Future Internet

37 pages, 2285 KB

Open AccessArticle

Secure and Reliable Data Exchange in Sensor Networks Utilizing Different Communication Technologies

by Svetozar Ilchev

Future Internet 2026, 18(7), 351; https://doi.org/10.3390/fi18070351 - 4 Jul 2026

The article discusses the development of a communication protocol that provides consistent security and reliability during data exchange in sensor networks. Different communication technologies are supported. The motivation for this work is presented against the background of contemporary communication technologies and capabilities. The [...] Read more.

The article discusses the development of a communication protocol that provides consistent security and reliability during data exchange in sensor networks. Different communication technologies are supported. The motivation for this work is presented against the background of contemporary communication technologies and capabilities. The article summarizes relevant application constraints. The capabilities of popular communication technologies are briefly analyzed. Typical sensor networks serve as examples. Work on the protocol design begins with identifying important network features that serve as requirements. The design and implementation work continues with establishing a suitable packet structure, packet processing strategies, and an overall communication flow between the nodes in the network. A concept for packet routing in large networks with different communication technologies is developed and presented. The strengths and weaknesses are summarized and discussed after testing and assessment. Future work will include enhancing protocol features to improve practical applicability in different scenarios. Full article

(This article belongs to the Special Issue Wireless Sensor Networks and Internet of Things—2nd Edition)

► Show Figures

Graphical abstract

3 pages, 127 KB

Open AccessEditorial

Advanced 5G and Beyond Networks

by Mohammad Rajiullah

Future Internet 2026, 18(7), 350; https://doi.org/10.3390/fi18070350 (registering DOI) - 2 Jul 2026

Abstract

The evolution of mobile and wireless communication systems has entered a decisive phase [...] Full article

(This article belongs to the Special Issue Advanced 5G and Beyond Networks)

37 pages, 6867 KB

Open AccessArticle

ITS-Vision: Autonomous Vehicles as Mobile Surveillance Nodes in Intelligent Transportation Systems—A Conceptual Framework and Proof-of-Concept Prototype

by Mirabela-Melinda Medvei, Denis Georgian Gurău and Mihai Coca

Future Internet 2026, 18(7), 349; https://doi.org/10.3390/fi18070349 - 1 Jul 2026

Abstract

Crime surveillance in urban environments faces increasing challenges due to dynamic conditions and the demand for real-time monitoring. This paper investigates the use of video data from autonomous vehicles to enhance situational awareness in public spaces through deep learning models optimized for edge [...] Read more.

Crime surveillance in urban environments faces increasing challenges due to dynamic conditions and the demand for real-time monitoring. This paper investigates the use of video data from autonomous vehicles to enhance situational awareness in public spaces through deep learning models optimized for edge processing. High-resolution vehicle-mounted cameras serve as mobile surveillance units capable of real-time object detection, human action recognition, and anomaly detection, bridging the gap between autonomous mobility and urban monitoring. Building on this vision, we introduce ITS-Vision, a generic framework that operationalizes these use cases, enabling autonomous vehicles to function as mobile, context-aware sensing platforms. To validate this approach, we develop prototypes for key ITS-Vision components: a fight detection module using a fine-tuned X3D model, suspect identification via MediaPipe for detection combined with FaceNet for embedding extraction, and a dangerous items detection module using a fine-tuned YOLOv11n model. Due to the limited availability of real-world autonomous vehicle video datasets, experiments were conducted in controlled laboratory environments, demonstrating the feasibility of the proposed architecture and algorithms under simulated conditions. Future work will focus on collecting dedicated datasets and advancing the models toward deployment in real urban scenarios. Full article

(This article belongs to the Section Smart System Infrastructure and Applications)

► Show Figures

Figure 1

41 pages, 1770 KB

Open AccessArticle

Repeat-Explore-Aware Grocery Next-Basket Recommendation with Time-Decayed LightGCN

by Zahra Mansouri, Salim Lahmiri and Rustam Vahidov

Future Internet 2026, 18(7), 348; https://doi.org/10.3390/fi18070348 - 30 Jun 2026

Abstract

Grocery shopping is highly repetitive, yet customers also introduce new products into their buying routine over time, creating a repeat-explore challenge for next-basket recommendation. This study examines this trade-off in offline grocery recommendation using the Dunnhumby Complete Journey and Ta-Feng datasets. The task [...] Read more.

Grocery shopping is highly repetitive, yet customers also introduce new products into their buying routine over time, creating a repeat-explore challenge for next-basket recommendation. This study examines this trade-off in offline grocery recommendation using the Dunnhumby Complete Journey and Ta-Feng datasets. The task is formulated at the household-product level and evaluated using overall, repeat-specific, and explore-specific Top-K metrics. We compare global and personal frequency baselines with GraphSAGE, GCN, LightGCN, and weighted and time-decayed graph variants, and propose two hybrid models: TL-HFE, which combines an exploitation-oriented LightGCN ranking over the learnable catalog with a history-filtered exploration head, and TL-PPR, which combines a non-parametric personal-popularity repeat head with a LightGCN-based exploration head and household-specific quota interleaving. On Dunnhumby, Personal Top-Frequency achieves Overall, Repeat, and Explore Recall@20 values of 0.20, 0.40, and 0.00, while TL-PPR achieves 0.15, 0.29, and 0.02. On Ta-Feng, TL-PPR achieves 0.19, 0.67, and 0.07, compared with 0.18, 0.80, and 0.03 for Personal Top-Frequency. Paired bootstrap tests confirm that TL-PPR significantly improves exploratory recommendation over Personal Top-Frequency, although repeat recovery remains stronger for the baseline. Overall, the findings show that grocery NBR should be evaluated through separate repeat and explore perspectives rather than aggregate accuracy alone, especially when product discovery is a practical objective. Full article

(This article belongs to the Section Big Data and Augmented Intelligence)

31 pages, 3087 KB

Open AccessArticle

Toward Secure Software-Defined Industrial Networks Through Asset Administration Shell Digital Twins

by Riccardo Bacca, Andrea Melis, Lorenzo Rinieri, Roberto Girau, Marco Prandini and Franco Callegati

Future Internet 2026, 18(7), 347; https://doi.org/10.3390/fi18070347 - 30 Jun 2026

Abstract

Industrial digitalization is moving from Industry 4.0 toward Industry 5.0’s emphasis on resilience, human-centric operation, and sustainability. This shift is enabled by the convergence of Operational Technology and Information Technology, but this integration also broadens the exposure of industrial infrastructures to cyber threats [...] Read more.

Industrial digitalization is moving from Industry 4.0 toward Industry 5.0’s emphasis on resilience, human-centric operation, and sustainability. This shift is enabled by the convergence of Operational Technology and Information Technology, but this integration also broadens the exposure of industrial infrastructures to cyber threats targeting communication integrity and process continuity. Mitigating these risks requires network control that is both programmable and aware of each asset’s operational context. However, there is still a lack of operational interfaces that translate the semantics of industrial assets into programmable, runtime-enforceable network behavior. In this paper, following a Design Science Research methodology, we introduce an asset-aware, closed-loop network control abstraction in which the industrial network itself is modeled as a managed asset through Asset Administration Shells. Asset state, lifecycle phase, and operational intent are translated into network policies enforced at runtime on programmable data planes, while in-network telemetry is exposed at the asset level and correlated with operational metrics. We validate the abstraction on a hybrid testbed that combines virtualized components with industrial-grade hardware and virtualized 5G connectivity, through three security-oriented use cases: (i) asset-driven customization of forwarding policies; (ii) human-centric secure maintenance with controlled remote access over 5G; and (iii) anomaly detection and isolation based on cross-layer telemetry correlation. The results show that asset-level operations can drive programmable network enforcement and make network telemetry available at the asset layer. Finally, the work outlines a first step toward standardizing network-oriented asset submodels by separating control-plane operations from data-plane state and telemetry. Full article

(This article belongs to the Special Issue Artificial Intelligence and Control Systems for Industry 4.0 and 5.0)

► Show Figures

Figure 1

27 pages, 3739 KB

Open AccessArticle

Credibility Context Improves Political Claim Classification on LIAR and Transfers to LIAR2 NEW

by Bushra Alkomah and Frederick Sheldon

Future Internet 2026, 18(7), 346; https://doi.org/10.3390/fi18070346 - 30 Jun 2026

Abstract

Automated screening of political claims is challenging because many statements are short, underspecified, and labeled with fine-grained truthfulness categories that are difficult to separate using claim text alone. This study examines whether lightweight credibility context, represented by pre-existing speaker fact-checking history counts, can [...] Read more.

Automated screening of political claims is challenging because many statements are short, underspecified, and labeled with fine-grained truthfulness categories that are difficult to separate using claim text alone. This study examines whether lightweight credibility context, represented by pre-existing speaker fact-checking history counts, can improve transformer-based political claim classification without external evidence retrieval. We evaluate the LIAR benchmark under both its native six-class formulation and a coarser three-class mapping using two strong pretrained encoders, DeBERTa-v3-large and RoBERTa-large. To isolate the effect of credibility context, we vary only one input factor: whether the five LIAR speaker-history counts (pants-fire, false, barely-true, half-true, mostly-true) are appended to the claim and standard metadata as structured text (Hist ON) or omitted (Hist OFF), while keeping the data split, model architecture, training pipeline, and evaluation protocol fixed. All experiments are repeated across five random seeds (7, 13, 42, 123, 2024) and reported as mean ± standard deviation. On LIAR, Hist ON improves macro-F1 and accuracy across both backbones and both label granularities, with the largest gains in the six-class setting where label ambiguity is highest. In the six-class task, macro-F1 increases from 0.3096 to 0.4773 for DeBERTa-v3-large and from 0.3127 to 0.4855 for RoBERTa-large. In the three-class task, the best Hist ON model reaches 0.6241 macro-F1. Because only five seeds are available, the minimum achievable two-sided paired Wilcoxon p-value is 0.0625; therefore, we do not claim conventional

α = 0.05

statistical significance and instead report paired mean differences, seed-level gains, and complementary prediction-level reliability analyses from saved test predictions. To assess whether the benefit is limited to the original LIAR test split, we further evaluate the LIAR-trained checkpoints directly on the full non-overlapping LIAR2 NEW test split without additional fine-tuning. This LIAR-to-LIAR2 NEW transfer evaluation shows that Hist ON improves macro-F1 over Hist OFF in all four backbone/granularity settings. The best absolute transferred macro-F1 is achieved by the three-class DeBERTa-v3-large setting (0.6462), whereas the largest Hist ON minus Hist OFF gain occurs in the six-class RoBERTa-large setting (+0.1433 macro-F1). These two values describe different quantities: absolute performance and improvement over the corresponding Hist OFF baseline. We frame the task as political claim classification, or screening, rather than evidence-grounded fact verification: the method uses speaker-level credibility priors and does not retrieve external evidence. The results support speaker-history credibility context as a low-cost, model-agnostic signal for improving claim screening, while the LIAR2 NEW findings should be interpreted as related-benchmark robustness rather than universal out-of-domain generalization. Full article

(This article belongs to the Special Issue Artificial Intelligence (AI) and Natural Language Processing (NLP))

► Show Figures

Figure 1

21 pages, 3311 KB

Open AccessArticle

A Hybrid CNN–LSTM Model for IoT Intrusion Detection: A Robustness Analysis Across Datasets

by Amir Muhammad Hafiz Othman, Mohd Faizal Ab Razak, Ahmad Firdaus, Hamid Tahaei and Mehdi Gheisari

Future Internet 2026, 18(7), 345; https://doi.org/10.3390/fi18070345 - 30 Jun 2026

Abstract

The rapid growth of Internet of Things (IoT) devices has led to security concerns due to increasing IoT attacks. Traditional intrusion detection systems (IDS) struggle to effectively detect attacks due to the evolving nature of threats and heterogeneous traffic patterns. Therefore, this study [...] Read more.

The rapid growth of Internet of Things (IoT) devices has led to security concerns due to increasing IoT attacks. Traditional intrusion detection systems (IDS) struggle to effectively detect attacks due to the evolving nature of threats and heterogeneous traffic patterns. Therefore, this study presents a structured and reproducible intrusion detection approach that integrates preprocessing and deep learning-based classification for binary detection in IoT networks. The datasets used are ToN_IoT and UNSW-NB15 datasets, which contain IoT network traffic data. This study deploys a meta-heuristic algorithm called Gray Wolf Optimizer (GWO) for feature selection. SMOTE is used for balancing the class sample, and MinMax and standard normalization for data scaling during preprocessing. A comparative analysis is performed across multiple deep learning models, including Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM), Multi-Layer Perceptron (MLP), Deep Neural Network (DNN), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN). Results show that the CNN–LSTM model demonstrates strong performance consistency across datasets, achieving 99.68% and 92.05% accuracy on ToN_IoT and UNSW-NB15, respectively. Threshold sensitivity analysis reveals key detection and false-positive trade-offs for edge IDS. Through extensive performance evaluation and sensitivity analysis, this study highlights the importance of combining preprocessing, model evaluation, and threshold analysis for reliable IoT intrusion detection. Full article

(This article belongs to the Special Issue Security and Privacy Issues in the Internet of Cloud—2nd Edition)

► Show Figures

Figure 1

23 pages, 311 KB

Open AccessArticle

Evaluating Adversarial Robustness of Deepfake Audio Detectors and Vocoder Fingerprint Detectors Against Universal Adversarial Perturbations

by Quang Minh Tran, Wei Zong, Yang-Wai Chow and Willy Susilo

Future Internet 2026, 18(7), 344; https://doi.org/10.3390/fi18070344 - 29 Jun 2026

Abstract

Audio deepfake and vocoder fingerprint detectors are increasingly used to identify synthetic speech and attribute it to its generating model. However, their robustness against adversarial perturbations remains unclear across attack algorithms, perturbation domains, detector representations, and vocoder types. This paper presents a focused, [...] Read more.

Audio deepfake and vocoder fingerprint detectors are increasingly used to identify synthetic speech and attribute it to its generating model. However, their robustness against adversarial perturbations remains unclear across attack algorithms, perturbation domains, detector representations, and vocoder types. This paper presents a focused, quality-aware evaluation of four representative adversarial attacks, namely the Fast Gradient Sign Method (FGSM), Basic Iterative Method (BIM), Projected Gradient Descent (PGD), and Carlini–Wagner (CW) attack, against audio deepfake and vocoder fingerprint detectors. Each attack is implemented in both the waveform domain and the short-time Fourier transform (STFT) magnitude domain. All attacks are optimized against Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks (AASIST) under a targeted fake-to-real objective and are evaluated on synthetic speech generated by HiFi-GAN, Fullband MelGAN, StyleMelGAN, and Parallel WaveGAN. Attack performance is first measured on the source AASIST detector, after which black-box transferability is assessed on three target detector families: ResNet with Linear Frequency Cepstral Coefficient (LFCC) features, LCNN with Constant-Q Cepstral Coefficient (CQCC) features, and a bidirectional long short-term memory (BiLSTM) detector. The results show that adversarial effectiveness depends strongly on perturbation domain and detector representation. STFT-magnitude PGD transfers strongly to LFCC-based ResNet detectors but has limited effect on CQCC-based and recurrent detectors. In contrast, waveform-domain attacks produce broader transferability across feature-based detectors, with different attacks showing distinct ASR–quality trade-offs. Under the chosen waveform-domain budget, FGSM and BIM preserve transcription-level intelligibility while retaining meaningful black-box transferability, whereas CW provides the strongest overall source-detector and black-box attack performance. To distinguish effective adversarial perturbations from destructive signal degradation, we evaluate audio quality and intelligibility using word error rate (WER) and signal-to-noise ratio (SNR). Overall, the findings show that robustness claims in audio deepfake and vocoder fingerprint detection are limited when adversarial perturbations, black-box transferability, and audio quality are jointly considered. Full article

(This article belongs to the Special Issue Adversarial Attacks and Cyber Security)

► Show Figures

Graphical abstract

43 pages, 2827 KB

Open AccessArticle

MS-SENet: A Multi-Scale Squeeze–Excitation Network for Deep-Learning-Based Automatic Modulation Classification in Cognitive Radio Systems

by Evelio Astaiza Hoyos, Héctor Fabio Bermúdez-Orozco and Nasly Cristina Rodriguez-Idrobo

Future Internet 2026, 18(7), 343; https://doi.org/10.3390/fi18070343 - 29 Jun 2026

Abstract

Automatic modulation classification (AMC) is a critical enabler of cognitive radio (CR) systems, allowing secondary users to identify primary user modulation schemes and adapt transmission parameters in real time. Traditional AMC approaches, based on likelihood functions or hand-crafted features, suffer from degraded performance [...] Read more.

Automatic modulation classification (AMC) is a critical enabler of cognitive radio (CR) systems, allowing secondary users to identify primary user modulation schemes and adapt transmission parameters in real time. Traditional AMC approaches, based on likelihood functions or hand-crafted features, suffer from degraded performance under low signal-to-noise ratio (SNR) conditions and realistic channel impairments. In this paper, we propose MS-SENet (Multi-Scale Squeeze–Excitation Network), a novel deep-learning architecture that integrates multi-scale convolutional feature extraction, squeeze-and-excitation channel attention, residual learning, bidirectional long short-term memory (BiLSTM) temporal modelling, and global attention pooling into a unified framework for robust AMC. The multi-scale convolution module employs parallel branches with kernel sizes of 3, 5, and 7 to capture both fine-grained phase transitions and coarse envelope patterns from raw in-phase/quadrature (I/Q) signal samples. Squeeze–excitation residual blocks perform channel-wise feature recalibration, enabling the network to emphasize informative feature maps while suppressing less relevant ones. A bidirectional LSTM layer models temporal dependencies across the signal sequence, and a global attention pooling mechanism performs weighted temporal aggregation prior to classification. We present a comprehensive taxonomy of deep-learning architectures for AMC organised along five axes—input representation, feature extraction, temporal modelling, regularization strategy, and architectural complexity—and conduct a rigorous comparative evaluation against ten baseline architectures on a RadioML-style synthetic dataset (110,000 samples, 11 modulation classes, and 20 SNR levels from −20 to +18 dB). The experimental results demonstrate that MS-SENet achieves a mean classification accuracy of 87.9% at SNR ≥ 0 dB (the average of the medium and high SNR regime averages: 86.06% for 0 ≤ SNR < 10 dB and 89.68% for SNR ≥ 10 dB) while maintaining a compact footprint of approximately 406 K parameters, making it suitable for deployment on resource-constrained edge devices. We further analyze the robustness of the proposed architecture to multipath fading, carrier frequency offset, and sample rate offset, confirming its resilience under practical operating conditions. MS-SENet is an architecture designed for automatic modulation classification of I/Q signals and is not related to the homonymous architecture for speech emotion recognition. Full article

5 pages, 143 KB

Open AccessEditorial

Advances in Smart Environments and Digital Twin: Current Trends and Future Directions

by Mariusz Żytniewski, Małgorzata Pańkowska, Mariia Rizun and Francesca Fallucchi

Future Internet 2026, 18(7), 342; https://doi.org/10.3390/fi18070342 - 29 Jun 2026

Abstract

Ubiquitous miniaturisation and increasing computer performance allow technical solutions to be used more often in everyday life to support end users [...] Full article

(This article belongs to the Special Issue Advances in Smart Environments and Digital Twin Technologies)

22 pages, 8185 KB

Open AccessArticle

Graph-Enhanced Transformer for Cross-Domain Sentiment Analysis: Integrating RoBERTa with Graph Attention Networks

by Moteechand Patel, Abhinav Shukla, Pritendra Kumar Malakar, R. Kanesaraj Ramasamy and Parul Dubey

Future Internet 2026, 18(7), 341; https://doi.org/10.3390/fi18070341 - 29 Jun 2026

Abstract

Sentiment analysis has become a critical task in natural language processing for extracting subjective insights from large-scale textual data across domains such as social media, e-commerce, and online reviews. However, existing methods often fail to simultaneously capture contextual semantics and structural relationships, particularly [...] Read more.

Sentiment analysis has become a critical task in natural language processing for extracting subjective insights from large-scale textual data across domains such as social media, e-commerce, and online reviews. However, existing methods often fail to simultaneously capture contextual semantics and structural relationships, particularly in cross-domain settings. This study proposes a hybrid RoBERTa–graph attention network (GAT) framework that integrates transformer-based contextual embeddings with graph-based relational learning. The methodology involves encoding text using RoBERTa, constructing token-level dependency graphs, and applying multi-head graph attention to model inter-token relationships. The model is evaluated on multiple benchmark datasets, including Twitter, Amazon, and IMDB reviews. The cross-domain results refer only to binary-harmonized positive/negative sentiment transfer and should not be interpreted as full three-class sentiment transfer, including the neutral class. The experimental results show that the proposed approach achieves consistent improvements over the selected baseline models in terms of accuracy, F1-score, MCC, and AUC. Statistical robustness analysis further supports the stability of these improvements across repeated runs. The findings highlight the effectiveness of combining semantic and structural learning, making the proposed framework suitable for robust cross-domain sentiment analysis applications. Full article

► Show Figures

Figure 1

38 pages, 1652 KB

Open AccessReview

Proximal Policy Optimization in 5G, B5G, and 6G Communication Systems: A Systematic Review

by Vijaya Kittu Manda, Bhukya Madhu and Theodore Tarnanidis

Future Internet 2026, 18(7), 340; https://doi.org/10.3390/fi18070340 - 27 Jun 2026

Abstract

Fifth-generation (5G), Beyond 5G (B5G), and sixth-generation (6G) wireless networks, along with the Internet of Things (IoT), are core communication infrastructure in smart cities. Their increased deployments create high-dimensional optimization and resource management challenges. Consequently, researchers have increasingly explored the use of Artificial [...] Read more.

Fifth-generation (5G), Beyond 5G (B5G), and sixth-generation (6G) wireless networks, along with the Internet of Things (IoT), are core communication infrastructure in smart cities. Their increased deployments create high-dimensional optimization and resource management challenges. Consequently, researchers have increasingly explored the use of Artificial Intelligence (AI) models for optimizing networks. The Proximal Policy Optimization (PPO) is one such algorithm that optimizes networks. This Systematic Literature Review (SLR) follows the PRISMA 2020 protocol to review 76 studies published between 2023 and 2026 to synthesize recent PPO-based approaches to optimize communication systems. This study examines key PPO variants in major communication domains. It outlines the primary obstacles to real-world deployment and provides a cross-domain classification. According to this study, PPO provides continuous action spaces with good training stability for AI models. Its stable policy-learning capabilities make it suitable for next-generation communication systems. However, sim-to-real transfer, reward design, and multi-agent scalability are a few key challenges encountered. Future directions emphasize robust, deployable PPO frameworks for 6G, IoT, and internet architecture. Full article

► Show Figures

Graphical abstract

30 pages, 531 KB

Open AccessArticle

Defeat Devices in AI Systems

by Emilio Ferrara

Future Internet 2026, 18(7), 339; https://doi.org/10.3390/fi18070339 - 27 Jun 2026

Abstract

AI systems increasingly exhibit behavior that differs systematically between evaluation and deployment contexts. Alignment faking, sandbagging, benchmark gaming, deceptive scheming, specification gaming, and trojans have each been documented separately, with each line of work characterizing one facet of what we argue is a [...] Read more.

AI systems increasingly exhibit behavior that differs systematically between evaluation and deployment contexts. Alignment faking, sandbagging, benchmark gaming, deceptive scheming, specification gaming, and trojans have each been documented separately, with each line of work characterizing one facet of what we argue is a single structural mechanism: we propose that this common mechanism is a defeat device, an engineering and regulatory concept long established in vehicle-emissions law and brought to broad public attention by the 2015 Volkswagen emissions case. A defeat device in an AI system has three necessary elements: a discriminator that detects evaluation context, a concealed swap that conditions behavior on detection, and a gap between eval-distribution and deployment-distribution performance on the stated evaluation criterion. We formalize this triadic test as a behavioral definition, organize documented cases along three taxonomic axes (origin, trigger, swap mechanism), propose Trigger-Axis-Aware Differential Probing (TADP) as a forensic detection protocol, and advance the claim that defeat devices can naturally emerge in current frontier AI systems without any operator engineering. We characterize naturally emerging defeat devices as potentially one of the harmful emerging phenomena that AI safety practice should monitor and test for systematically. An illustrative study applying TADP across eight open-weight models finds the discriminator to be near-universal (every model detects evaluation context well above chance), while the conditional swap is real but heterogeneous: it appears strongly as sycophantic stance-conditioning and as an evaluation-cued register shift, yet not as overt demographic discrimination, indicating that the mechanism’s discriminator generalizes even where individual swaps do not. Implications for evaluation methodology, post-training pipeline design, interpretability research priorities, and AI governance follow. Full article

(This article belongs to the Special Issue 2026 and 2027 Feature Papers from Future Internet’s Editorial Board Members)

► Show Figures

Figure 1

39 pages, 3377 KB

Open AccessArticle

International Digital System for Collective Food Security Support

by Maxim Logachev and Vitaliy Fomin

Future Internet 2026, 18(7), 338; https://doi.org/10.3390/fi18070338 - 26 Jun 2026

Abstract

(1) Background. Food sovereignty and local sustainability are ensured by large agro-industrial holdings and small-scale farms; this synergy forms a complementary model of the agrifood system. Maintaining this model’s balance requires the creation of a unified digital ecosystem that integrates all suppliers and [...] Read more.

(1) Background. Food sovereignty and local sustainability are ensured by large agro-industrial holdings and small-scale farms; this synergy forms a complementary model of the agrifood system. Maintaining this model’s balance requires the creation of a unified digital ecosystem that integrates all suppliers and consumers into production chains, thereby eliminating unnecessary intermediaries. (2) Methods. This study employs a comprehensive methodological framework, including systems analysis and mathematical modeling, to develop service algorithms. Object-oriented design and software engineering methods facilitated the development and implementation of a service-oriented architecture for the digital system. (3) Results. The study presents a multi-tier architecture featuring an integration bus based on a service-oriented approach. To implement direct supply-and-demand coupling strategies, the system integrates both internal services (microeconomic indicators) and external services (macroeconomic indicators). Additionally, a recommender system based on neural networks and mathematical models was developed to personalize consumer requests and manage product sales. (4) Conclusions. The software solution is consistent with the AgTech 4.0 concept and enables the creation of a seamless environment for interstate trade. The implementation of the system enhances the transparency of the “product footprint”, facilitates the redistribution of surpluses, and, consequently, contributes to price stabilization. Full article

(This article belongs to the Special Issue ICT and AI in Intelligent E-Systems—2nd Edition)

24 pages, 2593 KB

Open AccessArticle

Regional Strategy Composition: A Hierarchical-Action Reinforcement Learning Framework for Dynamic Smart-Meter Association over 5G NR mMTC Networks

by Muhammed Al-Ali, Esteban Inga, Juan Inga and Elias Yaacoub

Future Internet 2026, 18(7), 337; https://doi.org/10.3390/fi18070337 - 25 Jun 2026

Abstract

Advanced Metering Infrastructure (AMI) over 5G New Radio (NR) massive machine-type communication (mMTC) networks require efficient and adaptive communication mechanisms to support reliable data delivery for large numbers of smart meters under dynamic traffic and channel conditions. In this work, we propose a [...] Read more.

Advanced Metering Infrastructure (AMI) over 5G New Radio (NR) massive machine-type communication (mMTC) networks require efficient and adaptive communication mechanisms to support reliable data delivery for large numbers of smart meters under dynamic traffic and channel conditions. In this work, we propose a framework in which each smart meter chooses, at runtime, whether to transmit directly to the base station (BS) or via a nearby Data Aggregation Point (DAP). The optimal choice is dynamic and depends on DAP buffer occupancy, periodic congestion, channel quality, and packet deadline pressure. Formulating this as a per-meter binary decision yields an action space of size

2^{N}

for N meters, which is intractable for reinforcement learning (RL). We reformulate the problem as regional strategy composition: the RL agent selects one parameterized association strategy for each DAP region from a small library of interpretable rules, and a deterministic mapping expands the regional choice into per-meter modes. It reduces the policy action space from

2^{N}

to

K^{D}

, where D is the number of DAPs and K the number of strategies, while preserving meter-level control granularity. We evaluate Proximal Policy Optimization (PPO) and Deep Q-Network (DQN) controllers against eight meter-level baselines on a 5G NR-calibrated simulator with 1500 m, six DAPs, deadline-bounded delivery, stale channel-state information, and phase-offset congestion cycles. Across three traffic regimes and five random seeds, PPO improves packet delivery ratio (PDR) over the strongest heuristic by +0.63, +2.41, and +2.66 percentage points under baseline, high-load, and bursty-cycle conditions, respectively; all gains are statistically significant (paired t-test,

p < 0.001

; Cohen’s d up to 5.12), and the advantage grows with traffic stress. The results show that learned regional composition of classical heuristics outperforms any single fixed heuristic precisely when no individual rule is globally optimal. Full article

(This article belongs to the Special Issue Artificial Intelligence in Smart Grids)

► Show Figures

Figure 1

42 pages, 4791 KB

Open AccessArticle

Unpacking Internet-Based Social Engineering Victimisation on Social Networking Sites: An Interdisciplinary Qualitative Framework of Individual, Social, and Platform Factors

by Saad Saleh Alshammari, Ben Soh and Alice Li

Future Internet 2026, 18(7), 336; https://doi.org/10.3390/fi18070336 - 25 Jun 2026

Abstract

Despite extensive research on social engineering victimisation on social networking sites (SNSs) across the Internet, user susceptibility continues to increase, indicating that existing explanatory models remain incomplete. Previous studies have predominantly examined susceptibility through isolated factors, including individual traits, message characteristics, or source [...] Read more.

Despite extensive research on social engineering victimisation on social networking sites (SNSs) across the Internet, user susceptibility continues to increase, indicating that existing explanatory models remain incomplete. Previous studies have predominantly examined susceptibility through isolated factors, including individual traits, message characteristics, or source attributes, while often overlooking how evolving Internet-based SNS environments interact with human and social factors. To address this gap, this study presents an interdisciplinary qualitative investigation into emerging determinants of user susceptibility to social engineering cyberattacks (SECAs) on Internet-enabled SNS platforms. Drawing on in-depth interviews with 18 experts from cybersecurity, psychology, sociology, criminology, and linguistics, the study captures perspectives that are rarely integrated within a single analytical framework. Using NVivo 14 and inductive thematic analysis, six core themes and seven sub-themes were identified, revealing previously underexplored cognitive-emotional, social-relational, and platform-mediated mechanisms of victimisation. The key contribution of this research is not the identification of entirely new susceptibility factors, but the development of an interdisciplinary framework that integrates these previously disconnected dimensions. By foregrounding the role of SNS design affordances within the broader Internet ecosystem and their interaction with human cognition and social dynamics, this study advances current understanding beyond fragmented models of user vulnerability. The findings provide a novel conceptual foundation for future empirical research and inform the design of more effective, context-aware mitigation and awareness strategies for SECAs on Internet-based SNSs. Full article

(This article belongs to the Special Issue Adversarial Attacks and Cyber Security)

► Show Figures

Figure 1

44 pages, 2700 KB

Open AccessReview

Hybrid-Oriented Intelligent Operational and Architectural Foundations of IoT-Enabled Smart Grids: A System-Level Review and Challenge-Oriented Comparative Synthesis

by Grygorii Diachenko, Ivan Laktionov and Daniil Fainshtein

Future Internet 2026, 18(7), 335; https://doi.org/10.3390/fi18070335 - 24 Jun 2026

Abstract

The rapid digitalization of energy systems and the increasing integration of distributed energy resources, renewable energy technologies, and prosumer-oriented infrastructures have accelerated the development of IoT-enabled Smart Grids as a foundation for intelligent and adaptive energy management. Modern Smart Grids increasingly depend on [...] Read more.

The rapid digitalization of energy systems and the increasing integration of distributed energy resources, renewable energy technologies, and prosumer-oriented infrastructures have accelerated the development of IoT-enabled Smart Grids as a foundation for intelligent and adaptive energy management. Modern Smart Grids increasingly depend on the coordinated interaction of IoT architectures, artificial intelligence, distributed analytics, and decentralized control mechanisms to ensure reliability, scalability, and real-time operational flexibility. Despite extensive research activity, existing studies remain predominantly technology-centric, focusing on isolated architectural layers or individual intelligent methods without providing a unified system-level perspective on their coordinated operation and interoperability. This article presents a system-level integrative review and challenge-oriented comparative synthesis of intelligent operational and architectural foundations of IoT-enabled Smart Grids. The study analyzes data-driven, model-driven, knowledge-driven, agent-based, and hybrid-oriented intelligent paradigms within multi-layer IoT energy infrastructures. In addition, the research establishes a cross-layer mapping between Smart Grid operational challenges, enabling technologies, and corresponding analytical approaches while identifying interoperability constraints, scalability limitations, and coordination challenges associated with decentralized energy ecosystems. The conducted synthesis demonstrates that hybrid-oriented intelligent approaches represent the most promising direction for future Smart Grid evolution due to their ability to integrate AI, ML, digital twins, semantic reasoning, and decentralized multi-agent coordination within unified IoT architectures. The conducted comparative synthesis identifies the ongoing transition from isolated intelligent solutions toward integrated hybrid cyber–physical energy ecosystems and highlights key characteristics of future adaptive, interoperable, scalable, and explainable Smart Grid architectures. Full article

► Show Figures

Graphical abstract

35 pages, 4344 KB

Open AccessArticle

From Opaque Streams to Explainable Systems: Semantic MQTT Integration at the Edge

by Niklas Doerner and Maria Maleshkova

Future Internet 2026, 18(7), 334; https://doi.org/10.3390/fi18070334 (registering DOI) - 24 Jun 2026

Abstract

Industrial systems increasingly rely on MQTT-based message streaming to enable automated, data-driven production processes at the network edge. While semantic models such as the SSN/SOSA ontology enable machine-interpretable descriptions of observations and actuations, an explicit model of message transport is rarely considered. Consequently, [...] Read more.

Industrial systems increasingly rely on MQTT-based message streaming to enable automated, data-driven production processes at the network edge. While semantic models such as the SSN/SOSA ontology enable machine-interpretable descriptions of observations and actuations, an explicit model of message transport is rarely considered. Consequently, MQTT-based communication remains opaque, particularly regarding information processing, hindering the semantic analysis of application-specific topic structures and the behavior of transport protocols. To close this gap, this work introduces the revised MQTT4SSN ontology as a key contribution, extending existing semantic models with protocol-aware representations of MQTT entities, control packets, and transport-level interactions. MQTT4SSN enables end-to-end semantic traceability, from sensor observations and actuator controls to the underlying message transmission within distributed systems. Building on this contribution, the MQTT2RDF integration framework incorporates MQTT4SSN as its core to capture live MQTT traffic and represent both payload meaning and transport-level provenance within an RDF knowledge graph. This work presents a novel approach for representing edge computing and information processing over MQTT, addressing two key challenges. First, the framework supports semantic interpretation of topic hierarchies and provides configurable mappings between MQTT topics, payload structures, and observation or actuation semantics. This approach facilitates the setup of edge computing systems and enables context-aware subscription management and structured data formatting, thereby improving interoperability between heterogeneous deployments. Second, transport-level provenance analytics provide a semantic basis for query-based detection, classification support, and diagnostic analysis of malformed or incomplete MQTT communication. The approach provides explainable, traceable information processing through transport provenance, which is essential for safety-critical industrial environments. The contributions are validated through an industrial use case from a production environment, demonstrating its applicability for system monitoring, troubleshooting, and semantic analytics of MQTT-based infrastructures. Full article

(This article belongs to the Special Issue Intelligent Computing and Information Processing)

► Show Figures

Graphical abstract

32 pages, 2519 KB

Open AccessArticle

Feature Selection for Improving ANN and CNN Models for Attack Detection in Zeek Network Data

by Sikha S. Bagui, Mohamed Elbatouty, Dustin Mink and Subhash C. Bagui

Future Internet 2026, 18(7), 333; https://doi.org/10.3390/fi18070333 - 24 Jun 2026

Abstract

In the past few years, cyber-attacks have risen at an exponential rate across all sectors, and both private and public institutions have faced increasingly sophisticated threats. As this upward trend continues, the need for advanced and efficient threat detection systems is essential. This [...] Read more.

In the past few years, cyber-attacks have risen at an exponential rate across all sectors, and both private and public institutions have faced increasingly sophisticated threats. As this upward trend continues, the need for advanced and efficient threat detection systems is essential. This paper investigates the use of feature importance (FI) Coefficients to improve Artificial Neural Network (ANN) and Convolutional Neural Network (CNN) models, leveraging feature selection to enhance model interpretability and optimize performance. By systematically filtering out the weaker features, we examine the reduced features’ impact on model accuracy, precision, recall, and F1 score. Experiments were conducted on two new datasets, UWF-ZeekDataSum2025-1 and UWF-ZeekDataSum2025-2, using a baseline ANN/CNN architecture and multiple architectural variants. The results on UWF-ZeekDataSum2025-1 show a clear performance gain for certain feature importance thresholds, with models such as ANN-Minimal, ANN-Overfit-Wide, ANN-Shallow-Low-Optimization, CNN-Shallow, and CNN-Very-Shallow outperforming the baseline after reducing the feature space from seventeen features to fewer than four. For UWF-ZeekDataSum2025-2, improvements occur across a broader range of thresholds, with models including ANN-Deep-Sub-Conv, ANN-Shallow-Low-Opt, CNN-Shallow, CNN-Very-Shallow, and ANN-Minimal exceeding 95% performance around the 0.25–0.28 thresholds, with additional gains at 0.31–0.32 for some architectures. These findings demonstrate that by strategically leveraging feature importance coefficient thresholds, we can significantly enhance neural network intrusion detection systems, offering a reproducible pathway for adapting these methods on similar environments. Full article

(This article belongs to the Special Issue State-of-the-Art Future Internet Technology in USA 2026–2027)

► Show Figures

Figure 1

Journal Description

Future Internet

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI