Introduction to IoT

What is IoT?
The Internet of Things (IoT) is a network of smart devices embedded with sensors, software, and internet
connectivity to collect and share data.

IoT enables remote monitoring and control of devices, improving efficiency, accuracy, and automation.
Examples include smart home devices, healthcare wearables, connected vehicles, and industrial sensors. These
devices communicate and exchange data autonomously, enhancing various applications in daily life and
industries.

Genesis of IoT

 Start of IoT Era: Began around 2008-2009 when connected devices outnumbered the global
population.
 Term Coined: Kevin Ashton introduced the term “Internet of Things.”
 Technology Shift: Earlier, computers relied on human input; now, they use sensors to collect data
autonomously.
 Internet Evolution: IoT marks a key phase in the continuous growth of the Internet.

Current Phase of IoT

 Early Stage: Despite media coverage, IoT is still in its early phase, with 99% of objects yet to be
connected.
 Connectivity: Machines, objects, and humans are increasingly interconnected.
 Impact: Businesses and society are gaining vast data, insights, automation, and efficiency.
 Future Potential: IoT will transform the world, just like previous Internet advancements.
IoT and Digitization

 IoT connects objects and machines to a network like the Internet.

 Example: In malls, Wi-Fi tracking helps retailers understand customer movement.
 Digitization converts information into digital format.
 Example: Digital cameras replaced film cameras, changing the photography industry.

IoT Impact

1. Connected Roadways: Self-driving and regular vehicles communicate with roads and traffic systems.
2. Connected Factory: IoT improves automation, efficiency, and maintenance in factories.
3. Smart Connected Buildings: IoT enhances security, energy efficiency, and automation in buildings.
4. Smart Creatures: IoT devices monitor animals for health and environmental tracking.

1. Connected Roadways

 IoT helps self-driving cars interact with traffic systems and provide data to passengers.
 Connected roadways allow both driver and driverless cars to work with transportation infrastructure.
 The main goals of connected roadways are:
1. Safety – Prevent accidents and improve road security.
2. Mobility – Reduce traffic congestion and improve travel efficiency.
3. Environment – Lower vehicle emissions and improve air quality.

Benefits of Connected Roadways

 Reduced traffic jams and congestion.

 Fewer accidents and fatalities.
 Faster response times for emergency vehicles.
 Lower pollution from vehicles.

Example: Intersection Movement Assist (IMA)

 This system warns drivers or self-driving cars when it is unsafe to enter an intersection.
 It helps prevent collisions by detecting cars running stop signs or straying into other lanes.
 Communication between vehicles and traffic systems ensures quick and safe responses.

2.Connected Factory

Traditional factories face challenges due to disconnected systems. Managers struggle to track operations, fix
downtime issues, and improve efficiency.
Main Challenges in Factories:

 Launching new products and services quickly.

 Increasing production, quality, and uptime while reducing costs.
 Preventing unexpected downtime.
 Protecting factories from cyber threats.
 Reducing cabling costs.
 Improving worker safety and productivity.

Examples of Connected Factories:

1. Better Manufacturing Efficiency:

o Factories are using advanced technology to cut costs and improve flexibility.
o Automation and real-time monitoring help managers make quick decisions.

2. Smelting Facility Improvement:

o Operators previously had to walk long distances to control rooms for data.
o With IoT, real-time data is sent to mobile devices, saving time and improving production.

3. Real-Time Tracking in Factories:

o Small RFID tags track materials and production stages.
o Managers can monitor assembly lines in real time, adjusting speed and fixing bottlenecks.

IoT is transforming industries, making factories smarter, more efficient, and connected. This change is part of
the "Fourth Industrial Revolution."

3.Smart Connected Buildings

Buildings provide a safe and comfortable work environment. They need good lighting and the right temperature.
Fire alarms and security systems keep workers safe.

Modern buildings use smart systems to improve these functions. Temperature sensors help control airflow.
These sensors send data to the building management system (BMS).

Smart buildings are easier and cheaper to manage. Running a large building is expensive. Many people work in
these buildings every day. Managers want to make them more efficient.
Buildings use a communication protocol called BACnet. BACnet connects devices like HVAC, lighting, and
security systems. It allows these devices to communicate over an Ethernet network.

BACnet/IP helps connect the building network to the internet. This makes it easier to manage all systems from
one place.

Another important IoT technology in smart buildings is the digital ceiling.

The digital ceiling is more than a lighting system. It connects different building networks. These include
lighting, HVAC, blinds, CCTV, and security systems.All these systems work together on a single IP network.
This makes building management easier and more efficient.

The digital ceiling is mainly built around the lighting system.

The lighting industry is shifting to LEDs. LEDs use less energy and last longer than traditional lights.

4. Smart Creatures
IoT can also connect living things to the Internet. Sensors can be placed on animals and insects, just like
machines.

A well-known example is the connected cow. A Dutch company, Sparked, created a sensor placed in a cow’s
ear.

The sensor tracks the cow’s health and location. It sends 200 MB of data per year to farmers.

With this data, farmers can monitor the entire herd. They can see how the environment and diet affect the cows.

These sensors help detect diseases early. Cows eat less days before they show symptoms.

The sensors can even detect pregnancy in cows.

Convergence of IT and OT

IT (Information Technology) and OT (Operational Technology) used to be separate.

IT manages internet connections, data, and technology systems. It ensures secure data flow in an organization.

OT controls physical systems like assembly lines, utilities, and production facilities.

In the past, IT did not handle OT processes like production and logistics.

IT manages business systems like emails, databases, and file services.

OT manages industrial equipment like factory machines, meters, and SCADA systems.

IoT Challenges

IoT has great potential, but it also has challenges.

Many IoT technologies are already in use.

However, some obstacles must be solved for IoT to be used everywhere in industries and daily life.
IoT Network Architecture and Design

IT networks have evolved over the past 20 years and are well understood.

IoT network architecture is new and needs a different approach.

IT and IoT networks have some similarities, but their challenges and needs are very different.

IoT networks are often managed by OT, which handles operational systems.

IT networks focus on moving data, while IoT networks focus on collecting and using sensor data.

Drivers Behind New Network Architectures

IT networks support business applications like email and databases.

IoT networks focus on gathering, transporting, analyzing, and using sensor data.

New network architectures are needed to handle IoT’s unique requirements.

A comparison of IT and IoT networks helps identify the changes required for IoT.
Comparing IoT Architectures
Two of the best-known architectures are
1. oneM2M
2. The IoT World Forum

1.One M2M IoT Standardized Architecture

1. The European Telecommunications Standards Institute (ETSI) started the M2M Technical Committee
in 2008.
2. The goal was to standardize machine-to-machine (M2M) communication.
3. Over time, the scope expanded to include IoT.
4. oneM2M aims to create a common services layer for IoT devices.
5. This layer helps devices communicate with application servers.
6. oneM2M supports various IoT applications like:
o Smart meters
o Smart grids
o Smart cities
o E-health
o Connected vehicles

The oneM2M architecture has three main layers:

1. Application Layer
o Focuses on connecting devices and applications.
o Includes application-layer protocols.
o Standardizes northbound APIs for business intelligence (BI) systems.
o Applications are industry-specific and have different data models.
2. Services Layer
 o Works horizontally across different industries.
o Includes physical networks, management protocols, and hardware.
o Supports backhaul communications like cellular, MPLS, and VPNs.
o Has a common services layer with APIs and middleware for third-party services.

1. Goal:
o Create a common M2M Service Layer for connecting devices to cloud-based
applications.
o Ensure easy integration with different hardware and software.

2. Key Industries Involved:

o Transportation, healthcare, utilities, industrial automation, and smart homes.

3.Network Layer:

o Responsible for communication between IoT devices.

o Uses wireless technologies (IEEE 802.15.4, 801.11ah) and wired connections (IEEE
1901 power line communications).

IoT World Forum (IoTWF) Architecture

1. Created in 2014 by Cisco, IBM, and Rockwell Automation.

2. Seven-layer model for simplifying IoT architecture.
3. Includes edge computing, data storage, and access.
4. Data moves from sensors at the edge to a cloud or data center.
5. Benefits of the IoTWF model:
o Breaks IoT into smaller parts for better understanding.
o Identifies different technologies and their relationships.
o Allows different vendors to provide solutions for different layers.
o Defines interfaces that improve interoperability.
o Ensures security at different layers through a tiered model.

Layer 1: Physical Devices and Controllers

  This layer includes all IoT devices and sensors.
 Devices send and receive data over a network.
 The size of devices can vary from tiny sensors to large industrial machines.
 Their main role is to generate data and be controlled remotely.

Layer 2: Connectivity Layer

 This layer handles data transmission between devices and the network.
 It ensures data is sent reliably and on time.
 It connects Layer 1 devices to Layer 3 (Edge Computing). This layer includes all networking
components of IoT.

 It does not separate the last-mile network, gateway, or backhaul network.

 The last-mile network connects sensors/endpoints to the IoT gateway.
 The gateway helps route data between devices and the cloud.
 The backhaul network carries data from the gateway to the central system.

Edge Computing Layer (Layer 3)

 This layer reduces data before sending it to higher layers.

 It converts raw network data into useful information.
 Processing starts as early and as close to the edge as possible.
 This helps reduce network load and speeds up decision-making.

Upper Layers: Layers 4–7

The upper layers (Layers 4–7) manage and process IoT data from the lower layers. They store, analyze, and
share data for applications and decision-making. These layers ensure data is useful and accessible.
IT and OT Responsibilities in the IoT Reference Model

1. Demarcation Between IT and OT

o IoT architecture naturally separates responsibilities between IT and OT.
o IT handles applications, databases, and networks.
o OT manages physical devices, sensors, and industrial systems.
2. Boundaries Beyond Functional Layers
o OT controls the bottom layers, including sensors and industrial machinery.
o IT governs the top layers, including data processing and applications.
o Traditionally, IT and OT operated independently, but IoT bridges this gap.
3. Data Flow and Challenges
o OT devices generate real-time, high-volume data.
o IoT networks must handle massive data transfer efficiently.
o IT systems must process and store this data effectively for decision-making.

IoT integration is driving convergence between IT and OT, requiring better coordination and data management.
IT and OT in the IoT Reference Model

1. Separation of IT and OT
o OT manages physical devices like sensors in industries such as oil and gas.
o IT handles servers, databases, and applications.
2. IoT Bridges IT and OT
o Traditionally, IT and OT worked separately.
o IoT connects them, requiring better coordination.
3. Data Flow in IoT
o OT devices generate large amounts of real-time data.
o Data must be buffered and stored for IT applications.
o Data moves from "in motion" (OT) to "at rest" (IT).
4. Collaboration for Data Management
o IT and OT must work together to manage data.
5. Other IoT Reference Models
o Different industries use specific IoT models.
o Standards bodies support various frameworks.

Simplified IoT Architecture

1. Two Parallel Stacks

o IoT Data Management and Compute Stack
o Core IoT Functional Stack
2. Purpose of Simplification
o The model still includes all necessary details.
o It simplifies IoT into key building blocks.
o Helps in understanding design and deployment principles.
3. Functionality
o Covers all layers of complex models.
o Groups them into easy-to-understand functional blocks.
Core IoT Functional Stack

1. Better Visibility
o Separating layers improves understanding.
o The three-layer model shows foundational IoT building blocks.
2. Expanded View
o The Core IoT Functional Stack can be divided into more detailed sublayers.
3. Communications Layer Breakdown
o Divided into four sublayers:
 Access Network
 Gateways and Backhaul
 IP Transport
 Operations and Management
4. IoT Applications Layer
o Different from typical enterprise networks.
o Includes big data analytics and industry-specific IoT control systems.

Core IoT Functional Stack

1. Things Layer
o Physical devices must fit their environment while providing data.
2. Communications Network Layer (4 Sublayers)

1. Access Network Sublayer – The last mile of IoT using wired or wireless technologies like
Wi-Fi (802.11ah) and LoRa.
2. Gateways & Backhaul Sublayer – Connects smart objects to a central system via a gateway
that forwards data.
 3. Network Transport Sublayer – Uses IP and UDP for communication across different
devices.
4. IoT Network Management Sublayer – Uses protocols like CoAP and MQTT for data
exchange.
3. Application & Analytics Layer

o Processes collected data to control devices and make intelligent decisions.

Layer 1: Sensors and Actuators ("Things")

 IoT starts with connected physical objects (smart devices).

Classifications of Smart Objects:

1. Power Supply – Battery-powered or externally powered.

2. Mobility – Mobile or fixed in one place.
3. Reporting Frequency – Low (e.g., monthly) or high (frequent data updates).
4. Data Complexity – Simple (basic values) or rich (detailed sensor data).
5. Report Range – Distance between the sensor and the gateway.
6. Object Density – Number of devices connected per area.

layer 2: Communications Network Layer

1. Connecting Smart Objects

o Transmission depends on object form (range, data, frequency, density, and mobility).
2. Access Network Sublayer
o IoT network technology determines connectivity type and topology.
o Each technology is designed for specific use cases (what, where, how much, how often, and
how far to connect).
3. Technology Selection Factors
o Frequency band – Suitable for expected data transmission.
o Frame structure – Matches packet size and communication intervals.
o Network topology – Defines how devices are connected.
o Transmission range – Distance between the smart object and data collector.
IoT Communication and Data Management

Communication Network Layer

1. Transmission Distance & Network Types

o Cellular: Works best beyond 5 km but can work at 100 m.
o ZigBee: Works for a few meters but not over 10 km.
o Common Network Groups:
 PAN (Personal Area Network) – A few meters (e.g., Bluetooth).
 HAN (Home Area Network) – A few tens of meters (e.g., ZigBee, BLE).
 NAN (Neighborhood Area Network) – A few hundred meters.
 FAN (Field Area Network) – Hundreds of meters (outdoor, unsecured).
 LAN (Local Area Network) – Up to 100 m (Ethernet, Wi-Fi).
2. Gateways & Backhaul
o Data from sensors is sent to a central station.
o Gateways help transmit data from short-range networks to long-range ones.
o Example: Cars communicate with road sensors through gateways using Dedicated Short-
Range Communication (DSRC).
3. Network Transport Sublayer
o Different communication structures exist:
 Peer-to-peer (e.g., meter to meter).
 Point-to-point (e.g., sensor to central station).
 Point-to-multipoint (e.g., gateway to multiple sensors).
o Data transmission occurs over various media (wired, wireless, power lines, etc.).
4. IoT Network Management Sublayer
o Protocols like IP, TCP, UDP enable connectivity.
o Data transfer can be:
 Push model – Sensor sends data regularly.
 Pull model – Application requests data.
o XMPP (based on instant messaging) helps devices communicate but relies on TCP, which can
be limiting.

Applications & Analytics Layer

1. Types of IoT Applications

o Analytics Applications – Collect and process sensor data (e.g., weather predictions, business
dashboards).
 o Control Applications – Change the behavior of smart objects (e.g., adjusting a pump based on
sensor input).
2. Types of Analytics
o Data Analytics – Processes collected data to generate insights.
o Network Analytics – Monitors network performance to prevent failures.
3. Business Benefits of Analytics
o IoT systems must be flexible to allow future sensor additions.
o Example: Cisco Jasper helps manage vending machines efficiently.

IoT Data Management & Compute Stack

1. Challenges in IoT Data Handling

o Huge Data Volume – Most sensor data is unstructured and not always useful.
o Latency Issues – Fast processing is crucial for real-time decisions.
o Limited Bandwidth – Too much data can overload the network.
o Backhaul Limitations – Connectivity (3G/LTE, satellite) can be unreliable or expensive.
o Big Data Complexity – Cloud storage and analysis of all IoT data is impractical.
2. Solutions for Efficient Data Processing
o Process data close to the device to reduce latency.
o Minimize unnecessary data transmission to save bandwidth.
o Use localized responses rather than centralizing everything in the cloud.

Fog Computing

1. What is Fog Computing?

o It is a type of edge computing in IoT.
o Any device with computing, storage, and network capabilities can act as a fog node.
o Examples: Industrial controllers, switches, routers, embedded servers, and IoT gateways.
2. Why is Fog Computing Useful?
o Processes data close to where it is collected, reducing delays (latency).
o Reduces network congestion by keeping unnecessary data out of the main network.
o Keeps sensitive data local, improving security.
3. How Does It Work?
o Fog nodes analyze data before sending it to the cloud.
o They help make quick decisions without waiting for cloud processing.
o This improves performance, especially on limited (constrained) networks.
4. Fog Layer in IT Systems
o Traditional IT has only cloud and core networks.
o The fog layer is a new middle layer between devices and the cloud.
o It helps with faster data processing and better control.
 1. Fog computing happens close to IoT devices to process data near the source.
2. Fog nodes understand local sensors because they are physically close to them.
3. Example: A fog router on an oil rig analyzes sensor data before sending only important details to the
cloud.
4. Less data is sent to the cloud, reducing network congestion and costs.
5. Faster reactions to events without waiting for cloud processing.
6. Example: A fog node on a truck monitors tire pressure and engine sensors but only sends alerts if
there is a real issue.
7. Fog nodes monitor, analyze, and take action in real time (e.g., opening a valve, locking a door, or
sending alerts).

Key Features of Fog Computing

 Low latency & context awareness: Close to IoT devices for quick response.
 Geographically distributed: Unlike centralized cloud systems.
 Near IoT endpoints: Usually deployed where many IoT devices exist.
 Wireless communication: Best for scaling large IoT networks.
 Real-time processing: Focuses on instant data interactions, not batch processing.

Edge Computing

1. Pushes computing even further to the device itself (e.g., sensors and IoT devices).
2. Some IoT devices can process data independently with built-in computing power.
3. Example: A fire hydrant sensor detects a water pressure drop and alerts the system immediately.
4. Smart meters communicate with each other to monitor local power grids and report only critical data.
5. Edge computing works alongside fog computing, processing simple data locally while fog nodes
handle complex analysis.

Hierarchy: Edge, Fog, and Cloud

1. Edge computing (closest to devices) processes immediate data.

2. Fog computing (middle layer) analyzes and filters relevant information.
3. Cloud computing (highest layer) handles large-scale data storage and deep analysis.
Each layer reduces the load on the cloud, improving efficiency and response times.
Despite differences, all layers work together rather than replacing each other.

1. Fog nodes receive data from IoT devices at the network edge.
2. The fog application directs data based on urgency:
o Urgent data: Processed immediately on the fog node.
o Less urgent data: Sent to an aggregation node for further processing.
o Historical data: Sent to the cloud for long-term storage and analysis.
3. Example:
o A factory sensor detects overheating → Immediate action at the fog node.
o A city’s traffic data → Sent to an aggregation node for short-term analysis.
o Energy usage trends over a year → Stored in the cloud for big data analysis.

This structure ensures faster responses, efficient data management, and reduced cloud load.