2 - Domain Specific loTs

This Chapter Covers

lol;' Applicatins for:

• Home
• Cities
• Environment
• Energy Systems
• Retail
• Logistics
• Industry
• Agriculture
• Health & Lifestyle
54 Domain Specific loTs

2.1 Introduction
1:11~ Internet of Things (loT) applications span a wide, range of domains including ~ut not
retail, logistics, industfY, agnculture
limited to) homes, cities, environment, energy systems
and health. This chapter provides an overview of various types of IoT applications for each
s
of these domains. In the later chapters the reader is guided through detailed implementation
of several of these applications.

2.2 Home Automation

2.2.1 Smart Lighting
t
Smart lighting for homes helps in saving energy by adapting the lighting to the ambien
conditions and switching on/off or dimming the lights when needed. Key enabling technologies
for smart lighting include solid state lighting (such as LED lights) and IP-enabled lights. For
solid state lighting solutions both spectral and temporal characteristics can be configured
to adapt illumination to various needs. Smart lighting solutions for home achieve energy
savings by sensing the human movements and their environments and controlling the lights
accordingly. Wireless-enabled and Internet connected lights can be control led remotely
from IoT applications such as a mobile or web application. Smart lights with sensors for
occupancy, temperature, lux level, etc., can be configured to adapt the lighting (by changing
the light intensity, color, etc.) based on the ambient conditions sensed, in order to provide a
good ambiance. In [19] controllable LED lighting system is presented that is embedded with
ambient intelligence gathered from a distributed smart wireless sensor networ k to optimize
and control the lighting system to be more efficient and user-oriented. A solid state lighting
model is described in [20] and implemented on a wireless sensor networ k that provides
services for sensing illumination changes and dynamically adjusting lumina ry briohtn ess
according to user preferences. In chapter-9 we provide a case study on a smart lighting
system.

2.2.2 Smart Appliances

Modem homes have a nu~ber of appliances such as TVs, refrigerators, music systems,
washer/d~ers, etc. Managmg and controlling these appliances can be cumbe rsome with
each appliance ~aving its own controls or remote controls. Smart applian ces make the
management easier and also provide status information to the users remote ly For example
: ; :i::~ ::er: !: :n be controlled remotely and notify when the w~shing/drying'
• ermostats allow controlling the tempera ture remote ly and can
learn the user preferences [22] S .
(using RFID tags) and send u d t mart threfngerators can keep track of the items stored
p a es to e users when an item is low on stock. Smart

Bahga & Madisetti, © 2015

2.2 Home Automation 55

TVs allows users to search and stream videos and movies from the Internet on a local
storage drive, search TV channel schedules and fetch news, weather updates and other
content from the Intemet. OpenRemote [21] is an open source automation platform for
homes and buildings. OpenRemote is platform agnostic and works with standard hardware.
With OpenRemote, users can control various appliances using mobile or web applications.
OpenRemote comprises of three components - a Controller that manages scheduling and
runtime integration between devices, a Designer that allows you to create both configurations
for the controller and create user interface designs and Control Panels that allow you to
interact with devices and control them. An IoT-based appliance control system for smart
homes is described in [23], that uses a smart central controller to set up a wireless sensor
and actuator network and control modules for appliances.

Smart Lighting Smart Appliances Smoke/Fire Detector

~ \

Smart Thermostat Intrusion Detection

Figure 2.1: Applications of IoT for homes

Internet of Things - A Hands-On Approach

i
t
r·

56 Domain Specific loTs

2.2.3 Intrusion Detection

Home intrusion detection systems use security cameras and sensors (such as PIR sensors and
door sensors) to detect intrusions and raise alerts. Alerts can be in the form of an SMS or
an email sent to the user. Advanced systems can even send detailed alerts such as an image
grab or a short video clip sent as an email attachment. A cloud controlled intrusion detection
system is described in [24] that uses location-aware services, where the geo-location of each
node of a home automation system is independently detected and stored in the cloud. In
the event of intrusions, the cloud services alert the accurate neighbors (who are using the
home automation system) or local police. In [25], an intrusion detection system based on
UPnP technology is described. The system uses image processing to recognize the intrusion
and extract the intrusion subject and generate Universal-Plug-and-Play (UPnP-based) instant
messaging for alerts. In chapter-9 we provide a case study on an intrusion detection system.

2.2.4 Smoke/Gas Detectors

Smoke detectors are installed in homes and buildings to detect smoke that is typically an early
sign of fire. Smoke detectors use optical detection, ionization or air sampling techniques to
detect smoke. Alerts raised by smoke detectors can be in the form of signals to a fire alann
system. Gas detectors can detect the presence of harmful gases such as carbon monoxide
(CO), liquid petroleum gas (LPG), etc. A smart smoke/gas detector [22] can raise alerts in
human voice describing where the problem is, send or an SMS or email to the user or the
local fire safety department and provide visual feedback on its status (healthy, battery-low,
etc.). In [26], the design of a system that detects gas leakage and smoke and gives visual
level indication, is described.

2.3 Cities
2.3.1 Smart Parking
Finding a parking space during rush hours in crowded cities can be time consuming and
frustrating. Furthermore, drivers blindly searching for parking spaces create additional traffic
congestion. Smart parking make the search for parking space easier and convenient for
drivers. Smart parking are powered by IoT systems that detect the number of empty parking
slots and send the information over the Internet to smart parking application back-ends.
These applications can be accessed by the drivers from smart-phones, tablets and in-car
navigation systems. In smart parking, sensors are used for each parking slot, to detect
whether the slot is empty or occupied. This information is aggregated by a local controller
and then sent over the Internet to the database. In [29], Polycarpou et. al. describe latest
trends in parking availability monitoring, parking reservation and dynamic pricing schemes.

Bahga & Madisetti, © 2015

2.3 Cities 57

Design and implementation of a prototype smart parking system based on wireless sensor
netw_ork technology with features like remote parking monitoring, automated guidance, and
parking reservation mechan ism is described in [30]. In chapter-9 we provide a case study
on a smart parking system.

Str~ctural Healt~ Monitoring Surveillance Emergency Services

(Fire, Gas Leak, Water Leakage detection)

-
Smart Lighting Smart Roads-~ ~J Smart ~rking_ ~ .

Figure 2.2: Applications of IoT for cities

2.3.2 Smart Lighting

Smart lighting systems for roads, parks and buildings can help in saving energy. According
to an IEA report [27], lighting is responsible for 19% of global electricity use and around
6% of global greenhouse gas emissions. Smart lighting allows lighting to be dynamically
controlled and also adaptive to the ambient conditions. Smart lights connected to the Internet
can be controlled remotely to configure lighting schedules and lighting intensity. Custom
lighting configurations can be set for different situations such as a foggy day, a festival,
e
etc. Smart lights equipped with sensors can communicate with other lights and exchang
information on the sensed ambient conditions to adapt the lighting. Castro et. al. [28]
describe the need for smart lighting system in smart cities, smart lighting features and how
to develop interoperable smart lighting solutions.

Internet of Things - A Hands-On Approach

 -- -- •- -1 1 ~

Domain Specific loTs

58

2.3.3 Smart Roads el time

equ ipp ed with sen sors can pro vid e information on driving conditions, trav
Sm art roads Such
alerts in cas e of poo r driv ing con diti ons, traffic congestions and accidents.
estimates and s •. Inf~nnation
ds safer and help in reducing traffic jam
information can help in making the roa ns and
the roa ds can be com mu nic ated via Inte rne t to clo ud- bas ed applicat10
sen sed from l ], a
diss emi nate d to the driv ers who subscribe to suc h applications. In [3
social media and safety
and aut ono mo us sys tem of sen sor net wo rk nod es for imp rov ing driv ing
distributed with a
ds in pro pos ed. The sys tem can pro vid e the driv ers and pas sen ger s
on public roa dozen
of the roa d situ atio n a few hun dre d me ters ahe ad of the m or a few
consistent view
ential dangers ear ly enough.
miles away, so that they can react to pot

g
2.3.4 Structural Health Monitorin
r the vibration
s use a net wo rk of sen sor s to mo nito
Structural Health Monitoring system these sensors
and buildings. The dat a collected from
levels in the structures such as bridges ct
ss the hea lth of the stru ctur es. By analyzing the dat a it is pos sibl e to dete
is analyzed to asse the
han ical brea kdo wns , loca te the dam ages to a structure and also calculate
cracks and mec the
of the stru ctur e. Usi ng suc h sys tem s, adv anc e war nin gs can be given in
remaining life structural
inen t fail ure of the stru ctur e. An environmental effect removal bas ed
case of imm ce structural
mo nito ring sch eme in an IoT env ironment is pro pos ed in [32]. Sin
health powered by
mon itoring syst ems use larg e num ber of wireless sen sor nodes wh ich are
health to harvestino0
itio nal batt erie s, rese arch ers are exp loring energy harvesting tech nol ogi es
trad
rations, sunlight, and win d [33, 34].
ambient energy, such as mechanical vib

2.3.5 Surveillance
d to ensure
lanc e of ~nfr as~ructu _re, pub lic tran sport and events in citi es is req uire
Surveil e num ber of
ance infrastructure com pris ing of larg
s~fet_y and secunty. City wide surveill d. The video
eo surveillance cam era s can be cre ate
dist nbu ted and ~ntemet connected vid age solutions.
aggregated in cloud-based scalable stor
feeds from sur:eillance c~e ras ~an ~e t terns or spec1'fic
ed vid eo ana lyti cs apphca t10n s can be developed to search 1or
c
pa
Cloud-bas
f . is des crib ed that
] a smart city sur vei llan ce sys tem
events rom the video feeds. In [35
leverages benefits of cloud data stores.

2.3.6 Emergency Response

t · •• . •
d for monitoring the critical infrastr uc ure m cities such as bmldmgs,
1oT systems can be usees
gas and water pipelin bl• tr ions. loT s stem s for fire
. , pu ic ansport and pow er· substat • y . . . .
and wat er leak age dete ctio n can hel ting aler ts and mm 1mtzmg
detect10n, gas P m gen era

Bahga & Madisetti, © 2015

2·_4 Environment 59

their effects on the critical infrastructure. IoT systems for critical infrastructure monitoring
enable aggregation and sharing of information collected from large number of sensors. Using
cloud-based architectures, multi-modal information such as sensor data, audio, video feeds
can be analyzed in near real-time to detect adverse events. Response to alerts generated by
such systems can be in the form of alerts sent to the public, re-routing of traffic, evacuations
of the affected areas, etc. In [36] Attwood et. al. describe critical infrastructure response
framework for smart cities. A Traffic Management System for emergency services is
described in [37]. The system adapts by dynamically adjusting traffic lights, changing
related driving policies, recommending behavior change to drivers, and applying essential
security controls. Such systems can reduce the latency of emergency services for vehicles
such as ambulances and police cars while minimizing disruption of regular traffic.

2.4 Environment
2.4.1 Weather Monitoring
IoT-based weather monitoring systems can collect data from a number of sensor attached
(such as temperature, humidity, pressure, etc.) and send the data to cloud-based applications
and storage back-ends. The data collected in the cloud can then be analyzed and visualized
by cloud-based applications. Weather alerts can be sent to the subscribed users from such
applications. AirPi [38] is a weather and air quality monitoring kit capable of recording and
uploading information about temperature, humidity, air pressure, light levels, UV levels,
carbon monoxide, nitrogen dioxide and smoke level to the Internet. In [39], a pervasive
weather monitoring system is described that is integrated with buses to measure weather
variables like humidity, temperature and air quality during the bus path. In [40], a weather
monitoring system based on wireless sensor networks is described. In chapter-9 we provide
a case study on a weather monitoring system.
.
2.4.2 Air Pollution Monitoring
I, • ' ' •

• 1

IoTrbased air pollution monitoring systems can monitor emission of harmful gases (CO2, CO,
NO, N02 , etc.) by factories and automobiles using gaseous and meteorological sensors. The
collected data can be analyzed to make informed decisions on pollutions control approaches.
In.[41], a real-time air quality monitoring system is presented that·comprises of several
distributed monitoring stations that communicate via wireless with a back-end server using
machine-to-machine communication. In [42], an air pollution system is described that
integrates a single-chip microcontroller, several air pollution sensors, GPRS-Modem, and a
GPS module. In chapter-9 we provide a case study on an air pollution monitoring system.

Internet of Things - A Hands-On Approach

60 Domain Specific loTs

Air Pollution Monitorin g Noise Pollution Monitorin g

Weather Monitorin g

Forest Fire Detection River Flood Detection ·'

Figure 2.3: Applications of IoT for environment

2.4.3 Noise Pollution Monitoring

Due to growing urban development, noise levels in cities have increased and even become
to
alarmingly high in some cities. Noise pollution can cause health hazards for human s due
sleep disruption and stress. Noise pollution monitoring can help in genera ting noise maps
for cities. Urban noise maps can help the policy makers in urban plannin g and making
policies to control noise levels near residential areas, schools and parks. IoT based noise
at
pollution monitoring systems use a number of noise monitoring stations that are deployed
different places in a city. The data on noise levels from the stations is collect ed on servers
a
or in the cloud. The collected data is then aggregated to genera te noise maps. In [43],
noise mapping study for a city is presented which revealed that the city suffered from serious
the
noise pollution. In [44], the design of smart phone application is describ ed that allows
ed
users to continuously measure noise levels and send to a central server where all generat

Bahga & Madisetti, © 2015

2.5 Energy 61

information is aggregated and mapped to a meaningful noise visualization map.

2.4.4 Forest Fire Detection

Forest fires can cause damage to natural resources, property and human life. There can be
different causes offorest fires including lightening, human negligence, volcanic eruptions and
sparks from rock falls. Early detection of forest fires can help in minimizing the damage. IoT
based fore st fire detection systems use a number of monitoring nodes deployed at different
locations in a fore st. Each monitoring node collects measurements on ambient conditions
including temperature, humidity, light levels, etc. A system for early detection of forest
fires is described in [45] that provides early warning of a potential forest fire and estimates
the scale and intensity of the fire if it materializes. In [46], a forest fire detection system
based on wireless sensor networks in presented. The system uses multi-criteria detection
which is implemented by the artificial neural network (ANN). The ANN fuses sensing data
corresponding to multiple attributes of a fore st fire (such as temperature, humidity, infrared
and visible light) to detect forest fires.

2.4.5 River Floods Detection

River floods can cause extensive damage to the natural and human resources and human
life. River floods occur due to continuous rainfall which cause the river levels to rise and
flow rates to increase rapidly. Early warnings of floods can be given by monitoring the
water level and flow rate. IoT based river flood monitoring system use a number of sensor
nodes that monitor the water level (using ultrasonic sensors) and flow rate (using the flow
velocity sensors). Data from a number of such sensor nodes is aggregated in a server or in
the cloud. Monitoring applications raise alerts when rapid increase in water level and flow
rate is detected. In [47], a river flood monitoring system in described that measures river and
weather conditions through wireless sensor nodes equipped with different sensors. In [48], a
motes-based sensor network for river flood monitoring is described. The system includes a
water level monitoring module, network video recorder module, and data processing module
that provides flood information in the form of raw data, predicted data, and video feed.

2.5 Energy
2.5.1 ~ Smart Grids
Smart_Grid is a data communications network integrated with the electrical grid that c.ollec.ts
and analyzes data captured in near-real-time about power transmission, distribution, and
consumption. Smart Grid technology provides predictive information and recommendations
to· utilities, their suppliers, and their customers on how best to manage power.· Smart

Internet of Things - A Hands-On Approach

62 Domain Specific loTs

Grids collect data regarding electricity generation (centralized or distributed), co~su~pt~on

(instantaneous or predictive), storage (or conversion of energy into other forms), diS tnbution
and equipment health data. Smart grids use high-speed, fully integrated, two-way .
communication technologies for real-time information and power exchange. B~ usin~ loT
based sensing and measurement technologies, the health of equipment and th ~ integnty of
the grid can be evaluated. Smart meters can capture almost real-time consumptio_n, remotely
control the consumption of electricity and remotely switch off supply when requtred. Po~er
thefts can be prevented using smart metering. By analyzing the data on power generatto~,
transmission and consumption smart girds can improve efficiency throughout ~he elect~c
system. Storage collection and analysis of smarts grids data in the cloud can help in_ dy~anuc
optimization of system operations, maintenance, and planning. Cloud-based momtonng _of
smart grids data can improve energy usage levels via energy feedback to users coupl~d with
real-time pricing information. Real-time demand response and management strategies can
be used for lowering peak demand and overall load via appliance control and energy storage
mechanisms. Condition monitoring data collected from power generation and transmission
systems can help in detecting faults and predicting outages. In [49], application of IoT in
smart grid power transmission is described.

2.5.2 Renewable Energy Systems

Due to the variability in the output from renewable energy sources (such as solar and wind),
integrating them into the grid can cause grid stability and reliability problems. Variable
output produces local voltage swings that can impact power quality. Existing grids were
designed to handle power flows from centralized generation sources to the loads through
transmission and distribution lines. When distributed renewable energy sources are integrated
into the grid, they create power bi-directional power flows for which the grids were not
originally designed. IoT based system.s integrated with the transformers at the point of
interconnection measure the electrical variables and how much power is fed into the grid. To
ensure the grid stability, one solution is to simply cut off the overproduction. For wind energy
systems, closed-loop controls can be used to regulate the voltage at point of interconnection
which coordinate wind turbine outputs and provides reactive power support [52].

2.5.3 Prognostics
Energy systems (smart grids, power plants, wind turbine farms, for instance) have a large
number of critical components that must function correctly so that the systems can perform
their operations correctly. For example, a wind turbine has a number of critical components,
e.g., bearings, turning gears, for instance, that must be monitored carefully as wear and tear
in such critical components or sudden change in operating conditions of the machines can

Bahga & Madisetti, © 2015

2.5 Energy 63

Smart Grid

Renewable Energy Integration ........ Health

Prognostic , Management
..
~
,;.
- .: l ' -

Figure 2.4: Applications of IoT for energy systems

result in failures. In systems such as power grids, real-time information is collected using
specialized electrical sensors called Phasor Measurement Units (PMU) at the substations.
The information received from PMUs must be monitored in real-time for estimating the
state of the system and for predicting failures. Energy systems have thousands of sensors
that gather real-time maintenance data continuously for condition monitoring and failure
prediction purposes. IoT based prognostic real-time health management systems can predict
performance of machines or energy systems by analyzing the extent of deviation of a system
from its·normal operating profiles. Analyzing massive amounts of maintenance data collected
from sensors in energy systems and equipment can provide predictions for the impending
failures (potentially in real-time) so that their reliability and availability can be improved.
Prognostic health management systems have been developed for different energy systems.
OpenPDC [50] is a set of applications for processing of streaming time-series data collected
from Phasor Measurement Units (PMUs) in real-time. A generic framework for storage,
processing and analysis of massive machine maintenance data, collected from a large number

Internet of Things - A Hands-On Approach

64 Domain Specific loTs

of sensors embedded in industrial machines, in a cloud computing environment was proposed

in[51].

2.6 Retail
2.6.1 Inventory Management
Inventory management for retail has become increasingly important in the recent years with
the growing competition. While over-stocking of products can result in additional storage
expenses and risk (in case of perishables), under-stocking can lead to loss of revenue. loT
systems using Radio Frequency Identification (RFID) tags can help in inventory management
and maintaining the right inventory levels. RFID tags attached to the products allow them
to be tracked in real-time so that the inventory levels can be determined accurately and
products which are low on stock can be replenished. Tracking can be done using RFlD
readers attached to the retail store shelves or in the warehouse. loT systems enable remote
monitoring of inventory using the data collected by the RFID readers. In [53], an RFlD
data-based inventory management system for time-sensitive materials is described.

2.6.2 Smart Payments

Smart payment solutions such as contact-less payments powered by technologies such as
Near field communication (NFC) and Bluetooh. Near field communication (NFC) is a set of
standards for smart-phones and other devices to communicate with each other by bringing
them into proximity or by touching them. Customers can store the credit card information in
their NFC-enabled smart-phones and make payments by bringing the smart-phones near the
point of sale terminals. NFC maybe used in combination with Bluetooh, where NFC (which
offers low speeds) initiates initial pairing of devices to establish a Bluetooh connection while
the actual data transfer takes place over Bluetooh. The applications of NFC for contact-less
payments are described in [54, 55].

2.6.3 Smart Vending Machines

Smart vending machines connected to the Internet allow remote monitoring of inventory
levels, elastic pricing of products, promotions, and contact-less payments using NFC.
Smart-phone applications that communicate with smart vending machines allow user
preferences to be remembered and learned with time. When a user moves from one vending
machine to the other and pairs the smart-phone with the vending machine, a user specific
interface is presented. Users can save their preferences and favorite products. Sensors in
a smart vending machine monitor its operations and send the data to the cloud which can
be used for predictive maintenance. Smart vending machines can communicat e with other

Bahga & Madisetti, © 2015

2. 7 Logistics 65

vending machines in their vicinity and share their inventory levels so that the customers
can be routed to the nearest machine in case a product goes out of stock in a machine. For
perishable items, the smart vending machines can reduce the price as the expiry date nears.
New products can be recommended to the customers based on the purchase history and
preferences.

Inventory Monitoring Smart Shopping Applications

~, ,.Ll•
' i.
~J~·
~ ~~.
J3-
·~·~
;:.01~· :
-~
--

Smart Payment
~
Figure 2.5: Applications of IoT for retail

2.7 Logistics
2.7.1 Route Generation & Scheduling
Modem transportation systems are driven by data collected from multiple sources which
is processed to provide new services to the stakeholders. By collecting large amount of

Internet of Things - A Hands-On Approach

66 Domain Specific loTs

·
data f rom vanous sources and processing· the data i·nto useful information ' data-dn'ven
. . • uch as advanced route
. guidance. [62,
transportation systems can provide new services s
nd nd
63], dynamic vehicle routin [64], anticipating customer dema s for pickup a dehvery
d
. g · d h d ling systems can generate
problem, for instance. Route generation an sc e u . end-to-en
· comb'mat1on
routes using · of rou te pattems and transportation . modes and feasible. schedules
.
- b'l' f h'cles As the transportation network grows
. m size and
based on the ava11a 1Ity o ve i • . .
· the number of poss1'ble route combinations increases exponentially.
. IoT. based
complex1ty,
systems backed by the cloud can provide fast response to the route generation quenes and
can be scaled up to serve a large transportation network.

2.7.2 Fleet Tracking

Vehicle fleet tracking systems use GPS technology to track the locations of the vehicles in
real-time. Cloud-based fleet tracking systems can be scaled up on demand to handle large
number of vehicles. Alerts can be generated in case of deviations in planned routes. The
vehicle locations and routes data can be aggregated and analyzed for detecting bottlenecks
in the supply chain such as traffic congestions on routes, assignments an~ generation of
alternative routes, and supply chain optimization. In [58], a fleet tracking system for
commercial vehicles is described. The system can analyze messages sent from the vehicles
to identify unexpected incidents and discrepancies between actual and planned data, so that
remedial actions can be taken.

2. 7.3 Shipment Monitoring

Shipment monitoring solutions for transportation systems allow monitoring the conditions
inside containers. For exampl~, containers carrying fresh food produce can be monitored
to prevent spoilage of food. IoT based shipment monitoring systems use sensors such as
temperature, pressure, humidity, for instance, to monitor the conditions inside the containers
and send the data to the cloud, where it can be analyzed to detect food spoilage. The analysis
and interpretation of data on the environmental conditions in the container and food truck
positioning can enable more effective routing decisions in real time. Therefore, it is possible
to take remedial measures such as - the food that has a limited time budget before it gets
rotten can be re-routed to a closer destinations, alerts can be raised to the driver and the
distributor about the transit conditions, such as container temperature exceeding the allowed
limit, humidity levels going out of the allowed limit, for instance, and corrective actions
can be taken before the food gets damaged. A cloud-based framework for real-time fresh
food supply tracking and monitoring was proposed in [61]. For fragile products, vibration
levels during shipments can be tracked using accelerometer and gyroscope sensors attached
to IoT devices. In [59], a system for monitoring container integrity and operating conditions

Bahga & Madisetti, © 2015

2. 7 Logistics 67

Fleet Tracking
Shipment Monitoring
Remote Vehicle Diagnostics

Figure 2.6: Applications of IoT for logistics

is described. The system monitors the vibration patterns of a container and its contents
to reveal information related to its operating environment and integrity during transport,
handling and storage.

2.7.4 Remote Vehicle Diagnostics

Remote vehicle diagnostic systems can detect faults in the vehicles or warn of impending
faults. These diagnostic systems use on-board IoT devices for collecting data on vehicle
operation (such as speed, engine RPM, coolent temperature, fault code number) and status of
various vehicle sub-systems. Such data can be captured by integrating on-board diagnostic
systems with IoT devices using protocols such as CAN bus. Modem commercial vehicles
support on-board diagnostic (OBD) standards such as OBD-II. OBD systems provide
real-time data on the status of vehicle sub-systems and diagnostic trouble codes which
allow rapidly identifying the faults in the vehicle. IoT based vehicle diagnostic systems can
send the vehicle data to centralized servers or the cloud where it can be analyzed to generate
alerts and suggest remedial actions. In [60], a real-time online vehicle diagnostics and early
fault estimation system is described. The system makes use of on-board vehicle diagnostics

Internet of Things - A Hands-On Approach

l
•

68 Domain Specific loTs

device and expert system to achieve real-time vehicle diagnostics and fault warning.

2.8 Agriculture

2.8.1 Smart Irrigation

Smart irrigation systems can improve crop yields while saving water. Smart irrigation
systems use IoT devices with soil moisture sensors to determine the amount of moisture in
the soil and release the flow of water through the irrigation pipes only when the moisture
levels go below a predefined threshold. Smart irrigation systems also collect moisture level
measurements on a server or in the cloud where the collected data can be analyzed to plan
watering schedules. Cultivar's RainCloud [56] is a device for smart irrigation that uses water
valves, soil sensors and a WiFi enabled programmable computer.

Smart Irrigation - -- --
/

Green House Control

Figure 2.7: Applications of IoT for agriculture

Bahga & Madisetti, @2015

2.9 Industry 69

2.8.2 Green House Control

Green houses are structures with glass or plastic roofs that provide conducive environment for
growth of plants. The climatological conditions inside a green house can be monitored and
controlled to provide the best ~onditions for growth of plants. The temperature, humidity, soil
moisture, light and carbon dioxide levels are monitored using sensors and the climatological
conditions are controlled automatically using actuation devices (such as valves for releasing
water and switches for controlling fans). IoT systems play an important role in green
house control and help in improving productivity. The data collected from various sensors
is stored on centralized servers or in the cloud where analysis is perfonned to optimize
the control strategies and also correlate the productivity with different control strategies.
In [57], the design of a wireless sensing and control system for precision green house
management is described. The system uses wireless sensor network to monitor and control
the agricultural parameters like temperature and humidity in real time for better management
and maintenance of agricultural production.

2.9 Industry
2.9.1 Machine Diagnosis & Prognosis
Machine prognosis refers to predicting the perfonnance of a machine by analyzing the
data on the current operating conditions and how much deviations exist from the nonnal
operating conditions. Machine diagnosis refers to determining the cause of a machine
fault. IoT plays a major role in both prognosis and diagnosis of industrial machines.
Industrial machines have a large number of components that must function correctly for the
machine to perfonn its operations. Sensors in machines can monitor the operating conditions
such as (temperature and vibration levels). The sensor data measurements are done on
timescales of few milliseconds to few seconds, which leads to generation of massive amount
of data. IoT based systems integrated with cloud-based storage and analytics back-ends
can help in storage, collection and analysis of such massive scale machine sensor data.
A number of methods have been proposed for reliability analysis and fault prediction in
machines. Case-based reasoning (CBR) is a commonly used method that finds solutions to
new problems based on past experience. This past experience is organized and represented
as cases in a case-base. CBR is an effective technique for problem solving in the fields in
which it is hard to establish a quantitative mathematical model, such as machine diagnosis
and prognosis. Since for each machine, data from a very large number of sensors is collected,
using such high dimensional data for creation of case library reduces the case retrieval
efficiency. Therefore, data reduction and feature extraction methods are used to find the
representative set of features which have the same classification ability as the complete of

Internet of Things - A Hands-On Approach

70 Domain Specific loTs

Smart Lighting Indoor Air Quality Monitoring

Machine Diagnosis

~-==··"=~~=l

r::l -~'_,J.J

Prognostic Health Management Emergency Services

M2M Communication
(Fire, Gas Leak, Water Leakage detection)

Figure 2.8: Applications of IoT for industry

features. A CBR based machine fault diagnosis and prognosis approach is described in [51].
A survey on recent trends in machine diagnosis and prognosis algorithms is presented in [65].

2.9.2 Indoor Air Quality Monitoring

Monitoring indoor air quality in factories is important for health and safety of the workers.
Harmful and toxic gases such as carbon monoxide (CO), nitrogen monoxide (NO), Nitrogen
Dioxide (N02 ), etc., can cause serious health problems. IoT based gas monitoring systems
can help in monitoring the indoor air quality using various gas sensors. The indoor air
quality can vary for different locations. Wireless sensor networks based IoT devices can
identify the hazardous zones, so that corrective measures can be taken to ensure proper
ventilation. In [66] a hybrid sensor system for indoor air quality monitoring is presented,
which contains both stationary sensors (for accurate readings and calibration) and mobile

Bahga & Madisetti, © 2015

 ~-~ -----V•~~-=-----------

2.10 Health & Lifestyle 71

sensors (for coverage). In [67] a wireless solution for indoor air quality monitoring is
described that measures the environmental parameters like temperature, humidity, gaseous
pollutants, aerosol and particulate matter to determine the indoor air quality.

Fitness Monitoring

"c.- -

--
J ..

L•---· ~-- -

~~~
- - --~
m..,.=-,=------. - . -~~-=
-~•-~--a
....::.,,-=-- ......___,.""'5S

Vital Signs Monitoring

Figure 2.9: Applications of IoT for health

2.1 O Health & Lifestyle

2.10.1 Health & Fitness Monitoring
Wearable IoT devices that allow non-invasive and continuous monitoring of physiological
parameters can help in continuous health and fitness monitoring. These wearable devices
may can be in various forms such as belts and wrist-bands. The wearable devices form a

Internet of Things - A Hands-On Approach

 Domain Specific loTs
72

type of wireless sensor networks called body area networks in which the measurements from
a number of wearable devices are continuous sent to a master node (such as a smart-phone)
which then sends the data to a server or a cloud-based back-end for analysis and archiving.
Health-care providers can analyze the collected health-care data to determine any health
conditions or anomalies. Commonly uses body sensors include: body temperature, heart
rate, pulse oximeter oxygen saturation (SPo2), blood pressure, electrocardio gram (ECG),
movement (with accelerometers), and electroencephalogram (EEG). An ubiquitous mobiljty
approach for body sensor networks in health-care is proposed in [72]. In [73], a wearable
ubiquitous health-care monitoring system is presented that uses integrated electrocardiogram
(ECG), accelerometer and oxygen saturation (SpO2) sensors. Fitbit wristband [74] is a
wearable device that tracks steps, distance, and calories burned during the day and sleep
quality at night.

2.10.2 Wearable Electronics

Wearable electronics such as wearable gadgets (smart watches, smart glasses, wristbands,
etc.) and fashion electronics (with electronics integrated in clothing and accessories, (e.g.,
Google Glass or Moto 360 smart watch) provide various functions and features to assist
us in our daily activities and making us lead healthy ljfestyles. Smart watches that run
mobile operating systems (such as Android) provide enhanced functionality beyond just
timekeeping. With smart watches, the users can search the Internet, play audio/video files,
make calls (with or without paired mobile phones), play games and use various kinds of
mobile applications [68]. Smart glasses allows users to take photos and record videos, get
map directions, check flight status, and search the Internet by using voice commands [69].
Smart shoes monitor the walking or running speeds and jumps with the help of embedded
sensors and be paired with smart-phones to visualize the data [70]. Smart wristbands can
track the daily exercise and calories burnt [71 ].

Summary
In this chapters you learned about domain specific applications of Internet of Things (IoT).
For homes, IoT has several applications such as smart lighting that adapt the lighting to suit
the ambient conditions, smart appliances that can be remotely .monitored and controlled,
intrusion detection systems and smart smoke detectors. For cities, applications of IoT include
~mart ?arking systems that provide status updates on available slots, smart lighting that helps
m savmg energy, smart roads that provide information on driving conditions and structural
health monitoring systems. For environment, you learned about IoT applications including
weather monitoring, air and noise pollution, forest fire detection and river flood detection

Bahga & Madisetti, © 2015

2.10 Health & Lifestyle 73

systems. You learned about IoT applications for energy systems including smart grids, grid
integration of renewable energy sources and prognostic health management systems. For
retail domain, you learned about IoT applications such as inventory management, smart
payments and smart vending machines. For agriculture domain, you learned about smart
irrigation systems that help in saving water while enhancing productivity and green house
control systems. You learned about the industrial applications of IoT including machine
diagnosis and prognosis systems that help in pr~dicting faults and determining the cause
of faults and indoor air quality systems. You learned about IoT applications for health
and lifestyle such as health and fitness monitoring systems and wearable electronics. The
applications generate much value to the end users and also provide new revenue opportunities
to service and systems providers when integrated to rating, billing and financial applications.

Review Questions
1. Determine the IoT-levels for designing home automation IoT systems including smart
lighting and intrusion detection.
2. Determine the IoT-levels for designing structural health monitoring system.
3. Determine the various communication models that can be used for weather monitoring
system. Which is a more appropriate model for this system. Describe the pros and
cons.
4. Determine the types of data generated by a forest fire detection system? Describe
alternative approaches for storing the data. What type of analysis is required for forest
fire detection from the data collected?

Internet of Things - A Hands-On Approach