Workshop Series on Advanced

Communication Systems

Intelligent Reflecting Surfaces

Session I

Dr. Rajarshi Mahapatra

Associate Professor & Dean (Academics)
IIIT Naya Raipur
Email: rajarshi@[Link]
Introduction
Growing number of user
 Popularity of wireless communication
 Reachability, Mobility, Portability
 At Anytime in Anywhere with Anyone for Anything
 Number of mobile user is increasing rapidly
 Cross 480 crores (73 crores in India) worldwide by 2017
 Demand of high QoS is growing continuously
 High speed Internet (facebook, twitter, Whatsapp)
 Un-buffering Multimedia video (Youtube, BlueRay)
 Uninterrupted Video conference (Skype, video chatting)
 Smooth Voice call (conventional, IP phone, Whatsapp,
facebook)
 Easy E-Commerce (online transaction, shopping)
 Quick Browsing
 Others
World population: 760 crores

R. Mahapatra, IIITNR Intelligent Reflecting Surface 2

Evolution of Wireless Networks
Generation wise: Starting to present
Task Beginning Present Status
Distribut Femtoce
Technology 2G Digital Circuit 5G Digital, packet ed ll
Antenna
switched switched s

Throughput 8 kbps 10 Gbps Microcel

l
Client
Delay 30 ms less 1 ms Relay

Wired Network
BER 10-3 10-8

Backhaul
Picocell
WLAN
BW 200 KHz 100 MHz
Number of antenna Single Many Mobile Picocell
PAN
Sharing Individual Cooperation
Frequency Fixed Flexible Femtoce Relay
ll
Subscriber Few Many Wireless Access
Microcel Wireless backhau
l
Energy Not energy efficient energy efficient Wired backhaul

R. Mahapatra, IIITNR Intelligent Reflecting Surface 3

Requirements
To embrace the challenges
 High-capacity energy-efficient broadband communication
Improved spectrum efficiency/usage
Improved energy efficiency/usage
Proper cooperation and coordination among network entities
 Enhanced network deployment
Support of diverse user equipment, applications, and services
Seemless inter-working among multiple spectrums
Efficient support of hop spots and in-door environments
Traffic offloading from licensed spectrum to unlicensed spectrum
High capacity wired/wireless backhaul
Energy-efficient network deployment
Capability in self-healing and self-optimization

R. Mahapatra, IIITNR Intelligent Reflecting Surface 4

5G Communication
Objectives
5G Objectives 5G architecture themes
 It is expected that the connected devices will  Flexibility & Scalability
reach 20 billion, with 5 billion smartphone users  5G New Radio
in 2023  Fiber like performance
 Wide used for internet of things  5G is multi-RAT
 Incorporation of small cell  Network function virtualization
 Dense/Heterogeneous network  Software based: core and RAN
 Various services  Use of Cloud
 Real-time video  Network Slicing
 Data with high mobility  Self-contained, independent network partition
 Video conferences including all segments: core, radio, transport, edge
 Type of traffic  Multi-domain, multi-tenant
 Moving towards high volume service  Programmable network
 Virtual reality/Augmented reality,  Flexible orchestration of network resource and
infrastructure: RAN, core, transport
 others

R. Mahapatra, IIITNR Intelligent Reflecting Surface 5

5G Communication
Expectations  ultra-reliable low latency
communication (uRLLC),
 Capacity: 1000x (2020 vs 2010)
 massive machine type
 Data rate: communication (mMTC)
50 Gbps for low mobility users (50  enhanced mobile broadband (eMBB)
x 4G)
5 Gbps for high mobility user Bit rate

1 Gbps anywhere bits/s

109

 Latency eMBB 107

105

< 1 ms 103

1/10 of 4G
 Low energy consumption 1

10

1000x energy efficiency 102

103

UE battery life
days
R. Mahapatra, IIITNR Intelligent Reflecting Surface 6
Spectrum for 5G
 The 3GPP’s 5G New Radio (NR) specification includes  More spectrum will be needed to maintain 5G quality
traditional mobile bands as well as newer, wider bands of service and meet growing demand in the longer
designed for 5G. term (e.g. 3.3-4.2 GHz, 4.8 GHz and 6 GHz).
 It supports channel bandwidths ranging from 5 MHz to 100  Most focus has been on the 3.5 GHz range (i.e. 3.3-3.8
MHz for bands below 6 GHz, and channel sizes from 50 MHz GHz) to support initial 5G launches, followed by
to 400 MHz in bands above 24 GHz.
mmWave awards in the 26 GHz and 28 GHz bands.
 At least 80-100 MHz of contiguous spectrum per
operator in prime 5G mid-bands (e.g. 3.5 GHz) and 800  Europe has prioritized the 700 MHz band for wide area
MHz per operator in initial mmW bands (e.g. 26/28 5G and a growing number of countries globally are
GHz) for 5G launches supporting the 600 MHz band6 (including the US which
already uses it for 5G)
 Thee assigned 5G spectrum in three broad ranges:
 All 5G bands above 3 GHz – including the vital 3.5 GHz
 high bands (e.g. mmWave) which support the fastest 5G
speeds; range and mmWave bands – will adopt TDD.
 mid bands (e.g. 1-10 GHz) which offer a good mixture of  The Department of Telecommunications (DoT) has
coverage and capacity; and allotted 5G trial spectrum in the 700 Mhz, 3.5 Ghz and
• 3.3-3.8 GHz, e 1500 MHz, 1800 MHz, 2.1 GHz, 2.3 GHz and 2.6 26 Ghz bands, paving the way for Reliance Jio, Bharti
GHz Airtel and Vodafone Idea (Vi) to partner with non-
 low bands (e.g. below 1 GHz) which help provide strong wide Chinese network vendors and develop India-relevant
area and in-building coverage.
use cases on the next-gen fast wireless broadband
technology.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 7

Network Infrastructure
5G & Beyond
PE

IP/GMPLS OLT
Access Holographic Telepresence
PE PE ONU

OLT OLT
Data Center
OLT OTN
SWT SDN Applications
Networked
eNB Multidomain BW and
MIMO Network slicing
OLT PON
OLT
eNB
eNB DWDM Ring Augmented Reality/
OLT
Virtua Reality
OLT
eNB
PON

RRH RRH ROADM

OLT
5G-NR
OLT

CRAN
CoMP OLT

Picocell Relay 5G-NR

RRH
Massive
eNB 5G-NR
MIMO
Femtocell
Macrocell Microcel
mmWave
l

HetNet

R. Mahapatra, IIITNR Intelligent Reflecting Surface 8

5G New Radio  Flexible 5G NR slot structures supports
 TDD Self TDD Self TDD Self –Contained
• Opportunity for UL/DL scheduling, Opportunity for
 5G NR (New Radio) is a new radio access UL/DL scheduling, slot data
technology (RAT) developed by 3GPP for the  Data-centric
5G (fifth generation) mobile network. • More relaxed TDD timing configurations + FDD
 It was designed to be the global standard for operation
the air interface of 5G networks.  Mimi Slot:
• Optimized for shorter data transmissions, e.g. URLLC

Qualcomm presentation on 5GNR

R. Mahapatra, IIITNR Intelligent Reflecting Surface 9
Key Enabling Techniques
Physical Layers Adv. channel coding

Access perspective
TDMA, OFDMA, NOMA
BS perspective Scalable OFDM

Different sizes, different type, different RAT

Transmission perspective Microcell
Femtocell

Multi-user, multi-service, multi-provider

Antenna perspective
Multi-antenna, Multiple array, Massive array
Flexible slot-based framework

R. Mahapatra, IIITNR Intelligent Reflecting Surface 10

Wireless Network Technologies
Recent developments
 LTE and LTE Advanced
 Massive MIMO for macro and small cell
 Millimeter wave technology: 60 GHz
 M2M, V2V: device centric architecture
 Dense small cell deployment
Small cell cooperation (interference reduction)
Hyper-cellular architecture: separation of control & user plan
 Network virtualization
Core network: SDN
Access network: C-RAN
Backhaul network: flexible backhauling, combination of wired and wireless network and their
interoperability

R. Mahapatra, IIITNR Intelligent Reflecting Surface 11

Multiuser MIMO
From SU-MIMO
 MU-MIMO enables multiple
independent radio terminals to access a
system enhancing the communication
capabilities of each individual terminal
 MU-MIMO exploits the maximum
system capacity by scheduling multiple
users to be able to simultaneously
access the same channel
using the spatial degrees of freedom
offered by MIMO.
 MU-MIMO provides a methodology
whereby spatial sharing of channels can
be achieved with additional hardware
R. Mahapatra, IIITNR Intelligent Reflecting Surface 12
Massive MIMO
Introduction
 Massive MIMO is a multi-user MIMO technology where
each base station (BS) is equipped with an array of M
active antenna elements and utilizes these to  Massive MIMO uses:
communicate with K single-user terminals over the  Very large arrays with an order of magnitude
same frequency band higher number of sensors.
 M-MIMO in cellular systems brings improvements on  Deployment of devices (access points, mobile
 Increased data rate phones and tables) with a large number of
 Enhanced reliability antenna elements.
 Improved energy efficiency  Huge multiplexing gains allowing an order of
 Reduced interference and latency magnitude higher data rates.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 13

Massive MIMO
Characteristic  Protocols:
Scheduling and medium-access protocols
 Antennas: for numerous heterogeneous users.
 Reduction of RF chains and costs,
 Signal processing:
 Compact antennas and mitigation of
coupling effects. Transmit and receive processing,
 Electronic components: Scalability and hardware implementation,
 Low-cost components such as power Integration between signal processing and
amplifiers and RF components. RF devices to deal with impairments.
 Flexibility for different air interfaces and
replacement of coaxial cables.  Challenges for Massive MIMO
 Network architectures: Scaling data rates and interfaces to support
 Heterogeneous networks, small cells and large 𝑁𝐵𝑆
network MIMO, Low-latency for channel reciprocity (fast
 Cloud radio access networks to help switch from uplink to downlink)
different devices. Synchronizing radios across 𝑁𝐵𝑆 base
station antennas

R. Mahapatra, IIITNR Intelligent Reflecting Surface 14

Massive MIMO
Transmission
 Massive MIMO promises spectral efficiency
gains by spatial multiplexing of multiple
devices
User clustering, grouping, and beamforming
The first step of user clustering divides all
users connected to a BS into groups with
similar channel statistics.
In the second step, the user selection is done
for each group independently, with multi-
user precoding scheme
In the third step, the precoding weights for
simultaneous multi-user transmission are
obtained based on the effective channel

R. Mahapatra, IIITNR Intelligent Reflecting Surface 15

Massive MIMO
Architecture

R. Mahapatra, IIITNR Intelligent Reflecting Surface 16

Massive MIMO
If 𝑁𝑡 ≫ 𝑁𝑟 and 𝑁𝑡 → ∞
Channel Capacity
 Point to point MIMO

If 𝑁𝑟 ≫ 𝑁𝑡 and 𝑁𝑟 → ∞
 the scalar 𝜌 is the transmit power
 If we assume independent and identically distributed
(i.i.d.) Gaussian transmit signals and that perfect CSI is
available at the receiver, the instantaneous achievable
rate
MU-MIMO systems can obtain the promising
multiplexing gain of massive point-to-point
 When the propagation coefficients in the channel matrix MIMO systems while eliminating problems due
are normalized as 𝑇𝑟(𝑯𝑯𝐻 ) ≈ 𝑁𝑡 𝑁𝑟 , upper and lower to unfavorable propagation environments.
bounds on the capacity are derived with the help of
Jensen’s inequality

R. Mahapatra, IIITNR Intelligent Reflecting Surface 17

How Large is “Massive”?

R. Mahapatra, IIITNR Intelligent Reflecting Surface 18

R. Mahapatra, IIITNR Intelligent Reflecting Surface 19
Energy Efficient Network
Green Communication
 Global information communications technology (ICT)
ecosystem consumes around 1100 to 1800 Terawatt-hours
of electricity annually
 Worldwide telecommunication networks (which include
wire-line, wireless and core networks) are responsible for
more than one third of this energy consumption due to Estimation for coverage of different
technologies from the year 2010 to 2018
increased coverage and user subscriptions
 To deliver an hour of video on weekly basis to a smartphone
for one year will consume more power than the combined
annual power consumption of two household refrigerators
 BSs consume almost 60% of the total power consumption
of a typical mobile network
 Wireless networks can be made Green by reducing their
overall energy consumption by almost 90% in the year
2020 with significant economic and environmental impact
A. Mughees [Link], Towards Energy Efficient 5G Networks Using Machine Learning: Taxonomy, Research
Challenges, and Future Research Directions, IEEE Access, vol8, pp. 187498-187522, Oct 2020 Total energy consumption of 2018 and 2025
R. Mahapatra, IIITNR Intelligent Reflecting Surface 20
Energy Consumption
Estimation

estimation in communication
 Power consumed by system in 2013 and 2025

Usage
Location
Different service
Applications
CAPEX & OPEX
Sharing of fixed energy consumption Etc. Sharing of fixed energy consumption
between service categories.
between application
R. Mahapatra, IIITNR Intelligent Reflecting Surface 21
Energy Efficiency
Benefit-Cost Analysis
Cost Benefit
Energy Consumption Wireless Network Data Throughput
(Watt=Joule/sec) (Bits/sec)
(Energy cost (Joule/rupess)

Benefit-Cost Ratio Example: 4G Base station

Data throughput=28 Mbps
𝑏𝑖𝑡𝑠 𝐷𝑎𝑡𝑎 𝑇ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡 (𝑏𝑖𝑡𝑠/ sec)
𝐸𝑛𝑒𝑟𝑔𝑦 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = Energy consumption=1.35 KW
𝑗𝑜𝑢𝑙𝑒 𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 (𝑗𝑜𝑢𝑙𝑒/𝑠𝑒𝑐) EE=20 kbits/Joule
with Coverage
𝑏𝑖𝑡𝑠 𝐷𝑎𝑡𝑎 𝑇ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡 (𝑏𝑖𝑡𝑠 / sec/ 𝐾𝑚2 )
Area 𝐸𝑛𝑒𝑟𝑔𝑦 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 =
𝑗𝑜𝑢𝑙𝑒 𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 (𝑗𝑜𝑢𝑙𝑒 / sec / 𝑘𝑚2 )
𝑏𝑖𝑡𝑠 𝐷𝑎𝑡𝑎 𝑇ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡 (𝑏𝑖𝑡𝑠/ sec / 𝐻𝑧)
with 𝐸𝑛𝑒𝑟𝑔𝑦 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 / 𝐻𝑧 =
𝑗𝑜𝑢𝑙𝑒 𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 (𝑗𝑜𝑢𝑙𝑒 /𝑠𝑒𝑐)
frequency
R. Mahapatra, IIITNR Intelligent Reflecting Surface 22
 Energy Efficiency Energy efficient network
Understanding
Energy Consumption

Data Data Data

Energy Energy Energy

• Most likely solution for

Traffic Load
Require tradeoffs future network
• Consumes more energy,
but EE increases
Bandwidth Transmit power
Energy Efficiency

Pathloss
𝑏𝑖𝑡𝑠 𝑃𝑡𝑥 𝛽
𝐷𝑎𝑡𝑎 𝑇ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡 ( 𝑠𝑒𝑐 ) 𝐵 log 2 (1 + 𝐵𝑁 )
𝑜
= PSD of noise
𝑗𝑜𝑢𝑙𝑒 𝑃𝑡𝑥
𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 ( 𝑠𝑒𝑐 ) 𝜇 + 𝑃𝑐𝑖𝑟𝑐𝑢𝑖𝑡
Traffic Load Amplifier efficiency Circuit power

R. Mahapatra, IIITNR Intelligent Reflecting Surface 23

Green Metric
Measurement
 Component level  Energy Consumption Rating (ECR)
Antenna, PA, baseband, cooling etc  Energy Efficiency Rate (EER)
 Access Per Cycle (APC)
 Equipment level
 Power Usage Efficiency (PUE)
Device, processor etc
 Data Centre Efficiency (DCE)
 System/Network level  Telecommunication Equipment
Coverage, capacity, deployment etc Energy Efficiency Rating (TEEER)
 Greenness metrics  Telecommunication Energy Efficiency
Ratio (TEER)
Green Performance Indicator, Carbon
Emission Calculator, emission factors,  Energy Proportionality Index (EPI)
gCO2e/kWh  Data Center Infrastructure Efficiency

R. Mahapatra, IIITNR Intelligent Reflecting Surface 24

Cyber Physical Systems
Description
 A cyber-physical system (CPS) is a
system of collaborating
computational elements controlling Control
physical entities.
 CPS are physical and engineered
systems whose operations are
monitored, coordinated, controlled Information
and integrated by a computing and
communication core.
They allow us to add capabilities to Systems
physical systems by merging
computing and communication with
physical processes

R. Mahapatra, IIITNR Intelligent Reflecting Surface 25

Industry 4.0
Building Blocks
Simulation

Big data Autonomous

analytics Robots

Horizontal and
Augmented
reality Industry 4.0 vertical system
integration

Industrial
Cyber Security Internet of
Things

Additive Mfg

R. Mahapatra, IIITNR Intelligent Reflecting Surface 26

Emerging Services
In recent years
 Virtual reality/Augmented reality  Terahertz level Data
(VR/AR)  Extreme focus on Short-range
 Teleportation (Holographic way, 5- Communication
sense)
 AI is Inherent within the Network
 Integrated Terrestrial and Space
(Near to space)  Extremely High Radio Density
 High-Precision Service  Very High Network Heterogeneity
 Introduction of Industrial revolution (Extreme HetNets)
4.0 (Cyber physical system)  Transformed Radio Topology
Control, communication & computing (Mesh, Hops, and Peer-to-Peer
 Autonomous V2X

R. Mahapatra, IIITNR Intelligent Reflecting Surface 27

Intelligent Reflecting Surface (IRS)
Why 𝑯𝑃
 Have We Reached Shannon’s Capacity Limit? 𝐶 = log 2 (1 + 2 )
𝜎
 Yes, also No (as wireless channel 𝐻 is still random and uncontrolled)
 Can we make 𝐻 arbitrarily large, say from 𝐻 << 1 to 𝐻 → 1?
 Can we make 𝐻 less random, e.g., from Rayleigh fading to Rician fading?
 Existing wireless technologies (beamforming, power control, adaptive modulation,
etc.) only adapt to 𝐻, but have no control over it
 How to break this ultimate barrier to achieving ultra-high capacity and ultra-high
reliability in future wireless communications (e.g., 6G)?
 Promising new paradigm: Smart and Reconfigurable Wireless Environment
 Key enabling technology: Intelligent Reflecting Surface (IRS)
Other nomenclature: reconfigurable intelligent surface (RIS), software controlled
metasurface, passive intelligent mirror, smart reflect array, ….
WTC Online Seminar, 2020 by Dr. Rui Zhang

R. Mahapatra, IIITNR Intelligent Reflecting Surface 28

Intelligent Reflecting Surface (IRS) Also known as Reconfigurable
Intelligent Surface (RIS), large
Definition Copper
backplane
intelligent surface (LIS), software-
controlled metasurfaces
Control
circuit board
 Reconfigurable: Properties can be changed
 Tuning impedances, length of delay lines,
Phase-shifters
 Intelligent: Real-time programmable/controllable
 Surface: Two-dimensional array of scatterers

Communicate with other nodes in the

network (IRS, BS, terminal) and adjust
RIS the reflecting elements dynamically
Controller

 A digitally-controlled meta surface with massive low-cost passive reflecting elements

(each able to induce an amplitude/phase change in the incident signal)
 Low energy consumption (without the use of any transmit RF chains), high spectral
efficiency (full-duplex, noiseless reflection)
R. Mahapatra, IIITNR Intelligent Reflecting Surface 29
Intelligent Reflecting Surface
Advantages
 Easy to deploy  Spectral efficiency enhancement
 IRSs are nearly-passive devices, made of  IRSs are capable of reconfiguring the wireless
electromagnetic (EM) material propagation environment by compensating for
 IRSs can be deployed on several structures, the power loss over long distances
including but not limited to building facades,  Due to the intelligent deployment and design of
indoor walls, aerial platforms, roadside RISs, a software-defined wireless environment
billboards, highway polls, vehicle windows, as may be constructed, which, in turn, provides
well as pedestrians’ clothes due to their low potential enhancements of the received signal-
cost. to-interference-plus noise ratio (SINR).
 Environment friendly  Compatibility:
 IRSs are capable of shaping the incoming signal  IRSs support full-duplex (FD) and fullband
by controlling the phase shift of each reflecting transmission due to the fact that they only
element instead of employing a power amplifier reflect the EM waves
 Deploying IRSs is more energy-efficient and
environment friendly than conventional AF and
DF systems.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 30

Intelligent Reflecting Surface
Categories
 Waveguide IRS: The elements in the  Refracting IRS: It used an equivalent
metasurface are modeled as uncoupled impedance matrix model so that the
magnetic dipoles. tangential field components at the two
 The magnitude of each dipole element is sides of the metasurface are
proportional to the product of the appropriately optimized
reference wave and each element’s  Reflecting IRS: The elements in the
polarizability. By tuning the polarizability, metasurface contain varactor diodes with
the metasurface antenna can perform a tunable biasing voltage. By
beamforming. Each element on the predesigning several digitized biasing
metasurface serves as a micro-antenna. voltage levels, each element can apply
 Compared to conventional antenna discrete phase shifts and achieve
arrays, the compact waveguide beamforming for the reflected wave.
metasurface occupies less space and can
transmit towards wider angles.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 31

 EM reflection
using varactor-tuned resonators
 The EM characteristics of the IRS, such as phase discontinuity, can be reconfigured by
tuning the surface impedance.
 The choices of materials of the IRS include semiconductors and graphene.
 When the PIN diode is turned off, the incoming energy penetrates the surface and is
mostly absorbed. However, when the PIN diode is on, most of the incoming energy is
reflected
 Varactor-tuned resonators are also used for controlling the
signal's propagation
 When the bias voltage is applied to the varactor, a tunable phase
shift is attained.
 The phase shifts of the incoming signal can be controlled at
various locations of the meta-surface. Dynamic meta-surfaces
make up a tile that consists of a gateway, to which the controller
network acquires a slave/master relationship.
 The controller network records its running state and receives
instructions to change the current condition of the switching
elements through the gateway

R. Mahapatra, IIITNR Intelligent Reflecting Surface 32

 IRS Element
RIS

Pei X, Yin H, Tan L, Cao L, Li Z, Wang K, Zhang K, Björnson E. RIS-aided wireless communications: Prototyping, adaptive beamforming, and
indoor/outdoor field trials. IEEE Transactions on Communications. 2021 Sep 29;69(12):8627-40.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 33

Path Loss Model
Multipath Fading
 Baseband equivalent channel model (narrow-band)
 Assume isotropic reflection, and no mutual coupling
among reflecting elements

Two ray model

 Product-distance path loss model

R. Mahapatra, IIITNR Intelligent Reflecting Surface 34

IRS Path Loss Model
Product Distance or Sum Distance?
 Product-distance path loss model

 Sum-distance path loss model

 Applies to free-space propagation and infinitely

large perfect electric conductor (PEC) only
 Not applicable to IRS with finite-size elements
R. Mahapatra, IIITNR Intelligent Reflecting Surface 35
Signal Propagation
RIS

 When the signal reaches element 𝑛, it will be filtered

 Phase shift as external control variables 𝜃1 … 𝜃𝑁 . inside 𝑛 and then reradiated.
 For each of the N scattering elements of the IRS, the  Since the element is much smaller than the wavelength,
transmitted signal 𝑥𝑝𝑏 (𝑡) will propagate to it across an it can be modeled as a passive electric circuit. The
LTI channel represented by an arbitrary impulse passiveness implies there is no added noise within the
response 𝑎𝑛,𝑝𝑏 (𝑡) for element 𝑛 = 1 … 𝑁. circuit
 If its frequency response has a constant amplitude and  The amplitude loss depends on the frequency and
linear phase across the passband used by the signal (a capacitance. The losses are largest when tuning the RIS
narrowband channel.) to achieve zero phase response due to resonance in the
circuit.
R. Mahapatra, IIITNR Intelligent Reflecting Surface 36
Signal Propagation
 The impulse response of RIS, 𝜗𝑛,𝑝𝑏,𝜃𝑛 (𝑡) is
reconfigurable in the sense that it is
determined by an external stimulus
represented by the variable in 𝜃𝑛
 The signal that is reradiated from element n
propagates to the receiver across an LTI
channel with an arbitrary impulse response
𝑏𝑛,𝑝𝑏 (𝑡)
 Since the transmitted signal propagates via
element n across a cascade of three LTI filters,
the joint impulse response is the convolution
of their impulse responses:
 (𝑏𝑛,𝑝𝑏 ∗ 𝜗𝑛,𝑝𝑏,𝜃𝑛 ∗ 𝑎𝑛,𝑝𝑏 )(𝑡).
 This happens for all the N elements; thus, we
obtain the input–output relation
Björnson E, Wymeersch H, Matthiesen B, Popovski P, Sanguinetti L, de Carvalho E. Reconfigurable intelligent impulse response of the end-to-end system
surfaces: A signal processing perspective with wireless applications. IEEE Signal Processing Magazine. 2022
Feb 24;39(2):135-58.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 37

R. Mahapatra, IIITNR Intelligent Reflecting Surface 38
 IRS
in Multipath channels
 Suppose the channel from the transmitter to the 𝑛-th IRS element consists of
𝐿𝑎 propagation paths; then, the impulse response is modeled as
propagation loss delay of the 𝑙-th path

 Due to the product operation, each path is

the discrete-time impulse response very weak, but the large number of paths can
potentially lead to a good SNR.
 Each path is also associated with a phase shift
containing the accumulated delays
propagation loss of  The sinc function determines how the signal
each 𝑁𝐿𝑎 𝐿𝑏 paths energy carried by the path is divided between
from Tx to Rx the M taps of the FIR filter
R. Mahapatra, IIITNR Intelligent Reflecting Surface 39
Usages of IRS
 IRS is able to achieve several appealing
functions in wireless channel
reconfiguration, such as
1. creating virtual line-of-sight (LoS)
× link to bypass obstacles between
transceivers via smart reflection,
2) Improving channel Rank condition 2. adding extra signal paths toward
1) Coverage Extension desired direction to improve the
channel rank condition,
3. refining the channel
statistics/distribution by e.g.
transforming Rayleigh/fast fading to
Rician/slow fading for ultra-high
reliability,
4. suppressing/nulling co-channel/
inter-cell interference and so on.
3) Refining channel statistics 4) Interference suppression

R. Mahapatra, IIITNR Intelligent Reflecting Surface 40

IRS application/deployment scenarios
Future wireless networks

R. Mahapatra, IIITNR Intelligent Reflecting Surface 41

Advantages of IRS

Björnson E, Özdogan Ö, Larsson EG. Intelligent reflecting surface versus decode-and-forward: How large surfaces
are needed to beat relaying?. IEEE Wireless Communications Letters. 2019 Oct 31;9(2):244-8.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 42

 Why IRS
End-to End SNR

Improve propagation conditions

Large surfaces are needed More than just SNR gain!
to beat an elementary DF relay Ö. Özdogan, E. Björnson, E. G. Larsson, “Using Intelligent Reflecting
Surfaces For Rank Improvement in MIMO Communications”

R. Mahapatra, IIITNR Intelligent Reflecting Surface 43

 Channel Model
 The transmitter is not completely enclosed by
obstacles and a IRS is deployed to bypass these
obstacles and recover the service between Tx
and Rxs.
 Considering channel fading caused by the
× scatters in the environment,
the direct Tx-Rx links are pure NLoS links,
while the Tx-IRS-Rx links constitute both specular
(LoS) components and scattering (NLoS) components.
Xu J, Liu Y. A novel physics-based channel model for reconfigurable intelligent surface-assisted multi-user communication systems. IEEE Transactions on Wireless Communications. 2021 Aug 12.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 44

Channel Model
 Consider the wireless channel between a transmitter
and K receivers through an IRS
 The transmitted signal 𝑠 𝑡 = 𝑅𝑒{exp(𝑗2𝜋𝑓𝑐 𝑡)}
 The passband signal received by each receiver
𝑟(𝑡) = 𝑇𝑐 (𝑡)cos(𝜔𝑐 𝑡) − 𝑇𝑠 (𝑡)sin(𝜔𝑐 𝑡)
 the complex envelope of r(t),
෨
𝑅(𝑡) = 𝑇𝑐 (𝑡) + 𝑗𝑇𝑠 (𝑡)
×  For the IRS-assisted channel, assuming vertical
polarization for the wireless signal, the complex
received envelope
𝑅෨ 𝑡 = 𝑅෨𝑠𝐼𝑅𝑆 𝑡 + 𝑅෨𝑟𝐼𝑅𝑆 𝑡 + 𝑅෨ 𝐷 (𝑡)

LOS Tx-IRS-Rx link NLOS Tx-IRS-Rx link direct link

R. Mahapatra, IIITNR Intelligent Reflecting Surface 45

 Channel Model NLOS Tx-RIS-Rx link
𝑁1
LOS Tx-RIS-Rx link
𝑅෨𝑟𝐼𝑅𝑆 𝑡 = ෍ 𝑎𝑟 𝑒 𝑗(𝜔𝑟+𝜃𝑟 +𝜓𝑟)
𝟏 𝒆𝒋𝝎𝒊 𝒕+𝟐𝒋𝝅𝒅/𝝀
෩ 𝑰𝑹𝑺
𝑹 𝒔 𝒕, 𝜽𝒐𝒖𝒕 = ඵ ෩ 𝑭 𝜽𝒐𝒖𝒕
𝑼 𝒅∑ 𝑟=1
𝒋𝝀𝒄 ∑ 𝑰𝑹𝑺 𝒅
• 𝑁1 is the number of scatters considered in the NLoS
• 𝑈෩ is the aperture distributions of the specular component of the Tx-IRS-Rx link,
link at the IRS, • 𝑎𝑟 is the amplitude of the r -th multipath signal.
2𝜋𝑣
• 𝐹(𝜃𝑜𝑢𝑡 , 𝜑𝑜𝑢𝑡 ) is the leaning factor, • 𝜔𝑟 = 𝜆 cos(𝛾 − 𝛼𝑟 ) , where αr is the angle of arrival w.r.t x-
• 𝜔𝑖 = 2𝜋𝑣𝜆𝑐 𝑐𝑜𝑠(𝛾 − 𝛼𝑖 ) , 𝑣 is the velocity of the 𝐿𝑟
axis of this r -th multi-path signal, and 𝜃𝑟 = 2𝜋(𝑓𝑐 + 𝑓𝐷,𝑟 𝑐),
receiver, and γ denotes the angle of this 𝑣
𝑐
movement, 𝛼𝑖 is the angle of arrival w.r.t x-axis of where 𝑓𝐷,𝑟 = 𝜆 𝑐𝑜𝑠(𝛼𝑟 ) is the Doppler shift frequency of the
𝑐
signal from the i -th element, r-th multipath signal,
• 𝑑 is the distance between the IRS and the • 𝐿𝑟 is the total distance for the r-th multi-path signal to travel
receiver, Σ denotes the aperture area of the RIS. and c is the speed of light, and 𝜑𝑟 is the additional phase shift
direct link of the RIS acting on the r -th multipath signal.

𝑁2 is the number of considered multipath in the direct

link, 𝑏𝑛 is the amplitude of the n -th multipath signal,

R. Mahapatra, IIITNR Intelligent Reflecting Surface 46

Phase Shift in IRS
 IRS does not acquire perfect channel state information and its phase shifts do not change with the
instantaneous NLoS components of the Tx-IRS-Rx link.
 Consider IRS performing beam-steering for the specular component of the incident signal, the
configuration of the IRS is given by the co-phase condition
 In the continuous phase shift scenario, the It is also desirable to obtain the additional phase shift at
phase shifts 𝜙𝑚 can take on arbitrary values directions other than the targeted one
within [0, 2π).
 For convenience, we denote Φ𝑚 = 𝜙𝑚 − 𝜖𝑚
𝑝
 𝜖 = 2𝜋 𝜆𝑥 (sin 𝜃𝑜𝑢𝑡 − sin 𝜃𝑖𝑛 ).
𝑐
 As the angel of the receiver (𝜃𝑜𝑢𝑡 ) varies, the This means, the specular component 𝑅෨𝑠𝐼𝑅𝑆 𝑡 is deterministic.
additional phase shift associated with m-th On the contrary, 𝑅෨𝑟𝐼𝑅𝑆 𝑡 and 𝑅෨ 𝐷 (𝑡) are random variables.
element at 𝜃𝑜𝑢𝑡 , Φ𝑚 (𝜃𝑜𝑢𝑡 ) also changes.
 Implementing the co-phase condition for As a result, when 𝑁1 , 𝑁2 ≫ 1, meaning the numbers of the
phase shifts, a continuous phase shift RIS is multipath components are sufficiently large, 𝑅෨𝑟𝐼𝑅𝑆 𝑡 and 𝑅෨ 𝐷 (𝑡)
able to perfectly align all additional phase exhibits complex Gaussian distributions with zero means. Only
shifts at 𝜃𝑜𝑢𝑡 = 𝜃𝑡𝑎𝑟𝑔𝑒𝑡 . the deterministic 𝑅෨𝑠𝐼𝑅𝑆 𝑡 provides a non-zero mean.
R. Mahapatra, IIITNR Intelligent Reflecting Surface 47
IRS-Aided SISO System
Passive Beamforming
 consider the single-user single-input-single-output (SISO) case, i.e., 𝐾 = 1 and 𝑀𝑡 =
𝑀𝑟 = 1, and the narrow-band system with flat-fading channels.
 The baseband equivalent channels from the AP to the IRS, from the IRS to the user,
and from the AP to the user are denoted by 𝒈 ∈ 𝐶 𝑁×1 , 𝒉𝐻 ∈ 𝐶 𝑁×1 and h∗ ∈ 𝐶 𝑁×1
𝑟 d
the signal received at the user to maximize the achievable rate r (or SNR equivalently) by
optimizing the passive reflect beamforming at the IRS.

the user receive SNR

Wu Q, Zhang S, Zheng B, You C, Zhang R. Intelligent reflecting surface-aided wireless communications: A tutorial. IEEE Transactions on Communications. 2021 Jan 18;69(5):3313-51.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 48

IRS-Aided SISO System
 The optimal phase-shift solutions

∗
where 𝜙𝑛 , 𝜓𝑛 , 𝜁 are the phases of ℎ𝑟,𝑛 , 𝑔𝑛 , ℎ𝑑

 This is expected since the optimal phase shifts should

align all the signals reflected by the IRS regardless of
their strength with the signal coming directly from the
AP to achieve coherent combining and thus maximize
the received signal power at the user.
 if the AP-user direct link is negligible compared to the
IRS-reflected link

R. Mahapatra, IIITNR Intelligent Reflecting Surface 49

Receive Power Scaling With N
 The maximum receive SNR grows with the number of IRS
reflecting elements, N, when N becomes asymptotically large.
 the asymptotic receive power

 This result suggests that with sufficiently large N, the user

receive power increases quadratically with N
 This is because the IRS not only achieves the reflect
beamforming gain of O(N) in the IRS-user link, but also captures
an additional aperture gain of O(N) in the AP-IRS link.
 The path loss exponents of the three channel models are set
as 2, 2.4 and 2.8
 The achievable rate of the IRS-aided SISO system increases
about 2 bps/Hz by doubling N when N is sufficiently large, e.g.,
from N = 200 to N = 400,

The path loss model for both AP-IRS and IRS-user links is set as
𝑐0 𝑑0−𝑎 where 𝑑0 is the corresponding link distance in meter (m)
and 𝑎 denotes the path loss exponent. Other parameters are
R. Mahapatra,
= −30 𝑑𝐵, 𝑃𝑡 = 50 𝑚𝑊 and 𝜎 2 = Intelligent
set as 𝑐0 IIITNR −90 𝑑𝐵𝑚.Reflecting Surface 50
IRS-Aided MISO System:
Joint Active and Passive Beamforming
 we consider the single-user system but with multiple antennas at the AP, i.e., 𝑀𝑡 > 1, where the
downlink (MISO)/ uplink (SIMO)
 𝑮 and 𝒉𝐻𝑑 denote the channel matrix and channel vector from the AP to IRS and from the AP to user,
respectively.
w denotes the transmit beamforming
vector at the AP and 𝑃𝑡 is the maximum
transmit power.

 If the channel of the AP-user link, 𝒉𝐻

𝑑 , is much stronger than that of the AP-IRS link, G,
 it is preferable for the AP to beam toward the user directly, while in the opposite case, especially when the AP-
user link is severely blocked, the AP should steer its beamforming direction toward the IRS to leverage its
reflected signal to serve the user.
 In general, the transmit beamforming at the AP needs to be jointly designed with the phase shifts at the
IRS based on all the AP-IRS, IRS-user, and AP-user channels in order to maximize their cooperative gain.
 one can observe that if we fix the transmit beamforming vector w, the above problem is reduced to its
SISO, which thus motivates the alternating optimization (AO) approach to solve the above problem sub-
optimally
R. Mahapatra, IIITNR Intelligent Reflecting Surface 51
IRS-Aided MISO System
 with fixed 𝑤, we obtain the optimal phase shifts; while
with fixed 𝜃, the optimal transmit beamforming solution
is given by the maximum-ratio transmission (MRT),

 the achievable rate of the IRS-aided MISO system versus

the AP-user horizontal distance, d, for the same as earlier
setup with 𝑀𝑡 = 4 and N = 40. Rayleigh fading channels are
assumed for both the AP-user and IRS-user links, while the This demonstrates that the joint active and
free-space LoS channel is adopted for the AP-IRS link. passive beamforming design is essential to
 if the transmit beamforming is not designed properly strike an optimal balance between the
based on all the channels, the achievable rate by using the transmission toward the user directly and that
IRS may be even worse than the conventional MRT via the IRS reflection, to maximize the received
without the IRS signal power at the user.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 52

IRS-Aided MIMO
 The overall IRS-aided MIMO channel

 By considering continuous phase shifts

and maximum reflection amplitudes at all
IRS elements, the capacity optimization
problem is formulated as
the achievable rate of the AO-based algorithm is
significantly larger than the MIMO channel capacity
without IRS, as well as the achievable rate of a
benchmark scheme with the phase shift of each IRS
reflecting element independently and uniformly
𝑸 ∈ 𝐶𝑀𝑡×𝑀𝑡 denote generated in [0, 2) and the corresponding optimal
the transmit covariance matrix transmit covariance matrix.

R. Mahapatra, IIITNR Intelligent Reflecting Surface 53

RIS-Assisted
Wireless Network
 for enhancing spectrum sharing among  used as the signal receiver to enhance
multiple transceivers. system capacity
 use the IRS for steering the signal beams  multiple single-antenna devices
from different transmitters to enhance distributed in a three-dimensional (3D)
the useful signals and cancel the space transmit to a vertical plane
interference towards their respective metasurface (e.g., hung on the wall)
receivers.

Gong S, Lu X, Hoang DT, Niyato D, Shu L, Kim DI, Liang YC. Toward smart wireless communications via intelligent reflecting
surfaces: A contemporary survey. IEEE Communications Surveys & Tutorials. 2020 Jun 22;22(4):2283-314.
R. Mahapatra, IIITNR Intelligent Reflecting Surface 54
RIS-Assisted
Wireless Network

IRS-assisted secure communication

system with a friendly jammer

R. Mahapatra, IIITNR Intelligent Reflecting Surface 55