Lec2.pptx

Dr. Muhammad Faizan Khan
Department of Information Technology
The University of Haripur
@Dr. Muhammad Faizan Khan 1
Chapter 02: IoT Network Architecture & Design


Drivers Behind New Network
Architectures
Challenge Description IoT Architectural Change Required
Scale The massive scale of
IoT endpoints
(sensors) is far
beyond that of
typical IT networks
The IPv4 address space has reached exhaustion and
is unable to meet IoT’s scalability requirements.
Scale can be met only by using IPv6. IT networks
continue to use IPv4 through features like Network
Address Translation (NAT).
Security IoT devices,
especially those on
wireless sensor
networks (WSNs),
are often physically
exposed to the
world.
Security is required at every level of the IoT
network. Every IoT endpoint node on the network
must be part of the overall security strategy and
must support device-level authentication and link
encryption. It must also be easy to deploy with some
type of a zero-touch deployment model


Architectures
Devices and
networks
constrained
by power,
CPU, memory,
and link
speed
Due to the massive scale and
longer distances, the
networks are often
constrained, lossy,
and capable of supporting
only minimal data rates (tens
of bps to hundreds of Kbps)
New last-mile wireless technologies are
needed to support constrained IoT
devices over long distances. The
network is also constrained, meaning
modifications need to be made to
traditional network-layer
transport mechanisms
The massive
volume of
data generated
The sensors generate a
massive amount of data on a
daily basis, causing network
bottlenecks and slow
analytics in the cloud
Data analytics capabilities need to be
distributed throughout the IoT
network, from the edge to the cloud. In
traditional IT networks, analytics and
applications typically run only in the
cloud.


Architectures
Support
for legacy
devices
An IoT network often
comprises a collection of
modern, IP-capable
endpoints as well
as legacy, non-IP devices
that rely on serial or
proprietary protocols
Digital transformation is a long process
that may take many years, and IoT
networks need to support protocol
translation and/or tunneling mechanisms
to support legacy protocols over standards-
based protocols, such as Ethernet and IP
The need for
data to be
analyzed in
real time
Whereas traditional IT
networks perform
scheduled batch processing
of data, IoT data needs to
be analyzed and
responded to in real-time
Analytics software needs to be positioned
closer to the edge and should support real-
time streaming analytics. Traditional IT
analytics software (such as relational
databases or even Hadoop), are better
suited to batch-level analytics that
occur after the fact


Comparing IoT Architectures
In the past several years, architectural standards and frameworks have
emerged to address the challenge of designing massive-scale IoT
networks.
The foundational concept in all these architectures is supporting data,
process, and the functions that endpoint devices perform.
Two of the best-known architectures are those supported by
 oneM2M
 IoT World Forum (IoTWF)


The oneM2M IoT Standardized
Architecture


Architecture
The oneM2M architecture divides IoT functions into three major domains:
the application layer, the services layer, and the network layer. While this
architecture may seem simple and somewhat generic at first glance, it is very
rich and promotes interoperability through IT-friendly APIs and supports a
wide range of IoT technologies.
 Application Layer
 Services Layer
 Network Layer


Architecture
Application Layer:
The oneM2M architecture gives major attention to connectivity between
devices and their applications. This domain includes the application-layer
protocols and attempts to standardize northbound API definitions for
interaction with business intelligence (BI) systems. Applications tend to be
industry-specific and have their own sets of data models, and thus they are
shown as vertical entities.
Service Layer:
This layer is shown as a horizontal framework across the vertical industry
applications. At this layer, horizontal modules include the physical network
that the IoT applications run on, the underlying management protocols, and
the hardware. Examples include backhaul communications via cellular,
MPLS networks, VPNs, and so on.


Architecture
Service Layer, Cont’d:
Riding on top is the common services layer. This conceptual layer adds APIs
and middleware supporting third-party services and applications. One of the
stated goals of oneM2M is to “develop technical specifications which address
the need for a common M2M Service Layer that can be readily embedded
within various hardware and software nodes, and rely upon connecting the
myriad of devices in the field area network to M2M application servers,
which typically reside in a cloud or data center.” A critical objective of
oneM2M is to attract and actively involve organizations from M2M-related
business domains, including telematics and intelligent transportation,
healthcare, utility, industrial automation, and smart home applications, to
name just a few.


Architecture
Network Layer:
This is the communication domain for the IoT devices and
endpoints. It includes the devices themselves and the
communications network that links them. Embodiments of this
communications infrastructure include wireless mesh technologies,
such as IEEE 802.15.4, and wireless point-to-multipoint systems,
such as IEEE 801.11ah. Also included are wired device connections,
such as IEEE 1901 power line communications.


The IoT World Forum (IoTWF)
Standardized Architecture
In 2014 the IoTWF architectural committee (led by Cisco, IBM, Rockwell
Automation, and others) published a seven-layer IoT architectural reference
model. While various IoT reference models exist, the one put forth by the
IoT World Forum offers a clean, simplified perspective on IoT and includes
edge computing, data storage, and access. It provides a succinct way of
visualizing IoT from a technical perspective. Each of the seven layers is
broken down into specific functions, and security encompasses the entire
model.




The IoT Reference Model defines a set of levels with control flowing from the
center (this could be either a cloud service or a dedicated data center), to the
edge, which includes sensors, devices, machines, and other types of intelligent
end nodes. In general, data travels up the stack, originating from the edge,
and goes northbound to the center. Using this reference model, we are able to
achieve the following:
 Decompose the IoT problem into smaller parts.
 Identify different technologies at each layer and how they relate to one
another
 Define a system in which different parts can be provided by different
vendors
 Have a process of defining interfaces that leads to interoperability
 Define a tiered security model that is enforced at the transition points
between levels.


Layer 1: Physical Devices & Controllers Layer;
The first layer of the IoT Reference Model is the physical devices and
controllers layer. This layer is home to the “things” in the Internet of Things,
including the various endpoint devices and sensors that send and receive
information. The size of these “things” can range from almost microscopic
sensors to giant machines in a factory. Their primary function is generating
data and being capable of being queried and/or controlled over a network.
Layer 2: Connectivity Layer;
In the second layer of the IoT Reference Model, the focus is on connectivity.
The most important function of this IoT layer is the reliable and timely
transmission of data. More specifically, this includes transmissions between
Layer 1 devices and the network and between the network and information
processing that occurs at Layer 3 (the edge computing layer).


Layer 2: Connectivity
Layer, Cont’d;
The connectivity layer
encompasses all networking
elements of IoT and doesn’t
really distinguish between the
last-mile network (the network
between the sensor/endpoint
and the IoT gateway), gateway,
and backhaul networks.


Layer 3: Edge Computing
Layer;
Edge computing is the role of Layer
3. Edge computing is often referred
to as the “fog”. At this layer, the
emphasis is on data reduction and
converting network data flows into
information that is ready for storage
and processing by higher layers. One
of the basic principles of this
reference model is that information
processing is initiated as early and as
close to the edge of the network as
possible.




IT and OT Responsibilities in the IoT
Reference Model


A Simplified IoT Architecture


Expanded View of A Simplified IoT
Architecture


The End

Lec2.pptx

More Related Content

What's hot (20)

Similar to Lec2.pptx (20)

Recently uploaded (20)

Lec2.pptx