Electronics,
Год журнала:
2024,
Номер
13(21), С. 4185 - 4185
Опубликована: Окт. 25, 2024
The
Internet
of
things
(IoT)
presents
unique
challenges
for
the
deployment
machine
learning
(ML)
models,
particularly
due
to
constraints
on
computational
resources,
necessity
decentralized
processing,
and
concerns
regarding
security
privacy
in
interconnected
environments
such
as
cloud.
In
this
paper,
a
novel
ML
framework
is
proposed
IoT
characterized
by
wireless
communication,
dynamic
data
streams,
integration
with
cloud
services.
integrates
incremental
algorithms
robust
model
exchange
protocol,
ensuring
that
preserved,
while
enabling
devices
participate
collaborative
from
distributed
across
networks.
By
incorporating
gossip-based
communication
ensures
energy-efficient,
scalable,
secure
exchange,
fostering
effective
knowledge
sharing
among
devices,
addressing
potential
threats
inherent
cloud-based
ecosystems.
framework’s
performance
was
evaluated
through
simulations,
demonstrating
its
ability
handle
complexities
real-time
processing
resource-constrained
environments,
also
mitigating
risks
within
2022 International Conference on Inventive Computation Technologies (ICICT),
Год журнала:
2024,
Номер
unknown
Опубликована: Апрель 24, 2024
This
research
investigates
the
integration
of
bio-inspired
optimization
and
swarm
intelligence
principles
with
TinyML
for
development
energy-aware
Internet
Things
(IoT)
devices.
A
novel
model
algorithm,
termed
"BioSwarmML,"
is
introduced
evaluated
against
existing
algorithms
through
comprehensive
simulation
analyses
employing
suitable
metrics.
The
proposed
framework
aims
to
enhance
energy
efficiency
in
IoT
applications
by
leveraging
collective
derived
from
behaviors.
"BioSwarmML"
algorithm
designed
draw
inspiration
natural
processes,
incorporating
techniques
such
as
genetic
algorithms,
simulated
annealing,
evolutionary
strategies.
Concurrently,
are
integrated
emulate
decentralized
self-organizing
behaviors
observed
biological
systems.
amalgamation
optimize
consumption
models
on
devices,
facilitating
sustainable
adaptive
learning
processes.
Simulation
involve
a
comparative
study
established
evaluating
BioSwarmML
based
metrics
consumption,
accuracy,
latency.
results
demonstrate
efficacy
achieving
superior
while
maintaining
competitive
performance
terms
accuracy
responsiveness.
comparison
sheds
light
advantages
applications,
showcasing
its
potential
widespread
adoption
ecosystems.
contributes
advancement
energy-efficient
systems
introducing
algorithmic
paradigm
that
aligns
international
journal
standards.
showcases
promising
avenue
enhancing
sustainability
TinyML-driven
offering
valuable
addition
body
knowledge
field.
2022 International Conference on Communication, Computing and Internet of Things (IC3IoT),
Год журнала:
2024,
Номер
unknown, С. 1 - 5
Опубликована: Апрель 17, 2024
A
smart
solar
monitoring
system
using
IOT
describes
a
that
uses
various
sensors
and
devices
to
monitor
control
panels'
performance.
This
provides
real-time
data
on
energy
generation
consumption,
enabling
users
optimize
usage
make
informed
decisions
regarding
management.
The
consists
of
components,
including
sensors,
microcontrollers,
communication
protocols,
cloud
services.
These
components
work
together
collect
analyse
data,
provide
alerts
notifications,
generate
reports
use
technology
in
systems
improves
efficiency,
reduces
costs,
enables
remote
control,
making
it
an
ideal
solution
for
Electronics,
Год журнала:
2024,
Номер
13(21), С. 4185 - 4185
Опубликована: Окт. 25, 2024
The
Internet
of
things
(IoT)
presents
unique
challenges
for
the
deployment
machine
learning
(ML)
models,
particularly
due
to
constraints
on
computational
resources,
necessity
decentralized
processing,
and
concerns
regarding
security
privacy
in
interconnected
environments
such
as
cloud.
In
this
paper,
a
novel
ML
framework
is
proposed
IoT
characterized
by
wireless
communication,
dynamic
data
streams,
integration
with
cloud
services.
integrates
incremental
algorithms
robust
model
exchange
protocol,
ensuring
that
preserved,
while
enabling
devices
participate
collaborative
from
distributed
across
networks.
By
incorporating
gossip-based
communication
ensures
energy-efficient,
scalable,
secure
exchange,
fostering
effective
knowledge
sharing
among
devices,
addressing
potential
threats
inherent
cloud-based
ecosystems.
framework’s
performance
was
evaluated
through
simulations,
demonstrating
its
ability
handle
complexities
real-time
processing
resource-constrained
environments,
also
mitigating
risks
within
