UNIT-IV

IoT ARCHITECTURE AND PROTOCOLS

Definition & Characteristics of IoT - Challenges and Issues - Physical Design of IoT - Logical

Design of IoT - IoT Enabling Technologies – Domain Specific IoTs – IoT and M2M - IoT

Communication Models and APIs – IoT Protocols – LoRaWAN, 6LoWPAN, CoAP, MQTT.

IOT
The Internet of Things refers to the rapidly growing network of connected objects that can collect
and exchange data in real-time using embedded sensors.
Cars, lights, refrigerators, and appliances can all be connected to the IoT.

The Internet of things is a connecting bridge between the physical world and the cyber world and
Machine to Machine communication
Any physical object can be transformed into an IoT device if it can be connected to the internet to be
controlled or communicate information. A lightbulb that can be switched on using a smartphone app
is an IoT device, as is a motion sensor

What is a Sensor?
A Sensor is a device that is able to detect changes in an environment. Sensors are hardware components
that can detect events or changes in their surroundings, and then provide a corresponding output. Sensors
are the eyes and ears of any IoT Project.

What is Actuator?
An Actuator is a device that converts energy into motion. It is usually used to apply force on something.

Advantages & Disadvantages of IoT

Characteristics of the Internet of Things:
1. Connectivity — Connectivity is an important requirement of the IoT infrastructure. Things of
IoT should be connected to the IoT infrastructure. Anyone, anywhere, anytime can connect,
this should be guaranteed at all times. For example, the connection between people through
internet devices like mobile phones, and other gadgets, also a connection between Internet
devices such as routers, gateways, sensors, etc.

2. Intelligence and Identity — The extraction of knowledge from the generated data is very
important. For example, a sensor generates data, but that data will only be useful if it is
interpreted properly. Each IoT device has a unique identity. This identification is helpful in
tracking the equipment and at times for querying its status.

3. Scalability — The number of elements connected to the IoT zone is increasing day by day.
Hence, an IoT setup should be capable of handling the massive expansion. The data
generated as an outcome is enormous, and it should be handled appropriately.

4. Dynamic and Self-Adapting (Complexity) –IoT devices should dynamically adapt

themselves to changing contexts and scenarios. Assume a camera meant for surveillance. It
should be adaptable to work in different conditions and different light situations (morning,
afternoon, and night).

5. Architecture — IoT architecture cannot be homogeneous in nature. It should be hybrid,

supporting different manufacturers ‘ products to function in the IoT network. IoT is not
owned by anyone engineering branch. IoT is a reality when multiple domains come together.

6. Safety — There is a danger of the sensitive personal details of the users getting compromised
when all his/her devices are connected to the internet. This can cause a loss to the user.
Hence, data security is the major challenge. Besides, the equipment involved is huge. IoT
networks may also be at the risk. Therefore, equipment safety is also critical.

Challenges of IoT
1. Data Integration and Management
One of the biggest challenges to IoT analytics success is data integration & management. IoT
devices generate data in various formats & protocols, making it difficult to integrate & manage the
data. It is difficult to adequately interpret, store & manage the data created by IoT devices because
of its sheer amount and diversity. To guarantee that the data is precise, consistent, & secure,
organizations must have strong data integration and management systems in place.

2. Data Quality
Data quality is another significant challenge to IoT analytics success. The data generated by IoT
devices is often incomplete, inaccurate, or inconsistent. It is because IoT devices may malfunction
or experience network connectivity issues, leading to data loss or corruption. Furthermore, IoT
devices may generate data that is irrelevant to the analytics process, leading to noise in the data. To
guarantee that the data used for analytics is correct, consistent, and relevant, organizations need to
invest in data cleansing & validation processes.

3. Security and Privacy

IoT devices produce many useful yet sensitive data for an organization. The data may contain much
in-depth information also about the company. But as it is always connected to the internet there is a
huge issue of security. To guarantee that the data is secure and secret, organizations must implement
strong security & privacy measures.

4. Scalability
IoT analytics involves processing and analyzing massive amounts of data in real-time. It takes a
large amount of processing and storage power. Organizations may find it challenging to meet the
demand for computational power and storage space as they increase their IoT implementations.
Organizations need to have scalable analytics platforms in place to handle the growing volume of
data generated by IoT devices.

5. Interoperability
It means the ability of various devices communicates with one another. Interoperability is crucial in
the context of IoT analytics because IoT devices may produce data in different forms and protocols.
To guarantee that the data is integrated & analyzed properly, organizations must have systems in
place that can handle various data types and protocols.

6. Lack of Skilled Workforce

IoT analytics requires a skilled workforce that can analyze and derive insights from the massive
amounts of data generated by IoT devices. But there is a lack of knowledgeable data scientists and
analysts who can work with IoT data. Organizations need to invest in training and development
programs to build a skilled workforce that can handle IoT analytics effectively.

7. Data Governance
Data governance refers to the management & control of data assets in an organization. Data
governance is crucial in the context of IoT analytics to make sure that the data is appropriately
managed and used by legal standards. To ensure that the data created by IoT devices is managed
successfully, organizations must have strong data governance frameworks in place.

8. Real-time Analytics
IoT devices generate data in real time. Real-time data analysis is necessary for every organization
that wants to make educated judgments. Real-time analytics can be challenging as organizations
need to process & analyze data in real-time to derive insights. All Organizations must have real-
time analytics platforms in place. This real-time analytics platform can handle the volume & variety
of data IoT devices generate.

9. Cost Implementing
IoT analytics can be expensive, as it involves investing in hardware, software & skilled manpower.
Organizations need to consider the cost-benefit analysis before implementing IoT analytics. They
need to weigh the benefits of IoT analytics against the cost of implementation to ensure that it
makes business sense to invest in IoT analytics.

Physical Design of IoT

A physical design of an IoT system refers to the individual node devices and their protocols that are

utilized to create a functional IoT ecosystem. Each node device can perform tasks such as remote

sensing, actuating, monitoring, etc., by relying on physically connected devices. It may also be

capable of transmitting information through different types of wireless or wired connections.

The things/devices in the IoT system are used for:

 Building connections

 Data processing

 Providing storage

 Providing interfaces

 Providing graphical interfaces

The devices generate data, and the data is used to perform analysis and do operations for improving

the system. For instance, a moisture sensor is used to obtain the moisture data from a location, and the

system analyses it to give an output.

Things/Devices
Things/Devices are used to build a connection, process data, provide interfaces, provide storage, and

provide graphics interfaces in an IoT system. all these generate data in a form that can be analyzed by

an analytical system and program to perform operations and used to improve the system.

for example temperature sensor that is used to analyze the temperature generates the data from a

location and is then determined by algorithms.

Connectivity: Devices like USB hosts and ETHERNET are used for connectivity between the

devices and the server.

Processor: A processor like a CPU and other units are used to process the data. these data are further

used to improve the decision quality of an IoT system.

Audio/Video Interfaces: An interface like HDMI and RCA devices is used to record audio and

videos in a system.

Input/Output interface: To give input and output signals to sensors, and actuators we use things like

UART, SPI, CAN, etc.

Storage Interfaces: Things like SD, MMC, and SDIO are used to store the data generated from an

IoT device.

Other things like DDR and GPU are used to control the activity of an IoT system.
Logical Design of IoT
A logical design for an IoT system is the actual design of how its components (computers, sensors,

and actuators) should be arranged to complete a particular function. It doesn’t go into the depth of

describing how each component will be built with low-level programming specifics.

IoT logical design includes:

1. IoT functional blocks

2. IoT communications models

3. IoT communication APIs

IoT functional blocks

IoT systems include several functional blocks such as Devices, communication, security, services,
and application.

The functional blocks provide sensing, identification, actuation, management, and communication
capability. These functional blocks consist of devices that handle the communication between the
server and the host, enable monitoring control functions, manage the data transfer, secure the IoT
system using authentication and different functions, and provide an interface for controlling and
monitoring various terms.

The Functional blocks are:

Device: An IoT system comprises of devices that provide sensing, actuation, monitoring, and control
functions.
Communication: Handles the communication for the IoT system.
Services: services for device monitoring, device control service, data publishing services, and
services for device discovery.
Management: this block provides various functions to govern the IoT system.
Security: This block secures the IoT system and by providing functions such as authentication,
authorization, message and content integrity, and data security.
Application: This is an interface that the users can use to control and monitor various aspects of the
IoT system. The application also allows users to view the system status and view or analyze the
processed data.
IoT Communication Models
There are multiple kinds of models available in an Internet of Things system that is used for

communicating between the system and server, such as:

Request-Response Model
Request-response model is a communication model in which the client sends requests to the server

and the server responds to the requests. When the server receives a request, it decides how to respond,

fetches the data, retrieves resource representation, prepares the response, and then sends the response

to the client. Request-response is a stateless communication model and each request-response pair is

independent of the others.

HTTP works as a request-response protocol between a client and a server. A web browser may be the

client, and an application on a computer that hosts a website may be the server.
Example: A client (browser) submits an HTTP request to the server; then the server returns a

response to the client. The response contains status information about the request and may also

contain the requested content.

 Publisher-Subscriber Model — This model comprises three entities: Publishers, Brokers,

and Consumers.

Publishers are the source of data. It sends the data to the topic which is managed by the broker. They
are not aware of consumers.

Consumers subscribe to the topics which are managed by the broker.

Brokers' responsibility is to accept data from publishers and send it to the appropriate consumers.
The broker only has the information regarding the consumer to which a particular topic belongs
which the publisher is unaware.
  Push-Pull Model — The push-pull model constitutes data publishers, data consumers, and

data queues.

Publishers and Consumers are not aware of each other.

Publishers publish the message/data and push it into the queue. The consumers, present on the other

side, pull the data out of the queue. Thus, the queue acts as the buffer for the message when the

difference occurs in the rate of push or pull of data on the side of a publisher and consumer.

Queues help in decoupling the messaging between the producer and consumer. Queues also act as a

buffer which helps in situations where there is a mismatch between the rate at which the producers

push the data and consumers pull the data.

Exclusive Pair –
Exclusive Pair is the bi-directional model, including full-duplex communication between client and

server. The connection is constant and remains open till the client sends a request to close the

connection.

The Server has the record of all the connections which has been opened.

This is a state-full connection model and the server is aware of all open connections.

WebSocket-based communication API is fully based on this model.

IoT CommunicationAPIs:

a) REST based communication APIs (Request-Response BasedModel)

b) WebSocket based Communication APIs (Exclusive PairBasedModel)

a) REST based communication APIs: Representational State Transfer(REST) is a set of

architectural principles by which we can design web services and web APIs that focus on a
system‘s resources and have resource states are addressed andtransferred.
The REST architectural constraints: Fig. shows communication between client server with
REST APIs.

Client-Server: The principle behind client-server constraint is the separation of concerns.

Separation allows client and server to be independently developed and updated.

Stateless: Each request from client to server must contain all the info. Necessary to understand the
request, and cannot take advantage of any stored context on the server.

Cache-able: Cache constraint requires that the data within a response to a request be implicitly or
explicitly labeled as cache-able or non-cacheable. If a response is cache-able, then a client cache is
given the right to reuse that response data for later, equivalentrequests.

Layered System: constraints the behavior of components such that each component cannot see
beyond the immediate layer with which they are interacting.

User Interface: constraint requires that the method of communication between a client and a
server must be uniform.

Code on Demand: Servers can provide executable code or scripts for clients to execute in their
 context. This constraint is the only one that is optional.

Request-Response model used by REST:

RESTful webservice is a collection of resources which are represented by URIs. RESTful web API
has a base URI(e.g: http://example.com/api/tasks/). The clients and requests to these URIs using the
methods defined by the HTTP protocol(e.g: GET, PUT, POST or DELETE). A RESTful web service
can support various internet media types.
b) WebSocket Based Communication APIs: WebSocket APIs allow bi-directional,
full duplex communication between clients and servers. WebSocket APIs follow
the exclusive pair communicationmodel.
Difference between Rest API and Web Socket API :

Difference Between Physical and Logical Design of IoT

IoT Enabling Technologies
IoT-enabling technologies primarily focus on converting a standalone device into an IoT device by
giving it the additional possibility of connecting to the internet and exchanging information with it.

1) Wireless Sensor Network(WSN): Comprises of distributed devices with sensors which

are used to monitor the environmental and physical conditions. Zig Bee is one of the
most popular wireless technologies used byWSNs.

WSNs used in IoT systems are described as follows:

 Weather Monitoring System: in which nodes collect temp, humidity and other
data, which is aggregated and analyzed.
 Indoor air quality monitoring systems: to collect data on the indoor air quality and
concentration of various gases.
 Soil Moisture Monitoring Systems: to monitor soil moisture at variouslocations.
 Surveillance Systems: use WSNs for collecting surveillance data(motiondata
detection).
 Smart Grids : use WSNs for monitoring grids at variouspoints.

 Structural Health Monitoring Systems: Use WSNs to monitor the health of

structures(building, bridges) by collecting vibrations from sensor nodes deployed at
various points in thestructure.

2) Cloud Computing: Services are offered to users in differentforms.

 Infrastructure-as-a-service(IaaS):provides users the ability to provision computing
and storage resources. These resources are provided to the users as a virtual
machine instances and virtualstorage.
 Platform-as-a-Service(PaaS): provides users the ability to develop and deploy
application in cloud using the development tools, APIs, software libraries and
services provided by the cloud serviceprovider.
 Software-as-a-Service(SaaS): provides the user a complete software application or
the user interface to the applicationitself.

3) Big Data Analytics: Some examples of big data generated by IoTare

 Sensor data generated by IoTsystems.
 Machine sensor data collected from sensors established in industrial and energy
systems.
  Health and fitness data generated IoTdevices.
 Data generated by IoT systems for location and trackingvehicles.
 Data generated by retail inventory monitoringsystems.

4) Communication Protocols: form the back-bone of IoT systems and enable network
connectivity and coupling toapplications.
 Allow devices to exchange data overnetwork.
 Define the exchange formats, data encoding addressing schemes for device and
routing of packets from source todestination.
 It includes sequence control, flow control and retransmission of lostpackets.

5) Embedded Systems: is a computer system that has computer hardware and software
embedded to perform specific tasks. Embedded System range from low cost miniaturized devices
such as digital watches to devices such as digital cameras, POS terminals, vending machines,
appliancesetc.,

MQTT

Message Queuing Telemetry Transport (MQTT): The message query telemetry transport
protocol is a communication-based protocol that is used for IoT devices. This protocol is based on
the publish-subscribe methodology in which clients receive the information through a broker only
to the subscribed topic. A broker is a mediator who categorizes messages into labels before being
delivered.

Characteristics of the Protocol

 Light-weight and reliable: The MQTT message is compact, which can realize stable
transmission on severely limited hardware equipment and network with low bandwidth and
high delay.

 Publish/subscribe mode: Based on the publish/subscribe mode, the advantage of publishing

and subscribing mode is that the publisher and subscriber are decoupled: Subscribers and
publishers do not need to establish a direct connection or be online at the same time.

 Created for the IoT: It provides comprehensive IoT application features such as heartbeat
mechanism, testament message, QoS quality level
  Better ecosystem: It covers all-language platform's clients and SDKs, and it has mature Broker
server software, which can support massive Topic and ten-million-level device access and
provide rich enterprise integration capabilities.

CoAP
Constrained Application Protocol (COAP): The constrained application protocol is a client
server-based protocol. With this protocol, the COAP packet can be shared between different client
nodes which are commanded by the COAP server. The server is responsible to share the
information depending on its logic but has not acknowledged it. This is used with the applications
which support the state transfer model.

Characteristics of the Protocol

CoAP refers to many design ideas of HTTP, and it also improves many design details and adds many
practical functions according to the specific situation of limited resource-limited devices.

 It is based on message model

 Based on UDP Protocol, transport layer supports restricted devices

 It uses request/response model similar to HTTP request, and HTTP is text format, while CoAP
is binary format, which is more compact than HTTP
 It supports two-way communication
 It has the characteristics of light-weight and low power consumption
 It supports reliable transmission, data re-transmission, and block transmission to ensure reliable
arrival of data

 It supports IP multicast
 It supports observation mode
 It supports asynchronous communication

LoRaWAN
LoRaWAN refers to Long Rage Wide Area Network which is a wide area network protocol. It is
an optimized low-power consumption protocol design to support large-scale public networks with
millions of low-power devices. A single operator operates the LoRaWAN. The LoRaWAN network
is a bi-directional communication for IoT application with low cost, mobility, and security..
An end device can connect to a network with LoRaWAN in two ways:

 Over-the-air Activation (OTAA): A device has to establish a network key and an application session
key to connect with the network.
 Activation by Personalization (ABP): A device is hardcoded with keys needed to communicate with
the network, making for a less secure but easier connection.

Properties of LoRaWAN protocol

o Standard: LoRaWAN
o Frequency: Various
o Range: 2-5km (urban environment), 15km (suburban environment)
o Data Rates: 0.3-50 kbps.

6LoWPAN
The 6LoWPAN protocol refers to IPv6 Low Power Personal Area Network which uses a
lightweight IP-based communication to travel over low data rate networks. It has limited processing
ability to transfer information wirelessly using an internet protocol. So, it is mainly used for home
and building automation. The 6LoWPAN protocol operates only within the 2.4 GHz frequency range
with 250 kbps transfer rate. It has a maximum length of 128-bit header packets.

6LowPAN Security Measure

Security is a major issue for 6LowPAN communication Protocol. There are several attacks issues at
the security level of 6LoWPAN which aim is to direct destruction of the network. Since it is the
combination of two systems, so, there is a possibility of attack from two sides that targets all the
layer of the 6LoWPAN stack (Physical layer, Data link layer, Adaptation layer, Network layer,
Transport layer, Application layer).

Basic Requirements of 6LoWPAN:

1. The device should be having sleep mode in order to support the battery saving.
2. Minimal memory requirement.
3. Routing overhead should be lowered.
Features of 6LoWPAN:
1. It is used with IEEE 802.15,.4 in the 2.4 GHz band.
2. Outdoor range: ~200 m (maximum)
3. Data rate: 200kbps (maximum)
4. Maximum number of nodes: ~100
Advantages of 6LoWPAN:
1. 6LoWPAN is a mesh network that is robust, scalable, and can heal on its own.
2. It delivers low-cost and secure communication in IoT devices.
3. It uses IPv6 protocol and so it can be directly routed to cloud platforms.
4. It offers one-to-many and many-to-one routing.
5. In the network, leaf nodes can be in sleep mode for a longer duration of time.
Disadvantages of 6LoWPAN:
1. It is comparatively less secure than Zigbee.
2. It has lesser immunity to interference than that Wi-Fi and Bluetooth.
3. Without the mesh topology, it supports a short range.
Applications of 6LoWPAN:
1. It is a wireless sensor network.
2. It is used in home-automation,
3. It is used in smart agricultural techniques, and industrial monitoring.
4. It is utilised to make IPv6 packet transmission on networks with constrained power and
reliability resources possible.

Difference between COAP and MQTT protocols:

Basis of COAP MQTT

Message Queuing Telemetry

Constrained Application Protocol
Abbreviation Transport

Communication
It uses Request-Response model. It uses Publish-Subscribe model
Type
 Basis of COAP MQTT

This uses both Asynchronous and

This uses only Asynchronous
Messaging Mode Synchronous.

Transport layer This mainly uses User Datagram This mainly uses Transmission
protocol protocol(UDP) Control protocol(TCP)

It has 4 bytes sized header It has 2 bytes sized header

Header size

No it does not uses REST

Yes it uses REST principles
RESTful based principles

It supports and best used for live

It does not has such support
Persistence support data communication

It provides by adding labels to the

It has no such feature.
Message Labelling messages.

It is used in Utility area networks and It is used in IoT applications and is

Usability/Security has secured mechanism. secure

Effectiveness in LNN is excellent. Effectiveness in LNN is low.

Effectiveness

Communication Communication model is many-

Communication model is one-one.
Model many.