Chapter 2 IoT Architecture

Drivers Behind New Network Architectures

The key difference between IT and IoT is the data.While IT systems are mostly concerned with
reliable and continuoussupport of business applications such as email, web, databases,
CRMsystems, and so on, IoT is all about the data generated by sensors andhow that data is used.The
essence of IoT architectures thus involves how the data istransported, collected, analyzed, and
ultimately acted upon.

The IoT Architectural Drivers–

Scale
Security
Device and networks constrained by power
CPU, memory and link
The massive volume of data generated
Support for legacy devices
The need for data to be analyzed in real time

Scale

The scale of a typical IT network is on the order of several thousand devices—typically printers,
mobile wireless devices, laptops, servers, and so on.The traditional three-layer campus networking
model, supporting access,distribution, and core (with subarchitectures for WAN, Wi-Fi, data
center,etc.), is well understood.But now consider what happens when the scale of a network goes
from afew thousand endpoints to a few million. This kind of scale has only previously been seen by
the Tier 1 service providers.IoT introduces a model where an average-sized utility, factory,
transportation system, or city could easily be asked to support a network of this scale.Based on scale
requirements of this order, IPv6 is the natural foundation for the IoT network layer.

Security

One of the example is, targeted malicious attacks using vulnerabilities innetworked machines, such
as the outbreak of the Stuxnet worm, which specifically affected Siemens programmable logic
controller (PLC) systems.The frequency and impact of cyber attacks in recent years has increased
dramatically.Protecting corporate data from intrusion and theft is one of the mainfunctions of the IT
department.IT departments go to great lengths to protect servers, applications, and the network,
setting up defense-in-depth models with layers of security designed to protect the cyber crown
jewels of the corporation.However, despite all the efforts mustered to protect networks and data, the
hackers still find ways to penetrate trusted networks.

Refer notes ch2 by dr sasikala

https://www.coursehero.com/file/40674092/Ch-02pptx/

The IoT World Forum (IoTWF) Standardized Architecture :Layer 1-7

The discussion will be based on the seven layer architecture suggested by the Internet of Things
World Forum Steering Committee which is widely becoming popular according to current trends.
L1 layer (Physical Devices and Controllers)
The L1 layer consists of physical devices and controllers that might control multiple devices. These
are referred to as “Things” in IoT and they include of a wide range of devices that sends and receive
information. This information will include the status of the device and information detected by the
sensors, which is processed by a local controller and given for transmission using NFC. IoT
devices that are connected in the network are capable of A to D (Analog to digital) conversion when
required. These can generate data and can be queried or controlled over the net.

L2 layer (Connectivity)
L2 layer relays information between devices, across different networks and between L2 and lower
layer of L3. The important functions of L2 are:

 Communication between L1 devices.

 Reliable delivery across the network(s)
 Implementation of different protocols supported by the device
 Switching and routing
 Translation between protocols
 Security at the network level
 (Self-Learning) Networking Analytics
The device would be connected to internet using Wifi, Ethernet, 3G and other technologies. This
Connectivity layer enables the device to send data to cloud and vice versa.

L3 layer (Edge Computing)

The ‘Edge computing’ layer formats the data into information that can be understood by L4 layer.
The important function of L3 layer are:

 Data filtering, clean-up, aggregation & Validation evaluate if it is ok to be processed by L4.

 Formatting: Reformatting data for consistent higher-level processing
 Expanding/decoding
 Distillation/reduction: Compressing data to minimize the impact of data and traffic on the
network and higher-level processing systems
 Assessment: Event generation in case of any alerts.
In short, Edge computing layer formats the data to save in a database. It also receives the data from
internet and does filtering and normalization operations.

L4 layer (Data Accumulation)

As the sensor data keeps changing with time, L4 layer converts this data in motion to data in rest.
Data accumulation layer stores the data in an easy accessible format. The layer dramatically reduces
data through filtering and selective storing.

L5 layer (Data Abstraction)

Data abstraction layer or L5 layer takes the essential data that satisfies certain characteristics from
the stored data and supplies it to the application layer for processing. The layer focuses on rendering
data and its storage in ways that enable developing simpler, performance-enhanced applications.
L6 layer (Application)
Application layer or L6 layer processes data so that it will be accessible for every person or device
to coordinate with the activities. It links the software layer with physical layer. This layer interprets
data to generate reports for analytics and to apply business intelligence.

L7 layer ( Collaboration and Processes)

L7 or Collaboration and processes layer provides response or action to be taken against the data
provided. This action, for instance can even be an actuation of an electromechanical device upon
the instruction from the controller.
https://www.wemakeiot.com/internet-things-iot-architecture/

IT and OT Responsibilities in the IoT Reference Model

1 Introduction
The first major contribution of the IoT Architectural Reference Model (IoT ARM) is the IoT
Reference Model itself. Besides models, the IoT Reference Model provides the concepts and
definitions on which IoT architectures can be built.
IoT functional model – refer book ebook 1

The Core IoT Functional Stack

The IoT technology stack is nothing else than a range of technologies, standards and tools, which
lead from the simple connection of objects to the applications that use these connected things, the
data they gather and communicate and the different steps needed to power them.

https://www.i-scoop.eu/internet-of-things-iot/iot-technology-stack-devices-gateways-platforms/

This Thing Stack consists of three technology layers: sensors, microcontrollers and internet
connectivity, and service platforms.

 Layer One - Sensors are embedded in objects or the physical environment to capture
information and events for your company.
 Layer Two - Microcontrollers and internet connectivity share information captured by
sensors within your IoT objects and act based on this information to change the
environment.
 Layer Three - Through the aggregation and analysis of data, service platforms cater to your
customers. Service platforms also control your IoT product's end-to-end experience and
enable your customers to define system rules and update firmwar
Layer 1: Sensors

Sensors have been used for years in a number of different industry contexts like healthcare,
aviation, manufacturing and automotive. Now, sensors are so tiny and inexpensive they can be
embedded in all the devices you use personally and professionally. The sensor layer of the IoT tech
stack continues to expand as internet connected sensors are added to new products and services.
Besides incorporating sensors into your smartphones and tablets, your innovative competitors are
introducing new devices that contain numerous sensors. These sensors don't need to be connected
directly to the internet and they sync with your smartphone via Bluetooth LE to send captured data
to the cloud. For example, Sensordrone is a Kickstarter project that embeds many environmental
sensors into a small dongle. Some of the environmental sensors include color intensity, humidity,
gas, and pressure sensors. The dongle collects information about your physical environment and
sends this information to your smartphone or tablet via Bluetooth. This example demonstrates how
value can be created by new services that use sensors to gather new types of data from your
physical environment.
Another category of sensors are embedded in watches, contact lenses, wristbands, and fabrics.
Recently, Nokia patented a technique for magnetic tattoos that vibrate from incoming messages or
calls. This is possible by spray tattooing or stamping a material onto your skin. The material is then
paired with a smartphone or tablet, which lets you interact with the sensor. Microsoft and University
of Washington researchers are working on another way to connect sensors to your body through the
use of IoT contact lenses that provide real-time updates about the biochemical fluctuations in your
body. One use case is to help diabetes patients monitor their insulin levels while using these contact
lenses.

Layer 2: Micro-controllers and internet connectivity

The second layer in the Internet of Things technology stack allows for local storage, data processing
and internet connectivity. The Internet of Things needs internet connectivity to send collected data
to your cloud database. Because some sensors generate over 10,000 data points per second, it makes
sense to pre-process data locally before sending it to your cloud database. By analyzing, extracting
and summarizing your collected data before you send it to your cloud database, you reduce the
volume of unnecessary data you send to and store on your cloud database, saving you money on
data transfer and storage costs.
Your microcontroller is a small computer embedded within a chip and it helps your IoT device store
and pre-process collected data before it's synced to your cloud database. Your microcontroller
possesses a processor, a small amount of RAM to hold data, some kilobytes of EPROM or flash
memory to hold embedded software, and solid-date memory to cache data.
In some cases, your IoT device may need to use programmable microcontrollers to take action and
turn something on or off. In most cases, these decisions are made via a cloud application, but it
makes sense to use programmable microcontrollers when a sensor detects something that could
affect the health and safety of your end customers.
The main and most important capability of this layer is networking, which is either wireless or
wired. If a device is stationary and can access an external power source, a wired network is
sufficient, but a wired network doesn't make sense for many IoT use cases because physical cables
are needed to connect to the network. WiFi, wireless modems, and wireless mesh networks are the
most common ways IoT devices are connected to the internet.
If you plan to manufacture an IoT device you must keep in mind dependencies for your use cases. Is
your device mobile or fixed? Does your device need a battery or is it connected to a fixed power
supply? How much data do you need to transfer to your cloud database per hour? Should your
device's connectivity be episodic or continuous?
Devices you use to track your health and fitness while bicycling, running, and exercising store data
while you're active, and these devices use episodic connectivity. Your device then syncs with the
cloud when it's close to your smartphone or tablet. Compare this to the continuous connectivity
needed by Amazon Echo's voice based digital assistant who is always listening for your commands,
fetching answers from the internet the instant you ask a question. Depending on your IoT product's
use cases, you may need continuous connectivity.
When you research Think Stack vendors, you'll notice a wide range of different networking
protocols, hardware, software, and architectures are used to build IoT products. Due to the variation
in use cases and environments, you have many choices when it comes to adding networking and
computing capabilities your IoT device. While some vendors focus more on hardware components,
other vendors provide a system of integrated software and hardware. Sometimes, IoT software
solutions spill into the third layer of the Thing Stack, which is referred to as the service platform.

Layer 3: Service platforms

The first two layers for the Thing Stack embed sensor and microcomputers in your IoT device, but
your IoT product profits from the service platform layer. This layer delivers value to your customers
by automating processes and delivering rich data analytics. Your cloud application combines data
collected from numerous IoT sensors with your (or your customers) other business data to produce
insights that generate business value.
It's important for your service platform to create a feedback loop between your IoT devices and you
device management software, so you and your customers can upgrade, monitor, and maintain the
firmware on each the device. In most cases, service platforms operate on cloud infrastructure and
utilize a multi-tenant software architecture to deliver a seamless software-as-a-service (SaaS)
experience.
The convergence between our digital and physical worlds stresses your IT operations by increasing
demand for data management, storage, tagging and analysis. It's in your company's best interests to
build your IoT service platform on robust cloud infrastructure, so you can scale infinitely as your
business grows with your new IoT product.
While "software is eating the world" as Marc Andreessen said, consumers and companies still
purchase a lot of physical things. If you build a robust IoT service platform, you obtain insights
about how your customers use your IoT products and services. With a great IoT service platform
you can manage post-transaction relationships in new and engaging ways.
Because service platforms store and make decisions based on data collected from all types of IoT
devices, they are often considered the backbone of post-transactions relationships. What if you
reached out to customers who never powered on your IoT device and proactively on-boarded them?
What if you used anonymized data from customers getting the most value out of your product to
help other customers unlock more business value?
Now is the time to build your IoT product because sensors keep getting smaller and network
connectivity solutions keep getting better. Create a plan to engage with your customers and
facilitate goal-oriented post-transaction relationships, leading to new opportunities to turn a single
transaction into a strong relationship.

https://www.cognitiveclouds.com/insights/the-internet-of-things-iot-tech-stack-explained
Data Analytics in IoT
Internet of Things (IoT) analytics is a data analysis tool that assesses the wide range of data
collected from IoT devices. IoT analytics assesses vast quantities of data and produces useful
information from it.
IoT analytics are usually discussed in tandem with Industrial IoT (IIoT). Data is collected from a
wide range of sensors on manufacturing infrastructure, weather stations, smart meters, delivery
trucks, and all forms of machinery. IoT analytics can be applied to managing data centers and
applications handling retail and healthcare.
In many ways, IoT data is similar to big data. The key difference between the two is not just the
quantity of data but the range of sources it is obtained from. All this data has to be processed into
one comprehensible, single stream of data. Considering the several kinds of sources of information,
data integration becomes quite difficult, and this is where IoT analytics makes a difference, though
it can be tough to develop and implement.
Devices that power IoT Analytics
There are a wide range of IoT devices that collect data:

Wearables
Dedicated trackers such as Fitbit or other smartwatches have gone beyond tracking steps. You can
track your friends' fitness activities, compete with them, message, and even answer the phone by
connecting your devices through the Internet. This information is tracked by fitness companies,
enabling them to create customized packages if you sign up. This can include exercise routines,
diet, goals, and more. The newest smart watches even monitor heart rates and rhythms and have
accurately diagnosed heart problems in their wearers.

Smart Home
Smart homes have security systems you can access and control when you are away from home, to
appliances you can turn on and off with digital assistance. There is a wide range of devices that you
can incorporate into your home and a wide range of data that can be collected to assess usage
patterns, the efficacy of systems, and more.

Healthcare
Healthcare has a wide range of IoT devices. Bluetooth technology creates hearing aids, records
heart and blood pressure, and monitors pulse-based alarm systems that can call for help. This has
helped enhance healthcare to a large extent. The data collected is invaluable in terms of creating
newer and better technology.

Voice-Activated Everything
Digital assistants are a form of IoT devices. Alexa, Siri, and Google take notes, find information,
play music, order cabs, tell the weather, set alarms, and everything else. The internet regularly
updates these digital assistants to improve functionality. Their data helps companies tailor their
services for you, based on your everyday interaction with digital assistants.

How IoT Analytics Work and the Applications

With a wide range of devices, there is an endless stream of data in enormous quantities. IoT
analytics helps analyze this data across all connected devices without hardware or infrastructure. As
the needs of your organization change, computing power and data storage scales up or down
accordingly, ensuring your IoT analysis has the requisite capacity.
1. The first step is to collect data aggregated from a variety of sources, in a range of formats,
and at multiple frequencies.
2. This data is then processed with a wide range of external sources.
3. The information is then stored in a time-series for analysis.
4. The analysis can be done in multiple ways--with custom analysis systems, with standard
SQL queries, or with machine learning analysis techniques. The results can be used to make
a wide range of predictions.
5. With the information received, organizations can build several systems and applications to
ease business processes.

Business Use Cases of IoT Analytics

Smart agriculture: With IoT analytics, connected field machinery works based on information
derived from IoT analysis. Analysis factors include time, geographical location, weather, altitude,
and local environmental conditions. For example, irrigation systems can be optimized to deliver the
exact amount of water as rainfall predictions.
Regular restocking of supplies: Monitor inventories in real-time. A food vending company, with
connected machines, can have their machines request restocking based on the depletion of products.
This can be triggered when stocks in the machine reach a particular level.
Predictive maintenance: Varying infrastructure needs regular maintenance. With IoT analysis, pre-
set templates can help determine quality predictive maintenance models applied to specific needs.
For example, in long-distance transport vehicles with heating and cooling systems--IoT analytics
can determine when vehicles need an overhaul to ensure cargos are not damaged.
Process efficiency scoring: Every company works with a range of processes in place. IoT analytics
can measure the efficiency of these processes and make the necessary changes in them. Results
from IoT analytics can identify bottlenecks--both current and potential--and can increase
efficiencies.
https://www.tibco.com/reference-center/what-is-iot-analytics
Benefits of IoT analytics
IoT analytics offer several benefits for businesses that use them:
1. Visibility on the entire IoT network – IoT analytics enables businesses to oversee the
performance of their IoT network in real-time.
2. Fast identification and resolution of problems in business operations – Businesses can use
diagnostic analytic capabilities to quickly identify performance problems and use
prescriptive analytics to fix such problems.
3. Better asset utilization – Businesses can use IoT analytics to monitor the performance of
their assets, such as machinery, and tweak their utilization to ensure the long term health of
assets.
4. Cost optimization – IoT analytics help identify areas of cost reduction and steps to
implement to achieve such cost reduction.
5. Expansion into new markets – IoT analytics offer valuable insights on operations and
consumer behavior to ease expansion into new markets.
6. Improved product development – Businesses can study historical trends in product usage by
consumers to identify areas of improvement for future versions of their products.
7. Better customer experience – IoT analytics helps businesses identify customer problems in
real-time and act quickly to fix those problems, thus enhancing customer experience and
delight.

IoT Analytics: Challenges

Businesses may face challenges in using IoT analytics tools. Some of these challenges are –
1. Excessive data generation and storage requirements –The aggregate data generated by
thousands of IoT sensors are usually very large, thus making it expensive to manage and
store such data.
2. The complexity of data – Data from multiple IoT devices consists of different types, formats
and sizes, thus making it very complex and difficult to process and clean.
3. Security – Businesses, especially those dealing with consumer data, have to take various
security measures to protect the stored IoT network data against hacking attempts and leaks.
4. Inaccurate data – Faulty IoT devices lead to inaccurate measurements, thereby messing up
the analysis of such data. When you have faulty IoT devices at large, the insights offered by
IoT analytics tools become unreliable.
5. Building a competent data analysis team – Businesses need to hire data scientists and
analysts to run analytical techniques on the IoT data and derive actionable insights.
https://www.airtel.in/blog/business/iot-analytics-benefits-challenges-use-cases/

Network analytics in IoT and Control applications are pending

Cloud Computing
Cloud computing refers to the on-demand delivery of IT services/resources over the internet. On-
demand computing service over the internet is nothing but cloud computing. By using cloud
computing users can access the services from anywhere whenever they need.
Nowadays, a massive amount of data is generated every second around the globe. Businesses
collect and process that data from the people and get analytics to scale their business. When lots of
organizations access their data simultaneously on the remote servers in data centers, data traffic
might occur. Data traffic can cause some delay in accessing the data, lower bandwidth, etc. But
cloud computing technology alone is not effective enough to store and process massive amounts of
data and respond quickly.
For example, in the Tesla self-driving car, the sensor constantly monitors certain regions around the
car. If it detects an obstacle or pedestrian on its way, then the car must be stopped or move around
without hitting. When an obstacle is on its way, the data sent through the sensor must be processed
quickly and help the car to detect before it hits. A little delay in detection could be a major issue. To
overcome such challenges, edge computing and fog computing are introduced.

Edge Computing
Computation takes place at the edge of a device’s network, which is known as edge computing.
That means a computer is connected with the network of the device, which processes the data and
sends the data to the cloud in real-time. That computer is known as “edge computer” or “edge
node”.
With this technology, data is processed and transmitted to the devices instantly. Yet, edge nodes
transmit all the data captured or generated by the device regardless of the importance of the data.
Example of Edge computing:

 Autonomous vehicle edge computing devices collect data from cameras and sensors on the
vehicle, process it, and make decisions in milliseconds, such as self-parking cars.
 In order to accurately assess a patient’s condition and foresee treatments, data is processed
from a variety of edge devices connected to sensors and monitors.
Here Fog Computing was introduced and becomes an ideal solution.

Fog Computing
Fog computing is an extension of cloud computing. It is a layer in between the edge and the cloud.
When edge computers send huge amounts of data to the cloud, fog nodes receive the data and
analyze what’s important. Then the fog nodes transfer the important data to the cloud to be stored
and delete the unimportant data or keep them with themselves for further analysis. In this way, fog
computing saves a lot of space in the cloud and transfers important data quickly.

Fog computing is a decentralized computing infrastructure or process in which computing resources

are located between the data source and the cloud or any other data center. Fog computing is a
paradigm that provides services to user requests at the edge networks. The devices at the fog layer
usually perform operations related to networking such as routers, gateways, bridges, and hubs.
Researchers envision these devices to be capable of performing both computational and networking
operations, simultaneously. Although these devices are resource-constrained compared to the cloud
servers, the geological spread and the decentralized nature help in offering reliable services with
coverage over a wide area. Fog computing is the physical location of the devices, which are much
closer to the users than the cloud servers.

Difference Between Edge Computing and Fog Computing

S.NO. EDGE COMPUTING FOG COMPUTING
S.NO. EDGE COMPUTING FOG COMPUTING

Highly scalable when compared to edge

01. Less scalable than fog computing.
computing.
02. Billions of nodes are present. Millions of nodes are present.
Nodes are installed far away from the Nodes in this computing are installed closer to the
03.
cloud. cloud(remote database where data is stored).
Edge computing is a subdivision of fog Fog computing is a subdivision of cloud
04.
computing. computing.
The bandwidth requirement is very low. The bandwidth requirement is high. Data
05. Because data comes from the edge originating from edge nodes is transferred to the
nodes themselves. cloud.
06. Operational cost is higher. Operational cost is comparatively lower.
High privacy. Attacks on data are very
07. The probability of data attacks is higher.
low.
Edge devices are the inclusion of the
08. Fog is an extended layer of cloud.
IoT devices or client’s network.
The power consumption of nodes filter important
The power consumption of nodes is information from the massive amount of data
09.
low. collected from the device and saves it in the filter
high.
Edge computing helps devices to get Fog computing helps in filtering important
faster results by processing the data information from the massive amount of data
10.
simultaneously received from the collected from the device and saves it in the cloud
devices. by sending the filtered data.

https://www.geeksforgeeks.org/difference-between-edge-computing-and-fog-computing/
 Feature Cloud Computing Fog Computing
Cloud computing has high
Latency latency compared to fog Fog computing has low latency
computing
Cloud Computing does not
provide any reduction in data Fog Computing reduces the amount of
Capacity
while sending or transforming data sent to cloud computing.
data
Response time of the system is
Responsiveness Response time of the system is high.
low.
Cloud computing has less
Security security compared to Fog Fog computing has high Security.
Computing
Access speed is high depending High even more compared to Cloud
Speed
on the VM connectivity. Computing.
Multiple data sources can be Multiple Data sources and devices can
Data Integration
integrated. be integrated.
In cloud computing mobility is
Mobility Mobility is supported in fog computing.
Limited.
Partially Supported in Cloud
Location Awareness Supported in fog computing.
computing.
Number of Server Cloud computing has Few Fog computing has Large number of
Nodes number of server nodes. server nodes.
Geographical
It is centralized. It is decentralized and distributed.
Distribution
Services provided within the Services provided at the edge of the local
Location of service
internet. network.
Specific data center building Outdoor (streets,base stations, etc.) or
Working environment
with air conditioning systems indoor (houses, cafes, etc.)
Wireless communication: WLAN, WiFi,
Communication mode IP network 3G, 4G, ZigBee, etc. or wired
communication (part of the IP networks)
Dependence on the
Requires strong network core. Can also work in Weak network core.
quality of core network

https://www.geeksforgeeks.org/difference-between-cloud-computing-and-fog-
computing/?ref=rp