Security and Privacy Preservation via Interference Tolerant Fast Convergence Zeroing Neural Network With Reptile Search Optimization Algorithm in Fog‐Cloud Computing DOI

P Kumar,

Neha Verma, Shivani Gupta

и другие.

Transactions on Emerging Telecommunications Technologies, Год журнала: 2025, Номер 36(4)

Опубликована: Апрель 1, 2025

ABSTRACT More vertical service areas than only data processing, storing, and communication are promised by fog‐cloud computing. Due to its great efficiency scalability, distributed deep learning (DDL) across computing environments is a widely used application among them. With training limited sharing parameters, DDL can offer more privacy protection centralized learning. Nevertheless, still faces two significant security obstacles when it comes How ensure that users' identities not stolen outside enemies, prevent from being disclosed other internal participants in the process of training. In this manuscript, Interference Tolerant Fast Convergence Zeroing Neural Network for Security Privacy Preservation with Reptile Search Optimization Algorithm Fog‐Cloud Computing environment (SPP‐ITFCZNN‐RSOA‐FCC) proposed. ITFCZNN proposed preservation, Then (RSOA) optimize ITFCZNN, Effective Lightweight Homomorphic Cryptographic (ELHCA) encrypt decrypt local gradients. The SPP‐ITFCZNN‐RSOA‐FCC system attains better balance, efficiency, functionality existing efforts. implemented using Python. performance metrics like accuracy, resource overhead, computation overhead considered. approach 29.16%, 20.14%, 18.93% high 11.03%, 26.04%, 23.51% lower Resource compared methods including FedSDM: Federated dependent smart decision making component ECG at internet things incorporated Edge‐Fog‐Cloud (SPP‐FSDM‐FCC), A collaborative offloading dew‐enabled vehicular fog compute‐intensive latency‐sensitive dependence‐aware tasks: Q‐learning method (SPP‐FDQL‐FCC), fog‐edge‐enabled intrusion identification scheme grids (SPP‐FSVM‐FCC) respectively.

Язык: Английский

The applications of nature‐inspired algorithms in Internet of Things‐based healthcare service: A systematic literature review DOI
Zahra Mohtasham‐Amiri, Arash Heidari, Mohammad Zavvar

и другие.

Transactions on Emerging Telecommunications Technologies, Год журнала: 2024, Номер 35(6)

Опубликована: Май 21, 2024

Abstract Nature‐inspired algorithms revolve around the intersection of nature‐inspired and IoT within healthcare domain. This domain addresses emerging trends potential synergies between computational approaches technologies for advancing services. Our research aims to fill gaps in addressing algorithmic integration challenges, real‐world implementation issues, efficacy IoT‐based healthcare. We provide insights into practical aspects limitations such applications through a systematic literature review. Specifically, we address need comprehensive understanding healthcare, identifying as lack standardized evaluation metrics studies on challenges security considerations. By bridging these gaps, our paper offers directions future this domain, exploring diverse landscape chosen methodology is Systematic Literature Review (SLR) investigate related papers rigorously. Categorizing groups genetic algorithms, particle swarm optimization, cuckoo ant colony other approaches, hybrid methods, employ meticulous classification based critical criteria. MATLAB emerges predominant programming language, constituting 37.9% cases, showcasing prevalent choice among researchers. emphasizes adaptability paramount parameter, accounting 18.4% shedding light attributes, limitations, development, review contribute dynamic

Язык: Английский

Процитировано

46
ENVQA: Improving Visual Question Answering model by enriching the visual feature DOI
Souvik Chowdhury, Badal Soni

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 142, С. 109948 - 109948

Опубликована: Янв. 5, 2025

Язык: Английский

Процитировано

4
Machine Learning and Deep Learning Paradigms: From Techniques to Practical Applications and Research Frontiers DOI Creative Commons
Kamran Razzaq, Mahmood Shah

Computers, Год журнала: 2025, Номер 14(3), С. 93 - 93

Опубликована: Март 6, 2025

Machine learning (ML) and deep (DL), subsets of artificial intelligence (AI), are the core technologies that lead significant transformation innovation in various industries by integrating AI-driven solutions. Understanding ML DL is essential to logically analyse applicability identify their effectiveness different areas like healthcare, finance, agriculture, manufacturing, transportation. consists supervised, unsupervised, semi-supervised, reinforcement techniques. On other hand, DL, a subfield ML, comprising neural networks (NNs), can deal with complicated datasets health, autonomous systems, finance industries. This study presents holistic view technologies, analysing algorithms application’s capacity address real-world problems. The investigates application which techniques implemented. Moreover, highlights latest trends possible future avenues for research development (R&D), consist developing hybrid models, generative AI, incorporating technologies. aims provide comprehensive on serve as reference guide researchers, industry professionals, practitioners, policy makers.

Язык: Английский

Процитировано

4
Privacy-preserving approach for IoT networks using statistical learning with optimization algorithm on high-dimensional big data environment DOI Creative Commons
Fatma S. Alrayes, Mohammed Maray, Asma Alshuhail

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 27, 2025

Язык: Английский

Процитировано

3
Intelligent deep federated learning model for enhancing security in internet of things enabled edge computing environment DOI Creative Commons

Nasser Albogami

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Фев. 3, 2025

In the present scenario, Internet of Things (IoT) and edge computing technologies have been developing rapidly, foremost to development new tasks in security privacy. Personal information privacy leakage become main concerns IoT surroundings. The promptly IoT-connected devices below an integrated Machine Learning (ML) method might threaten data confidentiality. standard centralized ML-assisted methods challenging because they require vast numbers a vital unit. Due rising distribution many systems linked devices, decentralized ML solutions required. Federated learning (FL) was proposed as optimal solution discover these issues. Still, heterogeneity environments poses essential task when executing FL. Therefore, this paper develops Intelligent Deep Model for Enhancing Security (IDFLM-ES) approach IoT-enabled edge-computing environment. presented IDFLM-ES aims identify unwanted intrusions certify safety To accomplish this, technique introduces federated hybrid deep belief network (FHDBN) model using FL on time series produced by devices. Besides, uses normalization golden jackal optimization (GJO) based feature selection pre-processing step. learns individual distributed representation over databases enhance convergence quick learning. Finally, dung beetle optimizer (DBO) is utilized choose effectual hyperparameter FHDBN model. simulation value methodology verified benchmark database. experimental validation portrayed superior accuracy 98.24% compared other models.

Язык: Английский

Процитировано

3
Advancing Precision Agriculture: Enhanced Weed Detection Using the Optimized YOLOv8T Model DOI
Shubham Sharma, Manu Vardhan

Arabian Journal for Science and Engineering, Год журнала: 2024, Номер unknown

Опубликована: Авг. 2, 2024

Язык: Английский

Процитировано

8
Enhanced novelty approaches for resource allocation model for multi-cloud environment in vehicular Ad-Hoc networks DOI Creative Commons

R. Augustian Isaac,

P. Sundaravadivel,

V. Marx

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 19, 2025

As the number of service requests for applications continues increasing due to various conditions, limitations on resources provide a barrier in providing with appropriate Quality Service (QoS) assurances. result, an efficient scheduling mechanism is required determine order handling application requests, as well use broadcast media and data transfer. In this paper innovative approach, incorporating Crossover Mutation (CM)-centered Marine Predator Algorithm (MPA) introduced effective resource allocation. This strategic allocation optimally schedules within Vehicular Edge computing (VEC) network, ensuring most utilization. The proposed method begins by meticulous feature extraction from network model, attributes such mobility patterns, transmission medium, bandwidth, storage capacity, packet delivery ratio. For further analysis Elephant Herding Lion Optimizer (EHLO) algorithm employed pinpoint critical attributes. Subsequently Modified Fuzzy C-Means (MFCM) used vehicle clustering centred selected These clustered characteristics are then transferred stored cloud server infrastructure. performance methodology evaluated using MATLAB software simulation method. study offers comprehensive solution challenge Cloud Networks, addresses burgeoning demands modern while QoS assurances signifies significant advancement field VEC.

Язык: Английский

Процитировано

1
Federated Learning on Internet of Things: Extensive and Systematic Review DOI Open Access

Meenakshi Aggarwal,

Vikas Khullar, Sunita Rani

и другие.

Computers, materials & continua/Computers, materials & continua (Print), Год журнала: 2024, Номер 79(2), С. 1795 - 1834

Опубликована: Янв. 1, 2024

The proliferation of IoT devices requires innovative approaches to gaining insights while preserving privacy and resources amid unprecedented data generation.However, FL development for is still in its infancy needs be explored various areas understand the key challenges deployment real-world scenarios.The paper systematically reviewed available literature using PRISMA guiding principle.The study aims provide a detailed overview increasing use networks, including architecture challenges.A systematic review approach used collect, categorize analyze FL-IoT-based articles.A search was performed IEEE, Elsevier, Arxiv, ACM, WOS databases 92 articles were finally examined.Inclusion measures published English with keywords "FL" "IoT".The methodology begins an recent advances IoT, followed by discussion how these two technologies can integrated.To more specific, we examine evaluate capabilities talking about communication protocols, frameworks architecture.We then present comprehensive analysis number applications, smart healthcare, transportation, cities, industry, finance, agriculture.The findings from this services applications are also presented.Finally, comparative IID (independent identical data) non-ID, traditional centralized deep learning (DL) approaches.We concluded that has better performance, especially terms protection resource utilization.FL excellent because model training takes place on individual or edge nodes, eliminating need aggregation, which poses significant risks.To facilitate rapidly evolving field, presented intended help practitioners researchers navigate complex terrain IoT.

Язык: Английский

Процитировано

6
IntDEM: an intelligent deep optimized energy management system for IoT-enabled smart grid applications DOI

P. Ganesh,

B. Meenakshi Sundaram, Praveen Kumar Balachandran

и другие.

Electrical Engineering, Год журнала: 2024, Номер unknown

Опубликована: Июль 24, 2024

Язык: Английский

Процитировано

6
Detection of Real-Time Deep Fakes and Face Forgery in Video Conferencing Employing Generative Adversarial Networks DOI Creative Commons
Sunil Kumar Sharma, Abdullah M. Al‐Enizi, Manoj Kumar

и другие.

Heliyon, Год журнала: 2024, Номер 10(17), С. e37163 - e37163

Опубликована: Авг. 29, 2024

As facial modification technology advances rapidly, it poses a challenge to methods used detect fake faces. The advent of deep learning and AI-based technologies has led the creation counterfeit photographs that are more difficult discern apart from real ones. Existing Deep detection systems excel at spotting content with low visual quality easily recognized by artifacts. study employed unique active forensic strategy Compact Ensemble-based discriminators architecture using Conditional Generative Adversarial Networks (CED-DCGAN), for identifying real-time fakes in video conferencing. DCGAN focuses on video-deep features since creating convincing improving rapidly. first step towards recognizing DCGAN-generated images, split images into frames containing essential elements then use bandwidth train an ensemble-based discriminator as classifier. Spectra anomalies produced up-sampling processes, standard procedures GAN making large amounts data films. Ensemble (CED) concentrates most distinguishing feature between natural synthetic giving generators robust training signal. empirical results publicly available datasets show, suggested algorithms outperform state-of-the-art proposed CED-DCGAN technique successfully detects high-fidelity conferencing generalizes well when comparing other techniques. Python tool is implementing this accuracy obtained work 98.23 %.

Язык: Английский

Процитировано

6