Introduction to Internet of Things

Prof. Sudip Misra

Department of Computer Science & Engineering
Indian Institute of Technology, Kharagpur

Lecture - 06
Basics of IoT Networking-Part-II

So, we are now going to continue our discussions on the basic issues, basic aspects of the
networking in internet of things. So, we have already seen the different fundamental
issues that are out there. Now, we are going to look at the different protocols that are
there that can be used for something different purposes in IoT.

(Refer Slide Time: 00:44)

Now, based on the different functionalities, there are actually large numbers of protocols
that are proposed for use in IoT. So, based on the functionality, these protocols are
classified in this in this manner. So, this is just a classification that has been shown, but
this is not unique classification in anyway and it should be constituted in that manner.
So, this is just a classification attempt under different categories that is shown and here
as we can see the different protocols for something.

These classifiers you know a different issues are also mentioned in brackets like this. So,
it is basically not possible and also not required to go through all these different
protocols. So, we have selected only a few protocols from this you know red coloured
category and this is what we are going to discuss in this particular course.

93
(Refer Slide Time: 01:55)

So, we will start with the MQTT protocol first and this MQTT, the full form of this is
Message Queue Telemetry Transport. So, it is an ISO standard which is based on publish
subscribe model. So, basically you know what happens is there is some kind of
publishing of the data and then, the fetching of the data by the subscribers. So, this is
how this publish subscribe model works and MQTT basically what it has done is, this
publish subscribe model, it has been made lightweight through the use of this protocol,
so that this lightweight protocol can be used in conjunction with the TCP IP protocol
suit. This is what MQTT supports.

MQTT going back to the history was introduced in 1999 by IBM and is standardized in
2013 by Oasis. This is a standardization organization oasis. So, it has standardized in the
year 2013. So, this particular protocol does couple of things. One is offering connectivity
between different embedded devices between the applications and then, middle ware of
one device and network and communication on the other side of the device. So, we have
connectivity between applications and middle ware for one side and the networks and
communication on the other. This is what MQTT does.

94
(Refer Slide Time: 03:32)

So, in MQTT there are three concepts that are involved. The first we are going to go
through is the concept of a message broker and first, we are going to go through some of
these concepts and there after I am going to show you pictorially how MQTT functions.
So, we have a message broker; the concept of a message broker that basically serves like
a broker which takes control of publishing of the messages and subscription of the
messages. So, publish subscribe is basically controlled by the message broker, number 1.
Number 2 is there is a concept of topic and this is what the client is subscribed and based
on the updates. The data are sent to the clients by the message broker, this data are
distributed by the message broker to the clients who have subscribed to the services.

So, this is design for remote connections limited bandwidth environments and MQTT,
basically the advantage is that it provides every small code foot print. So, basically you
know by writing only a small piece of code, one would be able to achieve all these
different functions that I have just mentioned.

95
(Refer Slide Time: 04:48)

The different components of MQTT are as follows. We have three principle components.
The publishers which involve the different sensors, the subscribers and that means, those
entities, those applications, those units that are interested in the data that is published by
the sensors. Number 3 is the broker in between which helps the publishers and the
subscribers connect to one another and also help in classifying the sensor data into
different topics.

(Refer Slide Time: 05:29)

96
In MQTT there are a few different methods. One is connect, the second is disconnect,
subscribe, unsubscribe and publish. So, basically the connect method helps to connect
with the server, helps to connects this device with the server. Then, disconnect is the
opposite. Whenever it is no longer required to be to remain connected, the disconnect
method helps in disconnecting from the server from TCP IP service offerings and so on.
And then comes the subscribe which is basically subscribing to the services and
unsubscribe is the opposite that whenever it is no longer required to continue with
getting the different data offerings, the data services and so on.

The unsubscribe method can be executed and then, we have the publish method which is
basically publishing data for maybe you know publishing the data from these different
sensors or these different devices to the broker for it to be fetched by the different
application clients.

(Refer Slide Time: 06:49)

So, this is what I was referring to just a short while back. So, what we have is these
pictorial depictions of how these publish subscribe model works in the case of MQTT.
So, we have a temperature sensor, we have laptops, we have mobile devices. This
temperature sensor in this example publishes the temperature which is basically brokered
at this MQTT broker. So, what this fellow does is, it broker you know because it is a
broker between the different clients and different other application serving devices which

97
required the data which can subscribe to the data that is published by these different
sensors.

So, what it is going to do is, first it is going to get the subscription requests from these
different clients, the mobile device, the laptop and so on and so forth. So, first is the
subscribed and there after it is published. So, what is published? The temperature is
published. Where is the temperature residing? It is fetched from the sensor and it is being
brokered at the MQTT broker. So, from those devices, those clients which have basically
subscribed to the services, they are going to get this sensing updates, the sensor data
updates.

(Refer Slide Time: 08:27)

So, when we look at the communication, the architecture that is followed is publish
subscribe publish subscribe architecture and not the request response architecture that is
typically followed by traditional http which is used for the internet, this publish subscribe
model is event driven. That means, whenever there is an event, whenever there is a fire,
when the temperature increases, whenever a camera may be an IoT camera basically
observes some kind of change in the environment or whatever maybe there is an intruder
or whatever it is, so the central communication point.

So, it is sorry, I am sorry. So, what we have is it is event driven. So, whenever there is
some kind of an event and those data are basically pushed to the clients, so these
messages basically are pushed to the clients and then, we have this broker which is in-

98
charge of dispatching all the messages between the senders and receivers and each client
that publishes a message to the broker includes a topic into the message and this is the
topic which is very much important in the case of MQTT and this is the topic which is of
interest to these application clients.

(Refer Slide Time: 10:02)

So, each client that wants to receive the messages subscribes to a certain topic and the
broker delivers all the messages with in the matching topic to the client. So, essentially
what is happening is the clients, they do not have to know each other and what is
required is they only need to communicate with each other over the topic. So, they do not
know about each other and this architecture basically appears to be a scalable
architecture with a scalable solution. There is not much dependency between the
producers and the consumers of the data and that is where MQTT is very popular.

99
(Refer Slide Time: 10:43)

In MQTT we have the concept of the topic as I just mentioned a minute back and this
topic is nothing, but a simple string that can have more hierarchical levels which are
separated by a slash. In this manner a sample topic for sending the temperature data of
the living room could be marked in this manner, could be named in this manner. So, we
have house, a smart house kind of scenario. We have house and within the house we
have living room and the temperature that is collected from this particular living room is
of interest. So, this topic that is of interest and this is a sample, this is an example of how
the topic for sending temperature data of a living room looks like.

So, on one hand the client like a mobile device, a laptop or whatever can subscribe to the
exact topic or on the other hand, it can also use a wildcard. So, this is an example of an
exact topic. Wildcard could also be used.

100
(Refer Slide Time: 12:01)

Wildcards can be used in different ways based on the different levels the subscription to
house/+/temperature results in all messages sent to the previously mentioned topic,
house/living room/temperature as well as any topic with an arbitrary value in the place of
a living room such as house/kitchen/temperature.

So, basically you know this plus sign is a wildcard character which only allows arbitrary
values for one hierarchy. If more than one hierarchical level is required, the multilevel
wildcard is used. So, this is single level wildcard. The plus one is a single level wildcard
and this hash sign is as multilevel wide wildcard. It allows to subscribe to all underlying
hierarchical levels. For example, house/# is for subscribing to all topics beginning with
house.

101
(Refer Slide Time: 13:06)

Different applications that use MQTT, Facebook messenger for online chat, Amazon
website service use Amazon IoT with MQTT, Microsoft Azure IoT hub uses MQTT as
its main protocol. For telemetry messages, the EVRYTHNG IoT platform uses MQTT as
an M2M protocol for connecting several products and devices. Adafruit uses MT MQTT
cloud service for IoT experimenter’s caller Adafruit IO.

(Refer Slide Time: 13:45)

Finally after understanding the overall philosophy behind MQTT, let us now quickly
review the secured version of MQTT which is called the secure MTTM MQTT, the

102
secure MQTT or SMQTT in short. So, this is known in both these ways and this actually
to me is quite similar in notion to http and https, the secure http. So, we have http secure
http, we have MQTT secure MQTT.

So, secure MQTT is an extension of MQTT. So, basically it is an extension of the MQTT
that we just discussed by using different security features such as encryption and so on.
The advantage of such encryption is the broadcast encryption feature in which one
message is encrypted and delivered to multiple other nodes which is quite common in
IoT applications. In general, the algorithm consists of four main stages i.e. the setup
stage, the encryption stage, the publish stage and the decryption stage. So, setup
encryption publish and decryption.

(Refer Slide Time: 15:03)

In setup phase, the subscribers and publishers register themselves to the broker and get a
master secret key according to the developer’s choice of key generation algorithm. So,
depending on what key generation algorithm is used, the subscribers and the publishers
register themselves to the broker and get a master secret key according to that particular
algorithm.

So, when the data is published, it is encrypted and published by the broker which sends it
to the subscribers which is finally decrypted at the subscriber end having the same
master secret key. The key generation and encryption algorithms are not standardized
SMQ. SMQTT is proposed only to enhance the security aspects of MQTT.

103
So, in this part of the lecture on basic topics on IoT networks, what we have done is
primarily we have gone through two protocols. We have seen that there is an assortment
of different protocols that they have to support the networking of IoT, but in this
particular part of the lecture, we have focussed mostly on two protocols. One is the
SMTT and the other one is, sorry MQTT and the other one is SMQTT which is basically
the secure version of MQTT.

So, we are going to go through few other protocols in the next lecture and from that we
can use them finally to establish to build different platforms for IoT in a small scale or
even in the larger scale.

Thank you.

104
 Introduction to Internet of Things
Prof. Sudip Misra
Department of Computer Science & Engineering
Indian Institute of Technology, Kharagpur

Lecture - 07
Basics of IoT Networking- Part- III

So, we continue with our discussions about the different protocols that are used for
communication and networking of internet of things.

(Refer Slide Time: 00:34)

So, the next protocol that we are going to cover is the CoAP protocol and the full form of
which is Constrained Application Protocol.

105
(Refer Slide Time: 00:37)

This protocol is particularly used for web transfer and by web transfer I mean very
similar to the HTTP, but web transfer in the context of constrained networks resource,
constrained networks with nodes which are constrained with respect to different
resources, such as limited energy or power supply, limited computational resources,
limited communication resource, limited bandwidth environment and so on.

So, CoAP is sort of like an HTTP equivalent that can be used in the context of IoT and
the other thing that we have to understand is CoAP is strictly speaking a session layer
protocol. However, we can also contribute as an application layer protocol as well. So, in
IoT particularly when we taking to consideration the different applications involving
machine to machine communication for example, smart energy, smart environment you
know building automation and this kind of applications CoAP comes out to be very
much useful.

CoAP is based on a request response model. So, basically you know it will very clear
shortly about how CoAP works. So, at this point you know you have to understand that
there are two endpoints, the source and the destination and a request is sent and a
response is received back from the in the other end point and that means, the destination.
So, this is how CoAP works. So, there is sort of like a client server, kind of interaction
that goes on. So, there is the datagram that is sent from one endpoint to another and that

106
basically, it is an asynchronous kind of communication and it also works on top of in the
transport protocol UDP.

(Refer Slide Time: 02:55)

So, basically CoAP, we have to keep in mind that CoAP works on top of UDP. So, this
particular protocol is based on IETF RESTful Environment Working Group. So, there is
architecture, restful architecture and these people who have to post the restful
architecture, they have proposed the CoAP protocol. So, it is basically used sort of like a
lightweight equivalent of the HTTP and it is a standard REST. So, let us go back to the
rest protocol first.

So, REST is a standard interface between the HTTP clients and servers, but this REST
protocol is useful where there is no resource limitation because you know this is quite
resource hungry kind of protocol which communes lot of resource. REST basically is not
good for constrained environments like IoT.

So, CoAP is sort of like a protocol which is a lightweight equivalent of the rest
architecture and rest protocol and it helps to communicate with under low power
constraints.

107
(Refer Slide Time: 04:19)

So, this is how CoAP works. So, as I was telling you that it is a session layer protocol or
even we can think of it as an application layer protocol. So, it basically works on top of
the transport layer. So, session layer or application layer are on top of the transport layer
and the transport layer protocol that is used in the context of CoAP is that UDP. So,
CoAP basically has two main sub-layers, one is the messaging sub-layer and the other
one is the request response sub-layer and I will show you pictorially how it looks like
shortly.

So, in summary actually at a high level, we can think of the messaging sub-layer to be
responsible for functionality, such as reliability and duplication of avoidance of
duplication of messages while the request response sub-layer is responsible for the
communication, exact communication that is going to take place, the request being sent
and the response being received. So, these are two main sub-layers that are there in the
CoAP protocol or the CoAP architecture more specifically. So, there are different
messaging modes for CoAP, one is the confirmable mode, the second is the non-
confirmable mode, the third is the piggyback mode and the fourth is the separate.

108
(Refer Slide Time: 05:41)

So, as I was telling you in the protocol stack in terms of the layered architecture, CoAP is
a protocol of the session layer. Some can also think of it you know in some cases we try
to avoid the session layer. So, in that case, we can think of CoAP to be merged with the
application layer, but if session layer is considered, it is a protocol of the session layer.
So, session layer means that it lies between the transport layer and the application layer.

So, at the transport layer, we have the UDP protocol and different applications being run
in the application layer and CoAP basically sits in between. So, we have two sub-layers,
one is the request response and the other one is the messages. Messages is mostly
concerned about the reliability in sharing, reliability of the network, reliability in
communication whereas, request response is more to do with the exact communication in
terms of sending a request and getting a response back.

109
(Refer Slide Time: 06:49)

So, this is what I was mentioning earlier. So, we have different types of messages that
are used in CoAP. The first one is the confirmable message, the second is the non-
confirmable message, the third is the piggyback message and the fourth is the separate
message.

(Refer Slide Time: 07:11)

So, basically when we look at CoAP confirmable message, this is how it works. So, we
have CoAP is basically a connection between the client and the server in a resource

110
constraint environment. So, what happens is a message is sent and an acknowledgement
is received in the case of a confirmable message.

(Refer Slide Time: 07:48)

So, this message basically you know it gets an acknowledgement back. So, it is a
confirmable message and then, for non-confirmable message, there is no
acknowledgement from the server and then, we have the piggyback message which is
used for a client server direct communication where the server sends its response directly
after receiving the message. So, that basically you know what happens is along with the
acknowledgement message, the data is also sent, the response is also sent in the case of
piggyback messages.

And in the case of the separate mode, it is used when the server response comes in a
message separate from the acknowledgement and that basically may take some time to
be sent to the server, this particular message might you know because it is coming
separate from the acknowledgement. May be acknowledgement might be received and
the message might be received after a rewind back.

So, no sorry find thereafter I am sorry and so, similar to HTTP code basically utilizes
different functionalities, such as get functionality, get message, put message, push
message, delete message etcetera. So, basically get is for retrieval of some data, put is for
creation. So, you want to put some data or some message into the repository, so in that

111
case push or put the server. So, in that case that put message is used and then, we have
the update for that push is used and the delete message is for deletion purpose.

(Refer Slide Time: 09:30)

So, we have already looked at how the confirmable and the non-confirmable message
request response looks like. So, let us now look at pictorially how the piggyback
message request response model looks like. So, here basically as we can see first a
message is sent in piggyback in contrast to the previous two models. That means, the
confirmable and the non-confirmable models, rather the confirmable model. So, what we
have vary as we can see the data is basically piggybacked along with the
acknowledgement message. So, this is how the piggyback message request response
model functions.

Separate message, we have a message being sent and acknowledgement being received
and there is a wait period after which the data is going to be sent separately from the
server to the client and corresponding to that the client is going to send an
acknowledgement back to the server.

112
(Refer Slide Time: 10:52)

So, this is how separate messages look like so these basically CoAP as a whole and these
different message types, they together help to induce the overhead and the parsing
complexity of the network. So, there are different types of discovery of resources that are
supported by CoAP and we are going to go through them little bit further.

(Refer Slide Time: 11:23)

So, but so we now start with the XMPP protocol which is the next protocol to be
discussed.

113
(Refer Slide Time: 11:34)

The full form of XMPP is Extensible Messaging and Presence Protocol. So, it is a
message oriented middleware that is based on XML, whereas XML is particularly used
for unstructured data. XMPP is useful for real time exchange of structured data and it is
an open standard protocol.

(Refer Slide Time: 12:03)

So, XMPP uses a client server architecture, it uses a decentralized model meaning that
there is no server that is involved in the message transfer and it provides facilities for

114
discovery of messages which are residing locally or globally across the network and the
availability information of these services.

So, as we can now basically think about it, so it is well suited for cloud computing
environments, where virtual machines networks and firewalls are involved and would
otherwise present obstacles to the alternative service discovery and message based
solutions. So, you know think of it this way that with the help of XMPP, we can do
things very similar to like pin protocol. So, in the case of pin, basically when we have the
involvement of firewalls etcetera, so pin cannot be used as such, right.

So, in this particular case, in the case of XMPP, it basically removes all these constraints,
these barriers for having the discovery of the services and if it is the discovery of
services locally, then it is no problem, but if it is across the network and there is a
firewall in between, then XMPP can still work.

(Refer Slide Time: 13:34)

So, some of these highlights of the XMPP protocol, it is based on the concept of
decentralization where there is no central server and then, you know everybody can run
the XMPP server theoretically and it is based on open standard. So, there is no
involvement of royalties or granting permissions to implement the XMPP specifications,
different security features that the standard ones, such as authentication, encryption,
etcetera, can be implemented using XMPP on top of XMPP rather and XMPP also offers

115
flexibility in terms of supporting interoperability between different systems, different
devices, different protocols, and so on.

(Refer Slide Time: 14:26)

So, consequently I was giving you the analogy with the traditional pin protocol that is
used for internet and here we are trying to have something similar, but you know it is bit
different in this particular manner. So, it is now if you look at this particular figure what
we see is with the help of XMPP, not only it is possible to communicate with other
servers like in the case of the traditional internet, but also with other messaging platform
such as ICQ, AIM, Yahoo and so on. So, this is also possible. So, not only that this is
possible, but additionally it is also possible to communicate with other intranets, other
intranets.

116
(Refer Slide Time: 12:25)

So, XMPP basically helps in doing this. There are few core XMPP technologies, one is
the core technology which provides information about the core XMPP technologies for
XML streaming, then we have jingle which is used for multimedia signaling with the
help of voice you know wherever there is multimedia resources, such as voice, video, file
transfer etcetera, it can help in signaling jingle multi-user chat.

It is a flexible technology which can be used for multi-party communication. Pub sub is
Publish Subscriber. Publish subscribe model and publish basically alerts, this pub sub
model basically alerts and notifies for data syndication and the BOSH technology. It is
used for HTTP, binding for XMPP wherever there is required meant for HTTP binding
when using XMPP, this can be used.

117
(Refer Slide Time: 16:23)

There are different weaknesses as well of XMPP protocol. It does not support QOS, text
based communication including you know Higher Network Overheads are involved in
the use of XMPP. So, it is not good for text based communication.

(Refer Slide Time: 16:56)

Binary data must be first encoded to base 64 before it can be transmitted. The different
applications that use XMPP publish subscribe systems, pub sub systems, then signaling
for voice video file transfer, gaming applications, IoT applications such as smart, grid,
social networking and so on.

118
So, with this we have come to the end of two order protocols. We have discussed that
two order protocols. So, XMPP is a protocol that is very useful. So, we have discussed
about the CoAP protocol first with the session layer protocol which is useful for use in a
similar kind of platform, where rest is required for communication between the client
and the server. So, CoAP protocol and then, we have discussed about the XMPP
protocol.

Thank you.

119
 Introduction to Internet of Things
Prof. Sudip Misra
Department of Computer Science & Engineering
Indian Institute of Technology, Kharagpur

Lecture - 08
Basics of IoT Networking- Part- IV

So, we continue with few other protocols and now, we are going to look at one very
interesting protocol which is bit different from the previous protocol that we have been
discussing in the context of basics of networking, IoT networking.

(Refer Slide Time: 00:42)

So, this protocol is the AMQP protocol and the full form of which is Advanced Message
Queuing Protocol and this Advanced Message Queuing Protocol basically follows some
kind of an open standard which is based on the ISO. So, it follows, basically ISO IEC
1964 Standard.

So, this standard basically helps define how mists are going to be passed from
businesses, business applications or organizations. So, in other words, a particular
business is comprised of different systems and different processes, business processes.
So, a business can be conceived as a collection of different systems and business
processes. So, this particular standard helps in communicating between these systems
connecting rather connecting between these different systems and the business processes
of that particular business.

120
So, it is a binary application layer protocol. It is an application layer protocol and the unit
that is used for data transfer in this case is known as the frame. So, it is bit different from
the concept of frame in the link layer. So, you know here also it is known as frame, but it
is not exactly the same frame that we talked about in the context of link layer protocols.

(Refer Slide Time: 02:07)

So, let us now try to understand how this particular protocol works. So, we have clients
which are producers of different messages. We have a server and on the other hand, we
again have a layer of clients which are consumer of the messages. So, we have the
producers of the messages, we have the server which comprises of routers and filters and
queues for forwarding the message, for queuing and for buffering and forwarding the
messages and then, we have the consumers of the message. So, this server basically acts
as a broker, the server acts as a broker.

So, essentially what happens is these clients for example, different sensors they produce
different data. They produce different messages. So, these messages are sent to the
different routers and filters in this manner which again can be buffered in a queue and
with the intention of forwarding them again in the future and these data, the forwarded
data beyond the server are sent to other types of clients which are the consumers of these
messages. So, we have the producers, we have the server in between and we have the
consumers.

121
(Refer Slide Time: 03:48)

So, AMQP can be thought of as a collection of different features. These features

basically cut across things such as organization. So, basically AMQP can help in
connecting different organizations, connecting different technologies, connecting
different organizations with respect to time and that means, at different points of time
and the technologies and the organizations that are located in different locations
connecting them as well.

So, these are the main features of AMQP. So, connecting organizations technologies
time and space which could be distinct from one and so, distinct organizations, distinct
technologies you know with respect to time variations in the space AMQP basically
helps in connection.

122
(Refer Slide Time: 04:53)

Few other features from a different perspective AMQP supports security reliability
interoperability I think, security, and reliability are understood. I do not mean to
elaborate on them further, but in terms of interoperability, we are basically referring to
interoperability of not only devices, but also protocols, algorithms messages and so on.

So, this is the next feature and then routing of messages queuing of messages and that it
follows an open standard which is based on ISO.

(Refer Slide Time: 05:34)

123
So, there are different messages that are involved in this particular protocol and what is
required is to have message delivery guarantees. These message delivery guarantees can
be in three different forms. The first one is at most once which means that each message
is delivered at most once.

That means, once or never to the intended recipient. At least once each message is
certain to be delivered, but may be multiple times, but it should do at least words exactly
once messages will always certainly arrive and do. So, it is only once. So, these are the
three different messages, types of message, delivery types and guarantees that are
involved in this process.

(Refer Slide Time: 06:21)

So, the different frame types there are 9 AMQP frame types that are used for defining to
initiate control and tear down the transfer of messages between two peers. So, we have
the first one, the first frame type is the open frame type which is used for opening a
connection corresponding to this. We have the closed frame type close frame which is
for closing the connection closing the connection. Then, we have the begin frame type
which is for opening a session and correspondingly we have the end which is for closing
that particular session.

Then, we have the attach which is basically used for initiating a new link transfer for
sending actual messages flow, for controlling message flow rate and disposition that
informs the changes in state of transfer and then, we have the detach which is for

124
terminating the link. So, open, close, begin, end, attach, detach, and in between we have
transfer flow and disposition frame types.

(Refer Slide Time: 08:03)

So, the different components of this particular protocol include three things, three
primary things. One is exchange, the second is queue and a third is bindings. So, the
exchange component is basically part of the broker that is in that is task to receive
messages and route them to the queues. Queues are basically separate for different
business processes.

So, different separate queues are used for different or separate business processes and the
consumers receive the messages from the queues. Bindings are rules for distributing the
messages which means that who can determine who can access what message, the
destinations of the message and so on.

125
(Refer Slide Time: 08:54)

So, we have you know different exchange types in the case of AMQP. AMQP has
basically four different exchange types. One is the direct, the second is basically the fan
out, the third is the topic and fourth is the header. So, these are the four different AMQP
exchanges. So, direct exchange, fan out exchange, topic exchange and header exchange,
we are not going to go through these in detail just to keep things simpler and easily
remembering. So, that is the reason we have not discussed these in further detail.

(Refer Slide Time: 09:39)

126
So, these are some of the features of AMQP. It offers targeted QoS which basically
targets selective offering of QoS to certain links persistence which basically implies
MES offering message delivery guarantee, then we have the delivery of messages to
multiple consumers, the possibility of ensuring multiple consumption, possibility of
preventing multiple consumption and high speed.

(Refer Slide Time: 10:10)

In terms of the applications, it can be used for monitoring and global updates, sharing
applications, connecting different systems and processes to talk to one another, allowing
the servers to respond to immediate request quickly and delegate time consuming tasks
for little processing, distributing a message to multiple clients for consumption, enabling
offline clients to fetch data at any time and increasing the reliability and uptime of
application deployments.

So, this basically concludes the discussion about this protocol. So, with this we basically
come to an end of the series of lectures on the Basics of IoT networking and in the next
lecture, again we will look at few protocols, but the perspective from which we are going
to look at is going to be different. So, there we are to look at protocols that are very much
useful and that are used for establishing connectivity in IoT.

Thank you.

127
 Introduction to Internet of Things
Prof. Sudip Misra
Department of Computer Science & Engineering
Indian Institute of Technology, Kharagpur

Lecture – 09
Connectivity Technologies- Part-1

So far what we have gone through and have understood what the different basic concepts
that are involved in the networking aspects of IoT. So, we have gone through different
protocols and these protocols include XMPP protocol, MQTT protocol, CoAP protocol
cope and AMQP protocol. So, these are the different protocols that we have gone
through so far and these protocols are primarily involved with service offerings.

So, they are at the higher level and now we are going to understand, we are going to go
down and we are going to understand them more at the physical level. So, physical and
partially are the link layer. At physical level and link layer level what are these different
protocols; these protocol scan as whole offer connectivity to the different devices which
can help in physical establishment of the network.

(Refer Slide Time: 01:40)

So, we are now going to go through some of these protocols. So, when we talk about
these protocols, this can be used for both consumer IoTs and industrial IoT. So,
consumer IoT means like smart home, then different applications of smart home, then for
different serving, different applications for you know consumer devices, consumer based

128
systems and so on, then for industrial IoT it is like connecting different Machines,
industrial Machines, manufacturing Machines and so on offering different connectivity’s
and smart intelligence on top of those devices and so on.

So, all these different protocols that are listed in front of us for example, this IEEE 802.4
which is a protocol as well as standard as well this can be used ZigBee is very much
useful, 6 LoWPAN, wireless HART, Z-wave, ISA 100, Bluetooth NFC and RFID. So,
we are going to go through most of these protocols in the subsequent lectures.

So, we will start with the first one which is the 802.15.54 which is IEEE standard and
this basically is used for forming Wireless Personal Area Network.

(Refer Slide Time: 03:00)

So, personal skill means like you know in the skill of person. So, personal area network
are used for forming network at person skill. So, these are basically normally low data
rate networks. So, these basically are developed for low data rate monitoring and control;
so monitoring mean sensing low data, rate sensing, and then control maybe through
actuation and so on and in the process offering because it is low data rate and is used for
wireless personal area network.

These can help in offering extended life time to the networks extending the lifetime of
the network because of the reduced power consumption. So, this standard uses two sub-
layers, two layer sorry two layers. One is the physical layer and the other one is the MAC

129
layer plus the sub layers like logical link control, the SSCS which is the Service Specific
Convergence Sub-layer, etcetera to communicate with the upper layers, but this
particular standard it focuses on primarily two layers which is the physical and MAC.

So, physical MAC partly LLC and SSCS that is it about 802.15.4 as such. So, before we
go any further I would like to remind you that in this case, the communication takes
place in ISM band.

(Refer Slide Time: 04:46)

So, this is what I was explaining to you. So far this particular standard, this particular
protocol defines specifications. This gives specifications for operating in the physical
layer, MAC layer, SSCS and LLC sub-layers and how these are going to connect with
the network layer, but the main focus is on the physical layer and the data link layer of
the traditional OSI stack.

130
(Refer Slide Time: 05:26)

So, few features that we are going to just list over here and we are not going to go
through them in detail because here I am assuming that I mean you have some basic
background in networks and communication; if you have, you will be able to understand
these a little better. However, if you do not have you know this is just for you to
remember. We cannot go through the details of how these communication technologies,
these different schemes, they work. So, this 802.15.4, it is based on the modulation
schemes which is known as the DSSS which is Direct Sequence Spread Spectrum
Technique. So, this is the modulation scheme that it uses DSSS modulation scheme.

So, these basically are highly tolerant of the noise and interference and offers link
reliability improvement mechanisms. So, this particular standard is helpful in
environments which are noise prone and have lot of interferences and in a presence of
noise and interference, this particular standard can help in improving the reliability of the
network.

So, it has two different variants, two different versions. The low speed version basically
uses the BPSK and the high speed version uses what is known as the offset QPSK, O-
QPSK and for MAC communication, it uses CSMA-CA for channel access. That means
carrier senses multiple access and carrier sense multiple. CA stands for basically
collision avoidance. So, carrier sense multiple access with collision avoidance is used for
channel access and multiplexing basically allows multiple users of nodes to

131
communicate with one another in an interference free banner at different times over the
same channel.

(Refer Slide Time: 07:40)

So, power consumption using this particular protocol is minimized due to the
infrequently occurring very short transmissions with low duty cycle which is less than
one percent. So, the minimum power level that is defined in this particular standard is
minus 3 dBm or 0.5 microwatts. The transmission for most of these cases is line of sight
transmission, however non-line of sight transmission is also possible, but you get better
efficiency, better performance if line of sight communication is used.

The standard transmission range basically varies between 10 meters to 75 meters. 75

meters will particularly be obtained if it is used outdoors, but for indoor environments
typically like 10 meters I would say above 30-40 meters and so on. So, the best case
transmission that is received for outdoors can be even up to 1000 meter also in certain
cases, but typically you know. So, that is a theoretical possibility, but typically you do
not get that much of a transmission range even if it is outdoors that networking
technologies that are supported are the star topology and the mesh topology.

132
(Refer Slide Time: 09:03)

So, there are different variants of 802.15.4. So, within axel before we proceed further
802.15.4, the IEEE standard is useful for forming wireless personal area network. That
means, small range low data rate, low power consuming networks, this can use the
802.15.4. Now, this basically has different variants, the base variants are a and b, but
there are variants like the c variant which is used for China, the d for Japan, the e variant
is used for Industrial applications, the f variant for active, RFID uses the g for smart
utility networks such as smart grids.

So, these are the different variants of the 802.15.4 which are typically used for different
purposes in different parts of the world.

133
(Refer Slide Time: 10:10)

Now, 802.15.4 networks these networks can be classified into two types. The beacon
enabled network and the other one is non-beacon enabled network. So, we will look at
what is beacon enabled and non-beacon enabled shortly and also, these networks
basically use different types of devices.

One is FFD which stands for Fully Functional Device and the other one is the Reduced
Functional Device i.e. RFD. The fully functional devices include the PAN coordinator.
That means, a personal area network coordinator, the router or the device, they can act as
full functional device which can undertake all different types of functionalities, whereas
the reduce functional device can only send something very similar.

So, they have reduced functional. If they cannot route, they cannot switch, they cannot
send the packets, they cannot read a packets and so on. So, they can only do very simple
things. These are the RFD's Reduced Function Devices.

134
(Refer Slide Time: 11:22)

So, once again we have the fully functional device which can talk to all types of devices
and can support full protocols. The reduced functional devices can only talk to a fully
functional device and has lower power consumption and uses minimal CPU and RAM,
so minimal processing and storage.

(Refer Slide Time: 11:47)

So, there are different frame formats, frame types that are defined for 802.15.4. So, there
are typically five frames five frames that are defined for 802.15.4. So, we have the data

135
frame, the acknowledgement frame, the command frame, the MAC frame and the beacon
frame. So, I do not need to elaborate further because these are quite obvious.

The command frame is used for things such as different control functions, such as
associating a device with a PAN coordinator or disassociating a device or performing
different other control functions. So, there are command frames and then, the MAC
frames also do not need to elaborate further. This is standard and then, we have the
beacon frames.

This beacon frames physically what they do is this pan coordinator at regular intervals of
time, they basically sends these beacons which can basically advertise its present and the
different devices that are able to get the signal. That means the beacon form that is
broadcast by this particular PAN coordinator. These device they know that there is this
pan coordinator which is present. So, this is basically enabled, this particular
functionality is enabled with the help of this beacon frames.

(Refer Slide Time: 13:18)

So, going back to the previous classification, we have seen that we have a beacon
enabled network and the non-beacon enabled network. So, what is this beacon enabled
network? So, in beacon enabled network basically periodically pan coordinator, it is
going to send periodic transmissions of beacon messages. So, periodically beacon
messages are going to be transmitted which are going to be received by different devices
which one to get associated with the PAN and so on.

136
Then, the data frames are sent via slotted CSMA CA with a super frame structure that is
managed by the PAN coordinator. The beacons are used for synchronization and
association of other nodes with the coordinator and the scope of operation basically span
the whole network.

(Refer Slide Time: 14:11)

In the context of beacon enabled networks, in case of non-beacon enabled networks, the
data frames are sent via un-slotted CSMS CA. The previous one for beacon enabled
network, it was the slotted CSMA CA, but in the case of non-beacon enabled network, it
is the un-slotted CSMS CA. The beacons are used only for link layer discovery and that
means, whether there is any connectivity where is there is any link from one device to
another and so on. So, beacon messages basically will help in the discovery of these
different links from the PAN coordinator to the different devices and so on.

So, basically these network, this type of network that means the non beacon enabled
network request both the source and the destination ids. So, as 802.15.4 is primary mesh
protocol, it is primarily based on a mesh protocol. All the protocol addressing must
adhere to the mesh configuration. So, this is basically in a excel how the 802.15.4
protocol functions and the different features of it.

We are now going to look at the ZigBee protocol which is heavily used for IoT
applications for establishing connectivity between the different nodes and this ZigBee as
we will see shortly is basically, it works on top of the 802.15.4 for extending this

137
functionalities to the different other layers. So, if you recall 802.15.4 is useful for
establishing connections and function functionalities in the physical layer and the MAC
layer. ZigBee basically will take these functionalities to the higher layers network and
beyond network layer and beyond.

(Refer Slide Time: 16:08)

So, let us look at how ZigBee basically functions. So, ZigBee protocol is defined by the
layer 3 and above. So, it works on top of layers 1 and 2 of 802.15.4 and extends to layer
3 and above and ZigBee basically works on top of 802.15.4. So, this is something that
we have to understand and we have to remember that there is a difference between
802.15.4 and ZigBee.

This is what many people often confused. People think that ZigBee and 802.15.4 are one
and the same, but it is not. So, ZigBee is definitely based on the 802.15.4, but it has its
own distinct identity. So, the ZigBee basically uses layer 3 and layer 4 to define
additional communication enhancements and these enhancements include authentication
with valid nodes encryption for security and data routing and forwarding capability that
enables mesh networking and ZigBee is typically used in wireless sensor network
applications. It is heavily used in wireless sensor network applications, where mesh
topologies are formed with the help of ZigBee.

138
(Refer Slide Time: 17:27)

So, this is a diagraph diagram which basically shows the position of ZigBee with respect
to 802.15.4. So, where is 802.15.4, it focus is mostly on the 5 and the MAC layers
ZigBee takes it beyond MAC beyond network layer. So, network layer and the rest of the
layers all the way up to the application layer basically this extension or the enhancement
is made possible with the help of ZigBee over the 802.15.4 protocol.

(Refer Slide Time: 18:06)

In ZigBee there are primarily two different components. The first one is known as ZDO
which stands for ZigBee device object and it takes care of issues such as device

139
management, security provisioning policies and so on. So, these are the different
functionalities of ZDO and that means the ZigBee device object component.

The second component is APS which stands for Application Support Sub-layer which
takes care of services such as control services, interfacing bridging between network and
other layers and so on. So, these are the two distinct components with separate
functionalities as we have just gone through.

(Refer Slide Time: 19:02)

So, ZigBee basically supports star topology. So, this is the star topology that we see over
here and we do not need to elaborate on this further. So, we have this controller node and
we have these different age devices and these age devices, they form a star topology with
the coordinator node. Sorry it is not the control; it is a coordinator node along with the
coordinator.

Then, we have this coordinator node can be a gateway node also and this can be a simple
local area network or local area network equivalent. Then, we have the cluster tree
topology, where these are like different clusters that are formed with these different blue
colored cluster head which basically in reality are things such as routers and hubs.

So, these will act as routers and these routers will form a tree like structure with the
coordinator node. So, this is why this is known as the cluster tree topology and then, we
have the mesh topology which is forming a mesh network with the help of these different

140
routers. These different routers, they form a mesh backbone kind of network and to each
of these routers, these different devices are attached.

So, basically these green device are the end devices in this particular diagram and these
green devices attached to these blue devices which basically denote the the routers and at
the end of this mesh network on one end, we have this coding and then, node which acts
as a gateway and from this point on it offers connectivity to the outside network such as
the internet

(Refer Slide Time: 21:17)

So, in a mesh any node can communicate with any other node within its range. So, this is
the main advantage to offer faulty tolerance, reliability. Meshtopologies are very much
useful. So, if the nodes are not in the range, messages are relayed through intermediate
nodes. So, this allows the network deployment over large areas. So, using mesh topology
you can extend the network to larger areas, you can span across larger area. So, this is
possible with the help of the meshtopology.

141
(Refer Slide Time: 21:52)

So, the meshes have increased network reliability. For example, if nodes C and F are
down in this particular scenario, so let us say that initially we had this kind of a mesh.
Now, if the nodes C and F, these routers are down for one reason or another, the
messages can still go from A to Z using this particular path because an alternative path
was possible to have in because it is a mesh network.

Now, the ZigBee mesh networks are self-configuring and self-healing. Self-healing is
quite obvious because if there is some link failure or node failure or something, it is
possible to have other alternative routes and yes they can configure on their own, they
can form the network on their own. So, this is the advantage of the ZigBee mesh
network.

142
(Refer Slide Time: 22:49)

So, in ZigBee there are different entities. The first one is the ZigBee coordinator, ZC.
The ZigBee in the coordinator basically forms the root of the ZigBee network.

So, the entire network or the network tree has a route and these routes are known as the
ZigBee coordinator and form this coordinator. So, first of all there is a single coordinator
and from this coordinator, there is single hop connectivity to the end devices. So, this
coordinator basically stores information about the network which is under it and which is
outside it.

So, basically you know it is sort of buffers some of this information that I received from
these end devices and stores with it for certain time. It buffer for some certain time. So, it
also acts as a trust center and repository for the security keys.

143
(Refer Slide Time: 23:54)

Then comes the ZigBee router which is capable of running applications as well as
relaying information between the different nodes that are connected to it and then, we
have the end device which contains just enough functionality to talk to the parent node
and it cannot relay data from other devices. So, it has reduced functionality. So, it is a
reduced functionality device.

(Refer Slide Time: 24:21)

Now, ZigBee also incorporates a network layer. So, this network layer uses the ad hoc on
demand distance vector routing protocol which is the AODV protocol and it is been

144
popular in the case of ad-hoc networks. It is used mostly in the ad hoc networks for as a
routing protocol that operates in the network layer and it used to find the final
destination. So, how it can be found? AODV basically this particular protocol, it
broadcasts a route message to all its intermediate neighbors.

These neighbors basically they relay the same information to their neighbors intern and
eventually this message space across throughout the network. Upon discovering of
destination, a low cost path is calculated and is informed to the requesting device via the
unicast messaging. So, this is how this particular protocol functions.

(Refer Slide Time: 25:20)

So, ZigBee has different applications and it can be used for building automation smart
homes, smart health care, telecom services, offering link connectivity to led lighting
systems, then smart energy for home, energy monitoring, building automation. I think I
have already mentioned remote control and so on so forth.

These are the different applications where this ZigBee protocol can be used. So, we
come to an end of this. So, we have discussed two very important protocols IEEE
802.15.4 and the ZigBee protocol in this particular lecture. We have seen that where as
the 802.15.4 it is primarily restricted to the physical and the MAC layers ZigBee
basically extends it or enhances its functionality beyond network layer and all the way up
to application layer.

145
So, ZigBee basically uses the protocol AODV for routing purposes and it is one of the
very popular protocols that is used for sensor networking applications particularly using
the mesh topology and we have all seen that there are different types of topologies are
there i.e. star topology, cluster tree topology, mesh topology and so on. Mesh topology is
particularly useful when there is higher reliability that is required from the network
deployment for the application for which it is being used.

Thank you.

146
 Introduction to Internet of Things
Prof. Sudip Misra
Department of Computer Science & Engineering
Indian Institute of Technology, Kharagpur

Lecture - 10
Connectivity Technologies- Part-II

The next protocol that we are going to go through is also for offering connectivity
between different devices forming the connectivity and this protocol is the 6LoWPAN
protocol. So, this 6LoWPAN protocol is basically it runs over IPV6 that is from where
this 6 figure comes. So, it is from IPV6. So, it runs over IPV6. So, it offers radio
connectivity, radio linkages over IPV6 protocol. So, using IPV6 protocol and IPV6 as we
know is for addressing, it is an addressing protocol and it is very popular for use for
addressing in the case of IoT networks because of the large address space that is required
for IoT. So, let us look at this 6LoWPAN protocol.

(Refer Slide Time: 01:16)

So, it stands for 6LoWPAN stands for Low Power Wireless Personal Area Network over
IPV6 and it allows for the smallest devices and each of these devices having limited
processing ability to transmit information wirelessly over the internet protocol.

So, we have low power small devices limited processing capability as is typical of IoT
systems and wireless communication being present. So, it basically helps in establishing
connectivity in this kind of networks. So, it basically helps this 6LoWPAN protocol, it

147
helps and allows to have this low power devices to connect over the internet because
IPV6 is going to be used for addressing.

So, at the network layer, that is the reason why this protocol is useful for connecting this
IoT network. These low power devices to the internet, it is basically created out of the
IETF RFC 5933 and RFC 4911. So, these are two different RFC based on which this
6LoWPAN protocol is specified. So, this specification is available in these RFC's.

(Refer Slide Time: 02:48)

So, these are some of these features of 6LoWPAN and it allows IEEE 802.15.4 radios.
That means, in the previous lecture we have gone through this particular protocol which
is useful for setting up connectivity between the different nodes and it primarily operates
at the physical and the MAC layers. So, the radios of the 802.15.4 is used to carry 128 bit
addresses of the IPV6.

So, basically 6LoWPAN is an application or its joining you know conceptual joining of
802.15.4 radios with IPV6, but how it is made possible because you know 802.15.4, it is
low powers in light weight protocol and IPV6 is not lightweight. So, how it is made
possible? So, the header it is possible with the help of header compression and address
translation techniques that basically helps to convert 802.15.4 radios to access the
internet.

148
So, it will help this 802.15.4 radios to access the internet using header compression and
address translation techniques. IPV6 packets are compressed and reformatted to fit the
802.15.4 packet structure. So, this is what is done. IPV6 packets large in size, they have
to be compressed, they have to be reformatted and they have to be mapped with the
packet format of 802.15.4 which is primarily meant for low power networks, small scale
low power networks as is typical of IoT.

(Refer Slide Time: 04:58)

So, it can be used for, IoT 6LoWPAN can be used for IoT smart grid applications, smart
home applications, M2M applications and many other different, similar applications. So,
for addressing in 6LoWPAN, there are two types of addresses that are used i.e. 16 bit
short address which is for PAN specific communication. That means, it is assigned by
the PAN coordinator for communicating within the PAN, the personal area network and
64 bit extended address which is used for global unique connectivity, global unique
addressing throughout the network.

149
(Refer Slide Time: 05:45)

So, IPV6 multicast is not supported by 802.15.4 and IPV6 packets are carried as link
layer broadcast frames in the case of 6LoWPAN. So, we have in front of us the packet
format of 6LoWPAN. So, as we can see over here if you look very closely, we have
802.15.4 and IPV6 club together 802.15.4 radio and IPV6 for addressing over the
internet and these corresponding fields are also shown over here.

So, what we have for corresponding to IPV6? We have the source address, the
destination address and these different other IPV6 fills that are typical in this particular
protocol IPV6 protocol and for 802.15.4 as well, there is this source, the destination both
of which are 64 bits. That means, source and destination together will become 128 bits.

Then, we have this PAN ID because you know when we are talking about 15.4
networking mode go to personal area network. So, the PAN ID is basically stored in this
particular field. So, this is how the 6LoWPAN packet format looks like.

150
(Refer Slide Time: 07:06)

So, that is the packet format. Now, what about the header? There are three different types
of headers. One is known as the dispatch header, the second one is known as mesh
addressing header and the third one is known as fragmentation header.

(Refer Slide Time: 07:17)

So, let us look at these three different headers. The format of these headers is given over
here. So, how many bits we have? We have 8, 16, 24, and 32. So, the header is 32 bits
long, out of which the first two bits are used to identify the dispatch type and this

151
dispatch type basically helps in the dispatch communication initiating the
communication resetting the communication.

Now, this dispatch field has 6 bits. So, it is 6 bits long field and these basically identifies
the next header type and there after the next list of the bits are used to specify the type
specific header and that is determined by the dispatch header. So, then we have the mesh
addressing header and here basically the first two bits are used to store the ID of the
mesh addressing header.

The next field, the V field is 0. If the originator is 64 bit extended address and we have
seen that both of these are possible to have 16 bit as well as 64 bit address. The F field is
0 if the destination is 64 bit address and 1 if it is 16 bit address and hops left are
decremented by each node before sending to the next hop.

(Refer Slide Time: 09:49)

So, how many hops are left until the final destination node? This is basically stored in
this particular field and is decremented as I said hop by hop when one hop is over, it is
complete, that field the value is decremented by 1 and third fragment type is a
fragmentation header and the corresponding fields are shown over here. So, in this case,
the first fragment has this structure, the header has this structure as shown over here and
the subsequent fragments have this particular structure.

152
So, as you can see over here, the main difference between the first fragment and the
subsequent fragments is the inclusion of the datagram offset.

(Refer Slide Time: 10:42)

So this datagram offset basically shows that it will give the value of what are the
subsequent frames that are there. So, this will basically help to connect with the first
frames; so 6LoWPAN because it involves a strong network layer component and it takes
care of routing.

So, the most important type of routing is a mesh based routing and this mesh based
routing is used in the context of a PAN topology, Personal Area Network Topology. So,
routing is used, routing basically is based on the IPV6 protocol in the personal area
network domain and there are two protocols that are used in 6LoWPAN for routing.

153
(Refer Slide Time: 11:40)

One is the LOADng protocol and the other one is RPL protocol and as you can see over
here from this particular figure, we have this IPV6 domain and we have this personal
area network PAN and with the help of this coordinator of the gateway, it connects to the
IPV6 network. That means, the IPV6 based internet, the LOADng routing protocol, it is
primarily a derivation from the AODV protocol that is available and was proposed for ad
hoc networks and this has been used and extended for IoT networks.

So, this LOADng protocol has few different PDOs. The first one is the load request
PDOs and it is generated by a LOADng router, the originator for discovering a route to
the destination. So, the forwarding of such route requests take place until they reach the
destination LOADng router, then comes the route replies which is generated upon receipt
of the route request by the indicated destination and unicast hop by hop forwarding of
these route replies towards the originator.

154
(Refer Slide Time: 12:48)

There is also this route error message that is used to return errors to the originator of the
data in the event that there is some route breakage that take place. So, optimized routing
is supported reducing the overhead that is in cut by the route request generation and
flooding only the generation is permitted, sorry only the destination is permitted to
respond to the route request. Intermediate routers of LOADng are explicitly prohibited
from responding to the route request even if they have been very active in terms of
seeking routes and generating routes in the network.

(Refer Slide Time: 13:40)

155
The route request and route reply messages are generated by a given LOADng router and
this share a single unique monotonically increasing sequence number. Next comes RPL,
Routing Protocol which is based on the distance vector routing for lossy and low power
networks. So, this is where this l comes from this lossy. So, it is used for lossy networks
as well as low power networks for routing.

So, it maintains routing topology using low rate beaconing. Beaconing rate over here
increases on detecting inconsistencies with respect to situation such as load failure or
link failure. Routing information is included in the datagram itself. It uses two types of
routing proactive i.e. routing for maintaining, routing topology and reactive for
dissolving routing inconsistencies.

(Refer Slide Time: 14:31)

The RPL basically separates the packet processing and forwarding from the routing
optimization objective which helps in Low Power Lossy Networks, LLN.

So, this particular protocol, it supports features such as confidentiality, integrity,

ensuring integrity, validating data paths and detecting the presence of loops. The overall
optimization objectives of routing include minimizing the energy minimizing.

156
(Refer Slide Time: 15:16)

The latency and satisfying the constants with respect to node power bandwidth etcetera,
this RPL protocol operates using bi-directional links. So, there is bi-directional flow of
communication of messages.

So, in some LLN scenarios, that mean the lossy scenarios, these links may exhibit
asymmetric properties, right. So, basically asymmetric property means that while the
message is sent from the source to the recipient, it flows through one route may be
directly, but because of all these environmental situations or whatever the response or the
flow of message in the other direction does not take place through the same route. Maybe
it comes back through another route. So, it is asymmetric. So, it is required that the
reachability of a router is verified before the router can be used as a parent.

157
(Refer Slide Time: 16:05)

Finally for this particular lecture, we are going to go through the protocol RFID. RFID is
very popular. It is commonly used. It has been in use for long, it is still used in shopping
malls. What places you know? For example, the ID cards are fitted with RFID tags.
RFID tags can be scanned in the RFID readers. Similar things happen in the shopping
malls like when we go for purchase in certain items, for example, clothing sensors these
also are fitted with these RFID tags and these RFID ID tags can be used to scan against
RFID readers to get further information and so on.

So, how does RFID work? RFID is we have to remember that sensor networks and RFID
sensor networks we have not yet covered. We will cover in a subsequent lecture, but
RFIDs and sensor networks and also other technologies, such as zigbee, 802.15.4, for w
PAN, NFC also which is very similar to RFID. These are different other connectivity
offering mechanisms that are popularly used for IoT applications. So, these are the core
for establishing connectivity in IoT networks.

158
(Refer Slide Time: 17:36)

So, going back to RFID, we have first of all RFID, its an acronym for Radio Frequency
Identification, where the data is digitally encoded in these RFID tags and these data can
be read from the RFID tags. So, RFID tags are encoding the data and these data can be
scanned from RFID tags by the RFID reader. So, these are very similar to the bar coding
schemes and QR coding schemes.

So, in a barcode what happens in a barcode like in libraries ecetera, in a bar, bar coding
schemes are typically used to store information about the books and you know having
the identifiers for the books. So, the barcode basically are like vertical lines, right. So,
there is a barcode reader which can read those vertical lines. Similarly there is QR code
which is sort of like that square square kind of thing which is used for scanning the data,
right. So, these are the QR codes.

So, there is that reader which can take that image and it can basically process that image
to identify the data that is embedded in that particular code. So, RFID tags are also very
similar in operation to the barcodes and QR codes, but the functionality or the way these
operate are vastly different. So, let us now try to understand how RFID tags, the RFID
principle works.

159
(Refer Slide Time: 19:30)

So, every RFID tag consists of an integrated circuit and an antenna. So, basically it is a
very small tag. So, tag inside the tag, there is some circuitry that is there and a small
antenna which is embedded into it, inside it. So, this antenna, is going to be used it is
going to be used for communication with the outside world. That means, outside the tag
and the circuitry basically does number of things including storing the information in that
particular tag, may be the RFID tag could be for a smart card that can be used for storing
employee information.

So, you know employees in an organization have different identification for different
identifiers and those identifiers, the different data can be stored in electronic form inside
the chip that is built into this RFID tags. So, the RFID tag consists of integrated circuit
and an antenna. The tag is covered by a protective material. So, outside the tag is some
kind of a shield protective material which can also act as a shield against various
environmental effects.

The tags can be of two types. One is the passive tag; the other one is the active tag. So,
passive tags are more common and the way these passive tags are operated are through
the process of inductivity. So, inductively when these passive tags, when these tags come
in proximity to the RFID reader, there is some inductive effect, some magnetic force
fields are created due to which the information is transferred from the tag to the reader or
vice versa. So, it is from the tag to the reader and vice versa. Those are the passive tags.

160
(Refer Slide Time: 21:39)

On the other hand, the active tags, they have their own little source of power supply. The
working principle of RFID is similar to its predecessor which is called AIDC. The AIDC
full form is Automatic Identification and Data Capture Technology. So, it performs
object identification, object data collection and mapping of the collected data to
computer systems with little or no human intervention.

So, the concept of RFID is basically adopted from AIDC which is its predecessor. So,
AIDC is no longer very common, however the difference is RFID is mostly wireless, not
mostly it is fully wireless. On the other hand, AIDC uses wired communication. So,
RFID basically uses radio waves that mean wireless communication to perform different
functions which are also performed by AIDC.

161
(Refer Slide Time: 22:44)

So, let us try to understand how RFID works. Let us say that there is a tagged item like
clothing or something in a shopping mall, some kind of a cloth. So, this cloth is tagged
with this RFID reader, no sorry RFID tag. It is tagged with this RFID tag; it is attached
to the RFID tag. This RFID tag consists of the circuitry some kind of a quailing
mechanism and the cover.

This cover is some kind of a polymer, some plastic or some other polymer and the
circuitry is basically stored inside this particular tag. Then, we have this one if we look
over here. So, we have this part, we have this part which basically is for the reader. So,
this part is for the RFID tag and this part is for the RFID reader. As we can see over here,
this RFID reader has a software and a source of power supply and it also has a coil and
when you bring that reader which has a coil inside some magnetic coil, then there is this
magnetic inductive effect producing this magnetic lines of force are created.

162
(Refer Slide Time: 24:37)

So, this is how the data that is there in that small chip inside the RFID tag is transferred
to the RFID reader with the help of this force field magnetic force, field RFID tags.
RFIDs in general are useful for supporting different IoT applications such as inventory
management, asset tracking in an organization, personal tracking you know who is
coming when, who is living when in an organization, what is the attendance.

So, attendance tracking systems for example, controlling access to restricted areas. So,
you know whoever is authorized will be having an RFID tag and they can bring it in
close proximity to the RFID reader and if it is a valid tag, then the door is going to open
for that person and the person can get in. So, it is used for controlling access to the
restricted areas.

Id badging as basically you know identity badges, smart cards in an organization, it is

used for that supply chain management counterfeit prevention particularly in the
pharmaceutical industry. So, these are different applications of the RFID tags.

So, with this we come to an end of this particular lecture and there are few other
protocols that are also very much useful, and we are going to go through them in the next
lectures.

Thank you.

163