IEEE Transactions on Information Forensics and Security, Journal Year: 2024, Volume and Issue: 19, P. 7730 - 7743
Published: Jan. 1, 2024
Language: Английский
Journal of Systems Architecture, Journal Year: 2025, Volume and Issue: unknown, P. 103401 - 103401
Published: March 1, 2025
Language: Английский
Citations
0
IEEE Transactions on Intelligent Transportation Systems, Journal Year: 2024, Volume and Issue: 25(10), P. 14785 - 14802
Published: May 6, 2024
With
the
development
of
Intelligent
Transportation
Systems
(ITS),
edge
caching
has
gradually
emerged
as
a
critical
technology
to
reduce
transmission
delay
and
optimize
network
load.
However,
limited
storage
capacity
service
scope
individual
cache
servers
significantly
degrade
performance
caching.
To
address
this
issue,
we
propose
social-aware
assisted
collaborative
algorithm
based
on
Dueling
Double
Deep
Q-Network
Digital
Twin
Network
(SACTD-D3).
The
can
dynamically
adjust
decision
similarity
user
semantic
information
availability
services
fully
utilize
servers.
Firstly,
vehicle
clusters
are
formed
users'
similarity,
an
on-board
cloud
is
constructed
request
by
sinking
services.
Secondly,
establishment
three-layer
structure
macro
base
station,
roadside
units
cloud,
content
heat-based
policy
utilized
effectively
improve
hit
rate.
Moreover,
optimization
problem
formulated
maximize
overall
utility
system
subject
cost,
thus
optimal
solution
obtained
using
proposed
Language: Английский
Citations
2
Information, Journal Year: 2024, Volume and Issue: 15(4), P. 190 - 190
Published: March 29, 2024
This paper explores the practical implementation and performance analysis of distributed learning (DL) frameworks on various client platforms, responding to dynamic landscape 6G technology pressing need for a fully connected intelligence network Internet Things (IoT) devices. The heterogeneous nature clients data presents challenges effective federated (FL) techniques, prompting our exploration transfer (FTL) Raspberry Pi, Odroid, virtual machine platforms. Our study provides detailed examination design, implementation, evaluation FTL framework, specifically adapted unique constraints IoT By measuring accuracy across diverse clients, we reveal its superior over traditional FL, particularly in terms faster training higher accuracy, due use (TL). Real-world measurements further demonstrate improved resource efficiency with lower average load, memory usage, temperature, power, energy consumption when is implemented compared FL. experiments also showcase FTL’s robustness scenarios where users leave server’s communication coverage, resulting fewer less training. adaptability underscores effectiveness environments limited data, resources, contributing valuable information intersection edge computing DL IoT.
Language: Английский
Citations
1
IEEE Transactions on Services Computing, Journal Year: 2024, Volume and Issue: 17(6), P. 3640 - 3656
Published: July 25, 2024
Language: Английский
Citations
1
Electronics, Journal Year: 2024, Volume and Issue: 13(7), P. 1216 - 1216
Published: March 26, 2024
Due to the rapid development of low earth orbit satellite constellations, e.g., Starlink, OneWeb, etc., integrated satellite-terrestrial networks have been viewed as a promising paradigm globally provide internet services for users. However, when contents from ground data centers are provided users by networks, there will be high capital expenditures in terms communication delay and bandwidth usage. To this end, paper, cooperative-caching resource-allocation problem is investigated satellite–terrestrial networks. Popular contents, which cached on satellites centers, can accessed via inter-satellite cooperative way. The optimization formulated jointly minimize deployment costs storage resource usage network consumption. A caching allocation (CCRA) algorithm based neighborhood search proposed address problem. simulation results demonstrate that CCRA outperforms Greedy BFS reducing costs.
Language: Английский
Citations
1
IEEE Transactions on Computational Social Systems, Journal Year: 2024, Volume and Issue: 11(5), P. 5775 - 5788
Published: April 15, 2024
Federated learning is a distributed machine system that uses participants' data to train an improved global model. In federated learning, participants collaboratively model, and after the training completed, each participant receives model along with incentive. Rational try maximize their individual utility, they will not input high-quality truthfully unless are provided satisfactory payments based on contributions. Furthermore, benefits from cooperation of participants. Accordingly, how establish incentive mechanism both incentivizes inputting promotes cooperative contributions has become important issue consider. this article, we introduce sharing game for employ game-theoretic approaches design core-selecting by utilizing popular concept in games, core. core can be empty, resulting becoming infeasible. To address issue, our employs relaxation method simultaneously minimizes false all Meanwhile, reduce computational complexity mechanism, propose efficient sampling approximation only aggregates models sampled coalitions approximate exact result. Extensive experiments demonstrate incentivize truthful promote effectively, while it reduces overhead compared mechanism.
Language: Английский
Citations
0
IEEE Transactions on Information Forensics and Security, Journal Year: 2024, Volume and Issue: 19, P. 7730 - 7743
Published: Jan. 1, 2024
Language: Английский
Citations
0