Module 5 (6 Hours)

IoT privacy, security and vulnerabilities solutions, vulnerabilities, security requirements,

threat analysis, security tomography, layered attacker model, Identitymanagement, access
control, secure message communication.

Smart and Connected Cities-An IoT Strategy for Smarter Cities-Vertical IoT Needsfor
Smarter Cities, Global vs. Siloed Strategies-Smart City IoT Architecture-StreetLayer,
City Layer, Data Center Layer, Services Layer- Smart City Security Architecture - Smart
City Use-Case Examples — Street lighting, smart parking, smart traffic and air pollution
monitoring

Introduction:

• International organizations are making a number of efforts towards ensuring that IoT
design must ensure trust, data security and privacy.
• Trust in IoT context means dependability, accuracy, quality of data from multiple
sources for the intended applications and services.
• An organization, Open Trust Alliance established IoT Trustworthy Group (ITWG) for
recognizing the priority from the onset of product development and addressing
holistically.
• Security is important. For example, consider the ATM messages. They should
communicate on Internet securely. A smart city security is also important. The city deploys
smart health, public safety, transport, and deploys IoT and smart home applications and
services.
• Privacy is important. The video clips communicate on the Internet in a smart home
securityapplication.
• Message is a string that represents data or client-request or server-response which
communicates between sender and receiver objects.
• Hash
o Hash refers to a collection or bundle which gives an irreversible result after
many operations on data and the operations are just one way.
o {For example, when wheat crop is ripe and cut, hashing process separates out the
grains which are used for consumptionand the resultant waste is discarded. When data such as
user ID and password needs secret communication for the purpose of authentication,
then it is communicated after a set of standard operations on the usage of an
algorithm, called secure hash algorithm.}
o The algorithm generates a fixed size, say, 128, or 256-bit value using a secret key.
Only the hash value communicates. Receiver-end retrieves the hash value, and
compares that with stored hash value. If both are equal then the sender message is
authenticated.

• Digest is a process which gives the irreversible result involving many operations. A
standard algorithm called MD5 (Message Digest 5) is also used for digest, similar to the
hash value. Receiver-end stores the digest value expected to be obtained after the MD5
operations, and compares that with received value. If both are equal then the sender message
is authenticated.
• Encryption is a process of generating new data using a secret key known only to a receiver.
Before sending the encrypted data, the sender and receiver, both identify each other and
know the key that will be used by them. The encryption uses a 128, 192 or 256- bit key for
encrypting the data.

• Decryption is a process which retrieves the data from the encrypted data.

• Use Case

o Use case means a list of event steps or actions which define the interactions
between two ends, in which one is playing the role and other is the system.
o The steps accomplish a task or goal or mission. One end is called actor in
Unified Modelling Language (UML), while other end is the system.
o Use Case is a software engineering term.
o For example, an API is playing the role of obtaining inputs (events) and
generation of outputs which interact with the system such as a web server or
web API or service or web application using a callback () function as per
the output. Use cases define the required behaviour of software under
development. Use cases describe the details of usages of software and its normal
behaviour.

• Misuse Case
o Misuse case can be understood as reverse sense of Use case.
o Misuse case defines the behaviour which is not required from the software under
development.
o A Misuse case defines the behaviour which should not happen. This in turn
specifies the threats also.
o Misuse case gives information and renders help in identifying the requirement
of new Use cases for prevention of attack and find out what should not happen.

• Layer means a stage during a set of actions at which the action is taken as per the
specific protocol or method and then the result passes to the next layer until the set of
actions completes. A design using the layers’ model enables representation of a set of
systematic actions which are followed sequentially for accomplishing a task.

• Sublayer is a layer consisting of various sublayers in a model to provide set of actions

sequentially taking place at the layer.

• Firewall is a software interface, which interconnects networks with differing trusts, and
is immune to penetration and provides perimeter defence. It functions as a choke point
for controlling and monitoring. It does auditing and provides controlled accesses. It
allows only authorised traffic and imposes restrictions on network services. It can raise
alarms for abnormal behaviours.
Q) Explain the requirements of privacy and security, vulnerabilities from threats and
threat analysis in IoT

Privacy

• Message privacy means that the message should not reach into the hands of the
unrelated entities.
• When data or messages communicate from the things (device platforms), those are
meant only for the applications or services and for targeted goals only.
• Privacy also means no interference or disturbance from other.
• Consider an example of messages from embedded devices in an automobile using the
Internet to an automobile service centre. Privacy means the messages reach only the
centre and used only by the services of the centre. Another automobile company on
whose hands the data falls may face serious business consequences.
• IoT necessarily need privacy policy.
• A privacy policy needs to determine that ‘how much of the IoT devices data and which
data need absolute privacy and which need limited privacy’.
• Company authorities need the support for accessing the data which may be private for
individuals.
• The authorities also need to respect the individual customer needs of privacy and
understand that privacy is a legitimate human need. Privacy policy vendors should take
privacy seriously. They must respect their customers enough to understand that the
privacy is a legitimate human need.

Vulnerabilities of IoT

• Vulnerability means weak without complete protection, weakness to defend oneself or

can be easily influenced from surrounding unwanted things from itself.
• An IoT security article describes that there are many vulnerabilities, due to participation
of the number of layers, hardware sublayers and software in applications and services.
• The nature of IoT also varies. For example, sensors, machines, automobiles, wearables,
and so on. Each faces different kind of vulnerabilities and has complex security and
privacy issues.
• IoT network can be vulnerable to eavesdropping. Eavesdropper creates security issues.
• An eavesdropper, say E, listens to the messages and commands in the network during
communication and obtains confidential messages. A server at E sends fake commands
which a server S for the devices data assumes that are from the devices or applications.
• Open Web Application Security Project (OWASP) has undertaken the associated
security issues of IoT for the purpose of helping developers, manufacturers and
consumers.
• OWASP is open source and has free to use licensing policy. Project is community-
model based software development initiative. A community model is making collective
efforts and initiative by universities, organisations and institutions in an open source
project.
• OWASP has undertaken a number of security related subprojects, such as ones for
defining, the ‘Top Vulnerabilities’, ‘Attack Surface Areas’ and Testing Guides’.
• OWASP has identified top ten vulnerabilities in IoT applications/services as
follows:
• Insecure web interface
• Insufficient authentication or authorisation
• Insecure network services
• Lack of transport encryption/integrity verification
• Privacy concerns
• Insecure cloud interface
• Insecure mobile interface
• Insufficient security configurability
• Insecure software or firmware
• Poor physical security

Security Requirements
• IoT reference architecture means a guide for one or more concrete architects. IoT
reference architecture is a set of three architectural views—functional, information, and
deployment and operational.
• Security is one of the functional groups (FG) of the functional view. FG for security
consists of security functions between the application and device.

• Security FG contains five sets of functions which are required for ensuring
security and privacy. Large number of devices, applications and services
communicate in IoT. Five functional components (FCs) of security are defined in IoT
reference architecture.
• Following are five functional components (FCs):
• Identity management (IdM)
• Authentications
• Authorisation
• Key exchange and management
• Trust and reputation
 Threat Analysis

• A threat-analysis tool first generates the threats and analyses a system for threat(s).
• Threat analysis means uncovering the security design flaws after specifying the stride
category, data flow diagram, elements between that the interactions occurring during
the stride, and processes which are activated for analysis.
• Stride means a regular or steady course, pace or striding means, passing over or across
in one long step (set of statements when considering a threat to a software component).
• Stride means taking a long step for dainty little steps.
• Figure 10.2 shows an example of tool usage for threat analysis during a web service
interaction between the application and web.

Q) Outline security tomography of large networks and layered attacker model

Security Tomography

• Computational tomography means a computing method of producing a three-

dimensional picture of the internal structures of an object, by observation and recording
of the differences in effects on passage of energy waves impinging on those structures.

• Computational security in complex set of networks utilizes the network tomography

procedures of identifying the network vulnerabilities. This enables design of efficient
attack strategies. A complex set of networks may be distributed or collaborative.

• Network tomography refers to the study of vulnerabilities and security aspects for
network monitoring in a complex system, such as WSNs, RFIDs or IoT networks and
allocating resources and ensuring network reliability and security.

• Monitoring of individual nodes is not fast acting and is also impractical.

• Network tomography helps in observing each network section (for example, a WSN
nodes network between two access points) and subsections.

• The security tomography means finding attack vulnerable sections/subsections from the
observations for behaviours using a finite number of objects or threats in a complex set
of subsystems.

Layered Attacker Model

• Figure 10.4 shows a layered attacker model and possible attacks on the layers

• Following are the suggested solutions for mitigating the attacks on the layers (OSI
modified six layers IoT Architecture).

1. Layer 1 Attacks Solution

• Solution depends on the devices used.

• For example, link-level provisioning of security uses—BT LE link level AES-CCM
128 authenticated encryption algorithm for confidentiality and authentication, and
ZigBee at link-level security using AES-CCM-128.

2. Layer 2 Attacks Solution

• Programming the network switches to prevent internal node attacks during use of
DHCP or Spanning Tree Protocol (STP).
• Additional controls may include ARP inspection, disabling unused ports and enforcing
effective security on VLAN’s (Virtual LAN) to prevent VLAN hopping.

• VLAN refers to a group of end stations with a common set of requirements, independent
of a physical location. VLANs have the same attributes as a physical LAN but allow
you to group end stations even if they are not located physically on the same LAN
segment.
• LWM2M OMA (Lightweight M2M open mobile Alliance) specification for device
gateway to the Internet has provisions for MAS for security, root key data store, and
devices and data authentication.

3. Layer 3 Attacks Solution

• Use of temper resistant router, use of packet filtering and controlling routing messages
and packets data between layers 3 and 4 through a firewall reduces the risks.

4. Layer 4 Attacks Solution

• Port scanning method is a solution which identifies the vulnerable port. A solution is
the opening of network ports and configuring effectively the firewall, and locking down
ports only to those required.
• Another solution is DTLS between layers 5 and 4. The DTLS has provisions for three
types of security services, viz. integrity, authentication and confidentiality. A solution
is included SASL (Simple Authentication and Security Layer) for security when using
the XMPP protocol.

5. Layers 5 and 6 Attacks Solution

• Above layer 4, we are looking primarily at application-level attacks which are results
of poor coding practices. Assume an attacker injects the SQL input to extract data from
the database (e.g. SELECT * from USERS). When the application fails to validate the
injection, the query extracts the data.
• Web applications/services can use HTTPS communication link. The features of S-
HTTP (Secure HTTP) are as follows:
o Application-level security (HTTP specific)
o Content privacy domain header
o Allows use of digital signatures and encryption, various encryption options
o Server-client negotiations
o Cryptographic scheme is a property assigned for the link
o Specific algorithm is the value assigned
o Direction specification is done, one-way or two-way security

• CISCO suggested layered framework provisions for following solutions:

• Layers 1–6: Role-based security
• Layers 1–4 Anti-temper and detection-based security
• Layers 1–6: Data protection and confidentiality
• Layers 1–6: IP protection
Q) Examine functions for source identity management, identity establishment, device
messages access control, message integrity, messages non-repudiation and availability in
IoT applications and services

• Source of message needs to specify an identity (ID) when sending the messages.
• The receiver can thus know that from where the messages have been received. Number
of ways exist for specifying identity (ID).
• The messages can be from several sensors, actuators and platforms and those may be
for several applications and services. ID management and establishment for IoT are
therefore basic requirements.
• A MAC address can specify identity of a computing device platform.
• However, the platform may connect several sensors and actuators. An application layer
may consist of number of applications and services.
• An URI (Universal Resource Identifier) can be used on the Internet. Many devices
however do not use the URI. An Object Identifier (OID) in IoT can have the following
identifiers:
• Types of things (for example, streetlight, vehicle, ATM, WSN, RFID)
• Class identifier, since it refers to a class (or type, or category) of things; for example,
make and model
• Instance identifier; for example, VIN (Vehicle identity number) for vehicles

Identity Management and Establishment

• Identity Management (IDM) for the devices, applications and services is an FC of
security FG. IdM means managing different identities, pseudo-names, hierarchies of
group IDs as well as IDs for message senders and receivers. The FC anonymously
manages the IDs.

• Communication between the device and application/services is after each one

establishes the identity of the other securely, using authentication and authorisation and
other functions.

Access Control
• Three Functional Components (FCs) in a security Functional Group (FG) for ensuring
security and privacy are:

o Authentications
o Authorisation
o Key exchange and management

Authentication
o ID establishment and authentication are essential elements of access control.
o A hash function or MD5 gives the irreversible result after many operations on
that and the operations are just one way.
o The algorithm generates a fixed size, say, 128 or 256-bit hash or digest value
using authentication data and secret key.
o Only the hash or digest value communicates. The receiver-end receives the
value, and compares that with a stored value. If both are equal then the sender
is authenticated.
o Hash function characteristic features are pre-image resistance, hash function
 should not alter, before or after communication and should be as per the
previous image (original message), second pre-image resistance: hash function
should not be altered by an in between entity (called eavesdropper), should
remain the same as one for the previous image (original message) should be
collision-resistance and should not be the same for any form of altered message.

Authorisation

• Access control allows only an authorised device or application/service access to a

resource, such as web API input, IoT device, sensor or actuator data or URL.
Authorisation model is an essential element of secure access control. The standard
authorisation models are as follows:
o Access Control List (ACL) for coarse-grain access control
o Role-Based Access Control (RBAC) for fine-grain access control
o Attribute-Based Access Control (ABAC) or other capability-based fine grain
access control

• An access control server and data communication gateway can be centrally used to
control accesses between application/service and IoT devices. The server central
control can be on a cloud server. Each device can access the server and communicate
data to another server.

• Alternatively, a distributed architecture enables:

o Each device to request access to the server and the server grants
application/service access token
o Each application/service to request access to the server and the server grants
device access token for the device.

Key Exchange and Management

• Key of sender messages needs to be known to receiver for accessing the received data.
• Key of respondent of messages needs to be known to sender for accessing the responses.
• The keys, therefore, need to be exchanged before the communication of authentication
code, authorisation commands and encrypted messages.
• Since each application/service component and device data application or service may
need unique and distinct keys, an FC provisions for the functions of key management
and exchanges.
 Message-Integrity

• An important aspect of system design is message integrity (data integrity), which means
the message remains unaltered.

• A message should not be altered during its communication.

• The encrypted data after decrypting should be identical to one before encryption.
Message integrity check involves the following steps:

o Hash function or digest algorithm calculates 128 or 192 or 256 Hash value h0,
taking the message M0 and K as inputs
o Appends the h0 along with the message
o Communicate or store h0

Integrity Check

• Integrity check steps are as follows:

o Retrieve M any time later. Assume that retrieved message is M1.
o Calculate 128 or 192 or 256 Hash value h1, taking the message M1 and K as
inputs.
o Compare h1 and h0.
o Message is unchanged if h1 = h0, and integrity check passes else fails.
o Message or data integrity means maintaining and assuring the accuracy and
consistency over its entire lifecycle
 Non repudiation

• Non-repudiation means an assurance that the source of a message once having

communicated the data to a sender, cannot deny it later, that the message was not sent
from the source and is not the same as sent earlier.

• It means data is signed and the signature put at the source cannot be denied. Digital
signature is a method which ensures non-repudiation.

• The service provides proof of the message’s origin as well as its integrity.

• A digital certificate asserts the origin using a public key infrastructure. A digital
signature is certified by a trusted digital certifying service [trusted third party (TTP)
service].

• TTP protects the private (secret) key and issues the certificate that a message was sent
using this specific private (secret) key of source incase the private key is lost and used
by some other source of message. Only the TTP is permitted to be the repository for
public key certificates.

• Example gives the steps for signing, issue of digital certificates to a signed message and
verification of a signed message.

Message-Availability

• Message-availability affects when Denial-of-Service (DoS) attack occurs. This is

because source-end message (of device or network or application/service resource) is
unavailable to the intended destination-end on DoS.
• Examples of DoS attacks are:
o ICMP flooding which repeatedly sends control messages to the destination and
thus denies the path to source end
o An SYN flood means the attacker sends a flood of TCP/SYN messages
(packets) using a forged address, and the destination repeatedly sends TCP/SYN
packets assuming the packet is from an actual source. The service to the source
message becomes unavailable and the original message becomes unavailable at
the destination
o Peer-peer attack
o Application layer messages flooding
• An attack can be prevented using specific methods for specific type of attacks. Firewall
is one method for preventing attacks on messages from untrusted networks.
Smart and Connected Cities
An IoT Strategy for Smarter Cities
Managing a city bears some resemblance to managing a corporate enterprise. As the need for
efficiency increases, new tools help increase operational efficiency. For cities, just as for
businesses, digitization transforms the perspective on operations. New ideas emerge, bringing
different approaches to solving management issues. Scalable solutions utilizing information
and communications technology (ICT) can alleviate many issues urban centers face today by
increasing efficiency, which reduces costs and enhances quality of life.

Vertical IoT Needs for Smarter Cities

There are many differing approaches and solutions for city management. All these solutions
typically start at the street level, with sensors that capture data on everything from parking
space availability to water purity. Data analytics is also used extensively, for example, to reduce
crime or improve traffic flows. Citizens can use tools to leverage their smart mobile devices,
such as to report problems and make recommendations for improving urban life or locate
available parking spaces.
These connections create vast amounts of data, some of which has never been accessible
before. When this data is analyzed and used intelligently, the possibilities to correlate, analyze,
and optimize services and processes that deliver a better quality of life for people are practically
endless. However, the growth of IoT applications for urban centers not only delivers unique
benefits for each issue it solves but also enhances a city‘s ability to develop efficient services.
Cities are expected to generate almost two-thirds (63%) of IoT‘s overall civilian benefits
worldwide over the next decade. To maximize value, smart cities can combine use cases
through a shared-revenue business model together with special partners to monetize city
location services for retail and tourism, as well as city planning, parking, and water
management.
A recent Cisco study, as illustrated in Figure below, expects IoT to have the following
economic impact over a 10-year period:
■ Smart buildings: Smart buildings have the potential to save $100 billion by lowering
operating costs by reducing energy consumption through the efficient integration of
heating, ventilation, and air-conditioning (HVAC) and other building infrastructure
systems.
■ Gas monitoring: Monitoring gas could save $69 billion by reducing meter-reading
costs and increasing the accuracy of readings for citizens and municipal utility agencies.
There are also very important advantages in terms of safety, regardless of who operates
the utility. In cases of sudden consumption increase, a timely alert could lead to
emergency response teams being dispatched sooner, thus increasing the safety of the
urban environment.
■ Smart parking: Smart parking could create $41 billion by providing real-time
visibility into parking space availability across a city. Residents can identify and reserve
the closest available space, traffic wardens can identify noncompliant usage, and
municipalities can introduce demand- based pricing.
■ Water management: Smart water management could save $39 billion by connecting
household water meters over an IP network to provide remote usage and status
information. The benefit is obvious, with features such as real-time consumption
visibility and leak detection. A gate or a pump can be opened and closed remotely and
automatically in real time, based on a variety of flow input and output analytics data.
Vibrations can be measured to detect and predict potential equipment failures.
■ Road pricing: Cities could create $18 billion in new revenues by implementing automatic
payments as vehicles enter busy city zones while improving overall traffic conditions.
 Real-time traffic condition data is very valuable and actionable information that can also
be used to proactively reroute public transportation services or private users.

Figure: Key Use Cases for Smart Cities

Global vs. Siloed Strategies

The main obstacle in implementing smart solutions in today‘s traditional infrastructure is the
complexity of how cities are operated, financed, regulated, and planned. Cities attempting to
upgrade their infrastructure to match the growing needs of the citizen population often invest
in one problem at a time, and they do it independently. Even cities using IoT technology break
up city assets and service management into silos that are typically unable to communicate or
rely on each other. The independent investment model results in the following problems:
■ Isolation of infrastructure and IT resources
■ No sharing of intelligence and information, such as video feeds and data from sensors.
■ Waste and duplication in investment and effort
■ Difficulty scaling infrastructure management

This fragmented approach is not scalable, efficient, or economically viable, and it does not
benefit from cross-functional sharing of data and services.

Cities need to begin with a solution that can extend systems across vendors, technologies,
and data types, and they should approach their infrastructure investment with a horizontal
solution that addresses their issues cohesively.

City issues are typically large-scale. They require collection of large amounts of diverse data
sets in real time. For instance, managing traffic flows and congestion in a city involves
understanding patterns of traffic in real time. This means that data from traffic sensors, traffic
cameras, parking sensors, and more has to be collected and analyzed in real time so that
decision making can be optimized around signal timing, rerouting, and so on.

All these requirements pose technological challenges, including the following:

■ How do you collect the data? What are the various sources of data, including hardware
endpoints and software?
■ How do you make sure that any data collection devices, such as sensors, can be maintained
without high costs?
■ Where do you analyze the data? What data do you carry back to the cloud, and what data
do you analyze locally?
■ What kind of network connectivity is best suited for each type of data to c ollect?
■ What kind of power availability and other infrastructure, such as storage, is required?
■ How do you aggregate data from different sources to create a unified view?
■ How do you publish the data and make it available for applications to consume?
■ How do you make the end analysis available to specialized smart city personnel, such as
traffic operators, parking enforcement officers, street lighting operators, and so on at
their logical decision points?
■ How do you present the long-term analysis to city planners?

Smart City IoT Architecture

A smart city IoT infrastructure is a four-layered architecture, as shown in Figure below. Data
flows from devices at the street layer to the city network layer and connect to the data center
layer, where the data is aggregated, normalized, and virtualized. The data center layer provides
information to the services layer, which consists of the applications that provide services to the
city.

Figure: Smart Cities Layered Architecture

In smart cities, multiple services may use IoT solutions for many different purposes. These
services may use different IoT solutions, with different protocols and different application
languages. Therefore, data flow from sensor to application involves a translation process into
a normalized language that can be exposed through APIs for other service application
consumption. This translation ensures a single language for all devices in the cloud.
Street Layer
The street layer is composed of devices and sensors that collect data and take action
based on instructions from the overall solution, as well as the networking components needed
to aggregate and collect data.
A sensor is a data source that generates data required to understand the physical world. Sensor
devices are able to detect and measure events in the physical world. ICT connectivity
solutions rely on sensors to collect the data from the world around them so that it can be
analyzed and used to operationalize use cases for cities.

A variety of sensors are used at the street layer for a variety of smart city use cases. Here is a
short representative list:
■ A magnetic sensor can detect a parking event by analyzing changes in the surrounding
magnetic field when a heavy metal object, such as a car or a truck, comes close to it (or on
top of it).
■ A lighting controller can dim and brighten a light based on a combination of time-based
and ambient conditions.
■ Video cameras combined with video analytics can detect vehicles, faces, and traffic
conditions for various traffic and security use cases.
■ An air quality sensor can detect and measure gas and particulate matter concentrations to
give a hyper-localized perspective on pollution in a given area.
■ Device counters give an estimate of the number of devices in the area, which provides a
rough idea of the number of vehicles moving or parked in a street or a public parking area,
of pedestrians on a sidewalk, or even of birds in public parks or on public monuments—
for cities where bird control has become an issue.

For each type of data to collect, there are a variety of solutions and possible approaches.
The choice of sensor technology depends on the exact nature of the problem, the accuracy and
cost trade-offs appropriate for it, and any installation limitations posed by the physical
environment. Another consideration is the requirement to interact with other IoT systems in
the same physical space.

One of the key aspects to consider when choosing a sensing device is its lifetime maintenance
costs. Some sensors are mounted on city infrastructure, such as light poles. These sensors can
benefit from the power, and possibly the network connectivity, of their mounting location.

Another key aspect to consider when choosing the right technology for a smart city is edge
analytics. The many sensors and their data must be managed through the network in a way that
securely processes data with minimal delay—and often in real time.

Finally, for sensor characteristics, storage is a key consideration that depends on the method,
location, and length of time the data has to be archived. Data collection and storage also have
an important impact on privacy.

Regardless of the type of system chosen, sensor data is transported and processed by the IoT
system. Although IoT systems use common APIs and normalized language in the cloud, they
may use different network protocols.

Another issue that network planning must take into account is the required level of agnosticism
of smart city networks. LoRaWAN is growing as a major protocol for smart city sensors, across
multiple verticals. LoRaWAN is well adapted to the type of ranges required in an urban
environment and the types of data exchanges that most smart city sensors need.

City Layer
This layer aggregates all data collected by sensors and the end-node network into a single
transport network.
The city layer may appear to be a simple transport layer between the edge devices and the data
center or the Internet. However, one key consideration of the city layer is that it needs to
transport multiple types of protocols, for multiple types of IoT applications. Figure below
shows a Street Layer Resiliency.

Figure: Street Layer Resiliency

Data Center Layer
Data collected from the sensors is sent to a data center, where it can be processed and
correlated. Based on this processing of data, meaningful information and trends can be derived,
and information can be provided back. For example, an application in a data center can provide
a global view of the city traffic and help authorities decide on the need for more or less common
transport vehicles.

The key technology in creating any comprehensive smart solution with services is the cloud.
With a cloud infrastructure, data is not stored in a data center owned directly or indirectly by
city authorities.

The cloud model is the chief means of delivering storage, virtualization, adaptability, and
the analytics know-how that city governments require for the technological mashup and
synergy of information embodied in a smart city.

Figure below shows the vision of utilizing the cloud in smart solutions for cities. The cloud
provides a scalable, secure, and reliable data processing engine that can handle the immense
amount of data passing through it.
Figure: The Role of the Cloud for Smart City Applications
Smart city issues require not just efficient use of infrastructure, which the cloud helps enable,
they also require new data processing and management models. For example, cloud services
allow for Software as a Service (SaaS) models that create cyclical returns on investment.

Services Layer
The true value of ICT connectivity comes from the services that the measured data can provide
to different users operating within a city. Smart city applications can provide value to and
visibility for a variety of user types, including city operators, citizens, and law enforcement.
The collected data should be visualized according to the specific needs of each consumer of
that data and the particular user experience requirements and individual use cases.
For example, parking data indicating which spots are and aren‘t currently occupied can drive a
citizen parking app with a map of available spots, as well as an enforcement officer‘s
understanding of the state (utilization and payment) of the public parking space, while at the
same time helping the city operator‘s perspective on parking problem areas in the city at any
given time.

Smart City Use-Case Examples

There are multiple ways a smart city can improve its efficiency and the lives of its citizens.
Some of the applications commonly used as starting points to implement IoT in smart cities:
connected street lighting, smart parking, smart traffic control, and connected environment.

Connected Street Lighting

Of all urban utilities, street lighting comprises one of the largest expenses in a municipality‘s
utility bill, accounting for up to 40% of the total.
Maintenance of street lights is an operational challenge, given the large number of lights and
their vast geographic distribution.

Connected Street Lighting Solution

Cities commonly look for solutions to help reduce lighting expenses and at the same time
improve operating efficiencies while minimizing upfront investment. The installation of a
smart street lighting solution can provide significant energy savings and can also be leveraged
to provide additional services.

In this regard, light-emitting diode (LED) technology leads the transition from traditional street
lighting to smart street lighting:
■ LEDs require less energy to produce more light than legacy lights, and they have a much
 longer life span and a longer maintenance cycle.
■ A leading lighting company estimates that a complete switch to LED technology can
reduce individual light bills by up to 70%.
■ LEDs are well suited to smart solution use cases. For example, LED color or light intensity
can be adapted to site requirements.

Street Lighting Architecture

Connected lighting uses a light management application to manage street lights remotely
by connecting to the smart city‘s infrastructure. This application attaches to LED lights,
monitors their management and maintenance, and allows you to view the operational status of
each light. In most cases, a sensor gateway acts as an intermediate system between the
application and the lights (light control nodes).

The gateway relays instructions from the application to the lights and stores the local lights‘
events for the application‘s consumption. The controller and LED lights use the cloud to
connect to the smart city‘s infrastructure, as shown in Figure below.

Figure: Connected Lighting Architecture

A human or automated operator can use a cloud application to perform automated scheduling
for lights and even get light sensors to perform automated dimming or brightening, as needed.
The schedule can also impact the light intensity level and possibly the color, depending
on environmental conditions, weather, time of year, time of day, location within the city, and
so on.
Lighting nodes vary widely in the industry, especially with respect to elements such as what
communication protocol they use (for example, Wi-Fi, cellular, ZigBee, 802.15.4g [Wi-SUN],
LoRaWAN), level of ruggedization, and on-board sensor capabilities. These features are
optimized for different circumstances and conditions; no single lighting node can support all
environments ideally.
Smart Parking
Parking is a universal challenge for cities around the globe. According to urban planning
researchers, up to 30% of cars driving in congested downtown traffic are searching for parking
spaces. Ineffective parking access and administration make parking in urban areas a constant
struggle and affect cities in many ways.

Smart Parking Use Cases

Added traffic congestion is one consequence of drivers looking for parking space, and it has
several consequences:
■ Contributes to pollution: Tons of extra carbon emissions are released into the city‘s
environment due to cars driving around searching for parking spots when they could be
parked.
■ Causes motorist frustration: In most cities, parking spot scarcity causes drivers to lose
patience and waste time, leading to road rage, inattention, and other stress factors.
■ Increases traffic incidents: Drivers searching for parking spots cause increased
congestion in the streets and that, in turn, causes increased accidents and other traffic
[Link] loss is another consequence of drivers looking unsuccessfully for
parking space, and it also has various negative side effects:
■ Cities often lose revenue: As a result of inadequate parking meter enforcement and no-
parking, no-standing, and loading-zone violations, cities lose revenue.
■ Parking administration employee productivity suffers: Employees waste time
roaming the streets, attempting to detect parking rules offenders.
■ Parking availability affects income: Local shops and businesses lose customers because
of the decreased accessibility caused by parking space shortages.
Smart Parking Architecture
A variety of parking sensors are available on the market, and they take different approaches to
sensing occupancy for parking spots. Examples include in-ground magnetic sensors, which use
embedded sensors to create a magnetic detection field in a parking spot; video-based sensors,
which detect events based on video computing (vehicle movements or presence); and radar
sensors that sense the presence of vehicles (volumetric detection). Figure below shows the
Connected Parking Architecture.

Figure: Connected Parking Architecture.

Technology innovations are happening all the time, making the holistic ICT connectivity
architecture even more important. For example, new detection technologies rely on sensing
the radio emissions (Bluetooth and others) coming from a vehicle. The adoption of such new
technologies implies that the communication architecture is open enough to accommodate
the needs of these new systems.
Combining these technologies in innovative ways also expands the possibilities of the services
IoT systems can deliver; this certainly holds true for smart parking.

Regardless of the technology used, parking sensors are typically event-driven objects. A
sensor detects an event and identifies it based on time or analysis. The event is transmitted
through the device‘s communication protocol to an access point or gateway, which forwards
the event data through the city layer.

The following are some potential user experiences for these three user types:
■ City operators: These users might want a high-level map of parking in the city to maintain
perspective on the city‘s ongoing parking situation. They would also need information on
historical parking data patterns to understand congestion and pain points in order to be able
to effectively influence urban planning.
■ Parking enforcement officers: These users might require real-time updates on parking
changes in a certain area to be able to take immediate action on enforcement activities,
such as issuing tickets or sending warnings to citizens whose time is nearing expiration.
■ Citizens: These users might want an application with a map (such as a built-in parking
app in their car) showing available parking spots, reservation capabilities, and online
payment. Their focus would be on minimizing the time to get a parking spot and avoiding
parking tickets. The application could warn when parking duration limits approach,
allowing the driver to move the vehicle before the timer expires or pay a parking timer
extension fee without having to go back to the vehicle.

Smart Traffic Control

Traffic is one the most well-understood pain points for any city. It is the leading cause of
accidental death globally, causes immense frustration, and heavily contributes to pollution
around the globe. A smart city traffic solution would combine crowd counts, transit
information, vehicle counts, and so on and send events regarding incidents on the road so that
other controllers on the street could take action.

Smart Traffic Control Architecture

In the architecture shown in Figure below, a video analytics sensor computes traffic events
based on a video feed and only pushes events (the car count, or metadata, not the individual
images) through the network. These events go through the architectural layers and reach the
applications that can drive traffic services. These services include traffic light coordination and
also license plate identification for toll roads. Some sensors can also recognize abnormal
patterns, such as vehicles moving in the wrong direction or a reserved lane. In that case, the
video feed itself may be uploaded to traffic enforcement agencies.

Other types of sensors that are part of traffic control solutions include Bluetooth vehicle
counters, real-time speed and vehicle counters, and lighting control systems. These sensors
provide a real-time perspective while also offering data collection services for historical data
trending and correlation purposes.
Smart Traffic Applications
Traffic applications can be enabled to take immediate action with other sensors to manage
traffic and to reduce pain points. Historical data can be used to develop more efficient urban
planning to reduce the amount of traffic a city experiences.
A well-known remedy for stop-and-go traffic is to regulate the standard flow speed based on
car density. As density increases, car speed is forced down to avoid the wave effect.
An application that measures traffic density in real time can take action by regulating the street
light cycle duration to control the number of cars added to the flow of the main routes, thus
limiting or suppressing the wave effect.
Information can also be shared with drivers.
Understanding a city‘s real-time traffic patterns and being able to effectively mitigate traffic
issues can drive tremendous value for a city.

Connected Environment
As of 2017, 50% of the world‘s population has settled on less than 2% of the earth‘s surface
area. Such densely populated closed spaces can see spikes in dangerous gas molecules at
any given moment. More than 90% of the world‘s urban population breathes in
air with pollutant levels that are much higher than the recommended thresholds, and one out
of every eight deaths worldwide is a result of polluted air.

The Need for a Connected Environment

Most large cities monitor their air quality. Data is often derived from enormous air quality
monitoring stations that are expensive and have been around for decades. These stations
are highly accurate in their measurements but also highly limited in their range, and a city is
likely to have many blind spots in coverage.

To fully address the air quality issues in the short term and the long term, a smart city
would need to understand air quality on a hyper-localized, real-time, distributed basis at any
given moment. To get those measurements, smart cities need to invest in the following:
■ Open-data platforms that provide current air quality measurements from existing air quality
 monitoring stations.
■ Sensors that provide similar accuracy to the air quality stations but are available at much
lower prices.
■ Actionable insights and triggers to improve air quality through cross-domain actions.
■ Visualization of environmental data for consumers and maintenance of historical air
quality data records to track emissions over time.

Connected Environment Architecture

Figure below shows an architecture in which all connected environment elements overlay on
the generalized four-layer smart city IoT architecture.

As shown in Figure, at the street layer there are a variety of multivendor sensor offerings, using
a variety of communication protocols. Connected environment sensors might measure
different gases, depending on a city‘s particular air quality issues, and may include weather
and noise sensors. These sensors may be located in a variety of urban fixtures, such as in street
lights,

as explained earlier. They may also be embedded in the ground or in other structures or smart
city infrastructure. Even mobile sources of information can be included through connected
wearables that citizens might choose to purchase and carry with them to understand the air
quality around them at any given moment. Crowdsourcing may make this information available
to the global system.

Communication technologies depend on the location of the sensors. Wearables typically

communicate via a short-range technology (such as Bluetooth) with a nearby collecting device
(such as a phone).

Independent and standalone sensors typically use wireless technologies. In dense urban
environments, ZigBee and Wi-Fi are common. However, Wi-Fi is not very well adapted for
networks where reports are sporadic because Wi-Fi requires an 802.11 connection to be
maintained, which consumes battery resources.