Unit I

Introduction to Internet of Things

Introduction to loT: loT- loT terms and basic Definitions-Characteristics of loT- loT Eco
System –loT reference Model - Level 1 Physical Devices and Controllers-Level 2
Connectivity-Level 3 Edge Computing - Level 4 Data Accumulation - Level 5 Data
Abstraction - Level 6 Application - Level 7 Collaboration Process - Security in the loT.

1. Introduction to IoT
The Internet of Things (IoT) is a technology that IOT TERMS AND BASIC DEFINITIONS
connects physical objects such as sensors, THINGS
appliances, vehicles, machines, and wearable Physical objects connected to the Internet.
devices to the Internet. These smart devices collect Sensor
data, exchange information, and perform actions A device that detects physical conditions like
automatically with little or no human intervention. temperature, humidity, pressure, motion, etc.
IoT integrates sensing, communication, cloud Actuator
computing, and data analytics to create intelligent Converts electrical signals into physical actions
systems that improve efficiency, productivity, and such as opening a valve or switching on a motor.
quality of life. Gateway
Key Features Acts as an intermediary between IoT devices and the
 Connects physical objects through the cloud.
Internet. Cloud Computing
 Enables real-time monitoring and control. Provides storage, computing power, and services for
 Supports automation and smart decision- IoT applications.
making. Edge Computing
 Uses sensors and actuators to interact with Processes data near the source instead of sending
the environment. everything to the cloud.
 Improves operational efficiency and Connectivity
resource utilization. Communication between devices using protocols
2. What is IoT? like Wi-Fi, Bluetooth, ZigBee, LoRaWAN, NFC,
IoT refers to a network of interconnected devices and 5G.
capable of sensing, processing, and exchanging data 4. Disambiguation of IoT vs IoE vs M2M vs
over communication networks. Others
Definition Technology Description
Internet of Things (IoT) is a network of physical Connects smart devices
devices embedded with sensors, software, IoT
through the Internet.
processors, and communication technologies that IoE (Internet of Connects people, processes,
enable them to collect, share, and analyze data Everything) data, and devices together.
through the Internet. Direct communication between
Components of IoT M2M (Machine-
machines without human
 Smart Devices to-Machine)
involvement.
 Sensors
Integrates computing with
 Actuators CPS (Cyber
physical processes for real-time
 Communication Network Physical System)
control.
 Gateway
WSN (Wireless Network of sensors collecting
 Cloud Platform
Sensor Network) environmental information.
 Data Analytics
Difference
 User Applications
  IoT is Internet-based. Devices continuously change their operating
 IoE is broader than IoT. conditions.
 M2M mainly focuses on communication Scalability
between machines. Supports millions of connected devices.
 CPS combines computation and physical Heterogeneity
systems. Different hardware and communication
technologies work together.
CHARACTERISTICS OF IoT Self-Configuration
Connectivity Devices automatically configure themselves.
Devices communicate using wired or wireless Real-Time Communication
networks. Immediate data collection and processing.
Intelligence Security and Privacy
Collected data is analyzed to make smart decisions. Protects user information and prevents unauthorized
Dynamic Nature access
.
IOT ECOSYSTEM Edge devices are computing devices positioned
between IoT sensors and cloud servers. Instead of
sending all collected data directly to the cloud, edge
devices perform preliminary processing close to
where the data is generated. This reduces
communication delay, minimizes bandwidth usage,
improves response time, and enhances system
reliability. They also provide security functions
such as authentication and data filtering before
forwarding information to cloud platforms.

Examples: IoT Gateway, Raspberry Pi, Industrial

Gateway, Smart Router.

3. Embedded Systems
1.2.2 Enabling Technologies in IoT
An embedded system is a dedicated computer
The success of the Internet of Things depends on system designed to perform a specific task within a
several enabling technologies that allow smart larger electronic device. Unlike general-purpose
devices to sense, communicate, process, and computers, embedded systems are optimized for
analyze data. These technologies work together to reliability, low power consumption, and real-time
create intelligent systems capable of automation and operation. They consist of processors, memory,
real-time decision making. The major enabling input/output interfaces, sensors, and communication
technologies include sensors, edge devices, modules. Embedded systems form the intelligence
embedded systems, communication technologies, of smart devices by processing sensor data and
wireless sensor networks, cloud computing, digital controlling actuators.
twins, and big data analytics.
Applications: Smart TVs, Washing Machines,
1. Sensors Smart Watches, Medical Equipment, Industrial
Controllers.
Sensors are the most fundamental components of an
IoT system. They are electronic devices that detect 4. Communication Technologies
changes in the surrounding environment and
convert physical quantities into digital signals. Communication technologies provide connectivity
These signals are transmitted to processing units or between IoT devices and cloud platforms.
cloud servers through communication networks. Depending on application requirements,
Sensors enable smart devices to monitor communication may be wired or wireless. Wireless
temperature, humidity, pressure, light intensity, technologies are suitable for mobile and remote
motion, gas concentration, and many other applications, whereas wired communication offers
environmental parameters. Modern Micro-Electro- greater reliability and higher data transfer speeds.
Mechanical Systems (MEMS) technology has made
sensors smaller, cheaper, and more energy efficient, Wireless Technologies
allowing them to be embedded into wearable
devices, industrial machines, and household  Wi-Fi
appliances.  Bluetooth & BLE
 ZigBee
 RFID
Examples: Temperature sensor, Pressure sensor,
 NFC
Motion sensor, Humidity sensor, Gas sensor.
 LoRaWAN
 SIGFOX
2. Edge Devices
  Cellular Networks (3G, 4G, 5G)  Platform as a Service (PaaS) – Offers
development platforms for application
Wired Technologies creation.
 Software as a Service (SaaS) – Provides
 Ethernet software applications through the Internet.
 HomePlug
 HomeGrid 7. Digital Twin
 Telephone Network
A Digital Twin is a virtual representation of a
The selection of communication technology physical object or system. It continuously receives
depends on transmission range, power consumption, real-time data from sensors installed in the physical
bandwidth, and cost. device and mirrors its behavior in a digital
environment. Engineers can monitor equipment
5. Wireless Sensor Networks (WSN) performance, predict failures, and optimize
operations without interrupting the actual system.
A Wireless Sensor Network consists of a large Digital twins are widely used in manufacturing,
number of sensor nodes that communicate healthcare, aerospace, and smart city applications.
wirelessly to monitor environmental or physical
conditions. Each sensor node collects information 8. Big Data Analytics
and forwards it to a coordinator or gateway, which
then transmits the data to cloud servers for analysis. IoT devices continuously generate huge amounts of
WSNs are highly scalable, energy efficient, and structured and unstructured data. Big Data Analytics
suitable for remote monitoring applications. refers to the process of collecting, storing,
processing, and analyzing this massive volume of
Applications information to discover useful patterns and support
intelligent decision making. Organizations use
 Weather Monitoring analytics to improve operational efficiency, predict
 Air Quality Monitoring equipment failures, understand customer behavior,
 Smart Agriculture and automate business processes.
 Structural Health Monitoring
 Industrial Automation Big Data is characterized by the Three Vs:
 Military Surveillance
 Volume – Large amount of data generated.
6. Cloud Computing  Velocity – High speed of data generation
and processing.
Cloud computing enables IoT devices to access  Variety – Different formats such as text,
computing resources such as storage, processing images, videos, and sensor readings.
power, software, and databases through the Internet.
Since IoT devices generate enormous volumes of 1.2.3 Applications of IoT
data, cloud platforms provide scalable infrastructure
to store, process, and analyze this information. IoT has transformed almost every sector by enabling
Cloud computing also allows users to access IoT intelligent monitoring, automation, and remote
applications from anywhere at any time. management. In smart homes, IoT controls lighting,
security systems, and household appliances. In
Cloud services are generally classified into: healthcare, wearable devices monitor patients
continuously and support remote diagnosis. Smart
 Infrastructure as a Service (IaaS) – agriculture uses IoT for irrigation management, crop
Provides virtual hardware resources. monitoring, and soil analysis. Industrial IoT
improves manufacturing through predictive
maintenance and process automation.
Transportation systems use IoT for vehicle tracking, creating new business opportunities and
traffic management, and smart parking. transforming traditional industries.
Environmental monitoring applications measure
pollution levels, weather conditions, and natural 1.2.5 Vision of IoT
disasters. Retail businesses use IoT for inventory
management and customer analytics, while energy The vision of the Internet of Things is to create a
sectors employ smart grids and smart meters for world in which billions of physical objects are
efficient power distribution. intelligently connected through the Internet. These
smart objects can sense their environment,
1.2.4 Marketplace of IoT communicate with other devices, process
information, and perform actions without
The IoT marketplace is expanding rapidly due to continuous human intervention. A smart object
increasing demand for connected products and should possess a unique identity, communication
digital services. It consists of device manufacturers, capability, sensing ability, processing power, and
cloud service providers, software developers, networking functionality. The ultimate goal of IoT
communication service providers, research is to improve efficiency, automate routine tasks,
institutions, governments, entrepreneurs, and optimize resource utilization, support data-driven
consumers. Technology companies such as decision making, and enhance the quality of life
Amazon, Microsoft, Google, and Apple provide through intelligent services. As IoT continues to
cloud-based IoT platforms, while governments evolve, it will become the foundation for smart
invest in smart city projects and digital homes, smart industries, smart healthcare, and smart
infrastructure. Researchers develop new IoT cities.
technologies, and entrepreneurs create innovative
IoT products for healthcare, agriculture, industry,
and transportation. The rapid growth of IoT is
IoT REFERENCE MODEL  Secure communication

The IoT Reference Model is a seven-layer Level 3 – Edge (Fog) Computing

architectural framework that explains how data Edge computing performs local processing before
flows from physical devices to business sending data to the cloud. It filters unnecessary
applications and decision-making processes. It information, reduces latency, improves response
provides a standardized approach for designing, time, and minimizes bandwidth usage.
implementing, and managing IoT systems. Each Functions
layer performs a specific function, starting from  Data filtering
data collection at the device level and ending with  Packet inspection
business collaboration and intelligent decision-  Peer-to-peer communication
making. The model also supports scalability,  Local decision making
interoperability, flexibility, and easy maintenance  Data cleanup
of IoT applications.
Level 4 – Data Accumulation
Data Flow in IoT This layer stores incoming IoT data in databases or
 Upward Flow: Sensors → Network → cloud storage. Raw streaming data is converted
Edge → Storage → Information → into stored information that can be analyzed later.
Applications → Business Decisions. Functions
 Downward Flow: Commands generated by  Data storage
applications are sent back to physical  Sampling
devices for monitoring and control.  Filtering
 Many IoT systems support bidirectional  Aggregation
communication, enabling continuous  Data conversion
monitoring and automation.
Level 5 – Data Abstraction
Seven Layers of IoT Reference Model The stored data is processed and transformed into
meaningful information. Different data sources are
Level 1 – Physical Devices and Controllers integrated and made available for applications.
This is the lowest layer where IoT data is Functions
generated. It consists of sensors, actuators, RFID  Data aggregation
tags, robots, smart appliances, and embedded  Information integration
controllers. Sensors collect real-world information  Data access
such as temperature, pressure, humidity, and  Data transformation
motion, while actuators perform actions based on  Data visualization
received commands.
Examples: Temperature sensor, Arduino, Level 6 – Application
Raspberry Pi, RFID Reader, Smart Meter. This layer provides services to end users by using
processed IoT data. Applications display
Level 2 – Connectivity dashboards, analytics, reports, alerts, and control
This layer transfers data from physical devices to options.
higher layers through communication networks. It Functions
includes routers, gateways, switches, hubs, and  Device monitoring
communication protocols.  Business intelligence
Technologies: Wi-Fi, Bluetooth, ZigBee, NFC,  Data analytics
Ethernet, Cellular (4G/5G), MQTT, CoAP.  Smart dashboards
Functions  Mobile and Web applications
 Device communication
 Data transmission Level 7 – Collaboration and Processes
 Network connectivity
This is the highest layer where IoT information  People interaction
supports organizational decision-making and  Business processes
business operations. Insights generated by IoT  Workflow automation
applications are used to automate workflows,  Decision support
improve productivity, and support strategic  Enterprise management
planning.
Functions
LEVEL 2: CONNECTIVITY gateways or cloud servers. They reduce network
traffic and improve response time.
Introduction
2. Routers
Level 2 of the IoT Reference Model is known as
the Connectivity Layer. It provides the Routers connect different networks and determine
communication infrastructure required to transfer the best path for transmitting data packets. They
data generated by Level 1 devices (physical ensure efficient communication between local and
devices and controllers) to higher layers such as external networks.
Edge Computing, Cloud, and Applications. This
layer ensures reliable, secure, and efficient 3. Hubs
communication between IoT devices, gateways,
cloud platforms, and users. A hub connects multiple IoT devices within the
same network and forwards incoming data to all
The main purpose of the Connectivity Layer is to connected devices.
establish a communication path between devices
and the IoT system using wired or wireless 4. Switches
communication technologies. It acts as the bridge
that enables data exchange throughout the IoT Switches intelligently forward data only to the
ecosystem. intended destination, improving network efficiency
and reducing unnecessary traffic.
Objectives of Level 2
Functions of the Connectivity Layer
The major objectives of the Connectivity Layer
are: The Connectivity Layer performs several important
functions.
 To establish reliable communication
between IoT devices and the network.  Provides communication between IoT
 To transmit data generated by sensors to devices.
processing units.  Ensures reliable delivery of data.
 To enable communication among different  Implements different communication
IoT systems. protocols.
 To support communication between Level  Performs routing and switching of data
1 devices and Level 3 (Edge/Fog packets.
Computing).  Translates data between different protocols.
 To provide secure and uninterrupted data  Provides network-level security through
transmission. authentication and encryption.
 To support different communication  Supports scalable communication for large
protocols for interoperability. IoT deployments.
 Enables continuous communication
Communication and Processing Units between devices, gateways, and cloud
services.
Several networking components work together in
the Connectivity Layer. Communication Models in IoT

1. Edge Devices Different communication models are used

depending on the application requirements.
Edge devices receive data from sensors and
perform basic processing before forwarding it to 1. Device-to-Device (D2D) Communication
In this model, two or more IoT devices Applications
communicate directly without involving cloud
services. Communication takes place through  Smart home gateways
short-range wireless technologies such as  Industrial IoT
Bluetooth, ZigBee, or Z-Wave.  Smart buildings

Advantages 4. Back-End Data Sharing Model

 Fast communication In this model, cloud platforms share collected IoT

 Low power consumption data with multiple applications and service
 Suitable for home automation providers. Different organizations can access the
same data for analytics and decision-making.
Applications
Advantages
 Smart lighting
 Smart locks  Better analytics
 Home appliances  Data integration
 Multi-user access
2. Device-to-Cloud Communication
Applications
Here, IoT devices connect directly to cloud servers
through the Internet using Wi-Fi, Ethernet, or  Smart cities
Cellular networks. The cloud stores, processes, and  Traffic management
analyzes the collected data.  Healthcare systems

Advantages IoT Communication Protocols

 Remote monitoring The Connectivity Layer supports several wired and

 Large storage capacity wireless communication protocols.
 Easy device management
Bluetooth
Applications
Bluetooth is a short-range wireless communication
 Smart healthcare technology widely used in IoT applications.
 Weather monitoring Bluetooth Low Energy (BLE) provides low power
 Smart agriculture consumption, making it suitable for wearable
devices and health monitoring systems.
3. Device-to-Gateway Communication
Applications
In this model, IoT devices first communicate with
a local gateway (such as a smartphone, router, or  Smart watches
IoT gateway). The gateway performs protocol  Fitness bands
conversion, security checks, and forwards the data  Medical devices
to cloud servers.
ZigBee
Advantages
ZigBee is based on IEEE 802.15.4 and supports
 Improved security low-power mesh networking. It is suitable for
 Protocol translation connecting many devices with low data rates.
 Reduced network traffic
Applications  Passive RFID Tags
 Active RFID Tags
 Home automation
 Smart lighting Applications
 Industrial monitoring
 Inventory management
Wi-Fi  Toll collection
 Asset tracking
Wi-Fi provides high-speed wireless
communication over local area networks. It Z-Wave
supports Internet connectivity and is commonly
used when high data rates are required. Z-Wave is a low-power wireless communication
technology mainly used in smart home automation.
Applications
Applications
 Smart cameras
 Smart TVs  Smart locks
 Home automation  Smart lighting
 Home security
NFC (Near Field Communication)
LoWPAN
NFC enables communication between two devices
placed very close to each other (typically within 10 6LoWPAN enables IPv6 communication over low-
cm). It supports secure and quick data exchange. power wireless personal area networks. It allows
IoT devices to communicate directly over the
Applications Internet.

 Contactless payment Applications

 Access control
 Smart cards  Smart buildings
 Industrial automation
Cellular Networks (2G/3G/4G/5G)
Thread
Cellular technology provides wide-area
communication through mobile networks. It is Thread is an IP-based mesh networking protocol
suitable for IoT applications requiring long- designed for secure and reliable smart home
distance communication. communication.

Applications Applications

 Vehicle tracking  Home automation

 Smart transportation  Smart sensors
 Remote healthcare  Smart appliances

RFID (Radio Frequency Identification) Sigfox

RFID uses radio waves to identify and track Sigfox is a Low Power Wide Area Network
objects through RFID tags and readers. (LPWAN) technology designed for very low-
power IoT devices requiring long battery life.
RFID tags are of two types:
Applications  Platform independent
 Easy cloud integration
 Smart meters
 Environmental monitoring WebSocket APIs

Neul WebSocket provides full-duplex communication

between the client and server using a single
Neul uses TV White Space frequencies for long- persistent connection.
range communication while consuming very little
power. Unlike REST, WebSocket does not create a new
connection for every request.
Applications
Advantages
 Smart utility monitoring
 Industrial IoT  Real-time communication
 Low latency
LoRaWAN  Reduced network overhead
 Continuous data transfer
LoRaWAN is another LPWAN technology
designed for long-range communication with low Applications
power consumption.
 Live IoT dashboards
Applications  Smart surveillance
 Online monitoring systems
 Smart agriculture
 Environmental monitoring Importance of Level 2 Connectivity
 Smart cities
The Connectivity Layer is one of the most
RESTful APIs in IoT important layers in the IoT Reference Model
because it enables all IoT devices to communicate
REST (Representational State Transfer) is one of efficiently.
the most commonly used communication methods
in IoT. It provides:

REST uses HTTP methods:  Reliable communication

 Secure data transmission
 GET – Retrieve data  Protocol interoperability
 POST – Create new data  Cloud connectivity
 PUT – Update existing data  Device scalability
 DELETE – Remove data  Fast data exchange
 Network management
REST communication follows a request-response  Support for Edge/Fog Computing
model between the client and server.
Without the Connectivity Layer, IoT devices
Advantages cannot exchange information, making the entire
IoT system ineffective.
 Simple
 Scalable
Level 3: Edge (Fog) Computing To overcome these limitations, Edge Computing
processes the data locally or near the source before
Introduction forwarding only useful information to the cloud.

Level 3 of the IoT Reference Model is known as Functions of Level 3

the Edge (Fog) Computing Layer. This layer
receives data from the Connectivity Layer (Level The Edge Computing layer performs several
2) and performs preliminary processing before important operations before data reaches higher
forwarding the data to higher layers such as Data layers.
Accumulation (Level 4) and Data Abstraction
(Level 5). 1. Data Filtering

Unlike cloud computing, which processes data at Removes unwanted, duplicate, or irrelevant sensor
remote servers, Edge (Fog) Computing performs data.
computation close to the data source. This
reduces communication delay, decreases 2. Data Cleaning
bandwidth usage, improves response time, and
enables real-time decision making. Corrects missing or inconsistent values to improve
data quality.
The primary objective of this layer is to convert
network data flows into meaningful information 3. Data Aggregation
suitable for storage, analysis, and business
applications. Combines multiple sensor readings into
summarized information.

4. Packet Content Inspection

Examines the contents of incoming data packets

before forwarding them.

5. Network-Level and Data-Level Analysis

Performs analysis on both communication traffic

and sensor data.

6. Threshold Detection

Checks whether sensor values exceed predefined

Need for Edge (Fog) Computing
threshold limits.
IoT devices continuously generate huge amounts of
7. Event Generation
sensor data. Sending every data packet directly to
the cloud results in:
Generates alerts or notifications when abnormal
conditions occur.
 High network traffic
 Increased latency
8. Peer-to-Peer Communication
 Higher bandwidth consumption
 Slow response time
Allows nearby edge devices to exchange
 Increased cloud processing load
information directly without depending on the
cloud.

Data Transformation
One of the major responsibilities of Level 3 is Example:
transforming raw sensor data into meaningful Temperature = 24°C → Normal → Ignore
information. Temperature = 75°C → Abnormal → Forward

The transformation process includes: 2. Formatting

 Data Filtering Converts sensor data into a standard format

 Data Cleaning suitable for higher-level processing.
 Data Aggregation
 Packet Inspection 3. Expansion (Decoding)
 Data Analysis
 Event Detection Adds additional information such as device ID,
location, timestamp, and sensor type.
After transformation, the information becomes
understandable by higher IoT layers. 4. Distillation (Reduction)

Reduces unnecessary or repeated data.

Instead of sending 100 temperature readings every
minute,
Edge Device sends
This reduces network traffic.

5. Assessment

Checks whether sensor values exceed predefined

thresholds. If necessary, the data is redirected to
emergency systems or cloud servers.

Advantages of Edge (Fog) Computing

 Reduces communication latency

(The figure shows Level 3 positioned above  Faster decision making
Connectivity and Physical Devices. It illustrates  Lower bandwidth consumption
data filtering, cleanup, aggregation, packet  Reduces cloud workload
inspection, threshold detection, and conversion of  Improves reliability
raw data into meaningful information.)  Supports real-time applications
 Enhances privacy and security
Data Processing Activities  Enables local processing during Internet
failures
Since data arrives continuously from sensors, the
Edge Computing layer processes information Applications of Edge (Fog) Computing
packet-by-packet instead of waiting for complete
transactions. Edge Computing is widely used in:

The major processing activities are:  Smart Cities

 Smart Homes
1. Evaluation  Industrial IoT
 Autonomous Vehicles
Determines whether incoming data should be  Healthcare Monitoring
processed further or discarded.  Video Surveillance
 Agriculture
 Smart Grid
  Traffic Monitoring
 Manufacturing

Advantages over Cloud Computing

Edge (Fog)
Cloud Computing
Computing
Processes data near Processes data in remote
devices data centers
Very low latency Higher latency
Less bandwidth usage Higher bandwidth usage
Faster response Slower response
Suitable for real-time Suitable for large-scale
systems analytics
Level 4: Data Accumulation (Storage) 2. Event Comparison

Introduction The stored data is compared with previous records

to identify changes, trends, or abnormal behavior.
The Data Accumulation Layer (Level 4) is the This comparison helps applications detect
fourth layer of the IoT Reference Model. Its significant events.
primary responsibility is to store the processed
data received from the Edge (Fog) Computing 3. Event Joining for Complex Event Processing
Layer (Level 3). At this stage, the continuous (CEP)
stream of sensor data, known as data in motion, is
converted into data at rest, making it suitable for Information received from multiple sensors or
permanent storage and later analysis. devices is combined to produce meaningful events.
Complex Event Processing (CEP) enables the
Unlike the lower layers that focus on collecting and system to identify relationships between different
transmitting data in real time, Level 4 concentrates events occurring simultaneously.
on storing information in an organized manner so
that applications and users can access it whenever 4. Event-Based Rule Evaluation
required. The layer acts as a bridge between real-
time IoT operations and long-term business The stored data is evaluated according to
analytics. predefined business rules. When certain conditions
are satisfied, appropriate actions or notifications
Purpose of the Data Accumulation Layer are generated automatically.

IoT devices continuously generate enormous 5. Event Aggregation

volumes of data. Processing all this data
immediately is neither practical nor necessary. Multiple sensor readings are grouped together to
Therefore, the Data Accumulation layer create summarized information. Instead of storing
temporarily or permanently stores the processed every individual reading, the system stores
information so that it can be retrieved efficiently in meaningful summaries such as averages, totals,
the future. minimum, or maximum values.

This layer performs the transition from event- 6. Event Persistence

based processing to query-based processing,
allowing applications to retrieve historical data The processed data is stored permanently in
whenever required instead of processing data databases, file systems, or cloud storage. Persistent
continuously. storage ensures that historical information remains
available for future analysis and reporting.
Major Functions of Level 4
Converting Data in Motion into Data at Rest
The Data Accumulation layer performs several
important storage-related operations. One of the most important responsibilities of the
Data Accumulation layer is converting
1. Event Sampling and Filtering continuously flowing network data into stored
information.
Incoming data is sampled and filtered before
storage. Unnecessary or duplicate information is This conversion involves:
removed, reducing storage requirements and
improving database efficiency.  Receiving processed data from Level 3.
  Organizing the information into database Level 4 receives processed information from Level
structures. 3 and stores it for use by higher application layers.
 Storing the information permanently.
 Making the stored information available for Major Decisions Taken at Level 4
applications.
The Data Accumulation layer determines several
After storage, applications no longer process important aspects before storing information.
continuous data streams but instead retrieve
information through database queries. 1. Is the Data Useful?

Making Network Data Usable The layer first decides whether the incoming data
is important enough to be stored. Only valuable
The Data Accumulation layer performs several information is forwarded for permanent storage.
transformations that make network data useful for
higher-level applications. 2. Should the Data be Persisted?

Conversion of Data in Motion The system decides whether data should be:

Real-time streaming data is converted into stored  Stored permanently on disk,

records.  Stored temporarily in memory, or
 Discarded after processing.
Conversion of Network Packets
3. Selection of Storage Technology
Network packets are transformed into database
tables or structured records suitable for relational Depending on the application, different storage
or NoSQL databases. systems may be selected, including:

Transition from Event-Based to Query-Based  Relational Database Management Systems

Processing (RDBMS)
 NoSQL Databases
Instead of responding immediately to every sensor  Big Data Storage Systems
event, applications can retrieve stored information  Distributed File Systems
whenever required using database queries.  Cloud Storage

Selective Storage 4. Organization of Data

Only useful information is retained after filtering, The stored information must be organized
significantly reducing storage space and improving efficiently to support fast retrieval, indexing,
query performance. searching, and reporting.

Relationship with Other IoT Layers Recombination and Recomputation of Data

The position of Level 4 within the IoT Reference Sometimes stored data must be combined with
Model is illustrated below. information collected earlier or from external
systems.
Level 4 : Data Accumulation (Storage)
▲ The Data Accumulation layer supports:
Level 3 : Edge (Fog) Computing
▲  Combining current and historical data.
Level 2 : Connectivity  Integrating IoT and non-IoT information.
▲  Recomputing values when required.
Level 1 : Physical Devices and Controllers
  Generating new datasets for higher-level  Smart City Data Repositories
applications.  Healthcare Record Management
 Industrial Production Monitoring
Advantages of Data Accumulation  Smart Agriculture Data Storage
 Environmental Monitoring Systems
The Data Accumulation layer provides several  Energy Consumption Analysis
important benefits.  Smart Grid Monitoring
 Transportation and Traffic Analytics
 Converts continuous sensor streams into  Predictive Maintenance Systems
permanent records.  Cloud-Based IoT Platforms
 Supports historical data analysis.
 Reduces processing load on applications. Key Features of Level 4
 Enables efficient querying of stored
information.  Converts Data in Motion → Data at Rest.
 Reduces storage requirements through  Stores processed IoT data permanently.
filtering and aggregation.  Supports event sampling, comparison,
 Supports business intelligence and aggregation, and persistence.
analytics.  Converts network packets into structured
 Provides reliable long-term storage. database records.
 Improves application performance.  Enables query-based access to historical
information.
Applications of the Data Accumulation Layer  Acts as the foundation for higher-level data
abstraction and analytics.
The Data Accumulation layer is widely used in
various IoT applications, including:
Level 5: Data Abstraction (Aggregation 2. Distributed Data Storage
and Access)
Devices may be located in different geographical
Introduction regions, and the collected data is often stored locally
or in multiple cloud platforms.
The Data Abstraction Layer (Level 5) is the fifth
layer of the IoT Reference Model. After data has 3. Separation of Data Processing
been collected, processed, and stored in the previous
layers, this layer is responsible for integrating, Processing and storage operations are separated to
organizing, and presenting data in a form that improve performance. Real-time data generation
applications can easily understand and use. should not be delayed by database operations.

Large-scale IoT systems generate data from 4. Different Storage Technologies

millions of devices distributed across different
locations. This data may be stored in multiple IoT systems may simultaneously use:
databases, cloud platforms, enterprise systems, or
big data repositories. The Data Abstraction layer  Relational Databases (RDBMS)
hides these storage complexities by providing a  NoSQL Databases
unified data interface for applications.  Hadoop Distributed File System (HDFS)
 Cloud Storage
Instead of accessing multiple databases separately,  Data Warehouses
applications communicate with the Data
Abstraction layer, which gathers, transforms, and The abstraction layer hides these differences from
delivers the required information in a consistent applications.
format.
5. Different Processing Requirements
Purpose of the Data Abstraction Layer
Different applications require different types of
The main objective of Level 5 is to convert stored processing. For example:
IoT data into meaningful information that can be
used by business applications, dashboards, analytics  A smart factory dashboard needs live
platforms, and enterprise software. machine status.
 A management system requires daily
This layer performs aggregation, integration, production summaries.
normalization, transformation, and access  Predictive maintenance software requires
management, allowing applications to retrieve data historical sensor data.
without knowing where or how it is stored.
The abstraction layer prepares data according to the
Need for Data Abstraction needs of each application.

Modern IoT systems operate at enterprise or global Major Functions of Level 5

scales and produce enormous volumes of data.
Several challenges make the Data Abstraction layer The Data Abstraction layer performs several
essential. important functions.

1. Large Volume of Data 1. Data Aggregation

IoT devices continuously generate massive amounts Data collected from multiple IoT devices is
of information that cannot always be stored in a combined into a single logical view. Instead of
single database. retrieving information from each device
individually, applications receive integrated
information.
2. Data Integration Creating Schemas and Views

The layer combines information from multiple The abstraction layer creates database views and
sources such as: logical schemas that present information exactly as
required by applications.
 IoT Sensors
 Enterprise Resource Planning (ERP) Combining Multiple Data Sources
 Customer Relationship Management (CRM)
 Human Resource Management Systems Data collected from multiple databases and cloud
(HRMS) services is integrated into a single dataset.
 Cloud Databases
 External APIs Filtering and Projection

This creates a unified view of organizational data. Only useful attributes are selected while
unnecessary information is removed.
3. Data Transformation
Reconciliation of Data Differences
Different devices generate data in different formats.
The abstraction layer converts this information into The layer resolves differences in:
standardized formats that applications can easily
process.  Data format
 Data structure
4. Data Filtering and Selection  Data semantics
 Communication protocols
Only the required information is delivered to  Security policies
applications. Unnecessary data is filtered out,
reducing processing time and network traffic. Data Reformatting

5. Data Reformatting The integrated information is reformatted into a

consistent structure before being delivered to
The stored information is reorganized into applications.
application-specific structures such as:
Data Processing Activities
 Tables
 JSON The Data Abstraction layer performs several
 XML processing activities to improve application
 CSV performance.
 API responses
Data Normalization
6. Unified Data Access
Removes redundancy and organizes data efficiently.
Applications communicate with a single interface
instead of multiple databases, making application Data Denormalization
development simpler.
Creates optimized structures that improve query
Information Integration performance.

One of the major responsibilities of Level 5 is Data Indexing

Information Integration.
Indexes are created to enable faster searching and
This includes: retrieval.
Data Consolidation  Consolidating information into centralized
repositories.
Information from multiple databases is combined  Providing secure access through
using: authentication and authorization.
 Normalizing, denormalizing, and indexing
 ETL (Extract, Transform, Load) data for faster access.
 ELT (Extract, Load, Transform)
 Data Replication Advantages of the Data Abstraction Layer
 Data Virtualization
The Data Abstraction layer provides several
Authentication and Authorization benefits:

Access control mechanisms ensure that only  Simplifies application development.

authorized users and applications can access IoT  Hides storage complexity from users.
data.  Integrates multiple data sources.
 Provides consistent and standardized data.
Responsibilities of the Data Abstraction Layer  Supports enterprise-level scalability.
 Improves database performance.
The major responsibilities include:  Enables secure access to IoT information.
 Supports business intelligence and analytics.
 Reconciling multiple data formats from  Reduces application processing time.
different sources.  Makes IoT data easier to understand and use.
 Ensuring consistent semantics across all
datasets.
 Confirming that complete data is available
to higher-level applications.
Level 6: Application

Introduction  Provides graphical dashboards and reports.

 Enables monitoring and remote control.
Level 6 is the Application Layer of the IoT  Performs business intelligence and
Reference Model. It is the highest layer where the predictive analytics.
processed and integrated IoT data is finally used to  Offers user-friendly interfaces.
provide meaningful services to end users. At this  Supports mobile, desktop, and cloud
stage, the information generated by the lower layers applications.
is interpreted and converted into useful business
insights, monitoring dashboards, control systems, Types of IoT Applications
and decision-making applications.
The IoT Reference Model does not define a single
Unlike the lower layers that deal with sensing, application type. Instead, applications vary
communication, storage, and data processing, Level according to business requirements and industry
6 focuses entirely on how users interact with IoT needs.
systems. Since the required data has already been
accumulated (Level 4) and abstracted (Level 5), the Some applications focus on:
application layer does not need to process data at
network speed. Instead, it works with data at rest,  Device monitoring
enabling faster queries, visualization, reporting, and  Device control
intelligent decision making.  Data visualization
 Business analytics
Objectives of Level 6  Predictive maintenance
 Enterprise management
The Application Layer is responsible for:  Automation

 Presenting IoT information to users in an Examples of IoT Applications

understandable format.
 Supporting monitoring and control of IoT 1. Mission-Critical Business Applications
devices.
 Providing business intelligence and These applications support important business
analytics. operations where reliability is essential.
 Enabling enterprise decision making.
 Integrating IoT information with business Examples include:
software such as ERP and CRM.
 Delivering value-added services to  Enterprise Resource Planning (ERP)
industries, organizations, and consumers.  Supply Chain Management
 Manufacturing Execution Systems (MES)
Since the data is already filtered, integrated,  Hospital Information Systems
secured, and organized, applications can directly
access it for visualization, analytics, automation, or 2. Mobile Applications
control.
Mobile applications provide users with convenient
Characteristics of Level 6 access to IoT devices.

The Application Layer has the following Examples:

characteristics:
 Smart home control apps
 Works on processed data instead of raw  Smart lighting apps
sensor data.  Smart parking applications
 Supports multiple application domains.  Smart agriculture monitoring
Users can monitor sensors and control devices The Application Layer provides numerous
directly from smartphones. advantages:

3. Business Intelligence (BI) Applications  Converts processed data into useful

information.
Business Intelligence applications analyze stored  Simplifies decision making.
IoT data and convert it into meaningful reports.  Improves operational efficiency.
 Enables remote monitoring and control.
Features include:  Supports business intelligence.
 Enhances customer experience.
 Dashboards  Reduces operational costs.
 Charts  Enables predictive maintenance.
 KPI monitoring  Increases productivity.
 Trend analysis  Improves resource utilization.
 Performance reports
Real-Time Applications of Level 6
These applications help organizations make data-
driven decisions. Level 6 is widely used in various industries.

4. Analytical Applications Healthcare

Analytical applications use historical and real-time  Remote patient monitoring

data to identify patterns and predict future events.  Smart hospitals
 Medical diagnostics
Applications include:
Smart Home
 Predictive maintenance
 Demand forecasting  Lighting automation
 Energy optimization  Smart security
 Traffic analysis  Energy management
 Environmental monitoring
Smart Cities
These applications improve operational efficiency.
 Intelligent traffic systems
 Waste management
 Smart parking
5. System Management and Control  Air quality monitoring
Applications
Agriculture
These applications monitor and control the IoT
infrastructure itself.  Smart irrigation
 Crop monitoring
Functions include:  Livestock management

 Device configuration Manufacturing

 Firmware updates
 Fault detection  Predictive maintenance
 Remote diagnostics  Industrial automation
 Performance monitoring  Quality inspection
 Network management
Retail
Benefits of Level 6
  Inventory management Advantages
 Customer analytics
 Smart billing  Easy user interaction
 Supports intelligent decision making
Transportation  Provides graphical dashboards
 Enables remote device management
 Fleet monitoring  Integrates with enterprise applications
 Vehicle tracking  Improves business productivity
 Intelligent traffic control  Supports AI and Machine Learning
applications
Role of Level 6 in the IoT Reference Model  Enhances automation

Level 6 acts as the interface between IoT systems Limitations

and users. It transforms processed information into
actionable insights through dashboards, reports,  Depends on lower layers for accurate data.
alerts, and automation systems. It enables  Requires secure authentication and
businesses and individuals to monitor devices, authorization.
analyze data, make informed decisions, and control  Large-scale deployments require powerful
IoT environments effectively. application servers.
 Complex applications increase software
maintenance costs

Applications of Level 5  Supply Chain Management

 Smart Grid Monitoring
The Data Abstraction layer is widely used in:  Environmental Monitoring
 Business Intelligence Dashboards
 Smart City Management Systems  Cloud-based IoT Services
 Industrial IoT Platforms  Predictive Analytics Systems
 Healthcare Information Systems
 Smart Agriculture Platforms
Level 7: Collaboration and
Processes
IoT Security
Definition
Definition
IoT Security protects IoT devices, communication
Level 7 is the highest layer of the IoT Reference networks, applications, and data from cyber
Model. It focuses on connecting IoT information threats. It ensures secure operation across all layers
with people, business processes, and enterprise of the IoT architecture.
applications to support intelligent decision-
making. Objectives

Key Functions  Protect IoT devices from unauthorized

access.
 Collaboration: Enables multiple users to  Secure data during transmission and
share and use IoT information. storage.
 Business Process Automation: Triggers  Maintain confidentiality, integrity, and
actions and workflows based on IoT events. availability of data.
 Enterprise Integration: Connects IoT  Prevent cyber attacks and data breaches.
systems with ERP, CRM, and cloud
platforms. Security Across IoT Layers
 Decision Support: Provides reports,
dashboards, and alerts for better decisions.  Level 1: Secures physical devices and
hardware.
Applications  Level 2: Protects network communication.
 Level 3: Encrypts data and secures edge
Used in smart healthcare, smart manufacturing, communication.
smart cities, agriculture, logistics, and industrial  Levels 4 & 5: Secures stored data and
automation. access to databases.
 Levels 6 & 7: Provides authentication,
Advantages authorization, and identity management.

 Improves business efficiency. Common Security Mechanisms

 Enables better collaboration.
 Supports intelligent decision-making. Encryption, authentication, authorization, firewalls,
 Reduces manual effort through automation. secure boot, VPN, and Multi-Factor Authentication
(MFA) are commonly used to secure IoT systems.
Limitations
Benefits
 Complex to integrate with enterprise
systems.  Protects sensitive information.
 Depends on lower IoT layers.  Prevents unauthorized access.
 Requires strong security and reliable  Improves system reliability.
infrastructure.  Ensures user privacy and trust.