ICT Express最新文献

{"title":"Attestation-based verification of SBOM integrity via consumer-side reproducibility","authors":"Yiseul Choi,&nbsp;Junga Kim,&nbsp;Jun-Ho Hong,&nbsp;Seongmin Kim","doi":"10.1016/j.icte.2026.06.009","DOIUrl":"10.1016/j.icte.2026.06.009","url":null,"abstract":"<div><div>Software bills of materials (SBOMs) support supply chain transparency, but they do not prove that a delivered SBOM reproducibly corresponds to its software artifact. Existing signing and provenance mechanisms protect integrity and traceability, yet lack consumer-side reproducible verification. We propose an SBOM integrity verification framework combining procedure disclosure, consumer-side reproduction, authority-generated reference evidence, and digest comparison. A trusted authority records a reference digest, and consumers compare it with locally reproduced and delivered SBOM digests. Experiments on 100 real-world container images show detection of artifact tampering, SBOM substitution, distribution modification, and adaptive tampering beyond signature-based approaches.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 928-932"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ADIM-DETR: An adaptive dynamic hybrid convolution-based method for road defect detection","authors":"Yuanyuan Wang,&nbsp;Qiyuan Jiang,&nbsp;Jiahui Lei,&nbsp;Guxue Gao,&nbsp;Jibin Wang,&nbsp;Tingmei Ma,&nbsp;Zhiwen Zhuang,&nbsp;Yuchen Cai,&nbsp;Xinye Miao","doi":"10.1016/j.icte.2025.11.019","DOIUrl":"10.1016/j.icte.2025.11.019","url":null,"abstract":"<div><div>Road defect detection is challenged by the complex geometries of defects and the inefficiency of multi-scale feature fusion in existing models. To address this, we propose ADIM-DETR, an adaptive dynamic feature interaction detection model that employs a CSPNet backbone with an improved C2f_DCMB module for adaptive feature fusion, a TEAA encoder for efficient long-range dependency modeling, and a improved RetBlockC3 fusion module for enhanced positional awareness. Experiments on public and self-built datasets demonstrate gains in accuracy and efficiency, achieving up to 96.2 % [email protected] on UAV-PDD2023 and 68.2 % [email protected] on RCDV2, while reducing GFLOPs to 35.7.</div><div>© 2025 The Author(s). Published by Elsevier B.V. on behalf of The Korean Institute of Communications and Information Sciences. This is an open access article under the CC BY-NC<img>ND license (<span><span>http://creativecommons.org/licenses/by-nc-nd/4.0/</span><svg><path></path></svg></span>).</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 759-766"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147880228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"URLLC for 6G enabled industry 5.0: A taxonomy of architectures, cross layer techniques, and time critical applications","authors":"Abdikarim Mohamed Ibrahim,&nbsp;Rosdiadee Nordin,&nbsp;Yahya S.M. Khamayseh,&nbsp;Angela Amphawan,&nbsp;Muhammed Basheer Jasser","doi":"10.1016/j.icte.2026.05.008","DOIUrl":"10.1016/j.icte.2026.05.008","url":null,"abstract":"<div><div>Ultra reliable low latency communication (URLLC) is central to 6G systems that must support the applications of Industry 5.0. This paper surveys URLLC techniques and architectures and organizes them in a taxonomy covering objectives, mechanisms, challenges, and performance measures. We review cross layer methods, finite blocklength regimes, and scheduling and orchestration at the edge, and summarize evidence on latency and reliability trade-offs and determinism. The survey links techniques to industrial requirements and highlights where current assumptions fail. Finally, we outline open issues and near-term directions for secure, scalable, and energy-aware industrial URLLC.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 789-809"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multimodal data fusion in deep learning for diabetic retinopathy detection: A systematic review","authors":"Kartina Diah Kesuma Wardhani,&nbsp;Shahreen Kasim,&nbsp;Rohayanti Hassan,&nbsp;Rahmat Hidayat,&nbsp;Khaled Mahmud Sujon,&nbsp;Md Abdus Samad","doi":"10.1016/j.icte.2026.06.001","DOIUrl":"10.1016/j.icte.2026.06.001","url":null,"abstract":"<div><div>Multimodal data fusion has emerged as a promising approach for improving the accuracy and robustness of diabetic retinopathy (DR) detection by integrating complementary information from heterogeneous data sources. This review systematically examines recent deep learning-based multimodal approaches for DR detection, focusing on model architectures, modality combinations, fusion strategies, and reported performance outcomes. The analysis reveals that while convolutional neural networks (CNNs) remain dominant, hybrid architectures are increasingly adopted to handle heterogeneous inputs. Combinations such as fundus with OCT or electronic health records (EHR) are frequently associated with improved performance, although results remain highly dependent on dataset characteristics and experimental design. Across studies, early and joint fusion strategies are commonly applied, but no single approach consistently outperforms others, highlighting the context-dependent nature of fusion design. Overall, this review identifies key trends, methodological gaps, and emerging directions, providing a clearer foundation for developing robust and clinically applicable multimodal DR detection systems.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 899-916"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710179","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Federated learning in healthcare: A comprehensive survey on privacy, scalability and clinical applications","authors":"Abtahi Noor,&nbsp;Ummay Maimona Chaman,&nbsp;Md Saif Kabir","doi":"10.1016/j.icte.2026.05.011","DOIUrl":"10.1016/j.icte.2026.05.011","url":null,"abstract":"<div><div>The growing accessibility of medical data, coupled with advances in artificial intelligence, has accelerated the development of data-driven healthcare solutions. Deep learning models are highly effective in this context, as they are inherently designed to leverage data-driven approaches, with performance improving as larger datasets become available. While these models have demonstrated significant success, reliance on centralized data aggregation raises concerns regarding patient privacy, data heterogeneity, communication overhead, and system scalability. Federated Learning (FL) has emerged as a promising paradigm that enables collaborative model training across institutions while preserving data confidentiality. This survey presents a comprehensive review of 80 studies published between 2020 and 2026, highlighting the state of the art in FL applications for healthcare. It classifies research across multiple dimensions, including standard FL variants and aggregation strategies, communication-efficient methods, peer-to-peer FL, blockchain-based frameworks, explainable AI integration, deep learning and reinforcement learning architectures, hybrid optimization approaches, and specialized healthcare applications. Additionally, the survey examines recent directions in vertical and split FL, FL frameworks and platforms, and the integration of large language models (LLMs) and retrieval-augmented generation (RAG) within FL systems. Through this structured analysis, key challenges such as privacy risks, interoperability, communication bottlenecks, and lack of clinical validation are identified and mapped to potential solutions. Finally, a research roadmap is provided, outlining future directions in privacy-preserving techniques, scalability, data standardization, and clinical adoption. By bridging technological advances with healthcare requirements, this survey offers researchers and practitioners a concise reference for advancing secure, scalable, and intelligent healthcare systems through federated learning.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 825-839"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint acceleration and steering control via deep Q-network for high-precision path tracking of autonomous vehicles","authors":"Seung Geon Yang,&nbsp;Jeongyun Kim,&nbsp;Seung-Chan Lim","doi":"10.1016/j.icte.2026.06.008","DOIUrl":"10.1016/j.icte.2026.06.008","url":null,"abstract":"<div><div>This study presents a deep reinforcement learning framework for high-precision autonomous vehicle path tracking under stochastic driving environments. The path-tracking problem is formulated as a task-oriented Markov decision process (MDP) with a discretized dual-action structure that jointly models acceleration and steering decisions, enabling coordinated optimization of longitudinal and lateral vehicle dynamics within an unified control framework. The proposed joint deep Q-network (J-DQN) learns the control policy directly through interaction with the environment without relying on explicit system models. Experimental evaluations demonstrate that the proposed approach achieves superior tracking accuracy, stability, and robustness compared with representative baseline methods under realistic driving scenarios.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 923-927"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Forecasting before acting: Recent advances in learning-based traffic prediction and intelligent control in Open RAN","authors":"Thanh Thien-An Dang,&nbsp;Dongwook Won,&nbsp;Juyoung Kim,&nbsp;Viet Anh-Huy Le,&nbsp;Sungrae Cho","doi":"10.1016/j.icte.2026.07.007","DOIUrl":"10.1016/j.icte.2026.07.007","url":null,"abstract":"<div><div>Open Radio Access Network (O-RAN) disaggregates the RAN into standardized units coordinated by Near-Real-Time (Near-RT) and Non-Real-Time (Non-RT) RAN Intelligent Controllers (RICs), enabling closed-loop optimization. Proactive control requires accurate prediction of traffic demand, user equipment (UE) mobility, and resource consumption before congestion or service level agreement violations occur. Yet machine learning (ML)-based proposals remain fragmented across granularities, interface placements, and evaluation environments, with no unified framework for comparing designs or assessing evidence quality.</div><div>This paper surveys recent works on ML-based traffic and UE behavior prediction that explicitly feed O-RAN control loops. We classify each work along four axes: (i) ML tier, separating architectures with structural inductive bias from standard deep learning and classical methods; (ii) traffic target at UE, slice, and cell granularities; (iii) evaluation environment across live deployments, emulation, and simulation; and (iv) O-RAN interface placement across A1, E2, O1, and O2.</div><div>Our analysis reveals four headline findings. First, a <em>forecast-then-act</em> pipeline, composing a predictor with a downstream deep reinforcement learning or optimization controller, dominates closed-loop design across granularities. Second, standard deep learning predominates, whereas architectures with structural inductive bias remain a minority. Third, evaluation is dominated by custom, unstandardized simulators, with no surveyed work performing live closed-loop actuation on an operational deployment. Fourth, open-source reproducibility is near-absent, with only one complete release of code, real data, and evaluation scripts. We then identify open challenges in per-UE forecasting, standardized data exposure, and open-source evaluation infrastructure, and outline a research roadmap toward 6G predictive intelligence in O-RAN.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 1041-1061"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148709968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LEAP:VANET: A lightweight and efficient authentication protocol for intelligent transportation system using VANET","authors":"Ayman Yafoz,&nbsp;Raed Alsini,&nbsp;Alaa Omran Almagrabi","doi":"10.1016/j.icte.2026.03.013","DOIUrl":"10.1016/j.icte.2026.03.013","url":null,"abstract":"<div><div>Vehicular Ad-Hoc Networks (VANETs) are an important component of intelligent transportation systems (ITS). However, the wireless communication environment of the networks is exposed to impersonation, replay, and privacy attacks. The authentication protocols designed for the networks are associated with high computational costs, weak binding of the authentication protocol with the session key, and the use of plaintext identity. Hence, this paper introduces LEAP-VANET, a lightweight and efficient authentication protocol designed specifically for VANETs. The proposed protocol is based on the SHA-256 hashing algorithm and the XOR operation. The Real-or-Random (RoR) model is used for the security analysis of the protocol under the Canetti-Krawczyk (CK) model. The simulation results obtained by ProVerif show that the protocol is secure against replay attacks, man-in-the-middle attacks, and desynchronization attacks. The proposed protocol ensures unlinkability under the semi-honest model and provides forward secrecy. Performance evaluation of the scheme on heterogeneous hardware shows significant improvements in terms of 82.4% reduction in computation time (4.59 ms total time), 29.5% reduction in communication overhead (5024 bits), 43.5% reduction in storage requirements (2176 bits), and 99.9% reduction in energy consumption (0.16 mJ) compared to prior works. The comprehensive comparative analysis validates the performance consistency across different network conditions and demonstrates the effectiveness of LEAP-VANET as a viable solution for resource-constrained VANET environments without compromising security and privacy concerns.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 767-772"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Energy optimization in AIoT-based healthcare monitoring using reinforcement learning variational quantum algorithm","authors":"P Jose,&nbsp;P Tamilselvi,&nbsp;Shanmugasundaram Hariharan,&nbsp;J. Vijayalakshmi,&nbsp;A. Shameem Fathima,&nbsp;Shih-Yu Chen","doi":"10.1016/j.icte.2026.05.015","DOIUrl":"10.1016/j.icte.2026.05.015","url":null,"abstract":"<div><div>The integration of AIoT in healthcare allows continuous monitoring of patients and real-time clinical decision-making. Energy inefficiency of batteries in edge devices, as well as latency and accuracy limitations, are among the key concerns in the domain. In this study, a Reinforcement Learning (RL) driven Variational Quantum Algorithm (VQA)-based multi-objective approach (RL-VQA) to optimize AIoT for energy efficiency by incorporating quantum circuit parameter optimization is introduced. In this context, the intelligent optimization of the system-level design is attained through dynamic adjustment of variational parameters using reinforcement learning while considering energy efficiency, latency, and diagnostic accuracy. The problem setting is designed as a constrained optimization problem, allowing efficient search for possible solutions in the complex problem space through hybrid quantum-computational models. Results show that the approach yields up to 31.7% reduction in energy, 24 ms improvements in latency, and 1.1% higher accuracy in comparison to classical and static quantum approaches.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 840-845"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"UVFace: Utility Driven Video-based Face Recognition","authors":"Žiga Babnik,&nbsp;Peter Peer,&nbsp;Vitomir Štruc","doi":"10.1016/j.icte.2026.05.014","DOIUrl":"10.1016/j.icte.2026.05.014","url":null,"abstract":"<div><div>Face recognition methods are primarily designed for single-image analysis, even though video-based recognition has seen a dramatic increase in popularity in edge security and surveillance applications. Typically, a video template is constructed from the features of individual frames. Feature norms are commonly used as weights in the construction process, as they correlate well with the usefulness of samples for recognition. Classical training approaches directly optimize only the angular distances, in turn also guiding the feature norms. This can lead to suboptimal alignment between feature norms and the usefulness (utility) of samples, resulting in subpar video performance. Motivated by this insight, we propose the UVFace methodology, which presents an extended feature norm alignment branch. Through careful design of the quality ranking step, which produces feature norm labels and a new feature norm loss, UVFace improves performance over the reproduced AdaFace baseline on video-oriented benchmarks while retaining strong image-based performance. Code is available at <span><span>https://github.com/LSIbabnikz/UVFace</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"12 4","pages":"Pages 868-873"},"PeriodicalIF":4.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148710176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}