IOT TRAINING SESSION #3 | ISEP | © 2021-24 TG

INTERNET OF THINGS
 🥨 2

PREVIOUSLY

the last recap :-)

IOT: CONTEXT 🥨 3

IOT VALUE CHAIN

Connected IoT IoT Internet | VPN Business

Device Network Platform Connection Application
IOT: STANDARDS 🥨 4

MOBILE STANDARDS: BIG PICTURE

5G

Throughput
eMBB
4G 20 Gb/s
2021

Cellular 3G
2012+

critical
IoT
2004+ 2023+

NB-IoT
Cellular IoT 2G LTE-M

massive
2018
IoT
Energy savings

2025+
1992+

LPWAN / ISM bands

LoRa

SigFox
WAN
2016

Evolution
IOT WIRELESS NETWORKS 🥨 5

not to scale
⚠
Throughput

relative positions only

Cellular

3 Cellular IoT Satellite

LPWAN
Range
SUPPLY CHAIN 🥨 6

ASSET TRACKING: TECHNOLOGIES

cost savings
LPWAN

indoor outdoor
accuracy accuracy
UWB, WiFi GPS
RFID,
BLE

energy savings
LORAWAN 🥨 7

LORAWAN ARCHITECTURE

Device VPN Network VPN / Application

Server Internet Server

G/Ws: Gateways
LORAWAN 🥨 8

LORAWAN SF: SPREADING FACTOR

Source: ResearchGate.net
LORAWAN 🥨 9

ADR: ADAPTATIVE DATA RATE

Auto adaptative trade-off between QoS and Autonomy

Energy Settings
=
Repetitions Radio conditions
Spreading Factor
LORAWAN 🥨 10

GEOLOCATION
G/W

Network based
macro geolocation G/W
Device
< 10 km accuracy

RSSI vs. TDOA

G/W
SIGFOX VS. LORAWAN 🥨 11

SIGFOX LORAWAN

UNB: Ultra Narrow Band Spread Spectrum > motion, geoloc

12 Bytes / message at 100 b/s More than 50 Bytes / message
Up to 4 ACKs / day Bidirectional
No end-to-end encryption Secure by design (AES128)
Public & proprietary network Private & public networks / Alliance
LTE-M 🥨 12

LTE-M ESSENTIALS
LTE evolution for M2M / IoT
Cellular IoT: Cat-M1 at 800 MHz
Lower energy consumption
Mobility: handover
Roaming between operators
Standardized and 5G ready
NB-IOT 🥨 13

NB-IOT ESSENTIALS
Deep-indoor coverage:
key NB-IoT differentiator
Advanced energy savings
Up to 20+dB for underground
and enclosed spaces
DSSS modulation technology
vs. LTE-M spread technology
5G 🥨 14

5G ESSENTIALS

eMBB - enhanced Mobile Broadband

uRLLC - ultra Reliable Low Latency Communications
critical IoT
mMTC - massive Machine Type Communications
massive IoT
RedCap - aka NR Lite - Reduced Capacity

AGV
Mngmt
IOT PROTOCOLS 🥨 15

MQTT Message Queue Telemetry Transport: publish-subscribe protocol that facilitates

one-to-many communication mediated by brokers.
Clients can publish messages to a broker and/or subscribe to a broker to receive
certain messages. Messages are organized by topics, which essentially are
“labels” that act as a system for dispatching messages to subscribers.
IOT PROTOCOLS 🥨 16

COAP Constrained Application Protocol: client-server protocol not yet standardized.

A client node can command another node by sending a CoAP packet.
The CoAP server will interpret it, extract the payload, and decide
what to do depending on its logic and acknowledge it or not.
IOT PROTOCOLS 🥨 17

LWM2M
Application layer communication protocol
for IoT device management
an open standard from the
Open Mobile Alliance (OMA)
 🥨 18

AGENDA

Session 3 of 3
IOT: STANDARDS 🥨 19

AGENDA

IoT chips, sensors & actuators

IoT data & platforms

IoT security and AI

Use cases & workshop

 🥨 20

WEARABLE DEVICES

smart & small

 IOT: WEARABLES 🥨 21
SEGMENTS

FisherConnectors
IOT: WEARABLES 🥨 22

SMART-EVERYTHING

FisherConnectors
IOT: WEARABLES 🥨 23

IIOT WEARABLES

FisherConnectors
IOT: WEARABLES 🥨 24

IIOT
DIAG
 🥨 25

EMBEDDED DEVICES

Hardware is hard
EMBEDDED HARDWARE 🥨 26

IOT: ANATOMY OF EMBEDDED H/W

Sensors Antenna
I/O SEC RF
Actuators

IoT
MICRO CONTROLLER
Device
POWER Source
EMBEDDED HARDWARE 🥨 27

CHIPS … MODULES
Chips
SiP: System in Package
SoC: System on Chip
Modules
EMBEDDED HARDWARE 🥨 28

GROVE STARTER-KIT

https://seeeddoc.github.io/Grove-Starter_Kit_v3/
EMBEDDED HARDWARE 🥨 29

ENCLOSURE DESIGN
DIN rail mounted
Wall mounted
EMBEDDED HARDWARE 🥨 30

PROTECTION INDEX: IP..

IP: first digit IP: second digit
EMBEDDED HARDWARE 🥨 31

ATEX: EXPLOSION PROTECTION MARKING

Hazardous Area Classification
 🥨 32

SENSORS

digital view of an analog world

SENSORS 🥨 33

METERING

To measure the consumption

of water, gas, electricity…

Also to read existing values

on legacy, analog meters:
SENSORS 🥨 34

TEMPERATURE

To measure the amount

of heat energy in a source
Indoor & outdoor probes
AgTech: soil temperature
for crop growth.
SENSORS 🥨 35

HUMIDITY

To measure the amount of water vapor

in the atmosphere of air or other gases
Smart Industry/Buildings: in HVAC* systems
Smart City/Agri: in weather stations
Also to measure the conductivity
in various solids
* heating, vents and air conditioning
SENSORS 🥨 36

PRESSURE

To detect pressure fluctuations

Board-mount sensors
consider calibration, temperature compensation and amplification separately.

Transducers
compensated for temperature fluctuations, produces a voltage

Transmitters
compensated for temperature fluctuations, produces a 4-20 mA current

MEMS: Micro Electro Mechanical Systems

SENSORS 🥨 37

PRESSURE
Capacitive Strain-gauge Piezoresistive

capacitance change produced by foil or silicon strain gauges are arranged Piezoresistive sensing elements arranged in a bridge,
deflection of the membrane, as a Wheatstone bridge, in a diaphragm, attached to a flexible diaphragm. Resistance changes
which alters the capacitor geometry which deflects when pressure is applied according to the magnitude of the diaphragm deflection
SENSORS 🥨 38

PROXIMITY

Ultrasonic Infrared Inductive Capacitive

SENSORS 🥨 39

PROXIMITY

Photoelectric Magnetic LiDAR Light Detection and Ranging

Reflective, Through-beam, Low cost, long-range Excellent long range detection with fast update rates.
and Retro-reflective Magnets High end, expensive sensors
SENSORS 🥨 40

PROXIMITY
SENSORS 🥨 41

LEVEL Optical Floating switch Capacitance

Vibrating Ultrasonic Conductivity Microwave

tuning fork

SSTSensingLtd
SENSORS 🥨 42

LEVEL

level
matters
SENSORS 🥨 43

ACCELERATION

Acceleration / axis
SENSORS 🥨 44

ANGULAR ROTATION

Angle measurements
Inclinometers
SENSORS 🥨 45

GAS & PARTICLES

Air Quality:
up to 6 Gas sensors:
CO2, NH3, CO, H2S, FA...
and 4 Odor sensors
Particles:
PM 2.5 um and PM 10 um
SENSORS 🥨 46

DISTANT HEAT

To measure the a distant heat,

with infrared sensors
Security: differentiate humans,
animals vs. things
Smart building: people counting
SENSORS 🥨 47

LIGHT

To monitor lighting
in buildings and
tunnels for instance
 🥨 48

ACTUATORS

to act
ACTUATORS 🥨 49

RELAYS
ACTUATORS 🥨 50

CONTROL VALVES

to regulate
the flow
of fluids
ACTUATORS 🥨 51

MOTORS
AC Synchronous Motors
AC Induction Motors
DC Brushed Motors
DC Brushless Motors
…
Direct Drive Motors
Linear Motors
ACTUATORS 🥨 52

MOTORS - CONT’D
Servo Motors
Stepper Motors
ACTUATORS 🥨 53

VIBRATION MOTORS
 🥨 54

IOT PLATFORMS
IOT PLATFORMS 🥨 55

SO MANY IOT PLATFORMS

OPERATORS
PLATFORMS
IOT PLATFORMS
🥨
56
IOT PLATFORMS 🥨 57

PURE PLAYERS PLATFORMS

IOT PLATFORMS 🥨 58

IBM WATSON IOT + BLUEMIX

IOT PLATFORMS 🥨 59

IBM NODE RED

IOT PLATFORMS 🥨 60

MICROSOFT AZURE IOT

IOT PLATFORMS 🥨 61

BOSCH.IO
IOT PLATFORMS 🥨 62

SIEMENS MINDSPHERE
 🥨 63

AI

Essentials
AI: CONTEXT 🥨 64

LET’S MAKE STATISTICS GREAT (AGAIN)

STATISTICS 🥨 65

LINEAR REGRESSION
Regression analysis estimates the relationship between two or more variables
STATISTICS 🥨 66

LINEAR REGRESSION APPLICATIONS

Business: Relationship between advertising spending and new revenue.
Medecine: Relationship between drug dosage and blood pressure of patients.
Agriculture; Effect of fertilizer and water on crop yields.
Sports: Effect that different training regimens have on player performance.
Stocks: Prediction of how stocks will perform in the future based on past data
Retail: Prediction of consumer behavior
Etc.
STATISTICS 🥨 67

LINEAR REGRESSION ASSUMPTIONS

STATISTICS 🥨 68

STATISTICAL APPROACHES

Frequentist approach
Method which makes predictions on the underlying truths of the experiment,
using only data from the current experiment.

Bayesian approach
Method that encodes past knowledge of similar experiments into a statistical
device, known as prior. This prior is combined with current experiment data
to make a conclusion on the test. Essentially, you start out with a prior belief
and then update it in light of new evidence.
https://arbital.com/p/bayes_frequency_diagram/
STATISTICS 🥨 69

BAYESIAN STATISTICS
AI: CONTEXT 🥨 70

AI-WASHING…
AI: CONTEXT 🥨 71

SHORT HISTORY OF AI
AI: CONTEXT 🥨 72

PERFORMANCE EVOLUTION VS. HUMAN

AI: CONTEXT 🥨 73

MACHINE LEARNING
Supervised learning - system trained with data
examples with previously known results
> Regression: prediction of a continuous outcome,
e.g. level of manufacturing emissions.
> Classification: category (class) prediction,
e.g. whether an email is a spam or not

Self/Un-supervised learning - the system finds

internal relationships and patterns in the data.
In this case, the results for each example are unknown.

Reinforcement learning - approach in which the system is rewarded for correct actions
and penalized for wrong ones. As a result, the system learns to develop an algorithm in which
it receives the highest reward and the lowest penalty.
AI: CONTEXT 🥨 74

REINFORCEMENT LEARNING

PacMan example: where the goal of the agent (PacMan) is to

eat the food in the grid while avoiding the ghosts on its way.
The grid world is the interactive environment for the agent.
PacMan receives a reward for eating food and punishment
if it gets killed by the ghost (loses the game).
The states are the location of PacMan in the grid world and
the total cumulative reward is PacMan winning the game.
AI: NEURAL NETWORKS 🥨 75

LINEAR REGRESSION
Derived from statistics, there are multiple types
of regression algorithms:

๏ Simple Linear Regression

๏ Multiple Linear Regression
๏ Polynomial Regression
๏ Support Vector Regression
๏ Decision Tree Regression
๏ Random Forest Regression
AI: NEURAL NETWORKS 🥨 76

CLASSIFICATION
Classification Algorithms can be further divided
into the following types:

๏ Logistic Regression
๏ K-Nearest Neighbours
๏ Support Vector Machines
๏ Kernel SVM
๏ Naïve Bayes
๏ Decision Tree Classification
๏ Random Forest Classification
AI: NEURAL NETWORKS 🥨 77

NEURAL STRUCTURE
AI: NEURAL NETWORKS 🥨 78

STRUCTURE W/ HIDEN LAYERS

AI: NEURAL NETWORKS 🥨 79

Video: neural network examples

 🥨 80

AIOT = AI + IOT
AIOT: CONTEXT 🥨 81

Visual
recognition
AIOT: CONTEXT 🥨 82
AIOT: CONTEXT 🥨 83
AIOT: CONTEXT 🥨 84
AIOT: CONTEXT 🥨 85

https://www.visualcapitalist.com/aiot-when-ai-meets-iot-technology/
AIOT: MAIN MARKET SEGMENTS 🥨 86
AIOT: MAIN MARKET SEGMENTS 🥨 87
DATA 🥨 88

DEPTH OF INSIGHTS VS. SIMPLE IMMEDIACY

Speed of Actions

I/O G/W IoT Data

Sensor/ Edge Platform Center
Actuator

Depth of Insights
Tom Bradicich
 🥨 89

DISTRIBUTED LEDGERS

DCT, blockchain, smart contracts

BLOCKCHAIN 🥨 90

BLOCKCHAIN

Blockchain is essentially a shared database

filled with entries that must be
confirmed and encrypted.

- Business insider Intelligence

BLOCKCHAIN 🥨 91

HOW
BLOCKCHAIN
WORKS

https://www.ft.com/content/eb1f8256-7b4b-11e5-a1fe-567b37f80b64
BLOCKCHAIN 🥨 92

DISTRIBUTED LEDGER TECHNOLOGY (DLT)

Distributed ledger technology provides a trusted, immutable ledger

on which organizations can transmit and store valuable information
based on internal operations or interactions
with the organization’s environment,
including IoT devices.

https://www.iiconsortium.org/pdf/Digital_Transformation_in_Industry_Whitepaper_2020-07-23.pdf
https://www.iiconsortium.org/pdf/Distributed_Ledgers_in_IIoT_White_Paper_2020-07-22.pdf
BLOCKCHAIN 🥨 93

SMART CONTRACTS

Smart contracts are simply programs stored on a blockchain

that run when predetermined conditions are met.

They typically are used to automate the execution of an agreement

so that all participants can be immediately certain of the outcome,
without any intermediary's involvement or time loss.

- IBM
BLOCKCHAIN 🥨 94

SMART CONTRACTS

Efficiency, Sovereignty
Trust, Security, Savings
BLOCKCHAIN 🥨 95

https://youtu.be/n_XnEKJVKXo
https://youtu.be/iTGoS5IXrqQ
BLOCKCHAIN 🥨 96

https://youtu.be/n_XnEKJVKXo
 🥨 97

IOT SECURITY

…
IoT vulnerabilities & best practices
IOT SECURITY 🥨 98

SECURITY ALERT
IOT SECURITY 🥨 99

SECURITY ALERT

5.8b enterprise & automotive devices

were connected to the internet in 2020.

98% of all IoT device traffic is unencrypted.

57% of IoT devices are vulnerable

to medium or high severity attacks.

Sources: Gartner & Palo Alto networks, 2020

IOT SECURITY 🥨 100

THREATS ALONG THE VALUE CHAIN

Design faults
No updates unsecure
Bad lifecycle communications

policy no https code, lack of

management no VPNs penetration
… … testing…

DoS, policy management,

No/insecure updates
Physical attacks…
IOT SECURITY 🥨 101

CONNECTED CARS
multi-layered
vulnerabilities

McAfee
IOT SECURITY 🥨 102

IOT BOTNETS

TrendMicro
IOT SECURITY 🥨 103

SOME FAMOUS BOTNETS

Hydra (2008) DDoS via brute force attacks / routers
Aidra (2012) Cryptocurrency mining via Telnet attacks
Bashlite (2014) Attacks / Cameras, DVR, routers, etc.
Remaiten (2016) Sophisticated DDoS attacks via IoT devices
Mirai (2016) DNS attacks via 100,000 infected devices:
cameras, baby monitors, home routers, etc.
Persirai (2017) Camera control (1,000 models) via uPnP attacks
… TrendMicro
IOT SECURITY 🥨 104

MORE FAMOUS BOTNETS

IOT SECURITY 🥨 105

MIRAI ATTACK
Attack module
DoS methods
Process killer
telnet, SSH,
HTTP…
Scanner
random IP
addresses to
brute force

McAfee
IOT SECURITY 🥨 106

INDUSTRIES:
VIDEO CAMERAS

Nozomi
IOT SECURITY 🥨 107

AVOIDING WEAKEST LINK > E2E SECURITY

END TO END SECURITY

IOT SECURITY 🥨 108

GUIDANCE

Zero Trust approach

SASE approach
Strong Authentication
Security by design Strong Encryption
IoT SAFE (SIM based)
Etc.

Security by operations
IOT SECURITY 🥨 109

ALLIANCE
IOT SECURITY 🥨 110

BEST PRACTICES
IOT SECURITY 🥨 111

BEST PRACTICES
IOT SECURITY 🥨 112

APPS
PROFILE
IOT SECURITY 🥨 113

APPS PROFILE
IOT SECURITY 🥨 114

7 PRINCIPLES
to secure
connected
devices

Microsoft
IOT SECURITY 🥨 115

7 PRINCIPLES - CONT’D
to secure
connected
devices

Microsoft
 🥨 116

IOT USE CASES

Workshop
USE CASES 🥨 117

TRANSPORTATION
Please select ideal connectivity type(s) for

- handover at high speed > 300 km/h

- seamless roaming
- relatively high data rate (standard video)

?
LTE-cat1
LTE-M
USE CASES 🥨 118

SUPPLY CHAIN
Please select ideal connectivity type(s) for

- European roaming
- Geolocalisation
- Fleet management
- Updates over the air, etc.

?
LTE-M
USE CASES 🥨 119

ALARMS: LIFT OF DOUBT

Please select ideal connectivity type(s) for

- Large data volume

- Reliability, QoS
- Remote control, etc.

?
LTE-cat1
LTE-M
USES CASES 🥨 120

PERSONAL SAFETY
Please select ideal connectivity type(s) for

- Reliability
- Autonomy (edge computing)
- Remote control
- Indoor coverage, etc.

?
LTE-M
CELLULAR 🥨 121

ALERT BUTTONS
Please select ideal connectivity type(s) for

- Deep indoor coverage

- 2G sunset
- Reliability, QoS
- Etc.

?
LTE, LTE-cat1
LTE-M, NB-IoT
USE CASES 🥨 122

METERING / UTILITIES
Please select ideal connectivity type(s) for

- Deep indoor coverage

- Energy savings
- Over the air updates, etc.

?
NB-IoT
USE CASES 🥨 123

SUPPLY CHAIN / SHIPPING

Please select ideal connectivity type(s) for

Containers tracking
Global coverage

?
Solution hybride :
Cellulaire IoT + Satellite
 🥨 124

GLOSSARY

IoT
GLOSSARY 🥨 125

6LOWPAN Communication protocol which compresses IPv6 packages

Actuator Transforms electrical signal sinto different forms of energy - see sensor
ADR Adaptive Data Rate - to optimize LoRaWAN SF
API Application Programming Interface
BLE Bluetooth Low Energy
COAP Constrained Application Protocol
DCS Distributed Control System
Edge Where the action is, near data source
eMBB 5G's enhanced Mobile Broadband
EnOcean Energy harvesting wireless technology
ERP Enterprise Resources Planning
GSM Global System for Mobile communications
I2C Inter Integrated Circuits protocol
IaaS Infrastructure as a Service
IoT Internet Of Things
IP Internet Protocol - See TCP and UDP
IPv6 Version 6 of IP - allowing fixed addresses for all devices
LoRaWAN Long Range Wide Area Network
LTE-M Long Term Evolution : 4G
LTE-M Long Term Evolution / Cat M1
M2M Machine to Machine - part of IoT legacy
MES Manufacturing Execution System
ML Machine Learning
mMTC 5G's massive Machine Type Communications
MQTT Message Queuing Telemetry Transport
GLOSSARY 🥨 126

NB-IoT Narrow Band IoT

NFC Near Field Communication
PaaS Platform as a Service
PAN Personal Area Network
PLC Programmable Logic Controller
PoE Power over Ethernet
QoS Quality of Service
RFID Radio Frequency Identification
RSSI Received Signal Strength Indication
SCADA Supervisory Control and Data Acquisition
SaaS Software as a Service
Sensor Determine physical characteristic and transform them in electrical signals
SF Spreading Factor for LoRaWAN - SF7 (highest speed) to S12 (best coverage)
SLA Service Level Agreement
SNR Signal over Noise Ratio
TCP Transmission Control Protocol / IP - connected mode
UDP User Datagram Protocol / IP - non connected mode
UMTS 3G
URLLC 5G's Ultra Reliable Low Latency Communications
 🥨 127

IOT, NETWORKS
& TELECOM

let’s keep in touch:

[email protected]