Artificial intelligence based 5 g communication dr.k.thilagam

5G COMMUNICATION AND ITS
RESEARCH PERSPECTIVES
Dr.K.THILAGAM
ASSOCIATE PROFESSOR
DEPARTMENT OF ECE
VELAMMAL ENGINEERING COLLEGE

CONTENTS
• INTRODUCTION
• ABOUT 5G
• 5G CHALLENGES
• 5G REQUIREMENTS
• THE WAY TO 5G
• ABOUT 5G IN INDIA
• RESEARCH PERSPECTIVES
• 5G SOFTWARE TOOLS
• APPLICATIONS
• CONCLUSION
• REFERENCES

INTRODUCTION
• Nowadays, every individual is having mobile phone and
they are active in all the social medias like Whatsapp,
Face book, Twitter, Instagram, etc.
• Due to the proliferation of mobile devices and the
advancement in wireless technologies ,
• there is an increase in Demand for access to information
- anytime, anywhere and anyplace .
• This shows, how much the day today life is dependent
on radio spectrum and also on QoS.
• Moreover, Quality of service (QoS) is an important
consideration in all networks and it is also a significant
challenge

QUALITY OF SERVICE (QoS)
• Providing guaranteed QoS is a challenging issue.
• Reasons: Constraints and heterogeneities like power
constraints, bandwidth limitation, fading effects, different
protocols and standards.
• In addition to above said reasons, it should meet the
demands of multimedia applications.
4

PROVISIONING OF QoS
 Challenges still prevail in provisioning of QoS, to meet
(i) Improved Reliability
(ii) Higher capacity demands
(iii) Continuity of services
 Some of the critical Performance metrics are:
(i) Bit Error Rate
(ii) Spectral and Power efficiency
(iii) Outage Probability
5

POSSIBLE SOLUTIONS
Future generation networks
5G
6

COMPARISON OF 1G TO 5G TECHNOLOGIES

5G: SYMBIOTIC INTEGRATION OF EXISTING AND
NEW TECHNOLOGIES

Artificial intelligence based 5 g communication dr.k.thilagam

5G: KEY TECHNOLOGICAL COMPONENTS
• 5G wireless will support a heterogeneous set of integrated
air interfaces
• Seamless handover between heterogeneous wireless
access technologies will be the feature of 5G
• It uses simultaneous radio access technologies to increase
reliability and availability.
• 5G will ease and optimize network management
operations through proper algorithms

5G: KEY TECHNOLOGICAL COMPONENTS
• 5G will be driven by software.
• Network functions are expected to run over a unified
operating system in a number of points of presence,
especially at the edge of the network for meeting
performance targets.
• It relies on emerging technologies such as
(i)Software Defined Networking (SDN)
(ii) Network Functions Virtualization (NFV)
(iii) Mobile Edge Computing (MEC)
(iv) Fog Computing (FC)
to achieve the required performance.

The WAY TO 5G
• 5G Architecture
• Frame Structure
• Air Interface and Modulation
• Interference control
• Massive MIMO
• Spectrum and Propagation
• Millimeter wave communication
• Hardware Implementation

TO BE DISCUSSED
• 5G Architecture
• Artificial Intelligence (AI) -Based 5GArchitecture
• Massive MIMO
• Millimeter wave communication
• Ultra Dense Network (UDN)

ARTIFICIAL INTELLIGENCE-BASED 5G
• The International Telecommunication Union (ITU)
classifies 5G into three categories of usage scenarios:
• enhanced mobile broadband (eMBB)
• massive machine-type communication (mMTC)
• ultra-reliable and low latency communication
(URLLC)
• to account for more diverse services and resource
hungry applications.

• eMBB - addresses bandwidth-hungry applications,
such as massive video streaming and
virtual/augmented reality (VR/AR).
• mMTC - enabling massive sensing, monitoring, and
metering in support of the massive deployments of the
Internet of Things (IoT).
• URLLC -that supports latency sensitive services
including autonomous driving, drones and the tactile
Internet.

• artificial intelligence (AI)-defined 5G radio access networks
support those unprecedented requirements
• leverage the emergence of mobile edge computing and
caching, smart cities, and context-aware networking.

APPLICATION SCENARIOS OF AI-DEFINED 5G NETWORKS

• An AI-defined 5G network provides the base stations
(BSs)/Cloud with the capability to create a
• cognitive and comprehensive data repository by
splitting, processing and interpreting the operational
data.
• Massive amounts of real-time data are generated
across a large number of users and ranges from
channel state information (CSI) to IoT device
readings.
• The received data and geolocation databases are fused
to derive a complete understanding of the
environment.

• From a machine-centric communication perspective,
big data analytics are leveraged to extract
• massive patterns, especially at the physical (PHY) and
medium access control (MAC) layers and enable self
organizing operations.
• Neural networks can be used to redefine cognitive link
adaptation, resource scheduling, signal classification,
and carrier sensing/collision detection, among others.

• An RNN has also the capability to capture and
mitigate the imperfections and nonlinearities of radio
frequency (RF) components,
• such as high-power amplifiers (HPAs), which incur at
the PHY and can affect the network performance.
• DBN and CNN are better suited for resolving a range
of upper communication layer tasks such as network
optimization and resource management.

5G USE CASES, KEY TECHNOLOGY ENABLERS AND
TYPICALLY APPLICABLE AI PRINCIPLES

• Figure shows the critical 5G use cases and the
application of different types of neural networks as key
5G enablers.
• Massively parallel processing devices such as graphic
processing units (GPUs) and neural network processing
units (NPUs)
• It can be deployed in a 5G BS/Cloud.
• enable off loading the traditional data processing and
have shown to be highly energy-efficient for neural
network learning and inference .
• This shows the benefits of an AI framework to satisfy
the technical challenges of 5G.

Massive MIMO
• Its an extension of MIMO, which essentially groups
together antennas at the transmitter and receiver to
provide better throughput and spectrum efficiency

Massive MIMO
• Massive MIMO is widely perceived as a leading
candidate technology for 5G.
• The high number of BS antennas requires a large
number of PAs, one per antenna.
• The primary problem in RF PAs is known as the design
trade-off between linearity and efficiency:
• amplifiers can be designed to attain good linearity at the
cost of efficiency

• One of the main advantages of ANNs is their ability to
extract, predict, and characterize nonlinearities from
massive datasets.
• ANNs can capture variety of time-varying inputs and
outputs,
• such as complex nonlinear relationships or highly
dynamic relationships,
• They tackle nonlinearity problems at the PHY,
including PA nonlinearity tracking, predistortion and
impairment correction.

MMWAVE COMMUNICATIONS
• The demand for higher end-user data rates and higher
traffic density drive 5G networks to provide a
corresponding capacity boost.
• Millimeter wave (mmWave) systems have access to a huge
amount of available bandwidth in extremely high
frequency bands,
• between 30 and 300 GHz, to enable multi-gigabit-per-
second rate connectivity.

MILLIMETER (mm) WAVE TECHNOLOGY
(cm/mm Waves)

1890’s J.C. Bose Invented millimeter wave.
28 GHz and 38 GHz are the current frequencies that have low rainfall
attenuation & atmospheric attenuations.

• mmWave communications are facilitated by highly
directional communication links,
• where narrow beams can focus the energy to the
intended receivers, reduce PA requirements and
suppress interference from other transmitters.
• Whereas the mmWave spectrum may not be scarce, it
is also not very reliable due to its propagation
characteristics.
• At mmWave frequencies severe channel variations
occur due to blockage by cars, trees, solid building
material, and even human body parts.

• mm-wave frequencies could be used to augment the
currently saturated 700 MHz to 2.6 GHz radio spectrum
bands for wireless communications.
• By increasing the RF channel bandwidth for mobile radio
channels,
• the data capacity is greatly increased, while the latency
for digital traffic is greatly decreased.
• thus supporting much better internet based access and
applications that require minimal latency.
• Mmwave frequencies, due to the much smaller
wavelength, may exploit polarization
• and new spatial processing techniques, such as massive
MIMO and adaptive beam forming.

MAIN BENEFITS OF MILLIMETER WAVE
TECHNOLOGY
• The main benefit that millimeter Wave technology has
over RF frequencies is the spectral bandwidth of
5GHz being available in these ranges,
• resulting in current speeds of 1.25Gbps Full Duplex
with potential throughput speeds of up to 10Gbps Full
Duplex being made possible.
• Service providers can significantly expand channel
band width way beyond 20 MHz

NETWORK ULTRA-DENSIFICATION
• It is envisioned that the explosive data traffic
requirement in 5G will be largely dependent
• on the deployment of ultra-dense small cells the cell
densification is redeemed
• as one of the most aggressive ways to improve system
capacity as long as the interference can be efficiently
managed.
• Co-channel interference mitigation in 5G is naturally
enabled through massive MIMO beamforming and
mmWave transmission.

• AI-enabled interference mitigation based on
radio network prediction, evolution and
Optimization
• AI agent allows better fairness and end-user
experience

AI BASED architecture in 5G networks and
performance comparisons

• A reconfigurable deep learning framework based on
our proposed AI-aided 5G BS system is shown in Fig
for interpreting the signaling environment.
• It includes an offline training phase (including training
and fine-tuning) and an online spectrum access phase.
• This framework leverages the fact that the trained
model (especially the weights) can be efficiently
stored to facilitate online spectrum access decisions in
real time.

• To demonstrate this framework, we use a stack of
RBMs to train the DBN in a layer by- layer manner to
reduce the training complexity.
• The network first models the input datasets of
spectrum and regulation with the parameters of the
first layer of RBM.
• It then trains the subsequent layers of RBM with the
samples from the output of the previous layer.
• The contrastive divergence of the 1-step iteration (CD-
1) method is applied to update the
• weights in each layer thus allowing to find the
parameters for all layers of the RBM.

• In the online spectrum access phase, a spectrum
decision is made to generate proper transmission
schemes,
• that is, a suitable access method, carrier frequency and
bandwidth, modulation and coding scheme, and
transmission power.
• demonstrate how DBN is applied to detect the signals
in the 3.5 GHz band in high and low signal to noise
ratio (SNR) regimes.

ABOUT 5G IN INDIA
How are Indian operators preparing for 5G?
• India’s network operators have applied to the government
for permission to conduct field trials and selected one or
more equipment vendors with which to collaborate.
• They are also enhancing their network cores to prepare
for the additional traffic loads that 5G radio interfaces
will carry.

• In July 2020, Google said it would invest Rs
33,737 crore (USD 4.5 billion) in Jio to support the
operator’s upgrade.
• Jio also has support from Qualcomm and Intel for its
5G plan.
• The latter two are Jio’s hardware enablers: Chipmaker
Qualcomm is expected to offer deep technology know-
how and insights to drive the 5G vision.

• Airtel has said it will work with Huawei, ZTE,
Ericsson, and Nokia on its trials.
• It has deployed 100 hops of 5G technology
transmission equipment supplied by Huawei,
improving its backhaul capacity by a factor of four,
and has struck deals with Cisco and Ericsson to speed
up its core network in readiness for 5G service.
• Vodafone Idea will conduct its trials with Huawei,
ZTE, Ericsson, and Nokia, and is already using 5G AI
technology from Huawei to boost the capabilities of its
4G network.

What spectrum will 5G use in India?
• Days after announcing Google’s investment, Jio
approached the Department of Telecommunication
(DoT) to assign 5G spectrum for field trials.
• It has request 800 MHz of spectrum in the 24 GHz and
26 GHz bands and 100 MHz of spectrum in the 3.5
GHz band for running trials in Delhi and Mumbai.
• The Telecom Regulatory Authority of India (TRAI)
has recommended that spectrum be made available for
5G services is the 3300 to 3400 MHz and 3425 to
3600 MHz bands.

When will 5G services be available in India?
• While 5G phones are already available in India from
Realme and iQoo, both Chinese-owned.
• 5G networks were once expected to be launched in
India by late 2022.
• Telecom testing solutions provider Keysight
Technologies has inked a partnership with the Indian
Institute of Technology, Madras (IIT-M)
• to assist the Telecom Standards Development Society,
India (TSDSI) in developing India specific 5G
Standard.

• The telecom department (DoT) has initiated the funding of
large-scale ‘5G Testbed’ project with an aim to take an early
lead in the 5G technology and to build a testbed that resembles
real-world 5G deployment.

RESEARCH DEVELOPMENTS IN IMPLEMENTING 5G
NETWORK SOLUTIONS
• Multiple access and advanced waveform technologies
combined with coding and modulation algorithms
• Interference management
• Access protocols
• Service delivery architecture
• Massive MIMO
• Single frequency full duplex radio technologies
• 5G devices
• Virtualized and cloud-based radio access infrastructure

RESEARCH DEVELOPMENTS IN IMPLEMENTING 5G
NETWORK SOLUTIONS
• Seamless Spectrum Handovers
• Proactive Spectrum selection
• Energy Efficiency
• Service delivery architecture
• Network function virtualization, environmental awareness
and security challenges, etc., expected to have a
significant impact on ongoing AI-defined 5G
network research.

5G SIMULATION TOOLS
• Network Simulator –NS 2, NS 3
• OMNeT++ -Model for Ad hoc Network Simulations
• QualNet - Network Simulation Software
• Vienna 5G Link Level Simulator
• 5G Core and RAN Network Emulator
• MATLAB - communication simulator, NetSim - Network
level

5G SIMULATION TOOLS
• Matlab, Simulink and C++ programming.
SEAMCAT: http://www.seamcat.org/
• NetSim from Tetcos for simulation.
• NYUSIM: The Open Source 5G and 6G Channel
Model Simulator software
• 5G Antenna Designing Tool- IE3D, CST, HFSS,
etc.

CONCLUSION
• The successful migration from LTE-Advanced to 5G
and the effective integration of key radio access
technology is important.
• It is identified as the most promising for improving
system performance in both lower and higher
frequency bands.
• The built-in scalability and flexibility of the core
network are required to support wider range of QoS
and massive number of devices.

CONCLUSION
• 5G will not be just an evolution of 4G!
• A whole set of new Technologies are needed
• Architectural changes ,Novel Hardware , New
spectrum
• It is not going to be an easy task
• Good News is Plenty of research work to be done.
• Hopefully jobs, research grants, start-up opportunities,
etc...lots are there…

REFERENCES
[1] Miao Yao, Munawwar Sohul, Vuk Marojevic, and Jeffrey H. Reed, “Artificial
Intelligence-Defined 5G Radio Access Networks”, IEEE Communications
Magazine, March 2019.
[2] 5G Vision, NOKIA White paper 2015
[3]Ericsson, “5G Radio Access, Research and Vision,” white paper, 2013.
[4] Metis, “Scenarios, requirements and KPIs for 5G mobile and wireless
system,” ICT-317669 METIS project, May 2013.
[5]Ekram Hossain, Mehdi Rasti “Evolution Towards 5G Multi-tier Cellular
WirelessNetworks: An Interference Management Perspective”, February 2014
[6]Asvin Gohil and Hardik Modi, “5G Technology of Mobile Communication:A
Survey”, ISSP 2013
[7]. Roadmap and workplan on future technologies(2020) from 3GPP, ITU,
WRC, APT, CJK, China IMT2020, etc.

REFERENCES
[8]www.5g-ppp.eu/roadmaps
[9]NikiUpaddyay, “5G wireless
technology”Available:http://www.slideshare.net/upadhyayniki/5-
g-wireless-system
[10] Life emitates Arts:Mas and 5G
Available:http://aresproject.com/life-imitates-art-mas-and-5g/
[11] 5G mobile terminal network Available:
http://www.scribd.com/doc/84304292/Concepts-for-5g-Mobile-
Networks
[12] 5G mobile Technology Abstract
Available:http://www.seminarsonly.com/Labels/5g-Mobile-
Technology-Abstract.php

Artificial intelligence based 5 g communication dr.k.thilagam

More Related Content

What's hot (20)

Similar to Artificial intelligence based 5 g communication dr.k.thilagam (20)

Recently uploaded (20)

Artificial intelligence based 5 g communication dr.k.thilagam