The
effectiveness
of
Fuzzy
Inference
Systems
(FISs)
in
manipulating
uncertainty
and
nonlinearity
makes
them
a
subject
significant
interest
for
decision-making
embedded
systems.
Accordingly,
optimizing
FIS
hardware
improves
its
performance,
efficiency,
capabilities,
leading
to
better
user
experience,
increased
productivity,
cost
savings.
To
be
compatible
with
the
limited
power
budget
most
systems,
this
paper
presents
framework
realize
ultra-low
hardware.
It
supports
optimizations
both
conventional
arithmetic
as
well
MSDF-computing
highly
consistent
MSDF-based
sensors.
In
all
processes
fuzzification,
inference,
defuzzification
are
done
on
serially
coming
data
bits.
demonstrate
efficiency
proposed
framework,
we
utilized
Matlab,
Chisel3,
Vivado
implement
it
from
high-level
descriptions
synthesis.
We
also
developed
Scala
library
Chisel3
establish
connection
between
these
tools,
bridging
gap,
facilitating
design
space
exploration
at
level.
Furthermore,
realized
an
navigation
autonomous
mobile
robots
unknown
environments.
Synthesis
results
show
superiority
output
our
suggested
terms
resource
usage
energy
consumption
compared
Matlab
HDL
code
generator
output.
Symmetry,
Journal Year:
2025,
Volume and Issue:
17(2), P. 275 - 275
Published: Feb. 11, 2025
Despite
many
fuzzy
time
series
forecasting
(FTSF)
models
addressing
complex
temporal
patterns
and
uncertainties
in
data,
two
limitations
persist:
they
do
not
treat
crisp
as
a
unified
whole
for
analyzing
nonlinear
relationships
between
different
moments,
fail
to
effectively
capture
how
uncertainty
affects
predictions.
In
this
paper,
we
propose
an
FTSF
model
integrating
Bayesian
networks
overcome
the
limitations.
network
(BN)
structure
learning
is
employed
extract
fuzzy–crisp
dependencies
historical
fuzzified
data
predicted
alongside
within
data.
Integrating
logical
relationship
groups
(FLRGs)
BNs
representing
identifies
efficiently.
BN
parameter
occurrence
of
through
conditional
probability
distributions
BNs,
while
empirical
probabilities
quantify
elements
FLRGs.
The
defuzzification
stage
infers
value
using
fuzzy-empirical-probability
weighted
FLRGs
BN.
We
validate
performance
proposed
on
sixteen
diverse
series.
Experimental
results
demonstrate
competitive
compared
state-of-the-art
methods.
Mathematics,
Journal Year:
2025,
Volume and Issue:
13(7), P. 1156 - 1156
Published: March 31, 2025
This
study
introduces
the
GWO-FNN
model,
an
improvement
of
fuzzy
neural
network
(FNN)
architecture
that
aims
to
balance
high
performance
with
improved
interpretability
in
artificial
intelligence
(AI)
systems.
The
model
leverages
Grey
Wolf
Optimizer
(GWO)
fine-tune
consequents
rules
and
uses
mutual
information
(MI)
initialize
weights
input
layer,
resulting
greater
classification
accuracy
transparency.
A
distinctive
aspect
is
its
capacity
transform
logical
neurons
hidden
layer
into
comprehensible
rules,
thereby
elucidating
reasoning
behind
outputs.
model’s
were
rigorously
evaluated
through
statistical
methods,
benchmarks,
real-world
dataset
testing.
These
evaluations
demonstrate
strong
capability
extract
clearly
express
intricate
patterns
within
data.
By
combining
advanced
rule
mechanisms
a
comprehensive
framework,
contributes
meaningful
advancement
interpretable
AI
approaches.
