The
rapid
proliferation
of
Internet
Things
(IoT)
devices
has
led
to
an
increase
in
botnet
attacks
targeting
these
devices.
A
attack
is
a
cyber-attack
which
network
compromised
devices,
referred
as
"bots"
or
"zombies,"
utilized
execute
synchronized
attack.
These
can
result
substantial
harm
both
the
and
they
are
connected.
This
study
investigates
deployment
security
authentication
protocols
verify
identity
IoT
prior
connection.
also
evaluates
classification
accuracy
four
distinct
supervised
machine
learning
algorithms:
Random
Forest
(RF),
Naïve
Bayes
(NB),
DecisionTree
(DT),
eXtreme
Gradient
Boosting
(XGBoost).
It
was
foundXGBoost
best
performing
classifier
among
various
algorithms
tested,
terms
detecting
networks
using
Bot-IoT
dataset.
Research Square (Research Square),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Nov. 14, 2023
Abstract
This
research
focuses
on
developing
an
anomaly
detection
system
using
machine
learning
to
mitigate
Distributed
Denial
of
Service
(DDoS)
attacks
in
IoT
networks.
The
study
utilizes
a
diverse
dataset
from
environments
train
and
evaluate
algorithms
for
DDoS
detection.
includes
various
device
types,
communication
protocols,
network
configurations.
aims
achieve
several
objectives,
including
preprocessing,
feature
engineering,
model
selection,
detection,
performance
evaluation.
team
preprocesses
the
raw
Internet
Things
(IoT)
data
by
cleaning
transforming
it
prepare
analysis.
They
then
extract
relevant
features
effectively
characterize
normal
abnormal
behavior.
Multiple
are
evaluated
compared
determine
most
suitable
models
selected
used
identify
classify
traffic
patterns
associated
with
attacks.
developed
is
assessing
its
accuracy,
precision,
recall,
F1
score.
significance
this
lies
potential
enhance
security
networks
proactively
detecting
mitigating
By
leveraging
learning,
provide
robust
defense
mechanism
against
pervasive
threat,
ensuring
reliability
availability
services
applications.
2021 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech),
Journal Year:
2023,
Volume and Issue:
unknown, P. 0690 - 0697
Published: Nov. 14, 2023
The
increasing
equipment
of
cars
with
smart
systems
and
their
networking
other
devices
is
leading
to
a
growing
network
connected
vehicles.
Connected
are
Internet
Things
(IoT)
that
communicate
bidirectionally
systems,
enabling
internet
access
data
exchange.
Artificial
Intelligence
(AI)
offers
benefits
such
as
autonomous
driving,
driver
assistance
programs,
monitoring.
connectivity
also
brings
new
risks
users'
privacy.
Our
study
focuses
on
privacy
threats
in
from
user
perspective.
provides
comprehensive
threat
model
analysis
based
combination
STRIDE
LINDDUN.
We
analyze
the
various
vulnerabilities
arise
connecting
devices,
including
Vehicle-to-Vehicle
(V2V),
Vehicle-to-Vloud
(V2C),
Vehicle-to-Device
(V2D).
conduct
our
theoretical
modern-day
vehicle
another
study.
shows
several
types
can
negatively
impact
car
users.
This
encapsulates
potential
risks,
inadvertent
disclosure
personal
due
vehicle's
interconnectedness
smartphones,
subsequent
susceptibility
unauthorized
access,
while
highlighting
need
for
robust
security
measures
indicated
by
modeling,
safeguard
against
wide
array
identified
cybersecurity
threats.
The
concept
of
"smart
cities"
has
gained
a
lot
attention
in
recent
years
as
urban
regions
struggle
with
issues
including
population
development,
resource
management,
and
environmental
sustainability.
goal
this
research
paper
is
to
give
thorough
introduction
the
idea
smart
cities
by
examining
their
definition,
essential
elements,
prospective
advantages.
Smart
aim
improve
quality
life,
encourage
economic
growth,
guarantee
efficiency
utilizing
cutting-edge
technologies
data-driven
solutions.
This
also
covers
difficulties
factors
be
considered
when
putting
city
ideas
into
practice
provides
prominent
global
examples.
examines
security
attacks,
asset-related
threats,
proposes
countermeasures,
will
provide
understanding
cities,
setting
groundwork
for
future
how
they
might
change
ecosystems.
The
rapid
proliferation
of
Internet
Things
(IoT)
devices
has
led
to
an
increase
in
botnet
attacks
targeting
these
devices.
A
attack
is
a
cyber-attack
which
network
compromised
devices,
referred
as
"bots"
or
"zombies,"
utilized
execute
synchronized
attack.
These
can
result
substantial
harm
both
the
and
they
are
connected.
This
study
investigates
deployment
security
authentication
protocols
verify
identity
IoT
prior
connection.
also
evaluates
classification
accuracy
four
distinct
supervised
machine
learning
algorithms:
Random
Forest
(RF),
Naïve
Bayes
(NB),
DecisionTree
(DT),
eXtreme
Gradient
Boosting
(XGBoost).
It
was
foundXGBoost
best
performing
classifier
among
various
algorithms
tested,
terms
detecting
networks
using
Bot-IoT
dataset.
