file
5.4
Signal from SDF file is brought back to ADS 2002C to
49
perform the rest of simulation
5.5
Signal enters an AWGN channel before passing
51
through BS filter and downloading to ESG 4438C
5.6
Signal from SDF file is brought back to simulation
51
platform to perform the rest of the simulation with the
exclusion of band select filter and one of the LNAs
from RF receiver block
5.7
Simulation of LNA only for overall BER and PER
53
performance
5.8
First part of simulation 5 where the signal passes
54
through an AWGN channel before being downloaded
5.9
Second part of simulation 5 where signal that captured
54
as SDF file is measured for its BER and PER
performance
6.1
Result on RF receiver for BER performance (20 MHz)
61
6.2
Result on RF receiver for BER performance (10 MHz)
61
6.3
Result on RF receiver for BER performance (5 MHz)
61
6.4
Result on RF receiver for PER performance (20 MHz)
62
6.5
Result on RF receiver for PER performance (10 MHz)
62
6.6
Result on RF receiver for PER performance (5 MHz)
62
6.7
Comparison of BER performance between simulation
63
1, simulation 2 and simulation 3
6.8
Comparison of PER performance between simulation
63
1, simulation 2 and simulation 3
6.9
Comparison of BER performance between simulation 4
64
and simulation 5
6.10
Comparison of PER performance between simulation 4
and simulation 5
64
xiii
LIST OF SYMBOLS
Eb
-
Energy-per-bit
F
-
Noise Figure
Fcenter
-
Center Frequency
FIF
-
Frequency IF
fLO
-
Frequency Local Oscillator
FRF
-
Frequency RF
G
-
Gain
Kvco
-
Tuning Constant
N
-
Noise
NBW
-
Receiver Noise Bandwidth
NO
-
Noise Density
Pin
-
Input Power
Pout
-
Output Power
R
-
Data Rate
S
-
Signal
T
-
Temperature
Tb
-
Bit Time
Vcarrier
-
Carrier Voltage
Vout
-
Output Voltage
Vtune
-
Tuning Voltage
ωosc
-
Angular Frequency
xiv
LIST OF ABBREVIATION
ADS
-
Advanced Design System
AP
-
Access Point
AWGN
-
Additive White Gaussian Noise
BER
-
Bit Error Rate
BS
-
Band Select
CDMA
-
Code Division Multiple Access
CRC
-
Cyclic Redundancy Check
CS
-
Channel Select
DC
-
Direct Current
DSL
-
Digital Subscriber Line
DSSS
-
Direct Sequence Spread Spectrum
DUT
-
Device Under Test
ESG
-
Electronic Signal Generator
EDA
-
Electronic Design Automation
FET
-
Field effect transistor
FER
-
Frame Error Rate
GPIB
-
General Purpose Interface Bus
3G
-
Third Generation
GP
-
Good Packet
IEEE
-
Institute of Electrical and Electronics Engineer
IF
-
Intermediate Frequency
IMD
-
Intermodulation Distortion
xv
IPTV
-
Internet Protocol Television
IIP3
-
Input at Third Order Intercept Point
IR
-
Image Reject
LNA
-
Low Noise Amplifier
LO
-
Local Oscillator
LOS
-
Line of Sight
MAC
-
Medium Access Control
NF
-
Noise Figure
OFDM
-
Orthogonal Frequency Division Multiplexing
PE
-
Probability of Error
PER
-
Packet Error Rate
PSDU
-
Physical Sublayer Service Data Units
QPSK
-
Quadrature Phase Shift Keying
RF
-
Radio Frequency
SDF
-
Standard Data Format
SNR
-
Signal-to Noise Ratio
TD-
-
Time Division-Direct Sequence Code Division Multiple
SCDMA
Access
UNII
-
Unlicensed National Information Structure
USB
-
Universal Serial Bus
VOIP
-
Voice over Internet Protocol
VSA
-
Vector Signal Analyzer
WEP
-
Wired Equivalent Privacy
WiMAX
-
Worldwide Interoperability for Microwave Access
Wi-FI
-
Wireless Fidelity
WLAN
-
Wireless Local Area Network
WMAN
-
Wireless Metropolitan Area Network
xvi
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
A
LOW NOISE AMPLIFIER (LNA) Datasheet
70
B
Behavioural Model for RF Tranceiver Modeling
72
C
RF Transceiver Model
73
D
Instrument Setup
74
E
RF Tranceiver Prototype
75
CHAPTER 1
INTRODUCTION
1.1
Introduction
Due to the development of communication technology, the wireless
communication system grows rapidly to compete in the world market. Nowadays,
many systems in wireless communication have been introduced for a variety of
application such as wireless internet, 3G, Bluetooth, EDGE and many more. There
are many reasons why this technology becomes important today. For areas poorly
served by wired infrastructure, wireless is a good alternative to overcome this
limitation. Wireless MAN (IEEE 802.16a) is a standard for MANs (metropolitanarea network), as opposed to the LANs (local-area networks) served by the more
well-known Wi-Fi (Wireless Fidelity). Wireless MAN is introduced to improve
wireless LAN (IEEE 802.11a) standard by providing increased bandwidth and
stronger encryption.
With coverage of areas ranging up to 30 kilometers radius, wireless MAN
enables delivery of broadband services to residential and small-to-medium-sized
business customers, and large corporations in urban, suburban and rural areas
without requiring direct line-of-sight. Wireless MAN is not introduced to replace the
2
wireless LAN. The purpose of designated is to extend the wireless LAN application
that can connect IEEE 802.11a hotspots with each other and to other parts of internet.
Orthogonal Frequency Division Multiplexing (OFDM) is adopted in wireless
MAN to support high data rate up to 70Mbps for internet access. Wireless MAN uses
of OFDM and scheduled MAC allows wireless mesh network to be more robust and
reliable.
1.2
Project Background
Wireless MAN is a standards-based wireless technology providing highspeed data and voice services in networks covering long distances and wide ranges
without the need for direct line-of-sight with a base station. Therefore, a high
complexity in the digital system part as well as very accurate signal processing in the
analog RF subsystem is required. High transmission rates within band limited radio
channels affect the growing complexity of the devices and require the following [4]:
i) very high transmission frequencies must be used
ii) sophisticated modulation and coding technologies are used to achieve a high
spectral efficiency
iii) high requirements for the RF front-end (robustness against interferer, adjacent
channels and high linearity
The additional requirements such as low power consumption and low costs
must take into consideration for system designed. System level simulators like ADS
(Advanced Design System) or MATLAB can be implemented to build an executable
specification.
3
Wireless MAN can cover a large geographical area without line of sight with
higher data rate transmission by introducing OFDM technique.
The high
transmission data rate needs high requirements for RF front end. The performance of
the RF subsystem will be tested by using ADS (Advanced Design System)
simulation tool. The signal from ADS must be captured by ESG (Electronic Signal
Generator). The system must be verified by Vector Signal Analyzer.
Recently, the performance test of RF subsystem has been done by
undergraduate student from Wireless Communication Centre (WCC) of Universiti
Teknologi Malaysia (UTM) for wireless LAN standard. IEEE 802.16a extends this
coverage while offering the features consistent with the stringent demands of
operators in a wide variety of deployment scenarios. The Wireless MAN technology
fills a critical need in the end-to-end wireless network by bridging the gap between
IEEE 802.11 wireless LANs and the wide area network.
Wireless MAN standard published [12] on 1 April 2003 for urban area
coverage wireless access addresses frequencies from 2-11GHz including licensed
and unlicensed bands. This project will be focused on unlicensed band with upper
U-NII 5.725-5.825 GHz frequency. This band is allocated for the use of indoor
links.
The RF transceiver has been designed with a selected architecture of
superheterodyne receiver and two-step transmitter. The modeled of RF transceiver
has been analyzed in the Advanced Digital System (ADS) 2002C software for
system characteristic and performance [4]. This project extends the limited benefits
offered in Wireless LAN by doing some modification in the AWGN channel for
wireless LAN. The same concept with WLAN is applied. Wireless MAN is a new
wireless internet standard. ADS 2002C software does not support the Wireless MAN
system itself. Therefore, the same wireless LAN source will be used instead of
Wireless MAN source. The IEEE 802.16a standard specifies channel size ranging
from 1.25 up to 20MHz [1] with many options in between. The unique method is
implemented in this project by doing the simulation for the multiple channel
bandwidths which the 5MHz, 10MHz and 20MHZ are chosen to see the performance
4
of overall RF receiver. The best performance of these three bandwidth range is used
as a project based for the next steps of simulation.
1.3
Problem Statement
Major problem in RF System will certainly degrade the performance of RF
system are nonlinearity and Noise.
The nonlinearity phenomena are harmonic
generation, intermodulation distortion (IMD), gain compression and spurious
response.
The noises are thermal noise, phase noise and image noise.
These
problems will affect the RF system. So, the early intention of this project is to judge
how these nonlinearities and noises will affect the RF system by measuring the BER
and PER performance.
Normally, after a RF transceiver has been designed, the system will be tested
and verified. This is to ensure the standard of the system as well as its reliability. A
powerful instrument as well as simulation software will be implemented. In this
project a unique method need to be identified to complete the task of verification.
1.4
Objective
The objective of this project is to perform simulation and measurement on a
RF receiver including the system and subsystem level analysis as well as verification
of its subsystem in wireless MAN based on IEEE 802.16a standard
5
1.5
Project Scope
Generally in radio transmitter receiver, the system is divided into two
sections, analog section and digital section. Analog section consists RF part and IF
part.
Second section is digital part.
All the baseband processing such as
demodulating, channel decoding and deinterleaving is done in this part. This project
is focused on RF part consist of main components such as low noise amplifier,
mixer, amplifier, filters (band-select filter, image-reject filter and channel select
filter), and local oscillator.
A wireless MAN superheterodyne receiver might look likes a block diagram
shown in figure 1.1. The performance for the subsystem to be studied is LNA. This
project will cover some important features and specifications of Wireless MAN that
will be focused on fixed broadband access and the concepts and techniques of LNA
including the simulation by using ADS and testing to obtain the performance of the
system. LNA will be verified for its specification and standard.
Antenna
IF Cable
Figure 1.1: Superheterodyne Receiver
In IEEE 802.16a standard, three frequency bands are available for U-NII
band. There are lower U-NII band (5.150-5.250 GHz), middle U-NII band (5.2505.350 GHz) and upper U-NII band (5.725-5.825GHz) where the maximum allowable
6
output power are 40mW (16.02dBm), 200mW (23dBm) and 800mW (29dBm)
respectively. In wireless MAN, the transceiver under study is using the upper band
for its frequency operation.
1.6
Project Contribution
As mentioned earlier, this project is extended from undergraduate student by
doing some modification of Wireless LAN source. So, the same concept as WLAN
will be implemented in this project. This is continuative work of the previous
researcher which is to design, simulate and measure a RF transceiver operating at
5.725-5.825 GHz. This is then lead to the important purpose purpose of the project
which is to verify the real subsystem of the designed RF transceiver.
In short, at the end of the project, the overall system level performance of the
RF receiver will be obtained by simulation. The performance of BER and PER with
minimum noise must be achieved as a result of this WMAN system. After the
simulation has been done, the real subsystem of the RF receiver –low noise amplifier
will be verified for its specification and standard. So, the results in simulation will
then become the reference for the verification process.
Hence, a method of
verification is identified and this will become a very useful way for the other similar
design and development of such transceiver or other typical devices.
CHAPTER 2
LITERATURE REVIEW
2.1
Project Overview
Before we discuss more detail about wireless MAN technology, let us
compare this system with conventional wireless internet, wireless LAN technology.
Wireless LAN can provide the area of coverage less than 10km which the application
is between buildings to building.
In Wireless LAN, MAC uses contention access, so the device competes with
all other devices on the network for attention on random basis to pass data through.
Data sent and requested by devices closer to the network access point (AP)
constantly interrupt and even crowd out data sent and received by devices farther
away from the AP. Also, the more devices seeking access to the network, the lower
the quality of the signal. This means if we access the Internet with a Wireless router
hooked up to local cable TV company’s broadband service, for instance, and the
company also offers VoIP (Voice over Internet Protocol) phone service, the more
people who are online, watching TV, and talking on the phone in the access area, the
weaker the signal and the slower of the connection speed. VOIP & IPTV are
difficult to maintain for large number of users.
8
Encryption is not enabled in wireless LAN. This means the Access Point
typically default to an open (encryption-free) mode. Wireless LAN networks can be
monitored and used to read and copy data (including personal information)
transmitted over the network when no encryption is used.
The most common
wireless encryption standard, Wired Equivalent Privacy or WEP, has been shown to
be breakable even when correctly configured. Therefore, no security provided in
wireless LAN system.
Wireless LAN uses OFDM technique to support internet access data rate up
to 54Mbps. The required speeds defined in IEEE 802.11a are 6, 12 and 24 Mbps
with optional speeds up to 54Mbps [4].
Wireless MAN offers significant improvements over Wireless LAN, and
among the more important is the specification making 802.16a a system that uses a
scheduling MAC (Media Access Control).
The subscriber station only has to
compete once (for initial entry into the network). After that, it is allocated a time slot
by the base station. The time slot can enlarge and constrict, but it remains assigned
to the subscriber station meaning that other subscribers are not supposed to use it but
take their turn. Unlike 802.11a, this scheduling algorithm is stable under overload
and over-subscription. It is also much more bandwidth efficient. The scheduling
algorithm also allows the base station to control Quality of Service by balancing the
assignments among the needs of the subscriber stations.
The difference is distance and area of coverage. Wireless MAN provides up
to 30km (18.641 miles) in every direction for fixed stations and 3 to 10 miles for
mobile stations. It allows the connectivity between users without direct line of sight.
Wireless MAN will provide fixed, nomadic, portable and, eventually, mobile
wireless broadband connectivity [1].
Wireless can be used for a number of
applications including last mile broadband access, hotspot and cellular backhaul for
carrier infrastructure, and high speed enterprise connectivity.
9
Wireless MAN support the frequency range 2-11 GHz, of which most parts
are already unlicensed internationally and only very few still require domestic
licenses. The 802.16a specification improves upon many of limitation of 802.11a
standard by providing increased bandwidth and stronger encryption.
Wireless MAN specified the theoretical data rate up to 70Mbps. However, in
the real world test, the maximum data rate is between 50kbps and 2Mbps. It can be
achieved by providing OFDM technique as well as wireless LAN modulation
technique [1].
2.2
RF Receiver Architecture
It is the receiver that ultimately determines the performance of the wireless
link. Given a particular transmitter power, which is limited by the regulations, the
range of the link will depend on the sensivity of the receiver, which is not legally
constrainted. Of course, not all application require optimization of range, as some
are meant to operate at short distances of several meters or even centimeters. In
these cases simplicity, size and cost are the primary considerations [4].
The purpose of a radio frequency (RF) receiver is to process incoming energy
into useful information, adding a minimum distortion. How well receiver performs
its purpose is a function of the system design, its internal circuitry, and its working
environment. The acceptable amount and type of introduced distortion vary with the
application.
10
2.2.1
Superheterodyne Receiver
All modern digital radios have two sections. One section is analog part and
another part is digital. Among their other services, radio must provide frequency
conversion between antenna and the digital circuitry. In wireless MAN, the signal at
the antenna which operates in the GHz is converted to the IF by tuning the local
oscillator.
Ever since its invention by Edwin Armstrong in 1917 [5], the
superheterodyne architecture has dominated the design of practical radio.
This
architecture uses two frequency conversion steps: first the received frequency is
converted to an intermediate frequency, and after some amplification and filtering,
the IF is converted to baseband. Note that the final down-conversion is performed
twice, to produce I (in-phase) and Q (Quadrature) output. This is necessary because
the down-conversion operation cannot tell the difference between frequency above
and below the carrier, so they end up on top of each other when the carrier is
converted to zero frequency. The information about the sidebands can be preserved
by doing two frequency conversions using different phases of the carrier.
Figure 2.1: Verification of specification for the RF front-end subsystem
11
The receiver architecture that was designed for wireless MAN is shown in the
figure 2.1. The first component of the receiver, a band select (BS) filter attenuates
out of band signal received by the antenna. Therefore, the BS filter only selects the
RF band of 5.725-5.825 GHz. Two low-noise amplifiers (LNA) boost the desired
signal signal level while minimally adding the noise of the RF signal. Band Select
(BS) filter or Image Reject (IR) filter is designed to pass the RF band of interest and
reject signal from nearby bands. Image Reject (IR) filter is used to overcome the
image frequency problem by eliminates the image before down conversion. The
mixer downconverts the RF signal to lower IF by mixing the RF signal with a LO
signal.
Then, the channel select (CS) filter attenuates all unwanted frequency
components generated by the mixer and any signal in the adjacent channels. To
boost the IF signal to the desired level, the cascaded IF amplifier is used so that I and
Q demodulator (in Indoor Unit) can detect and downconvert the IF signal to the I and
Q signal [4]. Finally, the signals are further processed by OFDM baseband in order
to get actual baseband data.
2.3
Amplifier
Amplifiers are active component. The component consists of a solid state
device (transistor, FET, IMPATT, Gunn, etc) that generates a negative resistance
when it is properly biased. A positive resistance dissipates RF power and introduces
losses. In contrast, a negative resistance generates RF power from the DC bias
supplied to the active solid-state device.
The key properties of an amplifier are gain, bandwidth, noise, distortion and
power. Amplifier is a component that provides power gain to the input signal to the
amplifier. Gain is defined as the ratio of the size of the output to that of the input.
As shown in Fig 2.2, Pin is the input power and Pout is the output power. The power
gain is defined as
12
P
G  out
Pin
G (in dB) 10 log
(2.1)
Pout
Pin
(2.2)
G
Pin
Pout
Figure 2.2: Amplifier with power gain G
Amplifier can be cascaded to provide higher gain. For example, for two
amplifier with gain G1 and G2 in cascade, the total gain equal G 1G2. An amplifier
should not oscillate in the operating bandwidth. The stability of an amplifier is its
resistance to oscillation. An unconditionally stable amplifier will not oscillate under
any passive termination of input and output circuits.
Amplifiers must respond rapidly to be part of the RF chain; bandwidth is less
significant in the rest of the radio. Noise is what an amplifier should not add to the
signal but does. Distortion usually represents the main limitation on amplifier output
power.
13
2.3.1
Low Noise Amplifier (LNA)
The performance of a superheterodyne receiver is dominated by certain key
elements. The LNA is the main source of excess noise and typically determine the
sensitivity of the radio. This subsystem must increase the signal level while hardly
increasing the signal-to-noise ratio (SNR) of the incoming signal. These two tasks
are not always easily achieved, mainly because noise impedance matching and input
impedance matching are not always obtained for the same source impedance.
At the input of a receiver the signal level may be very low. In addition to
amplify the signal, we must exercise special care to minimize the unavoidable noise
contribution of the amplifier. The source termination (i.e. antenna) may vary during
operation, and the amplifier must function in spites of the changes. The output may
see highly reactive termination outside the passband, presented by filters that follow
the amplifier, and it must be stable for all those terminations.
The LNA is the first gain stage in the receiver path. Therefore, according to
Friss’s formula, the noise figure (NF) of this stage directly adds to that of the overall
system. The noise figure is a measure of the degradation of the signal-to-noise ratio
at the output of the LNA compared to that at the input. Another main performance
parameter of the LNA is its gain. The signal should be amplified as much as possible
with hardly lowering the SNR, while also maintaining linearity.
The last
performance parameter is therefore represented by IP3. Obviously, the obtained NF,
gain and IP3 should be within specification with minimum power consumption.
14
2.3.2
IF Amplifier
After the signal is raised well above the noise level, gain becomes a more
important factor than noise. Also, since the amplifier may face a wide range of
termination at various frequencies, RF stability is another key consideration. These
intermediate level amplifiers are designed for maximum gain, with simultaneously
matched input and output ports.
The IF amplifier is usually a high-gain stage. Its intercept point must be high
if it directly follows the mixer. If it follows one stage of IF filtering, the intercept
point requirements can be relaxed, because the IF filter offers some protection
against high-level, off-channel signals.
2.4
Mixer
A mixer combines an incoming transmitted or received signal with a nearly
pure sinusoid from a local Oscillator (LO). In the receive path, a low noise amplifier
is normally followed by one or more mixer to translate the RF frequency into a
baseband LO signal. The multiplication of these two distinctly different signals
produces harmonics, due to non-idealities, influencing the information signals. Up
conversion mixers are used in a similar way in the transmit path. Mixers are always
non-linear devices, but a good mixer is highly linear.
As shown in figure 2.3, a mixer is a three port device, which in addition to the
input (RF) signal port and output (IF) signal port, uses a third local oscillator (LO)
port to drive the mixer.
This driving action, sometimes called switching or
modulation because of its impact on the mixer device, is highly nonlinear and causes
15
either the device conductance or transconductance to switch between two states, one
with a low transconductance and the other with a high transconductance.
fRF
Three-terminal
nonlinear device
RF input
(small signal)
mfRF+nfLO
fLO
Local oscillator
(large signal)
Figure 2.3: A Generalized mixer model
Antenna
Mixer/buffer
fRF
RF LNA
Amplifier
fLO
IF filter
fIF = fRF- fLO
(low-side conversion)
= fLO- fRF
(high-side conversion)
LO1
Spectral
amplitude
LO
Image
IF
RF(signal)
Sum
Spectral
amplitude
Low-side LO
High-side LO
LO
RF(signal)
Image
IF
Figure 2.4: Example of a mixer down-converter system
IF
16
Antenna
Mixer
RF power
amplifier
fIF
RF filter
fIF = fRF- fLO
(low-side conversion)
= fLO- fRF
(high-side conversion
fLO
fRF
LO1
LO
RF(signal)
Difference
IF
Spectral
amplitude
Spectral
amplitude
Sum mixer
Difference mixer
LO
RF(signal)
Sum
IF
Figure 2.5: Example of a mixer up-converter system
2.5
Oscillator
Oscillators are intriguing blocks.
An oscillator is one of the very few
building blocks in a transceiver that has a built-in timing reference. When the power
supply is switch on, DC power is somehow translated into a periodic signal, which
forms the heartbeat of many systems. Ongoing world-wild research contributed to
the design of low phase noise oscillators, which make it more and more a science
rather than an art.
Advances in IC technology, especially in passives, have
simplified complete oscillator integration.
17
The basic function of an oscillator is to generate a periodic signal with certain
properties. An ideal oscillator generates a signal which only has wanted properties.
The output of an ideal harmonic oscillator with angular frequency ωosc [rad/s] and
peak amplitude Vcarrier [V] can be written as
Vout (t) = Vcarrier cos (ωosc t)
(2.1)
In the frequency domain, this is equivalent to a discrete spectral line with
height Vcarrier at angular frequency ωosc. This means that all carrier power is
located in an infinitely small bandwidth around ωosc. A tunable ideal oscillator can
be represented by:
Vout (t) = Vcarrier cos (2π(kvco Vtune +fcenter) t)
(2.2)
In (2.2) the initial phase of Vout(t) is assumed to be zero. Tuning voltage Vtune [V]
controls the frequency, and tuning canstant kvco [Hz/V] determines the tuning slope.
fcenter is the oscillation frequency with a zero tuning voltage.
In practice, anything that will change due to non-idealities. The oscillator
and its properties are no exception. An oscillator will never have the axact center
frequency required, due to the processing spread in the IC process. Some additional
tuning range will therefore always be required on top of the required range. Noise
from the oscillator circuitry and externally generated noise currupt the spectral purity
of an oscillator signal. This means that the carrier power is now distributed in a
finite bandwidth around ωosc and its harmonics. Figure 2.6 shows the fundamental
and two harmonics of a square wave.
18
Vcarrier
Phase noise
sidebands
Time domain
(V)
P(dBm)
t(s)
∆t jitter
ωosc
3ωosc
5ωosc
ω(rad/s)
Figure 2.6: The spectrum (fundamental and harmonics) of a square wave with phase
noise sidebands. The inset shows the effect of the sidebands in the time domain:jitter
2.6
Filters
Filters play a key role in rejecting undesired signals in a radio. Filters are
two-port networks used to control the frequency response in an RF or microwave
system by allowing transmission at frequencies within the passband of the filter, and
attenuation within the stopband of the filter. Common filter responses include lowpass, high-pass, band-pass and band-stop. Filters are indispensable components in
wireless systems, used in receivers for rejecting signals outside the operating band,
attenuating undesired mixer products, and for setting the IF bandwidth of the
receiver.
19
2.6.1
Band-Select (BS) Filter
BS-filter usually called the pre-selector and has three basic functions:
i)
to limit the bandwidth of spectrum reaching the RF amplifier and mixer to
minimize IM distortion
ii)
to attenuates receiver spurious responses (image and
1
IF are most
2
important)
iii)
to suppress local oscillator energy originating in the receiver
Attenuation of direct IF frequency pick up may also be a concern in receivers
with high first IF frequency. BS-filter may be a highly selective, cavity tuned filter,
cascaded with a low-pass filter to attenuate resonances at odd multiples of center
frequency (a property of all such filters).
RF amplifier noise figure, gain and intercept point are set by receiver
performance requirements. High reverse isolation is important to attenuate local
oscillator energy and to isolate BS-filter and CS-filter from each other, so that overall
selectivity is not destroyed. Low reverse isolation in the RF amplifier will cause the
filter to interact, with guaranteed degradation of RF selectivity at some frequencies.
2.6.2
Image-Reject (IR) Filter
The function of the IR-filter is to attenuate receiver spurious response
frequencies, attenuate direct IF frequency pickup, attenuate noise at the image
frequency originating in the low noise amplifier, which could potentially degrade
20
mixer second order intercept point. Depending on its bandwidth, IR filters can also
suppress local oscillator energy leaking back into the antenna. It is important for this
filter not to have any return responses at high frequency that may leak through the
system. This filter is frequently called the image-reject filter because it typically
rejects image noise by about 20dB. If good receiver sensitivity is not a requirement,
the LNA and image-reject filter may not be required.
2.6.3
Channel-Select (CS) Filter
The CS filter is the first IF filter which protects its following stages from
close-in IM signals, provides adjacent channel selectivity, and attenuates the second
image. Frequently, the second-image requirement is more stringent than the adjacent
channel selectivity requirement and determines the number of poles required to
obtain the required second image selectivity. The equivalent noise bandwidth of the
IF chain is an important receiver property, because it determines how much noise
reaches the detector, and it determines the modulation bandwidth that can be
received. Low group delay of the IF filters is particularly important for digital
communication. Group delay compensation is hardware and software can be used to
overcome the undesirable effects of group delay distortion, provided the group delay
is highly repeatable from unit to unit.
CHAPTER 3
METHODOLOGY
3.1
Introduction
Basically, this chapter discusses about various aspects concerning
methodology applied in simulation for BER and PER performance in this project.
As mentioned earlier, the methodology is almost the same as WLAN where the
objective is to verify the real LNA that has been built by the previous researcher. So,
the same instruments and tools will be applied. The methodology is very important
as it provides the steps to carry out a certain study in more structural way. It also
clarifies what are the hurdles in performing data collection as well as measures taken
to overcome them. By applying an effective and simple methodology, hopefully it
will bring positive results. Procedures, instrument and tools and concept diagram
will be covered in detail in this chapter.
22
3.1
Instrument and Tools
Instrument and tools are used when it involves difficult or complicated tasks
ahead for data collection. It is also used when time consuming is a matter of
concern. A good tool is able to provide accurate and needed readings in the given
environment where information is limited.
In this project, a powerful software
named Advanced Design System (ADS 2002C) is used for simulation of overall
system performance as well as subsystem performance. Besides, 89600S Vector
Signal Analyzer (VSA) and E4438C ESG Electronics Signal Generator will be
utilized in the interfacing between simulation software and real hardware. In other
words, these two instruments enable the sending of wireless MAN signal to the real
hardware that under test as well as capturing the signal back to the simulation
platform.
3.2.1
ADS 2002C
ADS offers electronic design automation (EDA) software for high frequency
system and circuit design. ADS is the industry leader in high frequency design. It
supports system and RF design engineers developing all types of RF designs, from
simple to most complex, from RF/microwave modules to integrate MMICs for
communications and aerospace/defense applications.
applied for the simulation process.
This powerful software is
23
3.2.2
89600S Vector Signal Analyzer
With today’s emerging broadband communication systems, the Agilent
Technologies 89600 Series VSAs are the indispensable tool for research, product
development, manufacturing and field-testing. Along with wide IF bandwidth (3639 MHz), the 89600 VSAs offer traditional RF spectrum display, baseband (I/Q)
analysis, signal capture memory, RF and IF triggering, a wide variety of analog and
digital demodulator, and an extensive set of time, frequency and modulation analysis
tools. These capabilities make 89600 VSAs ideal for evaluating narrow band and
broadband digital communication signals. Analyze a wide variety of standard and
non-standard formats.
Twenty-three standard-signal preset cover GSM, GSM
(EDGE), CDMAOne, CDMA2000, W-CDMA, 802.11a, 802.11b, 1xEV-D0, TDSCDMA, 802.16a and more. For emerging standards, the 89600 Series offers 24
digital demodulators which variable center frequency, symbol rate, filter type and
alpha/BT. A user-adjustable adaptive equalizer is also provided [4].
Figure 3.1: 89600S VSA
24
3.2.3
E4438C ESG Vector Signal Generator
The Agilent E4438C ESG Vector Signal Generator meets the needs of
engineers who are designing and developing the next generation of wireless
communication systems and is well suited for production test environment. An
assortment of a standard-based receiver and component test personalities for 3G and
emerging communications format are available to simplify the signal configuration
process.
The E4438C ESG Vector Signal Generator’s improved performance,
extended frequency range, increased memory for waveform playback and storage,
and application-specific personalities make it the clear choice for development and
manufacturing from the component to the system level.
Figure 3.2: Agilent E4438C ESG Vector Signal Generator
3.3
Procedures
The five fundamental steps of carrying out the project are:
i)
simulation on RF Receiver for BER and PER performance
ii)
simulation on RF Receiver for BER and PER performance using captured
signal (sdf file format)
iii)
simulation on RF Receiver (except LNA is real) for BER and PER
performance
25
iv)
simulation on the LNA only for BER and PER performance
v)
BER and PER measurement for LNA
3.3.1
Concept Diagram
ADS 2002C
Figure 3.3: Simulation on RF Receiver for BER and PER Performance
Initially, the overall RF receiver performance is measured and it is all done in
the simulation platform which is in the simulation software – ADS 2002C.
This is
the first step of simulation where the various frequency ranges are simulated in this
stage. The results from BER and PER measurements will become the reference for
the following simulation and measurement. Only one bandwidth which performs a
best performance will be used as a project based.
26
ADS 2002C
E4438C ESG
Vector Signal
Generator
89600S
Vector Signal
Analyzer
Figure 3.4: Simulation on RF Receiver for BER and PER Performance using
Captured Signal (sdf file format)
For the second step, the reliability of this two instrument which are ESG and
VSA need to be assured. Besides, is it also has to make sure the process of bridging
out as well as sending back the signal to simulation platform will cause minimum
errors. So, the step (ii) is the confirmation step before the testing can be done on the
real hardware. Before the simulated signal source goes through the RF receiver
block in simulation platform, it is brought out from simulation and without passing
through any real hardware under test, it is quickly being captured by VSA and thus
brought back to simulation platform without further delay. This is done so that the
raw signal is not attenuated by the environment of the nature.
27
This step is needed so that it can confirm that the necessary settings on the
two instruments are properly done and it only causes minimum errors which are then
can be neglected.
ADS 2002C
E4438C ESG
Vector Signal
Generator
DUT
89600S Vector
Signal Analyzer
Figure 3.5: Simulation on RF Receiver (except LNA is real) for BER and PER
Performance
After that, the real hardware testing and BER as well as PER measurement on
its begin. This is clearly shown in figure above. The wireless MAN signal is
generated and passed through band select filter. After passing through the filter, the
signal is brought out of simulation platform and sent to the ESG. It then passes
through the real hardware under test, LNA. Without further delay, it is then captured
by the ESG and VSA will bring the signal back to the simulation platform to go
through the following subsystems which are the same as previous steps. Only after
that the overall system performances are measured.
28
The results obtained will then be compared with the previous result. After
comparison, the real hardware will be verified whether it is up to the standard
specified or not.
ADS 2002C
LNA
Figure 3.6: Simulation on the LNA only for BER and PER performance
ADS 2002C
E4438C ESG
Vector Signal
Generator
DUT
89600S Vector
Signal Analyzer
Figure 3.7: BER and PER measurement for LNA
29
In addition to previous 3 steps, there are other two steps of simulation. These
fourth and fifth steps are just to reconfirm the previous 3 steps by doing simulation
and measurement on a subsystem only. Hence, the results will show exactly the
comparison performance between the model in simulation and the real hardware.
The only LNA is simulated by sending RF signal of Wireless MAN to the
unit. After passing through the LNA model, the signal will be received by a wireless
MAN receiver and demodulated so as to obtain the performance results. The carrier
frequency at the receiver for this case will be different with the receiver of previous
steps. This is because the signal is not passed through the mixer and LO for downconversion process.
Finally, the signal will be brought out of simulation platform. It will pass
through a real hardware of LNA which is under test before it is brought back to the
simulation for BER and PER performances measurement.
These five steps are the core activities throughout the study. The purpose of
showing these five steps is to give a clear picture of the overall performance study.
The simulation and measurement setup will be explained in chapter 5.
CHAPTER 4
RF RECEIVER PERFORMANCE MEASUREMENTS
4.1
Introduction
The performance of the system when it is corrupted by noise is the primary
considerations in the design of wireless MAN system. The performance measure for
digital system is the probability of error of the output signal and also called bit error
rate (BER).
Although there are several measurements used to test a receiver’s
performance, all of them measure the same quantity under different conditions. The
measured quantity is the BER. BER is the probability of a bit received in error that
is measured at the receiver output.
Different wireless give different names to various BER measurements such as
Minimum Input Power Sensitivity, Minimum Input Level Sensitivity, Adjacent
Channel Rejection, Adjacent Channel Selectivity, Reference Sensitivity Level,
Dynamic Range, Blocking, Intermod [4].
As mentioned earlier, all the above
measurements are BER measurements are BER measurements under different
conditions.
These different conditions include additive white gaussian noise
(AWGN), modulated interference signals, and CW interference signals.
The
interference signals can be in band and/or out of band. Typically, the standards
31
specify that the BER should not exceed a certain value for certain power levels of the
wanted and interfering signals, and a certain frequency offset (between the desired
signal’s channel frequency and the frequency of the interfering signals).
4.2
BER Definition
Bit error ratio (BER) is the number of bit received in error, divided by the
total number of bits received.
BER 
Number of bits received in error
Number of bits received
(4.1)
Bit error ratio (BER) is the most fundamental measure of system
performance. That is, it is a measure of how well bits are transferred end-to-end.
While this performance is affected by factors such as signal-to-noise and distortion,
ultimately it is the ability to receive information error free that defines the quality of
the link.
Quantization errors reduce BER performance, through incorrect or
ambiguous reconstruction of the digital waveform. This is also described by a
probability function that defines the likelihood that a digital transition or edge
detection error will occur. These errors are primarily a function of the accuracy of
the digital-to-analog and analog-to-digital conversion processes, and are related to
the number of bits used at these points in the circuit. The accuracy of the analog
modulation/demodulation process and the effects of filtering on signal and noise
bandwidth also affect quantization errors.
32
4.3
Eb/No Definition
Bit error rate (BER) and Frame and packet error rate (FER/PER) are typically
reported with respect to Eb/No. This note defines Eb/No and relates it to signal to
noise ratio (SNR). Distinction is made of local and system Eb/No. Eb /No is the
energy-per-bit divided by noise-density ratio at the receiver input.
S A,
NOA
Tx
Rx
DUT 1
A NF1, G1
SB,
NOB
B
SC ,
NOC
Rx DUT
2 NF 2,
G2
Detector
DSP and
BER Meas
C
R
Rx Antenna Ta
Figure 4.1: Typical RF Communication System Receiver Block Diagram
The initial two blocks represent the transmitted signal and the propagation
channel between the antenna at the transmitter and receiver. The transmitted signal
contains data with bit time, T b and bit rate R bits/sec. The propagation channel
includes significant attenuation and propagation effects (phase, amplitude, multi-path
fading, etc.). The explanation and of each block in figure 4.1 is simplified below:
-
A is the receiver antenna output
-
B is a mid point within the receiver system
-
C is the receiver system pre-detection point
-
Rx DUT 1 is the receiver RF front-end and contains any lossy lines before the
receiver and receiver front-end amplifiers, filters and mixers.
For this
discussion it is defined with gain in dB (G1 ) and noise figure in dB (NF1)
-
Rx DUT 2 is the receiver backend and contains content before detection and
includes amplifiers, filters, matched filter and sampler. For this discussion, it
is defined with gain in dB (G2) and noise figure in dB (NF2)
-
The BER is then measured, typically with suitable DSP algorithms
33
At each A, B and C point in the system, there is a measureable value for the
signal (SA, S B, S C) and noise density (NOA, NOB, NOC), where the signal is in Watts
(W) and noise density in Watts/Hz (W/Hz).
In this system, the received desired signal has additive thermal noise
contributions from the propagation path available at the receiver antenna output and
from the receiver noise figures.
Other noise contributors are ignored such as
interfering signals and nonlinear intermodulation products. Thermal noise receiver
antenna output is typically defined in terms of noise temperature in Kelvin and is call
as T A. Note that 290 K (16.85oC) corresponds to noise power density of -173.975
dBm/Hz value.
The receiver antenna output noise power density is:
NOA = k Ta , where k is Boltzmann’s constant.
Receiver noise figures can also be represented in terms of noise temperature in
Kelvin: T = 290 (F-1); F1 = 10 (NF/10)
The RF DUT 1 and 2 have associated noise temperatures at T1 and T2 respectively.
T1 = 290 (F 1-1); F1 = 10 (NF1/10)
T2 = 290 (F2-1); F2 = 10 (NF2/10)
T1 represents the equivalent noise temperature due to RF DUT 1 defined at
the input of RF DUT 1 and has associated noise power density: kT1. This results in
definition for NOB as:
NOB = G1 (kTa ) + G1 (kT 1) = G1 k (Ta +T1)
T1 represents the equivalent noise temperature due to RF DUT 2 defined at
the input of RF DUT 2 and has associated noise power density: kT2. This results in
definition for NOC as:
34
NOC = G1G2 (kTA) + G1G2 (kT1) + G 2 (kT2) = G1 G2 k (T a+T1+T2/G2)
SNR is related to Eb/No in the following way:
SNR = S/N = (Eb/Tb)/NONBW = EbR/NoNBW = Eb /No × R/NBW
Where,
SNR = signal-to-noise ratio (unitless)
S = signal power (W)
N= noise power (W)
Eb = bit energy (W/sec)
Tb = bit time (sec)
NBW = receiver noise bandwidth (Hz)
No noise power density = N/NBW (W/Hz)
R = data rate = 1/Tb (1/sec)
Eb/No = Eb over No (unitless)
To provide a signal-to-noise figure that is independent on the receiver noise
bandwidth, the signal-to-noise density is typically used.
S/No = (Eb/Tb)/No = EbR/No = Eb/No × R
Thus, we now see the relationship between Eb/No and S/No and S/N.
Eb/No = S/N × NBW/R = S/No/R
35
S/No and Eb/No values may be considered as local or system values. Local
values are specific to receiver system point where they are evaluated (point A, B, or
C in figure 4.1); system values are independent of the receiver system point where
they evaluated.
Local values of S/No and Eb/No are directly measurable at each point in the
system and are typically the preferred S/No and Eb/No values used by RF/analog
designers.
At points A, B and C, the local S/No values are:
SA/NOA = SA/(kTa)
SB/NOB = (S AG1)/(k(Ta+T1)G1) = S1 /(k(Ta +T1))
SC/NOC = (S AG1G2)/(G1G2 k(T a+T1)+G 2 kT2) = SA/(k (Ta+T1+T2 /G1)
System values of Eb/No anf S/No are directly measurable only at the
predetection system point (point C in figure 4.1). These are the system values
because they characterize the overall system performance. The system values are
typically the preferred S/N o and E b/No values used by System/DSP designers. At
point C, the local Eb/No and S/No values are the same as the system Eb/No and S/No
values.
In all cases,
Eb/No = S/N o/R
10 log (Eb /No) = 10 log (S) – 10 log (No) – 10 log (R)
10 log (No) = 10 log (S) – 10 log (R) – 10 log (Eb /No)
(4.2)
36
In general,
Noise Density (dB) = Signal Power (dB) – [Data Rate] – [Eb/No]
One way to lower the spectral noise density is to reduce the bandwidth, but it
is limited by the bandwidth required to transmit the desired bit rate (Nyquist
Criteria). The point to keep in mind, as with all wireless data Radio access Network
(RAN), is that the higher the data rate the more susceptible the signal will be to
interference and fading, and ultimately the shorter the range, unless output power is
increased [7].
Energy per bit can also be increased by using higher power
transmission, but interference with other systems can limit that option. A lower bit
rate increases the energy per bit, but we will lose capacity. Ultimately, optimizing
Eb/No is a balancing act among these factors.
4.4
PER Definition
For WMAN, packet error rate (PER) testing is more common than BER,
because it is easier to implement. The 802.16a standard defines the use of 32 bit
CRC that will detect erroneous packets. In IEEE 802.16 WMAN if any of the bits
are in error the Cyclic Redundancy Check (CRC) will detect the error and flag the
packet as being bad. This is what is referred to here as packet error. The receiver
must then request that the packet be retransmitted. This result is a decrease in the
WMAN throughput and an increase in network latency.
The probability of a packet in error is also often referred to as the packet error
rate. The WMAN packet Packet Error event, PE, means that the WLAN packet has
at least one bit error. It is often referred to a symbol error since not all WMANs use
binary modulation. For example, the 2 Mbps DSSS WMAN uses QPSK modulation,
37
and hence there are two bits per symbol. Of course, if there is a symbol error then
there is at least one bit error.
In fact, there are a lot of ways to evaluate and derive formula for PER. The
formula for PER will be different according to the modulation and type of system as
well as kinds of assumption made. However, formula derivation and evaluation is
not the main event in this project. The discussion here is just briefly and let the
reader has an overall idea of what PER is all about.
In general, the probability of WMAN packet error is one minus the
probability of a good WMAN packet.
P (PE) = 1- P (GP)
Commonly, PER is expressed in percentage (%). So, the formula will be as
follows.
PER = P (PE) × 100%
(4.3)
CHAPTER 5
SIMULATION AND MEASUREMENT SETUP
5.1
Introduction
This chapter mainly presents the necessary simulation and measurement setup
for the purpose of RF receiver verification. It includes the interfacing of two
instruments which are E4438C ESG Electronic Signal Generator and 89600 Vector
Signal Analyser with the simulation software – Advanced Design System (ADS
2002C). It also shows the real simulation platform appearance which is form the
simulation software – ADS 2002C. Besides, there is another softwaere called VSA
software which needed to configure as it is the linking bridge between 89600 VSA
and simulation software – ADS 2002C.
This chapter will explain the setup of the hardware and software tools. The
simulation and measurement setup are also covered in this chapter.
39
5.2
Setup on Hardware and Software
There are critical setups needed prior to simulation. This can be divided into
setup on hardware (E4438C ESG Vector Signal Generator and 89600S Vector Signal
Analyzer) and setup on software (VSA software and ADS 2002C). These are
important steps to ensure the signal can be captured properly and the desired result
can be obtained.
5.2.1
Setup on E4438C Vector Signal Generator
Basically, the role of the signal generator (E4438C ESG) used in this project
is to generate the WLAN signal instead of WMAN signal (as generated by RF signal
component in ADS) that has been used in the simulation for BER measurement. The
instrument is directly connected to the computer that having the ADS 2002C
simulation software by a General Purpose Interface Bus (GPIB) connector at
instrument and a Universal Serial Bus (USB) connector at computer. It permits the
transter of data from the software to the instrument.
The instrument will capture the data as an arbitrary waveform file format
(WFM1). The waveform segments reside in volatile memory, so when the
instrument is turned off then the signal captured will lost.
At the front panel of the instrument, there is one output called RF output
connector, this female type-N connector is the output for RF signals. It will be
connected directly to the vector signal analyzer or indirectly to the VSA where there
ia a device under test (DUT) in between these two instruments. The RF output must
40
shows the the RF ON when the signal is ready to be transmitted to the VSA. It also
can be switched by using the RF output button.
At the rear panel of the instrument, there is a connector called EVENT 1
connector. This female BNC connector outputs a pulse that can be used to trigger
the start of a data pattern, frame, or timeslot [4]. For this project, it will be used to
trigger the signal transmitted so that it can be captured by VSA without any delay.
Delay of the signal will cause inaccuracy in the measurement of BER performance.
The port will be connected to the external trigger of VSA.
Apart from that, there is also an important step which is the waiting time time
for the instrument to warm up. When it is just powered up, the display will shows
message indicating that the instrument is still cold. According to the manual, the
waiting time for the instrument is about 20 minutes to make sure that the instrument
gives the best accuracy in generating signal wanted.
5.2.2
Setup on 89600S Vector Signal Analyzer
There is not much setup in the instrument itself as it is a mostly controlled by
the accompanied VSA software. The only thing is that it has to be calibrated
continuously throughout the measurement and capturing of the signal whenever it is
required to do so. This is to ensure that the accuracy will not be affected by
temperature and other factors. The VSA will be calibrated according to the specific
calibration files that contain in the VSA software.
Besides, it is also have to make sure that the data sent is not overloaded by
seeing at the overload indicator on the front panel of the VSA. When the indicator is
41
lighting up, this indicates that the signal is overloaded and necessary step has to be
taken which is to adjust the level of signal to an appropriate level. This will be
discussed in the next subsection.
The external trigger connector will be linked to the EVENT 1 of E4438C
ESG for the purpose of trigerring signals so that the waveform will look static and
start transmit from 0 sec when signal recording takes its part.
5.2.3
Setup on Agilent 89600 VSA software
There are quite a number steps need to be done before the optimum results
are able to obtain. Firstly, the level of signals needs to be set correctly according to
the current level of signal. If the input range is set too low (more sensitive than
necessary), the analyzer’s ADC circuitry introduces distortion into the measurement.
But if the input range is set too high (less sensitive than necessary), there may be a
loss of dynamic range due to additional noise (in some cases), the increase in the
noise floor may even obscure low level frequency components. If OV1 or OV2
appears in the trace indicator, the ADC circuitry is being overdriven. In this case,
there is a need to change to a less sensitive range or reduce the input signal level.
Apart from that, so that a synchronous signal obtained, the input RF signal is
needed to be externally triggered. This is done when the external trigger connector is
connected to the EVEVT 1 of ESG 4438C and at the same time in the VSA software
itself, the option to set the the external trigger is selected following the path, Menu
Path: Input > Trigger > Type box.
42
After all the necessary settings have been done, it is needed to save this
setting as a file for reference. This file having the extension of .set will be used later
on the simulation in ADS 2002C.
5.2.4
Setup on ADS 2002C
For the simulation software ADS 2002C, it is important that the link between
the server of ADS 2002C and current simulation station is well established. This is
to ensure the continuous access of the simulation software. The server will permit
the usage of the simulation software by limiting the number of current users.
5.3
Simulation Setup
The initial stage of the project, the simulation is done totally by ADS 2002C
and it is not depend on the instruments to obtain the result. However, for the second
part, signal then is sent to the ESG 4438C so that it can model the signal used in the
first stage as identical as possible in order to test and verify the device under (DUT).
For this project, it is more focusing on testing the BER performance of RF receiver
front-end component which is the low noise amplifier (LNA). After that, signal
passing through the LNA need to be brought back into the simulation environment
for the rest of the simulation process and BER performance measurement.
43
5.3.1
Simulation 1
The first simulation is done so that the overall front-end receiver BER
performance is obtained. It can become guidance for other simulations later on. In
this project, three difference bandwidths are chosen to be simulated in this stage.
The optional bandwidths that have been chosen is this simulation is 5, 10 and 20
MHz. This is important procedure to see the different between WMAN and WLAN.
The best performance of this bandwidth is used for the next simulations
The WLAN 802.16a RF source is set to have the carrier frequency of 5.765
GHz. The data rate is 54 Mbps. Although the target of data rate in WMAN is higher
than 54Mbps, it cannot be set higher because the highest data rate can be supported
in ADS 2002C is 54Mbps. The modulator output power is fixed at -50dBm. The
modulation technique used is based on the IEEE 802.16a standard. The length is a
octet number of PSDU which is fixed at 1000.
Then the signal is sent through an adaptive White Gaussian Noise (AWGN)
channel model which is followed to provide proper channel noise. The signal is then
received by the front-end receiver model for downconverting purpose. The
intermediate frequency (IF) is set 815 MHz. Later on, the signal is demodulated by a
demodulator. The received data will then be compared with the original signal from
the WLAN 802.11a RF source for BER as well as PER performance.
The simulation will be done by fixing the transmitted power and raising the
bit energy to noise power density ratio in the AWGN channel gradually referring to
the formula 4.2. The BER and PER sink will measure the frame number 100 for
each value of Eb/No from 10 to 20.
Delay
D1
WLAN
WLAN
RF
Source
WLAN
1
2
Down
Converter
WLAN_80211a_RF
SignalSource
AddNDensit y
Noise
Modulator & OFDM Baseband
AWGN Channel
rx1_sub_timed
X1
Front End Reciever
BERPER
Rec eiver
With Freq Sync
WLAN_80211a_BERPER
BERPER
WLAN_80211a_RF_RxFSync
Receiver
Demodulator & OFDM Baseband
Figure 5.1: Simulation of RF receiver for overall BER and PER performance
5.3.2
Simulation 2
In this simulation, the signal from WLAN 802.11a RF source is sent to the
ESG 4438C. The output of the signal from the source is in timed format. However,
the signal type that can be read in the instrument is I & Q signal. So, in order to send
the signal to the instrument, the signal needs to be converted from timed format to
complex format then from complex format, it will be converted again to the needed
I&Q signal.
There is a very important step in sending the signal to the instrument that is a
number of samples that required. This needs to be set very carefully to keep the
signal sequence contents as an integer number of bursts. To do this, the following
steps need to be taken [4].
a)
Calculate the number of symbols per burst for WMAN data,
N = int ((16+8*Length+6)/NDBPS)
LL = 16+8*Length+6-NDBPS*N
KK = if (LL==0) then 0 else 1 endif
NSYM = N+KK
Where NDBPS is the number of data bits per OFDM symbol which is
predetermined and Length is the octet number of PSDU (physical layer convergence
procedure service data units). If there is a remainder after the division by NDBPS,
the number of symbols will be rounded up to the next interger.
b)
Total number of samples per burst,
SymbolRate = (20*2^(Order-6)) MHz
NSPB = ((20us+4us*NSYM)*SymbolRate+Idle)-1
c)
Decide start and stop. The waveform sent to the ESG can start from 0.
Burst = frame
ESG_Stop = (frame+1)*NSPB
46
The number of samples collected (Stop-Start+1) must be in the range of 16 to
1,048,576 and should be an even number of samples. The last sample is discarded if
Stop-Start+1 is odd.
Apart from that, for the type of data that are transmitted from WLAN 802.11a
source, it is needed to be set to zero. In other words, the probability of the data
consisting of all ‘0’ bits is 1.
After that, the signal is sent to ESG 4438C from ADS 2002C simulation
platform. The RF output of signal generator is now directly connected to the input of
89600VSA. So, at ADS 2002C again, the signal is then captured back to simulation
platform as Standard Data Format (SDF) file. The file will later be used to do the
rest of the simulation. The time step of simulation 1 is 12.5 ns. The time step is
equal to the sample clock or symbol rate in figure 5.2. The time step of simulation 2
is also 12.5 ns. However, the time step will be changed when the signal reaches
VSA. This is due to the different test instruments’ interpolation process [11]. The
different time step in simulation will cause error in measuring the BER and PER
performance. So, in order to recover the time step to its original value, a re-sampling
process is needed.
The time step of VSA can be adjusted in ADS 2002C. However, it can not be
set at as small as 12.5 ns. So, as obtain 12.5 ns for the time step in simulation,
initially the time step is set as 25 ns in VSA_89600_source component as in figure
5.3. After that, it is up-sampled to 12.5 ns.
After the signal is captured back as SDF file, the file is used as a source of
signal as shown in figure 5.4. However, the power of the signal is attenuated by the
cable that links the signal generator and vector signal analyzer. It is estimated that
the loss due to the cable is 4.5 dB. When the signal is brought back to the simulation
platform, it is needed to be restored to the original level of power by an amplifier
47
with gain of 4.5 dB. With the provided AWGN channel, the BER and PER
performances will be simulated. It is similar to simulation 1 at this stage.
WLAN
RF
Source
WLAN_80211 a_RF
SignalSource
Time dToCx
T1
Figure 5.2
CxToRect
C1
ESG_E4438C_Sin k
E1
Signal from source is downloaded to E4438C ESG
VSA
VSA_89600_Source
V1
USampleRF
U2
SDFWrite
S4
Signal from VSA saved as .sdf file
Figure 5.3
Signal from 89600S VSA is captured back to
49
Const
C1
Delay
D1
1
GainRF
G1
AddNDensity
Noise
AWGN Channel
WLAN
2
Down
Converter
WLAN
BERPER
rx1_sub_timed
X2
Front End Reciever
Rec eiver
With Freq Sy nc
WLAN_80211a_RF_RxFSync
Receiver
WLAN_80211a_BERPER
BERPER
Demodul ator & OFDM Baseband
SDFRead
S1
Modulator & OFDM Baseband
Figure 5.4
Signal from SDF file is brought back to ADS 2002C to perform the rest of simulation
5.3.3
Simulation 3
The overall simulation is shown in figure 5.5 and figure 5.6. In this
simulation, the LNA prototype will be tested and verified. To begin with the first
step, the data bits from the signal source must be set to ‘0’. Besides, there must be an
AWGN channel before the signal enters the RF receiver. LNA is the target
subsystem for testing but before it there is a subsystem called band select filter. So,
this filter will be just a model in this simulation. The signal generated from the
WLAN 802.111a RF source will pass through this filter before it is sent to E4438C
ESG. This is clearly shown in figure 5.5.
The real hardware of LNA is now connected between E4438C ESG and
89600S VSA. The step of capturing signal into SDF file will be the same as Figure
5.3.
After the signal is sent to ESG 4438C, passed through LNA and finally
captured by VSA as SDF file, it will then be used as a signal source to execute the
rest of the simulation. In this simulation, again the signal will be attenuated by the
cable and the step power restoration is needed. The RF receiver block will now
exclude the band select filter and one of the LNAs. Ultimately, the BER and PER
performance will be obtained for this real hardware.
WLAN
RF
Source
WLAN_80211a_RF
Signal Source
AddNDens ity
Nois e
Modul ator & OFDM Baseban d
Figure 5.5
BPF_ButterworthT imed
Filter2
AWGN Channel
T imedT oCx
T1
CxToRect
C1
ESG_E4438C_Sink
E1
Band Select Filter
Signal enters an AWGN channel before passing through BS filter and downloading to ESG 4438C
Const
C1
Delay
D1
WLAN
WLAN
1
GainRF
G1
BERPER
2
Down
Converter
rx1_sub_timed
X1
Front End Reciever
Receiver
With Freq Sy nc
WLAN_80211a_BERPER
BERPER
WLAN_80211a_RF_RxFSync
Receiver
Demodul ator & OFDM Baseband
SDFRead
S1
Modulator & OFDM Baseband
Figure 5.6
Signal from SDF file is brought back to simulation platform to perform the rest of the simulation with the exclusion of band select
filter and one of the LNAs from RF receiver block
5.3.4
Simulation 4
In this simulation, only one subsystem which is LNA will be measured for its
BER and PER performance. The signal from the source is passed through LNA and
is simulated by using ADS 2002C software only. This is done to double confirm the
results obtained from simulation 1, 2 and 3.
5.3.5
Simulation 5
The simulation is done so as to compare with the simulation 4. Basically, the
sending and capturing of signal are the same as simulation 3. The only difference is
that is to test only a subsystem that is LNA. The real hardware of LNA still will be
connected between E4438C ESG and 89600S VSA. Also, the power restoration will
be done in the second part of the simulation. This is to compensate the power loss
due to the attenuation of signal inside the cable.
Delay
D1
WLAN
WLAN
RF
Source
WLAN
BERPER
Receiver
WLAN_80211a_RF
SignalSource
Modulator & OFDM Baseband
Figure 5.7
WLAN_80211a_BERPER
BERPER
WLAN_80211a_RF_RxFSync
Receiver
With Freq Sync
AddNDensity
Noise
AWGN Channel
Gain RF
LNA2
Component: LNA
Demodulator & OFDM Baseband
Company: Hittite
Product ID: HMC320MS8G
Note: Medium Power
VSET=3V, Idd=25mA
Simulation of LNA only for overall BER and PER performance
54
WLAN
RF
Source
WLAN_80211a_RF
SignalSource
AddND ensity
Noise
Modulator & OFDM Baseband
Figure 5.8
TimedToCx
T1
Cx ToRect
C1
ESG_E4438C_Sink
E1
AWGN Channel
First part of simulation 5 where the signal passes through an AWGN channel before being downloaded to E4438C ESG
Const
C1
Delay
D1
WLAN
WLAN
BERPER
Receiver
With Freq Sync
GainRF
G1
WLAN_80211a_RF_RxFSync
Receiver
WLAN_80211a_BERPER
BERPER
SDFRead
S1
Modulator & OFDM Baseband
Figure 5.9
Demodulator & OFDM Baseband
Second part of simulation 5 where signal that captured as SDF file is measured for its BER and PER performance
5.4
Device Under Test (DUT) Setup
The device under test in this project is a low noise amplifier that can be
operated at 5-6 GHz. It is manufactured by Hittite. The specification of the LNA
will be attached in the appendix A.
To power up the unit, it needs 3V for Vdd and the value of Idd that permitted
will be in the range of 7mA till 40mA. So, a power supply is needed for the
functioning of this device
CHAPTER 6
RESULTS AND ANALYSIS
6.1
Introduction
In this chapter, the results obtained from simulation and measurement will be
discussed thoroughly. Before that, the result obtained is shown in the graph. This is
to make sure that the information from the results is clear.
6.2
Label of Graph
The results of simulation and measurement are in the form of graph. Below is the
meaning of each label. It consists of two parts which are trace type and simulation
step.
There are two types of trace in 4 graphs of results which are Performance P
and Performance Q. Besides, there are 5 simulations. There are Simulation 1,
Simulation 2, Simulation 3, Simulation 4 and Simulation 5.
57
Table 6.1 and Table 6.2 explain about the meanings of each label which
consists of alphabets and Arabic numbers. In order to easily identify the type of
trace and simulation that are referring to, Table 6.3 provides the better understanding
of the labeling of traces.
Table 6.1: Explanation on types of performance
Performance
Explanation
P
Bit Error Rate (BER)
Q
Packet Error Rate (PER)
Table 6.2: Explanation on simulations
Simulation
Explanation
1
Simulation on RF Receiver for BER and PER performance
2
Simulation on RF Receiver for BER and pER performance using
captured signal (sdf file format)
3
Simulation on RF Receiver (except LNA is real) for BER and PER
performance
4
Simulation on the LNA only for BER and PER performance
5
BER and PER measurement for LNA
Table 6.3: Labeling of traces according to the type of performance and simulation
Performance/Simulation
1
2
3
4
5
P
P1
P2
P3
P4
P5
Q
Q1
Q2
Q3
Q4
Q5
58
6.3
Analysis of Result
From the simulation 1, the result shows the performance of the different
bandwidth range of WMAN for BER and PER. In this case the 5 MHz, 10 MHz and
20 MHz is chosen for the measurement. In the graphs in figure 6.1, 6.2 and 6.3, the
result shows that the performance for 5 MHz bandwidth performs the best
performance in term of BER and PER. This result satisfied the formula to calculate
the Eb/No in chapter 4 sections 4.3 where the bandwidth is changes and the other
parameters remain constant. However, WMAN requires the larger bandwidth to go
to the higher data rate so, that it can covers a large coverage of area without LOS.
As mentioned earlier, the larger bandwidth, the noise also becomes higher. So, to
optimize the bandwidth with the minimum noise, the bandwidth of 10 MHz is used
as a project based.
From Figure 6.7, P2 and P3 are very near to P1. P3 is in the range of P1 and
P2. P1 is the simulation result on RF receiver for BER performance. Simulation 1 is
the modified from previous undergraduate student for WLAN. The modification is
in term of type of performance that is intended to look at. Previously, it is used to
simulate the minimum sensitivity level of the receiver. In this project, the sam
AWGN channel is used back. After some modification, Simulation 1 will perform
simulation on RF receiver for its BER performance. Hence, it becomes the reference
to the other two traces, P1 and P3.
P2 is the simulation results on RF receiver for BER performance using
captured signal. As the signal passing from E4438C ESG through cables and
connectors and before captured by using VSA, there is signal attenuation by the
cable and connector. Although the signal is restored back to the original power level
in the simulation later on, however, the power restoration may also increase the noise
as well at the same time. So, still there are differences in P1 and P2 in term of BER
performance. Another possible cause is that the E4438C ESG can not totally imitate
59
the signal pattern used in the simulation and generate the exact signal in terms of
amplitude and phase [4].
P3 is the simulation with the real LNA connected in between ESG and VSA.
So, the model of LNA is replaced by the real unit device under test. The result
shows that P2 is close to P1. It proves that the LNA prototype is within the
specification set in the model and up to standard stated. However, P3 is better than
P2 and approximately the same as P1. P3 has lower noise level that the power is
restored in the simulation.
PER performance also shows the same result as BER performance. It is
shown in Figure 6.5. The explanation is the same as above. Q1, Q2 and Q3 are very
close to each other resulting the LNA is within the specification and up to the
standard that specified.
From Figure 6.9, there are simulation 4 and 5. These steps are to double
confirming the finding in simulation 1, 2 and 3. The simulation 4 and 5 are to
measure the BER and PER performance of the LNA without other subsystems of RF
receiver. The LNA model used in simulation 4 is taken from simulation 1. Again, it
is proven that, the results of simulation 4 and 5 show the LNA is within the
specification and standard. P4 and P5 are very close to each other notifying that
BER performance of these two simulations is almost the same. Besides, from the
trace Q4 and Q5, PER performance of the two traces are approximately the same.
As a result, LNA prototype has successfully been verified. After all, the
LNA is proved to be within the specification and standard. This unique method of
verifying subsystem of RF receiver is found to be very useful. This is particularly
helpful when the only one subsystem need to be tested and verified where the other
components have not been fabricated.
60
As the summary of this solution that it creates simulation models from
hardware measurement. It can minimize the design time and expense. Apart from
that, it accelerates verification testing. We can find and fix issues earlier in the
design process. It also reduces the system integration risks. Also, it simplifies the
transition to prototype.
61
Result on RF receive r for BER performance (20MHz)
1
BER
1E-1
1E-2
1E-3
1E-4
10
12
14
16
18
20
EbNo
Figure 6.1: Result on RF receiver for BER performance (20 MHz)
Result on RF receiver for BER performance (10MHz)
1
BER
1E-1
1E-2
1E-3
1E-4
10
12
14
16
18
20
EbNo
Figure 6.2: Result on RF receiver for BER performance (10 MHz)
Result on RF receive r for BER performance (5MHz)
1
BER
1E-1
1E-2
1E-3
1E-4
10
12
14
16
18
20
EbNo
Figure 6.3: Result on RF receiver for BER performance (5 MHz)
62
Result on RF receive r for PER performance (20MHz)
1.0
PER
0.8
0.6
0.4
0.2
0.0
10
12
14
16
18
20
EbNo
Figure 6.4: Result on RF receiver for PER performance (20 MHz)
Result on RF receive r for PER performance (10MHz)
1.0
PER
0.8
0.6
0.4
0.2
0.0
10
12
14
16
18
20
EbNo
Figure 6.5: Result on RF receiver for PER performance (10 MHz)
Result on RF receive r for PER performance (5MHz)
1.0
PER
0.8
0.6
0.4
0.2
0.0
10
12
14
16
18
20
EbNo
Figure 6.6: Result on RF receiver for PER performance (5 MHz)
63
BER vs Eb/No
1.E+00
Trace2
Trace3
BER
1.E-01
Trace1
1.E-02
1.E-03
1.E-04
10
11
12
13
14
15
16
17
18
19
20
EbNo
Figure 6.7: Comparison of BER performance between simulation 1, simulation 2 and
simulation 3
PER vs Eb/No
1
PER
0.8
Trace Q2
0.6
Trace Q1
0.4
Trace Q3
0.2
0
10
12
14
16
18
20
EbNo
Figure 6.8: Comparison of PER performance between simulation 1, simulation 2 and
simulation 3
64
BER vs Eb/No
1.E+00
Trace P5
1.E-01
BER
Trace P4
1.E-02
1.E-03
1.E-04
10
11
12
13
14
15
16
17
18
19
20
EbNo
Figure 6.9: Comparison of BER performance between simulation 4 and simulation 5
PER vs Eb/No
1
PER
0.8
0.6
TraceQ4
0.4
TraceQ5
0.2
0
10
12
14
16
18
20
EbNo
Figure 6.10: Comparison of PER performance between simulation 4 and simulation 5
CHAPTER 7
CONCLUSION
In early discussion of this project, Wireless MAN technology is the wireless
broadband access as an alternative for cable and digital subscriber line (DSL). The
specification and standard of Wireless MAN have been studied. The research is
about designing RF receiver consist of subsystems such as band select filter, low
noise amplifier, image rejection filter, mixer, oscillator, channel select filter and
amplifier. However, this project is focused on the performance and testing of low
noise amplifier.
The ADS 2002C is adapted to the Wireless MAN system for simulation task.
The manual and simulation tool have been studied to ensure this project performs the
result successfully. This can be learned quickly by viewing and understanding those
examples from the library files that are provided by the software itself.
The
modification made are mainly referred to published papers entitled “Broadband
Wireless Access with WiMax/802.16: Current Performane Benchmarks and Future
Potential” by Arunabha Ghosh and David R. Wolter, SBC Laboratories Inc. Jeffrey
G. Andrews and Runha Chen, The University of Texas at Austin. Besides, the book
entitled 3G Wireless with WiMAX and Wi-FI are also referred as a main reference.
66
The system then will be verified and measured by using 89600 VSA and
E4438C ESG Vector Signal Generator to ensure the performance of the system is
within the specification and standard. The instruments must operate properly to
obtain the accurate result.
The correct ways of using the instruments are well
investigated by reading the hard copy of application manuals as well as soft copy
from the instrument supplier website.
Troubleshooting as well as trial and error is done throughout the simulations
so that the desired outcomes are achieved. Besides, consultant from application
engineers of Agilent Technologies is required along the study and this leads to the
success of implementing the connected solution method.
The result for simulation 1 shows the performance with the difference
frequency range. From the result, we can conclude that, if we need the better
performance in terms of BER and PER we must suffer for the small bandwidth. That
is the fact in wireless communication environment that we have to accept to design
the wireless MAN system. From the result of simulation 2, 3, 4 and 5, we can say
that this experiment is successful. From the comparison, we can conclude that the
device under test which is the low noise amplifier that operating at 5.0-6.0 GHz is
within the specification and standard.
Thus, a method to verify subsystems of WMAN is identified and this method
combines the usage of simulation software and operation instruments. It also
accelerates verification testing and minimizes the design time and expense by
applying the simple unique method.
67
7.1
Future Work
The performance of the subsystems not only LNA become most important in
wireless MAN system. The measurement and simulation setups can become a useful
reference in the process of testing and verifying on the other subsystems. The same
method can be applied to verify the other subsystems itself. This can be done
through proper modification and adjustment in the simulations.
Troubleshooting is the hard task in this project. In future, a short course
conducted by application engineers from Agilent Technologies should be taught to
operate the E4438C ESG and 89600S VSA properly. The knowledge from the
course is very useful and will accelerate the simulation process.
REFERENCES
[1]
Ghosh, A, Wolter, D.R, Andrews, J.G, Chen, R. Broadband Wireless Access
with WiMax/802.16: Current Performane Benchmarks and Future Potential.
In: IEEE. IEEE. Communications Magazine. USA: IEEE. 2005
[2]
Wang, C.S, Li, W.C and Wang, C.K. A Multi-Band Multi-Standard RF
Front-End for IEEE 802.16a and IEEE 802111.a/b/g Applications. 2006
[3]
Ommic. An Ultra Low Noise, High Linearity Amplifier. Microwave
Journal. 2006. 49 (4): 136
[4]
Eddy Seng Yan Tuck. Low Noise Amplifier Performance Study for
Wireless LAN Based on IEEE 802.11a Standard. Bachelor Degree.
Universiti Teknologi Malaysia: 2006
[5]
Dobkin, D.M. RF Engineering for Wireless Networks. USA: Elsvier Inc.
2005
[6]
Couch, L.W,II. Digital and Analog Communication Systems. 7th ed. New
Jersey: Pearson Education. 2007
[7]
Smith, C and Meyer, J. 3G Wireless with WiMAX and Wi-Fi. USA: Mc
Graw-Hills. 2005
[8]
Gu, Q. RF System Design of Tranceivers Wireless Communications. USA:
Springer Science + Business Media, LLC. 2005
69
[9]
Kenington, P.B. High-Linearity RF Amplifier Design. USA: Artech House
Inc. 2000
[10]
Egan, W.F. Practical RF System Design. New York: IEEE. 2003
[11]
Nor Azwan bin Shairi. Radio Frequency Tranceiver Design for Wireless
Local Area Network Bridge System at 5725-5825 MHz. Master Thesis.
Universiti Teknologi Malaysia: 2005
[12]
Institute of Electrical and Electronic Engineers. IEEE Standard for Local
and Metropolitan Area Networks. New York, SS95079. 2003
[13]
Marks, R. B. IEEE Standard 802.16 for Global Broadband Wireless Access.
The Future of Wireless. October, 14 2003. Geneva, Switzerland: IEEE.
2003. 1-8
[14]
Agilent Technologies. Low Noise Amplifier for 5.125-5.325 GHz and 5.7255.825 GHz Using the ATF-55143 Low Noise PHEMT. US: Application Note
1285. 2002
70
APPENDIX A
LOW NOISE AMPLIFIER (LNA) Datasheet
71
72
APPENDIX B
Behavioural Model for RF Tranceiver Modeling
Table B.1: Behavioral models under Agilent Ptolemy Simulator
Digital Signal Processing: Agilent Ptolemy Simulator
Components
Behavioral Model Name
Amplifier
Gain RF
Model
GainRF
G1
Gain=1
NoiseFigure=0
GCType=none
TOIout=3 W
dBc1out=1 W
Mixer
Mixer RF
MixerRF
M3
NoiseFigure=0
Type=RF plus LO
RfRej=-200
ImRej=-200
LoRej=-200
LComp="0 0 0"
Filter
BPF_ButterworthTimed
BPF_ButterworthTimed
B1
Loss=0.0
FCenter=1000000.0 Hz
PassBandwidth=2000.0 Hz
PassAtten=3.
StopBandwidth=1200000 Hz
StopAtten=50.
Local Oscillator
N_Tones
N_Tones
N2
Frequency1=999999 Hz
Power1=.010 W
Phase1=0.0
RandomPhase=No
PhaseNoiseData=
PN_Type=Random PN
73
APPENDIX C
RF Transceiver Model
Port
P1
BPF_ButterworthTimed
Num=1 B1
Loss=1.4
FCenter=5775 MHz
PassBandwidth=100 MHz
PassAtten=3.
StopBandwidth=725 MHz
StopAtten=20
BPF_ButterworthTimed
B3
Loss=9.55
FCenter=815 MHz
PassBandwidth=20 MHz
PassAtten=3.
StopBandwidth=40 MHz
StopAtten=30
GainRF
LNA1
Gain=dbpolar(12,0)
NoiseFigure=2.5
GCType=TOI+dBc1
TOIout=dbmtow(20)
dBc1out=dbmtow(9)
PSat=1 W
GainRF
LNA2
Gain=dbpolar(12,0)
NoiseFigure=2.5
GCType=TOI+dBc1
TOIout=dbmtow(20)
dBc1out=dbmtow(9)
GainRF
Amp1
Gain=dbpolar(16,0)
NoiseFig ure=6.5
GCType=TOI+dBc1
TOIout=dbmtow(27)
dBc1out=dbmtow(14)
BPF_ButterworthTimed
B2
Loss=4.5
FCenter=5775 MHz
PassBandwidth=120 MHz
PassAtten=3.
StopBandwidth=320 MHz
StopAtten=20
GainRF
Amp2
Gain=dbpolar(16,0)
NoiseFig ure=6.5
GCType=TOI+dBc1
TOIout=dbmtow(27)
dBc1out=dbmtow(14)
MixerRF
M2
NoiseFigure=6.5
Type=RF minus LO
RfRej=-200
ImRej=-200
LoRej=-200
LComp="13 -6.5 0"
GainRF
Amp3
Gain=dbpolar(20,0)
NoiseFigure=3.5
GCType=TOI+dBc1
TOIout=dbmtow(33)
dBc1out=dbmtow(20)
N_Tones
N1
Frequency1=(FCarrier-IF) MHz
Power1=dbmtow(13)
Phase1=0.0
RandomPhase=No
PN_Type=Random PN
Port
P2
Num=2
(a)
Port
P1
BPF_ButterworthTimed
Num=1 BPF1
Loss=3.6
FCenter=815 MHz
PassBandwidth=20 MHz
PassAtten=3.
StopBandwidth=40 MHz
StopAtten=10
MixerRF
M2
NoiseFigure=6.5
T ype=RF plus LO
RfRej=-200
ImRej=-200
LoRej=-200
LComp="13 -6.5 0"
BPF_ButterworthT imed
Filter1
Loss=1.4
FCenter=5775 MHz
PassBandwidth=100 MHz
PassAtten=3.
StopBandwidth=725 MHz
StopAtten=20
GainRF
Amp4
Gain=dbpolar(20,0)
NoiseFigure=6
GCT ype=T OI+dBc1
TOIout=dbmtow(43)
dBc1out=dbmtow(30)
GainRF
Amp2
Gain=dbpolar(20,0)
NoiseFigure=6
GCT ype=T OI+dBc1
TOIout=dbmtow(43)
dBc1out=dbmtow(30)
Port
P2
BPF_ButterworthT imed
Num=2
Filter2
Loss=1.4
FCenter=5775 MHz
PassBandwidth=100 MHz
PassAtten=3.
StopBandwidth=725 MHz
StopAtten=20
N_Tones
N1
Frequency1=(FCarrier-IF) MHz
Power1=dbmtow(13)
Phase1=0.0
RandomPhase=No
PN_T ype=Random PN
(b)
Figure C.1: RF transceiver model under Agilent Ptolemy simulator (a) RF receiver
model (b) RF transmitter model
74
APPENDIX D
Instrument Setup
Figure D.1: E4438 ESG Electronic Signal Generator and 89600S Vector Signal
Analyzer Setup
Figure D.2: Device Under Test Setup
75
APPENDIX E
RF Tranceiver Prototype
Figure E.1: RF transmitter prototype
LNA1
LNA2
Figure E.2: RF receiver prototype