UNIVERSITA’ DEGLI STUDI GUGLIELMO MARCONI
Master of Science in Data Science &
Artificial Intelligence (AI)
Mobile&
Wireless Networks – Evolution from 1980s to 5G and Beyond.
Academic Advisor Candidate
DHANABAL T AZRIN
ABDUL MAJID
Lithan Academy Rapheal
Azrin
MCSLT00084
UNIVERSITA’ DEGLI STUDI GUGLIELMO MARCONI
ACADEMIC YEAR
2025
Table of
Contents
Section |
Page |
Chapter 1 Introduction
& Research Methodology |
1 |
·
Research & Discussion on Possible Topics |
5 |
·
Brainstorming |
7 |
Chapter 2 - Literature
Review |
9 |
·
Thesis
Preface & Introduction |
11 |
·
Candidate Background |
12 |
Chapter 3 - Thesis Abstract |
13 |
·
Thesis Statement |
14 |
·
Dedication and Thanks |
15 |
Chapter 4 - Historical Timeline (1G to 5G and early 6G research) |
22 |
Chapter 5 - Spectrum
& Regulation (ASEAN/ANZ focus, 900 MHz, mmWave, etc.) |
32 |
·
Annex - AI & ML for
Propagator Prediction |
38 |
Chapter 6 - Modulation,
Coding, Multiple Access |
39 |
Chapter 7 - Network Topologies & Hierarchies |
44 |
Chapter 8 - Standards
by Generation (1G–5G, WiFi, WiMAX, LoRaWAN) |
51 |
Chapter 9-
Specialized Systems & Use Cases |
58 |
Chapter 10 - Core
Networks (MSC to 5GC with AI/ML enhancements) |
65 |
Chapter 11 - Security & Authentication |
71 |
Chapter 12 - Performance,
QoS, KPI Engineering (with Data Science) |
76 |
Chapter 13 - Deployment, Optimization, Testing (AI-driven SON, ML
analytics) |
81 |
Chapter 14 - Comparative
Case Studies (ASEAN & ANZ) |
86 |
Chapter 15 - 5G Advanced & Road to 6G |
91 |
Chapter 16 - Conclusion |
96 |
Chapter 17 - References & Bibliography |
101 |
Chapter
1 Introduction
This paper is to
demonstrate the competence level of the Postgraduate Masters Degree by
Coursework candidate, Mr Azrin Bin Abdul Majid aka RAPHEAL AZRIN,
to the competency level for the course on Data Science and Artificial
Intelligence, in the academic year of 2025.
The task assigned by the
teaching staff is to gather information, determine a sub-deviation of the
topics being tabled, and carefully expand and research into the subject
matters, in a miniature brief of 12,000 worded paper , with graphics et al.
A Long list of topics and
their Research Methodology were proposed.
Added to the ask is that,
the usage of AI ,DATASCIENCE, Cybersecurity et al, as the candidate is
attempting a Masters of Science in Data Science and AI under
Lithan-UniMarconi collaboration and executive Masters program as part of the
Skill future Career Transition Program, with 90% Funded by the Singapore
Government vis Skillfuture Credits + CoPayment by the student on the PGDip
segment.
The Objective of the
program is to ensure the student can demonstrate the ability to apply and use
case of the knowledge and skills learnt thru the SCTP Module , Data Science
Modelling, Deep Learning, AML & Methods, Data Science Principles, Python for
DS, DMV Data Modelling & Visualization, Generative AI & Project
Management.
Deliberations of the
modelling and reasoning brings us to the below mentioned. Following are the
proposed topics tabled by the senate as appropriate for the cohort for Data
Science, Security Management and Cloud Administration.
Generative and Agentic AI
Agent services were used in the research to expedite results and research,
cross referencing against the generative bias of agents such as ChatGPT5,
CoPilot365, Gemini Enterprise and DeepSeek alongside other
agentic workflow systems in place. CoPilot was also activated in
Office365 to assist in writing and editing of the paper. All AI Systems have
TRUTH MODE manually coded into the prompts, so no false information were
produced and fact checked on all counts.
However, the candidate believes these three cohorts could be
integrated into the overall master's program, allowing their skillsets to
intersect and complement each other.
Research
and Discussion on Possible Thesis Topics
Looking back at the
subject matter, these are the best options proposed by the cohort and teaching
staff for attempt, based on current capabilities, educational background and
course outline:
- AI-Driven
Cybersecurity in Mobile & Wireless Systems
How AI/ML enhances anomaly detection, intrusion prevention, and resilience in wireless networks. - Data Science for
Distributed Databases and Secure Data Management
Using predictive analytics, ML, and automated database management for secure, large-scale distributed data systems. - Intelligent
Error-Correcting Codes and Cryptography
Application of AI to optimize coding algorithms (e.g., convolutional codes, Reed-Solomon) and cryptographic schemes for secure communication.
Brainstorming
3 possible topics
After much deliberation,
the team members further refined and suggested the following topics for attempt:
- AI-Powered Intrusion
Detection Systems for Next-Generation Wireless Networks
- Data Science
Approaches to Secure and Efficient Distributed Database Management
- Machine Learning
Optimization of Error-Correcting Codes and Cryptographic Algorithms for
Cybersecurity
Thesis
Statements on the 3 possible topics
As candidate will be
attempting a PhD Styled thesis, these thesis statements would be used to
propose to the senate for the purpose of the attempted subjects.
Draft Statements for
MS-PhD level Abstract and synopsis :
- AI-Powered Intrusion
Detection Systems for Next-Generation Wireless Networks
This thesis investigates the integration of machine learning models into intrusion detection systems (IDS) for 5G/6G wireless infrastructures, aiming to enhance anomaly detection accuracy and reduce false positives, thereby strengthening cybersecurity in mobile environments. - Data Science
Approaches to Secure and Efficient Distributed Database Management
This research explores how advanced data science techniques, including predictive analytics and automated anomaly detection, can optimize performance, reliability, and security in distributed database systems handling sensitive and large-scale data. - Machine Learning
Optimization of Error-Correcting Codes and Cryptographic Algorithms for
Cybersecurity
This thesis develops AI-enhanced optimization models for coding algorithms and cryptographic schemes, with the goal of improving error resilience, throughput, and resistance against cyber threats in high-speed network communications.
END OF CHAPTER 1
Chapter
2 Literature Review
Each of the proposed
topics were presented with a set of Literature Review and Directional input as
to the how the research would be laid out.
Topic 1: AI-Powered
Intrusion Detection in Wireless Networks
- Foundations: Intrusion Detection
Systems (IDS), anomaly vs. signature-based detection.
- Current Research: ML in IDS (e.g.,
Random Forest, CNN, RNN for network traffic).
- Gap: High false positive
rates, adaptability in mobile/wireless contexts.
- Key Authors/Journals: IEEE Communications
Surveys, ACM Computing Surveys, Elsevier Computers & Security.
Topic 2: Data Science in
Distributed Databases
- Foundations: CAP theorem,
distributed DB management (MongoDB, Cassandra, etc.).
- Current Research: AI for query
optimization, automated fault detection.
- Gap: Balancing
performance, security, and reliability in real-time.
- Key Authors/Journals: VLDB Journal, IEEE
Transactions on Knowledge and Data Engineering.
Topic 3: AI-Optimized
Error-Correcting Codes & Cryptography
- Foundations: Cyclic codes,
convolutional codes, Reed-Solomon, block ciphers.
- Current Research: AI for decoding
optimization, cryptographic key prediction prevention.
- Gap: Integrating ML with
coding/crypto without introducing vulnerabilities.
- Key Authors/Journals: IEEE Transactions
on Information Theory, Journal of Cryptology.
However, the preliminary
subject that I will be attempting will be mobile and wireless networks
system as a primary subject.
Advisor also recommended
to keep the wording to 12,000 words in context at the MSc level, and expand
should I dedicate to have this paper be expanded further at the PhD level later
on in the year.
The take on TOPIC 1
is good, but given the density of the research, it would be good to do it at
the PhD Level, once this paper has been published on Google Scholar.
Background
of candidate
About Rapheal Azrin
My Profile as a Regional
IM/IT Digitalization & Technology Manager at Shell Petroleum
provides me a vast knowledge of know-how in the field of telecommunications and
I felt I should showcase my git into this paper.
Candidate is also an ICANN
IANA Working Stakeholder member, collaborating for RFC and IEEE technical
papers.
Candidate also has more
than 30 years of experience in the relevant field, specializing in Web Hosting,
Telecommunications and Intellectual Properties.
Something noteworthy is
that the candidate an IEEE Chartered Fellow in the field of WDRP Domain Names
and Dispute Arbitration, in the aspect of copyright & Intellectual Property
infringement.
Public information can be
gathered across search engines, Crunchbase and Linkedin.
https://www.linkedin.com/in/azrinmajid - Linkedin
Profile
https://icannwiki.org/Raphael_Azrin - ICANN WIKI
https://www.crunchbase.com/person/rapheal-azrin - Crunchbase
He is also an advocate for
crypto & digital assets in using blockchain as an encoding technology such
as Blake17,Blake19 shuffle et al.
END OF CHAPTER 2
Chapter 3 - Abstract
This thesis
presents a comprehensive, engineering-focused analysis of mobile and wireless
radio systems, charting their evolution from analog cellular networks of the
1980s through the era of 5G New Radio and into the initial phases of 6G
research. Emphasis is placed on the technical lineage of air
interfaces—progressing from FDMA and analog FM to TDMA, CDMA, OFDMA, and
massive MIMO—and situates these innovations within the broader contexts of
spectrum allocation, propagation fundamentals, and regulatory policy frameworks
across ASEAN, APAC, Europe, and ANZ.
Topological
perspectives are integrated, encompassing macro, micro, PICO, FEMTO, LPWAN, ad
hoc, and satellite constellation architectures. The study details the
progression of core network technologies, from circuit-switched Mobile
Switching Centers (MSCs) to 5G Service-Based Architectures (SBA) and the advent
of AI-native cores. Security and authentication mechanisms receive special
focus, tracing the shift from legacy analog protocols to modern SIM-based
authentication, mutual AKA, eSIM provisioning, SUCI identifier protection, and
quantum-safe, forward-looking approaches.
Comparative
case studies of deployments in Singapore, Malaysia, Japan, and Europe
illuminate how local geography, regulation, and socio-economic conditions have
shaped divergent paths of technology adoption. Special systems—including TETRA
for mission-critical voice, LoRaWAN for IoT, satellite backhaul (LEO/MEO/GEO),
WiMAX, Wi-Fi, Bluetooth, ZigBee, and UWB—are examined for their technical and
operational trade-offs. An addendum develops the Smart Nation case study,
highlighting Singapore’s PSA Tuas Mega Port, autonomous vehicles, and urban IoT
ecosystems as precursors for advanced research into smart infrastructures.
The thesis
also addresses the integration of Artificial Intelligence and Data Science,
covering predictive resource allocation, anomaly detection, self-optimizing
networks (SON), and machine learning-driven security as key enablers for
managing next-generation system complexity. Ultimately, the work argues that
the progression of mobile and wireless systems is not merely a pursuit of
speed, but a convergence of communication, computation, and control,
underpinning the Fourth Industrial Revolution. Concluding with a
forward-looking perspective, the thesis discusses developments in 5G-Advanced
and 6G research, including sub-THz spectrum, reconfigurable intelligent
surfaces (RIS), integrated sensing and communication (ISAC), and post-quantum
security, positioning this work as both a technical record and a strategic
roadmap for future scholarly and industrial exploration.
Thesis Statement
This thesis critically
examines the technical evolution, socio-economic implications, and prospective
development of mobile and wireless radio systems, tracing their advancement
from first-generation (1G) analog voice networks to contemporary 5G AI-native
infrastructures and emerging 6G paradigms. It posits that mobile networks have
transcended their original role as communication platforms, becoming essential
digital infrastructure that underpins smart nations, industrial automation, and
socio-economic growth. By integrating engineering analysis, comparative
regional case studies across ASEAN, APAC, Europe, and ANZ, and perspectives
from Artificial Intelligence and Data Science, this research demonstrates that
the design and operation of wireless systems represent a convergence of
spectrum policy, architectural innovation, and intelligent automation. The
thesis contends that the future of wireless connectivity will be defined by
AI-enhanced, secure, and context-aware architectures, capable of supporting both
mass-market services and mission-critical applications, thereby establishing
5G-Advanced and 6G as foundational enablers of the Fourth Industrial
Revolution.
Dedication
and Thanks
I am dedicating this
thesis to my beloved daughter, IZZATI MARDIAH AZRIN whom passed away on
10th September 2025 while I am writing this paper. Although it was a
crushing blow to my emotional state of mind and to our family during this
darkest moments, I strive to carry on to complete this paper , in whatever way
possible for the better of others. From Allah we come, To Him we return.
Nothing happens without the will of God.
Candidate is appreciative
for the support during these times to the following persons
JeyaShree Rajkumar – Senior Program Manager,
Lithan Academy for the inputs on the usage of Artificial Intelligence,
Prompting, Programming, Workflow designs which have been heavily utilized in
this program and the presentation of the thesis.
Andrew Toh Tze Chow – Adjuct Faculty Lecturer
- Artificial Intelligence Regression and Progression Modelling techniques https://www.linkedin.com/in/andrew-toh78/
Fan Yeng Loon – Adjuct Faculty Lecturer
– Machine Learning Technical techniques crucial insights to the technological
Data Science Analytics
Dhanabal T – Lithan Academy /
UniMarconi Adjuct Lecturer – Basis of Encyrptions in Data Communications
Dr Azian Razak – Zyraz Technology – CXO
– ZYRAZ Global Technologies – Scaleup systems in industrialised environments
optimized using Artificial Intelligence , Machine Learning and Data Science
Analytics. Linkedin : https://www.linkedin.com/in/azianrazak
Dr M Rezuwan Zakaria – Sentara group – Neste
Oils – Usage of Machine Learning Modelling in real time, JIT systems from
Research to full scale. https://www.linkedin.com/in/rezuwanzakaria
Also many thanks to fellow cohort mates of PDDS0524 of
the Lithan Academy:
Mahathir Humaidi - Collaboration classmate
& brainstorming of the subject matters – Program Manager, Lithan Academy
https://www.linkedin.com/in/mahathir-humaid
Aloysius Chia - Collaboration Classmate,AML/KYC Specialist –
HSBC Singapore
https://linkedin.com/in/aloysiuschia
Justin Neo - Collaboration Classmate – Business
Development – HSBC Singapore
https://www.linkedin.com/in/justin-neo-01140047
Jimmy Seah - Collaboration Classmate – Director – BECA
https://www.linkedin.com/in/jimmyseah/
Ng Yea Ling – Collaboration Classmate
– PICO Creative
https://www.linkedin.com/in/yea-ling-ng-578779171/
Research
& Scope
The
evolution of mobile and wireless communication systems over the last four
decades represents one of the most transformative engineering and societal
achievements in human history. From the humble beginnings of analog cellular
telephony in the 1980s (1G) to today’s highly sophisticated 5G networks and the
research frontier of 6G, mobile and wireless technologies have fundamentally
reshaped the global economy, connectivity paradigms, and the nature of human
interaction. This thesis presents a comprehensive engineering-oriented
overview of mobile and wireless radio systems, emphasizing their technical
evolution, underlying architectures, and deployment topologies, while
integrating perspectives from data science, artificial intelligence (AI),
and machine learning (ML).
The
study addresses both historical context and forward-looking
perspectives, incorporating systems such as Wi-Fi, WiMAX, LoRaWAN, TETRA
ATEX radios, ST GRID devices, and non-terrestrial networks including satellite
constellations such as Beidou, Gouwang, Qianfan, Starlink, and Inmarsat. It
also provides a comparative regional analysis focusing on ASEAN (Association
of Southeast Asian Nations) and ANZ (Australia and New Zealand)
markets, where diverse geographic, demographic, and regulatory conditions have shaped
unique deployment trajectories.
The
relevance of this topic is manifold. From an engineering standpoint, the
radio technologies that underpin mobile networks illustrate the interplay
between spectrum allocation, modulation techniques, hardware capabilities,
and topological design. From a data science and AI perspective,
modern networks have become too complex to manage solely through human
intervention; they rely increasingly on autonomous optimization, predictive
analytics, and AI-native architectures. From a socio-economic angle,
mobile and wireless networks have become essential infrastructures—particularly
evident during the COVID-19 pandemic when connectivity underpinned remote work,
education, and healthcare.
Thus,
this thesis aims to:
- Trace the technical lineage of
mobile and wireless systems from 1G to 5G and beyond, including parallel
technologies such as Wi-Fi, WiMAX, and LPWAN.
- Analyze the role of spectrum and regulation, with a specific emphasis on the historically significant 900 MHz
band and spectrum management policies in ASEAN and ANZ.
- Examine propagation fundamentals and
how link budgets drive engineering design.
- Investigate modulation, coding, and multiple access schemes as the building blocks of radio systems.
- Compare network topologies and hierarchies, from macro-cells to femtocells, as well as mesh and ad hoc
structures.
- Explore specialized systems
including mission-critical communications (TETRA ATEX, ST GRID) and
satellite networks.
- Assess the role of AI, ML, and Data Science in optimization, performance analysis, and next-generation
innovations.
- Provide regional case studies
(ASEAN and ANZ), highlighting opportunities and challenges.
- Conclude with a forward-looking view on 5G
Advanced, 6G research themes, and the growing AI-native network paradigm.
Research
Methodology
The
methodology employed in this thesis adheres to a mixed
qualitative–quantitative framework, designed to ensure both academic
rigor and practical engineering applicability.
Literature
Review and Bibliographic Analysis
- Primary Sources:
- 3GPP specifications (TS 36.xxx for LTE, TS
38.xxx for 5G NR).
- ITU reports (IMT-2000, IMT-Advanced, IMT-2020).
- IEEE standards (802.11 Wi-Fi, 802.16 WiMAX).
- LoRa Alliance technical documentation.
- Secondary Sources:
- Peer-reviewed academic journals (IEEE
Transactions on Wireless Communications, Elsevier Computer Networks, ACM
SIGCOMM).
- Industry white papers (Ericsson, Nokia, Huawei,
Qualcomm, Cisco).
- Regulatory reports from ASEAN/ANZ agencies
(e.g., ACMA in Australia, IMDA in Singapore, NTC in the Philippines).
- Grey Literature:
- Satellite operator documentation (Inmarsat
technical manuals, Starlink deployment updates, Chinese satellite
navigation whitepapers for Beidou/Gouwang/Qianfan).
- Operator field reports (Singtel, Telstra, Globe
Telecom).
The
literature review ensures triangulation of perspectives: academic theory,
industry practice, and regulatory frameworks.
Comparative
Case Study Method
The
thesis uses comparative case studies for ASEAN and ANZ to illustrate how
geographic, economic, and regulatory differences influence mobile system
deployment.
- ASEAN: diverse spectrum of markets, from advanced
(Singapore) to developing (Cambodia, Myanmar). Key themes: urban-rural
divide, affordability, spectrum scarcity.
- ANZ: large land masses, low population density,
leading to emphasis on satellite integration and rural coverage.
Case
studies will be supported by quantitative data (coverage, latency,
throughput, spectrum allocation) and qualitative analysis (policy,
regulation, operator strategies).
Engineering Analysis
The
thesis employs engineering evaluation methods, including:
- Link budget modeling: examining transmit
power, antenna gains, path loss models (COST-Hata, ITU-R).
- Spectral efficiency comparisons
across generations (bits/s/Hz).
- Latency and throughput trade-offs in
different access schemes (TDMA vs CDMA vs OFDMA).
- QoS/KPI assessment with AI-based predictive
modeling (supervised ML for anomaly detection, clustering for traffic
profiling).
AI and Data Science
Integration
Modern
networks rely heavily on AI/ML techniques. This thesis incorporates:
- AI-native RAN (Radio Access Networks): Machine learning for beam management, anomaly detection, and
interference prediction.
- Data Science for KPI Optimization: Predictive analytics for handover success rates, coverage holes,
and traffic distribution.
- Big Data in Core Networks: Cloud-native 5G cores leveraging ML-based orchestration.
- AI-enhanced Security:
ML-based intrusion detection and adaptive encryption.
Validation
To
ensure validity and reliability:
- Cross-verification of
technical details with multiple sources (academic + industry +
regulatory).
- Quantitative benchmarks aligned with ITU/3GPP standards.
- Transparency in assumptions (e.g., propagation models assume urban macro-cell unless stated
otherwise).
- Clear citation and footnoting, following UniMarconi’s guidelines.
Limitations
- Rapid pace of 5G/6G research may render certain
projections outdated.
- Some proprietary satellite system details are not
publicly disclosed.
- Case studies rely on available
regulatory/industry data, which may not be fully standardized across
regions.
- Thesis research is primarily focused on the APAC-ASEAN,
specifically Singapore, Malaysia, Japan, ANZ clustering with some
distinctive comparative legislations against the Euro EEC and Americas
- Machine Learning bias may be implied in the
research as agentic and deep research systems were used to derive the deep
dive data.
END OF CHAPTER 3
Chapter 4 - Historical timeline
– 1G to 5G and Beyond
The Evolution of Mobile
and Wireless communication networks were mainly defined by the CCITT or
Consultative Committee for International Telegraphs and Telephone, based out of
France and later on merged into the ITU or International Telecommunications Union
(www.itu.int) which is now based out of Geneva,
Swiss.
The earliest way of
communicating was using MORSE CODE through low band frequencies across the
Atlantic thanks to WW2 where early German Enigma machine cause heavy casualties
to the allied forces.
The
evolution of mobile communication networks from the 1980s to the present has
been marked by a sequence of “generational leaps” defined by standardization
bodies such as the International Telecommunication Union (ITU) and the 3rd
Generation Partnership Project (3GPP). Each generation (1G, 2G, 3G, 4G, and
5G) introduced new radio access technologies, modulation schemes, network
architectures, and service paradigms, with 6G already emerging as a
research frontier. These generational shifts were not merely technological;
they were shaped by regulation, spectrum availability, socio-economic
factors, and hardware capabilities.
The
following sections trace this timeline in detail, highlighting technical
characteristics, socio-economic impacts, and the role of AI and data science in
shaping or analyzing each phase.
4.1
First Generation (1G): The Analog Era (1981–1991)
4.1.1 Overview
1G systems marked the beginning of cellular telephony,
replacing limited-range radio phones with large-scale mobile systems. They were
fully analog, primarily designed for voice services, and lacked
encryption or advanced mobility features.
4.1.2 Key Standards
AMPS
(Advanced Mobile Phone System) – deployed in North
America, operating in the 800 MHz band with 30 kHz FM channels.
NMT
(Nordic Mobile Telephony) – Scandinavia, 450/900 MHz bands.
TACS
(Total Access Communication System) – UK,
Italy, other European countries.
4.1.3 Technical Features
Multiple
Access: FDMA (Frequency Division Multiple Access).
Bandwidth
Efficiency: <0.5 bits/s/Hz.
Handover:
Hard handoff, circuit-switched connections.
Security:
Minimal; cloning and interception common.
4.1.4 Limitations
·
Very low capacity relative to
spectrum usage.
·
No data services, SMS, or
encryption.
·
Expensive handsets with limited
battery life.
4.2 Second
Generation (2G++): Digital Cellular (1991–2001)
4.2.1 Overview
2G introduced digital modulation, which enabled greater
spectral efficiency, voice encryption, and SMS services. It was the first
generation where mobile phones became mass-market consumer devices.
4.2.2 Key Standards
GSM
(Global System for Mobile Communications) –
TDMA-based, widely deployed in Europe, Asia, Africa. Operated primarily in
900/1800 MHz bands.
IS-95
(cdmaOne) – First commercial CDMA system, pioneered by
Qualcomm, operating in the US.
IS-136
(D-AMPS) – North American digital TDMA system, less successful
than GSM or CDMA.
4.2.3 Technical Features
Multiple
Access: TDMA (GSM), CDMA (IS-95).
Encryption:
A5/1 cipher in GSM, improved security vs. 1G.
Data
Services: GPRS (General Packet Radio Service) → “2.5G”, EDGE
(Enhanced Data for GSM Evolution) → “2.75G”.
Peak
Speeds: 50–200 kbps.
4.2.4 Socio-Economic Impact
SMS
revolutionized communication; low-cost and widely adopted.
Mobile
penetration grew rapidly in ASEAN (e.g., Philippines’ “texting capital”
phenomenon).
In
ANZ, 2G enabled widespread rural voice coverage.
4.2.5
GPRS & EDGE
Typically known as 2.5G and 2.75G, both Global Packet Radio
Service and Enhanced Data for GSM Evolution plays a pivotal role in the next
step of evolution into 3G networks. It also enables Laptops (aka Mobile
Computers) to be really mobile so that it can be connected to the internet.
Usage includes wireless Credit Card EFTPOS devices and Taxi Booking MDT. (shown: MDT POS system for Comfort Taxi in
Singapore, circa 2000s)
4.3
Third Generation (3G): Multimedia & Internet (2001–2010)
4.3.1 Overview
3G was standardized under ITU’s IMT-2000 umbrella,
promising global roaming and broadband-like services.
4.3.2 Key Standards
UMTS/W-CDMA
– Europe and Asia, 5 MHz carriers.
CDMA2000
(1xRTT, EV-DO) – North America and parts of Asia-Pacific.
4.3.3 Technical Features
Multiple
Access: Wideband CDMA (UMTS), CDMA2000.
Peak
Speeds:
o
UMTS ~384 kbps,
o
HSPA (High Speed Packet Access) up
to 14 Mbps (later HSPA+).
Core
Network: Introduction of packet-switched cores alongside
circuit-switched.
4.3.4 Limitations
High
cost of licenses in some countries (e.g., Europe).
Spectrum
fragmentation slowed adoption in ASEAN.
ANZ
markets adopted 3G early (Telstra’s Next G network).
Different
Networks not cross compatible between US and Rest of World.
4.4
Fourth Generation (4G): The IP Broadband Era (2009–2019)
4.4.1 Overview
4G represented a paradigm shift: all-IP networks,
high-speed broadband, and LTE as the global standard.
4.4.2 Key Standards
LTE
(Long Term Evolution) – OFDMA downlink, SC-FDMA uplink.
WiMAX
(802.16e/m) – Considered 4G by ITU, but commercially outcompeted
by LTE.
4.4.3 Technical Features
Multiple
Access: OFDMA (downlink), SC-FDMA (uplink).
Bandwidth:
Up to 20 MHz per carrier; carrier aggregation up to 100 MHz.
Peak
Speeds: 100 Mbps (Release 8 LTE), later >1 Gbps with
LTE-Advanced.
Core
Network: Evolved Packet Core (EPC), all-IP, IMS/VoLTE for
voice.
4.4.4 Impact
Enabled
mobile broadband, video streaming, app economy.
ASEAN:
accelerated smartphone adoption, mobile financial inclusion.
ANZ:
Telstra and Optus pioneered early LTE rollouts for rural broadband.
4.5
Fifth Generation (5G): Ultra-Flexible Architecture (2019–Present)
4.5.1 Overview
5G is not just a faster network but a flexible,
software-defined platform supporting enhanced mobile broadband (eMBB),
ultra-reliable low-latency communication (URLLC), and massive machine-type
communication (mMTC).
4.5.2 Key Standards
3GPP Release 15 (2019) – 5G NR,
NSA/SA modes.
Release
16–17 – URLLC, industrial IoT, non-terrestrial networks
(NTN).
Release
18 (5G Advanced) – AI-native RAN, enhanced XR, positioning.
4.5.3 Technical Features
Multiple
Access: Flexible OFDM numerologies (15–240 kHz spacing).
Spectrum:
o
FR1 (sub-6 GHz, e.g., 3.5 GHz).
o
FR2 (mmWave, 24–40+ GHz).
Peak
Speeds: 10 Gbps in lab trials.
Latency:
<1 ms target for URLLC.
MIMO:
Massive MIMO (64T64R, >256 antennas), beamforming.
4.5.4 Use Cases
Smart
cities, autonomous vehicles, industrial automation.
ASEAN:
Singapore leads with 5G SA, others catching up.
ANZ:
Focus on industrial IoT and rural coverage with low-band 5G.
4.6 Towards Sixth Generation (6G): The
Research Frontier (2030+)
4.6.1 Emerging Concepts
Sub-THz
spectrum (100 GHz–1 THz) for Tbps data rates.( Terabits)
Joint
Communication and Sensing (JCAS).
Reconfigurable
Intelligent Surfaces (RIS).
Integrated
terrestrial–non-terrestrial networks (NTN).
AI-native
core and RAN.
4.6.2 Global Research Initiatives
Europe:
Hexa-X program.
China:
6G test satellite launches.
Japan/Korea:
Government-industry consortia for 6G by 2030.
ASEAN/ANZ:
Early research but more focused on 5G rollout.
4.7
Summary
From
analog voice (1G) to AI-native, cloud-driven architectures (5G and beyond), the
mobile communications timeline reflects both technical ingenuity and
socio-economic drivers. The integration of AI and Data Science is
not a late addition but a necessary response to the complexity of modern
networks, paving the way for 6G as a cognitive, sensing-aware
infrastructure.
END OF CHAPTER 4
Chapter
5 - Spectrum and Regulation (ASEAN/APAC Focus, 900 MHz and Beyond)
5.1
Introduction to Spectrum as a Resource
Spectrum
is the most fundamental and limited natural resource in wireless
communications. It is not consumable like fuel, but it is scarce and subject to
strict allocation, regulation, and reuse strategies. Unlike other
resources, spectrum cannot be manufactured; it must be efficiently shared and
reused using frequency planning, modulation efficiency, and multi-access
schemes.
The
management of spectrum has been overseen globally by the International
Telecommunication Union (ITU), while national regulators such as ACMA
(Australian Communications and Media Authority), IMDA (Singapore), NBTC
(Thailand), and others in the ASEAN region enforce localized frameworks.
Spectrum licensing, auctions, and refarming are key mechanisms through which
governments allocate frequencies to operators.
5.2
The Importance of the 900 MHz Band
The
900 MHz band has historically played a central role in mobile communications:
- 1G: Many early analog
systems operated at 900 MHz.
- 2G GSM: 900
MHz was the global anchor band, offering balance between coverage and
capacity.
- 3G UMTS: Some
refarming allowed deployment of 3G in 900 MHz.
- 4G LTE:
Operators in ASEAN/ANZ reused 900 MHz for LTE coverage in rural areas.
- 5G NR: While
primarily sub-6 GHz and mmWave, some early deployments use refarmed 900
MHz for coverage-layer NR.
5.3
Propagation Benefits
- Lower path loss vs. higher bands.
- Deeper indoor penetration.
- Larger cell radius → fewer base stations needed,
lowering CAPEX.
5.4
Trade-offs
- Narrower bandwidth compared to higher frequencies
(limiting capacity).
- Heavily congested in many regions due to legacy
users.
5.5 Spectrum Allocation by Generation
Generation |
Typical Bands |
Bandwidth Per Carrier |
Notes |
1G |
800–900 MHz |
25–30 kHz |
Analog
FDMA channels |
2G (GSM) |
900, 1800, 1900 MHz |
200 kHz |
900 MHz
dominant in ASEAN/ANZ |
3G (UMTS/ W-CDMA) |
2.1 GHz (primary), 900 MHz (refarmed) |
5 MHz |
Used dual
bands |
4G (LTE) |
700, 800, 900, 1800, 2600, 3500 MHz |
1.4–20
MHz (CA up to 100 MHz) |
ASEAN/ANZ
diversified |
5G (NR) |
FR1 (600–4200 MHz), FR2 (24–40 GHz) |
5–100 MHz
(FR1), 100–400 MHz (FR2) |
Auctions
across ANZ/ASEAN ongoing |
6G (Future) |
Sub-THz (100–1000 GHz) |
2–10 GHz |
Research
phase only |
5.6
Spectrum Licensing Approaches
5.6.1 Auction-Based Allocation
- Common in ANZ (e.g., Australian 3.6 GHz 5G auction, 2018).
- Generates government revenue but may increase operator costs →
higher end-user tariffs.
5.6.2 Administrative Assignment
- Seen in some ASEAN countries with less mature
regulatory markets.
- Spectrum directly assigned to incumbents.
5.6.3 Unlicensed Spectrum
- ISM bands (2.4 GHz, 5 GHz, 6 GHz) widely used for
Wi-Fi and LoRaWAN.
- ASEAN has mixed adoption of 6 GHz Wi-Fi 6E.
- ANZ allows wider unlicensed use due to advanced
regulatory frameworks.
5.6.4 Shared Spectrum Models
- CBRS (Citizens Broadband Radio Service)-style
models in the US inspire ASEAN/ANZ regulators.
- Potential use of shared spectrum for private 5G/IoT deployments.
5.7
Spectrum Refarming
Refarming
is essential to migrate legacy networks without wasting spectrum:
- ASEAN Example:
Philippines’ Globe Telecom reallocated 2G/3G bands to LTE and now to 5G.
- ANZ Example:
Telstra and Optus shut down 2G by 2016–2017, refarming spectrum to LTE.
Refarming
balances backward compatibility with forward innovation.
5.7.1 ASEAN
Spectrum Landscape
ASEAN nations are diverse in spectrum usage:
Singapore
(IMDA): Aggressive 5G rollout in 3.5 GHz and mmWave. 900 MHz
used for LTE fallback.
Malaysia
(MCMC): Single Wholesale Network for 5G (controversial).
Thailand
(NBTC): Multi-band 5G auctions (700 MHz, 2600 MHz, 26 GHz).
Indonesia:
Large geography, limited 3.5 GHz spectrum availability → reliance on 1800/2100
MHz refarming.
Philippines:
New entrant Dito Telecommunity licensed in 700 MHz for coverage.
Challenges:
Fragmented
policies across countries.
High
auction fees vs. affordability pressures.
Legacy
2G networks still operational in some regions.
5.7.2 ANZ Spectrum Landscape
Australia and New Zealand benefit from more coordinated
spectrum planning.
Australia
(ACMA):
o
700 MHz (“digital dividend”)
allocated for LTE and 5G.
o
3.6 GHz allocated to 5G.
o
mmWave auctions (26 GHz) completed
in 2021.
New
Zealand (MBIE):
o
600 MHz, 700 MHz for rural LTE/5G.
o
3.5 GHz as the primary 5G band.
Unique features:
Large
landmass → heavy reliance on low-band spectrum for rural coverage.
Early
adoption of satellite augmentation (Starlink, Inmarsat) to fill coverage
gaps.
5.8
Sub-THz and mmWave Challenges
While
mmWave (24–40 GHz) offers extreme bandwidth, coverage limitations are
significant:
- Propagation Loss:
Free-space path loss increases 6 dB per frequency doubling.
- Penetration Loss: High
absorption through walls, foliage, even rain.
- Deployment Cost:
Requires dense small-cell infrastructure.
ASEAN
context: Dense cities like Singapore may benefit. Rural ASEAN less suitable.
ANZ context: Limited mmWave beyond urban hotspots due to geography.
Sub-THz
for 6G introduces further challenges: atmospheric absorption, hardware cost,
regulatory uncertainties.
5.9
Spectrum Policy Trends
- Harmonization: ITU
encourages global harmonization (e.g., 3.5 GHz for 5G), but national
variations persist.
- Dynamic Spectrum Sharing (DSS): Used to allow LTE and 5G coexistence in same band.
- Satellite-Terrestrial Convergence: New ITU allocations for non-terrestrial networks (NTN) under
5G NR.
- AI for Spectrum Management: Cognitive radio and ML-driven spectrum sensing proposed for 6G.
5.10
Summary
Spectrum
regulation has been both an enabler and bottleneck of mobile technology
evolution. The 900 MHz band exemplifies the continuity of spectrum
reuse, while higher bands (3.5 GHz, mmWave) reflect the demand for capacity.
ASEAN’s fragmented policies contrast with ANZ’s coordinated planning, yet both
regions demonstrate innovation in refarming, auctions, and satellite
integration.
AI
and Data Science will increasingly be used for dynamic spectrum allocation,
interference detection, and demand prediction, making spectrum management
not only a regulatory challenge but also a computational one.
END
OF CHAPTER 5
Chapter
5.11 AI and ML for Propagation
Prediction
Traditional
models are deterministic/empirical. AI/ML methods use real measurement data:
- Supervised ML: Train models on
drive-test/KPI data to predict RSRP/RSRQ.
- Deep Learning: CNNs for RF fingerprinting.
- Reinforcement Learning: Adaptive beam
alignment in mmWave links.
- Big Data Analytics: Correlating traffic
patterns with propagation anomalies.
Operators
in Singapore and Australia have trialed AI for automatic coverage hole
detection.
Summary
Radio
propagation and link budgets are the foundation of mobile system design.
Lower frequencies (e.g., 900 MHz) provide wide-area coverage, while higher
frequencies (e.g., mmWave) require AI-optimized dense deployments. Modern
operators cannot rely solely on mathematical models—they require data-driven,
AI-augmented approaches to continuously optimize networks in real time.
Chapter
6 – Modulation, Coding, and Multiple Access Evolution
6.1
Introduction
The
evolution of modulation, coding, and multiple access techniques has been the engine
driving improvements in capacity, spectral efficiency, and reliability
across successive mobile generations. Each leap in wireless standards has
corresponded to advancements in how bits are represented on radio waves and how
multiple users are multiplexed onto finite spectrum.
This
chapter explores these advancements chronologically, from analog FM in 1G
to OFDMA and advanced coding in 5G, while considering the growing role
of AI and ML in waveform optimization, adaptive coding, and interference
management.
6.2
Modulation Evolution
1G:
Analog FM
- Frequency Modulation (FM):
Simple but inefficient.
- Channel bandwidth: 25–30 kHz.
- Susceptible to interference, poor spectral efficiency.
2G:
Digital Modulation
- GMSK (Gaussian Minimum Shift Keying) in GSM.
- Constant envelope →
efficient power amplifiers.
- Robust but limited spectral efficiency (~1
bit/s/Hz).
- π/4-QPSK in
IS-54 TDMA.
- QPSK/BPSK/CDMA spreading in IS-95.
2.5G
and 3G
- 8-PSK in
EDGE → 3 bits/symbol.
- QPSK, 16-QAM, 64-QAM in
W-CDMA/HSPA.
- Adaptive modulation introduced: switching based
on channel conditions.
4G
LTE (Long Term Evolution)
- OFDMA (downlink): Combats multipath with cyclic
prefix.
- SC-FDMA (uplink): Reduces Peak-to-Average Power
Ratio (PAPR).
- Adaptive modulation up to 256-QAM.
5G
NR
- Flexible numerologies (subcarrier spacing 15–240
kHz).
- Higher-order QAM (up to 1024-QAM in
trials).
- Beamformed modulation optimized for spatial
channels.
6G
Outlook
- Orbital Angular Momentum (OAM) waves under research.
- AI-designed waveforms: Neural networks optimize modulation schemes dynamically.
6.3
Channel Coding Evolution
Error-correcting
codes have evolved alongside modulation, crucial for reliability under
fading and interference.
- 1G: No channel coding.
- 2G GSM:
Convolutional coding + interleaving.
- 3G UMTS: Turbo
codes → near Shannon-limit performance.
- 4G LTE: Turbo
codes (legacy), later LDPC (in trials).
- 5G NR:
- LDPC (Low-Density Parity Check): For data channels.
- Polar Codes: For
control channels.
- 6G (Research):
Sparse graph codes, AI-assisted error correction.
6.4
Multiple Access Techniques
1G
– FDMA – Frequency Divided Multiple Access Technology
- Users separated by frequency.
- Simple but inefficient, poor scalability.
2G
– TDMA & CDMA – Time Divided & Code Division Multiple-Access
- TDMA (Time Division Multiple Access):
- GSM: 200 kHz channels, 8 time slots.
- Effective for circuit-switched voice.
- CDMA (Code Division Multiple
Access):
- IS-95, UMTS.
- Spreads signals with pseudo-random codes.
- Soft handoffs, interference-limited planning.
3G
– W-CDMA
- Wideband CDMA, 5 MHz carriers.
- Supported higher data rates but suffered from
self-interference.
OFDMA
- Orthogonal Frequency Division Multiple Access.
- Divides spectrum into orthogonal subcarriers.
- Flexible scheduling in time–frequency grid.
- Multipath resilience, high spectral efficiency.
5G
NR – Flexible OFDM
- Variable subcarrier spacing.
- OFDM extended to mmWave, massive MIMO.
- Sidelink and non-terrestrial integration.
- Support for URLLC (low latency), mMTC
(IoT), eMBB (broadband).
6G
Outlook
- NOMA (Non-Orthogonal Multiple Access):
- Users multiplexed by power domain.
- ML used for successive interference
cancellation.
- Rate-Splitting Multiple Access (RSMA).
- AI-driven resource allocation predicted to
dominate.
6.5
Synchronization and Reference Signals
Modern
systems require precise synchronization for coherent detection.
- 2G GSM:
Broadcast Control Channel (BCCH).
- 3G: Primary/Secondary
Synchronization Channels.
- 4G LTE:
Primary/Secondary Sync Signals, Cell-specific Reference Signals.
- 5G NR:
- PSS/SSS for
cell search.
- DM-RS/CSI-RS for
channel estimation in beamformed channels.
AI-enhanced
synchronization techniques are being explored, where ML algorithms dynamically
track Doppler and fading.
6.6
AI and Data Science in Modulation and Coding
- Adaptive Modulation & Coding (AMC):
Traditional link adaptation uses SNR (Signal Noise Ratio) thresholds. AI-based
methods optimize multi-dimensional trade-offs (throughput,
reliability, latency).
- Deep Learning for Decoding:
Neural decoders outperform LDPC in some cases.
- Reinforcement Learning for Multiple Access: Used for NOMA power allocation.
- Waveform Design: AI models search
modulation/coding combinations beyond human-designed standards.
6.7
Summary
From
simple analog FM (1G) to AI-optimized OFDM and coding in 5G and
beyond, modulation and multiple access have been central to increasing
capacity and efficiency. While human-designed techniques dominated until 5G,
the 6G era will increasingly rely on AI to co-design and optimize waveforms,
coding, and access schemes in real time.
END
OF CHAPTER 6
Chapter
7 - Network Topologies and Hierarchies
7.1
Introduction
A
mobile network is not only defined by its air interface or modulation scheme
but also by its topology — the physical and logical arrangement of base
stations, cells, and user devices. Over time, as demand for coverage,
capacity, and reliability has increased, mobile networks have evolved from
simple macrocellular deployments to heterogeneous networks (HetNets)
integrating macro, micro, pico, femto cells, mesh, and device-to-device
(D2D) topologies.
In
parallel, Low-Power Wide Area Networks (LPWANs) such as LoRaWAN
introduced a different topology optimized for low data rate, long-range IoT
connectivity.
This
chapter examines the evolution of topologies, their trade-offs, and the role of
AI and data science in self-organizing networks (SONs) and autonomous
topology management.
Definition
Macrocells
are large coverage areas served by high-power base stations, typically
transmitting at 20–40 W, mounted on towers or rooftops.
Characteristics
- Coverage radius: 2–35 km depending on frequency
and terrain.
- Primary layer of 1G–4G deployments.
- Cost-effective for rural areas (ANZ’s vast
geography).
Limitations
- Inefficient for indoor coverage in dense urban environments.
- High inter-cell interference in congested networks.
7.3 Microcells
Definition
Use
Cases
- Urban hotspots (shopping malls, downtown cores).
- Supplement macro coverage in ASEAN megacities
like Bangkok, Manila, Jakarta.
Trade-offs
- Lower transmit power → lower CAPEX per site.
- Increased OPEX due to denser site requirements.
6.4 Picocells
Very
small base stations with coverage radius of 10–200 meters.
Use
Cases
- Enterprises, office buildings, airports.
- Often used for indoor LTE/5G private networks.
ASEAN/ANZ
Deployment
- Singapore uses picocells for indoor 5G coverage
in MRT tunnels.
- Australia: mining companies deploy picocells for
private LTE in remote facilities.
6.5
Femtocells
Definition
Ultra-small,
consumer-deployed base stations (coverage radius 10–50 m).
Role
- Extend indoor coverage.
- Use residential broadband as backhaul.
Challenges
- Interference management.
- Security of user-deployed nodes.
6.6
Heterogeneous Networks (HetNets)
HetNets
combine macro, micro, pico, and femto cells into a layered architecture.
Benefits
- Macro layer ensures coverage.
- Small cells increase localized capacity.
- Offloading traffic improves spectrum efficiency.
6.7
SON (Self-Organizing Networks)
- Automates configuration and optimization.
- AI-based SONs adjust neighbor relations, power
levels, and handover thresholds dynamically.
- Widely used in 4G/5G deployments in ANZ
(Telstra’s AI-driven SON pilot).
6.8
Mesh and Ad Hoc Networks
Public
Safety and Tactical Use
- TETRA
systems use mesh/ad hoc extensions for first responders. (Used in
Singapore for Oil & Gas, specific to Bukom & Jurong Islands)
- ProSe (Proximity Services): Introduced in LTE Rel-12 for device-to-device.
Military
and Industrial IoT
- Mesh networks in mining (Australia) and oil rigs (Malaysia).
- D2D communication critical for autonomous vehicles.
Challenges
- Interference and scalability.
- Security of decentralized topologies.
6.9 LPWAN Topologies
LoRaWAN
– Low power Long Range Wide Area Network
Star-of-stars
topology: end devices → gateways → network server.
Gateways
connect to central servers via IP backhaul.
Optimized
for long-range, low-power IoT telemetry.
Alternatives
- Sigfox: Ultra-narrowband star
topology.
- NB-IoT / LTE-M: Cellular LPWAN alternatives.
ASEAN/ANZ
Adoption
Singapore:
nationwide LoRaWAN IoT deployments for smart city.
New
Zealand: farmers use LoRaWAN for livestock monitoring.
6.10 AI/ML in Topology
Optimization
Cell
Planning
- AI models trained on traffic data predict where
small cells should be deployed.
- Reinforcement learning for automatic tilt
optimization of antennas.
Traffic
Load Balancing
- Clustering algorithms (k-means, DBSCAN) detect overloaded cells.
- Load shifted via AI-controlled handover tuning.
Failure
Recovery
- Predictive maintenance using anomaly detection.
- Self-healing topologies reroute traffic around
failed nodes.
Topology |
Coverage Radius |
Strengths |
Limitations |
Macrocell |
2–35 km |
Wide area coverage, cost-effective |
Poor indoor penetration |
Microcell |
0.1–2 km |
Capacity boost, urban use |
Higher site density |
Picocell |
10–200 m |
Indoor coverage, enterprise |
CAPEX for enterprise |
Femtocell |
<50 m |
Consumer indoor use |
Security, interference |
Mesh/Ad Hoc |
Varies |
Resilient, decentralized |
Complexity, interference |
6.12 Summary
Network
topologies evolved from macro-only 1G deployments to sophisticated
multi-layered 5G HetNets and specialized IoT topologies. In ASEAN, dense
cities require pico/femto deployments, while ANZ emphasizes macrocells
and satellite integration.
AI
and Data Science enable dynamic optimization of these topologies, ensuring
real-time adaptability to changing traffic, failures, and propagation
environments.
END
OF CHAPTER 6
Chapter
7 - Standards by Generation
(1G to 5G, Wi-Fi, WiMAX, LoRaWAN)
7.1
Introduction
Telecommunications
systems evolve not in isolation but under the guidance of standards bodies,
which ensure interoperability, economies of scale, and global compatibility.
The most influential standards bodies are:
- 3GPP (3rd Generation Partnership Project): LTE, 5G NR, core networks.
- ITU (International Telecommunication Union): Defines IMT-2000, IMT-Advanced, IMT-2020 (frameworks for 3G, 4G,
5G).
- IEEE (Institute of Electrical and Electronics Engineers): Wi-Fi (802.11), WiMAX (802.16).
- LoRa Alliance: LoRaWAN standardization for
LPWAN.
This
chapter provides a generation-by-generation breakdown of standards, followed by
Wi-Fi/WiMAX and LoRaWAN ecosystems.
7.2 1G Standards
(Analog Cellular)
AMPS
(Advanced Mobile Phone System)
- Launched in the US in 1983.
- 30 kHz FM channels in the 800 MHz band.
- FDMA, no encryption, poor spectral efficiency.
NMT
(Nordic Mobile Telephony) – NoKIA/NorTEL
- First international cellular standard (Scandinavia, early 1980s).
- 450/900 MHz bands.
- Influenced global adoption.
TACS
(Total Access Communication System)
- UK adaptation of AMPS.
- Widely used in Europe and Asia before GSM.
Summary:
1G standards were fragmented, regional, and analog-only.
7.3 2G Standards (Digital Cellular)
GSM
(Global System for Mobile Communications)
- ETSI standard (Europe, 1991).
- TDMA, 200 kHz carriers.
- Global adoption, especially in ASEAN/ANZ.
- Introduced SMS and encryption.
IS-95
(cdmaOne)
- Qualcomm’s CDMA-based system.
- Spread spectrum, soft handoffs.
- Popular in US, Korea.
IS-136
(D-AMPS)
- US TDMA system.
- Transitional, eventually replaced by GSM/CDMA.
2.5G
Enhancements
- GPRS: Packet-switched data (~50 kbps).
- EDGE: Enhanced modulation (8-PSK), up to ~200 kbps.
7.4 3G Standards
(IMT-2000)
UMTS
/ W-CDMA
- Europe and Asia.
- 5 MHz carriers, 2.1 GHz band.
- HSPA/HSPA+ → up to 42 Mbps.
CDMA2000
- Evolution of IS-95.
- 1xRTT, EV-DO.
- Deployed in North America, some Asian markets.
Regional
Adoption
- ASEAN: uneven rollout due to cost.
- ANZ: Telstra’s “Next G” (850 MHz) was one of the
most advanced UMTS networks globally.
7.5 4G
Standards (IMT-Advanced)
LTE
(Long Term Evolution)
- 3GPP Release 8 (2009).
- OFDMA (DL), SC-FDMA (UL).
- Up to 20 MHz per carrier; carrier aggregation
(CA) enabled multi-carrier operation.
- LTE-Advanced (Rel-10+):
- Higher-order MIMO (8x8).
- Coordinated multipoint (CoMP).
- Peak rates >1 Gbps.
WiMAX
(IEEE 802.16e/m)
- Considered 4G by ITU, but lost to LTE due to
ecosystem dominance.
- Still deployed in some ASEAN regions for fixed
broadband.
7.6 5G Standards
(IMT-2020)
3GPP
Release 15–17
- 5G NR (New Radio).
- Flexible numerology (15–240 kHz spacing).
- Sub-6 GHz (FR1) and mmWave (FR2).
- Massive MIMO, beamforming.
- NSA (non-standalone, LTE anchor) and SA
(standalone, 5G core).
Release
18 and 5G Advanced
- AI-native RAN.
- Enhanced XR (Extended Reality) support.
- RedCap (Reduced Capability) devices for IoT.
- Non-terrestrial networks (NTN): satellites integrated with 5G.
7.7
Wi-Fi Standards (IEEE 802.11 Family)
Generation |
Standard |
Year |
Features |
Wi-Fi 1 |
802.11b |
1999 |
DSSS, 2.4 GHz, 11 Mbps |
Wi-Fi 2 |
802.11a |
1999 |
OFDM, 5 GHz, 54 Mbps |
Wi-Fi 3 |
802.11g |
2003 |
2.4 GHz, OFDM, 54 Mbps |
Wi-Fi 4 |
802.11n |
2009 |
MIMO, 2.4/5 GHz, 600 Mbps |
Wi-Fi 5 |
802.11ac |
2014 |
MU-MIMO, 160 MHz channels, Gb/s |
Wi-Fi 6 |
802.11ax |
2019 |
OFDMA, BSS Coloring, low latency |
Wi-Fi 7 |
802.11be |
2024 (ongoing) |
320 MHz channels, multi-link operation (MLO) |
ASEAN/ANZ:
Wi-Fi complements cellular for indoor connectivity; Wi-Fi 6E in 6 GHz is
adopted in Australia and Singapore.
LoRaWAN (Low-Power Wide
Area Network)
Characteristics
- Physical Layer: LoRa modulation (chirp spread
spectrum).
- Topology: Star-of-stars (end devices →
gateways → network servers).
- Bands: ISM (868/915 MHz in
ANZ/ASEAN).
Device
Classes
- Class A: Ultra-low power, scheduled
downlinks.
- Class B: Periodic beacon windows.
- Class C: Almost continuous receive.
Deployment
- ASEAN: Smart city (Singapore, Malaysia).
- ANZ: Agricultural IoT (livestock, irrigation monitoring).
7.8
Comparative Technology Matrix
Technology |
Access Method |
Peak
Data Rate |
Latency |
Strengths |
Limitations |
GSM/EDGE |
TDMA |
~0.2 Mbps |
~150 ms |
Wide
coverage |
Low data rates |
UMTS/HSPA |
W-CDMA |
14–42 Mbps |
50–100 ms |
Proven
tech |
Interference mgmt |
LTE-A |
OFDMA/SC-FDMA |
>1 Gbps |
20–50 ms |
Ecosystem,
efficiency |
Dense planning |
5G NR |
OFDMA |
>10 Gbps |
<1 ms |
Flexible,
slicing, URLLC |
Complexity, mmWave limits |
Wi-Fi 6/7 |
OFDMA |
Multi-Gbps |
~10 ms |
Unlicensed,
cheap |
Contention, interference |
LoRaWAN |
ALOHA-like |
kbps |
Seconds |
Long
range, ultra-low power |
Very low throughput |
7.9 Summary
Standards
have progressed from analog fragmentation (1G) to globally harmonized
digital ecosystems (5G NR). Wi-Fi and LoRaWAN evolved in parallel,
targeting different niches: Wi-Fi for indoor broadband and LoRaWAN
for ultra-low power IoT.
The
unifying trend is toward flexible, AI-native, software-defined standards
where modulation, access, and resource allocation can adapt in real time. This
lays the groundwork for 6G, where AI will likely play a direct role in waveform
and standard evolution.
END OF CHAPTER 7
Chapter
8 - Specialized Systems & Use Cases
(LoRaWAN,
TETRA ATEX, ST GRID, Satellites: Beidou, Gouwang, Qianfan, Starlink, Inmarsat)
8.1
Introduction
Beyond
mainstream mobile standards (1G–5G), specialized wireless systems have emerged
to meet unique requirements:
- Low-power IoT (LoRaWAN).
- Mission-critical communications (TETRA ATEX, ST
GRID).
- Satellite-based systems (Beidou, Gouwang,
Qianfan, Starlink, Inmarsat).
These
systems fill niches not well served by cellular broadband: industrial
safety, rural connectivity, global positioning, maritime/aviation
communications, and massive IoT telemetry.
8.2
LoRaWAN (Low Power, Long Range IoT)
Technical
Overview
- LoRa (physical layer): Chirp spread spectrum modulation.
- LoRaWAN (protocol):
Defines MAC and backend.
- Bands: ISM
(EU 868 MHz, US/AU 915 MHz, ASEAN varies).
- Data rate:
0.3–50 kbps.
- Range: 2–15
km urban, >50 km rural line-of-sight.
- Battery life: 5–10
years.
Use
Cases
- Smart cities: parking sensors, waste management.
- Agriculture: soil moisture, livestock tracking
(ANZ).
- Disaster response: temporary sensor networks.
Pros
and Cons
✅
Long range, ultra-low power, unlicensed spectrum.
❌
Low throughput, high latency (seconds).
8.3
TETRA ATEX Systems
TETRA
(Terrestrial Trunked Radio)
- ETSI standard for professional mobile radio.
- Used by police, emergency services, utilities.
- Digital trunking, encryption, direct mode (D2D).
- 446Mhz in Singapore
ATEX
Compliance
- Devices certified for explosive environments (oil
rigs, chemical plants, mining).
- Ruggedized walkie-talkies with intrinsically safe
design.
ASEAN/ANZ
Deployment
- ASEAN: TETRA
used in airports, seaports, public safety (Singapore Police, Jakarta MRT).
- ANZ: ATEX radios in mining
(Western Australia, Queensland coal sector).
Pros
and Cons
✅
Mission-critical reliability, D2D, encrypted.
❌
Limited bandwidth, not suited for broadband data. Transmitter is ranged up to
10km radius
8.4 - ST GRID Walkie-Talkie Phone
Overview
- Hybrid of mobile phone and push-to-talk (PTT)
radio.
- Developed for industrial/public safety
environments.
- Operates on both cellular and dedicated PMR
spectrum. 800/850x Mhz Freq in Singapore
Features
- Group calls with near-zero latency.
- Ruggedized design.
- Integration with LTE mission-critical
push-to-talk (MCPTT).
ASEAN/ANZ
Use Cases
- ANZ: Utilities and railway operators use ST GRID
devices for workforce communication.
- ASEAN: Oil & gas industry, metro
projects (e.g., Malaysia’s MRT).
Pros
and Cons
✅ Instant
communication, hybrid cellular/PMR.
❌
Niche, not scalable for broadband IoT.
Beidou
(China)
- Global Navigation Satellite System (GNSS).
- Provides positioning, navigation, timing (PNT).
- Coverage: Global since 2020 (BDS-3).
- ASEAN impact: integrated into agriculture and
logistics.
Gouwang
(China’s broadband LEO project)
- LEO constellation under development.
- Target: broadband satellite internet, competitor
to Starlink.
Qianfan
- Another Chinese LEO initiative for global
internet coverage.
- Focus on integration with Belt & Road
countries, including ASEAN.
Starlink
(SpaceX, USA)
- LEO constellation (>5,000 satellites deployed
as of 2024).
- Low latency (~25–40 ms).
- Broadband speeds 50–250 Mbps.
- ASEAN:
Pilots in Philippines, Malaysia.
- ANZ: Widely used in rural
Australia/New Zealand for broadband.
Inmarsat
(UK)
- Legacy GEO satellite operator.
- Provides maritime, aviation, emergency communications.
- Lower data rates, higher latency (~600 ms).
System |
Orbit/Layer |
Latency |
Strengths |
Limitations |
Beidou |
MEO/GEO |
~50–100 ms |
GNSS PNT |
Not broadband |
Gouwang/Qianfan |
LEO (planned) |
~25–40 ms |
National broadband strategy |
Still unproven |
Starlink |
LEO |
25–40 ms |
Broadband, rural coverage |
Expensive, high CAPEX |
Inmarsat |
GEO |
~600 ms |
Aviation, maritime |
High latency, limited throughput |
LTE/5G |
Terrestrial |
1–20 ms |
High capacity |
Limited rural reach |
LoRaWAN |
Terrestrial ISM |
Seconds |
Ultra-low power IoT |
Very low throughput |
TETRA ATEX |
PMR |
<100 ms |
Mission-critical voice |
No broadband |
8.6 Comparative Analysis: Satellite vs Terrestrial
AI/ML
in Specialized Systems
- LoRaWAN:
ML-based anomaly detection in IoT sensor data.
- TETRA ATEX:
AI-based predictive maintenance of mission-critical radios.
- Satellites:
- Beam-hopping optimization using AI.
- Traffic load prediction in LEO constellations.
- AI-based GNSS error correction (multipath
mitigation).
ASEAN
and ANZ Regional Use Cases
- ASEAN:
- Singapore: LoRaWAN smart nation projects.
- Philippines: Starlink trial for rural schools.
- Indonesia: Beidou adoption in
logistics/shipping.
- ANZ:
- Australia: Starlink integrated into Telstra
partnerships.
- New Zealand: LoRaWAN for agricultural IoT.
- Mining/oil industries: ATEX walkie-talkies.
8.7
Summary
Specialized
systems illustrate the breadth of wireless applications beyond mainstream
5G.
- LoRaWAN
dominates low-power IoT.
- TETRA ATEX / ST GRID serve
mission-critical industrial/public safety needs.
- Satellites (Beidou, Starlink, Inmarsat) extend coverage where terrestrial fails.
Together,
they form an ecosystem of complementary technologies. AI/ML ensures
their integration into a seamless connectivity fabric for ASEAN and ANZ.
END
CHAPTER 8
Chapter
9 - Core Networks: From MSC to 5GC
with AI Enhancements
9.1
Introduction
While
the Radio Access Network (RAN) defines how users connect to the network,
the core network provides intelligence, switching, and control that
enable mobility, authentication, and services. Core networks have evolved from circuit-switched
telephony systems in 1G/2G to the all-IP, cloud-native, service-based
architecture (SBA) of 5G.
This
chapter explores core network evolution, with emphasis on:
- MSC (Mobile Switching Centers) in 1G/2G.
- SGSN/GGSN in 3G.
- EPC (Evolved Packet Core) in LTE.
- 5GC (5G Core), cloud-native and AI-enabled.
We
also highlight AI, data science, and ML integration in traffic
engineering, security, and self-healing.
9.2
1G/2G Core Networks – Circuit Switching
Mobile
Switching Center (MSC)
- Centralized circuit switch.
- Handles call setup, routing, and handovers.
- Connected to PSTN for voice services.
Home/Visitor
Location Registers (HLR/VLR)
- HLR: Stores subscriber
identity and authentication keys.
- VLR: Temporary database for
roaming subscribers.
Authentication
Center (AuC)
- Stores secret keys for SIM-based authentication.
Short
Message Service Center (SMSC)
- Enabled SMS delivery.
Limitations:
No packet data, scalability issues, inflexible.
9.3
3G Core Networks – Dual Circuit and Packet
Serving
GPRS Support Node (SGSN)
- Routes data packets.
- Tracks mobility for packet-switched services.
Gateway
GPRS Support Node (GGSN)
- Connects packet data traffic to external networks
(e.g., internet).
Circuit/Packet
Split
- Dual domains coexisted: circuit (voice) + packet
(data).
Impact:
First step toward always-on data connections, though still limited compared to
LTE.
9.4
4G Core Networks – The Evolved Packet Core (EPC)
EPC
Architecture
- MME (Mobility Management Entity): Controls mobility, authentication.
- SGW (Serving Gateway): Routes user traffic.
- PGW (Packet Gateway):
Connects to external IP networks.
- PCRF (Policy and Charging Rules Function): Manages QoS and charging
All-IP
Architecture
- Voice over LTE (VoLTE) through IMS (IP Multimedia
Subsystem).
- Circuit switching eliminated.
Benefits
- Reduced latency.
- Scalability for broadband.
- Network sharing and virtualization enabled.
9.5G
Core (5GC) – Service-Based Architecture
Service-Based
Architecture (SBA)
- Components exposed as services via APIs.
- Cloud-native, microservices architecture.
Key
Functions
- AMF (Access and Mobility Management Function): Manages user access and mobility.
- SMF (Session Management Function): Session setup, QoS policies.
- UPF (User Plane Function): Handles user traffic forwarding.
- NRF (Network Repository Function): Service discovery.
- PCF (Policy Control Function): Policy management.
- AUSF (Authentication Server Function): Authentication and security.
Features
- Network Slicing:
Logical partitioning for different services (e.g., URLLC vs eMBB).
- Edge Computing (MEC):
Applications hosted at the edge to reduce latency.
- Non-Terrestrial Networks (NTN): Satellites integrated with 5GC.
9.6
Virtualization and Cloud-Native Cores
- NFV (Network Function Virtualization): Replaces proprietary hardware with virtualized functions.
- SDN (Software Defined Networking): Separates control/data planes, enabling programmability.
- Containers and Kubernetes: Used in 5GC for cloud-native deployments.
9.7
AI and Data Science in Core Networks
7.7.1 Traffic Prediction and Optimization
- ML models forecast traffic load, enabling dynamic
scaling of UPFs.
- Time-series models (ARIMA, LSTM) predict
busy-hour congestion.
7.7.2 Network Slicing Management
- AI orchestrators dynamically allocate slices.
- Reinforcement learning optimizes slice
performance.
7.7.3 Security and Anomaly Detection
- ML-based intrusion detection identifies unusual
traffic patterns.
- AI-based fraud prevention (SIM cloning, DDoS
mitigation).
7.7.4 Fault Management and Self-Healing
- Predictive maintenance of virtual network
functions (VNFs).
- AI-based failure prediction in data centers.
9.8
ASEAN and ANZ Deployment Trends
ASEAN
- Singapore: early adoption of 5GC standalone
networks.
- Malaysia: wholesale 5G model integrates 5GC for
multiple operators.
- Indonesia/Philippines: NSA rollouts dominate,
limited standalone due to cost.
ANZ
- Australia: Telstra deploying cloud-native 5GC
with edge computing for industrial IoT.
- New Zealand: Spark focusing on MEC for smart
agriculture.
9.9
Summary
The
core network has evolved from rigid circuit-switching (MSC) to flexible,
AI-native service-based architectures (5GC).
- 1G/2G: voice-centric, circuit-switched.
- 3G: dual circuit + packet.
- 4G: all-IP EPC.
- 5G: service-based, virtualized, sliceable,
AI-enabled.
Future
6G cores will be autonomous, cognitive, and self-optimizing, with
AI not as an overlay but as a native capability.
END
CHAPTER 9
Chapter
10 - Security and Authentication Across Generations
10.1
Introduction
Security
in mobile networks has evolved alongside the technologies themselves. While 1G
networks offered virtually no protection against eavesdropping and fraud,
today’s 5G networks provide end-to-end encryption, mutual authentication,
and subscriber privacy mechanisms.
However,
each generation introduced new attack surfaces: cloning in 1G, weak
ciphers in 2G, signaling storms in 3G, IMS vulnerabilities in 4G, and
virtualized network security risks in 5G. Looking ahead, AI/ML-enhanced
security frameworks are expected to underpin 6G.
10.2
1G: Insecure Beginnings
- Analog FM voice: No
encryption, easily intercepted by scanners.
- Authentication: None;
phone cloning rampant (copying ESN/IMSI).
- Vulnerabilities:
- Fraudulent calls billed to victims.
- Eavesdropping in sensitive scenarios (e.g., law
enforcement).
10.3
2G: Basic Digital Security
Authentication and Encryption
·
SIM-based authentication:
Introduced secret key (Ki) stored in SIM and AuC.
·
A3/A8 algorithms:
Generate session key (Kc).
·
Ciphering:
A5/1, A5/2 stream ciphers.
Weaknesses
·
A5/1 eventually cracked with
rainbow tables.
·
No mutual authentication (network
not authenticated by handset) → man-in-the-middle attacks.
10.4
3G: Stronger Mutual Authentication
UMTS Authentication and Key Agreement (AKA)
·
Mutual authentication
introduced.
·
128-bit keys.
·
Stronger algorithms (Kasumi
cipher).
Security Gains
·
Protected against fake base
stations.
·
Improved integrity protection.
Limitations
·
Still vulnerable to
denial-of-service and jamming.
·
Signaling overload attacks began to
emerge.
10.5
4G: IP-Based Security Challenges
EPS-AKA
·
Mutual authentication maintained.
·
Session keys for
encryption/integrity on IP bearer.
VoLTE and IMS Vulnerabilities
·
SIP protocol in IMS exposed to
spoofing, denial-of-service.
·
Interworking with internet services
increased attack surface.
LTE-Specific Threats
·
IMSI Catchers (Stingrays):
Exploited fallback to 2G/3G for interception.
·
Rogue eNodeBs:
Fake base stations used for surveillance.
10.6
5G: Enhanced Security and Privacy
5G Authentication and Key Agreement (5G-AKA)
·
Based on EPS-AKA but enhanced.
·
Introduced SUCI (Subscription
Concealed Identifier) to protect IMSI.
·
Home network authenticates user
directly, reducing reliance on roaming operator.
Integrity Protection
·
Extended to user plane as well as
control plane.
Network Slicing Security
·
Each slice may have independent
security policies.
·
AI used for adaptive policy
enforcement.
Virtualization and Cloud Risks
·
NFV/SDN components exposed to cloud
vulnerabilities.
·
Zero-trust architectures emerging.
10.7 AI and Data
Science for Security
Anomaly Detection
·
ML models trained on traffic
patterns detect botnets, DDoS, or malware.
·
Supervised and unsupervised
learning for intrusion detection.
Threat Prediction
·
Predictive analytics flag unusual
signaling storms.
·
Reinforcement learning enhances
adaptive firewalls.
Privacy-Preserving AI
·
Federated learning trains security
models without centralizing sensitive data.
10.8 ASEAN and ANZ
Security Considerations
ASEAN
·
Some ASEAN countries still operate
2G → vulnerabilities remain.
·
Increased risk from state and
non-state actors exploiting outdated protocols.
·
Singapore leading in 5G
cybersecurity regulation.
ANZ
·
Australia implemented 5G
security bans on certain vendors citing supply chain concerns.
·
New Zealand adopting cloud security
frameworks for 5GC deployments.
10.9
Comparative Evolution of Security
Generation |
Auth Method |
Encryption |
Weaknesses |
1G |
None |
None |
Cloning,
eavesdropping |
2G |
SIM-based (A3/A8) |
A5/1,
A5/2 |
No mutual
auth, weak ciphers |
3G |
AKA |
Kasumi |
DoS,
jamming |
4G |
EPS-AKA |
AES/Snow3G |
IMS
vulns, rogue eNodeB |
5G |
5G-AKA |
256-bit
AES, integrity |
NFV/SDN
risks, slice complexity |
10.10
Summary
Security
has advanced from non-existent (1G) to mutual, encrypted,
privacy-preserving (5G). Yet, vulnerabilities evolve with technology: virtualized,
cloud-native 5G cores create new risks. AI and ML are indispensable in threat
detection, anomaly analysis, and adaptive security.
In
ASEAN/ANZ, the coexistence of legacy 2G with cutting-edge 5G
creates unique regional security challenges.
END
CHAPTER 10
Chapter
11 – Performance, QoS, and KPI Engineering
11.1
Introduction
Performance
engineering ensures that mobile and wireless networks meet user expectations
and service-level agreements (SLAs).
This
involves measuring Key Performance Indicators (KPIs), enforcing Quality
of Service (QoS) classes, and optimizing resources to balance coverage,
capacity, and cost.
With
the advent of AI and big data analytics, performance management has shifted
from reactive monitoring to predictive, self-optimizing systems.
11.2
Fundamental Performance Metrics
11.2.1 Coverage Metrics
RSRP
(Reference Signal Received Power): Measures coverage strength.
RSRQ
(Reference Signal Received Quality): Signal
quality relative to interference.
SINR
(Signal-to-Interference-plus-Noise Ratio): Key for
throughput estimation.
11.2.2 Capacity Metrics
Cell
throughput (DL/UL).
Spectral
efficiency (bps/Hz).
Number
of active users per cell.
11.2.3 Reliability Metrics
Block
Error Rate (BLER).
Call
Drop Rate.
Packet
Loss.
11.2.4 Latency and Jitter
Critical
for real-time applications (VoIP, gaming, URLLC).
11.3
QoS Frameworks
11.3.1 2G/3G Era
Best-effort
packet data in GPRS.
Circuit-switched
QoS for voice.
11.3.2
4G LTE
QoS
Class Identifier (QCI): Defines priority, delay, packet
loss.
Differentiates
services (e.g., VoLTE vs video streaming).
11.3.3
5G NR
- 5QI (5G QoS Identifier): Flexible, service-specific.
Example:
- URLLC:
ultra-low latency, high reliability.
- eMBB: high
throughput.
- mMTC:
massive IoT connections.
11.4
KPI Engineering
9.4.1 Radio Access KPIs
Handover
success rate.
RRC
(Radio Resource Control) setup success.
PRB
(Physical Resource Block) utilization.
11.4.2 Core Network KPIs
Session
setup success rate.
Mean
session duration.
Signaling
load distribution.
11.4.3 Service KPIs
VoLTE
MOS (Mean Opinion Score).
Video
start time, buffering ratio.
IoT
device availability.
11.5
AI and Data Science for Performance Optimization
11.5.1 Predictive Analytics
ML
models (LSTM, ARIMA) forecast traffic loads.
Seasonal
patterns detected in ASEAN/ANZ (e.g., urban rush hour vs rural IoT).
11.5.2 Root Cause Analysis
Data
clustering isolates fault patterns.
AI
distinguishes between hardware failure vs interference vs misconfiguration.
11.5.3 Self-Optimizing Networks (SON)
Dynamic
PCI allocation.
Automatic
load balancing.
AI-based
antenna tilt and power optimization.
11.5.4 QoE (Quality of Experience) Prediction
Data
fusion: combine KPI data with crowdsourced app-level metrics.
Example:
Telstra’s ML-driven app experience scoring.
11.6
ASEAN and ANZ Deployment Context
11.6.1 ASEAN
Dense
cities (Jakarta, Manila, Bangkok): KPI engineering focused on congestion and
interference mitigation.
Smart
city rollouts (Singapore): QoS tuned for IoT and AR/VR applications.
11.6.2 ANZ
Rural
coverage: KPIs focused on coverage continuity and satellite integration.
Mining/industrial
IoT: SLAs prioritize machine telemetry reliability over throughput.
11.7
Example KPI Targets (5G in ASEAN/ANZ)
KPI |
Target (Urban) |
Target (Rural) |
DL throughput |
1 Gbps+ |
100 Mbps+ |
Latency |
<10 ms |
<20 ms |
Handover success |
>99% |
>97% |
Call drop rate |
<0.5% |
<1% |
IoT availability |
99.9% |
99% |
11.8
Case Studies
11.8.1 Singapore
AI-enhanced
KPI monitoring in 5G standalone networks.
Network
slicing deployed for autonomous vehicle trials.
11.8.2 Australia
Telstra
using AI to balance capacity between urban eMBB and rural fixed
wireless access (FWA).
Integration
with Starlink for outback connectivity.
11.9
Challenges
- Data Overload:
Billions of KPI data points daily → big data processing required.
- Heterogeneity:
LTE/5G/LoRaWAN coexistence complicates KPIs.
- QoE Alignment: User
experience often diverges from network-centric KPIs.
- Security Risks:
Exposing KPI data pipelines may create vulnerabilities.
11.10
Summary
Performance
engineering evolved from basic voice KPIs in 2G to multi-layered
AI-driven frameworks in 5G.
- 2G/3G: best-effort packet data.
- 4G: structured QoS via QCIs.
- 5G: ultra-flexible QoS via 5QIs, AI-driven KPI
monitoring.
ASEAN
and ANZ deployments highlight contrasting needs: urban congestion vs rural
coverage. AI and ML are now indispensable for ensuring QoE, optimizing
KPIs, and predicting failures.
END OF CHAPTER 11
Chapter
12 – Deployment, Optimization, and Testing
12.1
Introduction
Building
a mobile or wireless system is not just about spectrum and standards; deployment
and optimization are the decisive factors that determine whether a network
meets expectations.
From
site acquisition and radio planning to drive testing and AI-assisted
optimization, network engineering requires a structured methodology. With
5G and beyond, automation and data-driven intelligence dominate, reducing
reliance on manual fieldwork.
12.2
Deployment Fundamentals
12.2.1 Site Selection
Coverage-driven
(rural ANZ): Maximize cell radius, fewer sites.
Capacity-driven
(urban ASEAN): Densify with small cells and HetNets.
Consider
zoning, environmental impact, and backhaul availability.
12.2.2 Spectrum Planning
Lower
bands (<1 GHz): coverage.
Mid-bands
(2–4 GHz): balance.
High
bands (>24 GHz mmWave): hotspots.
12.2.3 Backhaul
Microwave
links (cost-effective in ASEAN rural).
Fiber
(preferred in cities like Sydney, Singapore).
Satellite
integration for remote ANZ deployments.
12.3
Optimization Techniques
12.3.1 Radio Optimization
Antenna
tilt and azimuth adjustments.
Power
control: Reducing interference, extending battery life.
Carrier
aggregation balancing.
12.3.2 Interference Management
ICIC/eICIC
(Inter-Cell Interference Coordination).
Coordinated
multipoint (CoMP).
Beamforming
calibration in 5G.
12.3.3 Load Balancing
Cell
reselection parameters tuned dynamically.
Traffic
steering between LTE, 5G, and Wi-Fi.
12.4
Testing Methodologies
12.4.1 Drive Testing
·
Traditional: engineers drive routes
collecting RSRP, SINR, throughput.
·
Tools: TEMS, Nemo, Rohde &
Schwarz scanners.
·
Challenge: costly, limited
coverage.
12.4.2 Crowdsourced Testing
·
App-based KPI collection (Ookla,
OpenSignal).
·
Offers broader coverage, lower
cost.
·
Privacy and accuracy remain issues.
12.4.3 Lab Testing
·
Pre-deployment functional tests
(conformance, stress, regression).
·
Simulated RAN and core
environments.
12.4.4 Field Trials
·
Phased rollout in test markets.
·
Regulatory and safety approval
required.
12.5
AI and Data Science in Optimization
12.5.1 Automated Parameter Tuning
·
AI systems self-adjust handover
thresholds, antenna tilt, and power levels.
·
Reinforcement learning optimizes
beamforming in dense urban ASEAN networks.
12.5.2 Predictive Network Planning
·
ML models forecast demand (e.g.,
holiday surges in Manila).
·
Clustering algorithms identify
underperforming cells.
12.5.3 Fault Detection
·
AI detects anomalies in KPI
streams.
·
Self-healing RAN reroutes traffic
automatically.
12.5.4 Testing Automation
·
Digital twins simulate network
rollout before deployment.
·
AI accelerates regression testing
of core functions.
12.6
ASEAN and ANZ Deployment Context
12.6.1 ASEAN
·
Singapore: strong reliance on indoor
small cells and HetNets.
·
Indonesia: challenge of archipelagic
geography → satellite backhaul + microwave.
·
Vietnam/Thailand: government-driven
5G pilots with state-owned carriers.
12.6.2 ANZ
·
Australia: Telstra and Optus focus
on nationwide coverage, hybrid satellite integration.
·
New Zealand: Spark targets agriculture
IoT with precision deployment of LoRaWAN and LTE-M.
12.7
Comparative Testing Approaches
Method |
Cost |
Coverage |
Accuracy |
AI Integration |
Drive Test |
High |
Limited |
High |
Low |
Crowdsourced |
Low |
Wide |
Moderate |
High (data science pipelines) |
Lab Testing |
Medium |
Simulated |
High |
High |
Field Trial |
Medium |
Targeted |
High |
Medium |
12.8
Challenges in Optimization
- Heterogeneous networks (HetNets): Complicated interference patterns.
- Multi-vendor environments: Interoperability issues.
- Cost constraints:
Especially in emerging ASEAN markets.
- AI trustworthiness:
Models must be explainable, not black-box.
12.9
Case Studies
12.9.1 Singapore 5G Standalone
·
Heavy reliance on AI-automated
optimization.
·
Digital twin used for Marina Bay 5G
rollout.
12.9.2 Australia Remote Connectivity
·
AI-driven planning of
satellite-cellular hybrid networks.
·
Starlink integrated into outback
schools with predictive traffic balancing.
12.10
Summary
Deployment
and optimization evolved from manual field-driven engineering to AI-powered,
data-driven automation.
- 2G/3G: manual drive testing and parameter tuning.
- 4G: semi-automated SONs.
- 5G: digital twins, AI-optimized RAN, predictive
deployment.
In
ASEAN, urban density and geographic fragmentation dominate deployment
challenges. In ANZ, vast rural areas and industrial IoT drive reliance
on automation and satellite integration.
END
OF CHAPTER 12
Chapter
13 – Case Studies and Comparative Analysis
13.1
Introduction
Case
studies illustrate how different mobile and wireless technologies have been applied
in real-world contexts. Comparative analysis highlights strengths,
weaknesses, and regional suitability across ASEAN and ANZ.
This
chapter reviews use cases in cellular generations (2G–5G), Wi-Fi, LPWAN
(LoRaWAN), mission-critical systems (TETRA ATEX, ST GRID), and satellite
networks (Beidou, Starlink, Inmarsat, etc.).
13.2
Case Study 1 – GSM/2G
in ASEAN (1990s–2000s)
Deployment
- GSM widely deployed in ASEAN from 1993 onward.
- 900/1800 MHz bands.
- Enabled SMS culture in Southeast Asia.
Impact
- First affordable digital voice in rural and urban
areas.
- Led to economic transformation (mobile banking in
Philippines).
Challenges
- Weak encryption (A5/1 cracked).
- Legacy persistence: 2G still active in parts of
Myanmar, Cambodia.
13.3
Case Study 2 – UMTS/3G
in ANZ (2005–2015)
Deployment
- Telstra’s Next G network (850 MHz) in
Australia was world-leading.
- Extended 3G coverage across rural Australia.
Impact
- Internet access in previously isolated communities.
- Enabled agricultural IoT pilot projects.
Challenges
- High CAPEX for nationwide deployment.
- 3G sunset (2024–2025) creates migration
challenges.
13.4
Case Study 3 – 4G LTE
in ASEAN Megacities
Deployment
- Dense rollouts in Bangkok, Jakarta, Manila.
- Spectrum used: 1800 MHz, 2100 MHz, 2600 MHz.
Impact
- Mobile broadband penetration surged (>80% smartphone adoption).
- Gig economy (Grab, Gojek) flourished with LTE connectivity.
Challenges
- Congestion in dense urban environments.
- Interference management critical.
13.5
Case Study 4 – 5G in
Singapore and Australia
Deployment
- Singapore: 5G standalone core with slicing
(2020+).
- Australia: Telstra and Optus launched 5G FWA
(Fixed Wireless Access).
Impact
- Singapore: 5G-enabled autonomous vehicles, smart
ports.
- Australia: rural households gain high-speed
internet via 5G FWA + Starlink.
Challenges
- ASEAN: patchy adoption, limited spectrum in some
markets.
- ANZ: mmWave rollout limited by cost and coverage.
13.6
Case Study 5 – Wi-Fi 6
in ASEAN Enterprises
Deployment
- Wi-Fi 6 deployed in airports, universities,
business parks.
- Complementary to 5G indoor deployments.
Impact
- Supports enterprise digitalization.
- Singapore: hybrid Wi-Fi/5G private networks in
Changi Airport.
Challenges
- Congestion in 2.4 GHz band.
- Limited indoor spectrum coordination with
cellular.
13.7
Case Study 6 – LoRaWAN
in New Zealand Agriculture
Deployment
- LoRaWAN deployed for sheep/livestock monitoring,
irrigation.
- Coverage: 10–15 km from base stations.
Impact
- Improved productivity and water efficiency.
- Reduced OPEX due to battery life (10 years+).
Challenges
- Low data rates (not suitable for multimedia
telemetry).
- Dependence on unlicensed spectrum (interference
risk).
13.8
Case Study 7 – TETRA
ATEX in ASEAN Oil & Gas
Deployment
- ATEX-certified TETRA radios used on offshore rigs
in Brunei, Malaysia.
- Mission-critical communication in hazardous
zones.
Impact
- Safety compliance in flammable environments.
- Robust voice communications under extreme
conditions.
Challenges
- Not suitable for data-heavy IoT.
- Proprietary devices with high cost.
13.9
Case Study 8 – Starlink
in Rural ANZ
Deployment
- Starlink terminals installed in Australian
Outback schools and New Zealand farms.
- Latency ~30–40 ms, throughput 50–200 Mbps.
Impact
- Bridged digital divide in rural education.
- Enabled telehealth and online education.
Challenges
- Cost of terminals (~USD 600) limits adoption.
- Weather sensitivity (rain fade).
11.10
Comparative
Technology Table
Technology |
Latency |
Throughput |
Strengths |
Limitations |
ASEAN/ANZ Use Case |
GSM/EDGE |
150+ ms |
~0.2 Mbps |
Coverage, maturity |
Low data |
SMS-driven culture in ASEAN |
UMTS/HSPA |
50–100 ms |
~14–42 Mbps |
Proven, flexible |
Interference |
Telstra Next G rural coverage |
LTE |
20–50 ms |
100s Mbps–Gbps |
Ecosystem, efficiency |
Dense planning |
ASEAN megacities broadband |
5G NR |
<10 ms |
>1 Gbps |
Low latency, slicing |
Complexity |
Singapore smart nation |
Wi-Fi 6 |
<10 ms |
Multi-Gbps |
Cost-effective |
Interference |
Airports, enterprises |
LoRaWAN |
Seconds |
kbps |
Long range, IoT |
Low throughput |
NZ agriculture |
TETRA ATEX |
~100 ms |
Narrowband |
Mission-critical |
Not broadband |
Oil & gas ASEAN rigs |
Starlink |
30–40 ms |
50–200 Mbps |
Rural coverage |
Cost, weather |
ANZ schools & farms |
13.11
AI/ML in Case Study Applications
- ASEAN megacities (LTE/5G): AI-driven interference coordination.
- ANZ rural Starlink: AI
optimizes satellite beam allocation.
- LoRaWAN NZ farms: ML
models predict soil irrigation schedules.
- Singapore 5G slicing:
Reinforcement learning allocates slices dynamically.
13.12
Summary
Case
studies demonstrate the complementary role of mobile, Wi-Fi, LPWAN,
mission-critical, and satellite systems:
- 2G/3G laid
foundations but persist as vulnerabilities.
- 4G/5G
dominate urban ASEAN/ANZ broadband.
- Wi-Fi 6
complements indoor capacity.
- LoRaWAN
thrives in agriculture IoT.
- TETRA ATEX
ensures mission-critical safety.
- Starlink
bridges rural broadband gaps.
AI/ML
consistently enhances performance, reliability, and security across all case
studies.
END OF CHAPTER 13
Chapter
15 – 5G Advanced and the Road to 6G
15.1
Introduction
While
5G networks are still being deployed globally, research and
standardization have already shifted toward 5G Advanced (3GPP Release 18 and
beyond) and 6G (expected 2030).
The
focus is on making networks:
- AI-native
(intelligence embedded into the protocol stack).
- Sustainable
(green networking, energy efficiency).
- Ubiquitous
(integrated terrestrial, aerial, and satellite).
- Convergent
(communications, sensing, and computing combined).
15.2
5G Advanced (Release 18–20)
15.2.1 Key Enhancements
·
Extended MIMO:
Extremely large antenna arrays with AI beam management.
·
Positioning Accuracy:
Sub-meter precision for industrial IoT and autonomous vehicles.
·
XR Support:
Optimized for AR/VR with latency <5 ms.
·
Uplink-Centric Design:
Prioritizing sensor-heavy IoT and metaverse devices.
·
Non-Terrestrial Networks (NTN):
Native satellite integration (LEO, GEO, HAPS).
15.2.2 AI-Native RAN
·
AI models trained to optimize
scheduling, resource allocation, and interference coordination.
·
Network functions exposed as APIs
for real-time ML training.
15.2.3 ASEAN/ANZ Applications
·
ASEAN: Smart
factories in Malaysia, Thailand → low-latency positioning.
·
ANZ:
Agriculture robotics with uplink-heavy IoT, powered by AI-optimized 5G.
15.3
Toward 6G (IMT-2030)
15.3.1 Target Capabilities
·
Data Rates:
Up to 1 Tbps peak.
·
Latency:
Sub-millisecond (0.1–0.5 ms).
·
Sensing:
Integrated radar-like capabilities in radio beams.
·
AI-Native Architecture:
AI/ML embedded at protocol design, not as an add-on.
·
Sustainability:
Net-zero carbon networks.
15.3.2 Frequency Ranges
·
Sub-THz (100–300 GHz):
Provides extreme bandwidth but limited range.
·
Visible Light Communication (VLC):
Integration with optical systems.
15.3.3 Key Enablers
·
Reconfigurable Intelligent Surfaces
(RIS): Smart reflective panels steering signals dynamically.
·
Extreme MIMO:
Thousands of antenna elements.
·
Joint Communication and Sensing
(JCAS): Networks double as radar/sensors.
·
Quantum-Safe Security:
Resistant to post-quantum attacks.
15.4
AI and Data Science in 6G
- AI-Centric Design:
Scheduling, waveform selection, and topology self-optimized.
- Digital Twins:
Entire networks simulated in real-time for predictive optimization.
- Federated Learning:
Privacy-preserving AI training across devices.
- Self-Evolving Networks: Reinforcement learning adapts protocols on the fly.
15.5
Non-Terrestrial Networks (NTN)
15.5.1 Satellite Integration
·
5G Advanced → 6G:
Direct-to-device satellite service.
·
LEO constellations integrated with
cellular core networks.
15.5.2 ASEAN/ANZ Relevance
·
ASEAN: Disaster
recovery networks after typhoons/earthquakes.
·
ANZ: Outback
IoT (mining, agriculture) seamlessly using NTN.
15.6
6G Use Cases
15.6.1 Industrial & Autonomous Systems
·
ASEAN: Autonomous
ports in Singapore, Malaysia.
·
ANZ: Robotic
farming with AI-powered real-time sensing.
15.6.2 Extended Reality (XR) & Metaverse
·
Holographic communication with Tbps
uplink/downlink.
·
Seamless global presence for work
and entertainment.
15.6.3 Smart Cities & Environmental Monitoring
·
ASEAN megacities:
Real-time air pollution sensing + traffic management.
·
ANZ: Wildfire
detection via UAV-6G integrated systems.
15.7
Comparative Timeline
Generation |
Peak Rate |
Latency |
Key Innovations |
4G LTE |
1 Gbps |
20–50 ms |
All-IP,
MIMO |
5G NR |
10 Gbps |
<1 ms |
Slicing,
massive MIMO, NTN |
5G
Advanced |
~20 Gbps |
<0.5 ms |
AI-native
RAN, XR, precise positioning |
6G |
1 Tbps |
0.1–0.5 ms |
RIS,
JCAS, sub-THz, AI-centric |
15.8
Challenges
- Hardware Constraints:
Sub-THz requires new semiconductor materials.
- Energy Consumption:
Ultra-dense antennas must be energy-efficient.
- Standardization:
Global consensus on IMT-2030 framework.
- Security:
Quantum-resistant cryptography needed.
15.9
ASEAN and ANZ Roadmaps
15.9.1 ASEAN
·
Singapore, Thailand, Malaysia
actively participating in ITU 6G studies.
·
Focus on smart cities, industrial
IoT.
15.9.2 ANZ
·
Australia and New Zealand
prioritizing agriculture, mining, and remote connectivity.
·
Early adoption of NTN for vast
territories.
15.10
Summary
5G
Advanced strengthens current 5G with AI-native features,
uplink optimization, and satellite integration.
6G envisions AI as the network itself: Tbps speeds, near-zero
latency, sensing + communications convergence.
ASEAN
and ANZ will apply these differently:
·
ASEAN → dense urban smart
cities, Industry 4.0.
·
ANZ → agriculture, mining,
remote IoT with satellite integration.
END OF CHAPTER 15
Chapter
17 – Conclusion
17.1
Overview
This
thesis traced the technical evolution of mobile and wireless systems
from the analog 1G era of the 1980s to 5G and the trajectory toward 6G,
while also exploring parallel ecosystems such as Wi-Fi, LoRaWAN, TETRA
ATEX, ST GRID devices, and satellite networks (Beidou, Starlink, Inmarsat,
Gouwang, Qianfan).
Across
ASEAN and ANZ, the comparative analysis demonstrated that technology adoption
reflects regional socio-economic and geographic realities:
- ASEAN megacities →
dense LTE/5G rollouts, smart city integration.
- ANZ rural/remote regions → satellite and hybrid systems critical for inclusion.
17.2
Key Findings
17.2.1 Generational Progression
- 1G: Introduced cellular
concept, but no security.
- 2G: Digitalization, SMS
culture, basic security.
- 3G: Mobile broadband
foundation, yet uneven global adoption.
- 4G LTE:
All-IP, high capacity, foundation for the gig economy.
- 5G NR:
AI-ready, ultra-low latency, slicing, satellite integration.
17.2.2 Specialized Systems
- LoRaWAN: Ideal
for long-range IoT, but limited throughput.
- TETRA ATEX / ST GRID:
Mission-critical, ruggedized comms for safety industries.
- Satellite Systems:
Complement terrestrial, crucial for rural ANZ and disaster-stricken ASEAN.
17.2.3 AI and Data Science
Now
central to:
Ø
Traffic prediction.
Ø
Security anomaly detection.
Ø
Automated optimization (SON,
digital twins).
Ø
QoS and QoE assurance.
17.3
Regional Insights
ASEAN
- Diversity in deployment maturity: Singapore (5G advanced),
Cambodia/Myanmar (legacy 2G still active).
- Smart city initiatives driving IoT adoption.
- Geographical fragmentation (archipelagos like Indonesia,
Philippines) → satellite + terrestrial hybrids.
ANZ
- Vast landmass and low population density → heavy
reliance on macrocells + satellite.
- Strong integration of LoRaWAN and LTE-M in
agriculture.
- Starlink + Telstra partnerships extend digital
equity in Australia’s Outback.
17.4
Future Outlook
17.4.1 5G Advanced
Ø
Strengthens positioning, uplink,
XR, and NTN.
Ø
AI becomes intrinsic to RAN/core
functions.
17.4.2 6G Vision
Ø
1 Tbps peak rates, 0.1 ms latency.
Ø
Integrated communication + sensing.
Ø Reconfigurable Intelligent Surfaces (RIS), sub-THz
spectrum.
Ø
Quantum-safe, AI-native networks.
ASEAN
will leverage 6G for urban industrial automation, while ANZ will apply
it to agriculture, mining, and remote connectivity.
17.5
Research Methodology Reflection
- Historical review:
Standards evolution analyzed from 3GPP, ITU, IEEE documentation.
- Comparative analysis:
Technologies benchmarked in terms of latency, throughput, security, and
use cases.
- Case studies:
ASEAN/ANZ deployments studied from operator whitepapers, government
policies, and academic research.
- AI/ML integration:
Evaluated through contemporary research on SONs, digital twins, and
anomaly detection.
17.6
Final Thoughts
The
central theme is convergence:
- Cellular, Wi-Fi, LPWAN, and satellite are not
competing but complementary.
- AI/ML ensures networks become adaptive,
autonomous, and predictive.
- Future 6G will be a cognitive fabric —
sensing, computing, and communicating seamlessly.
The
evolution from 1G to 5G and beyond is not merely technological but also socio-economic,
shaping digital inclusion in ASEAN and ANZ. The challenge for researchers and
policymakers is ensuring that technological sophistication aligns with
accessibility, affordability, and security.
REFERENCES & BIBIOGRAPHY
[1]
Canonical: 3GPP TS 23.501,
"System Architecture for the 5G System (5GS)"
URL:
https://www.3gpp.org/ftp/Specs/archive/23_series/23.501/
Used In Thesis Section(s):
Ch.8 Core Networks (MSC→EPC→5GC), Ch.6 Standards (5G), Ch.13 5G-Advanced
Referenced Statement: 5GC
is a cloud-native Service-Based Architecture (SBA) with functions such as AMF,
SMF, UPF, NRF, PCF communicating via service APIs.
[2]
Canonical: 3GPP TS 38.300,
"NR; NR and NG‑RAN overall description"
URL:
https://www.3gpp.org/ftp/Specs/archive/38_series/38.300/
Used In Thesis Section(s):
Ch.6 Standards (5G NR), Ch.8 Core Networks (RAN ↔ Core interfaces)
Referenced Statement: Overall
description of 5G NR/NG‑RAN, including architecture, protocol layers, and
reference signals.
[3]
Canonical: 3GPP TS 36.300,
"E‑UTRA and E‑UTRAN; Overall description; Stage 2"
URL:
https://www.3gpp.org/ftp/Specs/archive/36_series/36.300/
Used In Thesis Section(s):
Ch.5 Topologies (macro/micro/pico), Ch.6 Standards (LTE), Ch.4
Modulation/Access
Referenced Statement: LTE uses OFDMA
(downlink) and SC‑FDMA (uplink) and defines key E‑UTRAN procedures.
[4]
Canonical: ETSI GSM 05.05
(3GPP TS 45.005), "Radio transmission and reception"
URL:
https://www.etsi.org/deliver/etsi_gts/05/0505/05.05.01_60/ |
https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=1448
Used In Thesis Section(s): Ch.2
Timeline (2G), Ch.6 Standards (GSM)
Referenced Statement: GSM
carrier spacing is 200 kHz; baseline RF characteristics for GSM.
[5]
Canonical: 3GPP HSPA
(Release 5/6) — see TS 25.308 / TS 25.306; summary: "High Speed Packet
Access (HSDPA/HSUPA)"
URL:
https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=1411
Used In Thesis Section(s):
Ch.2 Timeline (3G), Ch.6 Standards (UMTS/HSPA)
Referenced Statement: HSDPA
introduced downlink peak rates on the order of ~14.4 Mb/s with link adaptation
and fast scheduling.
[6]
Canonical: 3GPP TR 38.901,
"Study on channel model for frequencies from 0.5 to 100 GHz"
URL:
https://www.3gpp.org/ftp/Specs/archive/38_series/38.901/
Used In Thesis Section(s):
Ch.3 Propagation & Link Budget, Ch.6 Standards (5G NR), Ch.13 5G‑Advanced
Referenced Statement: NR
channel modeling across sub‑6 GHz and mmWave, including path loss, delay
spread, and mobility (Doppler) effects.
[7]
Canonical: ITU‑R P.525‑4,
"Calculation of free-space attenuation"
URL:
https://www.itu.int/rec/R-REC-P.525/en
Used In Thesis Section(s):
Ch.3 Propagation & Link Budget
Referenced Statement: Free‑space
path loss (FSPL) formula used as baseline for link budgets.
[8]
Canonical: 3GPP TS 33.501,
"Security architecture and procedures for 5G System"
URL:
https://www.3gpp.org/ftp/Specs/archive/33_series/33.501/
Used In Thesis Section(s):
Ch.9 Security & Authentication, Ch.6 Standards (5G), Ch.8 Core Networks
Referenced Statement: 5G‑AKA,
SUCI (concealed SUPI), and mapping of integrity/confidentiality algorithms
(e.g., NEA1/2/3, NIA1/2/3).
[9]
Canonical: NIST
Cybersecurity White Paper — SUCI Overview (supporting 5G privacy)
URL:
https://csrc.nist.gov/publications
Used In Thesis Section(s):
Ch.9 Security & Authentication
Referenced Statement: Rationale
and mechanism for protecting subscriber identifiers via SUCI to mitigate
IMSI‑catcher attacks.
[10]
Canonical: LoRa Alliance,
"LoRaWAN® L2 1.0.4 Specification"
URL:
https://lora-alliance.org/resource_hub/lorawan-specification-1-0-4/ |
https://resources.lora-alliance.org/
Used In Thesis Section(s):
Ch.7 Specialized Systems (LoRaWAN), Ch.6 Standards (LPWAN)
Referenced Statement: Device
classes (A/B/C), MAC behavior, and certification references for LPWAN
deployments.
[11]
Canonical: IEEE 802.11ax
(Wi‑Fi 6) standard — overview via vendor technical white papers
URL:
https://www.cisco.com/c/en/us/solutions/collateral/enterprise-networks/802-11ax-solution/white-paper-c11-740788.html
Used In Thesis Section(s):
Ch.6 Standards (Wi‑Fi), Ch.10 Performance/QoS (dense deployments)
Referenced Statement: OFDMA,
BSS Coloring, and Target Wake Time (TWT) as key Wi‑Fi 6 features for efficiency
in dense environments.
[12]
Canonical: LANCOM Systems
Whitepaper (Wi‑Fi 6)
URL:
https://www.lancom-systems.com/wi-fi6-whitepaper/
Used In Thesis Section(s):
Ch.6 Standards (Wi‑Fi)
Referenced Statement: Practical
explanations of 802.11ax mechanisms (OFDMA, BSS Coloring) and deployment
guidance.
[13]
Canonical: IEEE 802.16e‑2005
(Mobile WiMAX) — mobility/QoS extensions
URL:
https://standards.ieee.org/standard/802_16e-2005.html |
https://ws680.nist.gov/publication/get_pdf.cfm?pub_id=150568
Used In Thesis Section(s):
Ch.6 Standards (WiMAX), Ch.2 Timeline (4G contenders)
Referenced Statement: Mobile
WiMAX extensions enabling mobility, QoS, and PHY/MAC enhancements; historical
4G context.
[14]
Canonical: 3GPP TR 38.821,
"Solutions for NR to support non‑terrestrial networks (NTN)"
URL:
https://www.3gpp.org/ftp/Specs/archive/38_series/38.821/
Used In Thesis Section(s):
Ch.6 Standards (5G/NTN), Ch.7 Specialized Systems (Satellites), Ch.13 Road to
6G
Referenced Statement: Architectural
and PHY/MAC adaptations for NR over LEO/HAPS, including timing/Doppler
considerations.
[15]
Canonical: 3GPP TS
38.211/38.213/38.214 (NR PHY, control, and procedures) — numerology
URL:
https://www.3gpp.org/ftp/Specs/archive/38_series/38.211/
Used In Thesis Section(s):
Ch.4 Modulation/Coding/Multiple Access, Ch.6 Standards (5G NR)
Referenced Statement: NR
subcarrier spacing Δf = 15×2^μ kHz, slot/frame structure, MCS tables and link
adaptation context.
[16]
Canonical: ETSI/3GPP
Algorithm specifications (Snow3G, AES‑CTR, ZUC) — mapped via TS 33.501 Annexes
URL:
https://www.3gpp.org/ftp/Specs/archive/35_series/ |
https://www.3gpp.org/ftp/Specs/archive/33_series/33.501/
Used In Thesis Section(s):
Ch.9 Security & Authentication
Referenced Statement: Families
of confidentiality (NEA1/2/3) and integrity (NIA1/2/3) algorithms used in
LTE/NR.
[17]
Canonical: ITU‑R P.838‑3
(specific attenuation model for rain) and P.840‑8 (rain height/cloud models)
URL:
https://www.itu.int/rec/R-REC-P.838/en
|
https://www.itu.int/rec/R-REC-P.840/en
Used In Thesis Section(s):
Ch.3 Propagation (mmWave rain fade), Ch.13 5G‑Advanced (FR2 planning)
Referenced Statement: Quantitative
rain attenuation models applicable to FR2/mmWave planning.
[18]
Canonical: RF Cafe
(historical engineering tables) — AMPS 30 kHz channelization
URL:
https://www.rfcafe.com/references/electrical/wireless-communications-standards.htm
Used In Thesis Section(s):
Ch.2 Timeline (1G), Ch.6 Standards (1G)
Referenced Statement: AMPS
channel spacing of 30 kHz and associated band plans (historical reference).
[19]
Canonical: 3GPP TS 23.401
(EPC); IMS/VoLTE (TS 24.229); 24/26 series for SIP/codec
URL: https://www.3gpp.org/ftp/Specs/archive/23_series/23.401/ |
https://www.3gpp.org/ftp/Specs/archive/24_series/24.229/ |
https://www.3gpp.org/ftp/Specs/archive/26_series/
Used In Thesis Section(s):
Ch.8 Core Networks (LTE/EPC), Ch.10 Performance/QoS
Referenced Statement: EPC
split (MME, SGW, PGW) and IMS/VoLTE SIP-call control with codec implications.
[20]
Canonical: 3GPP TS
36.212/36.213 (LTE PHY coding & modulation; link adaptation)
URL:
https://www.3gpp.org/ftp/Specs/archive/36_series/
Used In Thesis Section(s):
Ch.4 Modulation/Coding (4G), Ch.10 Performance/QoS
Referenced Statement: Coding/modulation
schemes, HARQ, higher-order QAM, and MCS tables used for LTE/LTE‑A throughput
claims.
[21]
Canonical: 3GPP TS
45.002/45.003 (GSM physical channel, modulation)
URL:
https://www.3gpp.org/ftp/Specs/archive/45_series/
Used In Thesis Section(s):
Ch.4 Modulation/Coding (2G), Ch.2 Timeline (2G)
Referenced Statement: GMSK
modulation, time-slot/multiframe structure for GSM; groundwork for GPRS/EDGE
evolution.
[22]
Canonical: 3GPP TS
25.331/25.211 (UMTS/W‑CDMA RRC & PHY)
URL:
https://www.3gpp.org/ftp/Specs/archive/25_series/
Used In Thesis Section(s):
Ch.4 Modulation/Coding (3G), Ch.2 Timeline (3G)
Referenced Statement: W‑CDMA
spreading/channelization and RRC procedures; HSPA context.
[23]
Canonical: IEEE 802.11be
(Wi‑Fi 7) – overview (IEEE/vendor)
URL:
https://standards.ieee.org/ieee/802.11be/7421/
|
https://www.intel.com/content/www/us/en/products/docs/wireless/wi-fi-7.html
Used In Thesis Section(s):
Ch.6 Standards (Wi‑Fi evolution), Ch.10 Performance/QoS
Referenced Statement: 320‑MHz
channels and Multi‑Link Operation (MLO) as Wi‑Fi 7 advancements.
[24]
Canonical: LoRa Alliance —
Regional Parameters (RP2‑1.0.4)
URL:
https://lora-alliance.org/resource_hub/rp2-1-0-4-lorawan-regional-parameters/
Used In Thesis Section(s):
Ch.7 Specialized Systems (LoRaWAN), Ch.3 Spectrum (ISM bands)
Referenced Statement: Regional
channel plans, duty‑cycle, and EIRP constraints for LoRaWAN.
[25]
Canonical: ITU IMT documents
(IMT‑Advanced & IMT‑2020 requirements/definitions)
URL:
https://www.itu.int/en/ITU-R/terrestrial/im/pages/im.aspx
Used In Thesis Section(s): Ch.2
Timeline (4G/5G definitions), Ch.6 Standards
Referenced
Statement: Formal generational requirements and performance frameworks for
IMT‑Advanced (4G) and IMT‑2020 (5G).
Image Credits
Credits to image owners
under Creative Commons or taken in full context
https://collection.sciencemuseumgroup.org.uk/objects/co430373/motorola-startac-mobile-phone-1997.
POST
GRADUATE PATHWAY
KEY TAKE AWAY IMAGES
Abstract
& Intro
- Timeline overview (1980s–5G).
- Conceptual diagram: radio systems ecosystem
(cellular, Wi-Fi, LPWAN, satellite).
- World map showing ASEAN & ANZ focus.
Chapter
1 – Introduction
- Cellular architecture basics.
- Comparative chart: wireless vs wired.
- Illustration of spectrum evolution.
Chapter
2 – Historical Timeline
- 1G–5G evolution timeline.
- Generational throughput/latency chart.
- Device evolution pictorial (brick phone →
smartphone → IoT).
Chapter
3 – Spectrum & Regulation
- 900 MHz propagation graph.
- ASEAN spectrum allocation table (visualized).
- ANZ spectrum licensing diagram.
Chapter
4 – Propagation & Link Budget
- Path loss models (free space vs Hata).
- Link budget block diagram.
- Antenna/MIMO illustration.
Chapter
5 – Network Topologies
- Macro/micro/pico/femto cell hierarchy.
- HetNet diagram.
- Mesh/LPWAN star-of-stars.
Chapter
6 – Standards by Generation
- Table: 1G–5G standards comparison.
- Wi-Fi evolution chart.
- LoRaWAN architecture.
Chapter
7 – Specialized Systems
- TETRA ATEX rugged radio pictorial.
- ST GRID hybrid phone diagram.
- Satellite constellations map (Beidou, Starlink,
Inmarsat).
Chapter
8 – Core Networks
- MSC → EPC → 5GC evolution.
- Service-based architecture diagram.
- NFV/SDN cloud-native illustration.
Chapter
9 – Security & Authentication
- Authentication methods evolution (1G–5G).
- Attack surface diagram (IMSI catcher, rogue base
station).
- AI-based intrusion detection pipeline.
Chapter
10 – Performance & QoS
- KPI dashboard mockup.
- QCI/5QI mapping table.
- Throughput/latency scatterplot.
Chapter
11 – Deployment & Testing
- Drive test vs crowdsourced testing diagram.
- Digital twin illustration.
- Optimization workflow (AI loop).
Chapter
12 – Case Studies
- ASEAN adoption map.
- ANZ rural broadband coverage.
- Comparative chart: LTE vs Starlink vs LoRaWAN.
Chapter
13 – 5G Advanced & 6G
- RIS (reconfigurable
intelligent surfaces) pictorial.
- Sub-THz spectrum chart.
- 6G vision roadmap (timeline).
Chapter
14 – Conclusion
- Pyramid diagram (1G foundation → 6G cognitive
fabric).
- ASEAN vs ANZ summary chart.
- Future integration flowchart (AI + satellite +
terrestrial).